Professional Documents
Culture Documents
Christopher W. Mastropietro
Kevin M. Valentine
Editors
ERRNVPHGLFRVRUJ
Editors
Christopher W. Mastropietro Kevin M. Valentine
Riley Hospital for Children Riley Hospital for Children
Indiana University School of Medicine Indiana University School of Medicine
Indianapolis, IN Indianapolis, IN
USA USA
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Foreword
v
vi Foreword
of the optimal therapies to tackle the big problems that we face while limit-
ing toxicity and other unwanted side effects, some of which we do not even
(yet) recognize. I believe that this textbook will help us to achieve that
important goal.
vii
Contents
ix
ERRNVPHGLFRVRUJ
x Contents
ERRNVPHGLFRVRUJ
List of Contributors
xi
ERRNVPHGLFRVRUJ
xii List of Contributors
ERRNVPHGLFRVRUJ
List of Contributors xiii
ERRNVPHGLFRVRUJ
xiv List of Contributors
ERRNVPHGLFRVRUJ
Part I
Respiratory Controversies
ERRNVPHGLFRVRUJ
Ventilator Management
for Pediatric Acute Respiratory 1
Distress Syndrome
Travis P. Vesel and Ira M. Cheifetz
ERRNVPHGLFRVRUJ
4 T. P. Vesel and I. M. Cheifetz
survivors progress to a chronic fibrosing alveoli- the term “acute lung injury” (ALI) was elimi-
tis, characterized clinically by chronic hypox- nated from the stratification scheme in the
emia, increased alveolar dead space, and 2015 PALICC definition.
decreased pulmonary compliance. The disease severity of PARDS is initially
stratified based on noninvasive mechanical venti-
lation or invasive mechanical ventilation.
Definition of Pediatric ARDS Considering the increased use of noninvasive
mechanical ventilation (i.e., CPAP or BiPAP), the
In 2015, members of the Pediatric Acute Lung PALICC definition includes patients supported in
Injury Consensus Conference (PALICC) this manner; however, these patients are not strat-
developed the first reported pediatric-specific ified as mild/moderate/severe. In patients sup-
definition of ARDS (Fig. 1.1) [3]. Earlier defi- ported with invasive mechanical ventilation,
nitions of acute respiratory distress syndrome disease severity is stratified using oxygenation
include the American European Consensus index (OI) and oxygen saturation index (OSI).
Conference [4] and Berlin [5] definitions and Considering pediatric patients are less likely to
do not include pediatric-specific criteria. The have arterial catheters as compared to adult
pediatric definition created by PALICC sought patients, diagnostic criteria and disease severity
to include the unique pathophysiology of stratification were expanded to include saturation
PARDS and include consideration of the by pulse oximetry. Previous definitions of ARDS
developmental factors that may influence lung relied on PaO2 by arterial blood gas to make the
pathology in children. It is important to note diagnosis of ARDS. By expanding this definition,
Chronic Lung Standard Criteria above for age, timing and origin of edema with chest imaging
Disease consistent with new infiltrate and acute deterioration in oxygenation from baseline
which meet oxygenation criteria above.3
Left Ventricular Standard Criteria for age, timing, and origin of edema with chest imaging changes
dysfunction consistent with new infiltrate and acute deterioration in oxygenation which meet
criteria above not explained by left ventricular dysfunction.
Fig. 1.1 2015 PALICC pediatric acute respiratory dis- index or oxygen saturation index should not be used for
tress syndrome (PARDS) definition. 1Use PaO2-based children with chronic lung disease supported with inva-
metric when available. However, if PaO2 is not available, sive mechanical ventilation at baseline or children with
wean FiO2 to maintain SpO2 ≤ 97% to calculate oxygen cyanotic congenital heart disease [3]. (Used with
saturation index or SpO2:FiO2 ratio. 2For non-intubated permission)
patients. 3Stratification of disease severity by oxygen
ERRNVPHGLFRVRUJ
1 Ventilator Management for Pediatric Acute Respiratory Distress Syndrome 5
ERRNVPHGLFRVRUJ
6 T. P. Vesel and I. M. Cheifetz
a irway clearance mechanisms. CPAP helps main- cardiopulmonary deterioration in patients with
tain airway and alveolar patency, thereby more severe disease. As noninvasive ventilation
preventing and/or improving atelectasis, a signif- is trialed, careful and rapid assessment of the
icant cause of shunt physiology and arterial patient’s response to therapy is necessary. Patients
hypoxemia. Additionally, adding inspiratory who will respond to therapy will likely show
pressure with BiPAP helps increase tidal volume improvement in respiratory distress and oxygen-
delivery in lungs with low compliance, improv- ation within the first 30–60 minutes. Clinical
ing alveolar ventilation and reducing PaCO2 [6]. vigilance is required to determine if a patient is
For most patients, noninvasive support devices adequately supported with noninvasive ventila-
are well tolerated, reduce the need for sedation, tion and whether invasive mechanical ventilation
and possibly prevent intubation and mechanical should be pursued.
ventilation, generally in patients with more mild
disease. Currently, there are only a few studies to
support the use of noninvasive respiratory sup- Lung-Protective Strategies
port in children. In one study of 50 children with
acute hypoxemic respiratory failure, predomi- In the modern era of mechanical ventilation,
nantly secondary to bronchiolitis, supported with much attention has been focused on what has
BiPAP or standard treatment (face mask oxygen), been coined “lung-protective strategies” to pre-
the patients supported with BiPAP showed a vent ventilator-induced lung injury (VILI). The
significantly decreased rate of intubation (28%) major focus of these strategies is reduction of
over those receiving standard therapy (60%, mechanical stresses on the alveoli, mainly over-
p = 0.045) [7]. This study showed noninvasive distension (volutrauma), cyclic opening and clos-
ventilation improved hypoxemia, tachycardia, ing of alveoli (atelectrauma), and excessive
and tachypnea as well as prevented some patients plateau pressure (barotrauma). Bedside goal-
from endotracheal intubation and invasive directed strategies, including tidal volume
mechanical ventilation. However, another study 5–8 ml/kg, positive end-expiratory pressures
comparing noninvasive positive-pressure ventila- (PEEP) 10–15 cm H2O, inspiratory plateau pres-
tion to inhaled oxygen post-extubation in chil- sure < 28 cm H2O [9], permissive hypercapnia
dren 28 days to 3 years of age showed no (pH > 7.25 without a specific target PaCO2), and
difference in re-intubation rates (9.1% vs 11.3%, permissive hypoxemia (SpO2 > 88%, PaO2
p > 0.05) [8]. These studies did not include selec- 55–80), are the mainstay of lung-protective ven-
tion criteria or stratification by ARDS criteria and tilator management strategies.
highlight the need for further studies in the ben-
efits and potential adverse events related to the
use of noninvasive respiratory support in the Tidal Volume Delivery: Volutrauma
PARDS population.
In light of the current lack of data in patients Prior to the early 2000s, the general approach to
with PARDS, noninvasive positive-pressure ven- mechanical ventilation targeted tidal volumes of
tilation may be a safe alternative for pediatric 10–15 ml/kg, normal PaCO2, and normal oxygen
patients with mild PARDS and can be considered saturations. It should be noted that the normal
to prevent intubation in some patients. It could be resting tidal volume in humans is generally
debated that noninvasive ventilation should only 6–8 ml/kg. In 2000, a landmark study by the
be considered in patients with less severe disease ARDS Network showed a significant decrease in
and not used in patients with moderate to severe mortality in adult ARDS patients with targeted
lung disease. The clinician must understand tidal volumes of 6 ml/kg (31%) as compared to
potential risks associated with these modalities, “traditional” tidal volumes of 12 ml/kg (39.8%,
including the risk of providing inadequate and p = 0.007) [10]. The results of this large adult
untimely respiratory support with subsequent study provided the basis for a significant shift in
ERRNVPHGLFRVRUJ
1 Ventilator Management for Pediatric Acute Respiratory Distress Syndrome 7
the mechanical ventilation management strate- on the findings of the initial adult studies, which
gies of ARDS patients. In practice, to achieve low have guided the clinical practice of ARDS with
tidal volumes and lower inspiratory pressures, a lower tidal volume goals. The studies in pediat-
deviation from the goals of normal PaCO2 and rics that show lower mortality related to higher
PaO2 (SpO2) was developed and coined permis- tidal volumes have suggested further study is
sive hypercapnia and permissive hypoxemia, likely warranted to assess a causal relationship
respectively. between tidal volume and outcome in those with
Although no pediatric study has confirmed a PARDS.
mortality benefit to low tidal volume ventilation
in PARDS, pediatric critical care clinicians, in
general, have been keen to adopt this strategy for PEEP Titration: Atelectrauma
its potential benefit. However, in contrast to the
outlined adult findings, it must be noted that During normal respiration, the vocal cords close
observational pediatric studies have shown a rela- at the end of expiration to maintain a low level of
tionship between higher tidal volumes and lower positive pressure in the airways and alveoli to
mortality [11] or no relationship between tidal prevent atelectasis. In ARDS, the functional
volume and mortality [12, 13]. Although they did residual capacity of the damaged alveoli
not find a relationship with mortality, Khemani decreases, causing atelectasis unless higher mean
and colleagues showed higher tidal volumes were airway pressure is applied. The use of higher pos-
associated with increased ventilator-free days. It itive end-expiratory pressure (PEEP) may help to
is important to note these pediatric studies were avoid repetitive collapse-opening-collapse injury
performed in the era of “lower than traditional” (atelectrauma).
targeted tidal volumes (i.e., <10 ml/kg); thus, a Determining the optimal PEEP at the bedside
comparison group to the “traditional” ARDS can to be a difficult task, with methods including
Network tidal volume group of >12 ml/kg is not incremental increases (decreases) in PEEP while
available. Considering the limitations of observa- monitoring lung compliance (estimated using
tional studies, it is likely these findings represent tidal volumes, drive pressure, and pressure/vol-
a heterogeneous severity of disease, with higher ume loops) and radiographic findings. During
tidal volumes seen in patients with better lung PEEP adjustment, especially at higher pressures,
compliance (less severe lung injury) with the use cardiopulmonary interactions and hemodynamic
of pressure-control ventilation mode. monitoring must be considered as elevated PEEP
Additionally, in patients with more severe lung (i.e., intrathoracic pressure) may adversely affect
injury, physicians likely targeted lower plateau central venous return and right ventricular after-
pressures to avoid barotrauma, resulting in lower load, therefore decreasing cardiac output.
tidal volumes. It should be noted that atelectrauma has only
Predicted body weight as compared to actual been shown in experimental studies [15]. In the
body weight is recommended when targeting a era of targeted low tidal volume, three adult trials
specific tidal volume as lung capacity is more in ARDS patients evaluating low PEEP vs. higher
closely related to height than weight [14]. PEEP showed no significant difference in mortal-
Targeting predicted body weight may decrease ity [16–18]; however, two systematic reviews and
the risk of over distension and volutrauma in meta-analyses suggested a small survival benefit
obese patients. of higher PEEP in patients with severe ARDS
The current recommendation for tidal volume [19, 20]. Interesting to the pediatric critical care
management for PARDS, as described by provider, a pediatric multicenter, retrospective
PALICC, is to target tidal volumes of 5–8 ml/kg analysis of 1134 patients with PARDS showed
predicted body weight and as low as 3–6 ml/kg that 26% of pediatric patients were managed with
in patients with poor respiratory system compli- lower PEEP than suggested by the ARDSnet pro-
ance [9]. This recommendation is based largely tocol based on FiO2. The investigators found an
ERRNVPHGLFRVRUJ
8 T. P. Vesel and I. M. Cheifetz
increased mortality in that group as compared to p ressure control is more complex than this single
the patients in which PEEP was within the proto- point.
col (OR 2.05, 95% CI 1.32, 3.17) [21]. Pediatric observational studies have shown
PALICC guidelines suggest maintaining both an association between high inspiratory
elevated levels of PEEP (10–15 cm H2O) with pressures and increased mortality [11, 12] and a
consideration of higher titration in severe lack of association between inspiratory pressure
ARDS with attention to limiting the plateau and mortality [13]. None of these studies were
pressure [9]. Considering no pediatric PEEP randomized or powered to determine the relation-
titration protocol has been studied prospec- ship between inspiratory pressure and mortality.
tively, controversy remains as to whether the A recent adult study in ARDS patients showed
ARDSnet adult PEEP/FiO2 titration chart is the drive pressure to be most predictive of mor-
optimal for both adult and pediatric patients tality [22]. Whether there is a relationship
with ARDS. between peak inspiratory, plateau, and/or drive
pressures and mortality in PARDS is yet to be
determined.
lateau Pressure and Drive Pressure
P Based on the available data and clinical exper-
(ΔP): Barotrauma tise, the PALICC recommendation is to maintain
plateau pressures <28 cm H2O, with consider-
Plateau pressure refers to the equilibrated static ation to increased pressure (28–32 cm H2O) in
pressure at the end of inspiration during an inspi- patients with increased chest wall elastance (i.e.,
ratory hold, which is a result of the tidal volume decreased chest wall compliance), such as those
delivered above PEEP without influence of air- with obesity, chest wall edema, or severely
ways resistance (flow). In pressure control mode increased abdominal pressure [9]. This recom-
of mechanical ventilation, peak inspiratory pres- mendation may be considered controversial to
sure (PIP) is controlled by the clinician, and ΔP some clinicians who argue that a higher plateau
(drive pressure) = PIP − PEEP. The drive pres- pressure (30–32 cm H2O) in those without
sure is influenced by: (1) airways resistance, (2) decreased chest wall compliance may be safe.
chest wall elastance, and (3) alveolar compliance, Further studies are needed to delineate a “safe”
whereas the plateau pressure reflects the compli- plateau pressure in those with PARDS with the
ance of the alveoli. The tidal volume is then shared goal to decrease secondary lung injury
dependent on the compliance of the lung, with caused by barotrauma.
worsening lung compliance resulting in lower
tidal volumes at the same inspiratory/plateau
pressure.
Elevated peak airway pressures may cause Clinical Case (Continued)
trauma simply by pressure injury to the lung The patient has been in the PICU for 72 h
parenchyma. Another mechanism suggested for and continues to have worsening hypox-
barotrauma is linked to the heterogeneous nature emia and progressive bilateral infiltrates
of ARDS, with some alveolar units more affected on chest radiograph. His viral panel is
than others, resulting in different compliance of positive for influenza. Despite attempts at
different lung segments. This may lead to low lung-protective ventilator strategies includ-
tidal volumes in poorly compliant lung segments ing increased PEEP, plateau pres-
and overdistension in more compliant (and sure < 28 cm H2O, and tidal volume 5–8 ml/
potentially healthier) lung segments. This con- kg ideal body weight, his oxygen satura-
cept supports the use of pressure control ventila- tions are consistently ~80–85%. He is on
tion modes in patients with PARDS, decreasing the conventional ventilator in pressure con-
the risk of over distension of healthier lung seg- trol mode with FiO2 0.80, PEEP 14 cm
ments, although the debate of volume control vs
ERRNVPHGLFRVRUJ
1 Ventilator Management for Pediatric Acute Respiratory Distress Syndrome 9
ERRNVPHGLFRVRUJ
10 T. P. Vesel and I. M. Cheifetz
Of the 2449 subjects enrolled in the trial, 353 ment maneuvers to be safe [32, 33] and improve
patients (14%) were supported with HFOV. After oxygenation [34] in patients with ARDS. No data
adjusting for risk category, the authors concluded exist on the effect of recruitment maneuvers on
early HFOV was associated with longer duration clinically relevant outcomes, such as mortality,
of mechanical ventilation but no association with morbidity, length of stay, or duration of mechani-
mortality. It is important to note this study was cal ventilation in pediatric patients [35].
not controlled or randomized to these groups, so In practice, there are several variations to per-
minimal definitive conclusions can be gained forming recruitment maneuvers. In the authors’
from this analysis. opinion, manual recruitment maneuvers are not
No conclusive evidence exists that high-fre- recommended as the pressure delivered via the
quency oscillatory ventilation is a superior mode bag can be highly variable and difficult to control
of ventilation as compared to “lung-protective” even with a manometer, risking the negative
conventional ventilation, with a large randomized effects of volutrauma and barotrauma on the
controlled adult trial showing it may be more lungs as well as decreased cardiac output
harmful. Despite this HFOV remains a com- (decreased venous return, increased right ven-
monly used modality in the respiratory manage- tricular afterload). Additionally, derecruitment is
ment of PARDS patients and a source of likely to occur when converting from the manual
controversy and debate. Inconsistent results sup- bag back to the ventilator circuit. Current recom-
porting negative effects, equipoise, and positive mendations support careful recruitment maneu-
benefit to its use leave the pediatric critical care vers to improve severe oxygenation impairment
clinician without guidance as no definitive trial of by using slow incremental and decremental PEEP
HFOV has yet been completed in the PARDS adjustment and recommend not using sustained
population. A randomized, controlled trial of insufflation maneuvers [9].
HFOV in patients with severe PARDS is cur-
rently being initiated. Hopefully, the role of
HFOV for PARDS will be known in the coming Prone Positioning
years. The PALICC recommendations at this
time support “consideration” of HFOV in patients Prone positioning may improve ventilation-per-
with hypoxemic respiratory failure in patients fusion matching due to shunt physiology related
whose plateau pressure exceeds 28 cm H2O in the to atelectasis by promoting blood flow to the
absence of clinical evidence of reduced chest more open anterior segments (i.e., creating zone
wall compliance [9] or 32 cm H2O in the pres- 3 conditions) and by mobilizing secretions. The
ence of reduced chest wall compliance. PROSEVA trial, a large adult randomized con-
trolled trial including 466 adults with severe
ARDS, showed improvement in 28-day (16.0%
Adjunctive Therapies vs 32.8%, p < 0.001) and 90-day mortality
(23.6% vs 41.0%, p < 0.001) with prone position-
Recruitment Maneuvers ing for at least 16 h/day [36]. Pediatric trials
showed improvement of oxygenation while in the
Recruitment maneuvers refer to intermittent prone position [37–39]; however, no change in
increases in airway pressure with the intent of mortality has been seen [40]. The largest pediat-
opening collapsed lung units. ARDS patients ric randomized controlled trial was stopped early
with predominant lung pathology of diffuse alve- due to futility, showing no change in ventilator-
olar collapse (as compared to focal consolida- free days (primary outcome) or secondary end-
tion) and inflammatory edema [30] and those points: time to recovery of lung injury, organ
without impairment of chest wall mechanics [31] failure-free days, cognitive impairment, overall
may benefit most from recruitment maneuvers. functional health at hospital discharge or on day
Pediatric and adult studies have shown recruit- 28, or mortality [41]. Systematic reviews showed
ERRNVPHGLFRVRUJ
1 Ventilator Management for Pediatric Acute Respiratory Distress Syndrome 11
improved oxygenation in patients with acute may be considered in patients with pulmonary
hypoxemic respiratory failure [42] and improved hypertension and right ventricular dysfunction
mortality in severe ARDS (PaO2/FiO2 ratio < 100) or, as a temporizing measure, while extracorpo-
[43], supporting consideration to prone position- real membrane oxygenation is mobilized in the
ing in this specific patient population. severely ill patient.
Considering the only large pediatric random-
ized controlled trial terminated early due to futil-
ity, prone positioning is not routinely Surfactant
recommended for PARDS by PALICC [44].
Surfactant is a mixture of protein and lipid pro-
However, this recommendation is debatable
duced by type II pneumocytes which helps main-
when considering the recent adult data showing
tain alveolar patency by decreasing surface
significant improvement in mortality in adults
tension. Proposed mechanisms for surfactant
with severe ARDS. Prone positioning could be
deficiency in ARDS are direct damage to type II
considered in severe PARDS patients (with P/F
pneumocytes and inactivation of surfactant by
ratio < 100) based on extrapolation from the
protein-rich pulmonary edema fluid during the
available adult-based data. A randomized con-
acute phase of ARDS. With the success of surfac-
trolled trial of prone positioning in severe PARDS
tant in the neonatal respiratory distress syndrome
is currently being initiated and will, hopefully,
population, much excitement has surrounded the
provide greater insight into this management
potential for restoration of the surfactant system
strategy.
to improve outcomes in the PARDS patient. Early
studies and randomized controlled trials showed
acute increases in oxygenation [49–52]. One of
Inhaled Nitric Oxide
three larger pediatric randomized controlled tri-
als showed an improved mortality [53], whereas
Inhaled nitric oxide (iNO) is a potent pulmonary
two others showed no effect on mortality [54,
vasodilator which has been evaluated for use in
55]. Interestingly, one study showed no improve-
patients with ARDS. The mechanism of action is
ment in oxygenation with surfactant administra-
relaxation of smooth muscle by increasing intra-
tion [55]. Current recommendations do not
cellular cyclic guanosine monophosphate. In
suggest the use of surfactant in the management
ARDS, delivery of iNO should theoretically
of PARDS [44].
preferentially vasodilate and increase perfusion
to well-ventilated healthy alveoli, thus possibly
decreasing intrapulmonary shunt physiology.
Pulmonary vasodilation also results in decreased Clinical Case (Continued)
pulmonary vascular resistance (i.e., right ventric- The patient was transitioned to HFOV on
ular afterload) when elevated due to hypoxic pul- PICU admission day 4 with a mild hypo-
monary vasoconstriction. Randomized controlled tension that responded to fluid resuscita-
trials in PARDS patients showed transient tion. He showed a sustained improvement
improvement in oxygenation [45] but no effect in both oxygen saturation and the bilateral
on mortality [46]. In 2011 a meta-analysis evalu- infiltrates over the following days. After
ating the use of iNO in 14 adult and pediatric discussions about adjunctive therapies for
studies showed transient improvements in oxy- PARDS, prone positioning was trialed;
genation but no reduction in mortality. The however, no improvement in oxygenation
authors noted that iNO may be harmful due to an was seen. Five days later, our patient was
increased rate of renal failure [47]. transitioned to conventional ventilation
Considering the data, iNO is not recom- and was successfully extubated several
mended for routine use in the management of days later to 2 lpm via nasal cannula.
children with ARDS [44, 48]. Inhaled nitric oxide
ERRNVPHGLFRVRUJ
12 T. P. Vesel and I. M. Cheifetz
ERRNVPHGLFRVRUJ
1 Ventilator Management for Pediatric Acute Respiratory Distress Syndrome 13
domized, controlled trial of noninvasive ventilation in tory distress syndrome: a randomized controlled trial.
pediatric acute respiratory failure. Pediatr Crit Care JAMA. 2008;299(6):646–55.
Med. 2008;9(5):484–9. 19. Briel M, Meade M, Mercat A, Brower RG, Talmor D,
8. Fioretto JR, Ribeiro CF, Carpi MF, Bonatto RC, Walter SD, et al. Higher vs lower positive end-expi-
Moraes MA, Fioretto EB, et al. Comparison between ratory pressure in patients with acute lung injury and
noninvasive mechanical ventilation and standard acute respiratory distress syndrome: systematic review
oxygen therapy in children up to 3 years old with and meta-analysis. JAMA. 2010;303(9):865–73.
respiratory failure after extubation: a pilot prospec- 20. Phoenix SI, Paravastu S, Columb M, Vincent JL,
tive randomized clinical study. Pediatr Crit Care Med. Nirmalan M. Does a higher positive end expiratory
2015;16(2):124–30. pressure decrease mortality in acute respiratory dis-
9. Rimensberger PC, Cheifetz IM, Pediatric Acute Lung tress syndrome? A systematic review and meta-analy-
Injury Consensus Conference Group. Ventilatory sup- sis. Anesthesiology. 2009;110(5):1098–105.
port in children with pediatric acute respiratory dis- 21. Khemani RG, Parvathaneni K, Yehya N, Bhalla AK,
tress syndrome: proceedings from the Pediatric Acute Thomas NJ, CJL N. PEEP Lower Than the ARDS
Lung Injury Consensus Conference. Pediatr Crit Care Network Protocol is Associated with Higher Pediatric
Med. 2015;16(5 Suppl 1):S51–60. ARDS Mortality. Am J Respir Crit Care Med.
10. Acute Respiratory Distress Syndrome Network,
2018;198:77.
Brower RG, Matthay MA, Morris A, Schoenfeld 22. Amato MB, Meade MO, Slutsky AS, Brochard L,
D, Thompson BT, et al. Ventilation with lower Costa EL, Schoenfeld DA, et al. Driving pressure and
tidal volumes as compared with traditional tidal survival in the acute respiratory distress syndrome. N
volumes for acute lung injury and the acute Engl J Med. 2015;372(8):747–55.
respiratory distress syndrome. N Engl J Med. 23. Ferguson ND, Cook DJ, Guyatt GH, Mehta S, Hand
2000;342(18):1301–8. L, Austin P, et al. High-frequency oscillation in early
11. Erickson S, Schibler A, Numa A, Nuthall G, Yung M, acute respiratory distress syndrome. N Engl J Med.
Pascoe E, et al. Acute lung injury in pediatric inten- 2013;368(9):795–805.
sive care in Australia and New Zealand: a prospective, 24. Bateman ST, Borasino S, Asaro LA, Cheifetz IM,
multicenter, observational study. Pediatr Crit Care Diane S, Wypij D, et al. Early high-frequency oscilla-
Med. 2007;8(4):317–23. tory ventilation in pediatric acute respiratory failure.
12. Khemani RG, Conti D, Alonzo TA, Bart RD 3rd, A propensity score analysis. Am J Respir Crit Care
Newth CJ. Effect of tidal volume in children with Med. 2016;193(5):495–503.
acute hypoxemic respiratory failure. Intensive Care 25. Arnold JH, Truog RD, Thompson JE, Fackler
Med. 2009;35(8):1428–37. JC. High-frequency oscillatory ventilation in pediatric
13. Zhu YF, Xu F, Lu XL, Wang Y, Chen JL, Chao JX, respiratory failure. Crit Care Med. 1993;21(2):272–8.
et al. Mortality and morbidity of acute hypoxemic 26. Rosenberg RB, Broner CW, Peters KJ, Anglin
respiratory failure and acute respiratory distress syn- DL. High-frequency ventilation for acute pediatric
drome in infants and young children. Chin Med J. respiratory failure. Chest. 1993;104(4):1216–21.
2012;125(13):2265–71. 27. Arnold JH, Hanson JH, Toro-Figuero LO, Gutierrez J,
14. Martin DC, Richards GN. Predicted body weight rela- Berens RJ, Anglin DL. Prospective, randomized com-
tionships for protective ventilation – unisex propos- parison of high-frequency oscillatory ventilation and
als from pre-term through to adult. BMC Pulm Med. conventional mechanical ventilation in pediatric respi-
2017;17(1):85. ratory failure. Crit Care Med. 1994;22(10):1530–9.
15.
Muscedere JG, Mullen JB, Gan K, Slutsky 28. Sud S, Sud M, Friedrich JO, Meade MO, Ferguson
AS. Tidal ventilation at low airway pressures can ND, Wunsch H, et al. High frequency oscillation in
augment lung injury. Am J Respir Crit Care Med. patients with acute lung injury and acute respiratory
1994;149(5):1327–34. distress syndrome (ARDS): systematic review and
16. Brower RG, Lanken PN, MacIntyre N, Matthay MA, meta-analysis. BMJ. 2010;340:c2327.
Morris A, Ancukiewicz M, et al. Higher versus lower 29. Young D, Lamb SE, Shah S, MacKenzie I, Tunnicliffe
positive end-expiratory pressures in patients with the W, Lall R, et al. High-frequency oscillation for
acute respiratory distress syndrome. N Engl J Med. acute respiratory distress syndrome. N Engl J Med.
2004;351(4):327–36. 2013;368(9):806–13.
17. Meade MO, Cook DJ, Guyatt GH, Slutsky AS,
30. Pelosi P, D’Onofrio D, Chiumello D, Paolo S, Chiara
Arabi YM, Cooper DJ, et al. Ventilation strategy G, Capelozzi VL, et al. Pulmonary and extrapulmo-
using low tidal volumes, recruitment maneuvers, nary acute respiratory distress syndrome are different.
and high positive end-expiratory pressure for Eur Respir J. 2003;42(Suppl):48s–56s.
acute lung injury and acute respiratory distress 31. Grasso S, Mascia L, Del Turco M, Malacarne P,
syndrome: a randomized controlled trial. JAMA. Giunta F, Brochard L, et al. Effects of recruiting
2008;299(6):637–45. maneuvers in patients with acute respiratory distress
18. Mercat A, Richard JC, Vielle B, Jaber S, Osman D, syndrome ventilated with protective ventilatory strat-
Diehl JL, et al. Positive end-expiratory pressure set- egy. Anesthesiology. 2002;96(4):795–802.
ting in adults with acute lung injury and acute respira-
ERRNVPHGLFRVRUJ
14 T. P. Vesel and I. M. Cheifetz
32. Cruces P, Donoso A, Valenzuela J, Diaz F. Respiratory cific ancillary treatment for pediatric acute respiratory
and hemodynamic effects of a stepwise lung recruit- distress syndrome: proceedings from the Pediatric
ment maneuver in pediatric ARDS: a feasibility study. Acute Lung Injury Consensus Conference. Pediatr
Pediatr Pulmonol. 2013;48(11):1135–43. Crit Care Med. 2015;16(5 Suppl 1):S61–72.
33. Povoa P, Almeida E, Fernandes A, Mealha R, Moreira 45.
Day RW, Allen EM, Witte MK. A random-
P, Sabino H. Evaluation of a recruitment maneuver ized, controlled study of the 1-hour and 24-hour
with positive inspiratory pressure and high PEEP in effects of inhaled nitric oxide therapy in children
patients with severe ARDS. Acta Anaesthesiol Scand. with acute hypoxemic respiratory failure. Chest.
2004;48(3):287–93. 1997;112(5):1324–31.
34. Badet M, Bayle F, Richard JC, Guerin C. Comparison 46. Dobyns EL, Cornfield DN, Anas NG, Fortenberry
of optimal positive end-expiratory pressure and JD, Tasker RC, Lynch A, et al. Multicenter ran-
recruitment maneuvers during lung-protective domized controlled trial of the effects of inhaled
mechanical ventilation in patients with acute lung nitric oxide therapy on gas exchange in children
injury/acute respiratory distress syndrome. Respir with acute hypoxemic respiratory failure. J Pediatr.
Care. 2009;54(7):847–54. 1999;134(4):406–12.
35. Halbertsma FJ, van der Hoeven JG. Lung recruitment 47. Afshari A, Brok J, Moller AM, Wetterslev J. Inhaled
during mechanical positive pressure ventilation in nitric oxide for acute respiratory distress syndrome
the PICU: what can be learned from the literature? and acute lung injury in adults and children: a sys-
Anaesthesia. 2005;60(8):779–90. tematic review with meta-analysis and trial sequential
36. Guerin C, Reignier J, Richard JC, Beuret P, Gacouin analysis. Anesth Analg. 2011;112(6):1411–21.
A, Boulain T, et al. Prone positioning in severe 48. Macrae DJ, Field D, Mercier JC, Moller J, Stiris T,
acute respiratory distress syndrome. N Engl J Med. Biban P, et al. Inhaled nitric oxide therapy in neo-
2013;368(23):2159–68. nates and children: reaching a European consensus.
37. Kornecki A, Frndova H, Coates AL, Shemie SD.
Intensive Care Med. 2004;30(3):372–80.
4A randomized trial of prolonged prone position- 49. Luchetti M, Casiraghi G, Valsecchi R, Galassini E,
ing in children with acute respiratory failure. Chest. Marraro G. Porcine-derived surfactant treatment
2001;119(1):211–8. of severe bronchiolitis. Acta Anaesthesiol Scand.
38. Bruno F, Piva JP, Garcia PC, Einloft P, Fiori R,
1998;42(7):805–10.
Barreto SM. Short-term effects of prone position- 50. Luchetti M, Ferrero F, Gallini C, Natale A, Pigna A,
ing on the oxygenation of pediatric patients sub- Tortorolo L, et al. Multicenter, randomized, controlled
mitted to mechanical ventilation. J Pediatr (Rio J). study of porcine surfactant in severe respiratory syn-
2001;77(5):361–8. cytial virus-induced respiratory failure. Pediatr Crit
39. Lopez-Herce Cid J, Garcia Sanchez E, Garcia
Care Med. 2002;3(3):261–8.
Sanz C, Ruperez Lucas M, Alcaraz Romero A, 51. Willson DF, Zaritsky A, Bauman LA, Dockery K,
Carrillo AA. Effects of prone position, inhaled James RL, Conrad D, et al. Instillation of calf lung
nitric oxide and surfactant in children with hypox- surfactant extract (calfactant) is beneficial in pediatric
emic pulmonary disease. An Pediatr (Barc). acute hypoxemic respiratory failure. Members of the
2003;58(2):106–14. Mid-Atlantic Pediatric Critical Care Network. Crit
40. Casado-Flores J, Martinez de Azagra A, Ruiz-Lopez Care Med. 1999;27(1):188–95.
MJ, Ruiz M, Serrano A. Pediatric ARDS: effect 52. Moller JC, Schaible T, Roll C, Schiffmann JH, Bindl
of supine-prone postural changes on oxygenation. L, Schrod L, et al. Treatment with bovine surfactant
Intensive Care Med. 2002;28(12):1792–6. in severe acute respiratory distress syndrome in chil-
41. Curley MA, Hibberd PL, Fineman LD, Wypij D,
dren: a randomized multicenter study. Intensive Care
Shih MC, Thompson JE, et al. Effect of prone posi- Med. 2003;29(3):437–46.
tioning on clinical outcomes in children with acute 53. Willson DF, Thomas NJ, Markovitz BP, Bauman
lung injury: a randomized controlled trial. JAMA. LA, DiCarlo JV, Pon S, et al. Effect of exog-
2005;294(2):229–37. enous surfactant (calfactant) in pediatric acute
42. Sud S, Sud M, Friedrich JO, Adhikari NK. Effect of lung injury: a randomized controlled trial. JAMA.
mechanical ventilation in the prone position on clini- 2005;293(4):470–6.
cal outcomes in patients with acute hypoxemic respi- 54. Thomas NJ, Guardia CG, Moya FR, Cheifetz IM,
ratory failure: a systematic review and meta-analysis. Markovitz B, Cruces P, et al. A pilot, randomized,
CMAJ. 2008;178(9):1153–61. controlled clinical trial of lucinactant, a peptide-
43. Sud S, Friedrich JO, Taccone P, Polli F, Adhikari NK, containing synthetic surfactant, in infants with acute
Latini R, et al. Prone ventilation reduces mortality hypoxemic respiratory failure. Pediatr Crit Care Med.
in patients with acute respiratory failure and severe 2012;13(6):646–53.
hypoxemia: systematic review and meta-analysis. 55. Willson DF, Thomas NJ, Tamburro R, Truemper E,
Intensive Care Med. 2010;36(4):585–99. Truwit J, Conaway M, et al. Pediatric calfactant in
44. Tamburro RF, Kneyber MC, Pediatric Acute Lung acute respiratory distress syndrome trial. Pediatr Crit
Injury Consensus Conference Group. Pulmonary spe- Care Med. 2013;14(7):657–65.
ERRNVPHGLFRVRUJ
1 Ventilator Management for Pediatric Acute Respiratory Distress Syndrome 15
56. Wong JJ, Jit M, Sultana R, Mok YH, Yeo JG, Koh J, 59. Budilarto SG, Frankowski BJ, Hattler BG, Federspiel
et al. Mortality in pediatric acute respiratory distress WJ. Flow visualization study of a novel respiratory
syndrome: a systematic review and meta-analysis. assist catheter. Artif Organs. 2009;33(6):411–8.
J Intensive Care Med. 2017:885066617705109. 60. Hattler BG, Lund LW, Golob J, Russian H, Lann MF,
57. Orwoll BE, Sapru A. Biomarkers in pediatric ARDS: Merrill TL, et al. A respiratory gas exchange catheter:
future directions. Front Pediatr. 2016;4:55. in vitro and in vivo tests in large animals. J Thorac
58. Conrad SA, Bagley A, Bagley B, Schaap RN. Major Cardiovasc Surg. 2002;124(3):520–30.
findings from the clinical trials of the intravascular
oxygenator. Artif Organs. 1994;18(11):846–63.
ERRNVPHGLFRVRUJ
Extracorporeal Membrane
Oxygenation for Acute Pediatric 2
Respiratory Failure
Matthew Friedman and Michael Hobson
ERRNVPHGLFRVRUJ
18 M. Friedman and M. Hobson
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 19
40–50 mmHg) are acceptable consequences of pARDS was significantly higher in non-survivors
these maneuvers, provided adequate systemic (0.31 vs 0.13), was independently associated
oxygen delivery and hemodynamics are main- with mortality, and functioned better as a predic-
tained [9]. In children with severe lung disease, tor of mortality than the initial PaO2/FiO2 ratio or
these lung-protective strategies may have to be oxygenation index [14]. This predictive value of
exceeded to provide adequate ventilation and dead space fraction however was not observed at
oxygenation, with progressive VILI as an untow- 24 h. Functionally, a single numerical value at
ard consequence. Initiating ECMO provides disease onset may not be practical from the stand-
respiratory support allowing for reduction in ven- point of clinical decision-making, as intensivists
tilator settings to non-toxic levels and possibly to may attempt other modalities and therapies (e.g.,
avoid further VILI. In either scenario, the most high-frequency oscillatory ventilation, prone
important principle when deciding upon ECMO position, inhaled nitric oxide, etc.) before pro-
suitability is to identify those children with a high ceeding with ECMO.
probability of mortality yet having potentially The PaO2/FiO2 (PF) ratio and the oxygenation
reversible lung disease. index (OI = [(mean airway pressure × FiO2 × 100)/
PaO2]) have both served as markers of lung dis-
ease severity in children with hypoxemic respira-
ypoxemic Respiratory Failure
H tory failure, but the OI has become preferred in
and Pediatric Acute Respiratory contemporary pediatric critical care practice, as it
Distress Syndrome incorporates the mean airway pressure (MAwP)
required to maintain oxygenation goals [15].
Pediatric acute respiratory distress syndrome Over the past decade, many retrospective and
(pARDS) is a clinical syndrome characterized by prospective studies have demonstrated an asso-
decreased lung compliance and difficulties with ciation between higher OI and mortality in chil-
oxygenation. Mortality from pARDS ranges dren with hypoxemic respiratory failure [16–19].
from 18% to 35% [10, 11]. Clinical predictors of Historically, an OI of greater than 20 has been
mortality from pARDS could help clinicians used as an indication to transition from conven-
identify children who would benefit from ECMO tional mechanical ventilation to high-frequency
support for refractory hypoxemic respiratory fail- oscillatory ventilation (HFOV) [20, 21]. In these
ure. Candidate predictors include alveolar dead studies, an OI greater than 20 is associated with
space fraction (utilized in the studies below as mortality rates of more than 40%. The trend in OI
[PaCO2 – end tidal CO2]/PaCO2), the PaO2/FiO2, value is likely more informative than any single
and the oxygenation index (OI). Nuckton et al. data point, as pARDS is an evolving disease pro-
prospectively measured dead space fraction in cess. For example, utilizing data from preexisting
adult patients with ARDS early in the course of cohorts with pARDS, the Pediatric Acute Lung
their illness and found increasing dead space Injury Consensus Conference evaluated the fol-
fraction to be an independent risk factor for mor- lowing variables as predictors of mortality: initial
tality [12]. From a pediatric perspective, in a ret- PF ratio, initial OI, worst PF ratio during the first
rospective review of 217 children requiring 3 days of mechanical ventilation, and worst OI
mechanical ventilation for acute hypoxemic values during the first 3 days of mechanical ven-
respiratory failure, the dead space fraction at dis- tilation. The worst (highest) OI value during the
ease onset and day one both correlated with mor- first 3 days of ventilation was the best discrimina-
tality, though not independently associated when tor for non-survival, with an area under the
controlled for severity of illness, 24-h maximal receiver operating characteristic curve of 0.75
inotrope score, and oxygenation index [13]. On [15]. Recent data suggests that incorporating an
the other hand, in a cohort of 266 children with inflammatory cytokine profile alongside the oxy-
pARDS, Yehya et al. recently showed that the genation index is superior in predicting outcomes
alveolar dead space fraction at the onset of in pARDS compared to the oxygenation index
ERRNVPHGLFRVRUJ
20 M. Friedman and M. Hobson
alone [22]. These data are intriguing, as they sug- v entilator-induced lung injury. Second, children
gest the possibility of a future biomarker array with evolving hypoxemic respiratory failure can
that could help identify patients with pARDS at develop hemodynamic instability. Etiologies for
risk for mortality and thus stratify those candi- this instability can be multifactorial, including
dates who should be considered for earlier impaired oxygen delivery to the myocardium,
ECMO support. Until then, it can be concluded impaired right ventricular preload from increased
that for patients with pARDS that progress to intrathoracic pressure, increased right ventricular
ECMO, higher pre-ECMO oxygenation index is afterload from hypercarbia and acidosis, and con-
at higher risk of mortality [23, 24]. current sepsis. Next, the cascade of inflammatory
Currently, the Extracorporeal Life Support cytokines released during ARDS as well as with
Organization recommends an OI sustained above VILI has been demonstrated to result in direct
40 as the indication for initiation of ECMO in organ dysfunction, fluid retention, and subse-
children with respiratory failure [25]. However, quent fluid overload [27]. All of the above factors
there are likely children with hypoxemic respira- may culminate in a progressive metabolic acido-
tory failure and OI values below this threshold sis, the effect of which is noteworthy in critically
that could benefit from ECMO’s potential ability ill children. In a retrospective cohort of children
to mitigate further VILI. For example, in an with hypoxemic respiratory failure supported
analysis of a cohort of children in our institution with HFOV, the presence of a metabolic acidosis
with hypoxemic respiratory failure who were was independently associated with a higher risk
ventilated with HFOV, an OI greater than 25 at of mortality [19]. Likewise, studies examining
48 h following the onset of pARDS conferred a pre-ECMO variables in this patient population
significantly higher odds of mortality (odds ratio: have shown that the presence of acidosis is asso-
5, 95% CI 1.3 to 16.7, p < 0.05) [19]. While the ciated with worse survival [23, 24]. For example,
predictive value of the oxygenation index contin- in a study by Dimico and colleagues containing
ues to grow in the arena of pediatric critical care, data from 1325 children within the ELSO regis-
to date, no definitive OI threshold exists above try, odds of survival to hospital discharged
which pediatric intensivists can be 100% certain increased by 15% for every 0.1 increase in pre-
regarding the optimal time point for the initiation ECMO pH.
of ECMO for pARDS. The Pediatric Acute Lung
Injury Consensus Conference from 2015 states
that “it is not possible to apply strict criteria for Hypercarbic Respiratory Failure
the selection of children who will benefit from
ECMO in pARDS” [26]. Though severe refractory hypoxemia is the most
In the absence of strict numerical criteria sur- common indication for pediatric respiratory
rounding oxygenation, intensivists are left to rely ECMO, there are patients who require ECMO
on the gestalt clinical picture when deciding upon due to severe ventilation impairment leading to
the initiation of ECMO to support children with hypercarbia. Acute hypercarbic respiratory fail-
hypoxemic respiratory failure. Key components ure is defined as a PaCO2 greater than 50 mmHg,
within this framework include trends in ventilator typically associated with respiratory acidosis.
support and oxygenation measures over time, Modern ventilation strategies are focused on lim-
hemodynamic stability, organ function, and acid- iting VILI and allowing for permissive hypercar-
base status. First, with regard to the trajectory of bia [28, 29]. The possible detrimental effects of
respiratory support and oxygenation, serial moderate hypercarbia (50–75 mmHg) have been
assessments serve more valuable than an evalua- debated, but they are inconsequential compared
tion at a single time point [26]. It is our practice to the risks associated with ECMO [30]. Thus,
to obtain arterial blood gases and calculate the ECMO is not advised for moderate hypercarbia
oxygenation index every 4 h. Concurrently, we (i.e., PaCO2 up to 75 mmHg). However, as PaCO2
also track trends in peak inspiratory pressures as rises and respiratory acidosis becomes more
a marker for the potential of evolving severe, dysfunction of other organ systems
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 21
becomes more significant, particularly that of the p ulmonologist and, if necessary, a regional lung
cardiovascular system. Therefore, severe hyper- transplant center.
carbic respiratory failure with acidosis and hemo-
dynamic instability is an indication for initiation
of ECMO [30]. ECMO can correct hypercarbia Airway Disorders
rapidly with a resultant improvement in acidosis
and organ function. During this process, clini- Patients with significant air leaks may not meet
cians must be aware of possible detrimental con- typical oxygenation or ventilation criteria for
sequences of rapid correction of PaCO2, which ECMO. Indications for ECMO support in patients
includes cerebral vasoconstriction and the devel- with air leak syndromes include recurrent pneu-
opment of alkalosis. mothoraces causing life-threatening events or
Status asthmaticus is the most common dis- persistent air leaks not improving with chest tube
ease causing severe hypercarbia requiring thoracostomy. Patients with broncho-pleural fis-
ECMO, but patients with status asthmaticus only tulae from pulmonary infection or surgery that do
represented 3% of the pediatric respiratory cases not heal spontaneously on mechanical ventilation
reported to ELSO from 2009 to 2015 [31, 32]. can also be successfully managed with ECMO,
Patients who require ECMO for status asthmati- which allows for reduction of peak airway pres-
cus typically have severe acidosis (pH < 7.0) and sure [34, 35]. ECMO can also be used intraopera-
severe hypercarbia (PaCO2 > 100 mmHg) despite tively for major airway surgeries and
maximal medical therapies [32]. Children with postoperatively to allow for healing of surgical
status asthmaticus on ECMO have very good sur- sites without the effects of high positive airway
vival, with 88% surviving to hospital discharge. pressures [36, 37].
Moreover, status asthmaticus improves relatively Our recommendations are a general frame-
quickly, with an average duration of ECMO of work to utilize when deciding upon ECMO sup-
92 h [32]. port for children with refractory respiratory
Cystic fibrosis (CF) with pulmonary exacer- failure who have a potentially reversible lung dis-
bation is another patient population that can ease and no absolute contraindication to ECMO
require ECMO for severe hypercarbic respiratory (Table 2.1).
failure. There were 73 ECMO runs in the ELSO
registry for cystic fibrosis from 1998 to 2013,
with 52% survival [33]. There has been debate if
patients with CF are candidates for ECMO due to Case Presentation
the progressive nature of their disease. Each OI was calculated to be 36 without
patient must be evaluated individually. First, the improvement over 12 h despite medical
reversibility of their respiratory failure should be therapy and use of high-frequency oscilla-
assessed. The primary determination in this tory ventilation. Her inspired oxygen con-
regard is if the child has a pulmonary infection centration could not be weaned below
that is potentially amenable to antibiotic therapy. 70%. At this point, her refractory hypox-
If lung disease is not thought to be reversible, emic respiratory failure secondary to influ-
then candidacy for lung transplantation should be enza and Staphylococcus aureus
considered. Cystic fibrosis should only be con- pneumonia was considered refractory to
sidered an absolute contraindication to ECMO if conventional therapy with no trajectory of
the patient is deemed not to have the potential for improvement. There was also concern that
recovery from the acute process and is not a can- her potentially injurious ventilator support
didate for lung transplantation. Candidacy should and high inspired oxygen concentration
be considered for any intubated cystic fibrosis would continue to worsen her underlying
patient to help inform decisions of whether or not lung disease. Based on these data, cannu-
to pursue ECMO. This candidacy should be lation for ECMO was being considered.
determined by working with your local
ERRNVPHGLFRVRUJ
22 M. Friedman and M. Hobson
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 23
ECMO showed an overall survival to hospital years seem to be particularly at risk. MRSA
discharge of 25% [52]. Survival remained con- pneumonia and sepsis may result in
stant across all decades, suggesting that the ful- thrombocytopenia-associated multiple organ fail-
minant nature of this particular infection should ure (TAMOF), and there is some evidence which
cause one to approach initiating ECMO with shows benefit from plasmapheresis performed
caution. during ECMO to reverse the sequelae of this dis-
Patients with active fungal infections prior to order [55]. Combining plasmapheresis to the
the initiation of ECMO are at risk of persistent ECMO circuit adds a layer of technical complex-
seeding and contamination of the ECMO circuit. ity, and there is also increased heparin clearance
Candida was the predominate fungal species during the procedure, thus necessitating careful
acquired both pre- and during ECMO in a study monitoring of anticoagulation. Despite these
of the ELSO registry. While the presence of a challenges, recent data have shown that therapeu-
fungal infection significantly increased the odds tic plasma exchange can improve organ function
of mortality in all age groups, 82–89% of patients in children with sepsis-induced multiorgan dys-
with fungal infections prior to the initiation of function who require ECMO support [56].
ECMO became culture negative at some point
into their ECMO course [53]. Similarly, the pres-
ence of a fungal infection did not lead to an Duration of Mechanical Ventilation
increase in circuit complications or circuit fail-
ure. Thus, the authors of this review concluded As late as the 1990s, children with respiratory
that while a fungal infection remains an important failure were not considered candidates for extra-
comorbidity to consider when initiating ECMO corporeal support after 7–10 days of mechanical
support, it, in and of itself, should not be consid- ventilation due to a significant reduction in sur-
ered an absolute contraindication to vival when ECMO was initiated beyond this
ECLS. However, caution must be exercised for timeframe [57, 58]. In this era, lung-protective
pediatric patients with respiratory failure second- ventilator strategies had not yet become standard
ary to fungal infections. In the aforementioned practice, and, perhaps, significant VILI that had
review by Zabrocki et al., fungal pneumonia was accrued in these children prior to ECMO initia-
the single diagnosis independently associated tion was irreversible. More contemporary data
with the highest risk of mortality [24]. It is likely suggests that the optimal timing for ECMO initi-
that the fungal infection alone does not portend to ation still lies within the first week of mechanical
a grim prognosis but rather is a surrogate marker ventilation [59–61], but outcomes for children
for patients with other diagnoses associated with ventilated beyond 7 days prior to ECMO have
poor outcomes on ECMO, such as patients with improved such that these children should still be
oncological diagnoses. considered appropriate candidates for extracor-
Community-acquired MRSA is an invasive poreal support. Domico et al. reviewed the rela-
bacterium that can cause skin and soft-tissue tionship of pre-ECMO mechanical ventilation
infections in previously healthy children. The duration and the outcomes of 1352 pediatric
invasive nature of this bacterium may lead to patients who required ECMO for respiratory fail-
severe necrotizing pneumonia and ARDS in ure from 1999 to 2008 [23]. In this analysis, a
some children. Concurrent with the rise of inva- significant reduction in survival was not observed
sive MRSA infections in the community over the until the pre-ECMO duration of ventilation
past 25 years, the use of ECMO to support chil- exceeded 14 days. Similarly, in Zabrocki’s review
dren with hypoxemic respiratory failure second- of pediatric respiratory failure, analysis of pre-
ary to MRSA pneumonia has also risen [54]. ECMO mechanical ventilation revealed that sur-
Mortality from this pathogen remains high, with vival remained 56% or higher for children who
an approximate mortality rate of 50% for chil- had ECMO initiated within the first 14 days of
dren who require ECMO; children older than 5 mechanical ventilation but declined to 38%
ERRNVPHGLFRVRUJ
24 M. Friedman and M. Hobson
beyond this timeframe [24]. Fourteen days is now 756 neonatal and pediatric ECMO patients, peak
often considered a “cutoff” for duration of venti- fluid overload, fluid overload at ECMO discon-
lation prior to ECMO; however, the outcomes for tinuation, and the change in fluid overload during
patients with certain diagnoses, such as viral ECMO were significantly higher in patients who
pneumonia, can be good even when ECMO is suffered mortality either while on ECMO or later
initiated after 2 weeks of ventilation [21, 22]. during their hospitalization [63].
Careful patient selection is therefore paramount The status of a patient’s immune system is a
to successful use of when ECMO if extracorpo- significant consideration when deciding upon a
real support is initiated after an extended course child’s suitability for extracorporeal support.
of mechanical ventilation. Factors that should be Historically, an immunocompromised condition
considered are the number of comorbidities, the was a relative contraindication for ECMO for
number of non-pulmonary organ failures, and the several reasons:
primary cause of pARDS.
• Leukopenia from chemotherapy or the disease
process itself confers a risk for subsequent
Patient Comorbidities super-infection.
• Concurrent thrombocytopenia and coagulopa-
Historically, the ideal ECMO candidate was one thy place patients at a higher risk for hemor-
with a known reversible illness, had single-organ rhagic complications.
failure and minimal comorbidities, was neuro- • Multiorgan failure is often present prior to the
logically intact and not developmentally delayed, initiation of ECMO.
and had minimal bleeding risk. Practically speak- • Doubt or uncertainty may exist regarding the
ing, very few children cared for in modern-day patient’s long-term prognosis with respect to
pediatric intensive care units fit this description. their underlying disease.
Over the past two decades, the use of ECMO for
respiratory support in children with comorbidi- A recent review examined the outcomes of
ties has increased markedly. In 1993, 19% of 107 children with underlying malignancies (73
children placed on ECMO had underlying comor- hematologic, 34 solid tumors) who received
bidities, compared to 47% in 2007 [24]. Not sur- ECLS for various disease processes (the majority
prisingly, children with comorbidities have lower of which were acute respiratory failure) over a
survival rates when compared to previously 13-year period [64]. Overall survival to hospital
healthy children [24]. Comorbidities which have discharge was 35% (36% for respiratory failure,
been shown to significantly increase mortality on 29% for cardiac failure), which is worse than for
ECMO include acute kidney injury, liver failure, other children receiving ECLS. Despite their
cancer, primary immunodeficiency, and pre- immunosuppressed state, only 19% of these chil-
ECMO cardiac arrest [24]. dren acquired a new infection while on
Acute kidney injury is a common occurrence ECMO. Children with malignancies did have a
in the intensive care setting, and its impact on higher rate of hemorrhagic complications, includ-
outcomes of children requiring ECMO is notable. ing cannula site bleeding, disseminated intravas-
Recently, the Kidney Intervention During cular coagulation, and CNS hemorrhages when
Extracorporeal Membrane Oxygenation Study compared to children without cancer. Currently,
Group showed that approximately 60% of chil- most pediatric ECMO centers offer ECMO sup-
dren requiring ECMO support have acute kidney port for this patient population. Of 118 centers
injury, and this comorbidity is associated with a surveyed, 78% did not view the presence of a
longer duration of ECMO and an increased risk malignancy as a contraindication, and 17% con-
of mortality [62]. Closely related to acute kidney sidered pediatric cancer as only a relative contra-
injury is fluid overload, which also impacts indication to ECMO [64]. The support for the use
ECMO patient outcomes. In a recent analysis of of ECMO in this patient population, despite only
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 25
a 35% overall survival, likely stems from the rhage, and five had support was withdrawn for
notion that these patients, the majority of whom others reasons [72]. Furthermore, only one of the
have leukemia, have a good prognosis with regard four remaining children alive after ECMO sur-
to their underlying malignancy if their acute vived to hospital discharge. A more recent ELSO
respiratory failure can be overcome. In this registry review of children placed on ECMO sup-
cohort, the median OI prior to initiation of ECMO port following hematopoietic stem cell transplant
was 52, reflecting that these patients likely have a also showed poor outcomes, with only 10% of
higher degree of lung injury prior to ECMO ini- patients surviving to hospital discharge [73].
tiation, relative to other patient populations. It Based on this experience, it can be concluded
remains to be seen if earlier initiation of ECMO that, at the present time, children undergoing
within this patient population could potentially HSCT are unlikely to survive refractory respira-
result in improved outcomes. Regardless, malig- tory failure requiring ECMO. Broadly speaking,
nancy should not be considered an absolute this population of children has had dismal out-
contraindication from ECMO. Close communi- comes when requiring intensive care admission,
cation with the child’s oncology team regarding particularly if mechanical ventilation is neces-
the long-term prognosis from their underlying sary, and thus ECMO should be avoided.
malignancy is an additional essential piece of In summary, while the historical approach
information to be considered in this decision- may have been to reserve ECMO for previously
making process. healthy “salvageable” patients, contemporary
Children undergoing hematopoietic stem cell ECMO utilization has shown that there are in fact
transplantation (HSCT) are another immunosup- very few conditions completely incompatible
pressed patient population where the decision to with ECMO. In the setting of pediatric respira-
implement extracorporeal support may be con- tory failure, extracorporeal support serves as
troversial. These children can develop life- either a bridge to patient recovery or a bridge to
threatening complications in the immediate consideration for lung transplantation in the event
posttransplant period (e.g., diffuse alveolar hem- of non-recovery. Unless a disease process is
orrhage, idiopathic pulmonary syndrome, various deemed irreversible or the child is not a candidate
infections) and in the later stages following trans- for lung transplantation, ECMO remains a realis-
plantation (e.g., graft versus host disease, bron- tic option to support most critically ill children
chiolitis obliterans), all of which can lead to with refractory respiratory failure.
respiratory failure refractory to conventional
care. Historical mortality rates for children
requiring mechanical ventilation following CMO Modality and Cannulation
E
HSCT are well over 50% [65–67]. More recent Strategy
data suggests that an earlier transition from con-
ventional ventilation to HFOV may improve sur- Once the decision has been made to initiate extra-
vival, but mortality remains high [68]. With corporeal support for the child with worsening
regard to the use of ECMO in this patient popula- respiratory failure, the next determination is
tion, there are sparse case reports documenting choice of ECMO modality. Venovenous (V-V)
the successful use of ECMO to support HSCT ECMO involves the removal of deoxygenated
patients through posttransplant complications, blood from the patient’s right atrium, oxygen-
including diffuse alveolar hemorrhage, [69] idio- ation and ventilation as the blood traverses
pathic pulmonary syndrome, [70] and sepsis sec- through the ECMO circuit, and then return of
ondary to neutropenic enterocolitis [71]. oxygenated blood into the child’s central venous
However, in an ELSO registry review of 19 chil- circulation through a different port, often within
dren undergoing ECLS following HSCT, 79% the same cannula. V-V ECMO provides gas
died during the ECMO course: seven developed exchange but no direct cardiac support. In con-
multiorgan failure, three had refractory hemor- trast, with veno-arterial (V-A) ECMO, oxygen-
ERRNVPHGLFRVRUJ
26 M. Friedman and M. Hobson
ated blood returning to the patient enters the vasopressor, and 41% required at least one ino-
arterial circulation, commonly through a differ- tropic infusion at the time of ECMO cannula-
ent cannula in the carotid or femoral artery. tion. After initiation of ECMO support, there
Completely bypassing the patient’s native heart was a significant reduction in vasoactive medica-
and lungs, V-A ECMO can provide complete car- tion requirements, and all patients were free
diopulmonary support [74]. from vasoactive infusions by day 6 of
The inherent risks between these two ECMO ECMO. Similarly, an analysis of 4332 ECMO
modalities should be considered. With V-A runs for children with sepsis showed signifi-
ECMO, cannulation of the femoral artery incurs cantly better survival for children receiving V-V
the risk of lower extremity ischemia [74], while support when compared to those on V-A ECMO
cannulation of the carotid artery carries a sub- [78]. This study was unable to account for pre-
stantial risk of stroke (23%) [75]. In an analysis ECMO severity of illness and also did not factor
of pediatric patients with respiratory failure, in the type of hemodynamic derangement that
the use of V-A ECMO was independently asso- characterized each patient’s type of sepsis (i.e.,
ciated with an increased risk of neurologic warm shock vs. cold shock). However, the fact
injury when compared to V-V ECMO [76]. that such a large number of children with sepsis
Other risks associated with V-A ECMO include were successfully supported with V-V ECMO
embolism of air or thrombi into the patient’s gives credence to the ability of this modality to
arterial circulation or increased systemic after- support critically ill children who have unstable
load leading to distension of the left ventricle, hemodynamics.
with subsequent risk of pulmonary hemorrhage. The use of V-V ECMO in pediatric critical
The primary disadvantage of V-V relative to care is continuing to increase [79], and based on
V-A ECMO is its inability to provide cardiac data from a large retrospective review, V-V
support for patients with hemodynamic insta- ECMO appears to confer a survival advantage
bility. However, there are important cardiac relative to V-A ECMO [24]. To date, there is no
benefits that indirectly result from initiation of definitive formula or inotrope score by which to
V-V ECMO for respiratory failure. First, with guide intensivists in choice of ECMO modality
the ability to wean mechanical ventilation, for children with acute respiratory failure and
intrathoracic pressure decreases, thereby associated hemodynamic instability. Our own
improving preload to the right ventricle. institutional practice relies greatly on echocar-
Second, the ability of V-V ECMO to correct diogram imaging of myocardial performance to
hypercarbia and acidosis can lead to a reduction aid in this decision. A child with mild to moder-
in pulmonary vascular resistance and a corre- ately depressed right ventricular function, which
sponding decrease in right ventricular after- is most often due to a combination of high pul-
load. Lastly, oxygenated blood returning from monary vascular resistance induced by lung dis-
the ECMO circuit passes through the pulmo- ease and respiratory acidosis, is a candidate for
nary vasculature, eventually making its way V-V ECMO, even if requiring a moderate amount
into the left ventricle. A portion of this blood of vasoactive medications for hemodynamic sup-
will enter the coronary circulation as it exits the port. V-V ECMO is recommended for these
aortic valve, providing a previously oxygen- patients, even in the setting of hemodynamic
deprived myocardium with a rich source of instability, as correction of respiratory acidosis
oxygen and a resultant improvement in ventric- and an increase in right ventricular preload that
ular function and hemodynamics. occur after ECMO initiation often improve ven-
In a single-center review of children requiring tricular function. In contrast, a child with an
ECMO for acute respiratory failure from 1991 echocardiogram showing significantly depressed
through 2002, Pettignano et al. illustrated the left ventricular or biventricular function due to
hemodynamic benefits of V-V ECMO [77]. In sepsis-induced myocardial depression should be
this cohort, 35% of patients required at least one cannulated for V-A ECMO.
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 27
ECMO cannula selection and configuration is ECMO flow interruption between the wire-rein-
an essential decision made in the process of initi- forced and non-wire cannulas within the first
ating V-V ECMO support. One option for V-V 72 h [81]. The two wire- reinforced VVDL
cannulation is a multisite configuration with ECMO catheters currently available for pediat-
single-
lumen ECMO catheters, the first being ric use are the OriGen® DL cannula (OriGen®
placed within the right internal jugular vein and Biomedical, Austin, TX) and Avalon® Elite
extending into the right atrium and the second Bicaval cannula (Avalon® Laboratories, LLC,
inserted into a femoral vein and extending Rancho Dominguez, CA). The OriGen® is
upward into the inferior vena cava. The alterna- placed in the right atrium and has a proximal
tive cannulation option utilizes a dual-lumen drainage hole and distal reinfusion hole. The
venovenous (VVDL) catheter inserted into the configuration of the bicaval Avalon® cannula
right internal jugular vein, with both drainage places the drainage holes within the SVC and
and reinfusion lumens of the cannula residing IVC, while the reinfusion port is located within
within the right atrium. One disadvantage of the the right atrium and is directed toward the tri-
two-cannula technique includes the need to cuspid valve, offering the theoretical advantage
access multiple venous sites, thus being a more of reducing the amount of recirculation. The use
invasive and time-consuming procedure, along of bicaval ECMO cannulas has been shown to
with increasing the sites for potential bleeding be effective and safe in the adult ECMO popula-
and infectious complications. Second, children tion [82, 83], but the smaller-sized venous and
weighing less than 15 kg typically do not have cardiac structures inherent to pediatric patients
large enough femoral veins to accommodate the may raise concern regarding an increased rate of
necessary-sized ECMO catheters. Given these mechanical complications and increased need
constraints of the two-site cannulation technique, for catheter repositioning. Several pediatric
the application of the VVDL cannulation strategy ECMO centers have published their institutional
to provide ECMO support to children with respi- experiences with the bicaval wire- reinforced
ratory failure has increased significantly over the cannulas and have noted minimal complications
past decade [24, 80]. Zamaro et al. analyzed the in both pediatric and neonatal patients [81, 84–
performance and complication rates of these two 86]. In an analysis of cannula complications by
cannulation strategies from 1323 pediatric V-V Zamaro and colleagues, there was no difference
ECMO runs [80]. Compared to multisite cannu- in rate of complications when wire-reinforced
lation, VVDL cannulation achieved greater and non-wire-reinforced catheters were com-
weight-adjusted ECMO flow but had a slightly pared [80]. Proper imaging techniques during
higher rate of mechanical (26.2% vs. 22.5%, p cannulation, including the combined use of
0.004) and cardiac (24.4% vs. 21.7%, p 0.03) echocardiography and fluoroscopy, have been
complications. Importantly, there was no signifi- shown to reduce complications and the need for
cant difference in survival between the two can- catheter repositioning [86]. Lastly, successful
nulation techniques. and safe percutaneous ECMO cannulation of
One drawback to early versions of VVDL pediatric patients performed by intensivists has
ECMO catheters was their potential to bend and recently been described [87]. A single-center
kink, thus raising the possibility of obstruction retrospective review of percutaneous ECMO
to blood flow and interruption of the ECMO cir- cannulations performed by intensivists included
cuit. The newer generation of VVDL catheters 18 pediatric patients cannulated for V-V
is manufactured with wire reinforcement to off- ECMO. In this cohort, the overall rate of suc-
set this potential complication. The only avail- cessful cannulation was 98% [87]. In our cur-
able pediatric literature comparing these two rent practice, we consider wire-reinforced
types of catheter designs was a single-center VVDL ECMO catheters safe for pediatric
retrospective study of 25 neonates and infants, patients and are the preferred modality by which
which found no difference in the incidence to provide extracorporeal respiratory support.
ERRNVPHGLFRVRUJ
28 M. Friedman and M. Hobson
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 29
ommendations regarding PEEP offer a wide vs 4–6 cmH2O) leads to shorter ECMO runs and
acceptable range – 5–15 cmH2O [25]. The use of fewer complications [93, 94]. Identifying the
PEEP at the higher end of this range in neonates ideal PEEP for a patient is a multifactorial deci-
with respiratory failure has been shown to lead to sion that includes evaluation of lung expansion
shorter duration of ECMO [93, 94]. Similarly, in on chest x-ray, hemodynamics, oxygenation,
adults on V-V ECMO, higher PEEP during the lung compliance, and other factors. In a recent
first 3 days on ECMO has been associated with study, the use of electrical impedance tomogra-
improved survival [95]. Conversely, higher PEEP phy for adult patients with respiratory failure on
over the whole ECMO course has been shown to ECMO showed that an optimal PEEP of
be associated with increased mortality [96]. It is 15 cmH2O best balanced overdistension and lung
hard to reconcile these seemingly incongruous collapse [100].
findings. However, one could hypothesize that The one consistent finding from the limited
PEEP later in the ECMO course is more related data published about mechanical ventilation for
to patient factors than ventilator strategy. In other adults on ECMO is that higher driving pressure
words, patients with improving lung disease will (peak inspiratory pressure minus PEEP) or pla-
have PEEP decreased over the course on ECMO, teau pressure is associated with increased mortal-
while patients who do not improve and ultimately ity [89, 101, 102]. Limiting peak or plateau
die will remain on higher PEEP throughout their pressure is advisable for adults on ECMO as it is
course. Additionally, high PEEP in the first for any adult with ARDS. However, the upper
3 days on ECMO is more likely to be a conscien- limit of what is reasonable for driving and pla-
tious strategy of maintaining lung recruitment teau pressures in children has not been well
and less so reflective of the patient’s severity of established in pARDS and might be higher than
lung disease. in adults [17, 18, 103].
The risks of maintaining a higher MAwP must
be weighed against the risks of allowing lung
collapse. High MAwP will cause elevated intra- Secretion Clearance on ECMO
thoracic pressure which may impair hemody-
namics due to impaired venous return. Elevated Invasive procedures on ECMO are avoided when-
static pressures, like those used in an open-lung ever possible due to the increased risk of hemor-
ventilation strategy or in APRV, do not to contrib- rhage associated with anticoagulation.
ute significantly to VILI [97]. There are also risks Bronchoscopy, however, is an invasive procedure
to a lung-rest approach resulting in complete that may be necessary for some patients sup-
lung collapse. Hemodynamically, lung collapse ported on ECMO. Though there are only a few
will lead to increased right ventricular afterload reports of bronchoscopy on ECMO in the litera-
by way of increased pulmonary vascular resis- ture [98–101], it is commonly implemented in
tance. On V-V ECMO, lung collapse will also clinical practice. Routine bronchoscopy of
lead to severe pulmonary venous desaturation, patients on ECMO is performed in 55% of pedi-
and systemic arterial saturations will suffer. atric ECMO centers and 76% of adult centers
Additionally, there is emerging evidence of the [90]. While the benefits of bronchoscopy on
damaging effects of recurrent atelectasis, often ECMO have not been established in the litera-
called atelectrauma [8, 98, 99]. If there is com- ture, they are often seen at the bedside in patients
plete lung collapse, the lungs will need to be re- with thick respiratory secretions that cannot be
recruited later in the ECMO course, which often mobilized through traditional suctioning tech-
can only be accomplished with high driving pres- niques. In the limited published data on pediatric
sure, which is the biggest contributor to VILI bronchoscopies on ECMO to date, minor bleed-
[97]. Lastly, the time necessary to re-recruit the ing episodes occur in up to 35% of patients in
lungs may prolong the ECMO course. Neonatal some report, the risk of significant pulmonary
data shows that higher PEEP (i.e.,12–14 cmH2O hemorrhage is 0–1.5% [104–107]. Bronchoscopy
ERRNVPHGLFRVRUJ
30 M. Friedman and M. Hobson
is therefore feasible and relatively safe in patients mechanical complications, obviating any possi-
on ECMO, with a small risk of pulmonary bility of extubation. It should also be noted that
hemorrhage. extubated patients might require reintubation for
HFPV is a combination of high-frequency procedures including bronchoscopy and decan-
ventilation and conventional mechanical ventila- nulation from ECMO.
tion principles. HPFV consists of high-frequency
sub-physiological tidal volumes superimposed
on low-frequency conventional tidal volume ven- Recommendations
tilation [108]. One institution published their
experience with the use of high-frequency per- Maintenance of lung aeration with modest venti-
cussive ventilation (HFPV) and increased fre- lator settings may be advantageous to their recov-
quency of bronchoscopy to target secretion ery. Lung aeration will facilitate higher arterial
clearance for pediatric respiratory ECMO. They oxygen saturations (SaO2), allow for some oxy-
showed an increase in ECLS-free days when genation if there is an emergency requiring sepa-
compared to historical controls [106]. It is not ration from the circuit, and possible shorten
clear if the benefit was due to HFPV, increased ECMO course by eliminating the time needed to
bronchoscopies, the combination of the two, or re-recruit collapsed lung. Our usual strategy is to
other changes to practice that were not measured. maintain lung aeration with high PEEP (10–
The use of HFPV and frequent bronchoscopy 14 mmHg) or APRV when possible. Due to our
promotes secretion clearance and is most likely concerns that strict adherence to the historical
to be beneficial in diseases that cause excessive “lung-rest” settings with low driving pressure
secretions such as bronchiolitis. (e.g.. 10 cmH2O) and complete lung collapse
may not be optimal and may prolong the ECMO
course, we often use slight higher MAwP (15–
Extubation on ECMO 20 mmHg) and driving pressure (12–18 mmHg)
than historical “lung rest.” On the other hand,
The practice of extubation on ECMO has becom- when there is complete lung collapse despite
ing increasingly utilized in the past several years. these moderate MAwP, we do not advise trying to
In a 2015 survey, only 10% of centers reported re-recruit alveoli in the acute inflammatory phase
extubating patients on ECMO, while a 2017 sur- of pARDS.
vey reported 41% of centers extubate patients Some patients will have complete lung col-
[90, 91]. ECMO is employed to allow for oxy- lapse in the face of high MAwP. In patients with
genation and ventilation without the risk of lung collapse and who are likely to have pro-
VILI. It has been suggested, therefore, that the longed ECMO courses, extubation is an option.
best way to limit VILI would be to remove the Patients that are most likely to be able to be extu-
“V” or ventilator. Extubating patients on V-V bated and benefit from extubation are school-
ECMO has been shown to be feasible and safe in aged or older children in whom a short ECMO
selected patients [90, 109]. In one retrospective run (<7 days) is not expected. Further, to success-
study, extubation on ECMO was associated with fully extubate patients on ECMO, the circuit
improved survival [109]. If extubation is feasible, must be able support oxygenation without the aid
there is the benefit of reducing sedation once the of mechanical ventilation, which requires high
noxious stimulus that is the endotracheal tube is flow rates without significant recirculation issues.
removed. Patients who are extubated and not We support aggressive airway clearance with
sedated will also be able to participate in more bronchoscopy for pediatric respiratory ECMO,
aggressive rehabilitation while on ECMO. Not particularly in patients with disease processes
all patients will be candidates for extubation on that lead to mucous plugging. Bronchoscopy is
ECMO. Younger patients, for instance, may likely most beneficial when performed early,
require paralysis or heavy sedation to prevent such as within the first week on ECMO, with the
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 31
goal of lung recruitment when the inflammatory the team must have an understanding of the
stage of pARDS is resolving. The need for fur- special needs of a patient on ECMO. Centers
ther bronchoscopy can be determined by the that perform a high volume of ECMO develop
secretion burden noted during the first bronchos- an institutional knowledge that can only
copy and the subsequent clinical course. develop over many years and many ECMO
Diagnostic bronchoscopy can also be performed runs.
to assess for infectious complications. Routine There have been multiple studies done to
reduction or interruption anticoagulation for evaluate if center volumes affect outcomes.
bronchoscopy is not suggested due to low risk of Children treated at sites with 20–49 ECMO runs
significant bleeding. per year and greater than 50 ECMO runs per
year (pediatric and neonatal combined) have
been shown to have lower odds of mortality
Case Presentation compared to centers with less than 20 runs per
After cannulation for venovenous ECMO, year [110]. The single cut-point that produced
the 2-year-old child with influenza and the most significant difference in mortality was
Staphylococcus pneumonia was managed 22 cases per year [110]. In another study, there
with a high PEEP strategy was a strong relationship between center volume
(i.e.,10–12 mmHg) and modest driving and mortality for pediatric cardiac patients [111].
pressure (i.e., pressure above PEEP Lastly, when survival was assessed based on
12–16 mmHg). Daily chest x-rays demon- age-specific center ECMO volumes, neonatal
strated modest lung aeration. During the and adult volumes were significantly associated
initial portion of her ECMO course, she with survival, and there was a nonsignificant
was receiving only relatively small tidal trend toward improved survival in higher-vol-
volumes (~2–3 mL/kg) with this approach. ume centers for pediatric patients [112]. The
Over the next 2 weeks, with the assistance effect of center volume is particularly difficult to
of an aggressive pulmonary toilet that answer concerning pediatric respiratory
included multiple bronchoscopies, her tidal ECMO. A medium-to-large sized pediatric
volumes gradually increased to 6–7 mL/kg, ECMO program may only have five respiratory
and she tolerated weaning of ECMO sup- cases per year, and few centers perform more
port. She was successfully decannulated than ten pediatric respiratory ECMO runs annu-
after 20 days of ECMO support with no ally. A recent international position paper on
detectable complications. ECMO for acute respiratory failure in adults
argues for centralization of ECMO and a mini-
mum of 20 ECMO cases per year, with 12 of
Centralization of ECMO them being for respiratory support [113]. This
and the Effect of Center Volume recommendation is based on the pediatric data
discussed above and expert opinion.
The relationship between ECMO volume and In the CESAR trial, adult patients with severe
the quality of ECMO care delivered at individ- ARDS were randomized to staying at commu-
ual centers are active topics of discussion and nity hospitals or being transferred to a center
study. There are theoretical benefits of central- capable of ECMO [5]. Patients who were trans-
ization of this highly complex care. ECMO is ferred had improved survival (63% vs. 47%),
relatively rare, even in the largest centers, and even though only 75% of transferred patients
centralization of care allows for a concentration were cannulated for ECMO [5]. Therefore, early
of expertise. ECMO requires a team approach transfer from centers that do not provide ECMO
between ECMO technicians, bedside nurses, can be recommended for patients with severe
respiratory therapists, surgeons, intensivists, ARDS with high OI. Volume alone however
and other medical personnel. All members of does not ensure quality care; education, training,
ERRNVPHGLFRVRUJ
32 M. Friedman and M. Hobson
60
50
40
Percent Survival
30
20
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 33
ECMO [116]. From 1993 to 2007, there were no known progressive lung diseases such as
survivors of the nine patients supported on CF. While consensus guidelines state that ECLS
ECMO for 52 days or more in the ELSO database is a relative contraindication to lung transplanta-
[116]. More recently, there have been many cases tion, 1.5% of lung transplant recipients are
reported of long-term respiratory ECMO in both bridged to transplant on ECMO [122, 123].
children and adults with good outcomes, includ- Patients bridged to lung transplant on ECMO
ing lung transplant or late lung recovery after up may have worse short-term survival, but adult
to 265 days of ECMO [116–121]. These reports lung transplant recipients who are spontane-
are remarkable as, historically, it has been thought ously breathing on ECMO have similar 3-year
that if native lung function was not improving or survival to patients on no support and better sur-
restored at 2–4 weeks of ECMO support, the vival than patients on mechanical ventilation
chance of recovery was remote [119]. As evi- alone or on ECMO with mechanical ventilation
denced by these case reports of successful pro- [123, 124].
longed ECMO, there may be previously
underappreciated regenerative capacity of the
lungs. The paradigm for lung recovery may be Considerations on Prolonged ECMO
more similar to recovery of kidney function after
acute kidney injury, where the chances of recov- When a patient is supported on prolonged
ery of native function decrease over time, but ECMO, other important considerations are reha-
there is still potential for return to normal func- bilitation and tracheostomy. Tracheostomy on
tion in some patients even weeks to months after ECMO is not common practice in pediatrics,
injury. with only 13% of pediatric centers reporting per-
The decision of when lung recovery becomes forming tracheostomies in these patients [90].
remote is difficult. There is a paucity of data to The placement of tracheostomy while on ECMO
help physicians prognosticate which patients on is more frequently performed in adults [90]. To
prolonged ECMO may recover. Despite the lack our knowledge, the largest report of tracheos-
of many survivors after 8 weeks on ECMO, tomy on ECMO at a pediatric institution con-
reports of successful outcomes beyond this time tains nine patients ages 7–25 years [125]. Two of
frame suggest there should be no absolute cutoff nine survived without lung transplantation, and
for length of ECMO duration in pediatric respi- four survived with lung transplantation. All
ratory failure. It is therefore imperative to care- patients had decreased sedation needs within
fully assess patients on prolonged ECMO 72 h. The biggest concern with performing tra-
support in regard to proximate cause of respira- cheostomy on ECMO is bleeding due to antico-
tory failure, premorbid lung function, potential agulation. Practices for anticoagulation vary
for lung recovery, presence of other organ dys- from holding of anticoagulation around the time
functions, and the goals of the patient or of procedure to varying reductions in heparin
decision-makers. dosing [90, 125]. In two adult case series of 168
ECMO patients in total, minor bleeding after tra-
cheostomy was common (approximately 30%),
ECMO as a Bridge to Lung Transplant while major bleeding occurred in only 2% and
8% of cases. No deaths were attributed to com-
Patients who are on prolonged ECMO without plications of tracheostomy [126, 127]. If trache-
recovery of native function may be candidates ostomy is to be considered, the timing of
for lung transplant. The evaluation process for tracheostomy remains a difficult decision. The
lung transplantation should be considered for average time on ECMO to tracheostomy in the
patients on ECMO for greater than 3 weeks pediatric study of nine patients was 10 days
without a trajectory toward improvement or [125]. Older pediatric patients with longer than
early in the ECMO course for patients with average expected ECMO durations or those on
ERRNVPHGLFRVRUJ
34 M. Friedman and M. Hobson
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 35
ERRNVPHGLFRVRUJ
36 M. Friedman and M. Hobson
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 37
26. Dalton HJ, Macrae DJ, Pediatric Acute Group. Pathobiology of acute respiratory distress
Lung Injury Consensus Conference Group. syndrome. Pediatr Crit Care Med. 2015;16(5 Suppl
Extracorporeal support in children with pedi- 1):S6–22.
atric acute respiratory distress syndrome: pro- 39. Askegard-Giesmann JR, Besner GE, Fabia R,
ceedings from the Pediatric Acute Lung Injury Caniano DA, Preston T, Kenney BD. Extracorporeal
Consensus Conference. Pediatr Crit Care Med. membrane oxygenation as a lifesaving modality in the
2015;16(5 Suppl 1):S111–7. treatment of pediatric patients with burns and respi-
27. Husain-Syed F, Slutsky AS, Ronco C. Lung-kidney ratory failure. J Pediatr Surg. 2010;45(6):1330–5.
cross-talk in the critically ill patient. Am J Respir 40. Skarda D, Henricksen JW, Rollins M. Extracorporeal
Crit Care Med. 2016;194(4):402–14. membrane oxygenation promotes survival in chil-
28. Santschi M, Randolph AG, Rimensberger PC, dren with trauma related respiratory failure. Pediatr
Jouvet P, Pediatric Acute Lung Injury Mechanical Surg Int. 2012;28(7):711–4.
Ventilation I, Pediatric Acute Lung I, et al. 41. Fortenberry JD, Meier AH, Pettignano R, Heard M,
Mechanical ventilation strategies in children with Chambliss CR, Wulkan M. Extracorporeal life sup-
acute lung injury: a survey on stated practice pattern. port for posttraumatic acute respiratory distress syn-
Pediatr Crit Care Med. 2013;14(7):e332–7. drome at a children’s medical center. J Pediatr Surg.
29. Acute Respiratory Distress Syndrome Network, 2003;38(8):1221–6.
Brower RG, Matthay MA, Morris A, Schoenfeld D, 42. Scalzo AJ, Weber TR, Jaeger RW, Connors RH,
Thompson BT, et al. Ventilation with lower tidal vol- Thompson MW. Extracorporeal membrane oxygen-
umes as compared with traditional tidal volumes for ation for hydrocarbon aspiration. Am J Dis Child.
acute lung injury and the acute respiratory distress 1990;144(8):867–71.
syndrome. N Engl J Med. 2000;342(18):1301–8. 43. Kimura D, Shah S, Briceno-Medina M, Sathanandam
30. Barnes T, Zochios V, Parhar K. Re-examining per- S, Haberman B, Zhang J, et al. Management of mas-
missive hypercapnia in ARDS: a narrative review. sive diffuse alveolar hemorrhage in a child with
Chest. 2017;154(1):185–95. systemic lupus erythematosus. J Intensive Care.
31. Barbaro RP, Paden ML, Guner YS, Raman L, Ryerson 2015;3:10.
LM, Alexander P, et al. Pediatric Extracorporeal Life 44. Zulian F, Martinez Toledo MM, Amigoni A, Martini
Support Organization Registry International Report G, Agosto C, Pettenazzo A. Successful use of extra-
2016. ASAIO J. 2017;63(4):456–63. corporeal membrane oxygenation for severe inter-
32. Hebbar KB, Petrillo-Albarano T, Coto-Puckett W, stitial lung disease in a child with dermatomyositis.
Heard M, Rycus PT, Fortenberry JD. Experience Intensive Care Med. 2007;33(9):1663–6.
with use of extracorporeal life support for severe 45. Lopez-Fernandez Y, Azagra AM, de la Oliva P,
refractory status asthmaticus in children. Crit Care. Modesto V, Sanchez JI, Parrilla J, et al. Pediatric
2009;13(2):R29. acute lung injury epidemiology and natural history
33. Hayes D Jr, Kopp BT, Preston TJ, Kirkby S, Tobias study: incidence and outcome of the acute respira-
JD, Papadimos TJ, et al. Survival of patients with cys- tory distress syndrome in children. Crit Care Med.
tic fibrosis on ECMO: analysis of the Extracorporeal 2012;40(12):3238–45.
Life Support Organization Registry. Int J Clin Exp 46. Halasa NB, Barr FE, Johnson JE, Edwards KM. Fatal
Med. 2014;7(5):1370–2. pulmonary hypertension associated with pertussis in
34. Daoud O, Augustin P, Mordant P, Lasocki S, infants: does extracorporeal membrane oxygenation
Al-Attar N, Maury JM, et al. Extracorporeal mem- have a role? Pediatrics. 2003;112(6 Pt 1):1274–8.
brane oxygenation in 5 patients with bronchial 47. Sawal M, Cohen M, Irazuzta JE, Kumar R, Kirton
fistula with severe acute lung injury. Ann Thorac C, Brundler MA, et al. Fulminant pertussis: a multi-
Surg. 2011;92(1):327–30. center study with new insights into the clinico-
35. Dolgner A, Bain J, Peterson-Carmichael SL, pathological mechanisms. Pediatr Pulmonol.
Turner DA, Rehder KJ. Extracorporeal mem- 2009;44(10):970–80.
brane oxygenation for refractory air leak in a child 48. Surridge J, Segedin ER, Grant CC. Pertussis requiring
presenting with bacterial tracheitis. Respir Care. intensive care. Arch Dis Child. 2007;92(11):970–5.
2014;59(10):e163–5. 49. Rowlands HE, Goldman AP, Harrington K, Karimova
36. Raake J, Johnson B, Seger B, Manning PB, A, Brierley J, Cross N, et al. Impact of rapid leukode-
Eghtesady P, Boesch P, et al. Extracorporeal mem- pletion on the outcome of severe clinical pertussis in
brane oxygenation, extubation, and lung-recruitment young infants. Pediatrics. 2010;126(4):e816–27.
maneuvers as rescue therapy in a patient with tra- 50. Stewart DL, Cook LN, Rabalais GP. Successful use
cheal dehiscence following slide tracheoplasty. of extracorporeal membrane oxygenation in a new-
Respir Care. 2011;56(8):1198–202. born with herpes simplex virus pneumonia. Pediatr
37. Hoetzenecker K, Klepetko W, Keshavjee S, Cypel Infect Dis J. 1993;12(2):161–2.
M. Extracorporeal support in airway surgery. J 51. Bonacchi M, Di Lascio G, Harmelin G, Pasquini
Thorac Dis. 2017;9(7):2108–17. A, Peris A, Sani G. Extracorporeal membrane oxy-
38. Sapru A, Flori H, Quasney MW, Dahmer MK, genation for refractory, life-threatening, and herpes
Pediatric Acute Lung Injury Consensus Conference simplex virus 1-induced acute respiratory distress
ERRNVPHGLFRVRUJ
38 M. Friedman and M. Hobson
syndrome. Our experience and literature review. Am oxygenation study group. Pediatr Crit Care Med.
J Emerg Med. 2012;30(6):1014.e3–e10. 2016;17(12):1157–69.
52. Prodhan P, Wilkes R, Ross A, Garcia X, Bhutta AT, 63. Selewski DT, Askenazi DJ, Bridges BC, Cooper DS,
Rycus P, et al. Neonatal herpes virus infection and Fleming GM, Paden ML, et al. The impact of fluid
extracorporeal life support. Pediatr Crit Care Med. overload on outcomes in children treated with extra-
2010;11(5):599–602. corporeal membrane oxygenation: a multicenter
53. Pluim T, Halasa N, Phillips SE, Fleming G. The retrospective cohort study. Pediatr Crit Care Med.
morbidity and mortality of patients with fungal 2017;18(12):1126–35.
infections before and during extracorporeal mem- 64. Gow KW, Heiss KF, Wulkan ML, Katzenstein HM,
brane oxygenation support. Pediatr Crit Care Med. Rosenberg ES, Heard ML, et al. Extracorporeal
2012;13(5):e288–93. life support for support of children with malig-
54. Creech CB, Johnson BG, Bartilson RE, Yang E, Barr nancy and respiratory or cardiac failure: the extra-
FE. Increasing use of extracorporeal life support in corporeal life support experience. Crit Care Med.
methicillin-resistant Staphylococcus aureus sepsis in 2009;37(4):1308–16.
children. Pediatr Crit Care Med. 2007;8(3):231–5; 65. Jacobe SJ, Hassan A, Veys P, Mok Q. Outcome of
quiz 47. children requiring admission to an intensive care
55. Bridges BC, Hardison D, Pietsch J. A case series unit after bone marrow transplantation. Crit Care
of the successful use of ECMO, continuous renal Med. 2003;31(5):1299–305.
replacement therapy, and plasma exchange for 66. Lamas A, Otheo E, Ros P, Vazquez JL, Maldonado
thrombocytopenia-associated multiple organ failure. MS, Munoz A, et al. Prognosis of child recipi-
J Pediatr Surg. 2013;48(5):1114–7. ents of hematopoietic stem cell transplantation
56. Kawai Y, Cornell TT, Cooley EG, Beckman CN, requiring intensive care. Intensive Care Med.
Baldridge PK, Mottes TA, et al. Therapeutic plasma 2003;29(1):91–6.
exchange may improve hemodynamics and organ 67. Keenan HT, Bratton SL, Martin LD, Crawford
failure among children with sepsis-induced mul- SW, Weiss NS. Outcome of children who require
tiple organ dysfunction syndrome receiving extra- mechanical ventilatory support after bone marrow
corporeal life support. Pediatr Crit Care Med. transplantation. Crit Care Med. 2000;28(3):830–5.
2015;16(4):366–74. 68. Rowan CM, Loomis A, McArthur J, Smith LS, Gertz
57. Moler FW, Palmisano J, Custer JR. Extracorporeal SJ, Fitzgerald JC, et al. High-frequency oscillatory
life support for pediatric respiratory failure: predic- ventilation use and severe pediatric ARDS in the
tors of survival from 220 patients. Crit Care Med. pediatric hematopoietic cell transplant recipient.
1993;21(10):1604–11. Respir Care. 2017;63:404.
58. Pranikoff T, Hirschl RB, Steimle CN, Anderson HL 69. Morris SH, Haight AE, Kamat P, Fortenberry
3rd, Bartlett RH. Mortality is directly related to the JD. Successful use of extracorporeal life support
duration of mechanical ventilation before the initia- in a hematopoietic stem cell transplant patient with
tion of extracorporeal life support for severe respira- diffuse alveolar hemorrhage. Pediatr Crit Care Med.
tory failure. Crit Care Med. 1997;25(1):28–32. 2010;11(1):e4–7.
59. Australia, New Zealand Extracorporeal Membrane 70. Liao WI, Tsai SH, Chiu SK. Successful use of extra-
Oxygenation Influenza Influenza Investigators, corporeal membrane oxygenation in a hematopoietic
Davies A, Jones D, Bailey M, Beca J, et al. stem cell transplant patient with idiopathic pneumo-
Extracorporeal membrane oxygenation for 2009 nia syndrome. Respir Care. 2013;58(2):e6–10.
influenza A(H1N1) acute respiratory distress syn- 71. Wolfson RK, Kahana MD, Nachman JB, Lantos
drome. JAMA. 2009;302(17):1888–95. J. Extracorporeal membrane oxygenation after
60. Nance ML, Nadkarni VM, Hedrick HL, Cullen stem cell transplant: clinical decision-making in
JA, Wiebe DJ. Effect of preextracorporeal mem- the absence of evidence. Pediatr Crit Care Med.
brane oxygenation ventilation days and age 2005;6(2):200–3.
on extracorporeal membrane oxygenation sur- 72. Gow KW, Wulkan ML, Heiss KF, Haight AE, Heard
vival in critically ill children. J Pediatr Surg. ML, Rycus P, et al. Extracorporeal membrane oxy-
2009;44(8):1606–10. genation for support of children after hematopoietic
61. Noah MA, Peek GJ, Finney SJ, Griffiths MJ, stem cell transplantation: the Extracorporeal Life
Harrison DA, Grieve R, et al. Referral to an extra- Support Organization experience. J Pediatr Surg.
corporeal membrane oxygenation center and mor- 2006;41(4):662–7.
tality among patients with severe 2009 influenza 73. Di Nardo M, Locatelli F, Palmer K, Amodeo A,
A(H1N1). JAMA. 2011;306(15):1659–68. Lorusso R, Belliato M, et al. Extracorporeal mem-
62. Fleming GM, Sahay R, Zappitelli M, King E, brane oxygenation in pediatric recipients of hemato-
Askenazi DJ, Bridges BC, et al. The incidence poietic stem cell transplantation: an updated analysis
of acute kidney injury and its effect on neonatal of the Extracorporeal Life Support Organization
and pediatric extracorporeal membrane oxygen- experience. Intensive Care Med. 2014;40(5):754–6.
ation outcomes: a multicenter report from the kid- 74. Brogan TVL, Lequier L, Lorusso R, MacLaren
ney intervention during extracorporeal membrane G, Peek G. Extracorporeal life support: the ELSO
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 39
red book. 5th ed. Ann Arbor: Extracorporeal Life b icaval dual-lumen catheter in neonates. J Pediatr
Support Organization; 2017. Surg. 2012;47(2):430–4.
75. Teele SA, Salvin JW, Barrett CS, Rycus PT, Fynn- 86. Jarboe MD, Gadepalli SK, Church JT, Arnold MA,
Thompson F, Laussen PC, et al. The association of Hirschl RB, Mychaliska GB. Avalon catheters in pedi-
carotid artery cannulation and neurologic injury in atric patients requiring ECMO: placement and migra-
pediatric patients supported with venoarterial extra- tion problems. J Pediatr Surg. 2018;53(1):159–62.
corporeal membrane oxygenation. Pediatr Crit Care 87. Conrad SA, Grier LR, Scott LK, Green R, Jordan
Med. 2014;15(4):355–61. M. Percutaneous cannulation for extracorporeal
76. Rollins MD, Hubbard A, Zabrocki L, Barnhart DC, membrane oxygenation by intensivists: a retrospec-
Bratton SL. Extracorporeal membrane oxygenation tive single-institution case series. Crit Care Med.
cannulation trends for pediatric respiratory failure 2015;43(5):1010–5.
and central nervous system injury. J Pediatr Surg. 88. Marhong JD, Telesnicki T, Munshi L, Del Sorbo
2012;47(1):68–75. L, Detsky M, Fan E. Mechanical ventilation dur-
77. Pettignano R, Fortenberry JD, Heard ML, Labuz ing extracorporeal membrane oxygenation.
MD, Kesser KC, Tanner AJ, et al. Primary use of the An international survey. Ann Am Thorac Soc.
venovenous approach for extracorporeal membrane 2014;11(6):956–61.
oxygenation in pediatric acute respiratory failure. 89. Marhong JD, Munshi L, Detsky M, Telesnicki T, Fan
Pediatr Crit Care Med. 2003;4(3):291–8. E. Mechanical ventilation during extracorporeal life
78. Skinner SC, Iocono JA, Ballard HO, Turner MD, support (ECLS): a systematic review. Intensive Care
Ward AN, Davenport DL, et al. Improved sur- Med. 2015;41(6):994–1003.
vival in venovenous vs venoarterial extracorporeal 90. Jenks CL, Tweed J, Gigli KH, Venkataraman R,
membrane oxygenation for pediatric noncardiac Raman L. An international survey on ventilator prac-
sepsis patients: a study of the Extracorporeal Life tices among extracorporeal membrane oxygenation
Support Organization registry. J Pediatr Surg. centers. ASAIO J. 2017;63(6):787–92.
2012;47(1):63–7. 91. Camporota L, Nicoletti E, Malafronte M, De Neef
79. Rehder KJ, Turner DA, Cheifetz IM. Extracorporeal M, Mongelli V, Calderazzo MA, et al. International
membrane oxygenation for neonatal and pediatric survey on the management of mechanical ventilation
respiratory failure: an evidence-based review of the during ECMO in adults with severe respiratory fail-
past decade (2002–2012). Pediatr Crit Care Med. ure. Minerva Anestesiol. 2015;81(11):1170–83, 77 p
2013;14(9):851–61. following 83.
80. Zamora IJ, Shekerdemian L, Fallon SC, Olutoye 92. Roy S, Habashi N, Sadowitz B, Andrews P, Ge L,
OO, Cass DL, Rycus PL, et al. Outcomes compar- Wang G, et al. Early airway pressure release ventila-
ing dual-lumen to multisite venovenous ECMO tion prevents ARDS-a novel preventive approach to
in the pediatric population: the Extracorporeal lung injury. Shock. 2013;39(1):28–38.
Life Support Registry experience. J Pediatr Surg. 93. Keszler M, Ryckman FC, McDonald JV Jr, Sweet
2014;49(10):1452–7. LD, Moront MG, Boegli MJ, et al. A prospec-
81. Subramanian S, Vafaeezadeh M, Parrish AR, tive, multicenter, randomized study of high ver-
McMullan DM. Comparison of wire-reinforced and sus low positive end-expiratory pressure during
non-wire-reinforced dual-lumen catheters for veno- extracorporeal membrane oxygenation. J Pediatr.
venous ECMO in neonates and infants. ASAIO J. 1992;120(1):107–13.
2013;59(1):81–5. 94. Keszler M, Subramanian KN, Smith YA, Dhanireddy
82. Javidfar J, Brodie D, Wang D, Ibrahimiye AN, R, Mehta N, Molina B, et al. Pulmonary manage-
Yang J, Zwischenberger JB, et al. Use of bicaval ment during extracorporeal membrane oxygenation.
dual-lumen catheter for adult venovenous extracor- Crit Care Med. 1989;17(6):495–500.
poreal membrane oxygenation. Ann Thorac Surg. 95. Schmidt M, Stewart C, Bailey M, Nieszkowska A,
2011;91(6):1763–8; discussion 9. Kelly J, Murphy L, et al. Mechanical ventilation
83. Teman NR, Haft JW, Napolitano LM. Optimal management during extracorporeal membrane oxy-
endovascular methods for placement of bica- genation for acute respiratory distress syndrome: a
val dual-lumen cannulae for venovenous extra- retrospective international multicenter study. Crit
corporeal membrane oxygenation. ASAIO J. Care Med. 2015;43(3):654–64.
2013;59(4):442–7. 96. Modrykamien AM, Hernandez OO, Im Y, Walters
84. Fallon SC, Shekerdemian LS, Olutoye OO, Cass RW, Schrader CL, Smith LE, et al. Mechanical
DL, Zamora IJ, Nguyen T, et al. Initial experience ventilation in patients with the acute respira-
with single-vessel cannulation for venovenous tory distress syndrome and treated with extra-
extracorporeal membrane oxygenation in pedi- corporeal membrane oxygenation: impact on
atric respiratory failure. Pediatr Crit Care Med. hospital and 30 day postdischarge survival. ASAIO
2013;14(4):366–73. J. 2016;62(5):607–12.
85. Lazar DA, Cass DL, Olutoye OO, Kim ES, Welty 97. Protti A, Andreis DT, Monti M, Santini A,
SE, Fernandes CJ, et al. Venovenous cannulation Sparacino CC, Langer T, et al. Lung stress and
for extracorporeal membrane oxygenation using a strain during mechanical ventilation: any d ifference
ERRNVPHGLFRVRUJ
40 M. Friedman and M. Hobson
between statics and dynamics? Crit Care Med. in patients undergoing extracorporeal life support.
2013;41(4):1046–55. Int J Artif Organs. 2017;40(12):696–700.
98. Chu EK, Whitehead T, Slutsky AS. Effects of cyclic 110. Freeman CL, Bennett TD, Casper TC, Larsen GY,
opening and closing at low- and high-volume ven- Hubbard A, Wilkes J, et al. Pediatric and neona-
tilation on bronchoalveolar lavage cytokines. Crit tal extracorporeal membrane oxygenation: does
Care Med. 2004;32(1):168–74. center volume impact mortality? Crit Care Med.
99. Tsuchida S, Engelberts D, Peltekova V, Hopkins N, 2014;42(3):512–9.
Frndova H, Babyn P, et al. Atelectasis causes alveo- 111. Karamlou T, Vafaeezadeh M, Parrish AM, Cohen
lar injury in nonatelectatic lung regions. Am J Respir GA, Welke KF, Permut L, et al. Increased extracor-
Crit Care Med. 2006;174(3):279–89. poreal membrane oxygenation center case volume is
100. Franchineau G, Brechot N, Lebreton G, Hekimian associated with improved extracorporeal membrane
G, Nieszkowska A, Trouillet JL, et al. Bedside con- oxygenation survival among pediatric patients. J
tribution of electrical impedance tomography to Thorac Cardiovasc Surg. 2013;145(2):470–5.
setting positive end-expiratory pressure for extracor- 112. Barbaro RP, Odetola FO, Kidwell KM, Paden
poreal membrane oxygenation-treated patients with ML, Bartlett RH, Davis MM, et al. Association
severe acute respiratory distress syndrome. Am J of hospital-level volume of extracorporeal
Respir Crit Care Med. 2017;196(4):447–57. membrane oxygenation cases and mortality.
101. Chiu LC, Hu HC, Hung CY, Chang CH, Tsai FC, Analysis of the extracorporeal life support orga-
Yang CT, et al. Dynamic driving pressure associ- nization registry. Am J Respir Crit Care Med.
ated mortality in acute respiratory distress syndrome 2015;191(8):894–901.
with extracorporeal membrane oxygenation. Ann 113. Combes A, Brodie D, Bartlett R, Brochard L, Brower
Intensive Care. 2017;7(1):12. R, Conrad S, et al. Position paper for the organiza-
102. Serpa Neto A, Schmidt M, Azevedo LC, Bein T, tion of extracorporeal membrane oxygenation pro-
Brochard L, Beutel G, et al. Associations between grams for acute respiratory failure in adult patients.
ventilator settings during extracorporeal membrane Am J Respir Crit Care Med. 2014;190(5):488–96.
oxygenation for refractory hypoxemia and outcome 114. Chan SY, Figueroa M, Spentzas T, Powell A,
in patients with acute respiratory distress syndrome: Holloway R, Shah S. Prospective assessment of
a pooled individual patient data analysis : mechani- novice learners in a simulation-based extracorporeal
cal ventilation during ECMO. Intensive Care Med. membrane oxygenation (ECMO) education pro-
2016;42(11):1672–84. gram. Pediatr Cardiol. 2013;34(3):543–52.
103. de Jager P, Burgerhof JG, van Heerde M, Albers 115. Zakhary BM, Kam LM, Kaufman BS, Felner
MJ, Markhorst DG, Kneyber MC. Tidal vol- KJ. The utility of high-fidelity simulation for train-
ume and mortality in mechanically ventilated ing critical care fellows in the management of
children: a systematic review and meta-anal- extracorporeal membrane oxygenation emergen-
ysis of observational studies. Crit Care Med. cies: a randomized controlled trial. Crit Care Med.
2014;42(12):2461–72. 2017;45(8):1367–73.
104. Kamat PP, Popler J, Davis J, Leong T, Piland SC, 116. Brogan TV, Zabrocki L, Thiagarajan RR,
Simon D, et al. Use of flexible bronchoscopy in pedi- Rycus PT, Bratton SL. Prolonged extracorpo-
atric patients receiving extracorporeal membrane real membrane oxygenation for children with
oxygenation (ECMO) support. Pediatr Pulmonol. respiratory failure. Pediatr Crit Care Med.
2011;46(11):1108–13. 2012;13(4):e249–54.
105. Prentice E, Mastropietro CW. Flexible bronchos- 117. Wiktor AJ, Haft JW, Bartlett RH, Park PK,
copy for children on extracorporeal membrane Raghavendran K, Napolitano LM. Prolonged VV
oxygenation for cardiac failure. Pediatr Crit Care ECMO (265 days) for ARDS without technical com-
Med. 2011;12(4):422–5. plications. ASAIO J. 2015;61(2):205–6.
106. Yehya N, Dominick CL, Connelly JT, Davis DH, 118. Moon SM, Lee H, Moon JH, Kim HK, Park JE,
Minneci PC, Deans KJ, et al. High-frequency Byeon S, et al. Prolonged maintenance of VV ECMO
percussive ventilation and bronchoscopy during for 104 days with native lung recovery in acute respi-
extracorporeal life support in children. ASAIO J. ratory failure. ASAIO J. 2016;62(2):e15–7.
2014;60(4):424–8. 119. Rosenberg AA, Haft JW, Bartlett R, Iwashyna TJ,
107. Karlson KH Jr, Pickert CB, Schexnayder SM, Huang SK, Lynch WR, et al. Prolonged duration
Heulitt MJ. Flexible fiberoptic bronchoscopy in ECMO for ARDS: futility, native lung recovery, or
children on extracorporeal membrane oxygenation. transplantation? ASAIO J. 2013;59(6):642–50.
Pediatr Pulmonol. 1993;16(4):215–8. 120. Iacono A, Groves S, Garcia J, Griffith B. Lung trans-
108. Lucangelo U, Fontanesi L, Antonaglia V, Pellis T, plantation following 107 days of extracorporeal
Berlot G, Liguori G, et al. High frequency percus- membrane oxygenation. Eur J Cardiothorac Surg.
sive ventilation (HFPV). Principles and technique. 2010;37(4):969–71.
Minerva Anestesiol. 2003;69(11):841–8, 8–51. 121. Gupta P, McDonald R, Chipman CW, Stroud M,
109. Bataillard A, Hebrard A, Gaide-Chevronnay L, Gossett JM, Imamura M, et al. 20-year experience
Martin C, Durand M, Albaladejo P, et al. Extubation of prolonged extracorporeal membrane oxygenation
ERRNVPHGLFRVRUJ
2 Extracorporeal Membrane Oxygenation for Acute Pediatric Respiratory Failure 41
in critically ill children with cardiac or pulmonary 129. Abrams D, Javidfar J, Farrand E, Mongero LB,
failure. Ann Thorac Surg. 2012;93(5):1584–90. Agerstrand CL, Ryan P, et al. Early mobilization of
122. Weill D, Benden C, Corris PA, Dark JH, Davis patients receiving extracorporeal membrane oxy-
RD, Keshavjee S, et al. A consensus document for genation: a retrospective cohort study. Crit Care.
the selection of lung transplant candidates: 2014 – 2014;18(1):R38.
an update from the Pulmonary Transplantation 130. Turner DA, Rehder KJ, Bonadonna D, Gray A, Lin
Council of the International Society for Heart and S, Zaas D, et al. Ambulatory ECMO as a bridge to
Lung Transplantation. J Heart Lung Transplant. lung transplant in a previously well pediatric patient
2015;34(1):1–15. with ARDS. Pediatrics. 2014;134(2):e583–5.
123. Schechter MA, Ganapathi AM, Englum BR, 131. Bain JC, Turner DA, Rehder KJ, Eisenstein EL,
Speicher PJ, Daneshmand MA, Davis RD, et al. Davis RD, Cheifetz IM, et al. Economic outcomes
Spontaneously breathing extracorporeal mem- of extracorporeal membrane oxygenation with and
brane oxygenation support provides the optimal without ambulation as a bridge to lung transplanta-
bridge to lung transplantation. Transplantation. tion. Respir Care. 2016;61(1):1–7.
2016;100(12):2699–704. 132. Marhong JD, DeBacker J, Viau-Lapointe J, Munshi
124. Toyoda Y, Bhama JK, Shigemura N, Zaldonis L, Del Sorbo L, Burry L, et al. Sedation and mobili-
D, Pilewski J, Crespo M, et al. Efficacy of extra- zation during venovenous extracorporeal membrane
corporeal membrane oxygenation as a bridge to oxygenation for acute respiratory failure: an interna-
lung transplantation. J Thorac Cardiovasc Surg. tional survey. Crit Care Med. 2017;45(11):1893–9.
2013;145(4):1065–70; discussion 70–1. 133. Ramanathan K, Cove ME, Caleb MG, Teoh KL,
125. Schwartz SP, Bonadonna D, Hartwig MG, Cheifetz Maclaren G. Ethical dilemmas of adult ECMO:
IM. Bedside tracheostomy on pediatric ICU subjects emerging conceptual challenges. J Cardiothorac
supported by extracorporeal membrane oxygen- Vasc Anesth. 2015;29(1):229–33.
ation. Respir Care. 2017;62:1447. 134. Peetz AB, Sadovnikoff N, O’Connor MF. Is
126. Kruit N, Valchanov K, Blaudszun G, Fowles JA, informed consent for extracorporeal life support
Vuylsteke A. Bleeding complications associ- even possible? AMA J Ethics. 2015;17(3):236–42.
ated with percutaneous tracheostomy insertion in 135. Bailly DK, Reeder RW, Zabrocki LA, Hubbard
patients supported with venovenous extracorpo- AM, Wilkes J, Bratton SL, et al. Development and
real membrane oxygen support: a 10-year insti- validation of a score to predict mortality in children
tutional experience. J Cardiothorac Vasc Anesth. undergoing extracorporeal membrane oxygenation
2017;32:1162. for respiratory failure: pediatric pulmonary rescue
127. Braune S, Kienast S, Hadem J, Wiesner O, with extracorporeal membrane oxygenation predic-
Wichmann D, Nierhaus A, et al. Safety of percu- tion score. Crit Care Med. 2017;45(1):e58–66.
taneous dilatational tracheostomy in patients on 136. Barbaro RP, Boonstra PS, Paden ML, Roberts
extracorporeal lung support. Intensive Care Med. LA, Annich GM, Bartlett RH, et al. Development
2013;39(10):1792–9. and validation of the pediatric risk estimate score
128. Zorowitz RD. ICU-acquired weakness: a rehabilita- for children using extracorporeal respiratory sup-
tion perspective of diagnosis, treatment, and func- port (Ped-RESCUERS). Intensive Care Med.
tional management. Chest. 2016;150(4):966–71. 2016;42(5):879–88.
ERRNVPHGLFRVRUJ
Weaning and Extubation
Readiness Assessment in Pediatric 3
Patients
Samer Abu-Sultaneh
and Christopher W. Mastropietro
ERRNVPHGLFRVRUJ
44 S. Abu-Sultaneh and C. W. Mastropietro
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 45
ERRNVPHGLFRVRUJ
46 S. Abu-Sultaneh and C. W. Mastropietro
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 47
ERRNVPHGLFRVRUJ
48 S. Abu-Sultaneh and C. W. Mastropietro
effective alternative [40]. Until more data are Table 3.1 Entry criteria for extubation readiness trial
available, practitioners must weigh the balance of Presence of spontaneous breathing
the potential benefits of these therapies against Adequate cough and gag reflex
the risks and expense, which includes the No planned procedures requiring general anesthesia
resources, personnel, and time necessary to within the next 24 h
Cardiopulmonary stability with no increase of
administer these therapies. vasoactive drips for at least 12 h
No increase of ventilator support for at least 12 h
Adequate lung compliance and gas exchange
xtubation Readiness Assessment
E
in Pediatric ICU
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 49
Fig. 3.2 Pressure rate Pressure.Rate Product before and after Extubation
product as a function All Patients, All endotracheal tube sizes, n=409
of peri-extubation Median 25%-75% Non-Outlier Range
respiratory support. 1000
(1) Pressure support of
10 over continuous
positive airway pressure 800
(CPAP) of 5 cmH2O. (2)
200
0
Pressure Support CPAP 5 minutes 60 minutes
Condition
ERRNVPHGLFRVRUJ
50 S. Abu-Sultaneh and C. W. Mastropietro
Cardiovascular assessment
No increase work of breathing or diaphoresis
No poor perfusion
No increase in heart rate > 40 beats per minute from baseline
No hypotension defined as value below minimal systolic blood pressure:
Sedation assessment
No apnea or bradypnea
Exhaled tidal volume ≥ 5 ml/kg of ideal body weight
glycopyrrolate can be used to decrease the ranges between 7 and 14% [14, 46]. Further, in
amount of secretions in this patient population regard to extubation failure, no clear benefit
[53, 54]. If a patient is deemed to have failed an could be gleaned from one modality as compared
ERT due to cardiopulmonary, sedation, or airway to the others. HFNC has increasingly been used
protection concerns, the patient is placed on pre- to support pediatric patients with respiratory dis-
ERT ventilator settings and screened again the tress and failure with a resultant decreased rate of
next day for extubation readiness [14, 21, 45, 46]. intubation [56, 57]. However, the prophylactic
ERT passing criteria are listed in Table 3.2. use of HFNC after extubation to prevent extuba-
tion failure has yet to be studied in general pedi-
atric ICU population. In a recent single-center
Anticipation of Post-Extubation prospective randomized trial of 108 mechanical
Respiratory Support ventilated children, NIPPV was not superior to
traditional oxygen therapy via NC in preventing
Theoretically, most patients should be able to be reintubation [58]. NIPPV is likely useful in pre-
extubated to room air or oxygen via traditional venting reintubation when applied electively for
nasal cannula (NC) if they passed an ERT. In patients at increased risk of extubation failure,
reality, respiratory support provided to patients including children with cerebral palsy, severe
upon extubation varies markedly across centers kyphoscoliosis, neuromuscular disorders, brain
[14, 15, 46]. In general, pediatric ICU population injury, and cardiac dysfunction [59–62].
post-extubation support using HFNC ranges The use of NIPPV provides an alternative to
between 21 and 36% [46, 55], while NIPPV use invasive ventilation with the advantage of
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 51
ERRNVPHGLFRVRUJ
52 S. Abu-Sultaneh and C. W. Mastropietro
On the other hand, patients with NIF values hospital length of stay, hospital cost, and mortal-
consistently greater than negative 15 cmH2O ity [6–8, 15]. The rate of extubation failure ranges
are less likely to be successfully extubated. between 2% and 20% depending on the study
Neurally adjusted ventilatory assist (NAVA) population, time period of the study, and the defi-
and maximum airway pressure during occlusion nition used for extubation failure [10]. The rate of
(aPiMax) are emerging as potentially useful extubation failure also varies across institutions
means of assessing respiratory muscle strength. [12, 15, 73, 74]. Extubation failure causes are not
NAVA catheters are used to assess electrical dia- well defined but commonly divided into the fol-
phragmatic activity using a multiple-array esoph- lowing categories: upper airway obstruction, pul-
ageal electrode catheter placed in the esophagus monary insufficiency, respiratory muscle
as a feeding tube. Pediatric patients who passed weakness, cardiac dysfunction, neurological
their ERT had a lower VT to delta electrical dia- impairment, and oversedation. In many cases, the
phragmatic activity, indicating better diaphragm etiology can be multifactorial. Upper airway
strength [26]. The accuracy of this method obstruction is frequently reported as the most
depends on the accurate positioning of the NAVA common cause of extubation failure in general
esophageal catheter. In a more recent study, maxi- pediatric critical care population, accounting for
mum airway pressure during occlusion (aPiMax) one- to two-thirds of extubation failure [14, 18,
was used to assess respiratory muscle strength 74]. Risk factors for extubation failure are young
prior to extubation using calibrated esophageal age (less than 48 months), longer duration of
manometry and respiratory inductance plethys- mechanical ventilation prior to the first extuba-
mography. Patients with aPiMax ≤30 cm H2O tion attempt, underlying airway disorders, genetic
were found to have higher rate of extubation fail- conditions, neurological disease, chronic respira-
ure when compared to patients with less muscle tory disease, and patients receiving chronic
weakness, i.e., aPiMax >30 cm H2O [70]. The NIPPV [7, 21, 70, 75].
utility of using these measures outside clinical
research depends on availability of the technology
and personnel needed to obtain such readings. trategies to Prevent Post-Extubation
S
Volumetric capnography has been used to cal- Upper Airway Obstruction
culate physiologic dead space to tidal volume
ratio (VD/VT) as a possible predictor of extuba- Because upper airway obstruction is the most
tion success or failure. In a prospective pediatric common cause of extubation failure in general
study of 45 mixed medical and surgical patients, pediatric critical care populations, pediatric criti-
VD/VT ≤ 0.50 was found to predict extubation cal care team should use an airway insertion and
success, while VD/VT > 0.65 was associated with maintenance bundle to decrease the risk. The
extubation failure [71]. In pediatric cardiac ICU, insertion bundle should focus on decreasing
dead space correlated with duration of mechani- injury to the airway during endotracheal tube
cal ventilation and predicted extubation success placement by choosing an age-appropriate endo-
in two ventricle, but not single-ventricle patients tracheal tube size according to the Pediatric
after cardiac surgery [72]. Volumetric capnogra- Advanced Life Support (PALS) guidelines and
phy requires specialized exhaled CO2 monitors limit the number of intubation attempts [76–78].
for its calculations, which are not yet used rou- Recommended sizes of endotracheal tubes by
tinely in many contemporary pediatric ICUs. age are provided in Table 3.3 [76, 77]. If notable
resistance to endotracheal tube advancement
occurs during placement, an endotracheal tube
Extubation Failure that is 0.5 mm smaller in diameter should be
used. Smaller endotracheal tubes should also be
Extubation failure is associated with increased used for patients with suspected airway narrow-
duration of mechanical ventilation, ventilator- ing such as patients with croup, thermal or chem-
associated pneumonia rates, intensive care and ical injuries, or previous history of subglottic
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 53
Table 3.3 Airway insertion and maintenance bundle m2) rather than a 3.5 uncuffed endotracheal tube
Airway insertion (17.5 mm/m2).
Use endotracheal tube (ETT) size according to Cuffed endotracheal tubes have many advan-
pediatric advanced life support (PALS) guidelines*: tages. They minimize air leak around the endo-
Uncuffed ETT size = 4 +
age tracheal tube, especially in patients requiring
4 higher peak and mean airway pressure due to
age poor respiratory system compliance. Cuff tubes
Cuffed ETT size = 3.5 +
4 can also provide a seal that should reduce aspi-
Infants up to 1 year of age: ration risk and improve the accuracy of VT and
3.5 uncuffed or 3.0 cuffed ETT EtCO2 monitoring. Further, with the of ability to
Children 1–2 years of age: change cuff volume using a simple small vol-
4.0 uncuffed or 3.5 cuffed ETT ume syringe, cuff volume and pressure can be
*Cuffed ETT is preferred
adjusted if subglottic diameter is affected by
Airway maintenance
Monitor cuff pressure and keep inflation pressure less
fluid balance or subglottic edema and inflamma-
than 20 cmH2O tion, thereby avoiding further airway injury.
Specifically, in an airway maintenance bundle,
endotracheal tube cuff pressure should be moni-
stenosis or tracheal surgical interventions. tored regularly to ensure that the inflation pres-
Patients with trisomy 21 have smaller subglottic sure is less than 20 cmH2O, which may help
diameter and are considered to be increased risk decrease the risk of post-extubation stridor [80].
for post-extubation stridor; practitioners should As an alternative, leak pressure can be moni-
therefore use an endotracheal tube 0.5 to 1 mm tored regularly such that there is only enough
smaller in diameter than what would typically be cuff inflation to allow audible air leak at
used for a genetically normal child of the same 25 cmH2O [76, 82].
age [79]. Data on the value of assessing cuff leak pres-
Contrary to what many clinicians have sure prior to extubation and peri-extubation corti-
thought, data have shown that cuffed endotra- costeroids administration in predicting or
cheal tubes are not associated with an increased preventing extubation failure, respectively, are
risk of post-extubation stridor as compared to conflicting. To obtain an accurate leak pressure,
uncuffed tubes [80]. In fact, using uncuffed the patient ideally should be neuromuscularly
endotracheal tube might increase the risk of blocked with the neck in neutral position and an
upper airway obstruction due to injury of the appropriately sized endotracheal tube for age in
subglottic epithelium that can occur when larger place. If no leak can be detected, especially when
uncuffed endotracheal tube is inserted to mini- higher airway pressures are used (>30 cmH2O),
mize leak [15, 81]. Indexing endotracheal tube caution regarding extubation should be exercised,
size to body surface area may help avoid this especially in neonates with small airway diame-
scenario. In a study of neonates recovering from ters (where only a small amount of subglottic
cardiac surgery, patients who failed extubation edema can result in upper airway obstruction)
had significantly larger inner diameter endotra- and patients with underlying airway anomalies.
cheal tubes (mean 16.3 mm/m2) compared to The caveat to the leak test is that secretions
patients who were successfully extubated (mean around endotracheal tube can affect the ability to
15.3 mm/m2), and this difference was largely detect a leak and can therefore cloud its interpre-
due to the size of the uncuffed tubes used in tation. For patients without a detectable leak at
patients who failed extubation (mean 17.6 mm/ reasonable airway pressures or patients with
m2). Based on these data, we recommend using underlying airway anomalies, use of systemic
endotracheal tubes with inner diameters less corticosteroids within 6–24 h prior to the planned
than 16 mm/m2. For example, in a neonate with extubation attempt may decrease the rate of post-
body surface area of 0.2 mm2, we would recom- extubation stridor [18, 83–85]. Routine use of
mend a 3.0 cuffed endotracheal tube (15 mm/ peri-extubation corticosteroids, on the other
ERRNVPHGLFRVRUJ
54 S. Abu-Sultaneh and C. W. Mastropietro
hand, based on current data, cannot be While post-extubation stridor and upper air-
recommended. way obstruction do frequently occur in children
with underlying cardiac disease, other causes of
extubation failure are more common. In two stud-
xtubation in Special Patient
E ies of neonates recovering from cardiac surgery,
Populations cardiac insufficiency was the most common
cause of extubation failure [13, 15]. Moreover,
ongenital or Acquired Cardiac
C risk factors for extubation failure in these studies
Disease were more specific to these neonates with cardiac
disease. For example, hypoplastic left heart syn-
drome, postoperative infection, prolonged open
Case Scenario sternotomy (greater than 4 days), and use of
A 2-year-old girl recently diagnosed with uncuffed endotracheal tubes have been found to
idiopathic dilated cardiomyopathy is be risk factors for extubation failure. Risk factors
admitted to the pediatric ICU on mechani- for extubation failure in general pediatric ICU
cal ventilation. Her respiratory status populations are also relevant to children with car-
improves gradually with diuretic therapy diac disease. In two studies from the Pediatric
and systemic afterload reduction using a Cardiac Critical Care Consortium, one of which
milrinone infusion. After 72 h of mechani- included a heterogeneous population of children
cal ventilation, she passes a 2 h extubation with medical and surgical cardiac disease; pro-
readiness trial, which used 10 cmH2O of longed duration of mechanical ventilation and
pressure support and 5 cmH2O of continu- underlying airway anomalies were identified as
ous positive airway pressure. She was extu- independent risk factors for extubation failure
bated to traditional nasal cannula without [12, 73]. Children who undergo cardiac surgery
any audible stridor. Her milrinone infusion are also at risk for recurrent laryngeal nerve
was stopped 12 h after extubation. Within injury and phrenic nerve injury, though vocal
the next 24 h, she was noted to have cord and diaphragm paresis, respectively, are not
increase work of breathing and hypoxemia typically apparent until after extubation [86, 87].
that did not respond to escalating respira- For patients who fail extubation secondary to
tory support to high-flow nasal cannula. post-extubation stridor and upper airway obstruc-
She was reintubated, and post-intubation tion following aortic arch intervention, where the
chest x-ray showed cardiomegaly and recurrent laryngeal nerve is located, evaluation
increased pulmonary edema. The bedside for vocal cord paresis via direct laryngoscopy
respiratory therapists ask if there are any should be pursued [88]. Likewise, for patients
tools available that could have alerted the with an elevated hemidiaphragm on chest x-ray
team to the risk of extubation failure in this following extubation failure or have no other dis-
patient. cernible causes for their extubation failure, evalu-
ation of diaphragm motion with ultrasound or
fluoroscopy is warranted.
Current data suggest that extubation failure in Though cardiac insufficiency commonly con-
cardiac critical care patients is around 6%, tributes to ventilator weaning and ERT failure in
though extubation failure is higher in neonates children with cardiac illness, interpretation of
recovering from cardiac surgery (around 12%). standard monitoring information such as changes
Similar to general pediatric ICU patients, most in vital signs and peripheral perfusion as surro-
extubation failures occur within the first 24 h gates of adequate cardiac output during an ERT
after extubation, and there is considerable varia- can be misleading. More objective measures such
tion in extubation failure rates across centers as arterial and venous blood gas measurements
[12, 15]. can be helpful, but these data are not always
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 55
obtainable (e.g., malfunctioning or discontinued (younger than 18 months of age) who underwent
arterial line, incorrect tip location of central cardiac surgery found that patient who were
venous catheter for accurate systemic venous placed on HFNC post-extubation had lesser use
oxygen saturation measurement). For these rea- of NIPPV when compared to patients who placed
sons, near-infrared spectroscopy (NIRS) moni- on conventional nasal cannula, but there was no
toring shows promise as an objective mean of difference of extubation failure between the two
evaluating the success or failure of an ERT [89]. groups [90].
NIRS provides real-time noninvasive data about Over the past decade, early extubation of
cerebral and somatic (most often renal) tissue patients who undergo surgery for congenital or
oxygen delivery and has become commonplace acquired heart disease in the operating room or
at many institutions. In one study, a 12% drop in shortly after arrival to the ICU has become stan-
somatic NIRS during an ERT was associated dard at many institutions. Benefits of early extu-
with an increased risk of extubation failure in this bation include decreased exposure to mechanical
population [89]. Though bedside clinicians ventilation and its associated complications
should not rely solely on NIRS measurements including sedation requirements; early initiation
when assessing extubation readiness in children of oral feeding; and decreased utilization of ICU
with cardiac disease, decreasing NIRS trends and hospital resources [91]. Additionally, sponta-
should give clinicians pause and prompt thor- neous breathing and early extubation have physi-
ough reassessment prior to extubation (Fig. 3.3). ologic benefits as well, as negative intrathoracic
Multiple respiratory modalities are being pressures are advantageous to most children with
used in pediatric cardiac ICU post-extubation. In cardiac disease, especially those with cavopul-
one study of 283 neonates who underwent car- monary connections [92, 93]. The benefits of
diac surgery, 55% were extubated to HFNC, early extubation must be weighed against the
31% were extubated to room air or NC, and 14% costs of extubation failure, which has been asso-
were extubated to NIPPV (Fig. 3.4) [15]. In a ciated with increased duration of mechanical
randomized, controlled trial of 89 patients ventilation, hospital length of stay, and mortality
1:00
2:00
2:59
3:59
4:59
5:59
6:59
7:59
8:59
9:59
10:59
ERRNVPHGLFRVRUJ
56 S. Abu-Sultaneh and C. W. Mastropietro
NC
HFNC
NIPPV
Fig. 3.4 Respiratory support provided upon extuba- v entilation (NIPPV). Each bar represents neonates suc-
tion. Number of neonates extubated to room air or oxygen cessfully extubated (green), neonates in which support
via nasal cannula (NC), oxygen via high-flow nasal can- was escalated to HFNC or NIPPV but not reintubated
nula (HFNC), and noninvasive positive-pressure (yellow), and neonates who required reintubation (red)
[7, 94]. In a recent report from the Virtual PICU Table 3.4 Relative contraindications for extubation in
Systems (VPS), LLC, database, 6810 (25%) of the operating room after pediatric congenital heart
surgery
27,398 children from 62 centers between 2009
and 2014 who underwent cardiac surgery were Neonatal age < 1 month
Cardiopulmonary bypass time > 150 min
extubated in the operating room [94]. Of the chil-
STAT mortality category 4 or 5
dren extubation in the operating room, 395 (6%)
Use of deep hypothermic circulatory arrest
required reintubation, and 44 patients (0.6%) Preoperative mechanical ventilation
died. Likewise, in a prospective observational Open sternotomy
single-center study of 1000 infants and children History of airway anomalies
who underwent surgery over an 8-month period Intraoperative respiratory issues (excessive secretions/
between 2012 and 2013, 45 of 871 (4.5%) of hypoxemia)
children extubated in the OR required reintuba- History of difficult intubation
tion, and 9 (1%) of these patients died [95]. Significant postoperative bleeding
Trisomy 21 with laryngomalacia
To identify optimal candidates for early extu-
Nonelective and emergent surgeries
bation, we recommend a multi-professional
Inotropic support (dopamine >5 mcg/kg/min, any dose
approach that can include pediatric cardiac inten- of epinephrine)
sive care providers, anesthesiologists, perfusion-
ists, cardiologists, cardiothoracic surgeons,
nursing, and respiratory therapists [95, 96]. Though this list is by no means absolute, we
Caudal analgesia; short-acting pain and sedative encourage individual institutions to develop simi-
agents like fentanyl, remifentanil, midazolam, or lar protocols to, at the very least, maintain a con-
medications that minimally suppress respiratory sistent approach to the practice of early extubation
drive like dexmedetomidine; and careful intraop- after cardiac surgery.
erative attention to fluid balance (i.e., modified
ultrafiltration) may help prepare patients for suc-
cessful extubation [91, 95–98]. At our institution, Traumatic Brain Injury
we utilized a multi-professional approach to
develop a list of evidenced-based relative contra- Patients with traumatic brain injury are at risk for
indications for extubation in the operating room, bulbar dysfunction, which can affect their ability to
which are listed in Table 3.4 [91, 94, 95, 97–102]. protect their airway, manage oropharyngeal secre-
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 57
ERRNVPHGLFRVRUJ
58 S. Abu-Sultaneh and C. W. Mastropietro
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 59
cal ventilator weaning protocols on respiratory respiratory tract disease. Pediatr Crit Care Med.
outcomes in infants and children: a randomized con- 2017;18(3):e106–e11.
trolled trial. JAMA. 2002;288(20):2561–8. 35. Curley MA, Wypij D, Watson RS, Grant MJ, Asaro
22. Restrepo RD, Fortenberry JD, Spainhour C, LA, Cheifetz IM, et al. Protocolized sedation vs
Stockwell J, Goodfellow LT. Protocol-driven ven- usual care in pediatric patients mechanically ven-
tilator management in children: comparison to tilated for acute respiratory failure: a randomized
nonprotocol care. J Intensive Care Med. 2004; clinical trial. JAMA. 2015;313(4):379–89.
19(5):274–84. 36. Riethmueller J, Borth-Bruhns T, Kumpf M, Vonthein
23. Jouvet P, Farges C, Hatzakis G, Monir A, Lesage F, R, Wiskirchen J, Stern M, et al. Recombinant human
Dupic L, et al. Weaning children from mechanical deoxyribonuclease shortens ventilation time in
ventilation with a computer-driven system (closed- young, mechanically ventilated children. Pediatr
loop protocol): a pilot study. Pediatr Crit Care Med. Pulmonol. 2006;41(1):61–6.
2007;8(5):425–32. 37. Prodhan P, Greenberg B, Bhutta AT, Hyde C,
24. Jouvet P, Eddington A, Payen V, Bordessoule A, Vankatesan A, Imamura M, et al. Recombinant
Emeriaud G, Gasco RL, et al. A pilot prospective human deoxyribonuclease improves atelectasis in
study on closed loop controlled ventilation and oxy- mechanically ventilated children with cardiac dis-
genation in ventilated children during the weaning ease. Congenit Heart Dis. 2009;4(3):166–73.
phase. Crit Care. 2012;16(3):R85. 38. Shein SL, Gallagher JT, Deakins KM, Weinert
25. Kim P, Salazar A, Ross PA, Newth CJ, Khemani DM. Prophylactic use of nebulized hypertonic
RG. Comparison of tidal volumes at the endotra- saline in mechanically ventilated children: a
cheal tube and at the ventilator. Pediatr Crit Care randomized blinded pilot study. Respir Care.
Med. 2015;16(9):e324–31. 2016;61(5):586–92.
26. Wolf GK, Walsh BK, Green ML, Arnold 39. Krause MF, Hoehn T. Chest physiotherapy in
JH. Electrical activity of the diaphragm during extu- mechanically ventilated children: a review. Crit Care
bation readiness testing in critically ill children. Med. 2000;28(5):1648–51.
Pediatr Crit Care Med. 2011;12(6):e220–4. 40. Deakins K, Chatburn RL. A comparison of intra-
27. Emeriaud G, Larouche A, Ducharme-Crevier L, pulmonary percussive ventilation and conven-
Massicotte E, Flechelles O, Pellerin-Leblanc AA, tional chest physiotherapy for the treatment of
et al. Evolution of inspiratory diaphragm activity in atelectasis in the pediatric patient. Respir Care.
children over the course of the PICU stay. Intensive 2002;47(10):1162–7.
Care Med. 2014;40(11):1718–26. 41. Mhanna MJ, Anderson IM, Iyer NP, Baumann
28. Lee EP, Hsia SH, Hsiao HF, Chen MC, Lin JJ, Chan A. The use of extubation readiness parameters: a
OW, et al. Evaluation of diaphragmatic function survey of pediatric critical care physicians. Respir
in mechanically ventilated children: an ultrasound Care. 2014;59(3):334–9.
study. PLoS One. 2017;12(8):e0183560. 42. Kaufman J, Rannie M, Kahn MG, Vitaska M, Wathen
29. Shah SK, Irshad M, Gupta N, Kabra SK, Lodha B, Peyton C, et al. An interdisciplinary initiative to
R. Hypophosphatemia in critically ill children: risk reduce unplanned extubations in pediatric critical
factors, outcome and mechanism. Indian J Pediatr. care units. Pediatrics. 2012;129(6):e1594–600.
2016;83(12–13):1379–85. 43. Klugman D, Berger JT, Spaeder MC, Wright A,
30. Williams S, Horrocks IA, Ouvrier RA, Gillis J, Ryan Pastor W, Stockwell DC. Acute harm: unplanned
MM. Critical illness polyneuropathy and myopathy extubations and cardiopulmonary resuscitation
in pediatric intensive care: a review. Pediatr Crit in children and neonates. Intensive Care Med.
Care Med. 2007;8(1):18–22. 2013;39(7):1333–4.
31. Arikan AA, Zappitelli M, Goldstein SL, Naipaul 44. da Silva PSL, Farah D, Fonseca MCM. Revisiting
A, Jefferson LS, Loftis LL. Fluid overload is asso- unplanned extubation in the pediatric intensive care
ciated with impaired oxygenation and morbidity unit: what’s new? Heart Lung. 2017;46(6):444–51.
in critically ill children. Pediatr Crit Care Med. 45. Curley MA, Arnold JH, Thompson JE, Fackler JC,
2012;13(3):253–8. Grant MJ, Fineman LD, et al. Clinical trial design –
32. Sinitsky L, Walls D, Nadel S, Inwald DP. Fluid effect of prone positioning on clinical outcomes in
overload at 48 hours is associated with respiratory infants and children with acute respiratory distress
morbidity but not mortality in a general PICU: ret- syndrome. J Crit Care. 2006;21(1):23–32; discus-
rospective cohort study. Pediatr Crit Care Med. sion −7.
2015;16(3):205–9. 46. Abu-Sultaneh S, Hole AJ, Tori AJ, Benneyworth
33. Ingelse SA, Wosten-van Asperen RM, Lemson J, BD, Lutfi R, Mastropietro CW. An Interprofessional
Daams JG, Bem RA, van Woensel JB. Pediatric Quality Improvement Initiative to Standardize
acute respiratory distress syndrome: fluid manage- Pediatric Extubation Readiness Assessment. Pediatr
ment in the PICU. Front Pediatr. 2016;4:21. Crit Care Med. 2017;18:e463.
34. Ingelse SA, Wiegers HM, Calis JC, van Woensel 47. Farias JA, Retta A, Alia I, Olazarri F, Esteban A,
JB, Bem RA. Early fluid overload prolongs Golubicki A, et al. A comparison of two methods to
mechanical ventilation in children with viral-lower perform a breathing trial before extubation in pedi-
ERRNVPHGLFRVRUJ
60 S. Abu-Sultaneh and C. W. Mastropietro
atric intensive care patients. Intensive Care Med. 60. Mayordomo-Colunga J, Medina A, Rey C, Concha
2001;27(10):1649–54. A, Menendez S, Los Arcos M, et al. Non invasive
48. Ouellette DR, Patel S, Girard TD, Morris PE, ventilation after extubation in paediatric patients: a
Schmidt GA, Truwit JD, et al. Liberation from preliminary study. BMC Pediatr. 2010;10:29.
mechanical ventilation in critically ill adults: an 61. Lum LC, Abdel-Latif ME, de Bruyne JA, Nathan
official American College of Chest Physicians/ AM, Gan CS. Noninvasive ventilation in a tertiary
American Thoracic Society clinical practice guide- pediatric intensive care unit in a middle-income
line: inspiratory pressure augmentation during country. Pediatr Crit Care Med. 2011;12(1):e7–13.
spontaneous breathing trials, protocols minimizing 62. Bach JR, Goncalves MR, Hamdani I, Winck
sedation, and noninvasive ventilation immediately JC. Extubation of patients with neuromuscular
after Extubation. Chest. 2017;151(1):166–80. weakness: a new management paradigm. Chest.
49. Ferguson LP, Walsh BK, Munhall D, Arnold JH. A 2010;137(5):1033–9.
spontaneous breathing trial with pressure support 63. Mortamet G, Amaddeo A, Essouri S, Renolleau S,
overestimates readiness for extubation in children. Emeriaud G, Fauroux B. Interfaces for noninvasive
Pediatr Crit Care Med. 2011;12(6):e330–5. ventilation in the acute setting in children. Paediatr
50. Khemani RG, Hotz J, Morzov R, Flink RC, Respir Rev. 2017;23:84–8.
Kamerkar A, LaFortune M, et al. Pediatric 64. Krawiec C, Carl D, Stetter C, Kong L, Ceneviva GD,
extubation readiness tests should not use pres- Thomas NJ. Challenges with implementation of a
sure support. Intensive Care Med. 2016;42(8): respiratory therapist-driven protocol of spontaneous
1214–22. breathing trials in the pediatric ICU. Respir Care.
51. Chavez A, dela Cruz R, Zaritsky A. Spontaneous 2017;62(10):1233–40.
breathing trial predicts successful extubation 65. Baumeister BL, el-Khatib M, Smith PG, Blumer
in infants and children. Pediatr Crit Care Med. JL. Evaluation of predictors of weaning from
2006;7(4):324–8. mechanical ventilation in pediatric patients. Pediatr
52. Grant MJ, Schneider JB, Asaro LA, Dodson BL, Pulmonol. 1997;24(5):344–52.
Hall BA, Simone SL, et al. Dexmedetomidine use in 66. Thiagarajan RR, Bratton SL, Martin LD, Brogan
critically ill children with acute respiratory failure. TV, Taylor D. Predictors of successful extubation in
Pediatr Crit Care Med. 2016;17(12):1131–41. children. Am J Respir Crit Care Med. 1999;160(5 Pt
53. Jongerius PH, van Tiel P, van Limbeek J, Gabreels 1):1562–6.
FJ, Rotteveel JJ. A systematic review for evidence 67. Finder JD, Birnkrant D, Carl J, Farber HJ, Gozal
of efficacy of anticholinergic drugs to treat drooling. D, Iannaccone ST, et al. Respiratory care of the
Arch Dis Child. 2003;88(10):911–4. patient with Duchenne muscular dystrophy: ATS
54. Parr JR, Buswell CA, Banerjee K, Fairhurst C, consensus statement. Am J Respir Crit Care Med.
Williams J, O’Hare A, et al. Management of drooling 2004;170(4):456–65.
in children: a survey of UK paediatricians’ clinical 68. Sharma GD. Pulmonary function testing in neu-
practice. Child Care Health Dev. 2012;38(2):287–91. romuscular disorders. Pediatrics. 2009;123(Suppl
55. Baudin F, Gagnon S, Crulli B, Proulx F, Jouvet 4):S219–21.
P, Emeriaud G. Modalities and complications 69. Hull J, Aniapravan R, Chan E, Chatwin M, Forton
associated with the use of high-flow nasal can- J, Gallagher J, et al. British Thoracic Society guide-
nula: experience in a pediatric ICU. Respir Care. line for respiratory management of children with
2016;61(10):1305–10. neuromuscular weakness. Thorax. 2012;67(Suppl
56. Schibler A, Pham TM, Dunster KR, Foster K, 1):i1–40.
Barlow A, Gibbons K, et al. Reduced intubation 70. Khemani RG, Sekayan T, Hotz J, Flink RC, Rafferty
rates for infants after introduction of high-flow GF, Iyer N, et al. Risk factors for pediatric Extubation
nasal prong oxygen delivery. Intensive Care Med. failure: the importance of respiratory muscle
2011;37(5):847–52. strength. Crit Care Med. 2017;45(8):e798–805.
57. Kawaguchi A, Yasui Y, de Caen A, Garros D. The 71. Hubble CL, Gentile MA, Tripp DS, Craig DM,
clinical impact of heated humidified high-flow nasal Meliones JN, Cheifetz IM. Dead space to tidal vol-
cannula on pediatric respiratory distress. Pediatr Crit ume ratio predicts successful extubation in infants
Care Med. 2017;18(2):112–9. and children. Crit Care Med. 2000;28(6):2034–40.
58. Fioretto JR, Ribeiro CF, Carpi MF, Bonatto RC, 72. Devor RL, Kang P, Wellnitz C, Nigro JJ, Velez
Moraes MA, Fioretto EB, et al. Comparison between DA, Willis BC. Pulmonary dead space fraction and
noninvasive mechanical ventilation and standard Extubation success in children after cardiac surgery.
oxygen therapy in children up to 3 years old with Pediatr Crit Care Med. 2018;19(4):301–9.
respiratory failure after extubation: a pilot prospec- 73. Benneyworth BD, Mastropietro CW, Graham EM,
tive randomized clinical study. Pediatr Crit Care Klugman D, Costello JM, Zhang W, et al. Variation
Med. 2015;16(2):124–30. in extubation failure rates after neonatal congenital
59. Cheifetz IM. Invasive and noninvasive pedi- heart surgery across pediatric cardiac critical care
atric mechanical ventilation. Respir Care. consortium hospitals. J Thorac Cardiovasc Surg.
2003;48(4):442–53; discussion 53–8. 2017;153(6):1519–26.
ERRNVPHGLFRVRUJ
3 Weaning and Extubation Readiness Assessment in Pediatric Patients 61
74. Baisch SD, Wheeler WB, Kurachek SC, Cornfield sis and surgical management. Pediatr Surg Int.
DN. Extubation failure in pediatric intensive care 2006;22(4):341–6.
incidence and outcomes. Pediatr Crit Care Med. 88. Lemmer J, Stiller B, Heise G, Hubler M, Alexi-
2005;6(3):312–8. Meskishvili V, Weng Y, et al. Postoperative phrenic
75. Fontela PS, Piva JP, Garcia PC, Bered PL, Zilles nerve palsy: early clinical implications and manage-
K. Risk factors for extubation failure in mechani- ment. Intensive Care Med. 2006;32(8):1227–33.
cally ventilated pediatric patients. Pediatr Crit Care 89. Foster CB, Spaeder MC, McCarter RJ, Cheng YI,
Med. 2005;6(2):166–70. Berger JT. The use of near-infrared spectroscopy
76. Newth CJ, Rachman B, Patel N, Hammer J. The use during an extubation readiness trial as a predic-
of cuffed versus uncuffed endotracheal tubes in pedi- tor of extubation outcome. Pediatr Crit Care Med.
atric intensive care. J Pediatr. 2004;144(3):333–7. 2013;14(6):587–92.
77. Kleinman ME, Chameides L, Schexnayder SM, 90. Testa G, Iodice F, Ricci Z, Vitale V, De Razza F,
Samson RA, Hazinski MF, Atkins DL, et al. Haiberger R, et al. Comparative evaluation of high-
Pediatric advanced life support: 2010 American flow nasal cannula and conventional oxygen therapy
Heart Association guidelines for cardiopulmonary in paediatric cardiac surgical patients: a randomized
resuscitation and emergency cardiovascular care. controlled trial. Interact Cardiovasc Thorac Surg.
Pediatrics. 2010;126(5):e1361–99. 2014;19(3):456–61.
78. Finn Davis K, Napolitano N, Li S, Buffman H, 91. Alghamdi AA, Singh SK, Hamilton BC, Yadava M,
Rehder K, Pinto M, et al. Promoters and barriers to Holtby H, Van Arsdell GS, et al. Early extubation
implementation of tracheal intubation airway safety after pediatric cardiac surgery: systematic review,
bundle: a mixed-method analysis. Pediatr Crit Care meta-analysis, and evidence-based recommenda-
Med. 2017;18(10):965–72. tions. J Card Surg. 2010;25(5):586–95.
79. Shott SR. Down syndrome: analysis of air- 92. Fogel MA, Durning S, Wernovsky G, Pollock AN,
way size and a guide for appropriate intubation. Gaynor JW, Nicolson S. Brain versus lung: hierarchy
Laryngoscope. 2000;110(4):585–92. of feedback loops in single-ventricle patients with
80. Weiss M, Dullenkopf A, Fischer JE, Keller C, superior cavopulmonary connection. Circulation.
Gerber AC, European Paediatric Endotracheal 2004;110(11 Suppl 1):II147–52.
Intubation Study Group. Prospective randomized 93. Morales DL, Carberry KE, Heinle JS, McKenzie
controlled multi-centre trial of cuffed or uncuffed ED, Fraser CD Jr, Diaz LK. Extubation in the operat-
endotracheal tubes in small children. Br J Anaesth. ing room after Fontan’s procedure: effect on practice
2009;103(6):867–73. and outcomes. Ann Thorac Surg. 2008;86(2):576–
81. Litman RS, Maxwell LG. Cuffed versus uncuffed 81. discussion 81–2.
endotracheal tubes in pediatric anesthesia: 94. Gupta P, Rettiganti M, Gossett JM, Yeh JC, Jeffries
the debate should finally end. Anesthesiology. HE, Rice TB, et al. Risk factors for mechani-
2013;118(3):500–1. cal ventilation and reintubation after pediat-
82. Schneider J, Mulale U, Yamout S, Pollard S, Silver ric heart surgery. J Thorac Cardiovasc Surg.
P. Impact of monitoring endotracheal tube cuff leak 2016;151(2):451–8 e3.
pressure on postextubation stridor in children. J Crit 95. Garg R, Rao S, John C, Reddy C, Hegde R, Murthy
Care. 2016;36:173–7. K, et al. Extubation in the operating room after car-
83. Anene O, Meert KL, Uy H, Simpson P, Sarnaik diac surgery in children: a prospective observational
AP. Dexamethasone for the prevention of postextu- study with multidisciplinary coordinated approach. J
bation airway obstruction: a prospective, random- Cardiothorac Vasc Anesth. 2014;28(3):479–87.
ized, double-blind, placebo-controlled trial. Crit 96. Mahle WT, Nicolson SC, Hollenbeck-Pringle D,
Care Med. 1996;24(10):1666–9. Gaies MG, Witte MK, Lee EK, et al. Utilizing a col-
84. Mhanna MJ, Zamel YB, Tichy CM, Super DM. The laborative learning model to promote early extuba-
“air leak” test around the endotracheal tube, as a pre- tion following infant heart surgery. Pediatr Crit Care
dictor of postextubation stridor, is age dependent in Med. 2016;17(10):939–47.
children. Crit Care Med. 2002;30(12):2639–43. 97. Mittnacht AJ, Thanjan M, Srivastava S, Joashi U,
85. Khemani RG, Randolph A, Markovitz Bodian C, Hossain S, et al. Extubation in the operat-
B. Corticosteroids for the prevention and treat- ing room after congenital heart surgery in children. J
ment of post-extubation stridor in neonates, chil- Thorac Cardiovasc Surg. 2008;136(1):88–93.
dren and adults. Cochrane Database Syst Rev. 98. Kin N, Weismann C, Srivastava S, Chakravarti S,
2009;3:CD001000. Bodian C, Hossain S, et al. Factors affecting the deci-
86. Smith BM, Ezeokoli NJ, Kipps AK, Azakie A, sion to defer endotracheal extubation after surgery
Meadows JJ. Course, predictors of diaphragm recov- for congenital heart disease: a prospective observa-
ery after phrenic nerve injury during pediatric car- tional study. Anesth Analg. 2011;113(2):329–35.
diac surgery. Ann Thorac Surg. 2013;96(3):938–42. 99. Winch PD, Nicholson L, Isaacs J, Spanos S, Olshove
87. Akay TH, Ozkan S, Gultekin B, Uguz E, Varan V, Naguib A. Predictors of successful early extuba-
B, Sezgin A, et al. Diaphragmatic paralysis after tion following congenital cardiac surgery in neonates
cardiac surgery in children: incidence, progno- and infants. Heart Lung Circ. 2009;18(4):271–6.
ERRNVPHGLFRVRUJ
62 S. Abu-Sultaneh and C. W. Mastropietro
100. DiNardo JA. Con: extubation in the operating room tion improves outcomes for neuromuscular disease
following pediatric cardiac surgery. J Cardiothorac patients with respiratory tract infections. Am J Phys
Vasc Anesth. 2011;25(5):877–9. Med Rehabil. 2005;84(2):83–8; discussion 9–91.
101. Miller JW, Vu D, Chai PJ, Kreutzer J, Hossain MM, 106. Fauroux B, Guillemot N, Aubertin G, Nathan
Jacobs JP, et al. Patient and procedural characteristics N, Labit A, Clement A, et al. Physiologic ben-
for successful and failed immediate tracheal extuba- efits of mechanical insufflation-exsufflation in
tion in the operating room following cardiac surgery children with neuromuscular diseases. Chest.
in infancy. Paediatr Anaesth. 2014;24(8):830–9. 2008;133(1):161–8.
102. Harris KC, Holowachuk S, Pitfield S, Sanatani S, 107. Racca F, Mongini T, Wolfler A, Vianello A, Cutrera
Froese N, Potts JE, et al. Should early extubation be R, Del Sorbo L, et al. Recommendations for
the goal for children after congenital cardiac surgery? anesthesia and perioperative management of patients
J Thorac Cardiovasc Surg. 2014;148(6):2642–7. with neuromuscular disorders. Minerva Anestesiol.
103. Cohn EC, Robertson TS, Scott SA, Finley AM, 2013;79(4):419–33.
Huang R, Miles DK. Extubation failure and tra- 108. Simon TD, Berry J, Feudtner C, Stone BL, Sheng
cheostomy placement in children with acute neu- X, Bratton SL, et al. Children with complex chronic
rocritical illness. Neurocrit Care. 2017; https://doi. conditions in inpatient hospital settings in the United
org/10.1016/j.cnc.2017.08.005. States. Pediatrics. 2010;126(4):647–55.
104. Miske LJ, Hickey EM, Kolb SM, Weiner DJ, Panitch 109. Benneyworth BD, Gebremariam A, Clark SJ,
HB. Use of the mechanical in-exsufflator in pediatric Shanley TP, Davis MM. Inpatient health care utiliza-
patients with neuromuscular disease and impaired tion for children dependent on long-term mechanical
cough. Chest. 2004;125(4):1406–12. ventilation. Pediatrics. 2011;127(6):e1533–41.
105. Vianello A, Corrado A, Arcaro G, Gallan F, Ori C,
Minuzzo M, et al. Mechanical insufflation-exsuffla-
ERRNVPHGLFRVRUJ
Management of Status
Asthmaticus in Critically Ill 4
Children
I. Federico Fernandez Nievas, Allison Fahy,
Michelle Olson, and K. J. S. Anand
ERRNVPHGLFRVRUJ
64 I. Federico Fernandez Nievas et al.
Asthma exacerbations not only produce fre- mixed granulocytic inflammation, carrying fea-
quent symptoms and increase medical resource tures of eosinophilic asthma and non-eosino-
utilization but are also associated with substantial philic asthma (see recent reviews [13, 14]).
disability, impaired quality of life, and avoidable The key pathophysiologic feature of status
deaths in children. Surveys from the International asthmaticus is inflammation of small airways
Study of Asthma and Allergies in Childhood leading to increased airway resistance and dra-
(ISAAC) found positive correlations between the matically extending the time required for full
prevalence of wheezing in childhood (6–7 years exhalation. Residual air remains “trapped” in
of age) with national mortality rates (r = 0.32, alveoli at the time of the next inhalation, thus
p < 0.05) and hospital admission rates (r = 0.73, alveolar volumes may increase progressively
p = 0.003) among 13–14-year-old children, with each breath and lead to increased end-alveo-
whereas severe wheezing at 6–7 years of age had lar and intrathoracic pressures. Consequently, the
stronger correlations with mortality at end-expiratory alveolar pressures are often two-
13–14 years (r = 0.42, p < 0.025) [10]. Given the to threefold higher than normal, increasing the
strong positive correlations between asthma required changes in pressure to reach the negative
symptom prevalence, hospital admissions, and alveolar pressures necessary to generate airflow
mortality, it is not surprising that status asthmati- by the patient [15, 16]. For example, continuous
cus is a leading source of critical illness in chil- positive airway pressure (CPAP) functionally
dren and the most common medical emergency normalizes this gradient and has been shown to
[6]. While evaluating these patients, clinicians reduce the respiratory load in the spontaneously
must maintain a high index of suspicion and breathing asthma patient [15]. Although changes
greater vigilance, since 13% of patients with in the respiratory system mainly reduce dynamic
near-fatal asthma present with their first-ever lung compliance due to increased airway resis-
attack of status asthmaticus [11]. Among those tance [17, 18], atelectasis develops around the
with a previous history of asthma, 63% had no overdistended alveoli to reduce static lung com-
prior hospital admissions for asthma in the year pliance as well. Greater resistance to airflow
preceding their presentation with near-fatal status increases work of breathing, presenting initially
asthmaticus, and 86% had no prior admissions to as increased distress and expiratory effort. Later,
the pediatric intensive care unit (PICU) [11]. worsening lung hyperinflation limits full dia-
phragmatic relaxation, thus reducing the effi-
ciency of diaphragmatic function and calling into
Pathophysiology: A Brief Précis play the use of accessory respiratory muscles
with increased work of breathing during inhala-
The National Institutes of Health (NIH) define tion and exhalation [19, 20].
asthma as a chronic inflammatory disorder of air- The progressively increasing lung volumes
ways in which many cells including mast cells seen in status asthmaticus can also affect cardiac
and eosinophils contribute to symptoms associ- ventricular function. Alveolar hyperinflation-
ated with variable airflow obstruction that is associated airway obstruction, increasing micro-
reversible either spontaneously or with medica- atelectasis, hypoxia-induced pulmonary
tions [12]. Although the detailed immunology of vasoconstriction in atelectatic areas, β-agonist
asthma is beyond the scope of this chapter, clini- and/or dehydration-induced metabolic acidosis,
cians must recognize that eosinophilic asthma and respiratory acidosis from impending respira-
may include patients with allergic and non-aller- tory failure all contribute to increases in right
gic eosinophilic inflammation, whereas non- ventricular afterload. Moreover, spontaneously
eosinophilic asthma may include patients with breathing patients during an asthma exacerbation
paucigranulocytic and neutrophilic inflamma- can have peak inspiratory pressures as extreme as
tion. Some patients may also present with a −35 cm H2O [21]. This negative intrathoracic
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 65
pressure directly accentuates left ventricular of their clinical assessment. The most vital aspect
afterload, with increasing likelihood of of clinical assessment in asthmatic patients is
pulmonary edema and worsening gas exchange serial physical exams by bedside clinicians at
[22]. Pulsus paradoxus is a physiological mani- least hourly or every couple of hours to appreci-
festation of the exaggerated variation in systolic ate changes in their clinical trajectory.
blood pressure associated with high intrathoracic Presenting symptoms usually include a his-
pressures during inspiration [23, 24]. Therapy tory of cough, increased respiratory rate,
with β2-agonists increases heart rate, contributing increased work of breathing and disordered
to progressively diminished ventricular filling breathing patterns. Auscultation of the chest will
time and consequently lower cardiac output. demonstrate turbulent airflow with diffuse
Switching from negative pressure ventilation to wheezing and a prolonged expiratory phase due
positive-pressure ventilation in patients who to air trapping by their hyper-reactive small bron-
require intubation is likely to result in acute chial airways. Children with mild-to-moderate
hypotension secondary to decreased venous status asthmaticus present with wheezing during
return [22]. In addition, many of the sedative the expiratory phase only, those with moderate-
agents used for intubation also have vasodilatory to-severe status asthmaticus manifest wheezing
and myocardial depressant effects, particularly during both inhalation and exhalation phases, and
among children, further affecting cardiac output patients with critical status asthmaticus may
and increasing the risks of cardiac arrest during present with a “silent chest” since wheezing is
or immediately after tracheal intubation. only appreciated if there is adequate airflow in
the small bronchi [24]. All patients with status
asthmaticus must be monitored closely in a pedi-
Clinical Assessment atric ICU, with serial physical exams being sup-
plemented with continuous cardiorespiratory
Patients suffering from status asthmaticus require monitoring and intermittent arterial blood gas
rapid and frequent assessments, watching for sampling.
signs of respiratory distress or impending respi-
ratory failure. A focused approach both for posi-
tive and negative findings on the physical exam Diagnostic Evaluation
will ensure that children are treated with the
required escalation of care as necessary. The The evaluation of children with status asthmati-
level of alertness is particularly important in their cus is mostly based on clinical findings, biplanar
neurological assessment, since lethargy may be chest radiographs, an arterial blood gas to evalu-
due to fatigue or due to hypercarbia, and this ate gas exchange, a complete hemogram to
observation may be confused with their natural exclude eosinophilia or other abnormalities,
sleep cycle during nighttime hours. Children who tests to exclude viral or atypical pneumonitis,
are lethargic due to fatigue or hypercarbia very and a basic metabolic profile to rule out dehydra-
likely have impending respiratory failure and tion or β2-agonist-induced hypokalemia. More
warrant close attention in the PICU. advanced testing is rarely required but may be
Children may also exhibit profoundly indicated to exclude parasitic, toxic, or environ-
increased work of breathing as demonstrated mental triggers for status asthmaticus. Bedside
with retractions or paradoxical thoracoabdominal asthma scores may facilitate communication
breathing. A prompt evaluation of the patients’ between members of the pediatric ICU team,
general appearance, airway patency, effective- though most clinical asthma scores lack suffi-
ness of respiratory effort (including both inhala- cient validation and are limited by the subjective
tion and exhalation), adventitious breath sounds, evaluation of the variables comprising these
and adequacy of circulation form the foundations scores [25, 26].
ERRNVPHGLFRVRUJ
66 I. Federico Fernandez Nievas et al.
Bronchoscopy may be indicated to rule out for- Inhaled nitric oxide is not indicated for the treat-
eign body aspiration and bilateral bronchoma- ment of status asthmaticus [32], but measure-
lacia or diffuse bacterial bronchitis, but the ments of exhaled nitric oxide may estimate the
vast majority of patients can be managed with- extent of inflammatory airways in asthmatic
out bronchoscopy. The risks versus benefits of patients. An increase in exhaled nitric oxide is
bronchoscopy must be weighed carefully known to accompany eosinophilic inflammation
because instrumenting hyper-reactive and [32]. Although measured concentrations of
inflamed airways may lead to significant clini- exhaled nitric oxide may help gage the patho-
cal deterioration, life-threatening hypoxemia, physiologic trajectory of patients with status
and cardiac arrest. In a single-center case series asthmaticus, the accuracy and prognostic value
of 44 ventilated asthmatic patients, bronchos- of this investigational test has not been estab-
copies revealed thick mucus plugs, secretions, lished in clinical studies as of yet [33]. Exhaled
and bronchial casts. Saline lavage of obstruc- nitric oxide is increased in steroid-naïve asth-
tive airways was well tolerated with demon- matic subjects during status asthmaticus,
strable improvements in pulmonary although this returns to baseline after appropriate
compliance, reduced duration of mechanical anti-inflammatory treatment is administered [33].
ventilation, but no differences in the PICU Additional studies are needed before testing for
length of stay [27]. Occasionally, fiberoptic exhaled nitric oxide demonstrates its effective-
bronchoscopy can be used to instill human ness to bedside clinicians.
recombinant DNase or other mucolytic agents
into plugged airways, but this practice is not
routine at most centers [28]. Case Scenario
An 11-year-old female with a history of
moderate-to-severe asthma presents to a
Xenon Ventilation Computed local emergency department with wheezing
Tomography progressing to severe respiratory distress
over the previous 24 h. Her respiratory rate
Recent studies have examined the usefulness of is 40 breaths per min, her heart rate is 125
xenon ventilation computed tomography in asth- breaths per minute, and her oxygen satura-
matic patients [29]. This technique is a relatively tion as determined by pulse oximetry is
new method to evaluate pulmonary functions and 96%. She receives three consecutive alb-
ventilation defects in asthmatic patients by exam- uterol nebulization treatments (2.5 mg
ining alteration in xenon trapping following each), one nebulization treatment with
administration of methacholine and salbutamol ipratropium bromide (500mcg), and one
[30]. Although this testing may potentially dose of intravenous methylprednisolone
unmask airway abnormalities contributing to (1 mg/kg). One hour later, she is assessed
ventilation-perfusion mismatch, its application to by a pediatric intern, who notes that her
pediatric patients with status asthmaticus remains vital signs and her respiratory effort have
controversial. The potential usage of xenon is not improved, and persistent prolonged
considered relatively benign since it is nonreac- expiration and wheezing are apparent on
tive in the body and disposed of from the lungs auscultation of her chest. After this assess-
without any systemic effects in critically ill ment, the pediatric intern asked her attend-
patients [29, 31]. ing physician what the next best step in the
management would be.
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 67
In a study including 13,552 children critically ill β-Agonists cause smooth muscle relaxation by
with asthma, marked clinical variability in phar- activating the β2-adrenergic receptor and increas-
macological management and mechanical sup- ing cyclic adenosine monophosphate (cAMP)
port was noted (21% were treated in Collaborative concentrations in smooth muscle cells, which
Pediatric Critical Care Research Network inhibits the release of calcium ion from intracel-
(CPCCRN) PICUs, 79% were treated in non- lular stores and reduces the membrane calcium
CPCCRN PICUs). Wide variations occurred in entry and its intracellular sequestration [35].
the frequency of medication use in CPCCRN Albuterol is a racemic mixture of R-albuterol
centers − ipratropium bromide 41–84% patients, and S-albuterol. The R-enantiomer is pharmaco-
terbutaline 11–74% patients, magnesium 23–64% logically active, and the S-enantiomer is inactive.
patients, and methylxanthines 0–46% patients − Levalbuterol is the pure R-enantiomer available
implying a lack of consensus with regard to the as a preservative-free solution. In comparative
pharmacological management of children with trials, the use of equivalent doses of levalbuterol
status asthmaticus [34]. We present the following was not superior to albuterol [36]. Albuterol
sections recognizing that different clinicians may remains the drug of choice for treatment of status
choose different elements from this menu based asthmaticus. Depending on different variables,
on the clinical features of specific patients, local approximately 10–20% of the albuterol dose will
institutional practices, resource availability, and reach the lungs. The National Asthma Expert
personal preference. We have also summarized a Panel recommends nebulized albuterol doses for
suggested algorithmic approach to the manage- asthma exacerbations in children younger than
ment of status asthmaticus in Fig. 4.1. 12 years of 0.15 mg/kg (minimum dose 2.5 mg)
Mechanical Ventila
ati
Ventilation
Status Asthmaticus: Mechanical Ventilation
A Stepwise Approach
Bolus dose 10 mcg/kg IV over 10-20 min; continuous infusion: typical
IV Terbutaline at starting dose 1 mcg/kg/min (max 4 mcg/kg/min); titrate by 0.5
Ventilation
Noninvasive Ven
mcg/kg/min every 15 min as needed
Load 5.7 mg/kg over 20 min followed by continuous infusion: < 9 y: 1 mg/kg/h;
IV Aminophylline 9-12 y: 0.89mg/kg/h; 12-16 y: 0.63 mg/kg/h; >16 yrs: 0.51 mg/kg/h
Check aminophylline level 1 hr after load; therapeutic goal: 10-15 µg/ml
IV Magnesium Bolus Dosing: < 30 kg: 50-75 mg/kg/dose; > 30 kg: 25-50 mg/kg/dose, infused over 30-45 min
.
Albuterol and Ipratropium All patients should have received these therapies in the ED prior to PICU admission
nebulization treatments,
IV Corticosteroids
ERRNVPHGLFRVRUJ
68 I. Federico Fernandez Nievas et al.
every 20 min for three doses, followed by 0.15– On the other hand, parasympathetic nerves
0.3 mg/kg (maximum 10 mg) every 1–4 h. supplying the lungs also have muscarinic recep-
For patients not showing clinical improve- tors. M2-receptors on postganglionic parasympa-
ment, continuous albuterol nebulization at 0.15– thetic nerves limit acetylcholine release by a
0.5 mg/kg/h is recommended. Larger doses, up negative feedback mechanism. Thus, blocking
to 30 mg/h, can be used for critical or near-fatal the M2-receptors on parasympathetic nerves with
asthma. Existing evidence supports the use of muscarinic antagonist will increase acetylcholine
continuous albuterol nebulization in pediatric release and potentiate vagally induced broncho-
patients with status asthmaticus and impending constriction. Parasympathetic neuronal
respiratory failure, leading to faster clinical M2-receptors are susceptible to viral infections
improvement and decreased duration of hospital and exposure to ozone (which decreases their
stay when compared with intermittent albuterol function) and are less functional in patients with
nebulization and decreased hospitalization rate asthma. The mechanism for this latter reduction
when continuous albuterol regimen is used in the in neuronal M2-receptor functions in multifacto-
emergency department. Additionally, continuous rial and involves the downregulation of receptor
albuterol treatment at these doses is safe and expression and blockade by endogenous antago-
well tolerated [37]. Reported doses of albuterol nists [40]. Anticholinergic drugs block M2 and
used in pediatric patients often exceed the expert M3 muscarinic receptors on the airway smooth
panel recommendations [38]. Data on the effec- muscle, glands, and nerves with similar affinity,
tiveness and safety of these higher doses com- thereby impairing smooth muscle contraction
pared to traditional recommended doses are and decreasing airway secretions while simulta-
sorely needed. neously augmenting acetylcholine release, sup-
porting the rationale to develop selective
M3-receptors medications [41].
Anticholinergic Agents Ipratropium bromide is a synthetic quaternary
ammonium derivative with an isopropyl group at
The parasympathetic nervous system contributes the N-carbon atom of atropine that limits its sys-
to airway resistance via acetylcholine-mediated temic absorption. Inhaled ipratropium targets the
airway smooth muscle contraction; regulation of muscarinic receptors in the bronchial airways
mucus secretion, ciliary beat frequency, and without the systemic effects of atropine, such as
mucus clearance via mucosal glands and epithe- tachycardia. The low oral absorption of ipratro-
lial cells; vasodilation by smooth muscle relax- pium is beneficial, since up to 90% of an aerosol-
ation in blood vessels; and modulation of ized dose may be swallowed. Ipratropium is a
inflammation [39]. There are five subtypes of nonselective muscarinic receptor inhibitor, which
muscarinic receptors (M1-M5), which belong to produces bronchodilation by the inhibition of
the larger group of G protein-coupled receptors. acetylcholine-mediated bronchospasm without
Acetylcholine stimulates these receptors. affecting the mucociliary clearance. Ipratropium
M3-receptors located on the airway smooth mus- has no impact on intraocular pressure or pupil-
cle mediate bronchoconstriction, and lary size even when up to four times the recom-
M3-receptors located on the submucosal cells mended dose is used; nevertheless, it can produce
regulate glandular secretion. M2 muscarinic prolonged pupillary dilatation when sprayed
receptors are also on the bronchial smooth mus- accidently into the eyes [42]. The half-life of
cle, which indirectly cause smooth muscle con- ipratropium is 3–4 h, the onset of action is 15 min,
traction by reducing β-adrenoceptor-mediated peak effects occur at 1–2 h after administration,
relaxation through inhibition of adenylate and duration of action is 4–6 h.
cyclase. Blockade of both M2 and M3-receptors Early administration of three or more repeated
on airway smooth muscle should therefore inhibit doses of inhaled ipratropium with β2-agonists has
bronchoconstriction. been shown to decrease the rate of hospital
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 69
admission for pediatric and adult patients with study of patients 2–18 years of age who presented
moderate-to-severe status asthmaticus by 30% to the emergency department with mild-to-mod-
[43]. Improvements in spirometry and clinical erate acute asthma exacerbations, Qureshi et al.
scores have also occurred with the use of multi- evaluated the efficacy of oral prednisone versus
dose protocols, without increasing side effects oral dexamethasone [49]. Compared to oral pred-
[43]. A double-blind, randomized study in adults nisone, oral dexamethasone did show similar
with severe asthma found that those receiving efficacy with improved compliance and fewer
ipratropium for 36 h or longer were discharged side effects when compared with oral predniso-
home earlier than those that receiving ipratro- lone [49]. The authors hypothesized that the
pium for 12 h [44]. In another study, patients improved compliance was observed because
using fenoterol/ipratropium versus fenoterol or dexamethasone is more palatable than oral
ipratropium alone were found to have greater prednisone with shorter prescription duration.
bronchodilator effects in children with acute Hydrocortisone, a short-acting corticosteroid
asthma [45]. with relatively less anti-inflammatory potency
Intermittent ipratropium therapy is recom- than the other corticosteroids, is not commonly
mended in hospitalized patients with acute used for acute asthma exacerbations.
asthma because of its high safety profile and doc- Corticosteroids that are highly potent with a
umented beneficial impact [46]. The effect of fast onset are the most desirable for patients with
ipratropium is dose-dependent, with the recom- status asthmaticus. Systemic corticosteroids are
mended dose range from 250 to 500 mcg [47]. In preferred and have also been shown to be supe-
one case report, a 13-year-old patient with status rior to inhaled corticosteroids for status asthmati-
asthmaticus refractory to β2-agonist treatments cus, with reduced need for hospitalization [50].
showed improvement after starting continuous Intravenous methylprednisolone is the most com-
ipratropium at 1000 mcg per hour [48]. Despite monly recommended due to its potent glucocorti-
this report, however, continuous ipratropium coid effects and limited mineralocorticoid effects
therapy has not been systematically investigated. [51]. Patients presenting with status asthmaticus
should receive intravenous methylprednisolone
2 mg/kg (maximum dose 80 mg) early in the
Anti-inflammatory Drugs course of their management, as its onset of action
occurs approximately 4 h after administration.
Corticosteroids are a cornerstone for both acute There are no significant added benefits from sys-
and long-term asthma management, as airway temic corticosteroids at doses above 80 mg/day
edema and inflammation are the most prominent or 2 mg/kg/day with regards to pulmonary func-
pathological features of the disease. Indeed, use tion, rate of hospital admission, or hospital length
of systemic corticosteroids to treat status asth- of stay [23].
maticus is not controversial and will therefore be
discussed briefly. Methylprednisolone, dexa-
methasone, prednisone, and hydrocortisone are Magnesium
the traditional corticosteroids used for critically
ill patients, but they differ in glucocorticoid Magnesium sulfate acts in the airway by blocking
potency, duration of onset, and mode of adminis- voltage-sensitive calcium channels, inhibiting
tration. For patients with mild-to-moderate calcium uptake and calcium-myosin interactions,
asthma exacerbations, oral prednisone is the most thus producing smooth muscle relaxation [52].
common therapy prescribed. Oral dexametha- Magnesium also stabilizes T-cells and inhibits
sone has also been used to treat patients with mast cell degranulation, consequently decreasing
asthma in the acute care setting. Dexamethasone histamine release and inflammatory mediators.
is a long-acting corticosteroid that has tradition- Other mechanisms of action include inhibition of
ally been used for children with croup. In one acetylcholine release by the cholinergic motor
ERRNVPHGLFRVRUJ
70 I. Federico Fernandez Nievas et al.
terminals and stimulation of nitric oxide and asthma exacerbations. In a systematic review
prostacyclin production. When intravenous mag- study including adult and pediatric patients,
nesium sulfate is added to β2-agonists and sys- inhaled magnesium has shown some clinical
temic corticosteroids, it improves pulmonary benefit in patients with acute severe asthma
function in children and adults and reduces hos- attacks with no apparent serious adverse effects
pital admissions by 30% in children and in a [57]. In another study containing adult asthma
lesser degree in adults [53]. patients, treatment with nebulized magnesium
Magnesium has a rapid onset of action and its sulfate improved pulmonary functions and
duration of action is limited by renal clearance. reduced hospital admissions in adults by 37%
As serum magnesium concentrations increase, [53]. In contrast, a study of pediatric patients
renal excretion of magnesium increases linearly, with moderate-to-severe asthma using 800 mg
potential hindering achievement of goal serum of nebulized magnesium failed to reduce their
concentrations. Specifically, serum magnesium time to discharge [58]. Similarly, a recent
concentrations of up to 4 mg/dL are thought to Cochrane review reported minimal clinical
be required for airway smooth muscle relax- improvement and no reduction in hospital
ation. In a cohort study of children with status admissions when nebulized magnesium was
asthmaticus, using continuous magnesium infu- used concomitantly with β2-agonists and ipratro-
sions at 40 mg/kg/h for 4 h after a loading dose pium therapy [59].
of 50–75 mg/kg was safe and attained magne- In conclusion, intravenous magnesium sulfate
sium levels of 4.4 mg/dL at the end of the infu- should be used in children, especially those with
sion [54]. Pediatric patients receiving magnesium life-threatening critical asthma and those not
sulfate in this study improved clinically, with responding to initial treatments, with low risk for
reductions in tachycardia and tachypnea. The severe adverse effects. Nebulized inhaled magne-
optimal dose of magnesium sulfate is currently sium therapy, however, cannot currently be
unclear, with recommended dose ranges of recommended.
25–100 mg/kg to a maximal dose of 2,000 mg,
independent of weight. In obese patients, mag-
nesium dose should be based on their ideal Methylxanthine Drugs
body weight. We recommend an initial bolus
dose of 50–75 mg/kg for children weighing The most commonly used methylxanthine in
<30 kg and 25–50 mg/kg for those weighing acute asthma exacerbation is theophylline.
>30 kg, infused over a period of 30–45 minutes Theophylline brings relief to asthmatic patients
to improve acute respiratory symptoms and due to its direct bronchodilator effect [23].
avoid hypotension. In cases of life-threatening Parenteral form of theophylline is aminophylline,
status asthmaticus refractory to standard treat- which is a 2:1 complex of theophylline and ethyl-
ment, a continuous infusion of 25–30 mg/kg/h enediamine (aminophylline is converted to the-
for children <30 kg and 15–20 mg/kg/h for ophylline systemically, such that 1 mg
children weighing >30 kg, up to a maximum aminophylline = 0.8 mg theophylline).
dose of 40 mg/kg/h can be added. Titration to Aminophylline is FDA approved as an adjunctive
the desired clinical effect should be based on treatment for acute asthma exacerbations in all
target serum magnesium concentrations of 3.5– age groups, including children older than 1 year.
4.5 mg/dL and tolerability [55]. On the other Theophylline pharmacokinetics are age-
hand, there is limited evidence that intravenous dependent, which affects its pediatric dosing rec-
magnesium is beneficial in asthma exacerba- ommendations. The elimination half-life of
tions of less severity (i.e., moderate-to-severe theophylline gradually decreases during the first
asthma presentations) [56]. year of life from ~ 24 h in term neonates to
Nebulized inhaled magnesium sulfate has between 2 and 10 h (mean 4 hs) in children 1 to
also been trialed as a therapeutic agent for 9 years old and between 3 and 16 h (mean 8 h) in
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 71
adults [60]. Theophylline’s primary actions are hospital cost was significantly less in patients
dose-dependent, as lower serum concentrations who received only theophylline: $280 US dol-
result in anti-inflammatory and immunomodula- lars compared to ~$4000 in each of the other
tory effects, whereas higher serum concentra- two cohorts.
tions show greater bronchodilator effects [61]. Prior to these two small but important trials,
Due to potential toxicity, serum theophylline several studies were published suggesting lack of
concentrations should be followed at least every benefit from theophylline therapy for patients
12 h, and repeated bolus doses or infusion rates with acute asthma exacerbations. Most of the
should be adjusted based on the target steady- patients in these studies however were not criti-
state serum concentrations, with a traditional cally ill [65–68]. Recently, a meta-analysis of 52
goal range of 10–15 μ/mL. Obese patients should study arms from 42 trials involving theophylline,
have their ideal body weight used for dosage cal- some of which included adults and others
culation [23]. Theophylline dosing is summa- included children, involving intravenous
rized in Fig. 4.1. theophylline concluded that, when given with
In a prospective, randomized, controlled trial bronchodilators and corticosteroids, theophylline
by Ream et al. published in 2001, 47 children can be helpful and represents a cost-effective and
admitted to a pediatric ICU with status asthmat- safe choice for patients with status asthmaticus
icus receiving traditional β-agonist, anticholin- [69]. We recommend initiation of intravenous
ergic, and corticosteroid therapy were aminophylline in patients who are not improving
randomized to receive IV theophylline or pla- or worsening despite inhaled β-agonists, sys-
cebo [62]. Twenty-three patients who received temic corticosteroids, ipratropium bromide, and
theophylline were compared with 24 controls, intravenous magnesium.
with cohorts having similar clinical asthma As previously concluded by Mahemuti et al.
scores (i.e., Wood-Downes score [63]) prior to given the low cost of theophylline, and its similar
study enrollment. Faster recovery times (defined efficacy and rate of side effects compared with
by clinical asthma scores ≤3) were noted in other drugs, we suggest that theophylline, when
both non-intubated and intubated patients given with bronchodilators and corticosteroids, is
receiving theophylline as compared to con- a cost-effective and safe choice for acute asthma
trols – 19 ± 3 h versus 31 ± 5 h for non-intubated exacerbations [69].
patients and 66 ± 10 h versus 191 ± 33 h for
intubated patients. Four years later, Wheeler and
coworkers examined the effects of intravenous Intravenous Albuterol
theophylline and terbutaline in a randomized
controlled trial in 36 patients with status asth- One study examined the advantage of combining
maticus receiving traditional β-agonist and cor- intravenous albuterol (salbutamol) to inhaled alb-
ticosteroid therapy [64]. More specifically, they uterol in children during the initial management
randomized patients to receive adjunctive ther- of severe acute asthma in the emergency depart-
apy with intravenous theophylline plus placebo, ment. Children receiving a single dose of 15 μg/
intravenous terbutaline plus placebo, or intrave- kg intravenous albuterol over 10 min with inhaled
nous theophylline plus terbutaline. The authors albuterol had a shorter recovery time and earlier
observed no differences in clinical asthma score discharge compared with a group of children who
over time, length of pediatric intensive care unit received inhaled albuterol alone. The intravenous
stay, or incidence of adverse events between the albuterol cohort had increased incidence of trem-
three groups, with the exception of a higher ors, but no other notable side effects [70]. These
incidence of nausea in children who received data support the use of a single-dose intravenous
both theophylline and terbutaline. Importantly, albuterol in addition to inhaled albuterol in the
in a cost analysis that included the cost of both emergency department in children with severe
drugs and the cost of theophylline levels, median asthma.
ERRNVPHGLFRVRUJ
72 I. Federico Fernandez Nievas et al.
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 73
h ospitalization or other important clinical out- of drugs, and the logistical issues of administer-
comes [78, 85]. Its role in mechanically venti- ing an inhaled anesthetic gas for long periods of
lated asthmatics is also likely limited. In a very time outside of the operating room setting have
recent prospective study of 13 adults with severe also prevented widespread use. If the capabilities
asthma or chronic obstructive pulmonary disease to administer inhaled isoflurane are available,
exacerbations requiring mechanical ventilation, close monitoring including invasive arterial pres-
use of heliox led to only modest reductions in sure measurement is mandatory, as observational
peak inspiratory pressure and partial pressure of studies have described frequent side effects, the
CO2 measurements and had little effect indices of most common of which is hypotension requiring
dynamic hyperinflation (e.g., plateau pressure vasoactive infusions (77%) [74]. Other reported
and total positive end-expiratory pressure) [86]. side effects include arrhythmias, neurologic
Use of heliox therapy is costly, has limited abnormalities, accumulation of inorganic
utility in patients with hypoxemia, and can be fluoride, tolerance during therapy, and abstinence
technically difficult to provide, especially in syndrome after discontinuation [89, 94, 95].
mechanically ventilated patients [87]. Many ven-
tilators are not equipped to deliver heliox safely
[88], and for mechanical ventilators that are Respiratory Support
capable of administering heliox, delivery of the
appropriate fractional oxygen component, vol- Oxygen Therapy
ume measurements, and valve functioning can
also be adversely affected [87]. Routine use of The goals of treatment as outlined by the
heliox for patients with status asthmaticus can National Asthma Education and Prevention
therefore not be recommended, and its use in Program (NAEPP) are to treat significant hypox-
mechanically ventilated patients should be emia, reverse the airflow obstruction, and reduce
avoided. On the other hand, it may be useful in the likelihood of future episodes [96]. Oxygen
select patients, such as children with status asth- should undoubtedly be applied to treat hypox-
maticus without significant hypoxemia, though emia, but high inspired oxygen concentrations
clinicians should be prepared for endotracheal are infrequently required for patients with status
intubation and intervene quickly if no improve- asthmatics. Thus, the need for higher inspired
ment or clinical worsening is noted. oxygen concentrations should raise concern for
other respiratory insults, most important and
life-threatening of which are pneumothoraces
Isoflurane [97, 98].
ERRNVPHGLFRVRUJ
74 I. Federico Fernandez Nievas et al.
breathing, less tachycardia, and shorter ED the use of NIPPV in children with status asthmat-
times in children with status asthmaticus [101, icus as a stop-gap measure to prevent worsening
102]. Despite low-grade evidence specific to respiratory failure and avoid the potential com-
this population supporting this modality, HFNC plications of endotracheal intubation and
as a safe and effective method of respiratory mechanical ventilation [106, 107].
support for patients with acute asthma exacerba-
tions is widely accepted and thus not
controversial. Tracheal Intubation
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 75
tension, relative hypovolemia from dehydration, observation of all intubated asthmatic patients is
β-agonist-induced metabolic acidosis, hypercar- required for the first few hours after initiating
bia from respiratory muscle fatigue, reduced left mechanical ventilation, since the risk of life-
ventricular filling from tachycardia, and increased threatening complications and unanticipated
left ventricular afterload from negative inspira- hemodynamic effects is highest in that period.
tory intrathoracic pressures [21]. Additional con-
cerns include instrumenting the hyper-reactive
airway for intubation, which accentuates respira- Mechanical Ventilation
tory obstruction; analgesic sedative drugs used
for intubation, which can cause systemic vasodi- Although there are no absolute criteria for
lation and reduce myocardial contractility; transi- mechanical ventilation, clinicians should con-
tioning to positive-pressure ventilation, which sider stepwise escalation in support from inhaled
further reduces venous return; and the risk of therapies to intravenous therapies and noninva-
barotrauma associated with mechanical ventila- sive mechanical support and, ultimately, culmi-
tion and hyperinflated lungs [112]. nating in invasive mechanical ventilation for
Prior to intubation, efforts must be made to refractory or rapidly progressive respiratory fail-
maximize management of status asthmaticus ure. Mechanical ventilation, either noninvasive or
using bronchodilators, intravenous corticoste- invasive, is generally designed to overcome the
roids, magnesium, aminophylline, judicious cor- dramatically increased work of breathing inher-
rection of hypovolemia with intravenous fluids, ent to status asthmaticus. Most clinicians agree
and, if possible, a trial of noninvasive ventilation that a low respiratory rate, long expiratory time
[97, 113–117]. Ketamine is the first choice for ventilator strategy is optimal to permit CO2 clear-
pre-intubation sedation because of its potentially ance. The amount of positive end-expiratory
advantageous bronchodilator and hemodynamic pressure (PEEP) to set on the ventilator, on the
effects [118–121], but the use of adjunctive short- other hand, is controversial [129–133]. After ini-
acting sedatives like midazolam, propofol, or tiation of mechanical ventilation, total PEEPtot
dexmedetomidine can also be helpful. Strategies should be measured using an expiratory pause
for achieving sedation and analgesia must take maneuver [134]. PEEPtot represents the sum of
into account their systemic vasodilatory, respira- the PEEP set by the ventilator and the PEEP gen-
tory depressant, myocardial contractility, and erated by air trapping, typically referred to as
other side effects [122]. Advanced airway skills auto-PEEP. While some clinicians support mea-
are essential, and neuromuscular blockade is suring auto-PEEP to regulate the ventilator PEEP
desirable to minimize the number of attempts in order to actually reduce alveolar hyperinflation
required for successful intubation [123]. and also recruit areas of atelectasis [133], others
Following intubation, chest rise and ausculta- do not support this concept and recommend using
tion of breath sounds may be difficult to elicit in minimal PEEP [112, 124]. The controversy sur-
the setting of severe airway obstruction. For this rounding the practice of higher PEEP settings in
reason, in-line end-tidal CO2 monitoring is essen- status asthmaticus is primarily due to reports of
tial to the verify correct placement of the endotra- paradoxical responses that lead to undesirable
cheal tube [124–127]. Manual ventilation increases in FRC in some patients with status
following intubation must limit the tidal volumes asthmaticus [129–131]. Interpretation of the data
and respiratory rates used, to avoid accentuating generated from these studies however has varied,
barotrauma and allowing complete exhalation and a consensus on optimal PEEP strategy for
between breaths. Attempts to normalize pH by status asthmaticus has yet to be reached [132].
correcting hypercapnia are unnecessary and We recommend careful titration of ventilator
potentially harmful [128]. Correction of hypox- PEEP close to the PEEPtot, which attempts to
emia and permissive hypercapnia (i.e., pH 7.2– maintain airway patency during expiration.
7.3) are reasonable goals. Close bedside Patients must be monitored closely, and v entilator
ERRNVPHGLFRVRUJ
76 I. Federico Fernandez Nievas et al.
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 77
11. Newth CJ, Meert KL, Clark AE, et al. Fatal and near-
cost and safe “second-tier” therapies for fatal asthma in children: the critical care perspective.
J Pediatr. 2012;161:214–21 e3.
patients with refractory status 12. Bousquet J, Michel FB. International consensus
asthmaticus. report on diagnosis and management of asthma.
• Routine use of intravenous terbutaline Allergy. 1992;47:129–32.
or heliox gas mixture cannot be recom- 13. Papi A, Brightling C, Pedersen SE, Reddel
HK. Asthma. Lancet. 2017;391(10122):783–800.
mended, but these therapies may have a 14. Russell RJ, Brightling C. Pathogenesis of asthma:
role in preventing intubation and implications for precision medicine. Clin Sci (Lond).
mechanical ventilation in select patients. 2017;131:1723–35.
• Compelling but not conclusive evidence 15. Martin JG, Shore S, Engel LA. Effect of continuous
positive airway pressure on respiratory mechanics
exists to support the use of noninvasive and pattern of breathing in induced asthma. Am Rev
positive pressure for children with status Respir Dis. 1982;126:812–7.
asthmaticus. 16. Martin JG, Shore SA, Engel LA. Mechanical load
and inspiratory muscle action during induced
asthma. Am Rev Respir Dis. 1983;128:455–60.
17. Dunn R, Szefler SJ. Severe asthma in pediatric
patients. Pathophysiology and unmet needs. Ann Am
References Thorac Soc. 2016;13(Suppl 1):S103–4.
18. King GG, James A, Harkness L, Wark
1. Vos T, Flaxman AD, Naghavi M, et al. Years lived PAB. Pathophysiology of severe asthma: we’ve only
with disability (YLDs) for 1160 sequelae of 289 dis- just started. Respirology. 2018;23:262–71.
eases and injuries 1990–2010: a systematic analy- 19. Thill PJ, McGuire JK, Baden HP, Green TP,
sis for the Global Burden of Disease Study 2010. Checchia PA. Noninvasive positive-pressure ven-
Lancet. 2012;380:2163–96. tilation in children with lower airway obstruction.
2. Asher I, Pearce N. Global burden of asthma among Pediatr Crit Care Med. 2004;5:337–42.
children. Int J Tuberc Lung Dis. 2014;18:1269–78. 20. Martin J, Powell E, Shore S, Emrich J, Engel
3. Ellwood P, Asher MI, Billo NE, et al. The Global LA. The role of respiratory muscles in the hyper-
Asthma Network rationale and methods for Phase inflation of bronchial asthma. Am Rev Respir Dis.
I global surveillance: prevalence, severity, manage- 1980;121:441–7.
ment and risk factors. Eur Respir J. 2017;49:1601605. 21. Stalcup SA, Mellins RB. Mechanical forces pro-
4. To T, Stanojevic S, Moores G, et al. Global asthma ducing pulmonary edema in acute asthma. N Engl J
prevalence in adults: findings from the cross-sec- Med. 1977;297:592–6.
tional world health survey. BMC Public Health. 22. Buda AJ, Pinsky MR, Ingels NB Jr, Daughters
2012;12:204. GT 2nd, Stinson EB, Alderman EL. Effect of
5. Asher MI, Montefort S, Bjorksten B, et al. intrathoracic pressure on left ventricular perfor-
Worldwide time trends in the prevalence of symp- mance. N Engl J Med. 1979;301:453–9.
toms of asthma, allergic rhinoconjunctivitis, and 23. Alangari AA. Corticosteroids in the treatment of
eczema in childhood: ISAAC Phases One and Three acute asthma. Ann Thorac Med. 2014;9:187–92.
repeat multicountry cross-sectional surveys. Lancet. 24. Nichols DG, Shaffner DH. Rogers’ textbook of pedi-
2006;368:733–43. atric intensive care. Philadelphia: Wolters Kluwer;
6. Bohn D, Kissoon N. Acute asthma. Pediatr Crit Care 2016.
Med. 2001;2:151–63. 25. van der Windt D. Promises and pitfalls in the
7. Leynaert B, Sunyer J, Garcia-Esteban R, et al. evaluation of pediatric asthma scores. J Pediatr.
Gender differences in prevalence, diagnosis and 2000;137:744–6.
incidence of allergic and non-allergic asthma: a pop- 26. Baxt WG. Prospective application of an asthma
ulation-based cohort. Thorax. 2012;67:625–31. severity rule. Acad Emerg Med. 2002;9:868–9.
8. Andersson M, Hedman L, Bjerg A, Forsberg B, 27. Maggi JC, Nussbaum E, Babbitt C, Maggi FE,
Lundback B, Ronmark E. Remission and persis- Randhawa I. Pediatric fiberoptic bronchoscopy as
tence of asthma followed from 7 to 19 years of age. adjunctive therapy in acute asthma with respiratory
Pediatrics. 2013;132:e435–42. failure. Pediatr Pulmonol. 2012;47:1180–4.
9. Raghavan D, Jain R. Increasing awareness of 28. Noizet O, Leclerc F, Leteurtre S, et al. Plastic bron-
sex differences in airway diseases. Respirology. chitis mimicking foreign body aspiration that needs
2016;21:449–59. a specific diagnostic procedure. Intensive Care Med.
10. Anderson HR, Gupta R, Kapetanakis V, et al. 2003;29:329–31.
International correlations between indicators of prev- 29. Bedi A, Murray JM, Dingley J, Stevenson MA, Fee
alence, hospital admissions and mortality for asthma JH. Use of xenon as a sedative for patients receiving
in children. Int J Epidemiol. 2008;37:573–82. critical care. Crit Care Med. 2003;31:2470–7.
ERRNVPHGLFRVRUJ
78 I. Federico Fernandez Nievas et al.
30. Jung JW, Kwon JW, Kim TW, et al. New insight tionship for nebulized ipratropium in asthmatic chil-
into the assessment of asthma using xenon ventila- dren. J Pediatr. 1984;105:1002–5.
tion computed tomography. Ann Allergy Asthma 48. Koumbourlis AC, Mastropietro C. Continuous
Immunol. 2013;111:90–5.e2. inhalation of ipratropium bromide for acute asthma
31. Reinelt H, Marx T, Kotzerke J, et al. Hepatic function refractory to beta2-agonist treatment. J Pediatr
during xenon anesthesia in pigs. Acta Anaesthesiol Pharmacol Ther. 2015;20:66–9.
Scand. 2002;46:713–6. 49. Qureshi F, Zaritsky A, Poirier MP. Comparative
32. Ashutosh K, Phadke K, Jackson JF, Steele D. Use of efficacy of oral dexamethasone versus oral pred-
nitric oxide inhalation in chronic obstructive pulmo- nisone in acute pediatric asthma. J Pediatr.
nary disease. Thorax. 2000;55:109–13. 2001;139:20–6.
33. Yates DH. Role of exhaled nitric oxide in asthma. 50. Rachelefsky G. Treating exacerbations of asthma
Immunol Cell Biol. 2001;79:178–90. in children: the role of systemic corticosteroids.
34. Bratton SL, Newth CJ, Zuppa AF, et al. Critical Pediatrics. 2003;112:382–97.
care for pediatric asthma: wide care variability 51. Expert Panel Report 3 (EPR-3): guidelines
and challenges for study. Pediatr Crit Care Med. for the diagnosis and management of asthma-
2012;13:407–14. summary report 2007. J Allergy Clin Immunol.
35. Johnson M. The beta-adrenoceptor. Am J Respir Crit 2007;120:S94–138.
Care Med. 1998;158:S146–53. 52. Lindeman KS, Hirshman CA, Freed AN. Effect
36. Qureshi F, Zaritsky A, Welch C, Meadows T, Burke of magnesium sulfate on bronchoconstriction
BL. Clinical efficacy of racemic albuterol versus lev- in the lung periphery. J Appl Physiol (1985).
albuterol for the treatment of acute pediatric asthma. 1989;66:2527–32.
Ann Emerg Med. 2005;46:29–36. 53. Shan Z, Rong Y, Yang W, et al. Intravenous and neb-
37. Papo MC, Frank J, Thompson AE. A prospective, ulized magnesium sulfate for treating acute asthma
randomized study of continuous versus intermittent in adults and children: a systematic review and meta-
nebulized albuterol for severe status asthmaticus in analysis. Respir Med. 2013;107:321–30.
children. Crit Care Med. 1993;21:1479–86. 54. Egelund TA, Wassil SK, Edwards EM, Linden S,
38. Arnold DH, Moore PE, Abramo TJ, Hartert Irazuzta JE. High-dose magnesium sulfate infusion
TV. The dilemma of albuterol dosing for acute protocol for status asthmaticus: a safety and phar-
asthma exacerbations in pediatric patients. Chest. macokinetics cohort study. Intensive Care Med.
2011;139:472. 2013;39:117–22.
39. Buels KS, Fryer AD. Muscarinic receptor antago- 55. Glover ML, Machado C, Totapally BR. Magnesium
nists: effects on pulmonary function. Handb Exp sulfate administered via continuous intravenous
Pharmacol. 2012;208:317–41. infusion in pediatric patients with refractory wheez-
40. Moulton BC, Fryer AD. Muscarinic receptor antag- ing. J Crit Care. 2002;17:255–8.
onists, from folklore to pharmacology; finding 56. Griffiths B, Kew KM. Intravenous magnesium sul-
drugs that actually work in asthma and COPD. Br J fate for treating children with acute asthma in the
Pharmacol. 2011;163:44–52. emergency department. Cochrane Database Syst
41. Scott GD, Fryer AD. Role of parasympathetic nerves Rev. 2016;4:CD011050.
and muscarinic receptors in allergy and asthma. 57. Knightly R, Milan SJ, Hughes R, et al. Inhaled mag-
Chem Immunol Allergy. 2012;98:48–69. nesium sulfate in the treatment of acute asthma.
42. Gross NJ. Ipratropium bromide. N Engl J Med. Cochrane Database Syst Rev. 2017;11:CD003898.
1988;319:486–94. 58. Alansari K, Ahmed W, Davidson BL, Alamri M,
43. Rodrigo GJ, Castro-Rodriguez JA. Anticholinergics Zakaria I, Alrifaai M. Nebulized magnesium for
in the treatment of children and adults with acute moderate and severe pediatric asthma: a randomized
asthma: a systematic review with meta-analysis. trial. Pediatr Pulmonol. 2015;50:1191–9.
Thorax. 2005;60:740–6. 59. Blitz M, Blitz S, Hughes R, et al. Aerosolized mag-
44. Brophy C, Ahmed B, Bayston S, Arnold A, McGivern nesium sulfate for acute asthma: a systematic review.
D, Greenstone M. How long should Atrovent be Chest. 2005;128:337–44.
given in acute asthma? Thorax. 1998;53:363–7. 60. Weinberger M, Hendeles L. Theophylline in asthma.
45. Watson WT, Becker AB, Simons FE. Comparison N Engl J Med. 1996;334:1380–8.
of ipratropium solution, fenoterol solution, and their 61. Magnussen H, Reuss G, Jorres R. Theophylline
combination administered by nebulizer and face has a dose-related effect on the airway response to
mask to children with acute asthma. J Allergy Clin inhaled histamine and methacholine in asthmatics.
Immunol. 1988;82:1012–8. Am Rev Respir Dis. 1987;136:1163–7.
46. Aaron SD. The use of ipratropium bromide for 62. Ream RS, Loftis LL, Albers GM, Becker BA,
the management of acute asthma exacerbation in Lynch RE, Mink RB. Efficacy of IV the ophylline
adults and children: a systematic review. J Asthma. in children with severe status asthmaticus. Chest.
2001;38:521–30. 2001;119:1480–8.
47. Davis A, Vickerson F, Worsley G, Mindorff C, Kazim 63. Wood DW, Downes JJ, Leeks HI. A clinical scoring
F, Levison H. Determination of dose-response rela- system for the diagnosis of respiratory failure: pre-
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 79
liminary report on childhood status asthmaticus. Am 77. Otis AB, Bembower WC. Effect of gas density on
J Dis Child. 1972;123:227–8. resistance to respiratory gas flow in man. J Appl
64. Wheeler DS, Jacobs BR, Kenreigh CA, Bean JA, Physiol. 1949;2:300–6.
Hutson TK, Brilli RJ. Theophylline versus terbuta- 78. Kudukis TM, Manthous CA, Schmidt GA, Hall
line in treating critically ill children with status asth- JB, Wylam ME. Inhaled helium-oxygen revisited:
maticus: a prospective, randomized, controlled trial. effect of inhaled helium-oxygen during the treat-
Pediatr Crit Care Med. 2005;6:142–7. ment of status asthmaticus in children. J Pediatr.
65. DiGiulio G, Kercsmar C, Krug S, Alpert S, Marx 1997;130:217–24.
C. Hospital treatment of asthma: lack of benefit from 79. Haynes JM, Sargent RJ, Sweeney EL. Use of heliox
theophylline given in addition to nebulized albuterol to avoid intubation in a child with acute severe asthma
and intravenously administered corticosteroid. J and hypercapnia. Am J Crit Care. 2003;12:28–30.
Pediatr. 1993;122:464–9. 80. Austan F. Heliox inhalation in status asthmaticus
66. Carter E, Cruz M, Chesrown S, et al. Efficacy of and respiratory acidemia: a brief report. Heart Lung.
intravenously administered theophylline in chil- 1996;25:155–7.
dren hospitalized with severe asthma. J Pediatr. 81. Carter LER, Webb CCR, Moffitt CDR. Evaluation
1993;122:470–6. of heliox in children hospitalized with acute severe
67. Strauss RE, Wertheim DL, Bonagura VR, et al. asthma: a randomized crossover trial. Chest.
Aminophylline therapy does not improve outcome 1996;109:1256–61.
and increases adverse effects in children hospital- 82. Kim IK, Phrampus E, Venkataraman S, et al.
ized with acute asthmatic exacerbations. Pediatrics. Helium/oxygen-driven albuterol nebulization in
1994;93:205–10. the treatment of children with moderate to severe
68. Goodman DC, Littenberg B, O’Connor GT, asthma exacerbations: a randomized, controlled
et al. Theophylline in acute childhood asthma: a trial. Pediatrics. 2005;116:1127–33.
meta-analysis of its efficacy. Pediatr Pulmonol. 83. Rivera ML, Kim TY, Stewart GM, Minasyan L,
1996;21:211–8. Brown L. Albuterol nebulized in heliox in the ini-
69. Mahemuti G, Zhang H, Li J, Tieliwaerdi N, Ren tial ED treatment of pediatric asthma: a blinded,
L. Efficacy and side effects of intravenous the- randomized controlled trial. Am J Emerg Med.
ophylline in acute asthma: a systematic review 2006;24:38–42.
and meta-analysis. Drug Des Devel Ther. 84. Bigham MT, Jacobs BR, Monaco MA, et al. Helium/
2018;12:99. oxygen-driven albuterol nebulization in the manage-
70. Browne GJ, Penna AS, Phung X, Soo M. Randomised ment of children with status asthmaticus: a random-
trial of intravenous salbutamol in early manage- ized, placebo-controlled trial. Pediatr Crit Care Med.
ment of acute severe asthma in children. Lancet. 2010;11:356–61.
1997;349:301–5. 85. Rodrigo G, Rodrigo C, Pollack C, Rowe B. Helium-
71. Nievas IF, Anand KJS. Severe acute asthma exac- oxygen mixtures for non-intubated acute asthma
erbation in children: a stepwise approach for esca- patients. Cochrane Library. Issue 4. Oxford: Update
lating therapy in a pediatric intensive care unit. J Software; 2001.
Pediatr Pharmacol Ther. 2013;18:88–104. 86. Leatherman JW, Romero RS, Shapiro RS. Lack of
72. Doymaz S, Schneider J, Sagy M. Early administra- Benefit of Heliox During Mechanical Ventilation of
tion of terbutaline in severe pediatric asthma may Subjects With Severe Air-Flow Obstruction. Respir
reduce incidence of acute respiratory failure. Ann Care. 2018;63:375–9.
Allergy Asthma Immunol. 2014;112:207–10. 87. Hashemian SM, Fallahian F. The use of heliox in
73. Bogie AL, Towne D, Luckett PM, Abramo TJ, Wiebe critical care. Int J Crit Illn Inj Sci. 2014;4:138.
RA. Comparison of intravenous terbutaline versus 88. Gainnier M, Forel J-M. Clinical review: use of
normal saline in pediatric patients on continuous helium-oxygen in critically ill patients. Crit Care.
high-dose nebulized albuterol for status asthmaticus. 2006;10:241.
Pediatr Emerg Care. 2007;23:355–61. 89. Shankar V, Churchwell KB, Deshpande
74. Travers A, Jones AP, Kelly K, Barker SJ, Camargo JK. Isoflurane therapy for severe refractory sta-
CA, Rowe BH. Intravenous beta2-agonists for acute tus asthmaticus in children. Intensive Care Med.
asthma in the emergency department. Cochrane 2006;32:927–33.
Database Syst Rev. 2001;2:CD002988. 90. Johnston RG, Noseworthy TW, Friesen EG,
75. Travers AH, Milan SJ, Jones AP, Camargo Yule HA, Shustack A. Isoflurane therapy for sta-
CA Jr, Rowe BH. Addition of intravenous tus asthmaticus in children and adults. Chest.
beta(2)-agonists to inhaled beta(2)-agonists 1990;97:698–701.
for acute asthma. Cochrane Database Syst Rev. 91. Otte RW, Fireman P. Isoflurane anesthesia for the
2012;12:CD010179. treatment of refractory status asthmaticus. Ann
76. Travers AH, Jones AP, Camargo CA Jr, Milan SJ, Allergy. 1991;66:305–9.
Rowe BH. Intravenous beta(2)-agonists versus intra- 92. Rice M, Hatherill M, Murdoch IA. Rapid response
venous aminophylline for acute asthma. Cochrane to isoflurane in refractory status asthmaticus. Arch
Database Syst Rev. 2012;12:CD010256. Dis Child. 1998;78:395–6.
ERRNVPHGLFRVRUJ
80 I. Federico Fernandez Nievas et al.
93. Carrie S, Anderson TA. Volatile anesthetics for 108. Malmstrom K, Kaila M, Korhonen K, et al.
status asthmaticus in pediatric patients: a compre- Mechanical ventilation in children with severe
hensive review and case series. Paediatr Anaesth. asthma. Pediatr Pulmonol. 2001;31:405–11.
2015;25:460–7. 109. Bratton SL, Odetola FO, McCollegan J, Cabana MD,
94. Arnold JH, Truog RD, Molengraft JA. Tolerance Levy FH, Keenan HT. Regional variation in ICU
to isoflurane during prolonged administration. care for pediatric patients with asthma. J Pediatr.
Anesthesiology. 1993;78:985–8. 2005;147:355–61.
95. Arnold JH, Truog RD, Rice SA. Prolonged 110. Bratton SL, Roberts JS. Variation in the use of
administration of isoflurane to pediatric patients mechanical ventilation for asthma: how big a gap?
during mechanical ventilation. Anesth Analg. Pediatr Crit Care Med. 2007;8:186–7.
1993;76:520–6. 111. Carroll CL, Smith SR, Collins MS, Bhandari A,
96. Camargo CA Jr, Rachelefsky G, Schatz M. Managing Schramm CM, Zucker AR. Endotracheal intuba-
asthma exacerbations in the emergency department: tion and pediatric status asthmaticus: site of origi-
summary of the National Asthma Education and nal care affects treatment. Pediatr Crit Care Med.
Prevention Program Expert Panel Report 3 guide- 2007;8:91–5.
lines for the management of asthma exacerbations. 112. Tuxen DV, Lane S. The effects of ventilatory pat-
J Emerg Med. 2009;37:S6–S17. tern on hyperinflation, airway pressures, and cir-
97. Carroll CL, Zucker AR. The increased cost of culation in mechanical ventilation of patients with
complications in children with status asthmaticus. severe air-flow obstruction. Am Rev Respir Dis.
Pediatr Pulmonol. 2007;42:914–9. 1987;136:872–9.
98. Sarnaik AP, Daphtary KM, Meert KL, Lieh-Lai MW, 113. Carroll CL, Sala KA. Pediatric status asthmaticus.
Heidemann SM. Pressure-controlled ventilation in Crit Care Clin. 2013;29:153–66.
children with severe status asthmaticus. Pediatr Crit 114. Cox RG, Barker GA, Bohn DJ. Efficacy, results,
Care Med. 2004;5:133–8. and complications of mechanical ventilation in
99. Nielsen KR, Ellington LE, Gray AJ, Stanberry LI, children with status asthmaticus. Pediatr Pulmonol.
Smith LS, DiBlasi RM. Effect of high-flow nasal 1991;11:120–6.
cannula on expiratory pressure and ventilation in 115. Dworkin G, Kattan M. Mechanical ventila-
infant, pediatric, and adult models. Respir Care. tion for status asthmaticus in children. J Pediatr.
2018;63:147–57. 1989;114:545–9.
100. Coletti KD, Bagdure DN, Walker LK, Remy
116. Leatherman J. Mechanical ventilation for severe
KE, Custer JW. High-flow nasal cannula uti- asthma. Chest. 2015;147:1671–80.
lization in pediatric critical care. Respir Care. 117. Zimmerman JL, Dellinger RP, Shah AN, Taylor
2017;62:1023–9. RW. Endotracheal intubation and mechani-
101. Baudin F, Buisson A, Vanel B, Massenavette B, cal ventilation in severe asthma. Crit Care Med.
Pouyau R, Javouhey E. Nasal high flow in manage- 1993;21:1727–30.
ment of children with status asthmaticus: a retro- 118. Denmark TK, Crane HA, Brown L. Ketamine to
spective observational study. Ann Intensive Care. avoid mechanical ventilation in severe pediatric
2017;7:55. asthma. J Emerg Med. 2006;30:163–6.
102. Powell CV. Acute severe asthma. J Paediatr Child 119. Rock MJ, Reyes de la Rocha S, L’Hommedieu CS,
Health. 2016;52:187–91. Truemper E. Use of ketamine in asthmatic children
103. Mayordomo-Colunga J, Medina A, Rey C, et al. to treat respiratory failure refractory to conventional
Non-invasive ventilation in pediatric status asth- therapy. Crit Care Med. 1986;14:514–6.
maticus: a prospective observational study. Pediatr 120. Jat KR, Chawla D. Ketamine for management of
Pulmonol. 2011;46:949–55. acute exacerbations of asthma in children. Cochrane
104. Basnet S, Mander G, Andoh J, Klaska H, Verhulst Database Syst Rev. 2012;11:CD009293.
S, Koirala J. Safety, efficacy, and tolerability of 121. Jones BP, Paul A. Management of acute asthma in
early initiation of noninvasive positive pressure the pediatric patient: an evidence-based review.
ventilation in pediatric patients admitted with status Pediatr Emerg Med Pract. 2013;10:1–23.
asthmaticus: a pilot study. Pediatr Crit Care Med. 122. deBacker J, Hart N, Fan E. Neuromuscular blockade
2012;13:393–8. in the 21st century management of the critically ill
105. Gupta P, Tang X, Gossett JM, et al. Association of patient. Chest. 2017;151:697–706.
center volume with outcomes in critically ill children 123. Tarquinio KM, Howell JD, Montgomery V, et al.
with acute asthma. Ann Allergy Asthma Immunol. Current medication practice and tracheal intuba-
2014;113:42–7. tion safety outcomes from a prospective multicenter
106. Silva Pde S, Barreto SS. Noninvasive ventilation in observational cohort study. Pediatr Crit Care Med.
status asthmaticus in children: levels of evidence. 2015;16:210–8.
Rev Bras Ter Intensiva. 2015;27:390–6. 124. Leatherman J. Life-threatening asthma. Clin Chest
107. Rehder KJ. Adjunct therapies for refractory Med. 1994;15:453–79.
status asthmaticus in children. Respir Care. 125. Paret G, Kornecki A, Szeinberg A, et al. Severe acute
2017;62:849–65. asthma in a community hospital pediatric intensive
ERRNVPHGLFRVRUJ
4 Management of Status Asthmaticus in Critically Ill Children 81
care unit: a ten years’ experience. Ann Allergy 131. Tuxen DV. Detrimental effects of positive end-expi-
Asthma Immunol. 1998;80:339–44. ratory pressure during controlled mechanical ven-
126. Roberts JS, Bratton SL, Brogan TV. Acute severe tilation of patients with severe airflow obstruction.
asthma: differences in therapies and outcomes Am Rev Respir Dis. 1989;140:5–9.
among pediatric intensive care units. Crit Care Med. 132. Stewart TE, Slutsky AS. Occult, occult auto-PEEP in
2002;30:581–5. status asthmaticus. Crit Care Med. 1996;24:379–80.
127. Deho A, Lutman D, Montgomery M, Petros A, 133. Banner MJ, Downs JB, Kirby RR, Smith RA,
Ramnarayan P. Emergency management of children Boysen PG, Lampotang S. Effects of expiratory flow
with acute severe asthma requiring transfer to inten- resistance on inspiratory work of breathing. Chest.
sive care. Emerg Med J. 2010;27:834–7. 1988;93:795–9.
128. Darioli R, Perret C. Mechanical controlled hypoven- 134. Reddy VG. Auto-PEEP: how to detect and how
tilation in status asthmaticus. Am Rev Respir Dis. to prevent – a review. Middle East J Anaesthesiol.
1984;129:385–7. 2005;18:293–312.
129. Caramez MP, Borges JB, Tucci MR, et al. Paradoxical 135. Briassoulis GC, Venkataraman ST, Vasilopoulos
responses to positive end-expiratory pressure in AG, Sianidou LC, Papadatos JH. Air leaks from the
patients with airway obstruction during controlled respiratory tract in mechanically ventilated children
ventilation. Crit Care Med. 2005;33:1519–28. with severe respiratory disease. Pediatr Pulmonol.
130. Qvist J, Andersen JB, Pemberton M, Bennike 2000;29:127–34.
KA. High-level PEEP in severe asthma. N Engl J
Med. 1982;307:1347–8.
ERRNVPHGLFRVRUJ
Part II
Cardiovascular Controversies
ERRNVPHGLFRVRUJ
Medical Management of Acute
Fulminant Myocarditis 5
Fabio Savorgnan and Paul A. Checchia
ERRNVPHGLFRVRUJ
86 F. Savorgnan and P. A. Checchia
Table 5.1 Common etiology of myocarditis observed in 7.7%, and cough was observed in
Infectious Inflammatory 17% of patients studied. Further, the study found
Enterovirus (Coxsackie B) Kawasaki disease that gastrointestinal (GI) tract symptoms – spe-
Adenovirus Rheumatologic cifically nausea, vomiting, abdominal pain, and
disease
diarrhea – were more frequent than cardiopulmo-
Parvovirus B19
nary symptoms at the onset of the disease (45%
HHV6 (human herpesvirus)
Lyme (Borrelia) Toxins vs. 25%, P = 0.01) [4]. Regarding cardiovascular
Cytomegalovirus (CMV) Cocaine manifestations at admission, this study also found
Epstein-Bar virus (EBV) Iron congestive heart failure (CHF) was present in
(hemochromatosis) 36% of children, arrhythmias were present in
Hepatitis C Copper (Wilson’s) 22%, syncope was noted in 10%, and cardiogenic
Chagas (South America, Arsenic shock occurred in 13% [4].
presents with right heart
failure)
Myocarditis can be confused with acute myo-
cardial infarction when presenting with chest
Ginsberg and Parrillo [1]
pain, S-T changes on electrocardiogram (ECG),
elevations of troponin, and creatine kinase-MB
t hereafter, stops the inflammatory process, would (CK-MB). Assessment of the coronary arteries
not injure the myocardium in a way that would performed by echocardiogram, computed tomog-
result in fulminant myocarditis. Patients with an raphy (CT) scan, and/or magnetic resonance
uncontrolled immune response suffer damage to imaging (MRI) can help distinguish myocarditis
the myocardium due to persistent inflammation
and ultimately progress to fulminant myocarditis
[5, 12–14].
Case Scenario
In phase 3, patients develop to chronic dilated
cardiomyopathy. This end result is likely due to After returning from summer vacation, a
(1) massive damage during the initial phase, (2) 15-year-old, previously healthy girl pres-
persistent inflammation, or (3) no eradication of ents to the emergency department with a
the virus leading to ongoing infection and inflam- 3-day history of stomach pain and vomiting
mation [5, 12–14] (Fig. 5.1). and a 1-day history of labored breathing
and decreased energy. One week ago, she
was seen by the primary care doctor com-
Clinical Presentation plaining of sore throat and fever. She had
white spots in her posterior oropharynx.
A retrospective study performed at 7 tertiary Testing for group A streptococcal infection
pediatric hospitals reviewed 171 patients with was negative.
myocarditis. The median age of the patients was Her vital signs include heart rate 160
13.1 years (25%, 75%: 2.1, 15.9) with a bimodal beats per minute, blood pressure
distribution: 24% were less than 2 years and 46% 70/40 mmHg, respiratory rate 50 breaths
were between 13 and 18 years [15]. Many patients per minute, and oxygen saturation as deter-
present with heart failure contemporaneous with mined by pulse oximetry 90%. Her physical
a recent viral illness. Myocarditis may present examination was notable for the following:
with a variety of symptoms from abdominal pain moderate respiratory distress with rales
to cardiogenic shock and sudden death [1]. In a throughout her lung fields, prominent S3
nationwide survey to determine the clinico-epi- gallop auscultated on cardiac exam, liver
demiological features of myocarditis in Japanese was palpable 5 centimeters below the right
children and adolescents observing initial non- costal margin, and cool and clammy
specific symptoms, fever was observed in 48%, extremities with diminished pulses. Her
nausea or vomiting was observed 30%, abdomi- face was also notably swollen.
nal pain was observed in 9.5%, diarrhea was
ERRNVPHGLFRVRUJ
5 Medical Management of Acute Fulminant Myocarditis 87
Viral replication
Receptor-mediated binding and Abnormalities in
Phase 1
Infiltration of inflammatory
(IFN, IL-1, TNF)
cells into myocardium
Release of nitric oxide
Chronic myocarditis
Yes
Phase 4
Fig. 5.1 Pathogenesis of myocarditis. Abbreviations: CHF congestive heart failure, IFN interferon, IL-1 interleukin,
TNF tumor necrosis factor. (Gupta et al. [5])
from myocardial infarction [16, 17]. Arrhythmias patients will have electrocardiogram (ECG)
are also commonly seen as a consequence of abnormalities, though typically these abnor-
myocardial infection and inflammation. Sinus malities are nonspecific and, in some cases,
tachycardia, atrioventricular conduction delays, subtle (Fig. 5.2a, b). Echocardiography is
and/or supraventricular and ventricular arrhyth- essential to diagnose and quantitate regional
mias are the most frequent types of arrhythmias and global left and right ventricular function,
in fulminant myocarditis [17]. the presence of pericardial effusion, and valvu-
lar regurgitation. Importantly, fulminant myo-
carditis usually presents with a non-dilated left
Diagnosis ventricle with severe systolic dysfunction as
well as increased wall thickness reflecting myo-
Patients with acute fulminant myocarditis can cardial edema. In contrast, acute but non-fulmi-
present with tachycardia, hypotension, S3 gal- nant myocarditis presents with dilated left
lop resulting from severe heart failure, elevated ventricle but normal wall thickness. Right ven-
jugular venous pressure, facial edema, and tricular systolic dysfunction is more common in
other signs of severe heart failure including fulminant myocarditis and can be a sign of poor
pulmonary rales and wheezing with impending prognosis [1, 17–19].
respiratory failure [1]. C-reactive protein, In a study to evaluate right ventricular, sys-
erythrocyte sedimentation rate, troponin, and tolic function in patients with active myocardi-
CK-MB values are frequently elevated. B-type tis, 23 patients with biopsy-proven myocarditis
natriuretic peptide and N-terminal pro-B-type were studied. The patients were divided into
natriuretic peptide are also commonly elevated those with normal right ventricular function
in patients with severe heart failure [1]. Most (normal right ventricular descent: the descent
ERRNVPHGLFRVRUJ
88 F. Savorgnan and P. A. Checchia
Fig. 5.2 (a) ECG of a patient with myocarditis demon- tions, most notable in leads I, aVL, and the left precordial
strating a low-voltage rhythm (requiring double standard leads (V4–V6). (b) ECG of patient with myocarditis dem-
calibration) along with diffuse, nonspecific S-T eleva- onstrating subtle nonspecific S-T changes in leads II, III,
and aVF
of the base right ventricle in systole of more likelihood of an adverse outcome (defined as
than 1.9+/−0.1 cm) and those with abnormal death or need for cardiac transplantation) was
right ventricular function (abnormal right ven- greater in patients with abnormal right ventric-
tricular descent: the descent of the base right ular function (right ventricular descent
ventricle in systole of 0.8 +/− 0.1 cm). The ≤1.7 cm) [19].
ERRNVPHGLFRVRUJ
5 Medical Management of Acute Fulminant Myocarditis 89
Cardiac magnetic resonance imaging (CMR) tion (condition for which there is evidence and/
has been increasingly utilized as part of the or general agreement that a given procedure is
diagnostic evaluation for myocarditis. CMR beneficial, useful, and effective) to differentiate
can evaluate anatomy and function of the right lymphocytic myocarditis from giant cell myo-
and left ventricles. In addition, CMR can evalu- carditis and eosinophilic myocarditis because
ate the myocardium for edema, hyperemia, and immunosuppressive therapy is indicated in the
fibrosis. CMR is more sensitive when per- latter two conditions [27, 28].
formed more than 7 days after the onset of
symptoms [20–23]. Because inflammation
associated with myocarditis is patchy in nature,
the CMR can help to guide the endomyocardial
biopsy to minimize the possibility of sampling Case Scenario: Continued
error [23]. Three CMR criteria for myocarditis The patient was admitted to the pediatric
include: intensive care unit and the following stud-
ies were obtained:
1. Enhancement on T2-weighted imaging con-
sistent with edema • Chest x-ray: diffuse bilateral haziness
2. Early enhancement on T1-weighted imaging throughout the lungs fields.
consistent with hyperemia • Laboratory studies: (normal values).
3. Late gadolinium enhancement (LGE) also
–– C-reactive protein 20 mg/L (<3.0 mg/
consistent with myocardial fibrosis dl).
–– Erythrocyte sedimentation 70 mm/h
Studies have demonstrated 85% diagnostic (0–29 mm/h).
accuracy for myocarditis by CMR when any of –– Troponin I 3.5 ng/ml (<0.01 ng/dl).
the two criteria are present [1, 24]. –– CK-MB 40 IU/L (5-25IU/L).
The gold standard for diagnosis of myocar- –– Viral PCR and bacterial culture were
ditis is endomyocardial biopsy [25]. The Dallas negative.
Criteria were created in order to have a stan- • Electrocardiogram: Subtle nonspecific
dardized pathologic definition of myocarditis. S-T changes in leads II, III, and aVF
This criteria requires evidence of cellular (Fig. 5.2b).
inflammatory process in the myocardium and • Echocardiogram: Dilated left ventricle
myocardial necrosis [25]. In the last 10 years, with severe systolic dysfunction, EF
progress in immunohistochemistry techniques 25%. The left ventricle demonstrated
has enhanced sensitivity in the detection of increased wall thickness, as well as
myocarditis [9, 14, 26]. There are disadvan- moderate mitral regurgitation. A mild
tages associated with endomyocardial biopsy pericardial effusion is also seen (Fig.
including different interpretations by different 5.3a and b).
pathologists, sampling error due to scattered • CMR: Diffuse decrease in left ventric-
focal presentation of the disease, and risk for ular function is observed. The left ven-
cardiac perforation during the procedure [25]. tricle also demonstrated regional
Currently, endomyocardial biopsy is recom- edema on T2-weighted imaging. A
mended in patients with fulminant myocarditis, mild pericardial effusion is present
severe ventricular arrhythmias, or advanced (Fig. 5.4).
heart block according to recommendations in • Biopsy: It was not performed in this
the scientific statement from the American case. The decision not to perform endo-
Heart Association/American College of myocardial biopsy was a clinical deci-
Cardiology/European Society of Cardiology sion in this particular patient.
[27]. Endomyocardial biopsy is a class I indica-
ERRNVPHGLFRVRUJ
90 F. Savorgnan and P. A. Checchia
ERRNVPHGLFRVRUJ
5 Medical Management of Acute Fulminant Myocarditis 91
and reasonable blood pressure. Caution need (ELSO) Registry from 1995 to 2006 showed
to be exercised in the use of milrinone in the myocarditis as a reason for ECMO deploy-
setting of low blood pressure due to its after- ment in 1.3% of ECMO runs. The survival to
load-reducing effect, which can worsen the discharge was 61% in this cohort of infants,
low blood pressure. children, and young adults [39]. In another
• Oral heart failure therapies: Standard oral study utilizing the ELSO database with the
heart failure therapies such as angiotensin- purpose of analyzing data outcomes of extra-
converting enzyme (ACE) inhibitors, beta- corporeal cardiopulmonary resuscitation
blockers (carvedilol), and aldosterone (eCPR) in patients with structurally normal
antagonists should be considered when hemo- heart, the overall survival to hospital discharge
dynamic stability improves but, in fact, have was 32%. From this study, 20 patients pre-
little role in the acute presentation of fulmi- sented with myocarditis, and 15 patients sur-
nant myocarditis. vived to hospital discharge [48].
• Mechanical ventilation: Patients who do not
improve with medical therapy and present Immunomodulatory treatments have also
with worsening respiratory distress may ben- become an important part of the management of
efit from the use of mechanical ventilation. myocarditis at many centers. These therapies
This strategy reduces the afterload on the left include:
ventricle and therefore improves the function
of the left ventricle [35]. • Corticosteroids: Corticosteroids decrease
• Advanced life support: Some patients with inflammation and temper increased capillary
myocarditis fail medical therapy and progress permeability often seen in acute myocarditis.
to refractory, end-stage shock with extremely In a Cochrane Database review of 8 random-
poor cardiac performance. In those unfortu- ized clinical trials (RCTs) which include 719
nate patients, the only tool available may be patients with myocarditis, there was no statis-
mechanical circulatory support (MCS) to res- tical difference in mortality in patients who
cue the patients and bridge them to recovery received corticosteroids versus placebo. This
or cardiac transplantation. Early deployment study includes adult and pediatric patients.
of circulatory support can improve end results Patients who received corticosteroids however
in patients with fulminant myocarditis. In did have better ventricular function and lower
patients with fulminant myocarditis, survival cardiac enzyme measurements in comparison
rates associated with extracorporeal mem- to placebo. Of note, the studies in this
brane oxygenation (ECMO) and ventricular Cochrane Database review were determined
assist devices (VAD) vary in different studies. to be of poor quality in general [42]. Further,
In the studies reviewed, worse survival rates because corticosteroids are commonly used in
were typically associated with studies that combination with other therapy that alter the
were done a long time ago, that were per- immune system, it is challenging to isolate
formed in facilities with less experience, and their benefit in the treatment of myocarditis
that concerned patients with concomitant, patients from other treatments given simulta-
multi-organ dysfunction [3, 6, 36–47]. In the neously [49–51].
aforementioned retrospective analysis of 514 • Immunoglobulin (IVIG): In the treatment of
patients with myocarditis utilizing the PHIS pediatric myocarditis, the use of IVIG is
database, 26% died and 13% underwent heart highly controversial. To date, much of the
transplantation. Ninety-five patients (account- evidence is based on retrospective data. No
ing for 18% of the patients studied) required randomized controlled trial (RCT) in the
ECMO, and the use of ECMO was indepen- pediatric population has yet to be conducted.
dently associated with death or transplant [3]. In one retrospective chart review of 171 adult
The Extracorporeal Life Support Organization patients who presented with myocarditis,
ERRNVPHGLFRVRUJ
92 F. Savorgnan and P. A. Checchia
ERRNVPHGLFRVRUJ
5 Medical Management of Acute Fulminant Myocarditis 93
diagnosed with myocarditis clinically with or these patients, EMB and immunomodulatory
without biopsy confirmation (n = 372) to children therapies should be considered only if clinical
who were diagnosed with idiopathic dilated car- improvement does not occur within a reasonable
diomyopathy (n = 1123). The study found better period of time (i.e., 12–2 4 weeks) or the patient’s
outcomes (death, transplantation, and echocar- condition worsens. In these cases, however,
diographic normalization 3 years after presenta- immunomodulatory therapies should be utilized
tion) in patients with myocarditis in comparison only if persistent inflammation is noted without
to patients with idiopathic dilated cardiomyopa- active viral infection.
thy (P = 0.003) [58]. Another report, a recent ret-
rospective chart review from 7 tertiary care
pediatric hospitals that included 171 patients Key Points
with myocarditis, found that patients presenting • Myocarditis is a disease characterized
with worsening left ventricular function at admis- by inflammation in the myocardium that
sion (ejection fraction less than 30% or shorten- results in cardiac-histologic and ana-
ing fraction less than 14%) had higher B-type tomical alterations accompanied by
natriuretic peptide but had lower troponins at functional derangements secondary to
admission. The study also showed that patients myocardial destruction.
with GI symptoms (42%) and patients with mod- • Parvovirus B19 and human herpesvirus
erate to severe ventricular dysfunction with less 6 have been identified as the most fre-
than 21% and lower shortening fraction on echo- quent causes of viral myocarditis.
cardiogram (40%) were more likely to die or • Myocarditis may present with a variety
require transplantation. Specifically, patients of symptoms from abdominal pain to
with GI symptoms and/or lower shortening frac- cardiogenic shock and sudden death.
tion were at increased risk for death (9.5% vs. • Echocardiography is essential to diag-
0%) or transplantation (16% vs. 3.3%) [15]. nose and quantitate regional and global
left and right ventricular function,
though cardiac magnetic resonance
Summary imaging (CMR) is increasingly utilized
as part of the diagnostic evaluation for
A reasonable approach to diagnosis and manage- myocarditis.
ment of myocarditis has been recently published • The gold standard for diagnosis of myo-
by Kindermann and colleagues, which is provided carditis is endomyocardial biopsy; CMR
in Fig. 5.5 [59]. The diagnosis and management can inform an endomyocardial biopsy in
are dictated in large part by the clinical presenta- order to avoid sampling error.
tion. Patients who present with acute fulminant • We recommend use of corticosteroids and
myocarditis requiring intensive care therapies and IVIg in patients who present with acute
mechanical support should go under EMB. This fulminant myocarditis, especially those
step is important, as patients with fulminant car- patients who require mechanical support
diovascular collapse without myocarditis are due to failure of medical therapy.
unlikely to recover and, thus, may need to be con- • Patients with less severe myocarditis
sidered for heart transplantation sooner rather with minimal improvement with medi-
than later. Though not included in the algorithm, cal therapy over a reasonable period of
we also recommend corticosteroids and IVIg as time warrant EMB, and those found to
immunomodulatory therapies as discussed in the have evidence of persistent inflamma-
clinical case scenario for patients with acute ful- tory cells in their myocardium and an
minant myocarditis. For patients with less severe absence of viral organisms by immuno-
acute myocarditis, biomarkers (e.g., B-type natri- histochemistry may benefit from the use
uretic peptide, troponins), echocardiography, and of immunotherapy.
CMR should be used to make the diagnosis. For
ERRNVPHGLFRVRUJ
Patient with suspected myocarditis
94
No myocarditis Myocarditis
Recovery within
Control: ECG, Echo, Control: ECG, Echo, 2-4 weeks
Biomarker, cMRI Biomarker, cMRI
ERRNVPHGLFRVRUJ
Myocarditis No myocarditis
Heart Failure Treatment
Immunosuppressive or
Not evidence based, Antiviral
Immunomodulatory treatment only Treatment
Treatment according to study
protocols
Proposed diagnostic and therapeutic algorithm for patients with suspected acute myocarditis considering biomarkers, cardiac magnetic resonance Imaging (cMRI), and
endomyocardial biopsy (EMB). BI-VAD = biventricular assist device; Circ. = circulatory; ECMO = extracorporeal membrane oxygenation; LV = left ventricular; LVAD = left
ventricular assist device.
Fig. 5.5 Proposed diagnostic and therapeutic algorithm for suspected BI-VAD biventricular assist device, Circ circulatory, ECMO extracorpo-
myocarditis. Proposed diagnostic and therapeutic algorithm for patients real membrane oxygenation, LV left ventricular, LVAD left ventricular
F. Savorgnan and P. A. Checchia
with suspected acute myocarditis considering blomarkers, cardiac mag- assist device. (Kindermann et al. [59]. Acknowledgment: Jack Price, MD
netic resonance imaging (cMRI), and endomyocardial blopsy (EMB). and Kristina Jovanovic, RN)
5 Medical Management of Acute Fulminant Myocarditis 95
References 19. Mendes LA, Dec GW, Picard MH, Palacios IF, Newell
J, Davidoff R. Right ventricular dysfunction: an inde-
pendent predictor of adverse outcome in patients with
1. Ginsberg F, Parrillo JE. Fulminant myocarditis.
myocarditis. Am Heart J. 1994;128:301–7.
Crit Care Clin. 2013;29:465–83.
20. Childs H, Friedrich MG. Cardiovascular magnetic
2. Klugman D, Berger JT, Sable CA, He J, Khandelwal
resonance imaging in myocarditis. Prog Cardiovasc
SG, Slonim AD. Pediatric patients hospitalized with
Dis. 2011;54:266–75.
myocarditis: a multi-institutional analysis. Pediatr
21. Ellis CR, Di Salvo T. Myocarditis: basic and clinical
Cardiol. 2010;31:222–8.
aspects. Cardiol Rev. 2007;15:170–7.
3. Ghelani SJ, Spaeder MC, Pastor W, Spurney CF,
22. Nelson KH, Li T, Afonso L. Diagnostic approach and
Klugman D. Demographics, trends, and outcomes
role of MRI in the assessment of acute myocarditis.
in pediatric acute myocarditis in the United States,
Cardiol Rev. 2009;17:24–30.
2006 to 2011. Circ Cardiovasc Qual Outcomes.
23.
Friedrich MG, Sechtem U, Schulz-Menger J,
2012;5:622–7.
et al. Cardiovascular magnetic resonance in myo-
4. Saji T, Matsuura H, Hasegawa K, et al. Comparison
carditis: a JACC white paper. J Am Coll Cardiol.
of the clinical presentation, treatment, and outcome
2009;53:1475–87.
of fulminant and acute myocarditis in children. Circ J.
24. Abdel-Aty H, Boye P, Zagrosek A et al. Diagnostic
2012;76:1222–8.
performance of cardiovascular magnetic
5. Gupta S, Markham DW, Drazner MH, Mammen
resonance in patients with suspected acute myo-
PP. Fulminant myocarditis. Nat Clin Pract Cardiovasc
carditis: comparison of different approaches. J
Med. 2008;5:693–706.
Am Coll Cardiol. 2005;45:1815–22.
6. Teele SA, Allan CK, Laussen PC, Newburger JW,
25. Ilback NG, Fohlman J, Friman G. Exercise in cox-
Gauvreau K, Thiagarajan RR. Management and out-
sackie B3 myocarditis: effects on heart lymphocyte
comes in pediatric patients presenting with acute ful-
subpopulations and the inflammatory reaction. Am
minant myocarditis. J Pediatr. 2011;158:638–643 e1.
Heart J. 1989;117:1298–302.
7. Feldman AM, McNamara D. Myocarditis. N Engl J
26. Parrillo JE. Inflammatory cardiomyopathy (myo-
Med. 2000;343:1388–98.
carditis): which patients should be treated with anti-
8. Grun S, Schumm J, Greulich S, et al. Long-term
inflammatory therapy? Circulation. 2001;104:4–6.
follow-up of biopsy-proven viral myocarditis: predic-
27. Cooper LT, Baughman KL, Feldman AM, et al. The
tors of mortality and incomplete recovery. J Am Coll
role of endomyocardial biopsy in the management of
Cardiol. 2012;59:1604–15.
cardiovascular disease: a scientific statement from the
9. Kindermann I, Kindermann M, Kandolf R, et al.
American Heart Association, the American College of
Predictors of outcome in patients with suspected myo-
Cardiology, and the European Society of Cardiology.
carditis. Circulation. 2008;118:639–48.
Circulation. 2007;116:2216–33.
10. Molina KM, Garcia X, Denfield SW, et al. Parvovirus
28. Wu LA, Lapeyre AC 3rd, Cooper LT. Current role of
B19 myocarditis causes significant morbidity and
endomyocardial biopsy in the management of dilated
mortality in children. Pediatr Cardiol. 2013;34:390–7.
cardiomyopathy and myocarditis. Mayo Clin Proc.
11. Pollack A, Kontorovich AR, Fuster V, Dec GW. Viral
2001;76:1030–8.
myocarditis – diagnosis, treatment options, and cur-
29. Kantor PF, Lougheed J, Dancea A, et al. Presentation,
rent controversies. Nat Rev Cardiol. 2015;12:670–80.
diagnosis, and medical management of heart failure in
12.
Cooper LT Jr. Myocarditis. N Engl J Med.
children: Canadian cardiovascular society guidelines.
2009;360:1526–38.
Can J Cardiol. 2013;29:1535–52.
13. Liu PP, Mason JW. Advances in the understanding of
30. Kirk R, Dipchand AI, Rosenthal DN, et al. The
myocarditis. Circulation. 2001;104:1076–82.
International Society for Heart and Lung Transplantation
14.
Magnani JW, Dec GW. Myocarditis: current
guidelines for the management of pediatric heart fail-
trends in diagnosis and treatment. Circulation.
ure: executive summary. [Corrected]. J Heart Lung
2006;113:876–90.
Transplant. 2014;33:888–909.
15. Butts RJ, Boyle GJ, Deshpande SR et al.
31. Elkayam U, Bitar F, Akhter MW, Khan S, Patrus S,
Characteristics of clinically diagnosed pediat-
Derakhshani M. Intravenous nitroglycerin in the treat-
ric myocarditis in a contemporary multi-center
ment of decompensated heart failure: potential ben-
cohort. Pediatr Cardiol. 2017;38:1175–82.
efits and limitations. J Cardiovasc Pharmacol Ther.
16. Hufnagel G, Pankuweit S, Richter A, Schonian
2004;9:227–41.
U, Maisch B. The European study of epidemiol-
32. O’Connor CM, Starling RC, Hernandez AF, et al.
ogy and treatment of cardiac inflammatory diseases
Effect of nesiritide in patients with acute decompen-
(ESETCID). First epidemiological results. Herz.
sated heart failure. N Engl J Med. 2011;365:32–43.
2000;25:279–85.
33. Jefferies JL, Denfield SW, Price JF, et al. A prospec-
17. Elamm C, Fairweather D, Cooper LT. Pathogenesis
tive evaluation of nesiritide in the treatment of pediat-
and diagnosis of myocarditis. Heart. 2012;98:835–40.
ric heart failure. Pediatr Cardiol. 2006;27:402–7.
18. Schultz JC, Hilliard AA, Cooper LT Jr, Rihal
34. O’Connor CM, Gattis WA, Uretsky BF, et al.
CS. Diagnosis and treatment of viral myocarditis.
Continuous intravenous dobutamine is associated with
Mayo Clin Proc. 2009;84:1001–9.
an increased risk of death in patients with advanced
ERRNVPHGLFRVRUJ
96 F. Savorgnan and P. A. Checchia
heart failure: insights from the Flolan International 47. Madden K, Thiagarajan RR, Rycus PT, Rajagopal
Randomized Survival Trial (FIRST). Am Heart J. SK. Survival of neonates with enteroviral myocardi-
1999;138:78–86. tis requiring extracorporeal membrane oxygenation.
35. Bronicki RA. Cardiopulmonary interactions in Pediatr Crit Care Med. 2011;12:314–8.
children with heart failure. Curr Cardiol Rev. 48. Conrad SJ, Bridges BC, Kalra Y, Pietsch JB, Smith
2016;12:104–6. AH. Extracorporeal cardiopulmonary resuscita-
36. Reiss N, El-Banayosy A, Arusoglu L, Blanz U,
tion among patients with structurally normal hearts.
Bairaktaris A, Koerfer R. Acute fulminant myocardi- ASAIO J. 2017;63:781–6.
tis in children and adolescents: the role of mechanical 49. Mason JW, O’Connell JB, Herskowitz A, et al. A
circulatory assist. ASAIO J. 2006;52:211–4. clinical trial of immunosuppressive therapy for myo-
37. Wu ET, Huang SC, Chen YS, Wang JK, Wu MH, Ko carditis. The myocarditis treatment trial investigators.
WJ. Children with fulminant myocarditis rescued N Engl J Med. 1995;333:269–75.
with extracorporeal membrane oxygenation. Heart. 50. Jones SR, Herskowitz A, Hutchins GM, Baughman
2006;92:1325–6. KL. Effects of immunosuppressive therapy in
38. Nahum E, Dagan O, Lev A, et al. Favorable outcome biopsy-proved myocarditis and borderline myo-
of pediatric fulminant myocarditis supported by extra- carditis on left ventricular function. Am J Cardiol.
corporeal membranous oxygenation. Pediatr Cardiol. 1991;68:370–6.
2010;31:1059–63. 51. Chen HS, Wang W, Wu SN, Liu JP. Corticosteroids
39. Rajagopal SK, Almond CS, Laussen PC, Rycus PT, for viral myocarditis. Cochrane Database Syst Rev.
Wypij D, Thiagarajan RR. Extracorporeal membrane 2013;18:CD004471.
oxygenation for the support of infants, children, and 52. McNamara DM, Holubkov R, Starling RC, et al.
young adults with acute myocarditis: a review of the Controlled trial of intravenous immune globulin in
extracorporeal life support organization registry. Crit recent-onset dilated cardiomyopathy. Circulation.
Care Med. 2010;38:382–7. 2001;103:2254–9.
40. Wilmot I, Morales DL, Price JF, et al. Effectiveness of 53. Drucker NA, Colan SD, Lewis AB, et al. Gamma-
mechanical circulatory support in children with acute globulin treatment of acute myocarditis in the pediat-
fulminant and persistent myocarditis. J Card Fail. ric population. Circulation. 1994;89:252–7.
2011;17:487–94. 54. Prasad AN, Chaudhary S. Intravenous immuno-
41. Mody KP, Takayama H, Landes E, et al. Acute
globulin in children with acute myocarditis and/
mechanical circulatory support for fulminant myocar- or early dilated cardiomyopathy. Indian Pediatr.
ditis complicated by cardiogenic shock. J Cardiovasc 2014;51:583–4.
Transl Res. 2014;7:156–64. 55. Canter CE, Simpson KE. Diagnosis and treatment of
42. Chen YS, Wang MJ, Chou NK, et al. Rescue for acute myocarditis in children in the current era. Circulation.
myocarditis with shock by extracorporeal membrane 2014;129:115–28.
oxygenation. Ann Thorac Surg. 1999;68:2220–4. 56. Schultheiss HP, Kuhl U, Cooper LT. The management
43. Duncan BW, Bohn DJ, Atz AM, French JW, Laussen of myocarditis. Eur Heart J. 2011;32:2616–25.
PC, Wessel DL. Mechanical circulatory support for 57. Ammirati E, Cipriani M, Lilliu M, et al. Survival
the treatment of children with acute fulminant myo- and left ventricular function changes in fulminant
carditis. J Thorac Cardiovasc Surg. 2001;122:440–8. versus nonfulminant acute myocarditis. Circulation.
44. Lee KJ, McCrindle BW, Bohn DJ, et al. Clinical
2017;136:529–45.
outcomes of acute myocarditis in childhood. Heart. 58. Foerster SR, Canter CE, Cinar A, et al. Ventricular
1999;82:226–33. remodeling and survival are more favorable for
45.
Kuhn B, Shapiro ED, Walls TA, Friedman myocarditis than for idiopathic dilated cardiomy-
AH. Predictors of outcome of myocarditis. Pediatr opathy in childhood: an outcomes study from the
Cardiol. 2004;25:379–84. Pediatric Cardiomyopathy Registry. Circ Heart Fail.
46. Topkara VK, Dang NC, Barili F, et al. Ventricular 2010;3:689–97.
assist device use for the treatment of acute viral myo- 59. Kindermann I, Barth C, Mahfoud F, et al. Update on
carditis. J Thorac Cardiovasc Surg. 2006;131:1190–1. myocarditis. J Am Coll Cardiol. 2012;59:779–92.
ERRNVPHGLFRVRUJ
Pediatric Cardiac Transplantation
and Mechanical Assist Devices 6
Juan M. Lehoux, Kimberly D. Beddows,
and Jacqueline M. Lamour
ERRNVPHGLFRVRUJ
98 J. M. Lehoux et al.
50
Fig. 6.1 Kaplan-Meier survival of pediatric heart trans- than exact rates because the time of death is not known for
plants performed between Jan. 1982 and June 2015. Since all patients. The median survival is the estimated time
many patients are still alive and some patients have been point at which 50% of all of the recipients have died
lost to follow-up, the survival rates are estimates rather
16 days versus 7 days, respectively [5]. The eras, but a significant proportion of these patients
differences in length of stay are likely related, at will develop end-stage heart failure that require
least in part, to the disparity in out-of-hospital heart transplantation [7]. As is the case in all solid
options for young children as compared to adults. organ transplantation, the demand for organs
For instance, there are limited mechanical assist exceeds the supply. So, although transplant
device options for children that would allow for remains the best treatment option for children
discharged home. Although mortality rates for with end-stage heart failure, waitlist mortality
both pediatrics and adults with cardiomyopathy remains an issue. Mechanical circulatory support
have decreased over time, overall mortality is provides a temporary solution to the shortage of
worse in children. Infants have the highest mor- donor hearts [8]. In the adult population, there
tality rate of any age group, including patients has been significant investment by industry in the
greater than 70 years of age [5]. research and development of ventricular assist
The prevalence of children born with congeni- devices. These devices have revolutionized the
tal heart disease worldwide is approximately 1% treatment of advanced heart failure in adults [9].
[6]. Advances in surgical technique in infants The field of mechanical circulatory support in
with congenital heart disease have palliated children has lagged behind but in the recent era
patients that would have otherwise died in earlier has made great strides.
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 99
ERRNVPHGLFRVRUJ
100 J. M. Lehoux et al.
Fig. 6.2 Mechanical assist devices. (a) Berlin Heart (c) HeartWare HVAD console and pump. (d) Thoratec
EXCOR biventricular support and console with varying HeartMate II LVAD. (e) HeartMate 3 LVAD
sizes of pumps. (b) PediMag device and console.
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 101
Fig. 6.2 (continued)
ERRNVPHGLFRVRUJ
102 J. M. Lehoux et al.
used off label for much longer periods as part of in over 20,000 patients. In the adult population,
the ECMO circuit or as paracorporeal ventricular the HeartMate II has been shown to improve both
assist devices. Some centers, in an effort to survival and quality of life with an improved side
reduce the cost associated with frequent pump effect profile, when compared to older pulsatile
exchanges often required with the Berlin Heart designs [7]. The device is implanted via a median
EXCOR, use these continuous-flow devices as an sternotomy and placed in a surgically created
alternative. Use of the CentriMag and PediMag preperitoneal pocket. A single drive line exits the
devices connected via Berlin heart cannulas has abdomen and connects the device controller and
been described in small case series [12–14]. batteries. Due to its size, this device can only be
used in adult-sized patients (BSA greater than 1).
Despite the decreased incidence of adverse events
HeartWare HVAD compared to older, pulsatile devices, there are
still significant issues with morbidity associated
The HeartWare HVAD is a fully implantable, with this device including stroke, gastrointestinal
continuous centrifugal flow device (Fig. 6.2c). bleeding, intractable drive line infections and
The pump’s only moving part is a magnetically pump thrombosis [17].
stabilized rotor. A single drive line exits the body
and connects to an external controller device and
batteries. There are no mechanical bearings mak- Thoratec HeartMate 3
ing it highly durable. The device has been exten-
sively used as bridge to transplant in adults and The HeartMate 3 LVAS (left ventricular assist
has recently been approved for destination ther- system) is the most recent ventricular assist
apy [15]. Given the better side effect profile asso- device approved for use as bridge to transplant
ciated with continuous-flow devices versus older (Fig. 6.2e). The device, which features a com-
pulsatile flow pumps, the HVAD is preferred to pletely magnetically levitated rotor that provides
the Berlin Heart EXCOR in larger children. wide spaces for blood flow, is designed for
Though designed to be used in adult patients, the improved hemocompatibility and reduced pump-
use of the HeartWare HVAD is possible in chil- related morbidity. The HeartMate 3 has proven to
dren with BSA greater than 0.6 m2 with modifica- be highly resistant to pump thrombosis. Its design
tions to the implant technique (i.e., preperitoneal and reduced size makes it easier to implant [18,
pocket vs. intra-pericardial) [16]. Patients and 19]. Given its proven benefits and recent approval
their guardians can be trained in the day-to-day as bridge to transplant, the HeartMate 3 is our
management of the HVAD. Children supported device of choice in the adolescent population.
with this device have been able to go to school The HeartMate 3 is larger than the HeartWare
and lead a relatively normal life while waiting for HVAD, which may make implant in smaller chil-
heart transplantation. dren more challenging.
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 103
the adult population, given the availability of function [21]. Banding the pulmonary artery will
newer-generation devices with a more favorable increase the right ventricular pressure and shift
side effect profile, the decision to proceed with the interventricular septum leftward. This recon-
VAD implant is simplified. In the current era, figuration of the septum can reduce mitral regur-
continued medical management in the setting of gitation by reducing mitral valve annulus
worsening functional status and end organ dys- diameter and reportedly improve cardiac output.
function is no longer indicated and, furthermore, Pulmonary artery banding in very small children
is associated with worse outcomes even if implant with dilated cardiomyopathy may be an accept-
of a VAD is eventually undertaken. Dependence able alternative to VAD allowing the child to
on intravenous inotropic support is the usual indi- grow to sufficient size for a safer VAD implanta-
cation that prompts VAD implantation. There are, tion or to be transplanted [21].
of course, some exceptions to this rule, such as A comprehensive evaluation to rule out revers-
favorable blood type with short transplant wait ible causes of heart failure accompanied by mul-
times in the bridge-to-transplant patient. tidisciplinary management discussions should
In pediatrics, the decision to proceed with take place for every child admitted in heart fail-
VAD implantation is complicated by several fac- ure. If deterioration progresses despite inotropic
tors including patient size, device availability, support, mechanical ventilation is the next step in
blood type, expected transplant wait time, etiol- escalation of care. Intubation should be done in a
ogy of heart failure, and overall condition of the controlled setting with surgical consultation
patient. There is significant variability in practice immediately available should ECPR need to be
across the world that considers the abovemen- deployed. It is not uncommon for a child whose
tioned factors with no approach being supported status is deteriorating to arrest while attempting
by evidence. Decisions on the use of mechanical intubation. In this scenario, ECPR with prompt
support in the pediatric patient with heart failure restoration of cardiac output can be lifesaving
must therefore be based on physician experience [22]. Mechanical ventilatory support decreases
and sound physiologic rationale. cardiac preload and afterload in the failing heart
and also reduces the effort made by a child with
limited cardiopulmonary reserve [23]. Ventilator
Patient Size dependence should trigger VAD implantation,
with the goal of liberation from mechanical ven-
Small Children (BSA < 0.6 m2) tilation. Long-term mechanical ventilation and
its required sedation cause progressive decon-
There are limited mechanical support options for ditioning, which can affect posttransplant out-
small children in heart failure [20]. ECMO is comes. Being on mechanical ventilation is a
commonly used in conjunction with CPR (ECPR) known risk factor for poor outcome after heart
or when short duration of support is anticipated. transplantation [24, 25]. Successfully implanting
Long-term mechanical support options currently a durable VAD that restores adequate cardiac out-
available are the Berlin Heart EXCOR and the put and allows the patient to be mobilized, reha-
PediMag LVAD connected to Berlin Heart can- bilitate, and gain weight is worthwhile despite
nulas. Due to the limited options and high mor- the risks of surgery.
bidity associated with mechanical support in Right heart catheterization can be helpful in
small children, we seek to minimize the child’s assessing the right heart function prior to implan-
exposure to a device as long as it is reasonable to tation but should be weighed against the risk.
do so. In children who are less than 5 kg, every Echocardiographic evaluation of the right heart
attempt is made to delay VAD implantation. can often lead to concerns of post-VAD implant
Pulmonary artery banding has been reported as a right heart failure. Because of high left-sided fill-
temporizing measure for patients with dilated ing pressures and high pulmonary artery pres-
cardiomyopathy with preserved right ventricular sures, the right heart can appear to be severely
ERRNVPHGLFRVRUJ
104 J. M. Lehoux et al.
dysfunctional. Once the LVAD is implanted and • Avoid BiVAD implant if possible.
the filling pressures of the left ventricle improve, • Do not delay in RVAD implant if evidence of
often what appeared to be a failing right ventricle right heart dysfunction develops post LVAD
can provide adequate cardiac output to fill the implant.
left-sided device. There is evidence that the use
of biventricular VADs (BiVADs) is associated
with worse outcomes [26]. Avoidance of biven- Larger Children (BSA > 0.6 m2)
tricular support at all costs however is also ill
advised. A child struggling in low cardiac output Children whose body surface area is >0.6 m2
due to RV failure after VAD implantation can become candidates for the HeartWare
develop worsening end organ dysfunction. With HVAD. This device, as described above, is
the Berlin EXCOR, evidence of LVAD under fill- designed for use in adults and approved for long-
ing, low cardiac output, and high right-sided fill- term support. In adults, the HVAD has a signifi-
ing pressures should prompt RVAD implant as cantly better side effect profile than older
soon as possible. Under filling of the left-sided paracorporeal pulsatile devices that are designed
device also causes wrinkles to form on the pump similar to the Berlin Heart EXCOR [11]. In pedi-
diaphragm, providing a nidus for clot formation atrics, there has been great interest in using
even if anticoagulation is adequate. Clot forma- continuous-flow devices in the hope of replicat-
tion increases the risk of embolus and its associ- ing the results seen in adults. There is currently
ated neurologic and vascular complications. little evidence to support using implantable
When right heart function is marginal, under continuous-flow devices rather than the Berlin
filling of the Berlin Heart EXCOR can occur, Heart EXCOR, but given the reduced incidence
increasing the risk of pump thrombosis and its of adverse events reported in the adult literature,
associated morbidities including stroke. For this many centers favor VAD implant earlier in the
reason, some centers will implant the Berlin disease course [27]. It is our practice to consider
Heart EXCOR cannulas in the usual fashion and implant of the HeartWare HVAD in larger chil-
connect a PediMag continuous-flow pump dren as soon as the child becomes dependent on
instead of the pulsatile Berlin Heart EXCOR one or more positive inotropic drugs (e.g., milri-
when RV dysfunction is present. This approach none, dobutamine, etc.). Regional wait times,
allows the patient to recover from the initial post- blood type, and overall condition of the patient
operative right ventricular dysfunction without will factor into the decision to implant the device
the associated risk of an under-filled Berlin Heart or to continue to wait for transplant on inotropic
EXCOR device. Once the child is extubated, the infusions. Restoration of adequate cardiac output
marginal right ventricular function usually before the onset of end organ dysfunction has
improves, allowing for the patient to be transi- been shown to improve VAD outcomes in adults.
tioned to a Berlin Heart EXCOR device for long- The HeartWare HVAD is connected via a single
term support. It is important to note that this drive line to a small controller and batteries. This
approach requires close monitoring for progres- design makes it possible for patients to resume
sive right ventricular dysfunction. Marginal many normal activities that improve the physical
LVAD flows with evidence of end organ dysfunc- and psychological condition of the child. Implant
tion in the setting of right ventricular dysfunction of the HVAD is not free of the complications that
should prompt RVAD implant before further affect all newer-generation continuous-flow
clinical deterioration ensues. devices such as drive line infections, gastrointes-
Key points in small children (BSA < 0.6 m2): tinal bleeding, and stroke. Despite these possible
complications, it is thought that the benefits of
• VAD team evaluation once inotropic support earlier VAD support outweigh these concerns. It
is started. is not inconceivable that as VAD technology
• VAD implant if ventilator dependent. improves and adverse effects decrease, VAD
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 105
implant will become an option for patients who conditions, central ECMO cannulation for post-
are in significant heart failure but not yet inotrope cardiotomy shock avoids cannulating the ventri-
dependent. cle as is frequently necessary for VAD implant. If
Adult-sized adolescents can be implanted a reasonable period of time (1–2 weeks) has
with any VAD currently on the market. The passed with little evidence of recovery, then alter-
HeartMate 3 device was recently approved as a nate, longer-term support options should be dis-
bridge-to-transplant device. The HeartMate 3 cussed. Conversion to a long-term VAD while
device has been designed for improved hemo- awaiting transplant will depend on the antici-
compatibility in an effort to reduce adverse pated wait time on the heart transplant list.
events. The device has been widely used in Centers must take regional and patient-specific
Europe and has been implanted may times in the factors into account when deciding when to aban-
United States as part of the Momentum 3 trial don the short-term support strategy in favor of a
[18]. In both the European and US experience longer-term device.
with this device, there has been a dramatic reduc-
tion in pump thrombosis and need for pump
exchange. Pump thrombus has been a significant Bridge to Transplant
source of morbidity and mortality in patients sup-
ported on VADs. The resistance to thrombosis In patients whose heart failure etiology is unlikely
demonstrated by the HeartMate 3 LVAD opens to recover, VAD implantation is done as a bridge
up exciting possibilities for future changes in the to transplantation. The benefits of VAD support
anticoagulation management that will hopefully while awaiting transplant are significant in both
decrease the rate of bleeding complications. FDA the adult and pediatric population [28]. A com-
approval of the HeartMate 3 makes it our device mon scenario is a child with heart failure who
of choice in adult-sized adolescents over the acutely deteriorates and requires emergent
older HeartMate II. ECMO cannulation or ECPR. Once the child is
Key points in larger children (BSA > 0.6 m2): hemodynamically stable on ECMO and recovery
of end organ dysfunction has been proven, the
• Implant when patient is dependent on inotro- decision between waiting for heart transplanta-
pic infusions. tion on ECMO and transitioning the patient to a
• Use centrifugal continuous-flow devices. more durable VAD must be made. The decision
• HeartMate 3 preferred device when the child’s will depend on several factors. Blood type can
BSA is greater than 1 (i.e., adult size) due to significantly affect wait times. If the child is a
its relative resistance to pump thrombosis. candidate for ABO-incompatible heart transplant
• May discharge home on VAD support with or if the blood type is AB, which are associated
adequate patient and caregiver education. with the shorter wait times in some regions, it
may be reasonable to avoid the insult of VAD
implant. Wait times also vary widely by geogra-
Anticipated Duration of Support phy, and because listing across blood groups in
infants is an accepted practice, listing across
Bridge to Recovery blood groups does not necessarily shorten wait-
list times. It is important to be familiar with the
Heart failure due to a potentially reversible etiol- local organ procurement organization (OPO) to
ogy, like viral myocarditis or arrhythmia-induced assist with decision-making and estimate wait
cardiomyopathy, is often treated with mechanical times. If the anticipated wait time is greater than
support once medical management becomes several weeks, it is reasonable to transition the
untenable. ECMO support provides adequate patient to a durable VAD.
short-term support and avoids more invasive Once the VAD is implanted, the physiological
options. In patients with surgically correctable impact of the procedure must be evaluated to
ERRNVPHGLFRVRUJ
106 J. M. Lehoux et al.
decide when to activate the patient on the trans- patients in various stages of palliation. Studies
plant list. Some centers will inactivate recently have shown dismal outcomes when single-
implanted patients for several weeks to wait for ventricle patients with shunt physiology undergo
recovery. We believe that the decision to make a VAD therapy, with slightly better results in
recently implanted patient active on the transplant patients that have undergone second and third
list must be made on a case-by-case basis. In a stage of the single-ventricle palliation [31]. Given
small child recently implanted with a Berlin the available evidence, we would not offer VAD
Heart, who is doing well several days post implant therapy to a single ventricle before the last stage
with no evidence of end organ dysfunction, sig- of palliation. In these cases, we would support
nificant inflammation, or fluid retention, it is rea- the patient with ECMO as bridge to transplant. In
sonable to proceed with transplant if an adequate patients with failure of the Fontan circulation, if
heart becomes available. The risk of continued VAD therapy is being considered, it is critical to
exposure to the VAD should be weighed against determine the mechanism of failure. Cardiac
the risk of performing a heart transplant on a catheterization should be performed to document
debilitated patient who has just undergone a major the ventricular filling pressure and confirm anat-
operation. In older patients who have been omy. If the patient has failed Fontan physiology
implanted with a continuous-flow device, it is rea- with normal ventricular filling pressure, a VAD
sonable to wait a longer period before reactivation implant is unlikely to improve outcomes and the
on the transplant list. The lower risk of adverse patient should be transplanted. If there is high
events with newer continuous-flow devices shifts ventricular filling pressure, then a VAD may
the risk/benefit analysis toward waiting for the improve the patient’s symptoms [32].
patient to recover and transition from a catabolic There are other risk factors and comorbidities
to an anabolic state. that have to be considered when considering
VAD placement in a Fontan patient as a bridge to
transplant. Multiple sternotomies cause signifi-
Destination Therapy cant scar formation that can make the operation
technically challenging. Patients with failing
In patients who are not candidates for transplan- Fontan physiology are also commonly debilitated
tation but are suffering from heart failure, there by protein-losing enteropathy and have limited
are several devices that are FDA approved for immunologic and hepatic reserve to tolerate the
long-term support. In the pediatric population, insult of a major operation. Due to these poten-
destination therapy is not a common indication tially complicating factors, VAD therapy has not
for implant. There are several reports of implants become commonplace as a bridge to transplant in
in patients with progressive degenerative condi- this population, even if there is objective evi-
tions that disqualify them for heart transplant [29, dence of possible benefit. Multidisciplinary eval-
30]. These cases have so far been the exception uation that includes cardiology, cardiac surgery,
rather than the rule. We expect that as device hepatology, and anesthesia should be completed
technology improves, destination therapy may before any surgical procedure is undertaken.
become a viable alternative to heart transplanta-
tion in pediatric patients.
Anticoagulation After VAD Implant
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 107
Active XII
VII
XI
Ca2+
+ Active XI Tissue factor +
(III) and
IX active VII
Ca2+ positive feedback
Active IX
VIII
Ca2+
X
phospholipids (PL)
Active X
COMMON PATHWAY
Prothrombin
Ca2+,
positive feedback V, PL
Thrombin
Fibrinogen
Cross-linked fibrin
postoperative bleeding resolves. Heparin binds II, VII, IX, and X in addition to regulatory fac-
to the enzyme inhibitor antithrombin III which tors, proteins C and S (Fig. 6.3).
then inactivates thrombin and factor Xa. Aspirin Smaller children supported with the EXCOR
is started 48–72 hours after the patient returns device are especially prone to embolic complica-
from the operating room. Aspirin irreversibly tions primarily because the pumps must be run at
blocks the formation of thromboxane A2 in lower rates. Lower flow through the device makes
platelets preventing platelet aggregation for the thrombus formation more likely. In this high-risk
life of the affected platelet. Warfarin is then population, management of anticoagulation post-
started in preparation for discharge once the operatively is especially important. In young
patient is tolerating a regular diet. The interna- patients, anticoagulation is challenging for a vari-
tional normalized ratio (INR) goal is 2–3. ety of reasons. Hemostasis is a complex process
Warfarin inhibits the synthesis of clotting factors involving many proteins, and the level of proteins
ERRNVPHGLFRVRUJ
108 J. M. Lehoux et al.
involved in hemostasis changes significantly with Table 6.2 Edmonton anticoagulation protocol for Berlin
age. An example of this variation is the enzyme Heart EXCOR
inhibitor antithrombin III (AT3). In children, nor- Initiation
mal AT3 levels are less than 50% of adult levels. parameters Goal anticoagulation
24–48 hrs Plt > 20,000, TEG Start UFH, goal
This relative AT3 deficiency can pose a chal-
MA > 46 anti-Xa 0.35–0.5
lenge, given that AT3 is the pharmacologic target 2–4 days No bleeding, Transition Lo LMWH
of heparin, the most commonly used anticoagu- normal renal eventual anti-Xa
lant for VAD patients both intra- and postopera- function 0.6–1
tively. Because of these developmental variations >1 week >12 months old, no Warfarin with goal
in hemostasis, clotting and bleeding can be bleeding, tolerating INR 2.7–3.5 bridge
enteral feeding with LMWH if
unpredictable. INR < 2.7
Once anticoagulation is started, monitoring Plt platelets, MA maximum amplitude, TEG thromboelas-
practices vary widely. Many tests are often tography, UFH unfractionated heparin, LMHW low
ordered with sometimes contradictory results. molecular weight heparin, INR international normalized
Adding to the difficulty, it is unclear what value ratio
of a given test indicates adequate anticoagula-
tion therapy in the pediatric population [34]. important downsides when used in the pediatric
Early in the Berlin Heart IDE EXCOR trial, a patient. The reason heparin is called unfraction-
protocol was put in place to standardize the ated is because it has molecules of varying sizes
management of the anticoagulation across all in a single vial. Because of the variable molecu-
patients in the trial (Tables 6.1 and 6.2). Referred lar size, there is variable activity of the molecule
to as the Edmonton protocol, it is still the stan- against thrombin and factor Xa. The amount of
dard for the management of anticoagulation for the 18-saccharide unit that is active and binds to
patients on the EXCOR VAD [33]. Despite the AT3 is variable. Heparin also has a propensity to
use of a standardized anticoagulation protocol, adhere to positively charged plasma proteins
the Berlin Heart IDE trial had very high rates of that can alter the bioavailability of the drug. All
stroke, pump thrombosis, and bleeding. These these factors make for a nonlinear response to
complications are related to multiple patient and heparin dosing. Chronic exposure to heparin
device factors including size and design of the also causes osteopenia in already debilitated
pump. Surgeons depend on heparin anticoagula- children and, although less common than in
tion for cardiopulmonary bypass and are there- adults, can cause heparin-induced thrombocyto-
fore familiar with the drug and comfortable with penia (HIT) [35, 36].
its use in children. Unfractionated heparin has The challenges that arise when using heparin
as the principal drug in an anticoagulation regi-
Table 6.1 Edmonton antiplatelet protocol for Berlin men have prompted some in the field of pediatric
Heart EXCOR heart failure and mechanical support to try alter-
Initiation parameters
Goal antiplatelet nate strategies with more predictable drugs.
>48 hrs Plt > 40,000 ADP Start dipyridamole VanderPluym et al. have championed the use of
inhibition <70% (4 mg/kg/day divided direct thrombin inhibitors in mechanically sup-
MAckh>56 mm in 4 doses) titrate to ported children [36]. The ideal alternate to hepa-
TEG ADP inhibition
rin would be a drug that is reliable, with highly
4–7 days All drains removed Start ASA (1 mg/kg/
AA inhibition <70% day divided in 2 predictable dosing, fast onset, and a short half-
MAckh >72 mm doses) titrate to TEG life. Ideally, the drug would not require other fac-
AA inhibition tors or plasma proteins to get the job done and
Plt platelets, ADP thromboelastography adenosine would not be impacted by renal or hepatic dys-
diphosphate pathway, AA thromboelastography arachi- function. Direct thrombin inhibitors meet most if
donic acid pathway, MAckh maximum amplitude, citrated
blood sample activated with kaolin and heparinase, ASA not all of these requirements. The most commonly
acetylsalicylic acid; aspirin used direct thrombin inhibitor in mechanically
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 109
supported children has been bivalirudin. adults and children with HIT [38–40]. Experience
Bivalirudin directly inhibits thrombin, which in with the use of bivalirudin with the EXCOR is
turn is responsible for cleaving fibrinogen into limited but so far encouraging. Rutledge et al.
fibrin and activating factor XIII, which stabilizes reported six patients supported with the Berlin
a thrombus by fibrin cross-linking. Bivalirudin Heart EXCOR. These patients were switched to
has linear pharmacokinetics, with a dose- and bivalirudin due to heparin-associated complica-
concentration-dependent activity in prolonging tions including HIT and pump thrombosis. In this
the activated clotting time (ACT), activated par- small series, one patient had a stroke with com-
tial thromboplastin time (aPTT), prothrombin plete recovery. The rest of the patients had no
time (PT), and thrombin time. It has quick onset complications while on the drug and five were
with almost immediate effect and does not bind successfully transplanted [41]. Studies are
to plasma proteins. Bivalirudin does not depend actively underway to definitively confirm the
on AT3. The drug is also metabolized by prote- superiority of direct thrombin inhibitors over the
olysis and less than 20% of it is excreted by the Edmonton protocol. There is currently no set
kidney. The half-life of the drug is 25 min, which pediatric dosing for bivalirudin in these clinical
mitigates some of the concern about the lack of scenarios. The Boston group recommends a bolus
antidote. In the case of significant bleeding or (0.1–0.2 mg/kg) if urgent increase in anticoagu-
prior to a procedure, turning off the infusion is lation is needed followed by ACT measurement
usually sufficient for reversal of anticoagulation. [36]. If the ACT is greater than 225 seconds fol-
Like all drugs however, bivalirudin has several lowing the bolus, an infusion is started at 0.15–
associated risks. Renal dysfunction will increase 0.5 mg/kg/hr, always starting at lower doses in
its half-life. When on bivalirudin, blood stasis patients with renal dysfunction. Once bivalirudin
must be avoided; proteolysis will degrade the is initiated, therapy is titrated based on aPTT
drug in static blood and allow clotting. If a proce- measurements, with target levels of 1.5–3 times
dure in which stasis is expected such as a pump that of baseline measurements, depending of the
weaning trial, transition to heparin anticoagula- patient’s individual risk of bleeding (e.g., within
tion should be strongly considered. For unclear the early postoperative period, abnormal platelet
reasons, chronic use of bivalirudin requires function) versus clotting risk (e.g., fibrin visible
increasing the dose over time. Although there is in the pump, systemic infection with increased
no antidote for the drug, fresh frozen plasma inflammation). All bleeding must stop in the
(FFP) and activated factor VIIa may be used if early postoperative period before starting bivali-
life-threatening bleeding were to occur. The need rudin. Once the infusion is started, an aPTT is
to reverse bivalirudin acutely is rarely necessary measured. The aPTT is also measured every 4 h
due to its short half-life, though half-life will be after dose change. Checking daily aPTT and INR
prolonged in patients with severe renal dysfunc- is recommended, in addition to weekly thrombo-
tion. Importantly, bivalirudin is dialyzable. elastography (TEG) with platelet mapping, lac-
There is experience in using bivalirudin in tate dehydrogenase, C-reactive protein, and full
mechanically supported children. Bivalirudin has coagulation studies (Table 6.3) [36].
been used successfully in children on ECMO
when there is concern for HIT and when antico-
agulation with heparin becomes difficult, which Heart Transplantation
can be indicated by increasing doses to maintain
adequate anticoagulation or the need for multiple Early posttransplant survival has improved over
doses of AT3. In some published series, the use of the eras. Infant and adolescent median survivals
bivalirudin on ECMO has shown no difference in are 22.3 and 13.1 years, respectively (Fig. 6.1).
rates of thromboembolism or bleeding compared Patients who survive the first year after transplant
to heparin [37]. The drug has also been used suc- have a median survival of 15 years in all age
cessfully during cardiopulmonary bypass for groups [2]. As survival improves, the focus in
ERRNVPHGLFRVRUJ
110 J. M. Lehoux et al.
Table 6.3 VanderPluym et al.’s recommendations for Table 6.4 Adverse effects of immunosuppression
starting and titrating bivalirudin
Drug Adverse effects
Bivalirudin dosing Starting Starting Tacrolimus Hyperkalemia, hypomagnesemia,
bolus infusion dose hyperglycemia, metabolic acidosis,
0.1–0.2 mg/kg 0.15–0.5 mg/ elevated transaminases,
kg/hr nephrotoxicity, tremors, hypertension,
Bivalirudin monitoring and titration headaches, leg cramps, hair loss
aPTT results Adjustment Mycophenolate Myelosuppression, gastrointestinal
1–15 seconds out of +/−0.2 mg/kg/hr. from initial side effects, headaches, viral
target range infusion rate reactivation infections (CMV, EBV),
lymphoma, pregnancy loss, and fetal
16–30 seconds outside +/− 0.5 mg/kg/hr. from initial
malformations
of target range infusion rate
Cyclosporine Hyperkalemia, hypomagnesemia,
VanderPluym [36] hyperglycemia, metabolic acidosis,
hyperlipidemia, hypertension,
pediatric transplant medicine can change to find nephrotoxicity, tremors, seizures,
gingival hyperplasia, hypertrichosis
ways to enhance long-term survival and improve
Sirolimus Gastrointestinal side effects (nausea,
quality of life by decreasing the morbidity that is diarrhea, stomach cramps),
inherent with this therapeutic modality. hyperlipidemia, proteinuria, impaired
wound healing, mouth ulcers,
myelosuppression, elevated
transaminases, pneumonitis,
Immunosuppression headaches, acne, leg cramps,
hypertension
Immunosuppression is the mainstay of transplan- Azathioprine Myelosuppression, gastrointestinal
tation management, but each of the drugs used side effects, elevated transaminases,
rash
can have adverse effects (Table 6.4). Combinations
Prednisone Hypertension, hyperglycemia,
of drugs that have evolved over the years have gastrointestinal side effects, weight
decreased the incidence of rejection while mini- gain hirsutism, edema, irritability,
mizing toxicity by avoiding the need to use high insomnia, acne, osteoporosis, growth
suppression, poor wound healing,
doses of any single drug. However, it is difficult
adrenal suppression
to say which regimen is ideal due to lack of pedi-
atric randomized, controlled trials. Without such
trials, we cannot adequately account for selection of patients receiving anti-thymocyte globulin.
bias and the numerous covariates that affect Despite the changes in clinical practice, no sur-
transplant outcomes. Clinical practice has vival benefit has been shown with any of these
changed over the years as newer drugs that are changes [2].
more immunosuppressive with less cosmetic side Given the side effects of corticosteroid use,
effects or that target different inhibitory path- not using them for chronic therapy would be
ways of T- and B-cell replication have become preferable. Since the 1980s, single-center series
available. The 2017 International Society for have reported low rejection rates and compara-
Heart and Lung Transplantation (ISHLT) registry ble survival outcomes to registry data using ste-
report shows that over the eras, cyclosporine and roid avoidance maintenance regimens [44–46].
azathioprine use has decreased, while tacrolimus Each of these reports, however, had different
and mycophenolate mofetil (MMF) use has immunosuppression protocols – some varied
increased. Similarly, as more pediatric studies over time within the same center and used echo-
suggest that induction therapy may decrease risk cardiography as the primary surveillance tool
of early rejection while not increasing the risk for for detection of rejection. Moderate cellular
infection and malignancy, its use in clinical prac- rejection by endomyocardial biopsy is not nec-
tice has changed [42, 43]. In the recent era, 70% essarily associated with echocardiographic
of pediatric heart transplant recipients received changes and therefore may underestimate cellu-
some form of induction therapy with the majority lar rejection [47].
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 111
More recently, Singh et al. reported on 55 con- mTOR use, and leucopenia and aphthous ulcers
secutive patients from 2 centers who received the in 32% of patients. It is thought that mTOR
same immunosuppression protocol consisting of inhibitors may reduce the development of graft
induction with thymoglobulin and a maintenance vasculopathy due to its anti-proliferating effects,
regimen of tacrolimus and MMF. Rejection sur- though this has yet to be clearly demonstrated in
veillance used endomyocardial biopsy at frequent pediatric heart transplant. A double-blind study
intervals during the first year posttransplant. An of 634 de novo adult heart transplant recipients
87% freedom from rejection at 1 year was randomized to either high everolimus, low-dose
reported, which at the time was lower than that everolimus, or azathioprine showed a lower inci-
reported in the International Society for Heart dence of graft vasculopathy at 6 months by intra-
and Lung Transplantation (ISHLT) registry. coronary ultrasound and a lower rate of CMV
There were 15 patients considered not eligible infection in both everolimus groups compared to
for the protocol at the time of transplant due to azathioprine [53]. A similar randomized, open-
high risk of antibody-mediated rejection. label trial using sirolimus in de novo adult heart
Excluding these patients may have lowered the transplant patients showed a reduction in acute
incidence of early rejection in this cohort. This rejection episodes and graft vasculopathy at
report was the first dual center study in pediatric 2 years [54]. In contrast, in a recent study using
heart transplantation to have a standardize immu- the Pediatric Heart Transplant Society database,
nosuppression protocol and rejection surveil- no difference was found in time to rejection, hos-
lance. Auerbach et al. (2014), using the Organ pitalization for infection, renal insufficiency,
Procurement and Transplantation Network graft vasculopathy, or survival between patients
(OPTN) database and the Pediatric Heart on sirolimus at 1 year posttransplant and
Transplant Society (PHTS) database, used pro- propensity-matched controls [2]. A similar study
pensity matching to mimic randomization and looking at early initiation of mTOR inhibitors did
were able to show no difference in graft survival not show a reduction in graft vasculopathy or sur-
between steroid-free patients and those on main- vival benefit, but patients treated with mTOR
tenance steroids at 1 year [48]. As is frequently inhibitors had a higher rate of rejection in the first
the case with large registry databases, the comor- year [42].
bidities of steroid use, such as hypertension and The multiple single-center protocols for
diabetes, were not analyzed due to incomplete induction and maintenance immunosuppression
data sets. Additionally, baseline immunosuppres- make it difficult to make comparisons and recom-
sion was not able to be analyzed in either study. mendations about the ideal immunosuppressive
Sirolimus and everolimus are classes of drug regimen. Expansion of the evidence base relating
that inhibit the mechanistic target of rapamycin to the efficacy and safety of these drugs in pediat-
(mTOR). mTOR regulates cellular metabolism, ric heart transplant recipients is necessary and
growth, and proliferation. Their use in pediatrics imminent. The TEAMMATE Trial (Tacrolimus/
remains very low, with less than 2% of patients Everolimus against Tacrolimus/MMF) recently
on one of these drugs at the time of transplant funded by the Department of Defense is the first
discharge [2]. There is evidence in the pediatric randomized, multicenter trial in pediatric heart
heart literature that conversion from a calcineurin transplant to compare the efficacy and safety of
inhibitor to an mTOR inhibitor as primary immu- two drug regimens in preventing major adverse
nosuppression or its use with a lower dose of cal- events from 6 to 36 months after transplant (clini-
cineurin inhibitor can improve renal function caltrials.gov NCT03386539). Additionally, the
[49–52]. However, the adverse effects may make prospective, observational, multicenter Clinical
its use challenging. Chinnock et al. reported Trials in Organ Transplantation in Children,
hyperlipidemia in 50% of patients, anemia and alloantibodies in children, funded by the National
neutropenia in 40%, and aphthous ulcers in 15%. Institute of Allergy and Infectious Diseases
Asante-Korang et al. reported a significant (NIAID) will give the pediatric heart transplant
increase in cholesterol and triglycerides with community a unique opportunity to look at a
ERRNVPHGLFRVRUJ
112 J. M. Lehoux et al.
cohort of pediatric heart transplant recipients would do poorly after a second transplant. There
treated with the same immunosuppression proto- have been several studies looking at large regis-
col [55]. Enrolled patients were started on a tries to assess outcome after retransplantation and
steroid-sparing protocol including 5-day thymo- to identify risk factors for poor outcome, the most
globulin induction followed by maintenance recent of which uses data from the ISHLT registry
therapy with tacrolimus and MMF. Target levels [57–59]. One-year survival after retransplantation
of tacrolimus based on time from transplant were was similar to primary transplant, but long-term
suggested in addition to suggested guidelines for survival was worse. Survival after primary trans-
treating sensitized patients perioperatively [55]. plant was 84%, 72%, and 60% at 1, 5, and 10 years
and in the retransplant group 81%, 63%, and 46%,
respectively. The median survival in primary
Retransplantation transplant recipients was 15 years compared to
8.7 years for retransplanted children [57].
Pediatric retransplantation accounts for 5% of Graft vasculopathy is the most common indi-
total pediatric transplants [56]. The lack of unifor- cation for retransplantation, accounting for more
mity in patient selection, comorbidities, and than 50% of cases [57]. It has better survival than
length of follow-up, along with small numbers in those retransplanted for other reasons (Fig. 6.4).
single-center series, make it difficult to distin- Survival after retransplant nears that of primary
guish appropriate candidates from those that transplants but only if retransplanted longer than
75
Survival (%)
50
25
Coronary Artery Disease (N=273) No pair-wise comparisons were
Primary Failure (N=38) significant at p < 0.05 except Coronary
Artery Disease vs. Primary Failure
Rejection (N=78)
0
0 1 2 3 4 5 6 7 8 9
Years
Only patients who were less than 18 years old
at the time of retransplant are included.
Fig. 6.4 Kaplan-Meier survival rates in pediatric heart rates are estimates rather than exact rates because the time
retransplant by reason. Since many patients are still alive of death is not known for all patients
and some patients have been lost to follow-up, the survival
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 113
5 years after the original transplant. Patients regarding the role of retransplantation. Transplant
retransplanted less than 1 year after the initial programs have a responsibility to their patients
transplant, presumably for graft failure, have the and need to be responsible stewards of donor
worst survival (Fig. 6.5). Multiple risk factors for organs. A careful assessment of why the first
poor outcome reflecting disease acuity while transplant failed, particularly if early after trans-
waiting for retransplant have been reported. They plant, is necessary to maximize the potential for a
include being in the ICU, need for intubation, successful second transplant and appropriate use
dialysis or cardiac operation prior to retransplant, of donor organs. Being able to risk stratify candi-
or developing an infection prior to retransplant dates who would derive the most benefit from
[57–59]. retransplantation is imperative.
In addition to inferior survival in children who
receive second transplants, there is also more
morbidity associated with retransplantation. An Rejection Surveillance
increased rate of late rejection, graft vasculopa-
thy, and renal failure has been reported [57]. Rejection is a major cause of morbidity and mor-
Given these findings and the known shortage of tality after heart transplant. Fortunately, there has
organs and waitlist mortality in patients awaiting been a decrease in the percentage of patients
primary transplant, controversy will remain being treated for rejection early after transplant
75
Survival (%)
50
25
Comparison of survival for
retransplant groups: p < 0.0001
0
0 1 2 3 4 5 6 7 8 9 10
Time (years) since most recent transplant
Fig. 6.5 Kaplan-Meier survival rates by inter-transplant is not known for all patients. A significant p-value means
intervals. Since many patients are still alive and some that at least one of the groups is different than the others,
patients have been lost to follow-up, the survival rates are but it doesn’t identify which group it is
estimates rather than exact rates because the time of death
ERRNVPHGLFRVRUJ
114 J. M. Lehoux et al.
ERRNVPHGLFRVRUJ
6 Pediatric Cardiac Transplantation and Mechanical Assist Devices 115
device exemption trial. Semin Thorac Cardiovasc 27. Nassar MS, Hasan A, Chila T, et al. Comparison
Surg. 2013;25:100–6. of paracorporeal and continuous flow ventricular
11. Strueber M, Larbalestier R, Jansz P, et al. Results of assist devices in children: preliminary results. Eur J
the post-market registry to evaluate the HeartWare left Cardiothorac Surg. 2017;51:709–14.
ventricular assist system (ReVOLVE). J Heart Lung 28. Chen JM, Richmond ME, Charette K, et al. A decade
Transplant. 2014;33:486–91. of pediatric mechanical circulatory support before
12. Conway J, Al-Aklabi M, Granoski D, et al. Supporting and after cardiac transplantation. J Thorac Cardiovasc
pediatric patients with short-term continuous-flow Surg. 2012;143:344–51.
devices. J Heart Lung Transplant. 2016;35:603–9. 29. Char DS, Lee SS, Ikoku AA, Rosenthal D, Magnus
13. Gerrah R, Charette K, Chen JM. The first successful D. Can destination therapy be implemented in chil-
use of the Levitronix PediMag ventricular support dren with heart failure? A study of provider percep-
device as a biventricular bridge to transplant in an tions. Pediatr Transplant. 2016;20:819–24.
infant. J Thorac Cardiovasc Surg. 2011;142:1282–3. 30. Villa CR, Lorts A. Cardiac destination therapy in
14. Maat AP, van Thiel RJ, Dalinghaus M, Bogers
pediatrics – are we there yet? Pediatr Transplant.
AJ. Connecting the Centrimag Levitronix pump to 2016;20:738–9.
berlin heart Excor cannulae; a new approach to bridge 31. Weinstein S, Bello R, Pizarro C, et al. The use of
to bridge. J Heart Lung Transplant. 2008;27:112–5. the Berlin Heart EXCOR in patients with func-
15. Ibrahim N. FDA letter to Medtronic, Inc. re: Heartware tional single ventricle. J Thorac Cardiovasc Surg.
HVAD system. (2017). https://www.accessdata.fda. 2014;147:697–704; discussion 704–5.
gov/cdrh_docs/pdf10/P100047S090a.pdf 32. Morales DL, Adachi I, Heinle JS, Fraser CD Jr. A new
16. Adachi I, Guzman-Pruneda FA, Jeewa A, Fraser CD era: use of an intracorporeal systemic ventricular assist
Jr, McKenzie ED. A modified implantation technique device to support a patient with a failing Fontan circu-
of the HeartWare ventricular assist device for pediat- lation. J Thorac Cardiovasc Surg. 2011;142:e138–40.
ric patients. J Heart Lung Transplant. 2015;34:134–6. 33. Steiner ME, Bomgaars LR, Massicotte MP, Berlin
17. Slaughter MS, Rogers JG, Milano CA, et al. Advanced Heart EPVADIDEsi. Antithrombotic therapy in a pro-
heart failure treated with continuous-flow left ventric- spective trial of a pediatric ventricular assist device.
ular assist device. N Engl J Med. 2009;361:2241–51. ASAIO J. 2016;62:719–27.
18. Mehra MR, Naka Y, Uriel N, et al. A fully magneti- 34. Young G, Male C, van Ommen CH. Anticoagulation
cally levitated circulatory pump for advanced heart in children: making the most of little patients and little
failure. N Engl J Med. 2017;376:440–50. evidence. Blood Cells Mol Dis. 2017;67:48–53.
19. Netuka I, Sood P, Pya Y, et al. Fully magnetically 35. Newall F, Johnston L, Ignjatovic V, Monagle
levitated left ventricular assist system for treating P. Unfractionated heparin therapy in infants and chil-
advanced HF: a multicenter study. J Am Coll Cardiol. dren. Pediatrics. 2009;123:e510–8.
2015;66:2579–89. 36. VanderPluym C. Alternative anticoagulation strate-
20. Lorts A, Zafar F, Adachi I, Morales DL. Mechanical gies for Berlin heart EXCOR. Finding solutions from
assist devices in neonates and infants. Semin Failure. Berlin Heart EXCOR user training, October
Thorac Cardiovasc Surg Pediatr Card Surg Annu. 22–24. Orlando: Nemours Children’s Hospital; 2017.
2014;17:91–5. 37. Pieri M, Agracheva N, Bonaveglio E, et al. Bivalirudin
21. Schranz D, Rupp S, Muller M, et al. Pulmonary artery versus heparin as an anticoagulant during extracorpo-
banding in infants and young children with left ven- real membrane oxygenation: a case-control study. J
tricular dilated cardiomyopathy: a novel therapeutic Cardiothorac Vasc Anesth. 2013;27:30–4.
strategy before heart transplantation. J Heart Lung 38. Gates R, Yost P, Parker B. The use of bivalirudin for
Transplant. 2013;32:475–81. cardiopulmonary bypass anticoagulation in pediatric
22. Barbaro RP, Paden ML, Guner YS, et al. Pediatric heparin-induced thrombocytopenia patients. Artif
extracorporeal life support organization registry inter- Organs. 2010;34:667–9.
national report 2016. ASAIO J. 2017;63:456–63. 39. Dyke CM, Smedira NG, Koster A, et al. A compari-
23. Steingrub JS, Tidswell M, Higgins TL. Hemodynamic son of bivalirudin to heparin with protamine reversal
consequences of heart-lung interactions. J Intensive in patients undergoing cardiac surgery with cardio-
Care Med. 2003;18:92–9. pulmonary bypass: the EVOLUTION-ON study. J
24. Pietra BA, Kantor PF, Bartlett HL, et al. Early pre- Thorac Cardiovasc Surg. 2006;131:533–9.
dictors of survival to and after heart transplantation 40. Dragomer D, Chalfant A, Biniwale R, Reemtsen B,
in children with dilated cardiomyopathy. Circulation. Federman M. Novel techniques in the use of bivali-
2012;126:1079–86. rudin for cardiopulmonary bypass anticoagulation
25. Auerbach SR, Richmond ME, Chen JM, et al.
in a child with heparin-induced thrombocytopenia.
Multiple risk factors before pediatric cardiac trans- Perfusion. 2011;26:516–8.
plantation are associated with increased graft loss. 41. Rutledge JM, Chakravarti S, Massicotte MP, Buchholz
Pediatr Cardiol. 2012;33:49–54. H, Ross DB, Joashi U. Antithrombotic strategies in
26. Zafar F, Jefferies JL, Tjossem CJ, et al. Biventricular children receiving long-term Berlin Heart EXCOR
Berlin Heart EXCOR pediatric use across the United ventricular assist device therapy. J Heart Lung
States. Ann Thorac Surg. 2015;99:1328–34. Transplant. 2013;32:569–73.
ERRNVPHGLFRVRUJ
116 J. M. Lehoux et al.
42. Castleberry C, Pruitt E, Ameduri R, et al. Risk stratifi- 54. Keogh A, Richardson M, Ruygrok P, et al. Sirolimus
cation to determine the impact of induction therapy on in de novo heart transplant recipients reduces acute
survival, rejection and adverse events after pediatric rejection and prevents coronary artery disease at
heart transplant: a multi-institutional study. J Heart 2 years: a randomized clinical trial. Circulation.
Lung Transplant. 2017;4:458–66. 2004;110:2694–700.
43. Gajarski RJ, Blume ED, Urschel S, et al. Infection and 55. Zuckerman WA, Zeevi A, Mason KL, et al. Study ratio-
malignancy after pediatric heart transplantation: the nale, design and pre-transplant alloantibody status: a
role of induction therapy. J Heart Lung Transplant. first report of clinical trials in organ transplantation in
2011;30:299–308. children-04 (CTOTC-04) in pediatric heart transplan-
44. Dionigi B, Razzouk AJ, Hasaniya NW, Chinnock RE, tation. Am J Transplant. 2018;18(9):2135–47.
Bailey LL. Late outcomes of pediatric heart trans- 56. Dipchand AI, Edwards LB, Kucheryavaya AY, et al.
plantation are independent of pre-transplant diagnosis The registry of the International Society for Heart
and prior cardiac surgical intervention. J Heart Lung and Lung Transplantation: seventeenth official
Transplant. 2008;27:1090–5. pediatric heart transplantation report – 2014; focus
45. Leonard H, Hornung T, Parry G, Dark JH. Pediatric theme: retransplantation. J Heart Lung Transplant.
cardiac transplant: results using a steroid-free mainte- 2014;33:985–95.
nance regimen. Pediatr Transplant. 2003;7:59–63. 57. Conway J, Manlhiot C, Kirk R, Edwards LB,
46. Smith RR, Wray J, Khaghani A, Yacoub M. Ten
McCrindle BW, Dipchand AI. Mortality and mor-
year survival after paediatric heart transplantation: bidity after retransplantation after primary heart
a single centre experience. Eur J Cardiothorac Surg. transplant in childhood: an analysis from the reg-
2005;27:790–4. istry of the International Society for Heart and
47. Rosenthal DN, Chin C, Nishimura K, et al. Identifying Lung Transplantation. J Heart Lung Transplant.
cardiac transplant rejection in children: diagnostic 2014;33:241–51.
utility of echocardiography, right heart catheteriza- 58. Mahle WT, Vincent RN, Kanter KR. Cardiac retrans-
tion and endomyocardial biopsy data. J Heart Lung plantation in childhood: analysis of data from the
Transplant. 2004;23:323–9. united network for organ sharing. J Thorac Cardiovasc
48. Auerbach SR, Gralla J, Campbell DN, Miyamoto
Surg. 2005;130:542–6.
SD, Pietra BA. Steroid avoidance in pediatric heart 59. Chin C, Naftel D, Pahl E, et al. Cardiac re-
transplantation results in excellent graft survival. transplantation in pediatrics: a multi-institutional
Transplantation. 2014;97:474–80. study. J Heart Lung Transplant. 2006;25:1420–4.
49. Chinnock TJ, Shankel T, Deming D, et al. Calcineurin 60.
Daly KP, Marshall AC, Vincent JA, et al.
inhibitor minimization using sirolimus leads to Endomyocardial biopsy and selective coronary
improved renal function in pediatric heart transplant angiography are low-risk procedures in pediat-
recipients. Pediatr Transplant. 2011;15:746–9. ric heart transplant recipients: results of a mul-
50. Matthews K, Gossett J, Kappelle PV, Jellen G, Pahl ticenter experience. J Heart Lung Transplant.
E. Indications, tolerance and complications of a 2012;31:398–409.
sirolimus and calcineurin inhibitor immunosuppres- 61. Zhorne D, Petit CJ, Ing FF, et al. A 25-year experience
sion regimen: intermediate experience in pediatric of endomyocardial biopsy safety in infants. Catheter
heart transplantation recipients. Pediatr Transplant. Cardiovasc Interv. 2013;82:797–801.
2010;14:402–8. 62.
Stendahl G, Bobay K, Berger S, Zangwill
51. Behnke-Hall K, Bauer J, Thul J, et al. Renal function S. Organizational structure and processes in pediatric
in children with heart transplantation after switching heart transplantation: a survey of practices. Pediatr
to CNI-free immunosuppression with everolimus. Transplant. 2012;16:257–64.
Pediatr Transplant. 2011;15:784–9. 63. Godown J, Harris MT, Burger J, Dodd DA. Variation
52. Asante-Korang A, Carapellucci J, Krasnopero D,
in the use of surveillance endomyocardial biopsy
Doyle A, Brown B, Amankwah E. Conversion from among pediatric heart transplant centers over time.
calcineurin inhibitors to mTOR inhibitors as primary Pediatr Transplant. 2015;19:612–7.
immunosuppressive drugs in pediatric heart trans- 64. Castleberry C, Ziniel S, Almond C, et al. Clinical
plantation. Clin Transpl. 2017;31:e13054. practice patterns are relatively uniform between pedi-
53. Eisen HJ, Tuzcu EM, Dorent R, et al. Everolimus for atric heart transplant centers: a survey-based assess-
the prevention of allograft rejection and vasculopa- ment. Pediatr Transplant. 2017;21:e13013.
thy in cardiac-transplant recipients. N Engl J Med.
2003;349:847–58.
ERRNVPHGLFRVRUJ
Surgical Management
of Hypoplastic Left Heart 7
Syndrome
Peter Sassalos and Richard G. Ohye
ERRNVPHGLFRVRUJ
118 P. Sassalos and R. G. Ohye
Fig. 7.2 Current traditional surgical management of HLHS consists of staged palliation to a Fontan circulation
Based on this standard management, out- was 15.6%, the hemi-Fontan or bidirectional
comes have dramatically improved. The aggre- Glenn procedure was 2.1%, and the Fontan pro-
gate outcomes of staged palliation of all cedure was 1.4%. The aggregate average post-
participants of the Society of Thoracic Surgeons operative length of stay for the Norwood
Congenital Heart Surgery Database (STS procedure was 42.2 days, the hemi-Fontan or
CHSD) were reported in 2016. The aggregate bidirectional Glenn procedure was 13.8 days,
operative mortality for the Norwood procedure and the Fontan procedure was 13.4 days [10]. In
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 119
experienced centers, hospital survival following Table 7.1 Practice pattern variation in perioperative care
the Norwood procedure has been reported of neonates undergoing the Norwood procedure using the
Single Ventricle Reconstruction (SVR) trial dataset. The
greater than 90% [11]. range is the differences in practice between each center.
Despite these advances, there is still signifi- This demonstrates significant variability in preoperative,
cant morbidity and mortality associated with intraoperative, and postoperative variables. (Adapted
HLHS. This has led to investigation into new and from Pasquali et al. [13]) (DHCA deep hypothermic circu-
latory arrest, ECMO extracorporeal membrane oxygen-
alternative therapies. As a result, there is substan- ation, ICU intensive care unit, HCT hematocrit, RCP
tial practice pattern variation among institutions retrograde cerebral perfusion)
and between individual surgeons, cardiologists, Practice pattern variation
and intensivists. Preoperative variables
Wernovsky et al. conducted an online survey Fetal diagnosis 55–85%
in 2007 evaluating management of HLHS in 52 Intubation 29–91%
centers worldwide thought to manage 1000 neo- Intraoperative variables
nates with HLHS annually. The results demon- Total bypass support time 74–189 min
strated considerable variability in most Aortic cross-clamp time 33–73 min
parameters. Some results favored a consensus DHCA only 3–100%
RCP only 3–93%
opinion, whereas others favored equivocal or
Lowest HCT 22–41%
even controversial conclusions. Of note, the
Classic arch reconstruction 41–100%
type of intensive care unit in which patients Postoperative variables
were managed, both before and after surgery, ICU stay 9–44 days
varied widely among centers [12]. Pasquali Hospital stay 19–44 days
et al. then conducted a study in 2012 to evaluate Ventilator time 4–16 days
practice pattern variation in perioperative care Open sternum 35–100%
of neonates undergoing the Norwood procedure ECMO 7–35%
using the Single Ventricle Reconstruction (SVR) Enteral tube 2–100%
trial dataset. This also demonstrated significant Home monitoring 1–100%
Death or transplant during Norwood 7–39%
variability in preoperative, intraoperative, and hospitalization
postoperative variables (Table 7.1) [13].
However, of particular interest, significant dif-
ferences also existed for in-hospital mortality Another controversial topic is regionalization
and transplantation between centers. One may of care to centers of excellence. Using the 2003
then theorize that practice pattern variation and Kids’ Inpatient Database, hospital mortality for
current controversies in the management of the Norwood procedure and arterial switch oper-
these patients may be partially responsible for ations were studied as a function of institutional
these differences. Therefore, if best practices volume. A total of 624 Norwood procedures per-
can be identified, hopefully outcomes for HLHS formed at 60 hospitals, with a case range of 1–31
patients may improve. This has already been operations a year at each site, were evaluated. An
demonstrated in adult cardiac surgery with the inverse relationship was demonstrated between
Michigan Society of Thoracic and institutional volume and mortality with 35%,
Cardiovascular Surgery (MSTCVS) [14] and 26%, and 17% mortality in hospitals performing
Northern New England Cardiovascular Disease 2 per year, 10 per year, and 20 per year, respec-
Study Group [15]. Through adoption of prac- tively [17]. Using the STS CHSD, 2555 patients
tices used by high-performing centers, variation undergoing the Norwood procedure at 53 centers
in care was reduced, outcomes were improved, by a total of 111 surgeons were also evaluated.
and hospital costs lowered. Similar efforts are This demonstrated that lower center and surgeon
now being made through the National Pediatric volume were associated with higher mortality
Cardiology Quality Improvement Collaborative [18]. To better understand these differences, a
(NPC-QIC) [16]. large analysis of the 2006–2009 STS CHSD was
ERRNVPHGLFRVRUJ
120 P. Sassalos and R. G. Ohye
Table 7.2 The current controversies in the surgical man- Fetal Cardiac Intervention
agement of HLHS which will be the focus of this chapter
Current controversies Prenatal diagnosis of HLHS has increased. Since
Fetal cardiac intervention 2012, 82% of Norwood procedures performed at
First-stage palliation
the University of Michigan had a prenatal diag-
Norwood procedure
nosis. This allows surveillance of high-risk
Shunt type
Deep hypothermic circulatory arrest versus regional lesions, the opportunity for counseling, optimal
cerebral perfusion timing and location of delivery, and better transi-
Delayed sternal closure tion to postnatal care. Although it has not clearly
Hybrid Norwood demonstrated improved survival [21], there is
Postoperative management evidence of decreased morbidity [22]. In addi-
Second-stage palliation tion, it offers the consideration for fetal cardiac
Third-stage palliation intervention (FCI) performed in select quaternary
Role of a fenestration
referral centers.
Mechanical circulatory support
Transplantation
Fetal cardiac interventions are reserved at this
time for HLHS patients with either aortic valve
stenosis or a restrictive/intact atrial septum. It has
performed including 40,930 patients at 72 cen- been proposed that aortic valve stenosis in a fetus
ters. Interestingly, there was no difference in with an initially normal-sized left ventricle can
complication rates between high, middle, and progress to HLHS. Therefore, the hope is that
low mortality hospitals. However, low mortality fetal balloon valvuloplasty can prevent this pro-
hospitals had the lowest failure to rescue rate (the gression. It is currently selectively utilized in
probability of death after a complication) [19]. patients who are not yet thought to have HLHS or
This has now become a well-known phenomenon in patients who may have potential for biventric-
that has been demonstrated across many other ular repair [23–33].
surgical specialties [20]. A restrictive or intact atrial septum is a known
As is clearly evident, multiple current contro- risk factor for HLHS [34]. It leads to pulmonary
versies exist regarding surgical management of venous obstruction and irreversible pulmonary
HLHS (Table 7.2). They will therefore be dis- vascular changes that are unfavorable to a patient
cussed in the remainder of this chapter. destined for staged Fontan palliation. Therefore,
the hope is that fetal balloon atrial septostomy or
atrial septal stent placement can prevent these
Patient Scenario changes [35, 36]. Although these therapies have
A 2.5 kg male born at 35 weeks gestational shown some promising results, additional work
age with a postnatal diagnosis of HLHS is needed as they carry high risk with an esti-
presents from an outside hospital in shock. mated procedural fetal loss at approximately
The patient is intubated and begun on pros- 10–15% [21].
taglandin infusion and vasoactive support.
A transthoracic echocardiogram demon-
strates a nonrestrictive atrial septum, First-Stage Palliation
HLHS consisting of mitral stenosis and
aortic stenosis, a 3 mm ascending aorta, Norwood Procedure
and no evidence of coronary sinusoids. The
resident physician caring for this baby asks First-stage palliation for HLHS has traditionally
what surgical options are available as well been the Norwood procedure [9]. This procedure
as if something could have prevented or is performed within the first 7–14 days of life via
mitigated the severity of this condition. a median sternotomy using cardiopulmonary
bypass. The three goals of the procedure are to
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 121
ERRNVPHGLFRVRUJ
122 P. Sassalos and R. G. Ohye
Table 7.4 The theoretical advantages and disadvantages Despite this landmark trial, the controversy
of the right ventricle-to-pulmonary artery shunt at the regarding shunt type still exists. The initial sur-
Norwood procedure. (PA, pulmonary artery; PBF, pulmo-
nary blood flow)
vival benefit of the RVPAS has statistically been
lost in longer-term studies; however, the trend
Theoretical advantages and disadvantages of the right
ventricle-to-pulmonary artery shunt still is present. Overall survival for patients with
Advantages Disadvantages HLHS needs to improve, and continued investi-
Noncontinuous shunt flow Less PBF gation into ideal shunt type at the Norwood pro-
in systole cedure remains an important clinical question.
No diastolic runoff More cyanosis
Improved diastolic Less PA growth
systemic perfusion
Improved coronary blood More PA interventions eep Hypothermic Circulatory Arrest
D
flow Versus Regional Cerebral Perfusion
Improved end-organ Right ventriculotomy
perfusion Cardiopulmonary bypass technique during the
Hemodynamic stability Decreased systemic Norwood procedure differs between surgeons
ventricular function
and institutions, particularly pertaining to aortic
Improved survival Increased
arrhythmogenicity arch reconstruction. Options performed include
standard cardiopulmonary bypass with variable
degrees of systemic cooling, deep hypothermic
MBTS or RVPAS at the time of a Norwood pro- circulatory arrest (DHCA), regional cerebral per-
cedure. Of note, the SVR trial was the first ran- fusion (RCP), or maintenance of total body per-
domized control trial comparing two operations fusion. In general, there is a 50% metabolic
in congenital heart surgery. The primary outcome reduction for every 10° Fahrenheit (7 °C)
was transplant-free survival at 12 months, which decrease in temperature, known as the metabolic
was statistically better at 74% for the RVPAS as reduction Q10 rule. Cooling the patient allows
compared to 64% for the MBTS. Secondary out- decreased cardiopulmonary bypass flow and sys-
comes were perioperative morbidity after the temic oxygen delivery with end-organ preserva-
Norwood procedure, unintended cardiovascular tion. For DHCA, the head is placed in ice, and the
interventional procedures, right ventricular func- patient is cooled to 18 °C for at least 20 min to
tion by echocardiography, pulmonary arterial ensure even cooling. The pump is then turned off
size by angiography, and neurodevelopment at and the patient exsanguinated into the venous
14 months [44]. reservoir for the period of aortic arch reconstruc-
The longer-term results at both 3 and 6 years tion. This is our preference. In contrast, RCP can
have now also been reported. At 3 years, trans- be performed by retrograde flow through the
plant-free survival was not statistically different superior vena cava (SVC) or antegrade into the
at 67% for RVPAS as compared to 61% for the head vessels, typically the innominate artery,
MBTS. The RVPAS group had more catheter either by direct cannulation or through a graft.
interventions and worse RV ejection fraction This can be performed during DHCA to maintain
[45]. At 6 years, transplant-free survival was not continuous cerebral perfusion. A few groups have
statistically different at 64% for RVPAS as com- also described perfusing the brain using the
pared to 59% for the MBTS. The RVPAS group above RCP techniques, as well as the lower body
had more catheter interventions pre-Fontan, and by cannulating the descending aorta.
there was no difference in either RV ejection The Boston Circulatory Arrest Study was a
fraction or complications between the two single-center randomized control trial that evalu-
groups. However, there was overall significant ated perioperative neurologic effects associated
morbidity in both groups [46]. In addition to with DHCA versus low-flow cardiopulmonary
these results, the SVR database has led to a pleth- bypass during the arterial switch operation for
ora of other important PHN studies as well [47]. transposition of the great arteries. This demon-
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 123
strated higher risk of clinical seizures, higher risk onstrated practice pattern variation with regard to
of ictal activity on EEG monitoring, and greater delayed sternal closure and that centers with
release of creatine kinase brain isoenzyme asso- greater use had higher postoperative infection
ciated with DHCA [48]. Long-term neurodevel- and length of stay [52]. Some institutions and
opmental outcomes at 4 and 8 years of age were surgeons choose to electively leave all patients
then evaluated. At 4 years of age, there was no open, whereas others selectively decide based on
difference in IQ or overall neurologic status; individual patient factors (Table 7.2). Our prefer-
however, DHCA patients had worse motor coor- ence is the latter based on an intraoperative deci-
dination and planning [49]. At 8 years of age, sion by the surgeon.
both groups were associated with increased risk
of neurodevelopmental abnormalities; however,
the DHCA patients generally had greater func-
tional deficits [50]. Patient Scenario
The University of Michigan then conducted a A hybrid Norwood procedure was recom-
single-center randomized control trial that evalu- mended given the size, age, and presenta-
ated neurodevelopment outcomes associated tion of the patient. Once the patient was
with DHCA versus RCP in patients undergoing stabilized with end-organ recovery, he was
the Norwood procedure. Neurodevelopment was taken to the operating room for this proce-
measured prior to second-stage palliation and at dure performed jointly by a congenital
1 year of age. This study did not suggest improved heart surgeon and interventional cardiolo-
outcomes with RCP [51]. At this time, both strat- gist. A median sternotomy was performed.
egies will continue to be used until further multi- Bilateral PA bands were constructed from
center studies show clear consensus on which 3.0 millimeter Gore-Tex graft and applied.
technique is related to superior outcomes. Catheter-based distal PA pressures of
approximately 15 mmHg were achieved. An
angiogram was then performed to assess
Delayed Sternal Closure ductal and arch anatomy. A PDA stent was
placed with a completion angiogram per-
Delayed sternal closure refers to temporary patch formed. The chest was closed and the
closure of the skin with the sternum left open at patient transferred to a dedicated pediatric
the time of the Norwood procedure and other cardiothoracic intensive care unit.
complex congenital heart surgeries. The sternum
is then closed as a separate operation typically
several days later either in the intensive care unit
or operating room. The proposed advantage is to Hybrid Norwood
provide more space in the setting of decreased
function, myocardial edema, and coagulopathy. Given the significant morbidity and mortality
This can minimize the effects of diastolic dys- associated with the Norwood procedure, collabo-
function and elevated filling pressures, potential ration between surgeons and interventional cardi-
compression of anterior structures such as the ologists led to a less invasive alternative strategy
RVPAS, increased vasoactive requirements with first described by Gibbs et al. in 1993 [53]. It has
hemodynamic instability, tamponade physiology, since been advanced by teams in Giessen,
and cardiac arrest. The proposed disadvantages Germany, and Columbus, Ohio. This hybrid
are infectious and wound healing risks, need for Norwood procedure, also known as hybrid stage
an additional operation and anesthesia, and pro- I palliation, has become an evolving therapy in
longed mechanical ventilation. A study using the the armamentarium for repair of HLHS.
STS CHSD evaluated 1283 infants undergoing The hybrid Norwood procedure achieves the
the Norwood procedure from 45 centers. It dem- same goals of the traditional Norwood procedure
ERRNVPHGLFRVRUJ
124 P. Sassalos and R. G. Ohye
without cardiopulmonary bypass (Table 7.3). ensure coronary perfusion in HLHS patients with
Unobstructed pulmonary venous return is aortic atresia [56].
achieved by balloon atrial septostomy or atrial The surgical indications for the hybrid
septal stent placement. Unobstructed systemic Norwood procedure are also quite variable. Some
outflow is accomplished by placement of a patent centers have adopted this for all HLHS patients,
ductus arteriosus (PDA) stent, either percutane- and others have selectively used it only for high-
ous or transthoracic via the main pulmonary risk subgroups. Our preference to date has been
artery, or by continuous prostaglandin infusion. the latter. High-risk patients are defined as less
Adequate, but restricted, pulmonary blood flow is than 2.5 kg [57], less than 34 weeks gestation,
achieved by placement of bilateral PA bands. intact or highly restrictive atrial septum, severe
Despite the same goals, multiple strategies have tricuspid regurgitation, severe right ventricular
been employed to achieve this result (Fig. 7.3). dysfunction, severe noncardiac medical or
The Giessen technique is a median sternotomy genetic conditions, renal dysfunction, intracra-
for placement of bilateral PA bands followed by nial hemorrhage or neurologic injury, contraindi-
percutaneous PDA stent placement and atrial cation to cardiopulmonary bypass, severe
septal intervention [54]. The Columbus technique ascending aortic hypoplasia (<2 millimeters),
is a median sternotomy for placement of bilateral coronary sinusoids (mitral stenosis, aortic atre-
PA bands and transthoracic PDA stent placement sia), and postnatal cardiac arrest or shock. Of
followed by delayed percutaneous atrial septal note, the hybrid Norwood has also been used for
intervention, except in cases of restrictive or potential biventricular patients with Shone’s
intact atrial septum [55]. An additional technique complex and high-risk features, interrupted aor-
described is placement of a reversed MBTS to tic arch with high-risk features, and critical aortic
Hybrid
Norwood
Traditional
Second-Stage
Traditional
Third-Stage
Fig. 7.3 Hybrid Norwood treatment algorithms and subsequent pathway options
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 125
ERRNVPHGLFRVRUJ
126 P. Sassalos and R. G. Ohye
maintain adequate systemic oxygen delivery. or bidirectional Glenn procedure (Fig. 7.2).
This is assessed by clinical exam, routine vital Both create a superior cavopulmonary connec-
signs, intracardiac pressure monitoring, continu- tion as the source of pulmonary blood flow
ous mixed venous oximetry, near-infrared spec- while volume unloading the ventricle. The pre-
troscopy (NIRS), continuous pulse oximetry, vious systemic-to-pulmonary artery shunt is
arterial and venous blood gas sampling, and removed.
markers of end-organ function. Typically, the The bidirectional Glenn procedure creates
goal is a mean arterial blood pressure approxi- this connection by an end-to-side anastomosis
mately 40–45 mmHg, an oxygen saturation between the divided SVC and a longitudinal
70–75%, an arteriovenous difference of 20%, ipsilateral branch pulmonary arteriotomy. In
normal pH, a PCO2 40 mmHg, a PO2 contrast, both the original and modified hemi-
30–40 mmHg, normal lactic acid levels without Fontan procedures create this connection by
significant base deficit, and a hematocrit greater suturing a right atriotomy to the central pulmo-
than 40%. To achieve this physiologic balance, nary arteries which are augmented with an
the systemic (SVR) and pulmonary vascular allograft patch. The right atrium, which is now a
resistance (PVR) can be manipulated to control common atrium due to the previous atrial septec-
respective blood flows. The SVR can be increased tomy, is partitioned by patch. Therefore, the
by systemic vasoconstrictors such as vasopressin, SVC return enters the partitioned superior por-
norepinephrine, epinephrine, or high-dose dopa- tion of the common atrium to flow into the pul-
mine infusions. The SVR can be decreased by monary arteries. The inferior vena cava (IVC)
systemic vasodilators such as milrinone, direct return enters the partitioned inferior portion of
arterial vasodilators, or alpha-antagonists. The the common atrium to enter the right ventricle
PVR can be increased by increasing PCO2, either [59, 60].
through decreased minute ventilation or addition There remains controversy as to the appropri-
of inhaled CO2, or decreasing PO2, either by low- ate second-stage palliation. Excellent results
ering the FiO2 and PEEP or administration of have been demonstrated with both procedures.
sub-ambient O2. The PVR can be decreased by The choice becomes institution and surgeon
decreasing PCO2, increasing PO2, or adding pul- dependent largely based on experience. The bidi-
monary vasodilators such as inhaled nitric oxide rectional Glenn procedure is currently the more
or oral sildenafil. In addition, optimization of commonly performed second-stage palliation at
medical therapy is important with temperature most centers. Advocates favor this approach for
control, appropriate pain and sedation control, the technical ease and ability to perform without
possible neuromuscular blockade, inotropic sup- cardiac arrest or even without cardiopulmonary
port, acid-base management, and adequate oxy- bypass [61, 62]. However, the hemi-Fontan is our
gen carrying capacity with blood transfusion as procedure of choice at the University of Michigan
needed. In general, the goal for the balance unless anatomically not feasible, such as some
between SBF and PBF is generally a ratio of 1:1, cases of anomalous pulmonary venous connec-
which maintains both adequate peripheral oxy- tions, select cases of abnormal relationship and
gen saturation (~75%) and systemic cardiac position of the atria to the ventricles, and some
output. forms of heterotaxy with anomalous systemic
venous connections. If bilateral SVC is present,
we elect to perform a right modified hemi-Fontan
Second-Stage Palliation with a left bidirectional Glenn procedure.
Although technically more challenging, it is
This stage is typically performed between 4 and favored because it is felt for many reasons to
6 months of age. Once deemed an appropriate make patients more suitable Fontan candidates
candidate, the options include the hemi-Fontan [59]. Optimal PA anatomy is ensured through
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 127
routine augmentation of the branch pulmonary the atrium surrounding the orifice of the IVC
arteries. This more complex operation simplifies and any additional hepatic veins. The conduit is
the lateral tunnel Fontan when the postoperative then brought through the atriotomy which is
hemodynamics are more demanding. The entire closed around the conduit. The completion of
cardiac output, with the exception of a fenestra- the conduit is then performed in a similar
tion, passes through the lungs at the Fontan stage. fashion to the extracardiac conduit described
Therefore, longer anesthetic and cardiopulmo- above [66, 67].
nary bypass times can negatively impact the In contrast, the lateral tunnel is performed
lungs which more seriously affect a Fontan through a right atriotomy from the inferior
patient. Lastly, mathematical modeling has dem- cavoatrial junction to just inferior to the previ-
onstrated that the hemi-Fontan with lateral tunnel ously placed hemi-Fontan patch (which is later
Fontan circulation has more favorable flow pat- removed). A lateral tunnel the width of the IVC
terns with less energy loss and more equal distri- is created with a PTFE patch. The patch is
bution of IVC blood flow, as compared to the sutured around the internal orifice of the IVC,
bidirectional Glenn with extracardiac conduit anterior to the right pulmonary veins, around the
Fontan circulation [63]. orifice of the SVC, and the anterior edge of the
patch is then incorporated into the atriotomy clo-
sure. This creates a lateral tunnel pathway within
Third-Stage Palliation the common atrium where SVC and IVC return
is directed into the pulmonary arteries [59, 60,
This stage is typically performed between 18 and 68, 69].
48 months of age depending on the type of Fontan There also remains controversy as to the
performed. Once deemed an appropriate candi- appropriate third-stage palliation. Excellent
date, the options include the intra-atrial lateral results have been demonstrated with both proce-
tunnel Fontan, otherwise known as the lateral dures [70–82]. Once again, the choice becomes
tunnel Fontan, the extracardiac conduit Fontan, institution and surgeon dependent largely based
or the intra-/extracardiac conduit Fontan on experience. The extracardiac conduit Fontan
(Fig. 7.2). Each completes the Fontan circulation is currently the more commonly performed third-
by directing the IVC blood directly to the lungs. stage palliation. Advocates favor this approach
Following this stage, the entire deoxygenated for technical ease, ability to perform without car-
systemic venous return will drain directly into the diac arrest or even cardiopulmonary bypass [71,
pulmonary arteries, driven only by central venous 83, 84], decreased arrhythmogenicity due to less
pressure. The oxygenated pulmonary venous atrial suture lines and the atrium excluded from
return drains into the common atrium to be deliv- higher venous pressures [67, 85], and less PA
ered to the systemic circulation via the systemic reconstruction if transplantation required in the
right ventricle. future [86].
The extracardiac conduit Fontan is performed However, the lateral tunnel Fontan is our pro-
by placement of an interposition graft, typically cedure of choice at the University of Michigan
an 18–20 mm stretch PTFE graft, between the when anatomically possible for several reasons.
divided IVC and either the SVC or an arteriot- The procedure is performed with technical ease
omy on the inferior aspect of the PA involved in following the hemi-Fontan procedure. There are
the bidirectional Glenn anastomosis. There is more favorable flow patterns as described above
controversy though as to the ideal size of the [63]. Less prosthetic material is used which pre-
conduit [64] and location of the latter anastomo- serves growth potential and possibly decreases
sis [65]. The intra-/extracardiac conduit Fontan thrombogenicity [69]. Fenestration is easily per-
is a modification of this where an anastomosis is formed, and more ready percutaneous catheter
performed between the end of the conduit and access to the common atrium is maintained. In
ERRNVPHGLFRVRUJ
128 P. Sassalos and R. G. Ohye
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 129
ERRNVPHGLFRVRUJ
130 P. Sassalos and R. G. Ohye
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 131
fetuses with severe aortic valve obstruction. Am J 36. Marshall AC, van der Velde ME, Tworetzky W, Gomez
Cardiol. 2000;85(10):1230–3. CA, Wilkins-Haug L, Benson CB, et al. Creation of an
26. Kohl T, Witteler R, Strumper D, Gogarten W, Asfour atrial septal defect in utero for fetuses with hypoplas-
B, Reckers J, et al. Operative techniques and strategies tic left heart syndrome and intact or highly restrictive
for minimally invasive fetoscopic fetal cardiac inter- atrial septum. Circulation. 2004;110(3):253–8. https://
ventions in sheep. Surg Endosc. 2000;14(5):424–30. doi.org/10.1161/01.CIR.0000135471.17922.17.
27. Kohl T, Strumper D, Witteler R, Merschhoff G,
37. Tweddell JS, Mitchell ME, Woods RK, Spray TL,
Alexiene R, Callenbeck C, et al. Fetoscopic direct Quintessenza JA. Construction of the right ventricle-
fetal cardiac access in sheep: an important experimen- to-pulmonary artery conduit in the Norwood: the
tal milestone along the route to human fetal cardiac “Dunk” technique. Oper Tech Thorac Cardiovasc
intervention. Circulation. 2000;102(14):1602–4. Surg. 2012;17(2):81–98. https://doi.org/10.1053/j.
28.
Tworetzky W, Wilkins-Haug L, Jennings RW, optechstcvs.2012.05.003.
van der Velde ME, Marshall AC, Marx GR, et al. 38. Mascio CE, Spray TL. Distal dunk for right ventri-
Balloon dilation of severe aortic stenosis in the cle to pulmonary artery shunt in stage 1 palliation.
fetus: potential for prevention of hypoplastic left Ann Thorac Surg. 2015;100(6):2381–2. https://doi.
heart syndrome: candidate selection, technique, org/10.1016/j.athoracsur.2015.05.024.
and results of successful intervention. Circulation. 39. Norwood WI, Lang P, Casteneda AR, Campbell
2004;110(15):2125–31. https://doi.org/10.1161/01. DN. Experience with operations for hypoplas-
CIR.0000144357.29279.54. tic left heart syndrome. J Thorac Cardiovasc Surg.
29. Marshall AC, Tworetzky W, Bergersen L, McElhinney 1981;82(4):511–9.
DB, Benson CB, Jennings RW, et al. Aortic valvulo- 40. Kishimoto H, Kawahira Y, Kawata H, Miura T,
plasty in the fetus: technical characteristics of suc- Iwai S, Mori T. The modified Norwood pallia-
cessful balloon dilation. J Pediatr. 2005;147(4):535–9. tion on a beating heart. J Thorac Cardiovasc Surg.
https://doi.org/10.1016/j.jpeds.2005.04.055. 1999;118(6):1130–2. https://doi.org/10.1016/
30. Selamet Tierney ES, Wald RM, McElhinney DB,
S0022-5223(99)70118-2.
Marshall AC, Benson CB, Colan SD, et al. Changes 41. Sano S, Ishino K, Kawada M, Arai S, Kasahara S,
in left heart hemodynamics after technically success- Asai T, et al. Right ventricle-pulmonary artery shunt
ful in-utero aortic valvuloplasty. Ultrasound Obstet in first-stage palliation of hypoplastic left heart syn-
Gynecol. 2007;30(5):715–20. https://doi.org/10.1002/ drome. J Thorac Cardiovasc Surg. 2003;126(2):504–
uog.5132. 9; discussion 9–10.
31. McElhinney DB, Marshall AC, Wilkins-Haug LE,
42. Sano S, Ishino K, Kado H, Shiokawa Y, Sakamoto K,
Brown DW, Benson CB, Silva V, et al. Predictors of Yokota M, et al. Outcome of right ventricle-to-pulmo-
technical success and postnatal biventricular outcome nary artery shunt in first-stage palliation of hypoplastic
after in utero aortic valvuloplasty for aortic steno- left heart syndrome: a multi-institutional study. Ann
sis with evolving hypoplastic left heart syndrome. Thorac Surg. 2004;78(6):1951–7; discussion 7–8.
Circulation. 2009;120(15):1482–90. https://doi. https://doi.org/10.1016/j.athoracsur.2004.05.055.
org/10.1161/CIRCULATIONAHA.109.848994. 43. Sano S, Ishino K, Kawada M, Honjo O. Right
32.
Mizrahi-Arnaud A, Tworetzky W, Bulich LA, ventricle-pulmonary artery shunt in first-stage pal-
Wilkins-Haug LE, Marshall AC, Benson CB, et al. liation of hypoplastic left heart syndrome. Semin
Pathophysiology, management, and outcomes of Thorac Cardiovasc Surg Pediatr Card Surg Annu.
fetal hemodynamic instability during prenatal cardiac 2004;7:22–31.
intervention. Pediatr Res. 2007;62(3):325–30. https:// 44. Ohye RG, Sleeper LA, Mahony L, Newburger JW,
doi.org/10.1203/PDR.0b013e318123fd3a. Pearson GD, Lu M, et al. Comparison of shunt types
33.
Vogel M, Wilkins-Haug LE, McElhinney DB, in the Norwood procedure for single-ventricle lesions.
Marshall AC, Benson CB, Silva V, et al. Reversible N Engl J Med. 2010;362(21):1980–92. https://doi.
ductus arteriosus constriction due to maternal indo- org/10.1056/NEJMoa0912461.
methacin after fetal intervention for hypoplastic left 45. Newburger JW, Sleeper LA, Frommelt PC, Pearson
heart syndrome with intact/restrictive atrial sep- GD, Mahle WT, Chen S, et al. Transplantation-
tum. Fetal Diagn Ther. 2010;27(1):40–5. https://doi. free survival and interventions at 3 years in the
org/10.1159/000268290. single ventricle reconstruction trial. Circulation.
34. Rychik J, Rome JJ, Collins MH, DeCampli WM,
2014;129(20):2013–20. https://doi.org/10.1161/
Spray TL. The hypoplastic left heart syndrome with circulationaha.113.006191.
intact atrial septum: atrial morphology, pulmonary 46. Newburger JW, Sleeper LA, Gaynor JW, Hollenbeck-
vascular histopathology and outcome. J Am Coll Pringle D, Frommelt PC, Li JS, et al. Transplant-free
Cardiol. 1999;34(2):554–60. survival and interventions at 6 years in the single ven-
35. Marshall AC, Levine J, Morash D, Silva V, Lock JE, tricle reconstruction trial. Circulation. 2018; https://
Benson CB, et al. Results of in utero atrial septoplasty doi.org/10.1161/CIRCULATIONAHA.117.029375.
in fetuses with hypoplastic left heart syndrome. Prenat 47. Si MS, Pearson GD, Ohye RG. Shunt choice in sin-
Diagn. 2008;28(11):1023–8. https://doi.org/10.1002/ gle right ventricle patients: an update. Expert Rev
pd.2114.
ERRNVPHGLFRVRUJ
132 P. Sassalos and R. G. Ohye
ERRNVPHGLFRVRUJ
7 Surgical Management of Hypoplastic Left Heart Syndrome 133
patients. Pediatr Cardiol. 2007;28(6):426–32. https:// tion: association between early outcome and type
doi.org/10.1007/s00246-007-9002-5. of cavopulmonary connection. Ann Thorac Surg.
70. Hosein RBM, Clarke AJB, McGuirk SP, Griselli M, 2012;93(4):1254–60; discussion 61. https://doi.
Stumper O, De Giovanni JV, et al. Factors influenc- org/10.1016/j.athoracsur.2012.01.060.
ing early and late outcome following the Fontan pro- 81. Ono M, Kasnar-Samprec J, Hager A, Cleuziou J,
cedure in the current era. The ‘two commandments’? Burri M, Langenbach C, et al. Clinical outcome fol-
Eur J Cardiothorac Surg. 2007;31(3):344–53. https:// lowing total cavopulmonary connection: a 20-year
doi.org/10.1016/j.ejcts.2006.11.043. single-centre experience. Eur J Cardiothorac Surg.
71. Petrossian E, Reddy VM, Collins KK, Culbertson CB, 2016; https://doi.org/10.1093/ejcts/ezw091.
MacDonald MJ, Lamberti JJ, et al. The extracardiac 82. Ravishankar C, Gerstenberger E, Sleeper LA, Atz
conduit Fontan operation using minimal approach AM, Affolter JT, Bradley TJ, et al. Factors affecting
extracorporeal circulation: early and midterm out- Fontan length of stay: results from the single ven-
comes. J Thorac Cardiovasc Surg. 2006;132(5):1054– tricle reconstruction trial. J Thorac Cardiovasc Surg.
63. https://doi.org/10.1016/j.jtcvs.2006.05.066. 2016;151(3):669–75 e1. https://doi.org/10.1016/j.
72. Hirsch JC, Goldberg C, Bove EL, Salehian S, Lee jtcvs.2015.09.061.
T, Ohye RG, et al. Fontan operation in the current 83. Burke RP, Jacobs JP, Ashraf MH, Aldousany A, Chang
era: a 15-year single institution experience. Ann AC. Extracardiac Fontan operation without cardiopul-
Surg. 2008;248(3):402–10. https://doi.org/10.1097/ monary bypass. Ann Thorac Surg. 1997;63(4):1175–
SLA.0b013e3181858286. 7. https://doi.org/10.1016/s0003-4975(97)00191-4.
73. Pundi KN, Johnson JN, Dearani JA, Pundi KN, Li Z, 84. McElhinney DB, Petrossian E, Reddy VM, Hanley
Hinck CA, et al. 40-year follow-up after the Fontan FL. Extracardiac conduit fontan procedure with-
operation: long-term outcomes of 1,052 patients. J out cardiopulmonary bypass. Ann Thorac Surg.
Am Coll Cardiol. 2015;66(15):1700–10. https://doi. 1998;66(5):1826–8. https://doi.org/10.1016/
org/10.1016/j.jacc.2015.07.065. s0003-4975(98)00928-x.
74. d’Udekem Y, Iyengar AJ, Galati JC, Forsdick V,
85. Backer CL, Deal BJ, Kaushal S, Russell HM,
Weintraub RG, Wheaton GR, et al. Redefining expec- Tsao S, Mavroudis C. Extracardiac versus intra-
tations of long-term survival after the Fontan proce- atrial lateral tunnel fontan: extracardiac is better.
dure: twenty-five years of follow-up from the entire Semin Thorac Cardiovasc Surg Pediatr Card Surg
population of Australia and New Zealand. Circulation. Annu. 2011;14(1):4–10. https://doi.org/10.1053/j.
2014;130(11 Suppl 1):S32–8. https://doi.org/10.1161/ pcsu.2011.01.019.
CIRCULATIONAHA.113.007764. 86.
Bradley SM. Extracardiac conduit fontan pro-
75. Gentles TL, Mayer JE Jr, Gauvreau K, Newburger JW, cedure. Oper Tech Thorac Cardiovasc Surg.
Lock JE, Kupferschmid JP, et al. Fontan operation in 2006;11(2):123–40. https://doi.org/10.1053/j.
five hundred consecutive patients: factors influencing optechstcvs.2006.03.005.
early and late outcome. J Thorac Cardiovasc Surg. 87. Thompson LD, Petrossian E, McElhinney DB,
1997;114(3):376–91. Abrikosova NA, Moore P, Reddy VM, et al. Is it nec-
76. Stamm C, Friehs I, Mayer JE, Zurakowski D,
essary to routinely fenestrate an extracardiac Fontan?
Triedman JK, Moran AM, et al. Long-term results J Am Coll Cardiol. 1999;34(2):539–44. https://doi.
of the lateral tunnel Fontan operation. J Thorac org/10.1016/s0735-1097(99)00228-4.
Cardiovasc Surg. 2001;121(1):28–41. https://doi. 88. Pretre R, Dave H, Mueller C, Kassem K, Kretschmar
org/10.1067/mtc.2001.111422. O. A new method to fenestrate the Fontan circula-
77. Tweddell JS, Nersesian M, Mussatto KA, Nugent M, tion. J Thorac Cardiovasc Surg. 2012;144(1):273–5.
Simpson P, Mitchell ME, et al. Fontan palliation in the https://doi.org/10.1016/j.jtcvs.2011.12.057.
modern era: factors impacting mortality and morbid- 89. Michel-Behnke I, Luedemann M, Bauer J, Hagel
ity. Ann Thorac Surg. 2009;88(4):1291–9. https://doi. KJ, Akintuerk H, Schranz D. Fenestration in extra-
org/10.1016/j.athoracsur.2009.05.076. cardiac conduits in children after modified Fontan
78. Brown JW, Ruzmetov M, Deschner BW, Rodefeld operation by implantation of stent grafts. Pediatr
MD, Turrentine MW. Lateral tunnel Fontan in the Cardiol. 2005;26(1):93–6. https://doi.org/10.1007/
current era: is it still a good option? Ann Thorac Surg. s00246-004-0693-6.
2010;89(2):556–62; discussion 62–3. https://doi. 90. Amin Z, Danford DA, Pedra CA. A new Amplatzer
org/10.1016/j.athoracsur.2009.10.050. device to maintain patency of Fontan fenestra-
79. Rogers LS, Glatz AC, Ravishankar C, Spray TL,
tions and atrial septal defects. Catheter Cardiovasc
Nicolson SC, Rychik J, et al. 18 years of the Interv. 2002;57(2):246–51. https://doi.org/10.1002/
Fontan operation at a single institution: results ccd.10308.
from 771 consecutive patients. J Am Coll Cardiol. 91. Salazar JD, Zafar F, Siddiqui K, Coleman RD,
2012;60(11):1018–25. https://doi.org/10.1016/j. Morales DL, Heinle JS, et al. Fenestration during
jacc.2012.05.010. Fontan palliation: now the exception instead of the
80. Stewart RD, Pasquali SK, Jacobs JP, Benjamin DK, rule. J Thorac Cardiovasc Surg. 2010;140(1):129–36.
Jaggers J, Cheng J, et al. Contemporary Fontan opera- https://doi.org/10.1016/j.jtcvs.2010.03.013.
ERRNVPHGLFRVRUJ
134 P. Sassalos and R. G. Ohye
ERRNVPHGLFRVRUJ
Part III
Gastrointestinal Controversies
ERRNVPHGLFRVRUJ
Nutritional Support
in the Pediatric ICU 8
Kimberly I. Mills and Nilesh M. Mehta
ERRNVPHGLFRVRUJ
138 K. I. Mills and N. M. Mehta
provide recommendations to guide bedside controlled trials and other pragmatic study
practice. designs become available, these unanswered
questions should coalesce into uniform evidence-
based guidelines.
Introduction
ERRNVPHGLFRVRUJ
ETIOLOGY &
ANTHROPOMETRY MECHANISM IMBALANCE OF NUTRIENTS OUTCOMES
CHRONICITY
STARVATION
Anorexia, socio-
NON-ILLNESS RELATED
Parameters economic, LOSS OF LEAN BODY
Behavioral, socioeconomic
latrogenic feeding MASS
or environmental
INTAKE
Weight, height interruptions, or
or length, skin intolerance MUSCLE WEAKNESS
folds, mid
OR
upper arm DEVELOPMENTAL
8 Nutritional Support in the Pediatric ICU
ERRNVPHGLFRVRUJ
Reference CHRONIC (≥3 months) IMBALANCE HEALING
HYPERMETABOLISM
charts e.g.: Cystic Fibrosis, Energy expenditure
Chronic lung disease, NUTRIENT REQUIREMENT PROLONGED
WHO MGRS MICRONUTRIENT HOSPITAL STAY
Cancer
(0-2 yrs) DEFICIENCIES
+/-
CDC 2000 Altered utilization
(2 – 20 yrs) of nutrients
INFLAMMATION
Fig. 8.1 ASPEN’s new definition for pediatric malnutrition in hospitalized children. (Reprinted with permission [16]. Abbreviations: WHO World Health Organization, MGRS
Multicenter Growth Reference Study, CDC Centers for Disease Control and Prevention)
139
140 K. I. Mills and N. M. Mehta
Table 8.1 Available screening tools to evaluate the presence and severity of malnutrition in pediatric patients upon
admission. Abbreviations: mo months old, hrs hours, yo years old
Screening tool Variables Population Outcome
Pediatric Subjective Food intake Pediatric patients >1 mo Weight loss >2% during
Global Nutritional Ability to eat admitted to medical or admission
Assessment (SGNA) Difficulty retaining food surgical ward for ≥48 hrs
[18] Pain
Disease severity
Pediatric Nutritional Weight and height Pediatric patients >1 mo Major/minor infectious
Risk Score (NRS) Ideal body weight and < 18 yo requiring complications
[19] BMI-for-age major elective surgery Major/minor noninfectious
MUAC complications
Triceps skinfold thickness Postoperative LOS
Mid-arm muscle area Non-prophylactic antibiotic use
Handgrip strength Unplanned reoperation
Albumin Readmission
Transferrin
Hemoglobin
Total lymphocyte count
Pediatric Yorkhill BMI Pediatric patients 1 to 16 Compare PYMS score to full
Malnutrition Score History of recent weight loss yo admitted to medical or dietitian’s assessment of
(PYMS) [20] Changes in nutritional intake surgical ward malnutrition risk
Current medical condition’s
effect on nutritional status
Screening Tool for Subjective clinical assessment Pediatric patients 1 mo to Weight-for-length/height
Risk of Impaired High-risk disease 18 yo admitted to z-score
Nutritional Status and Nutritional intake medical or surgical ward Prevalence of acute
Growth Weight loss malnutrition
(STRONGkids) [21] Hospital LOS
Screening Tool for the Diagnosis’ impact on nutrition Pediatric patients 2–17 Compare STAMP score to full
Assessment of Dietary intake yo admitted to medical or dietitian’s assessment of
Malnutrition in Weight and height surgical ward for >24 hrs malnutrition risk
Pediatrics (STAMP)
[22]
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 141
Europe [23]. The study demonstrated that the Increased counter-regulatory hormones, such as
identification and classification of malnutrition glucagon, cortisol, and epinephrine, induce insu-
risk varied across the screening tools and were lin and growth hormone resistance in response to
unable to detect a considerable portion of stress after injury, infection, surgery, or trauma
undernourished children. Based on these find- [25]. This neuroendocrine response drives the
ings, the authors recommended that none of the catabolism of endogenous protein, carbohydrate,
screening tools could be utilized in clinical and fat (Fig. 8.2) [27]. Protein catabolism is the
practice. In the absence of a validated formal sine qua non of the metabolic stress response.
screening tool, most centers rely on admission The continuous degradation and decreased syn-
weight-for-age or BMI-for-age z-scores to thesis of muscle protein, resulting in a net nega-
identify those at risk for nutritional deteriora- tive nitrogen balance, result in a large pool of
tion in the ICU. This approach is reasonable free amino acids. The free amino acids are redis-
and necessary, as clinical outcomes (i.e., rate of tributed, from visceral proteins (i.e., albumin),
infectious complications, length of stay, dura- which comprise erythrocytes, granulocytes, lym-
tion of mechanical ventilation, and mortality phocytes, and other solid tissue organs, to
rate) in the ICU have been associated with poor inflammatory response proteins (i.e., C-reactive
nutritional status at admission using various protein, fibrinogen, haptoglobin) that aid in
anthropometric measurements [4, 5, 11, 12]. wound healing and tissue repair. The remaining
Specifically, one multicenter, retrospective free amino acids are shuttled to the liver to par-
cohort study demonstrated admission BMI take in gluconeogenesis. In addition, carbohy-
z-score-predicted mortality for children receiv- drate breakdown leads to an increase in glucose
ing mechanical ventilation [13]. Although there oxidation and thus gluconeogenesis [28].
are challenges with obtaining accurate anthro- Gluconeogenesis is essential in critical illness as
pometrics upon admission to the ICU, the asso- it ensures adequate energy reserves for glucose-
ciation of malnutrition with poor clinical dependent organs such as the brain, red blood
outcomes should prioritize procurement of cells, and renal medulla. Finally, the metabolic
these measurements. stress response increases fatty acid oxidation as
The development of a validated pediatric well, providing ketones as a secondary fuel
nutrition screen specific for critically ill children source for the brain [29].
is therefore paramount for the assessment of The provision of protein, carbohydrate, and
nutritional risk in a timely and accurate manner. fat does not suppress the metabolic stress
Until an appropriate screening tool is established, response during critical illness as it does during
the development and implementation of a nutri- starvation [30, 31]. As a result, protein, carbohy-
tion support team (i.e., interdisciplinary team drate, and lipid catabolism continue despite nutri-
comprised of physicians, dietitians, nurses, and ent intake. Protein breakdown often exceeds
pharmacists with specialty training in nutrition) protein synthesis and if unmatched by adequate
in the ICU should be considered, as they have concomitant intake can result in loss of lean body
been shown to improve surveillance for those at mass and nutritional deterioration [31]. The loss
risk for malnutrition and aid in individualized of muscle mass is not isolated to skeletal muscle
nutritional prescriptions [24]. alone, but may affect cardiac and diaphragmatic
muscles resulting in cardiorespiratory insuffi-
ciency. Likewise, the provision of carbohydrate
Metabolic Stress Response does not stop gluconeogenesis but instead results
in “stress hyperglycemia” [32]. Finally, increased
A basic understanding of the metabolic stress lipid demand in the setting of limited fat stores
response can assist in the accurate assessment of and inadequate provision can lead to essential
energy expenditure and help tailor individual- fatty acid deficiency, especially in preterm infants
ized nutritional prescriptions in the critically ill. [33, 34].
ERRNVPHGLFRVRUJ
142 K. I. Mills and N. M. Mehta
Acute inflammatory
Loss of lean
Lipolysis Proteins
body mass
↑Fatty
Trauma Acids
Protein synthesis
Sepsis
Critical Muscle AMINO
illness break down ACIDS
Gluconeogenesis
Burn
Urea
Surgery
Fig. 8.2 Pathways of the metabolic stress response during critical illness. (Reprinted with permission [26])
Determining Energy Requirements late the respiratory quotient (RQ), which is calcu-
lated as RQ = VCO2/VO2. RQ values range from
The metabolic state during critical illness is 0.6 to 1.4 based on the type of substrate utilized
dynamic and unpredictable, ranging from hypo- by the patient. Carbohydrate oxidation results in
metabolism (<90% of predicted measured resting higher carbon dioxide production and therefore
energy expenditure) as a result of sedation, higher RQ, whereas lipolysis is associated with
mechanical ventilation, and targeted temperature comparatively lower VCO2 measurements and
management to hypermetabolism (>110% of pre- hence a lower RQ. Mixed fuel utilization results
dicted measured resting energy expenditure) as in typical RQ ranging from 0.8 to 1.2. Although
seen in severe burn injuries [35–39]. Inaccurate carbohydrate excess may increase the RQ value,
energy estimates can result in underfeeding or the use of RQ as a measure of overfeeding is not
overfeeding with potential negative clinical con- recommended [49]. IC has several limitations as
sequences [26, 40–43]. Underfeeding can lead to it is not reliable in children that weigh less than
poor wound healing, impaired oxygen utilization, 5 kg, those supported with an inspired O2 concen-
increased infection risk, poor neurodevelopmental tration greater than 60%, or in patients with a
outcomes, and increased mortality, while over- sizeable air leak (i.e., around endotracheal tube,
feeding can result in hypertriglyceridemia, hyper- chest tube).
glycemia, hepatic steatosis and cholestasis, Though IC is deemed the gold standard for
increased carbon dioxide production, and uremia measuring energy expenditure in critically ill
[44–46]. children, the majority of ICUs lack the resources
Indirect calorimetry (IC) remains the gold and expertise to operationalize IC in their daily
standard and current ASPEN/SCCM guideline clinical management [50–53]. When IC is not
recommendation to measure resting energy available and despite substantial evidence against
expenditure in critically ill children [43, 47, 48]. their accuracy, clinicians utilize predictive equa-
IC, which is typically performed using a meta- tions based on patient demographics to estimate
bolic cart, measures oxygen consumption (VO2) resting energy expenditure (Table 8.2) [39, 43,
and carbon dioxide production (VCO2) to calcu- 52–57]. If predictive equations are utilized, the
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 143
ERRNVPHGLFRVRUJ
144 K. I. Mills and N. M. Mehta
Fig. 8.3 Relation 15
between enteral protein 14
intake adequacy and Theoretical Curve (Logistic Regression)
13
60-day mortality in 95% Confidence Interval
mechanically ventilated 12 Likelihood ratio test = 9.16, P = 0.002
children (n = 1245). 11
Probability of Mortality (%)
(Reproduced with 10
permission [6])
9
8
7
6
5
4
3
2
1
0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Protein Adequacy (%)
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 145
[75]. These findings were similar to those result in a delay in initiation of EN [85–87].
represented in a secondary analysis from a simi- However, postpyloric feeds may be beneficial in
lar adult randomized controlled trial [76]. Critical patients who suffer from feeding intolerance and
review of the study cautions against a change in are at risk for aspiration [88, 89]. One random-
daily clinical practice however, as the study was ized controlled trial demonstrated reduced gas-
observational in nature and not developed as a tric residual volumes (GRVs) in patients who
dosing study, unique clinical outcomes were were fed postpyloric compared to gastric,
developed as primary outcome measures, refer- although two randomized controlled trials have
ence macronutrient doses used in the study were not demonstrated a reduction in the rate of aspi-
higher than recommended by the ASPEN/SCCM ration [85, 86].
guidelines, and there was no examination of the Another consideration when initiating EN is
interaction between different macronutrient lev- to whether to begin with continuous versus inter-
els. Hence, until further studies are available to mittent feeds. Existing data is currently conflict-
clarify the conflicting data, use of the ASPEN/ ing and insufficient for the ASPEN/SCCM
SCCM guidelines for protein delivery in criti- guidelines to recommend one practice as opposed
cally ill patients is appropriate. to the other. The only evidence currently avail-
able consists of two randomized controlled trials
that demonstrated no difference in EN tolerance
etermining the Delivery Route
D between continuous and intermittent feeds [90,
of Nutrition 91]. Based on these data, the delivery method for
enteral nutrition can be determined by provider
Enteral Nutrition preference.
Once EN is initiated, maintenance of EN
Enteral nutrition (EN) is the preferred mode of remains challenging, as interruptions are com-
nutrient delivery in critically ill children. mon [92, 93]. Barriers to optimal EN include
Regardless of most diagnoses, sedative, and delayed initiation, mechanical feeding tube
vasoactive use, EN has been shown to be safe and issues, perceived feeding intolerance, noninva-
beneficial [77, 78]. As timing of EN initiation has sive positive-pressure ventilation use, and pro-
been associated with nutritional adequacy, initia- longed fasting around procedures including
tion of EN within 24–48 h of ICU admission, intubation and extubation (Fig. 8.4) [81, 92, 94].
known as “early EN,” is preferred according to A prospective cohort study found that over half
the ASPEN/SCCM guidelines [6, 79–81]. of the interruptions to EN in the PICU were
Furthermore, achieving two thirds of the pre- avoidable [92]. These avoidable interruptions
scribed energy and protein goals via EN within were associated with a threefold increase in par-
the first week of critical illness may be associated enteral nutrition (PN) use and a significant delay
with improved clinical outcomes [3, 6]. Early EN in achieving the prescription goal; thus an effort
has demonstrated a lower risk of infection, to minimize interruptions is of paramount impor-
reduced LOS, improved anthropometrics, and tance. Methods to minimize avoidable interrup-
improved survival when compared to EN initi- tions include careful consideration regarding
ated later [61, 79, 82–84]. timing of procedures, guideline development and
When initiating EN, the question remains adherence around duration of fasting, and a dedi-
whether to begin with gastric or postpyloric cated team of nurses and support from interven-
feeds. Currently, initiating feeds via the gastric tional radiology to assist in the successful and
route is preferred and physiologic; however there expedient placement of feeding tubes.
is no evidence to support this recommendation Once EN is initiated, there is no uniform
from the ASPEN/SCCM guidelines. Considering method to advancing EN. A stepwise algorithmic
postpyloric feeds requires available technical approach to advancing EN in the ICU has been
expertise in placing the feeding tube and may shown to improve time to goal prescription,
ERRNVPHGLFRVRUJ
146 K. I. Mills and N. M. Mehta
0 10 20 30 40 50
Percentage of all EN interruptions
Fig. 8.4 Reasons for interruptions to enteral nutrition, both avoidable and unavoidable [92]. (Reproduced with
permission)
increase the percent of patients who achieve their The benefits of EN have been demonstrated in
prescription goal, reduce interruptions to nutri- both human and animal studies. Gastrointestinal
tion, decrease PN use, and improve nutritional mucosal integrity and motility may improve
and clinical outcomes (Fig. 8.5) [80, 95–98]. when EN is prescribed [102]. These beneficial
Devising an algorithm for use in the ICU should effects of EN are likely related to engaged gut-
provide guidance on detecting and managing associated lymphoid tissue (GALT), mucosal
intolerance to ensure appropriate and expedient immunity, and improved gastrointestinal blood
EN advancement [50]. flow [103–106]. Additional studies are required
Perceived feeding intolerance is one of the pri- to further understand the benefits of providing
mary reasons for interrupting EN. Currently, early EN. Universally advanced and clearer step-
feeding intolerance lacks a uniform description wise algorithms need to be developed and should
and could possibly refer to gastroesophageal be supported by evidence considering gastric ver-
reflux, vomiting, constipation, diarrhea, or mal- sus postpyloric, continuous versus intermittent,
absorption. Traditionally, gastric residual volume and methods to obviate interruptions to EN.
(GRV) was used to define feeding intolerance;
however its accuracy has been questioned, and it
is no longer recommended in adult ICUs [73, Parenteral Nutrition
99–101]. As there are no comparable pediatric
studies to support this move, the use of GRVs is When enteral nutrition fails, parenteral nutrition
cautiously recommended in the most recent (PN) is advised [15, 107]. In addition, when EN
ASPEN/SCCM guidelines [15]. Despite a lack of is not feasible or contraindicated, such as follow-
definitive data in pediatrics, many centers use ing major abdominal surgery, when there are
prokinetic agents (i.e., erythromycin, metoclo- concerns for intestinal ischemia or in a low car-
pramide), antiemetics, acid suppression, antidiar- diac output state, PN should be considered.
rheals, and laxatives as adjuncts to EN. Furthermore, if a patient is severely malnour-
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 147
Is patient able to
meet nutrition
goals orally
Yes No
Exit Algorithm
Is patient able to
be fed enterally?
Yes No
Yes No
--Record baseline AG
--Record baseline abdominal girth (AG)
--GRV is measured before initiation and at each
--Gastric residual volume (GRV) is not measured
advancement step
AFTER 4 HOURS
Measured GRV and assess for signs of intolerance
Does patient
have GRV > 3ml/kg or
evidence of EN
intolerance?
Yes No
Yes No
Reassess after 1
hour for signs of
intolerance - Review energy and protein adequacy Consider the following:
- Consider increasing density of formula - Promotility agent
- Monitor weight - Post-pyloric feeds (if Gastric fed)
- Consider Indirect Calorimetry - If PN is indicated
Does patient - Implement Bowel Management Guideline - Implement Bowel Management Guideline
still have signs of EN No - Monitor for signs of overfeeding - Anti-diarrheal agents
intolerance or GRV >
3ml/kg?
Yes
Fig. 8.5 Example of a stepwise algorithm for initiating and advancing enteral nutrition [95]. (Reproduced with
permission)
ERRNVPHGLFRVRUJ
148 K. I. Mills and N. M. Mehta
ished, at high risk for nutritional deterioration Following that recommendation, the timing of
during their hospitalization (i.e., severe burn supplemental PN needs to be made on an indi-
injury), or a neonate (<30 days old) and not able vidualized basis and should take in consideration
to achieve energy and protein goals via EN, PN the nutritional and clinical status of the patient.
should be initiated. The macronutrient composition of PN and
The optimal timing for initiation of PN particularly the alternative lipid emulsions are
remains controversial. Adult studies have being extensively studied. Recommendations for
reported a potential benefit when PN is initiated protein intake mirror the current enteral recom-
after day 3 if nutritional goals are not met by EN mendations, although further research into the
but inferior clinical outcomes if PN is initiated route of protein supplementation and its effects
earlier [108–111]. Prior to the publication of the on clinical outcomes is needed [15]. With the
PEPaNIC trial, there was a dearth of randomized, recent Food and Drug Administration’s (FDA)
controlled trials addressing the effects of PN on approval of alternative lipid formulations, emerg-
clinical outcomes in children [112, 113]. The ing literature has indicated benefits in utilizing
PEPaNIC trial was a three-center trial in criti- olive oil- and fish oil-based lipids. Non-soy-based
cally ill children who were randomized to receive lipid formulations have demonstrated a trend
either an early (<24 h) or late (>7 days) PN strat- toward improved survival, shorter duration of
egy. The study demonstrated improved outcomes mechanical ventilation, and ICU length of stay
in the children who received the late PN strategy, [116, 117]. These clinical improvements are
specifically by lowering the rate of new infec- thought to be secondary to higher antioxidant
tions, decreasing ICU length of stay, shortening content, immune modulating, and less inflamma-
the duration of mechanical ventilation, and tory properties [118]. As an additional benefit,
decreasing renal replacement therapy utilization. these lipid formulations have been shown to
Several issues regarding the study methods need reduce the incidence and possibly reverse
to be reviewed: the portion of the calories that PN-associated liver disease in patients with short
were provided via PN was small; energy goals gut syndrome and PN dependence [119].
were calculated by predictive equations, putting Additional studies are required to determine
the subjects at risk for overfeeding; and subjects the optimal timing for PN initiation and the role
at risk for malnutrition were treated similarly to of supplemental PN for critically ill children in
those well-nourished and identified using the general. Ongoing research regarding the potential
STRONGkids screening tool, which has not been benefits of alternative lipid formulations may
previously validated in the ICU population [114]. lead to a uniform recommendation in the future.
Hence, the current ASPEN/SCCM guidelines
recommend exercising caution when applying
these results broadly in clinical practice, particu- Role of Micronutrients
larly in vulnerable newborns and severely mal- as Immunonutrition
nourished children [15]. Furthermore, the recent
publication of NUTRIREA-2, a study examining The role of micronutrients as immunomodulators
the safety of early enteral versus parenteral nutri- in critically ill patients surfaced as an area of
tion in mechanically ventilated adults with shock, research over a decade ago. Micronutrients and
demonstrated no difference in hospital-acquired antioxidants were hypothesized to diminish
infections among the two groups and not surpris- inflammation or replete nutrients depleted by
ingly demonstrated an increased rate of feeding stress. Glutamine, arginine, selenium, copper,
intolerance in the enteral group [115]. In sum- and zinc are a few of the studied micronutrients
mary, the ASPEN/SCCM guidelines advise to date. Several randomized controlled trials
against initiating PN within the first 24 h of comparing various forms of immunonutrition
admission and to consider a delayed PN approach have been undertaken and have yet to demon-
in children who are not severely malnourished. strate any clinical benefit [120–123]. Furthermore,
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 149
ERRNVPHGLFRVRUJ
150 K. I. Mills and N. M. Mehta
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 151
identify children at risk of malnutrition. Am J Clin ated with increased variability. Crit Care Med.
Nutr. 2000;72:64–70. 2000;28:2655–6.
19. Secker DJ, Jeejeebhoy KN. Subjective global nutri- 34. Coss-Bu JA, Klish WJ, Walding D, Stein F, Smith
tional assessment for children. Am J Clin Nutr. EO, Jefferson LS. Energy metabolism, nitrogen bal-
2007;85:1083–9. ance, and substrate utilization in critically ill chil-
20. Gerasimidis K, Keane O, Macleod I, Flynn DM, dren. Am J Clin Nutr. 2001;74:664–9.
Wright CM. A four-stage evaluation of the paediat- 35. Duggan C, Bechard L, Donovan K, et al. Changes in
ric yorkhill malnutrition score in a tertiary paediatric resting energy expenditure among children undergo-
hospital and a district general hospital. Br J Nutr. ing allogeneic stem cell transplantation. Am J Clin
2010;104:751–6. Nutr. 2003;78:104–9.
21. Hulst JM, Zwart H, Hop WC, Joosten KF. Dutch 36. Li J, Zhang G, Herridge J, et al. Energy expendi-
national survey to test the STRONGkids nutritional ture and caloric and protein intake in infants follow-
risk screening tool in hospitalized children. Clin ing the Norwood procedure. Pediat Crit Care Med.
Nutr (Edinburgh). 2010;29:106–11. 2008;9:55–61.
22. McCarthy H, Dixon M, Crabtree I, Eaton-Evans MJ, 37. Mehta NM, Costello JM, Bechard LJ, et al. Resting
McNulty H. The development and evaluation of the energy expenditure after Fontan surgery in children
Screening Tool for the Assessment of Malnutrition with single-ventricle heart defects. JPEN J Parenter
in Paediatrics (STAMP(c)) for use by healthcare Enteral Nutr. 2012;36:685–92.
staff. J Hum Nutr Diet. 2012;25:311–8. 38. Suman OE, Mlcak RP, Chinkes DL, Herndon
23. Chourdakis M, Hecht C, Gerasimidis K, et al. DN. Resting energy expenditure in severely burned
Malnutrition risk in hospitalized children: use of 3 children: analysis of agreement between indirect cal-
screening tools in a large European population. Am J orimetry and prediction equations using the Bland-
Clin Nutr. 2016;103:1301–10. Altman method. Burns. 2006;32:335–42.
24. Gurgueira GL, Leite HP, Taddei JA, de Carvalho 39. Coss-Bu JA, Jefferson LS, Walding D, David Y,
WB. Outcomes in a pediatric intensive care unit Smith EO, Klish WJ. Resting energy expenditure
before and after the implementation of a nutri- and nitrogen balance in critically ill pediatric patients
tion support team. JPEN J Parenter Enteral Nutr. on mechanical ventilation. Nutrition (Burbank).
2005;29:176–85. 1998;14:649–52.
25. de Groof F, Joosten KF, Janssen JA, et al. Acute 40. Faisy C, Lerolle N, Dachraoui F, et al. Impact of
stress response in children with meningococcal sep- energy deficit calculated by a predictive method
sis: important differences in the growth hormone/ on outcome in medical patients requiring pro-
insulin-like growth factor I axis between nonsur- longed acute mechanical ventilation. Br J Nutr.
vivors and survivors. J Clin Endocrinol Metab. 2009;101:1079–87.
2002;87:3118–24. 41. Klein CJ, Stanek GS, Wiles CE 3rd. Overfeeding
26. Mehta NM, Duggan CP. Nutritional deficien- macronutrients to critically ill adults: metabolic
cies during critical illness. Pediatr Clin N Am. complications. J Am Diet Assoc. 1998;98:795–806.
2009;56:1143–60. 42. Mehta NM, Bechard LJ, Dolan M, Ariagno
27. Weissman C. The metabolic response to stress: K, Jiang H, Duggan C. Energy imbalance and
an overview and update. Anesthesiology. the risk of overfeeding in critically ill children.
1990;73:308–27. Pediatr Crit Care Med. 2011;12:398–405.
28. Forchielli ML, McColl R, Walker WA, Lo 43. Mehta NM, Bechard LJ, Leavitt K, Duggan
C. Children with congenital heart disease: a nutri- C. Cumulative energy imbalance in the pedi-
tion challenge. Nutr Rev. 1994;52:348–53. atric intensive care unit: role of targeted indi-
29. Tilden SJ, Watkins S, Tong TK, Jeevanandam rect calorimetry. JPEN J Parenter Enteral Nutr.
M. Measured energy expenditure in pediatric 2009;33:336–44.
intensive care patients. Am J Dis Child (1960). 44. Askanazi J, Rosenbaum SH, Hyman AI, Silverberg
1989;143:490–2. PA, Milic-Emili J, Kinney JM. Respiratory changes
30. Parekh NR, Steiger E. Percentage of weight loss induced by the large glucose loads of total parenteral
as a predictor of surgical risk: from the time nutrition. JAMA. 1980;243:1444–7.
of Hiram Studley to today. Nutr Clin Pract. 45. Grohskopf LA, Sinkowitz-Cochran RL, Garrett DO,
2004;19:471–6. et al. A national point-prevalence survey of pediatric
31. Hulst JM, Joosten KF, Tibboel D, van Goudoever intensive care unit-acquired infections in the United
JB. Causes and consequences of inadequate sub- States. J Pediatr. 2002;140:432–8.
strate supply to pediatric ICU patients. Curr Opin 46. MacIntyre NR, Cook DJ, Ely EW Jr, et al. Evidence-
Clin Nutr Metab Care. 2006;9:297–303. based guidelines for weaning and discontinuing
32. Long CL, Kinney JM, Geiger JW. Nonsuppressability ventilatory support: a collective task force facilitated
of gluconeogenesis by glucose in septic patients. by the American College of Chest Physicians; the
Metab Clin Exp. 1976;25:193–201. American Association for Respiratory Care; and the
33. Chwals WJ, Bistrian BR. Predicted energy expen- American College of Critical Care Medicine. Chest.
diture in critically ill children: problems associ- 2001;120:375s–95s.
ERRNVPHGLFRVRUJ
152 K. I. Mills and N. M. Mehta
47. Dokken M, Rustoen T, Stubhaug A. Indirect calo- mine energy expenditure in ventilated critically ill
rimetry reveals that better monitoring of nutrition children. Clin Nutr (Edinburgh). 2017;36:452–7.
therapy in pediatric intensive care is needed. JPEN 61. Wong JJ, Han WM, Sultana R, Loh TF, Lee
J Parenter Enteral Nutr. 2015;39:344–52. JH. Nutrition delivery affects outcomes in pediatric
48. Jotterand Chaparro C, Laure Depeyre J, Longchamp acute respiratory distress syndrome. JPEN J Parenter
D, Perez MH, Taffe P, Cotting J. How much protein Enteral Nutr. 2017;41:1007–13.
and energy are needed to equilibrate nitrogen and 62. de Neef M, Geukers VG, Dral A, Lindeboom R,
energy balances in ventilated critically ill children? Sauerwein HP, Bos AP. Nutritional goals, prescrip-
Clin Nutr (Edinburgh). 2016;35:460–7. tion and delivery in a pediatric intensive care unit.
49. Guenst JM, Nelson LD. Predictors of total par- Clin Nutr (Edinburgh). 2008;27:65–71.
enteral nutrition-induced lipogenesis. Chest. 63. Taylor RM, Preedy VR, Baker AJ, Grimble
1994;105:553–9. G. Nutritional support in critically ill children. Clin
50. Martinez EE, Bechard LJ, Mehta NM. Nutrition Nutr (Edinburgh). 2003;22:365–9.
algorithms and bedside nutrient delivery prac- 64. Mtaweh H, Smith R, Kochanek PM, et al. Energy
tices in pediatric intensive care units: an interna- expenditure in children after severe traumatic brain
tional multicenter cohort study. Nutr Clin Pract. injury. Pediat Crit Care Med. 2014;15:242–9.
2014;29:360–7. 65. Bechard LJ, Feldman HA, Venick R, et al.
51. Kyle UG, Arriaza A, Esposito M, Coss-Bu JA. Is Attenuation of resting energy expenditure follow-
indirect calorimetry a necessity or a luxury in the ing hematopoietic SCT in children. Bone Marrow
pediatric intensive care unit? JPEN J Parenter Transplant. 2012;47:1301–6.
Enteral Nutr. 2012;36:177–82. 66. Geukers VG, Dijsselhof ME, Jansen NJ, et al. The
52. van der Kuip M, Oosterveld MJ, van Bokhorst-de effect of short-term high versus normal protein
van der Schueren MA, de Meer K, Lafeber HN, intake on whole-body protein synthesis and balance
Gemke RJ. Nutritional support in 111 pediatric in children following cardiac surgery: a random-
intensive care units: a European survey. Intensive ized double-blind controlled clinical trial. Nutr J.
Care Med. 2004;30:1807–13. 2015;14:72.
53. Sion-Sarid R, Cohen J, Houri Z, Singer P. Indirect 67. de Betue CT, van Waardenburg DA, Deutz NE, et al.
calorimetry: a guide for optimizing nutritional sup- Increased protein-energy intake promotes anabo-
port in the critically ill child. Nutrition (Burbank). lism in critically ill infants with viral bronchiolitis:
2013;29:1094–9. a double-blind randomised controlled trial. Arch Dis
54. Hardy CM, Dwyer J, Snelling LK, Dallal GE, Child. 2011;96:817–22.
Adelson JW. Pitfalls in predicting resting energy 68. de Betue CT, Joosten KF, Deutz NE, Vreugdenhil
requirements in critically ill children: a comparison AC, van Waardenburg DA. Arginine appearance and
of predictive methods to indirect calorimetry. Nutr nitric oxide synthesis in critically ill infants can be
Clin Pract. 2002;17:182–9. increased with a protein-energy-enriched enteral for-
55. Meyer R, Kulinskaya E, Briassoulis G, et al. The mula. Am J Clin Nutr. 2013;98:907–16.
challenge of developing a new predictive formula to 69. Verbruggen SC, Coss-Bu J, Wu M, et al. Current
estimate energy requirements in ventilated critically recommended parenteral protein intakes do not sup-
ill children. Nutr Clin Pract. 2012;27:669–76. port protein synthesis in critically ill septic, insulin-
56. Framson CM, LeLeiko NS, Dallal GE, Roubenoff resistant adolescents with tight glucose control. Crit
R, Snelling LK, Dwyer JT. Energy expenditure Care Med. 2011;39:2518–25.
in critically ill children. Pediat Crit Care Med. 70. van Waardenburg DA, de Betue CT, Goudoever JB,
2007;8:264–7. Zimmermann LJ, Joosten KF. Critically ill infants
57. White MS, Shepherd RW, McEniery JA. Energy benefit from early administration of protein and
expenditure in 100 ventilated, critically ill children: energy-enriched formula: a randomized controlled
improving the accuracy of predictive equations. Crit trial. Clin Nutr (Edinburgh). 2009;28:249–55.
Care Med. 2000;28:2307–12. 71. Mehta NM, Compher C. A.S.P.E.N. Clinical guide-
58. Picolo MF, Lago AF, Menegueti MG, et al. Harris- lines: nutrition support of the critically ill child.
benedict equation and resting energy expenditure JPEN J Parenter Enteral Nutr. 2009;33:260–76.
estimates in critically Ill ventilator patients. Am J 72. Botran M, Lopez-Herce J, Mencia S, Urbano J,
Crit Care. 2016;25:e21–9. Solana MJ, Garcia A. Enteral nutrition in the criti-
59. Mehta NM, Smallwood CD, Joosten KF, Hulst JM, cally ill child: comparison of standard and protein-
Tasker RC, Duggan CP. Accuracy of a simplified enriched diets. J Pediatr. 2011;159:27–32.e1.
equation for energy expenditure based on bedside 73. McClave SA, Taylor BE, Martindale RG, et al.
volumetric carbon dioxide elimination measure- Guidelines for the provision and assessment of
ment – a two-center study. Clin Nutr (Edinburgh). nutrition support therapy in the adult critically Ill
2015;34:151–5. patient: Society of Critical Care Medicine (SCCM)
60. Kerklaan D, Augustus ME, Hulst JM, van Rosmalen and American Society for Parenteral and Enteral
J, Verbruggen S, Joosten KFM. Validation of Nutrition (A.S.P.E.N.). JPEN J Parenter Enteral
ventilator-derived VCO2 measurements to deter- Nutr. 2016;40:159–211.
ERRNVPHGLFRVRUJ
8 Nutritional Support in the Pediatric ICU 153
74. Moreno YMF, Hauschild DB, Martins MD, Bechard 88. Lopez-Herce J, Mencia S, Sanchez C, Santiago MJ,
LJ, Mehta NM. Feasibility of Enteral Protein Bustinza A, Vigil D. Postpyloric enteral nutrition
Supplementation in Critically Ill Children. JPEN J in the critically ill child with shock: a prospective
Parenter Enteral Nutr. 2018;42:61–70. observational study. Nutr J. 2008;7:6.
75. Vanhorebeek I, Verbruggen S, Casaer MP, et al. 89. Lopez-Herce J, Sanchez C, Carrillo A, et al.
Effect of early supplemental parenteral nutrition in Transpyloric enteral nutrition in the critically
the paediatric ICU: a preplanned observational study ill child with renal failure. Intensive Care Med.
of post-randomisation treatments in the PEPaNIC 2006;32:1599–605.
trial. Lancet Respir Med. 2017;5:475–83. 90. Horn D, Chaboyer W, Schluter PJ. Gastric resid-
76. Casaer MP, Wilmer A, Hermans G, Wouters PJ, ual volumes in critically ill paediatric patients: a
Mesotten D, Van den Berghe G. Role of disease and comparison of feeding regimens. Aust Crit Care.
macronutrient dose in the randomized controlled 2004;17:98–100, 2–3.
EPaNIC trial: a post hoc analysis. Am J Respir Crit 91. Horn D, Chaboyer W. Gastric feeding in critically
Care Med. 2013;187:247–55. ill children: a randomized controlled trial. Am J Crit
77. Panchal AK, Manzi J, Connolly S, et al. Safety of Care. 2003;12:461–8.
enteral feedings in critically Ill children receiving 92. Mehta NM, McAleer D, Hamilton S, et al.
vasoactive agents. JPEN J Parenter Enteral Nutr. Challenges to optimal enteral nutrition in a multi-
2016;40:236–41. disciplinary pediatric intensive care unit. JPEN J
78. King W, Petrillo T, Pettignano R. Enteral nutrition Parenter Enteral Nutr. 2010;34:38–45.
and cardiovascular medications in the pediatric 93. Rogers EJ, Gilbertson HR, Heine RG, Henning
intensive care unit. JPEN J Parenter Enteral Nutr. R. Barriers to adequate nutrition in critically ill chil-
2004;28:334–8. dren. Nutrition (Burbank). 2003;19:865–8.
79. Mikhailov TA, Kuhn EM, Manzi J, et al. Early 94. Leong AY, Cartwright KR, Guerra GG, Joffe AR,
enteral nutrition is associated with lower mortality in Mazurak VC, Larsen BM. A Canadian survey of
critically ill children. JPEN J Parenter Enteral Nutr. perceived barriers to initiation and continuation of
2014;38:459–66. enteral feeding in PICUs. Pediat Crit Care Med.
80. Petrillo-Albarano T, Pettignano R, Asfaw M, 2014;15:e49–55.
Easley K. Use of a feeding protocol to improve 95. Hamilton S, McAleer DM, Ariagno K, et al. A step-
nutritional support through early, aggressive, wise enteral nutrition algorithm for critically ill chil-
enteral nutrition in the pediatric intensive care dren helps achieve nutrient delivery goals*. Pediat
unit. Pediatr Crit Care Med. 2006;7:340–4. Crit Care Med. 2014;15:583–9.
81. Canarie MF, Barry S, Carroll CL, et al. Risk factors 96. Yoshimura S, Miyazu M, Yoshizawa S, et al.
for delayed enteral nutrition in critically Ill children. Efficacy of an enteral feeding protocol for providing
Pediat Crit Care Med. 2015;16:e283–9. nutritional support after paediatric cardiac surgery.
82. Marik PE, Zaloga GP. Early enteral nutrition in Anaesth Intensive Care. 2015;43:587–93.
acutely ill patients: a systematic review. Crit Care 97. Briassoulis GC, Zavras NJ, Hatzis MT. Effectiveness
Med. 2001;29:2264–70. and safety of a protocol for promotion of early intra-
83. Heighes PT, Doig GS, Sweetman EA, Simpson gastric feeding in critically ill children. Pediatr Crit
F. An overview of evidence from systematic reviews Care Med. 2001;2:113–21.
evaluating early enteral nutrition in critically ill 98. Meyer R, Harrison S, Sargent S, Ramnarayan
patients: more convincing evidence is needed. P, Habibi P, Labadarios D. The impact of
Anaesth Intensive Care. 2010;38:167–74. enteral feeding protocols on nutritional sup-
84. Khorasani EN, Mansouri F. Effect of early enteral port in critically ill children. J Hum Nutr Diet.
nutrition on morbidity and mortality in children with 2009;22:428–36.
burns. Burns. 2010;36:1067–71. 99. Martinez EE, Pereira LM, Gura K, et al. Gastric
85. Kamat P, Favaloro-Sabatier J, Rogers K, Stockwell emptying in critically Ill children. JPEN J Parenter
JA. Use of methylene blue spectrophotometry to Enteral Nutr. 2017;41:1100–9.
detect subclinical aspiration in enterally fed intu- 100. Elke G, Felbinger TW, Heyland DK. Gastric residual
bated pediatric patients. Pediat Crit Care Med. volume in critically ill patients: a dead marker or still
2008;9:299–303. alive? Nutr Clin Pract. 2015;30:59–71.
86. Meert KL, Daphtary KM, Metheny NA. Gastric vs 101. Ozen N, Tosun N, Yamanel L, Altintas ND, Kilciler
small-bowel feeding in critically ill children receiv- G, Ozen V. Evaluation of the effect on patient param-
ing mechanical ventilation: a randomized controlled eters of not monitoring gastric residual volume in
trial. Chest. 2004;126:872–8. intensive care patients on a mechanical ventilator
87. Montejo JC, Grau T, Acosta J, et al. Multicenter, receiving enteral feeding: a randomized clinical trial.
prospective, randomized, single-blind study com- J Crit Care. 2016;33:137–44.
paring the efficacy and gastrointestinal complica- 102. Fukatsu K, Zarzaur BL, Johnson CD, Lundberg AH,
tions of early jejunal feeding with early gastric Wilcox HG, Kudsk KA. Enteral nutrition prevents
feeding in critically ill patients. Crit Care Med. remote organ injury and death after a gut ischemic
2002;30:796–800. insult. Ann Surg. 2001;233:660–8.
ERRNVPHGLFRVRUJ
154 K. I. Mills and N. M. Mehta
103. Ikeda S, Kudsk KA, Fukatsu K, et al. Enteral feed- with shock: a randomised, controlled, multicentre,
ing preserves mucosal immunity despite in vivo open-label, parallel-group study (NUTRIREA-2).
MAdCAM-1 blockade of lymphocyte homing. Ann Lancet (London). 2018;391:133–43.
Surg. 2003;237:677–85; discussion 85. 116. Manzanares W, Dhaliwal R, Jurewitsch B, Stapleton
104. Kudsk KA, Stone JM, Carpenter G, Sheldon RD, Jeejeebhoy KN, Heyland DK. Alternative
GF. Enteral and parenteral feeding influences mor- lipid emulsions in the critically ill: a system-
tality after hemoglobin-E. coli peritonitis in normal atic review of the evidence. Intensive Care Med.
rats. J Trauma. 1983;23:605–9. 2013;39:1683–94.
105. Li J, Kudsk KA, Gocinski B, Dent D, Glezer J, 117. Nehra D, Fallon EM, Potemkin AK, et al. A compar-
Langkamp-Henken B. Effects of parenteral and ison of 2 intravenous lipid emulsions: interim analy-
enteral nutrition on gut-associated lymphoid tissue. sis of a randomized controlled trial. JPEN J Parenter
J Trauma. 1995;39:44–51; discussion −2. Enteral Nutr. 2014;38:693–701.
106. Sano Y, Gomez FE, Kang W, et al. Intestinal poly- 118. Larsen BM, Field CJ, Leong AY, et al. Pretreatment
meric immunoglobulin receptor is affected by type with an intravenous lipid emulsion increases plasma
and route of nutrition. JPEN J Parenter Enteral Nutr. eicosapentanoic acid and downregulates leukotriene
2007;31:351–6; discussion 6–7. b4, procalcitonin, and lymphocyte concentrations
107. Koletzko B, Goulet O, Hunt J, Krohn K, Shamir after open heart surgery in infants. JPEN J Parenter
R. 1. Guidelines on Paediatric Parenteral Enteral Nutr. 2015;39:171–9.
Nutrition of the European Society of Paediatric 119. Hojsak I, Colomb V, Braegger C, et al. ESPGHAN
Gastroenterology, Hepatology and Nutrition Committee on Nutrition Position Paper. Intravenous
(ESPGHAN) and the European Society for Lipid Emulsions and Risk of Hepatotoxicity in
Clinical Nutrition and Metabolism (ESPEN), Infants and Children: a Systematic Review and
Supported by the European Society of Paediatric Meta-analysis. J Pediatr Gastroenterol Nutr.
Research (ESPR). J Pediatr Gastroenterol Nutr. 2016;62:776–92.
2005;41(Suppl 2):S1–S87. 120. Jacobs BR, Nadkarni V, Goldstein B, et al.
108. Casaer MP, Mesotten D, Hermans G, et al. Early ver- Nutritional immunomodulation in critically ill chil-
sus late parenteral nutrition in critically ill adults. N dren with acute lung injury: feasibility and impact
Engl J Med. 2011;365:506–17. on circulating biomarkers. Pediatr Crit Care Med.
109. Singer P, Anbar R, Cohen J, et al. The tight calo- 2013;14:e45–56.
rie control study (TICACOS): a prospective, ran- 121. Mayes T, Gottschlich MM, Kagan RJ. An evaluation
domized, controlled pilot study of nutritional of the safety and efficacy of an anti-inflammatory,
support in critically ill patients. Intensive Care Med. pulmonary enteral formula in the treatment of pedi-
2011;37:601–9. atric burn patients with respiratory failure. J Burn
110. Heidegger CP, Berger MM, Graf S, et al. Care Res. 2008;29:82–8.
Optimisation of energy provision with supplemental 122. Briassoulis G, Filippou O, Hatzi E, Papassotiriou I,
parenteral nutrition in critically ill patients: a ran- Hatzis T. Early enteral administration of immunonu-
domised controlled clinical trial. Lancet (London). trition in critically ill children: results of a blinded
2013;381:385–93. randomized controlled clinical trial. Nutrition
111. Doig GS, Simpson F. Early parenteral nutrition in (Burbank). 2005;21:799–807.
critically ill patients with short-term relative contra- 123. Carcillo JA, Dean JM, Holubkov R, et al. The ran-
indications to early enteral nutrition: a full economic domized comparative pediatric critical illness stress-
analysis of a multicenter randomized controlled induced immune suppression (CRISIS) prevention
trial based on US costs. ClinicoEcon Outcome Res. trial. Pediatr Crit Care Med. 2012;13:165–73.
2013;5:369–79. 124. Wischmeyer PE, Dhaliwal R, McCall M, Ziegler
112. Fivez T, Kerklaan D, Mesotten D, Verbruggen TR, Heyland DK. Parenteral glutamine supplemen-
S, Joosten K, Van den Berghe G. Evidence for tation in critical illness: a systematic review. Crit
the use of parenteral nutrition in the pediat- Care (London). 2014;18:R76.
ric intensive care unit. Clin Nutr (Edinburgh). 125. Heyland D, Muscedere J, Wischmeyer PE,
2017;36:218–23. et al. A randomized trial of glutamine and anti-
113. Fivez T, Kerklaan D, Mesotten D, et al. Early oxidants in critically ill patients. N Engl J Med.
versus Late Parenteral Nutrition in Critically Ill 2013;368:1489–97.
Children. N Engl J Med. 2016;374:1111–22. 126. Bertolini G, Iapichino G, Radrizzani D, et al. Early
114. Mehta NM. Parenteral nutrition in critically Ill chil- enteral immunonutrition in patients with severe
dren. N Engl J Med. 2016;374:1190–2. sepsis: results of an interim analysis of a random-
115. Reignier J, Boisrame-Helms J, Brisard L, et al. Enteral ized multicentre clinical trial. Intensive Care Med.
versus parenteral early nutrition in ventilated adults 2003;29:834–40.
ERRNVPHGLFRVRUJ
Medical Management of Acute
Liver Failure 9
Heli Bhatt and Girish S. Rao
ERRNVPHGLFRVRUJ
156 H. Bhatt and G. S. Rao
Up to 40–50% patients with PALF lack a spe- including electrolytes, blood urea nitrogen
cific etiological diagnosis partly due to lack of (BUN), creatinine (Cr), and albumin, liver
thorough diagnostic evaluation [8]. In infants, enzymes, total and fractionated bilirubin,
infections and metabolic diseases are the most gamma-glutamyl transferase (GGT), coagula-
common known etiologies for PALF [7, 8]. tion profile along with prothrombin time (PT)
Herpes simplex virus is the most commonly with international normalized ratio (INR), a
identified infectious etiology in these children. complete blood count with differential and
Galactosemia, tyrosinemia, and fatty acid oxida- platelets, and a reliable serum ammonia level
tion defect are the commonly identified meta- should be obtained in all the patients at diagno-
bolic disorders in this age group, while neonatal sis and thereafter monitored closely. Additional
hemochromatosis was the most common cause tests to assess for etiology of liver failure should
of liver failure in the neonatal period [7]. In chil- be tailored to the age and presentation of the ill-
dren older than 7 months of age with ALF, drug ness. These should be targeted to evaluate for
toxicity, especially with acetaminophen over- infectious, metabolic, and autoimmune liver
dose, and autoimmune hepatitis are the most diseases in addition to drug or other toxin expo-
commonly identified etiologies [7, 8]. In devel- sures (Table 9.1) [9]. A complete abdominal
oping countries, infectious etiologies remain the ultrasound with Doppler should be obtained in
most common identified cause of acute liver fail- all patients with PALF, but additional imaging
ure through all age groups. Hepatitis A viral studies like CT, MRI, and/or MRCP should be
infection is the most common infectious agent in considered based on the individual case. Liver
these countries [8]. biopsy should be considered early in the course
of PALF to assess for the cause of liver failure
and the extent and pattern of hepatocyte injury.
Diagnostic Evaluation Transjugular approach is preferred in the setting
of severe coagulopathy with liver failure. The
All children with ALF should undergo thorough level of necrosis might be underestimated on
systematic evaluation for the underlying etiol- liver biopsy [10]. Submassive or massive liver
ogy of acute liver failure along with the assess- necrosis is associated with poor prognosis [11].
ment for liver injury, dysfunction, and
multisystem complications. This diagnostic
evaluation should be individualized and directed Management
toward the age-specific causes of acute liver fail-
ure. A detailed medical history including onset Management of acute liver failure is very chal-
of symptoms and neurological changes; history lenging due to multiple reasons. The multisystem
of exposure to infections or medications; family organ involvement with potential for rapid dete-
medical history including history of liver dis- rioration and death in absence of timely liver
ease, consanguinity, genetic, metabolic, or auto- transplant makes this disease one of the most dif-
immune diseases; and/or sibling death should be ficult diseases to manage. At the same time, there
obtained in all patients with acute liver failure. A is a tremendous potential for self-recovery with-
thorough physical examination with special out transplant and unnecessary morbidity can be
attention to mentation and neurological status is avoided by preventing transplants in these cases.
imperative. There is a lack of adequate data on management
Radiological and laboratory tests in PALF of PALF due to its rarity and currently, the major-
should be obtained to evaluate underlying etiol- ity of our clinical practice guidance is derived
ogy, to assess the extent of liver injury and fail- from adult studies. Hence, many principles of
ure, or to monitor for potential complications of management of this disease are currently unclear
liver failure. A comprehensive metabolic panel and controversial.
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 157
ERRNVPHGLFRVRUJ
158 H. Bhatt and G. S. Rao
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 159
timely decision-making for critical aspects of olism that is generated from the breakdown of
patient care and interventions including trans- glutamine by glutaminase, an enzyme within the
plantation and its evaluation. enterocytes of the small intestine and colon, and
by urease-producing bacteria that inhabit the gut.
This gut-derived ammonia enters the urea cycle
Central Nervous System to be detoxified into urea and excreted in the
urine. Ammonia that bypasses this detoxification
Hepatic encephalopathy (HE) is a neuropsychiat- is converted to glutamine in hepatocytes, skeletal
ric syndrome associated with liver failure in the myocytes, and astrocytes in brain. Astrocytes are
absence of a preexisting brain disease. It is char- the most abundant type of cells in the brain. They
acterized by progressive but reversible deteriora- are very sensitive to a rapid increase in ammonia,
tion of behavior, cognition, and mentation in which subsequently leads to cellular edema due
patients with PALF. The clinical features of to increased influx of water secondary to osmotic
hepatic encephalopathy can range from overt gradient created by increased intracellular gluta-
coma to irritability to minor changes in behavior mine. Although hyperammonemia has been
and motor or cognitive skills. In children, fea- implicated to play a pivotal role in development
tures of HE can be subtle and difficult to assess of hepatic encephalopathy, a consistent correla-
and range from mild irritability to inactivity to tion between the plasma concentration of ammo-
coma. Table 9.3 describes the clinical stages of nia and clinical manifestation of HE has not been
encephalopathy originally developed to assess established [14]. In addition to hyperammone-
patients with cirrhosis but is used in ALF for the mia, increased proinflammatory circulatory cyto-
lack of a better clinical tool. In the Pediatric kines like IL-1β, IL-6, and TNF-α can modulate
Acute Liver Failure study group database, the cerebral blood flow and cellular permeability to
majority of the patients (75%) had grade 1–2 have direct or permissive effects on development
hepatic encephalopathy, and grade 3 and 4 and progression of HE into cerebral edema [15].
encephalopathy were seen in 17% and 7%
patients, respectively. Out of 348 patients
included in this study group, more than half Management of Hepatic
developed hepatic encephalopathy [3]. Encephalopathy
The exact pathogenesis of hepatic encepha-
lopathy is complex, is not completely understood, Given the potential for rapid neurological deteri-
and involves a number of interrelated factors [4, oration in patients with PALF and HE, early rec-
13]. Ammonia is a by-product of nitrogen metab- ognition and prompt management of HE are
ERRNVPHGLFRVRUJ
160 H. Bhatt and G. S. Rao
necessary to decrease morbidity and mortality in Currently, there is no data to support their use in
these patients. As mentioned above, these patients PALF. The benefit of continuous renal replace-
should be closely monitored in a critical care set- ment therapy (CRRT) in reducing serum ammo-
ting with frequent neurological assessments. nia and improving 21-day transplant-free survival
There should be minimal stimulation, and unnec- has been demonstrated in a recent cohort study
essary interventions should be avoided. from US ALF study group registry [21].
Endotracheal intubation for airway protection or
controlled ventilation in advanced stages of
encephalopathy should be considered. Head of Management of Intracranial
the bed should be elevated to 20–30°, and the Hypertension and Cerebral Edema
patient’s head should be maintained in the mid-
line to provide optimal CSF drainage and jugular The goal in the management of cerebral edema
venous outflow. Aggressive efforts to maintain and intracranial hypertension is maintaining ade-
normothermia are crucial as fever and shivering quate cerebral perfusion pressure (CPP) while
may exacerbate intracranial pressure (ICP). lowering and maintaining intracranial pressure
Ventilation, oxygenation, and mean arterial pres- (ICP) to less than 20 mm Hg to assure adequate
sures should be meticulously monitored and perfusion of the brain [4, 22]. Close clinical mon-
maintained. itoring is strongly recommended but highly chal-
Lactulose and nonabsorbable oral antibiotics lenging in pediatric patients, especially if they
like neomycin and rifaximin have been used for have progressed to grade 3–4 HE. Cushing’s triad
prophylaxis and management of HE in patients of irregular breathing, systemic hypertension,
with chronic liver disease and cirrhosis. Lactulose and bradycardia is not uniformly present.
is a synthetic nonabsorbable disaccharide which Intracranial pressure monitoring can be used to
can be used to decrease intraluminal pH in the assess CPP to avoid hypoxic brain injury.
colon and prevent uptake of ammonia from the Multiple studies have been done to evaluate the
gastrointestinal lumen. Given the central role of safety and efficacy of invasive intracranial pres-
increased arterial ammonia levels in the patho- sure monitoring in management of ALF, but there
genesis of hepatic encephalopathy, one could has been no demonstrated improvement in sur-
assume that ammonia-lowering strategies might vival [23–25]. The use of invasive intracranial
be effective in halting the progression of neuro- monitoring in PALF is controversial owing to the
logical deterioration. While the abovementioned lack of sufficient evidence for its routine use and
agents have a role in preventing progression of safety. Maintaining systemic blood pressure by
HE associated with cirrhosis in patients with adequate volume resuscitation and use of vasoac-
chronic liver disease, there are no controlled tri- tive medications as well as aggressively treating
als to support the use of these medications to treat fluid overload with CRRT are crucial to maintain
hepatic encephalopathy in ALF [16–18]. In one adequate CPP.
nonrandomized retrospective series, there was no The principal therapy to reduce cerebral
improvement in outcome with lactulose therapy edema and increased intracranial pressure is
[19]. Lactulose use should be avoided in PALF as administration of osmotic agents like hypertonic
it has the potential to cause intravascular volume saline and mannitol [26]. Mannitol is a hyperos-
depletion, hypernatremia, and bowel distension molar agent, which works by promoting move-
or megacolon, which can be dangerous during ment of water from astrocytes into the serum by
transplant [16, 17]. Neomycin is also not recom- increasing osmolality of serum. Mannitol also
mended due to increased risk of nephrotoxicity decreases viscosity of blood causing vasocon-
[16]. L-Ornithine L-aspartate (LOLA) and striction, which leads to less cerebral blood vol-
L-ornithine phenyl acetate (LOPA) have shown ume and decreased ICP. It is used as a first-line
promising results in adult trials [20]. These work agent in management of increased ICP in adults
by increasing renal excretion of ammonia. with ALF [16, 17]. The recommended dose for
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 161
use is 0.25–1.0 gm/kg IV bolus that can be acute life-threatening mannitol-refractory wors-
repeated once or twice [16, 17]. The majority of ening of intracranial pressure to delay impending
information on mannitol in PALF is extrapolated cerebral herniation, but sustained hyperventila-
from adult literature. The effect of mannitol is tion should be avoided in patients with ALF.
transient, and it is difficult to achieve sustained Hypothermia prevents cerebral edema by
reduction of ICP to acceptable levels with man- decreasing cerebral metabolism, neuronal inflam-
nitol alone. The use of mannitol is not recom- mation, and oxidative stress. It also decreases
mended in presence of renal failure, hypovolemia, ammonia level and improves cerebral hemody-
or serum osmolality >320 mOsm/L [16, 17]. namics. Therapeutic hypothermia to 32–35° has
Hypertonic saline (3–30%) decreases ICP by been used in adults with ALF to reduce ICP for
mechanisms similar to mannitol. In addition, it successfully bridging these patients to liver trans-
also stabilizes cerebral endothelial cell volume plantation. There have been a few reports of ben-
and improves cerebral circulation. In a random- eficial effects of therapeutic hypothermia in adult
ized controlled trial from King’s College, patients patients with ALF [31, 32]. However, there have
who received hypertonic saline had decreased been complications reported with therapeutic
ICP from baseline in the first 24 h, and the inci- hypothermia too. These include cardiac dys-
dence of ICP >25 mm Hg or greater was signifi- rhythmias, increased risk of infection, coagulop-
cantly lower than control group; however, it did athy, electrolyte disturbances, hyperglycemia,
not show improved survival in patients treated and theoretical decreased hepatic regeneration
with HTS [27]. The use of hypertonic saline in [31, 32]. A multicenter retrospective cohort anal-
treatment of elevated ICP in PALF has not been ysis of 97 patients enrolled in the US ALF study
studied. In patients with PALF who have elevated group did not find a difference in 21-day mortal-
ICP, it is reasonable to maintain serum sodium ity as well as transplant-free survival with or
level around 145–150 mmol/L, especially now without the use of therapeutic hypothermia [33].
that hypertonic saline has been established as a There is no data to support the use of therapeutic
standard of care in management of pediatric hypothermia for neuroprotection. However,
patients with traumatic brain injury [28]. There hyperthermia should be aggressively managed to
are no randomized clinical trials regarding the avoid worsening of ICP. Currently, active normo-
use of hypertonic saline or mannitol in children. thermia (36–37°) remains the standard of care as
Hypertonic saline provides all benefits of hyper- it offers the best risk-benefit ratio [4].
osmolar agent without the hemodynamic side Early identification of neurological decline
effects associated with mannitol. According to and timely therapeutic interventions to minimize
2012 guidelines for the medical management of neurological morbidity are crucial in ALF
severe traumatic brain injury in children, use of because neurological morbidity is a major deter-
hypertonic saline is favored over the use of man- minant of outcome in ALF [4]. However, clinical
nitol for management of ICH [29]. assessment of neurological status in pediatric
Patients with ALF hyperventilate due to the patients can be difficult. Head imaging with com-
metabolic milieu associated with ALF, and this, puterized tomography (CT) scan is used to
in turn, helps restore cerebral autoregulation, exclude intracranial hemorrhage as a cause of
vasoconstriction, and reduction of ICP. Effects of sudden decline in neurological status. However,
hyperventilation are temporary and continuous in a recent single-center retrospective pediatric
hyperventilation offers no survival benefit in study assessing the role EEG in management of
patients with ALF [30]. Moreover, it has the PALF by Hussain et al. [34], CT and magnetic
potential to worsen cerebral edema due to cere- resonance imaging (MRI), even though abnormal
bral hypoperfusion. According to American in 13% of patients, failed to demonstrate consis-
Association for the Study of Liver Disease tent abnormalities to suggest the presence of
(AASLD) position paper for management of cerebral edema. There was no association
ALF, hyperventilation may be used to manage between EEG and CT/MRI findings in this study.
ERRNVPHGLFRVRUJ
162 H. Bhatt and G. S. Rao
However, there was increased mortality in cially deleterious in advanced stages of hepatic
patients with certain EEG abnormalities. These encephalopathy. Pain management and sedation
included moderate to severe slowing, epilepti- are important components of critical care man-
form discharge, and electrographic seizure. EEG agement of children with ALF. Sedating non-
seems to be a very sensitive tool to screen not intubated patients may be necessary, and
only for subclinical seizures but also declining anxiolytics should be used after carefully weigh-
neurological status in patients progressing to ing the benefit of reducing agitation versus blunt-
grade 3 or 4 encephalopathy or with unexplained ing the signs of neurological deterioration and
clinical deterioration [34, 35]. Transcranial exacerbating encephalopathy [4]. Also, numer-
Doppler ultrasonography has been shown to be ous drugs used for sedation and analgesia have
helpful in measuring dynamic changes in ICP in hepatic or renal clearance. There is a lack of suf-
a small retrospective study in adults with ALF ficient data for the use of standard sedative or
[36]. There are no studies for this diagnostic analgesic agents in PALF. Short-acting agents,
modality in PALF. with appropriate dose adjustments for liver dys-
Seizure activity worsens cerebral edema by function, should be used. Benzodiazepines can
increasing the oxygen requirement of the astro- have prolonged sedative effect when used in
cytes [35]. The true frequency of seizures in patients with hepatic impairment and should be
patients with ALF may be underestimated with- avoided. Furthermore, benzodiazepines and pro-
out continuous EEG. In the study by Hussain pofol have the potential to worsen HE by increas-
et al. 11% of patients had clinical seizures; how- ing gamma-aminobutyric acid (GABA)
ever, almost 5% of patients had subclinical non- neurotransmission [38]. Propofol in limited doses
convulsive seizures [34]. Continuous EEG should and for short period of time may be used in older
be used in patients with grade 3 or 4 HE or with children without mitochondrial disease, due to
acute decline in neurological status to rule out shorter recovery time and neuroprotective effect
subclinical seizures [4, 35]. Aggressive antiepi- through decreased cerebral blood flow and
leptic therapy should be implemented to control decrease in ICP [4, 39]. Opioid agents with short
seizures to prevent further neurological morbid- half-life such as fentanyl and remifentanil can be
ity. Phenytoin can be used for prompt control of used concurrently to improve cardiovascular sta-
epileptiform activity, and short-acting benzodiaz- bility [16]. Dose adjustments are recommended
epines can be used in phenytoin-refractory cases while using dexmedetomidine for its sedative and
[17]. Valproic acid should be avoided if mito- analgesic effect in PALF as it is metabolized pri-
chondrial disease is suspected as the underlying marily in the liver [40]. Atracurium and cisatra-
etiology for ALF. Prophylactic phenytoin has curium are the preferred agents for neuromuscular
been tried to suppress subclinical epileptiform blockade in PALF. These undergo ester hydroly-
activity in adults with ALF; however, a subse- sis and Hoffman elimination, and their duration
quent trial demonstrated no benefit of its use in of action in liver failure is similar to the same in
preventing seizures, brain edema, or improving normal liver function [41]. Vecuronium and
survival [35, 37]. There are no pediatric trials to rocuronium should be avoided in ALF as they
support the use of prophylactic phenytoin in undergo hepatic metabolism.
management of PALF. Prophylactic phenytoin,
therefore, cannot be recommended in PALF at
this time. Although there are rare reports of drug- Cardiovascular
induced liver injury with levetiracetam, it can be
safely used in management of seizures in PALF. Hyperdynamic circulatory failure with low mean
Pain can arise from multiple diagnostic and arterial pressure occurs in ALF due to peripheral
therapeutic interventions and procedures in vasodilation caused by elevated circulatory cyto-
patients with PALF. Psychomotor agitation is kines. This significantly decreases peripheral tis-
known to increase ICP, and this may be espe- sue oxygenation exacerbating multi-organ
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 163
failure. Depleted intravascular volume due to pressor/fluid refractory hypotension may benefit
decreased intake as well as increased transuda- from a trial of systemic corticosteroid adminis-
tion into extravascular space adds to this hemo- tration [16, 45, 46].
dynamic instability.
As with any patient with hypotension, intra-
vascular volume status should be assessed and Respiratory
replenished with adequate volume replacement
[17]. If the patient remains hypotensive after fluid In adults, about 20–30% patients with ALF are
resuscitation, vasopressors should be initiated to diagnosed with acute respiratory distress syn-
maintain mean arterial pressures within the age- drome (ARDS) [47]. Exact incidence of ARDS in
appropriate normal range. This is crucial to PALF is unknown. According to the PALF study
assure adequate cerebral perfusion pressure. group data, almost 40% children with PALF
Norepinephrine has been the preferred agent in required ventilator support [3]. Endotracheal
adults, mainly, because it provides a more consis- intubation may be required in PALF either for
tent and predictable increase in cerebral perfu- airway protection in patients with hepatic
sion while minimizing tachycardia and preserving encephalopathy or for management of respiratory
splanchnic circulation [16]. Despite the lack of failure secondary to sepsis, fluid overload-associ-
adequate pediatric data on choice of vasopres- ated pulmonary edema, pulmonary hemorrhage,
sors, norepinephrine does seem to be a reason- or ARDS.
able choice to optimize organ perfusion in PALF There are no pediatric trials directing mechan-
[4]. Vasopressin and its analogues can be used in ical ventilation in children with PALF. Mechanical
volume- and norepinephrine-refractory cases to ventilation strategies in PALF should aim at
potentiate its effects; however, these should be decreasing ventilator associated lung injury while
used cautiously due to the potential direct cere- providing maximum neuroprotection in the set-
bral vasodilatory effect that may worsen intracra- ting of elevated intracranial pressure.
nial hypertension [16, 17]. Echocardiography can Conservative tidal volume ventilation (5–8 ml/kg
be used to assess for systolic and diastolic dys- of predicted body weight) with moderately ele-
function in patients not responding to volume and vated positive end expiratory pressures should be
vasoactive support. titrated to maintain normocapnia and avoid
hypoxemia [48]. As mentioned above, sustained
hyperventilation should be avoided as the effects
Adrenal of hyperventilation on intracranial pressure are
temporary, and there is a potential risk of worsen-
Although there is a discrepancy in the definition, ing cerebral edema by causing cerebral hypoxia
relative adrenal insufficiency/hepatoadrenal syn- [17]; however, it may be used briefly in sudden
drome has been well described in septic shock as life-threatening worsening of intracranial pres-
well as ALF [42]. A third of adults with ALF may sure that is refractory to osmotic agents to delay
develop relative adrenal insufficiency, and its impending cerebral herniation.
incidence seems to be directly proportional to the
severity of liver failure [43]. Low HDL levels
with increased circulatory endotoxins and proin- Renal
flammatory markers like TNF-α lead to impaired
cortisol secretion and impaired adrenal function Acute kidney injury (AKI), and subsequent renal
that can depress sensitivity to catecholamines failure, is a relatively common complication of
[43, 44]. No data is available to define this condi- ALF. In large retrospective review using patients
tion in PALF or to guide diagnosis and manage- in the US Acute Liver Failure study group
ment of relative adrenal insufficiency in (ALFSG), AKI was seen in almost 47% patients
PALF. However, children with PALF and vaso- with ALF [49]. In this study, there was decreased
ERRNVPHGLFRVRUJ
164 H. Bhatt and G. S. Rao
overall survival in patients with AKI versus cient data. RRT (hemofiltration or dialysis) can
patients without AKI. Also, there was decreased help correct electrolyte imbalances, worsening
transplant-free survival in patients needing renal acidosis, fluid overload, and hyperammonemia.
replacement therapy (RRT) or with advanced The degree of kidney injury, electrolyte distur-
AKI versus those without AKI [49]. Although bance, and fluid imbalance should be integrated
exact incidence of AKI in PALF is unknown, the into the decision to start RRT in patients with
prospective PALF study reported the need for PALF [4]. Continuous renal replacement therapy
hemofiltration in nearly 10% of patients [3]. AKI is preferred over intermittent hemodialysis due to
has been reported in 15–20% of children with lower risk of hemodynamic instability and wors-
ALF where it was associated with decreased sur- ening ICP [17, 55]. In a recent study on a large
vival [50]. cohort of patients enrolled in ALFSG, serum
AKI in ALF can be multifactorial, and causes ammonia modulation with CRRT improved
include hypovolemia, sepsis, acute tubular necro- 21-day transplant-free survival [21]. Although
sis, nephrotoxic medications, acetaminophen- there is inadequate data in PALF regarding use of
induced renal injury, and functional renal failure CRRT in management of HE, early use of CRRT
[51, 52]. Functional renal failure can arise from a should be considered in patients at a greater risk
mechanism similar to hepatorenal syndrome in of progression of HE (e.g., high-grade encepha-
chronic liver disease. Intrarenal vasoconstriction lopathy, advanced AKI, vasopressors, etc.).
leads to decreased renal perfusion and subse- AKI in PALF resolves after restoration of liver
quent kidney injury [51–53]. function in majority of cases. However, in certain
Assessment of renal dysfunction in chil- circumstances, simultaneous liver and kidney
dren with ALF may be augmented by using transplantation must be considered. In adults,
multiple criteria such as the combination of indications for concurrent liver/kidney transplan-
serum creatinine (sCr), urinary output, and tations are based on degree and duration of renal
fluid balance aid in diagnosing AKI. SCr by injury and are strongly considered if patient has
itself may overestimate renal function, and needed dialysis for 8–12 weeks [57]. However,
change of SCr over baseline is more relevant there is minimal data, mainly based on single-
over a single value to assess progression of center experience, available for guidance regard-
renal injury [54, 55]. Urinary biomarkers like ing liver/kidney transplant in children [58].
neutrophil gelatinase-associated lipocalin,
IL-18, kidney injury molecule-1, and liver-
type fatty acid-binding protein are emerging Fluid, Electrolytes, and Nutrition
for evaluation of pediatric AKI; however, these
have not been studied in PALF [56]. Acid-base imbalances, electrolyte abnormalities,
Management of AKI in PALF should be and metabolic derangements are common in
focused on interventions to reduce kidney injury PALF and need to be identified and corrected fas-
and prevent progression to renal failure. Adequate tidiously due to their life-threatening potential.
hydration, avoiding excessive diuresis, maintain- Serum electrolytes should be monitored fre-
ing adequate renal perfusion pressure, and mini- quently and corrected meticulously to prevent
mizing the use or adjusting the dose of intravenous mortality and decrease morbidity in these criti-
contrast and nephrotoxic medications are some cally ill patients. Alkalosis and acidosis both may
measures to prevent AKI in ALF [16, 17, 52]. occur in ALF and should be managed by correct-
Intravenous fluid challenge should be considered ing the cause of acid-base imbalance [17].
in patients with suspected prerenal azotemia, but Hyponatremia and hypokalemia can be second-
volume overload should be avoided in patients ary to ascites, dilution from aggressive volume
with PALF [4]. resuscitation, and urinary losses from diuretic
The criteria for initiation or discontinuation of use. Hyponatremia should be strictly avoided as
RRT in PALF are ill-defined due to lack of suffi- it can exacerbate cerebral edema.
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 165
ERRNVPHGLFRVRUJ
166 H. Bhatt and G. S. Rao
bleeding in patients with ALF is mainly stress no clear guidelines for the use of prophylactic
induced or acid related. The AASLD guidelines on antibiotics or antifungals for liver failure in
management of ALF recommend H2 blockers or adults or children, and these should be avoided
proton pump inhibitors (PPI) for prophylactic pre- [16, 17]. Pulmonary, urinary, and blood stream
vention of this type of GI bleeding [17]. Variceal infections are the most common sites for bacte-
bleeding is rare in patients with ALF. Ascites may rial infection, and gram-positive cocci like
develop in a minority of patients with ALF, and staphylococci and streptococci and enteric gram-
spironolactone is the drug of choice to manage negative bacilli are the most commonly isolated
ascites of hepatic origin. Diuresis is indicated only organisms [66]. Obtaining appropriate evalua-
if there is respiratory compromise or significant tions in patients with ALF and signs/symptoms
discomfort due to abdominal distension. of infection should not be delayed. This may
Aggressive diuresis should be avoided to prevent include chest x-ray and urine and blood cultures
precipitation of hepatorenal syndrome [5]. with any suspicion of infection, SIRS, refractory
hypotension, or worsening encephalopathy [16,
17]. Empiric antibiotics should be initiated in
Infection: SIRS patients exhibiting SIRS and should have ade-
quate coverage for the abovementioned bacterial
The liver is involved in multiple immune-related infections.
functions which are disturbed in ALF making
these patients more susceptible to infection.
Additionally, these patients have defects in many Liver Support Systems
other host defense mechanism pathways which
decrease their ability to fight against infections. In acute liver failure, CRRT is highly effective in
As a result, infection and systemic inflammatory the removal of smaller molecules of water-solu-
response syndrome contribute to significant mor- ble toxins, i.e., urea, ammonia, etc. However, the
bidity and mortality in these patients. Bacterial larger or albumin-bound non-water-soluble mol-
infections account for 10–37% of mortality in ecules like cytokines, bile acids, bilirubin, and
adults with ALF [69, 70]. In a retrospective metabolites of aromatic amino acids and medium
review, the incidence of bacterial infection in chain fatty acids are not successfully removed
children with PALF was about 25%, and bacterial during hemodialysis. These large albumin-bound
infections were associated with increased mor- non-soluble molecules accumulate and contrib-
bidity in these patients [71]. Systemic inflamma- ute to progression of liver failure [73, 74]. High-
tory response syndrome (SIRS) has been reported volume hemofiltration has been used to remove
in 50–60% of adults with ALF [69]. Increased circulatory cytokines and was associated with
SIRS components are directly correlated to improved hemodynamics and encephalopathy
increasing mortality and are strongly associated [75]. In a randomized control trial in adults with
with worsening encephalopathy. Encephalopathy ALF, high-volume nonselective plasmapheresis
has been shown to progress in majority of patients demonstrated improved transplant-free survival
with infection and in 50% of patients with greater when compared to a control group along with
than two SIRS components versus 25% of decreased SIRS score and SOFA score [76].
patients without SIRS [72]. There are no retro- High-volume plasmapheresis dampened innate
spective or prospective studies to assess the asso- immune response, and early use of plasmapher-
ciation or incidence of infection or SIRS with esis might provide a window of homeostasis for
outcomes in PALF. the liver to regenerate [76]. However, a small ret-
In spite of several studies examining the use rospective pediatric study failed to show survival
of prophylactic antibiotics in management of benefit with the use of plasma exchange [77].
ALF, the results remain inconclusive. There are
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 167
ERRNVPHGLFRVRUJ
168 H. Bhatt and G. S. Rao
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 169
ERRNVPHGLFRVRUJ
170 H. Bhatt and G. S. Rao
an opportunity for quality improvement. J Pediatr. mortality in acute liver failure. Hepatology. 2017;
2009;155(6):801–6.e1. https://doi.org/10.1016/j. https://doi.org/10.1002/hep.29488.
jpeds.2009.06.005. 22. Bucuvalas J, Yazigi N, Squires RH Jr. Acute liver fail-
9. Bhatt H, Rao GS. Management of Acute Liver ure in children. Clin Liver Dis. 2006;10(1):149–68.,
Failure: a pediatric perspective. Curr Pediatr Rep. vii. https://doi.org/10.1016/j.cld.2005.10.006.
2018; https://doi.org/10.1007/s40124-018-0174-7. 23. Vaquero J, Fontana RJ, Larson AM, Bass NM, Davern
10. Singhal A, Vadlamudi S, Stokes K, Cassidy FP,
TJ, Shakil AO, et al. Complications and use of intra-
Corn A, Shrago SS, et al. Liver histology as pre- cranial pressure monitoring in patients with acute
dictor of outcome in patients with acute liver fail- liver failure and severe encephalopathy. Liver Transpl.
ure. Transpl Int. 2012;25(6):658–62. https://doi. 2005;11(12):1581–9. https://doi.org/10.1002/
org/10.1111/j.1432-2277.2012.01470.x. lt.20625.
11. Miraglia R, Luca A, Gruttadauria S, Minervini MI, 24. Rajajee V, Fontana RJ, Courey AJ, Patil PG. Protocol
Vizzini G, Arcadipane A, et al. Contribution of tran- based invasive intracranial pressure monitoring in
sjugular liver biopsy in patients with the clinical pre- acute liver failure: feasibility, safety and impact on
sentation of acute liver failure. Cardiovasc Intervent management. Crit Care. 2017;21(1):178. https://doi.
Radiol. 2006;29(6):1008–10. https://doi.org/10.1007/ org/10.1186/s13054-017-1762-6.
s00270-006-0052-5. 25. Kamat P, Kunde S, Vos M, Vats A, Gupta N, Heffron
12. Ostapowicz G, Fontana RJ, Schiodt FV, Larson
T, et al. Invasive intracranial pressure monitoring is a
A, Davern TJ, Han SH, et al. Results of a prospec- useful adjunct in the management of severe hepatic
tive study of acute liver failure at 17 tertiary care encephalopathy associated with pediatric acute liver
centers in the United States. Ann Intern Med. failure. Pediatr Crit Care Med. 2012;13(1):e33–8.
2002;137(12):947–54. https://doi.org/10.1097/PCC.0b013e31820ac08f.
13. Scott TR, Kronsten VT, Hughes RD, Shawcross
26. Richardson D, Bellamy M. Intracranial hyperten-
DL. Pathophysiology of cerebral oedema in acute liver sion in acute liver failure. Nephrol Dial Transplant.
failure. World J Gastroenterol. 2013;19(48):9240–55. 2002;17(1):23–7.
https://doi.org/10.3748/wjg.v19.i48.9240. 27. Murphy N, Auzinger G, Bernel W, Wendon J. The
14. Bernal W, Hall C, Karvellas CJ, Auzinger G, Sizer E, effect of hypertonic sodium chloride on intra-
Wendon J. Arterial ammonia and clinical risk factors cranial pressure in patients with acute liver fail-
for encephalopathy and intracranial hypertension in ure. Hepatology. 2004;39(2):464–70. https://doi.
acute liver failure. Hepatology. 2007;46(6):1844–52. org/10.1002/hep.20056.
https://doi.org/10.1002/hep.21838. 28. Bell MJ, Kochanek PM. Pediatric traumatic brain
15. Butterworth RF. The concept of "the inflamed brain" injury in 2012: the year with new guidelines and com-
in acute liver failure: mechanisms and new therapeu- mon data elements. Crit Care Clin. 2013;29(2):223–
tic opportunities. Metab Brain Dis. 2016;31(6):1283– 38. https://doi.org/10.1016/j.ccc.2012.11.004.
7. https://doi.org/10.1007/s11011-015-9747-0. 29. Kochanek PM, Carney N, Adelson PD, Ashwal
16. Stravitz RT, Kramer AH, Davern T, Shaikh AO, S, Bell MJ, Bratton S, et al. Guidelines for the
Caldwell SH, Mehta RL, et al. Intensive care of acute medical management of severe traumatic
patients with acute liver failure: recommenda- brain injury in infants, children, and adoles-
tions of the U.S. acute liver failure study group. cents—second edition. Pediatr Crit Care Med.
Crit Care Med. 2007;35(11):2498–508. https://doi. 2012;13(Suppl 1):S1–82. https://doi.org/10.1097/
org/10.1097/01.ccm.0000287592.94554.5f. PCC.0b013e31823f435c.
17. Lee WM, Larson AM, Stravitz RT. AASLD posi- 30.
Ede RJ, Gimson AES, Bihari D, Williams
tion paper: the management of acute liver failure: R. Controlled hyperventilation in the prevention
update 2011. AASLD September. 2011. of cerebral oedema in fulminant hepatic failure. J
18. Kodali S, McGuire BM. Diagnosis and Management Hepatol. 1986;2(1):43–51. https://doi.org/10.1016/
of Hepatic Encephalopathy in fulminant hepatic fail- S0168-8278(86)80007-1.
ure. Clin Liver Dis. 2015;19(3):565–76. https://doi. 31. Vaquero J. Therapeutic hypothermia in the man-
org/10.1016/j.cld.2015.04.006. agement of acute liver failure. Neurochem Int.
19. Alba L, Hay JE, Angulo P, Lee WM. Lactulose
2012;60(7):723–35. https://doi.org/10.1016/j.
therapy in acute liver failure. J Hepatol. 2002;36:33. neuint.2011.09.006.
https://doi.org/10.1016/S0168-8278(02)80097-6. 32.
Stravitz RT, Larsen FS. Therapeutic hypo-
20. Stravitz RT, Gottfried M, Durkalski V, Fontana RJ, thermia for acute liver failure. Crit Care Med.
Hanje AJ, Koch D, et al. Safety, tolerability and phar- 2009;37(7 Suppl):S258–64. https://doi.org/10.1097/
macokinetics of L-ornithine Phenylacetate in patients CCM.0b013e3181aa5fb8.
with acute liver injury/failure and Hyperammonemia. 33. Karvellas CJ, Cavazos J, Battenhouse H, Durkalski
Hepatology. 2017; https://doi.org/10.1002/ V, Balko J, Sanders C, et al. Effects of antimicrobial
hep.29621. prophylaxis and blood stream infections in patients
21. Cardoso FS, Gottfried M, Tujios S, Olson JC,
with acute liver failure: a retrospective cohort study.
Karvellas CJ. Continuous renal replacement therapy Clin Gastroenterol Hepatol. 2014;12(11):1942–9.e1.
is associated with reduced serum ammonia levels and https://doi.org/10.1016/j.cgh.2014.03.011.
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 171
ERRNVPHGLFRVRUJ
172 H. Bhatt and G. S. Rao
59. Plauth M, Cabre E, Riggio O, Assis-Camilo M, Pirlich 72. Vaquero J, Polson J, Chung C, Helenowski I, Schiodt
M, Kondrup J, et al. ESPEN guidelines on enteral FV, Reisch J, et al. Infection and the progression
nutrition: liver disease. Clin Nutr. 2006;25(2):285–94. of hepatic encephalopathy in acute liver failure.
https://doi.org/10.1016/j.clnu.2006.01.018. Gastroenterology. 2003;125(3):755–64.
60. Plauth M, Cabre E, Campillo B, Kondrup J, Marchesini 73. Struecker B, Raschzok N, Sauer IM. Liver sup-
G, Schutz T, et al. ESPEN guidelines on parenteral port strategies: cutting-edge technologies. Nat Rev
nutrition: hepatology. Clin Nutr. 2009;28(4):436–44. Gastroenterol Hepatol. 2014;11(3):166–76. https://
https://doi.org/10.1016/j.clnu.2009.04.019. doi.org/10.1038/nrgastro.2013.204.
61. Schutz T, Bechstein WO, Neuhaus P, Lochs H, Plauth 74. Rademacher S, Oppert M, Jorres A. Artificial extra-
M. Clinical practice of nutrition in acute liver fail- corporeal liver support therapy in patients with severe
ure—a European survey. Clin Nutr. 2004;23(5):975– liver failure. Expert Rev Gastroenterol Hepatol.
82. https://doi.org/10.1016/j.clnu.2004.03.005. 2011;5(5):591–9. https://doi.org/10.1586/egh.11.59.
62. Stravitz RT, Ellerbe C, Durkalski V, Schilsky M,
75. Chevret L, Durand P, Lambert J, Essouri S, Balu
Fontana RJ, Peterseim C, et al. Bleeding complications L, Devictor D, et al. High-volume hemofiltration in
in acute liver failure. Hepatology. 2018;67(5):1931– children with acute liver failure*. Pediatr Crit Care
42. https://doi.org/10.1002/hep.29694. Med. 2014;15(7):e300–5. https://doi.org/10.1097/
63. Stravitz RT, Lisman T, Luketic VA, Sterling RK,
pcc.0000000000000172.
Puri P, Fuchs M, et al. Minimal effects of acute liver 76. Larsen FS, Schmidt LE, Bernsmeier C, Rasmussen
injury/acute liver failure on hemostasis as assessed by A, Isoniemi H, Patel VC, et al. High-volume plasma
thromboelastography. J Hepatol. 2012;56(1):129–36. exchange in patients with acute liver failure: an open
https://doi.org/10.1016/j.jhep.2011.04.020. randomised controlled trial. J Hepatol. 2016;64(1):69–
64. Agarwal B, Wright G, Gatt A, Riddell A, Vemala V, 78. https://doi.org/10.1016/j.jhep.2015.08.018.
Mallett S, et al. Evaluation of coagulation abnormali- 77. Singer AL, Olthoff KM, Kim H, Rand E, Zamir G,
ties in acute liver failure. J Hepatol. 2012;57(4):780– Shaked A. Role of plasmapheresis in the manage-
6. https://doi.org/10.1016/j.jhep.2012.06.020. ment of acute hepatic failure in children. Ann Surg.
65. Pereira SP, Rowbotham D, Fitt S, Shearer MJ, Wendon 2001;234(3):418–24.
J, Williams R. Pharmacokinetics and efficacy of oral 78. Khuroo MS, Khuroo MS, Farahat KL. Molecular
versus intravenous mixed-micellar phylloquinone adsorbent recirculating system for acute and acute-on-
(vitamin K1) in severe acute liver disease. J Hepatol. chronic liver failure: a meta-analysis. Liver Transpl.
2005;42(3):365–70. https://doi.org/10.1016/j. 2004;10(9):1099–106. https://doi.org/10.1002/
jhep.2004.11.030. lt.20139.
66. Shami VM, Caldwell SH, Hespenheide EE, Arseneau 79. Saliba F, Camus C, Durand F, Mathurin P, Letierce
KO, Bickston SJ, Macik BG. Recombinant activated A, Delafosse B, et al. Albumin dialysis with a non-
factor VII for coagulopathy in fulminant hepatic cell artificial liver support device in patients with
failure compared with conventional therapy. Liver acute liver failure: a randomized, controlled trial.
Transpl. 2003;9(2):138–43. https://doi.org/10.1053/ Ann Intern Med. 2013;159(8):522–31. https://doi.
jlts.2003.50017. org/10.7326/0003-4819-159-8-201310150-00005.
67. Pavese P, Bonadona A, Beaubien J, Labrecque P,
80. Lexmond WS, Van Dael CM, Scheenstra R, Goorhuis
Pernod G, Letoublon C, et al. FVIIa corrects the coag- JF, Sieders E, Verkade HJ, et al. Experience with
ulopathy of fulminant hepatic failure but may be asso- molecular adsorbent recirculating system treatment in
ciated with thrombosis: a report of four cases. Can J 20 children listed for high-urgency liver transplanta-
Anaesth. 2005;52(1):26–9. https://doi.org/10.1007/ tion. Liver Transpl. 2015;21(3):369–80. https://doi.
bf03018576. org/10.1002/lt.24037.
68. Munoz SJ, Stravitz RT, Gabriel DA. Coagulopathy of 81. Demetriou AA, Brown RS Jr, Busuttil RW, Fair J,
acute liver failure. Clin Liver Dis. 2009;13(1):95–107. McGuire BM, Rosenthal P, et al. Prospective, ran-
https://doi.org/10.1016/j.cld.2008.10.001. domized, multicenter, controlled trial of a bioarti-
69. Rolando N, Wade J, Davalos M, Wendon J, Philpott- ficial liver in treating acute liver failure. Ann Surg.
Howard J, Williams R. The systemic inflammatory 2004;239(5):660–7. discussion 7-70
response syndrome in acute liver failure. Hepatology. 82. Jain V, Dhawan A. Extracorporeal liver support
2000;32(4 Pt 1):734–9. https://doi.org/10.1053/ Systems in Paediatric Liver Failure. J Pediatr
jhep.2000.17687. Gastroenterol Nutr. 2017;64(6):855–63. https://doi.
70. Rolando N, Harvey F, Brahm J, Philpott-Howard J, org/10.1097/mpg.0000000000001500.
Alexander G, Gimson A, et al. Prospective study of 83. Squires RH, Ng V, Romero R, Ekong U, Hardikar
bacterial infection in acute liver failure: an analysis of W, Emre S, et al. Evaluation of the pediatric
fifty patients. Hepatology. 1990;11(1):49–53. patient for liver transplantation: 2014 practice
71. Godbole G, Shanmugam N, Dhawan A, Verma
guideline by the American Association for the
A. Infectious complications in pediatric acute liver Study of Liver Diseases, American Society of
failure. J Pediatr Gastroenterol Nutr. 2011;53(3):320– Transplantation and the north American Society
5. https://doi.org/10.1097/MPG.0b013e318222b0cd. for Pediatric Gastroenterology, hepatology and
ERRNVPHGLFRVRUJ
9 Medical Management of Acute Liver Failure 173
nutrition. Hepatology. 2014;60(1):362–98. https:// 87. Barshes NR, Lee TC, Udell IW, O’Mahoney CA,
doi.org/10.1002/hep.27191. Karpen SJ, Carter BA, et al. The pediatric end-stage
84. Baliga P, Alvarez S, Lindblad A, Zeng L. Posttransplant liver disease (PELD) model as a predictor of survival
survival in pediatric fulminant hepatic failure: the benefit and posttransplant survival in pediatric liver
SPLIT experience. Liver Transpl. 2004;10(11):1364– transplant recipients. Liver Transpl. 2006;12(3):475–
71. https://doi.org/10.1002/lt.20252. 80. https://doi.org/10.1002/lt.20703.
85. Farmer DG, Venick RS, McDiarmid SV, Duffy JP, 88. Lu BR, Zhang S, Narkewicz MR, Belle SH, Squires
Kattan O, Hong JC, et al. Fulminant hepatic failure RH, Sokol RJ. Evaluation of the liver injury unit
in children: superior and durable outcomes with liver scoring system to predict survival in a multina-
transplantation over 25 years at a single center. Ann tional study of pediatric acute liver failure. J Pediatr.
Surg. 2009;250(3):484–93. https://doi.org/10.1097/ 2013;162(5):1010–6.e1-4. https://doi.org/10.1016/j.
SLA.0b013e3181b480ad. jpeds.2012.11.021.
86. Sundaram V, Shneider BL, Dhawan A, Ng VL, Im 89. Lu BR, Gralla J, Liu E, Dobyns EL, Narkewicz MR,
K, Belle S, et al. King’s college hospital criteria Sokol RJ. Evaluation of a scoring system for assess-
for non-acetaminophen induced acute liver fail- ing prognosis in pediatric acute liver failure. Clin
ure in an international cohort of children. J Pediatr. Gastroenterol Hepatol. 2008;6(10):1140–5. https://
2013;162(2):319–23.e1. https://doi.org/10.1016/j. doi.org/10.1016/j.cgh.2008.05.013.
jpeds.2012.07.002.
ERRNVPHGLFRVRUJ
Part IV
Renal Controversies
ERRNVPHGLFRVRUJ
Diagnosis and Management
of Acute Kidney Injury in Critical 10
Illness
ERRNVPHGLFRVRUJ
178 T. N. Webb et al.
the pediatric ICU. Neonates, and those requiring Children who undergo surgery to correct con-
renal replacement therapy, had the highest mor- genital heart lesions commonly develop cardiac
tality rates. In addition to this cross-sectional surgery-associated AKI (CS-AKI), with an inci-
study, our understanding of AKI in the pediatric dence of up to 50% postoperatively and with an
ICU has greatly expanded with the recent multi- even higher incidence in neonates [10, 11].
national, multicenter prospective study entitled Emerging data now reveals that factors such as
AWARE (Assessment of Worldwide Acute prolonged cardiopulmonary bypass (CPB) time,
Kidney Injury, Renal Angina and Epidemiology young age, and higher RACHS-1 (Risk Adjusted
in Critically Ill Children) [7]. classification for Congenital Heart Surgery) cat-
The AWARE study examined pediatric and egory are less likely to be independently associ-
young adult patients admitted to the pediatric ated with CS-AKI. The increase in CS-AKI is
ICU and children admitted to the pediatric car- likely attributed to the increased complexity of
diac ICU (but not after surgery for congenital heart surgeries that are performed along with
heart disease). This study provides the most com- increased survival of patients with congenital
prehensive analysis of the epidemiology of AKI heart lesions. CS-AKI is also associated with
with recruitment from over 30 pediatric ICUs increased length of ICU hospital stay and
from four continents. Findings revealed that dur- mortality [11, 12].
ing the first 7 days of ICU admission, AKI
occurred in approximately one-fourth of the
patients and severe AKI occurred in approxi- Outcomes Ascribed to AKI
mately 12%. Even after controlling for multiple
potential confounders and severity of illness AKI is common in adults, children, and neonates
scores, severe AKI and receipt of renal replace- admitted to intensive care units [1, 7, 13]. Even
ment therapy (RRT) were significant predictors after controlling for numerous potential con-
of death by 28 days of admission. Severe AKI founders, those with AKI have higher mortality,
was also associated with increased use of prolonged mechanical ventilation, and increased
mechanical ventilation, RRT, and longer ICU ICU length of stay. In addition, growing evidence
length of stay. These findings correlate with suggests that AKI is not only associated with
aforementioned adult data, specifically with short-term but also long-term consequences such
AKI-EPI. as chronic kidney disease (CKD) and end-stage
Defining AKI in the neonatal population is renal disease (ESRD) [14, 15]. Even if an AKI
challenging due to confounders including the pres- episode seems to resolve and the sCr returns to
ence of maternal serum creatinine (sCr) and imma- baseline, there is evidence that these patients may
turity of the proximal tubules. For these reasons, have “subclinical CKD” and are at higher risk to
investigations are ongoing to identify biomarkers progress to CKD [16]. In a prospective study, the
to assist with AKI diagnosis in neonates. A 24 cen- association between children with AKI in the
ter, multinational study was recently performed: pediatric and cardiac ICU and the incidence of
the AWAKEN (Assessment of Worldwide Acute CKD at 1–3 years after AKI was evaluated, and it
Kidney injury Epidemiology in Neonates) study was found that 10% developed CKD (eGFR
[8]. Of the over 2000 infants studied, 30% (605) of <60 mL/min/1.73m2) and an additional 50%
the patients were found to have AKI with the were at risk for development of CKD (measured
majority of those being less than 29 weeks gesta- GFR 60–90 mL/min/1.73 m2, hypertension, or
tion followed by those greater than 36 weeks ges- hyperfiltration) [15]. In the kidney transplant
tation. AKI was defined by the neonatal modified population, those who develop AKI within 3
KDIGO criteria [9]. Like the AKI-EPI and the years of kidney transplantation are at an increased
AWARE studies, even after adjusting for multiple risk for development of CKD and graft failure
potential confounders, those with AKI had longer [17]. Unfortunately, although guidelines for care
length of hospital stay and higher mortality. suggest that all patients should have kidney
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 179
f ollow-up within 3 months of hospital discharge, tions. The AWARE study demonstrated that not
very few programs have a systematic method to including oliguria as a criterion for AKI failed to
follow these patients, and in some reports, only identify a significant number of patients with
40% of those who develop AKI may actually fol- AKI and oliguria in of itself was associated with
low up with a nephrologist [18]. an increased risk of mortality [7]. Unfortunately,
obtaining both urine and sCr can be difficult in
children in the critical care setting. If an indwell-
Recognition: Does It Matter? ing bladder catheter is not in place, the clinician
then must depend on reports of number of voids
Current AKI criteria include both sCr and urine or weighing of diapers and not having a true
output for diagnosis; however, until recently, hourly urine flow rate. Of course, risks and ben-
many studies on AKI in the pediatric ICU did not efits must be considered in the decision of main-
include oliguria. The importance of inclusion of taining an indwelling bladder catheter with
both for diagnosis remains in question. There are concerns of increased risk of infection.
quite a few studies that suggest omitting oliguria More individuals are diagnosed with AKI by
fails to identify a significant number of individu- incorporating urine output criteria than by using
als with AKI. sCr alone [21–24]. In a prospective observational
study of over 300 critically ill patients, the
authors demonstrated that the diagnosis of AKI
Serum Creatinine as a Metric for AKI occurred earlier in patients with oliguria in com-
parison to those without [24]. In another prospec-
Serum creatinine currently remains the gold stan- tive observational study in critically ill adults, the
dard for AKI diagnosis as it is often readily avail- RIFLE (risk, injury, failure, loss of kidney func-
able and inexpensive. Importantly, sCr changes act tion, and end-stage kidney disease) criteria with
as a biomarker of kidney function, not injury. sCr alone was compared to sCr plus urine output.
Serum creatinine changes are often seen days after Using the RIFLE criteria with sCr alone failed to
the initial injury, thus delaying diagnosis and early recognize as many patients with AKI, failed to
management. Serum creatinine is not the ideal bio- identify the maximum AKI severity, and led to
marker in the detection of AKI as it is affected by delays in AKI diagnosis. Incorporation of urine
many other nonrenal factors such as gender, mus- output into the definition was associated with
cle mass, and fluid balance. It is extremely impor- higher mortality [23]. The added value of urine
tant to be mindful of those patients who are fluid output was also recently explored in a retrospec-
overloaded and its effects on the measurement of tive analysis of over 30,000 hospital admissions
sCr, likely masking the severity of AKI. Prior stud- over 8 years. In this cohort, approximately 75%
ies have demonstrated that failure to correct sCr of the patients developed AKI. Individuals who
for fluid balance underestimates the prevalence of met both sCr and urine output criteria for AKI
AKI, therefore suggesting that in some cases, oli- had worse outcomes than those who met only one
guria may be a better indicator of AKI [19, 20]. criterion [25]. Collectively, these studies strongly
support the need for both sCr and urine output in
defining AKI in the ICU population.
Urine Output as a Metric for AKI
ERRNVPHGLFRVRUJ
180 T. N. Webb et al.
Table 10.1 RIFLE classification of AKI Table 10.3 KDIGO classification of AKI
RIFLE classification criteria KDIGO classification criteria
Class Serum creatinine or GFR Urine output Stage Serum creatinine Urine output
Risk Increase in serum Less than 1 1.5–1.9 × baseline Less than
creatinine × 1.5 or GFR 0.5 mL/kg/h or ≥ 0.3 mg/dL increase 0.5 mL/kg/h
decrease >25% for more than for 6–12 h
6 h 2 2–2.9 × baseline Less than
Injury Increase in serum Less than 0.5 mL/kg/h
creatinine × 2 or GFR 0.5 mL/kg/h for ≥12 h
decrease >50% for more than 3 3 × baseline or increase in Less than
12 h serum creatinine ≥4 mg/dL or 0.3 mL/kg/h
Failure Increase in serum Less than initiation of renal replacement for ≥24 h or
creatinine × 3 or serum 0.3 mL/kg/h therapy or, in patients anuria for
creatinine >4 mg/dL with for 24 h or <18 years, decrease in eGFR to ≥12 h
an acute rise >0.5 mg/dL anuria for <35 mL/min/1.73 m2
or GFR decrease >75% 12 h
Loss Persistent acute renal
failure (complete loss of
kidney function a combination of the RIFLE, pRIFLE, and AKIN
>4 weeks) definitions and encompasses both adult and pedi-
End- End-stage renal disease atric criteria [27].
stage >3 months
A recent study in pediatrics compared the inci-
kidney
disease dence of AKI in both the ICU and non-ICU set-
tings between pRIFLE, AKIN, and KDIGO
according to creatinine changes [28]. This retro-
Table 10.2 AKIN classification of AKI spective study revealed that both AKI incidence
AKIN classification criteria
and staging varied among all three definitions. By
Stage Serum creatinine Urine output detecting the most stage 1 cases, pRIFLE gener-
1. Increase in serum creatinine of Less than ated the largest AKI cohort. In reference to AKI
≥0.3 mg/dL or increase 0.5 mL/kg/h diagnosis, AKIN and KDIGO were the two that
≥150–200% (1.5–2-fold) from for more than corresponded most accurately. These findings
baseline 6 h
demonstrate that while these definitions are simi-
2. Increase in serum creatinine Less than
>200–300% (>2–3-fold) from 0.5 mL/kg/h lar, there are differences significant enough to
baseline for more than cause variation in AKI staging. There should be a
12 h constant pursuit to find methods to improve the
3. Increase in serum creatinine Less than ability to predict meaningful outcomes. It is pos-
>300% (>3-fold) from 0.3 mL/kg/h
sible that incorporation of fluid overload and bio-
baseline or ≥ to 4 mg/dL with for 24 h or
an acute increase of at least anuria for markers may improve our ability to properly detect
0.5 mg/dL or on renal 12 h AKI. For now, KDIGO-defined AKI should be the
replacement therapy standardized criteria used for AKI diagnosis.
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 181
factors to better guide appropriate management. cer, and hypoplastic left heart syndrome, was
Based upon their specific physiological charac- evaluated in a single-center study. Specified cut-
teristics, biomarkers can be divided into catego- off values for interpretation of AKI risk were
ries of markers of tubular injury, glomerular used based upon cutoff values generated by
filtration rate (GFR), inflammation, and cell cycle their clinical laboratory. The trend of serial
arrest [29]. Some of the common biomarkers uNGAL values provided both predictive and
investigated include neutrophil gelatinase-prognostic value and served as a means of sup-
associated lipocalin (NGAL), cystatin C (CysC), port for clinical decision-making in their popu-
kidney injury molecule-1 (KIM-1), IL-18, liver- lation [37].
type fatty acid-binding protein (L-FABP), tissue Cystatin C is a marker of GFR. It is an endog-
inhibitor of metalloproteinase-2 (TIMP-2), and enous cysteine protease inhibitor that is pro-
insulin-like growth factor-binding protein 7 duced in all nucleated cells and is not affected by
(IGFBP7). While these biomarkers are promising gender or muscle mass [38]. CysC has a rela-
for AKI diagnosis, reference ranges for pediatrics tively short half-life of 2 h and responds rapidly
and widespread availability continue to be a chal- to changes in GFR. A recent meta-analysis of 13
lenge and require more investigation. studies evaluated the ability of CysC to predict
NGAL is a marker of tubular injury and is AKI. Of the 13 studies, most were adult studies
one of the most extensively studied AKI bio- and involved individuals post-cardiac surgery.
markers with over 200 studies in the medical Serum CysC had an AUC of 0.96 for predicting
literature. Both serum and urine NGAL AKI. However, subgroup analysis revealed that
(uNGAL) are upregulated following nephro- only when measured within 24 hour of renal
toxic and ischemic injury such as ischemia- injury or ICU admission was serum CysC of
reperfusion injury, drug toxicity, hypoxia, and diagnostic value [39]. A multicenter prospective
bacterial infections [30–32]. In both the neona- study of almost 300 children undergoing cardiac
tal and pediatric population requiring CPB, surgery evaluated whether measuring pre- and
NGAL measured within 2 h after initiating CPB postoperative serum CysC improved the predic-
was found to be an excellent early predictor of tion of AKI in comparison to sCr. Postoperative
AKI [33]. In a single-center, case-control study serum CysC measured within 6 h of CPB
of pediatric patients, the utility of multiple uri- strongly predicted the development of AKI with
nary biomarkers of AKI after CPB was evalu- an AUC of 0.89 (AKI was defined by sCr AKI).
ated. Urine NGAL was the only biomarker Postoperative serum CysC also predicted longer
elevated at 2 h post initiation of CPB with an ICU length of stay and longer duration of venti-
area under the operating curve (AUC) of >0.9 lation [40]. There is apprehension about utilizing
for AKI predictive ability. It was not until 12 h CysC alone in detecting AKI because it is
that the combination of NGAL with other bio- affected by multiple factors including corticoste-
markers improved the AUC for the prediction of roids, thyroid function, and CRP levels. These
AKI [34]. While NGAL has been extensively elements make its value questionable in accu-
studied and its utilization continues to increase, rately detecting AKI. Serum CysC in combina-
there are still some ongoing concerns such as tion with urine NGAL was investigated in a
confounders in which NGAL is affected includ- retrospective analysis of 345 pediatric patients
ing sepsis [35] and urinary tract infections [36]. who underwent CPB. Combining both serum
Furthermore, determining the cutoff values for CysC and urine NGAL at 2 h post-CPB was
uNGAL in varying age groups such as adult vs superior to sCr alone in predicting both AKI
pediatric vs neonatal population remains an severity and duration [41].
ongoing issue. The impact of uNGAL over a KIM-1 is also a marker of tubular injury [42].
variety of hospitalized patients with ages rang- KIM-1 was evaluated in 40 pediatric patients
ing from 4 months to 25 years, with various after CPB and was found to be elevated at 12 h
diagnoses including nephrotic syndrome, can- post-CPB with an AUC of 0.83 in the individu-
ERRNVPHGLFRVRUJ
182 T. N. Webb et al.
als who developed AKI [42]. Another study set significantly increased at 4 h post-CPB with an
out to characterize the patterns of KIM-1 and AUC of 0.81 [50].
uNGAL in the pediatric ICU and assess their While there continues to be ongoing investi-
properties in identifying those at risk for the gation of biomarkers for early AKI detection,
development of AKI. KIM-1 was not found to incorporating their use into routine clinical
be as reliable in identifying those at risk for AKI practice remains a challenge. Expecting a single
development and peaked between 12 and 24 h biomarker to replace sCr and urine output for
post-ICU admission with an AUC of 0.74. The investigating renal function is not realistic. The
patients with sepsis had higher levels of both combination of novel biomarker(s) with current
uNGAL and KIM-1, irrespective of develop- standards of assessing renal function will likely
ment of AKI [43]. As demonstrated, these stud- prove more effective. For example, combining
ies yield conflicting results for the accuracy of sCr, CysC, and uNGAL can help delineate glo-
KIM-1 in predicting AKI. merular from structural tubular damage. Aside
IL-18 is a pro-inflammatory cytokine that is a from CysC, uNGAL is the most studied and
mediator of ischemic renal injury [44]. By sys- readily available biomarker but does not yet
temic review and meta-analysis, the utility of bio- have widespread availability. Its use is advanta-
markers in predicting the need for RRT in geous in comparison to other biomarkers as
critically ill patients was evaluated. IL-18 had an uNGAL has been shown to be elevated within 2
AUC of 0.66 in predicting the need for RRT [45]. h of injury and is not removed by dialysis, so it
In a prospective, multicenter cohort study of chil- can also be used as a measure of renal recovery
dren with congenital cardiac lesions, it was found even in those patients receiving dialysis and
that IL-18 peaked at 6 h post-cardiac surgery. appears to be closer to validation in the pediatric
IL-18 along with uNGAL improved risk predic- population.
tion for severe AKI including the need for dialy-
sis, mechanical ventilation, and length of
hospitalization; however, it was only moderately Furosemide Stress Test
accurate in diagnosing severe AKI with an AUC
of 0.72 [46]. Another analysis demonstrated that While not labeled as a biomarker, furosemide
in non-septic critically ill children, IL-18 has been examined as a means to determine renal
increased before sCr and predicted the severity of tubular function. Furosemide is a highly protein-
AKI as well as mortality [47]. bound loop diuretic that is not filtered at the
L-FABP is induced in the proximal tubule glomerulus and is actively transported to the
early after AKI. In a single-center study of tubular lumen. Its use results in natriuresis by
pediatric patients post-CPB, L-FABP was inhibiting active chloride transport in the thick
found to increase 6 h post-CPB with an AUC of ascending limb of the loop of Henle. Therefore,
0.77 in predicting AKI [48]. In contrast, a pro- the urinary response to furosemide provides a
spective multicenter study consisting of chil- functional assessment of renal tubular function.
dren and adults undergoing CPB found that There are an increasing number of adult studies
L-FABP was not associated with AKI develop- that have examined the kidney’s response to
ment [49]. furosemide as a marker of renal functional
IGFBP7 and TIMP-2 are markers of cell cycle reserve in AKI in what is known as the furose-
arrest [29]. There are limited pediatric studies on mide stress test (FST). What makes this test ideal
the use of [TIMP-2]*[IGFBP7] for the prediction is that often in the setting of oliguric AKI, many
of AKI. In a case-control study evaluating 50 are “challenged” with a dose of furosemide in
patients at high risk for AKI development post- order to determine if the patient will have a uri-
CPB, it was found that [TIMP-2]*[IGFBP7] was nary response. In many cases, multiple doses are
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 183
given to no avail thus delaying the initiation of early assessment of renal function and serve as a
RRT. Standardization of the FST in the pediatric guide for clinical decision-making.
ICU population will assist in predicting those
who will likely have progression of AKI thus
allowing earlier intervention such as timely ini- Risk Stratification
tiation of RRT.
In an adult study, it was hypothesized that the The concept of renal angina was devised to
FST could predict which patients would have apply objectivity in the assessment of AKI risk
progression of AKI. This was done by measuring analogous to the components for angina pectoris
urine volume and flow after the administration of [54]. While there are no specific symptoms of
1–1.5 mg/kg of furosemide. The sum of the urine AKI such as chest pain for angina pectoris, clin-
volume at the first 2 h after receiving furosemide ical signs such as oliguria and fluid overload
had the best predictive ability for progression to were utilized. Renal angina therefore identifies
AKIN stage 3 within 14 days of performing the those at higher risk of AKI and guides the use of
FST. Urine volume of less than 200 mL at 2 h additional diagnostic evaluation for those who
offered the best sensitivity and specificity for will benefit from additional biomarker assays.
predicting AKI progression [51]. The FST was From this idea, the renal angina index (RAI)
combined with other AKI biomarkers in the pre- was derived, which is a product of AKI risk and
vious study cohort. The combination of the FST signs of injury, with a value of >/=8 as fulfill-
with uNGAL increased the prediction of pro- ment of renal angina (Fig. 10.1). This model
gression to AKIN stage 3, receipt of RRT during was found to be useful in the pediatric critical
admission, and inpatient death [52]. These find- care population of detecting likelihood of severe
ings suggest that in combination with uNGAL, AKI development 3 days post-ICU admission
the FST may improve risk stratification in early [55]. Additionally, urinary biomarkers com-
AKI. bined with RAI improved AKI prediction [56].
The FST was evaluated in neonates at risk for The authors illustrate that these findings provide
CS-AKI. Neonates and infants less than 90 days of a potential model for AKI risk stratification
age who received furosemide within 24 h of CPB upon early ICU admission (Fig. 10.2).
were included in a single-center, retrospective
study [53]. Hourly and cumulative urine output for
6 h after the initial postoperative furosemide dose Management Options
was evaluated. The maximum urine output occurred
in the first hour with almost half of the cumulative Understanding the etiology of AKI allows for the
urine output in the first 2 h with an average urine elimination of offending agents and reversible eti-
output of 1.6 mL/kg/h. Cumulative urine output ologic factors which include low oncotic pressure,
was lower in patients with CS-AKI. Furosemide low hydrostatic pressure, abdominal compartment
response had significant areas under the curve pre- syndrome, bladder obstruction, and nephrotoxic
dictive of CS-AKI, prolonged peritoneal dialysis, medications. To date, there are no medications or
prolonged mechanical ventilation, and peak fluid therapies that prevent or treat AKI, and manage-
overload greater than 15%. Unlike the adult data, a ment is largely based upon early detection,
specified cutoff point for cumulative urine output removal or mitigation of the offending agent and
after furosemide was unable to be determined for supportive therapy. Modifications that can improve
CS-AKI prediction. While prospective studies will outcomes in those with AKI include prevention of
be needed for validation of furosemide stress test- worsening kidney injury, nutrition optimization,
ing in this population, current data suggests com- minimization of fluid overload, and optimization
bining the FST with novel AKI b iomarkers may aid of acid/base and electrolyte balance.
ERRNVPHGLFRVRUJ
184 T. N. Webb et al.
Fig. 10.1 Renal angina index (RAI). (Copyright permis- eter between change in estimated creatinine clearance
sion obtained and adapted from Basu et al. [55]) from baseline and % fluid overload was used to yield an
Renal angina index (RAI) – Based on existing pediatric injury score. The RAI index score can range from 1 to 40.
AKI literature, tiered AKI risk strata were assigned point A cutoff value of >/= 8 is used to determine fulfillment of
values for “risk” and “signs” of injury. The worse param- renal angina (from Basu et al. [55] with permission)
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 185
Substitute avoidable
nephrotoxins
Limit nephrotoxins
Avoid Fluid Overload> 15%
Avoid contrast
Consider RRT Initiation
Hemodynamic support
AKI Biomarker
Assessment +
r
Renal Angina r ke
Assessment o ma
Bi
Conservative Fluid
Bi
in a+ om Management
A ng ar
ke
Adjust nephrotoxins
al r
Ren -
ICU
Admit Re
na
lA
ng
ina
-
Standard Management for Critically III Patients
Resuscitation
Stabilization
Fig. 10.2 Schema of use of RAI for AKI stratification Represented is a potential trial of prospective evaluation
after ICU admission. (Copyright permission obtained and on outcome based on the use of the RAI for AKI risk strat-
adapted from Menon et al. [56]) ification after ICU admission (from Menon et al. [56] with
permission)
ERRNVPHGLFRVRUJ
186 T. N. Webb et al.
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 187
ERRNVPHGLFRVRUJ
188 T. N. Webb et al.
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 189
gent fluid removal later on. New technology has made 14. Devarajan P, Jefferies JL. Progression of chronic kid-
significant advancements and promises to improve the ney disease after acute kidney injury. Prog Pediatr
safety of RRT in neonates or small infants who would Cardiol. 2016;41:33–40.
have otherwise not been eligible for these therapies. 15. Mammen C, Al Abbas A, Skippen P, Nadel H, Levine
D, Collet JP, et al. Long-term risk of CKD in chil-
dren surviving episodes of acute kidney injury in the
intensive care unit: a prospective cohort study. Am J
References Kidney Dis. 2012;59(4):523–30.
16. Goldstein SL, Jaber BL, Faubel S, Chawla LS. AKI
1. Bellomo R, Kellum JA, Ronco C. Acute kidney injury. transition of care: a potential opportunity to
Lancet. 2012;380(9843):756–66. detect and prevent CKD. Clin J Am Soc Nephrol.
2. Fortenberry JD, Paden ML, Goldstein SL. Acute kid- 2013;8(3):476–83.
ney injury in children: an update on diagnosis and 17. Mehrotra A, Rose C, Pannu N, Gill J, Tonelli M,
treatment. Pediatr Clin N Am. 2013;60(3):669–88. Gill JS. Incidence and consequences of acute kidney
3. Rewa O, Bagshaw SM. Acute kidney injury- injury in kidney transplant recipients. Am J Kidney
epidemiology, outcomes and economics. Nat Rev Dis. 2012;59(4):558–65.
Nephrol. 2014;10(4):193–207. 18. Askenazi DJ, Feig DI, Graham NM, Hui-Stickle S,
4. Susantitaphong P, Cruz DN, Cerda J, Abulfaraj M, Goldstein SL. 3–5 year longitudinal follow-up of
Alqahtani F, Koulouridis I, et al. World incidence pediatric patients after acute renal failure. Kidney Int.
of AKI: a meta-analysis. Clin J Am Soc Nephrol. 2006;69(1):184–9.
2013;8(9):1482–93. 19. Basu RK, Andrews A, Krawczeski C, Manning P,
5. Hoste EA, Bagshaw SM, Bellomo R, Cely CM, Wheeler DS, Goldstein SL. Acute kidney injury
Colman R, Cruz DN, et al. Epidemiology of acute based on corrected serum creatinine is associated
kidney injury in critically ill patients: the mul- with increased morbidity in children following the
tinational AKI-EPI study. Intensive Care Med. arterial switch operation. Pediatr Crit Care Med.
2015;41(8):1411–23. 2013;14(5):e218–24.
6. Sutherland SM, Ji J, Sheikhi FH, Widen E, Tian 20. Liu KD, Thompson BT, Ancukiewicz M, Steingrub
L, Alexander SR, et al. AKI in hospitalized chil- JS, Douglas IS, Matthay MA, et al. Acute kidney
dren: epidemiology and clinical associations injury in patients with acute lung injury: impact of
in a national cohort. Clin J Am Soc Nephrol. fluid accumulation on classification of acute kid-
2013;8(10):1661–9. ney injury and associated outcomes. Crit Care Med.
7. Kaddourah A, Basu RK, Bagshaw SM, Goldstein 2011;39(12):2665–71.
SL. Epidemiology of acute kidney injury in criti- 21. Prowle JR, Liu YL, Licari E, Bagshaw SM, Egi M,
cally ill children and young adults. N Engl J Med. Haase M, et al. Oliguria as predictive biomarker of
2017;376(1):11–20. acute kidney injury in critically ill patients. Criti Care.
8. Jetton JG, Boohaker LJ, Sethi SK, Wazir S, Rohatgi 2011;15(4):R172.
S, Soranno DE, et al. Incidence and outcomes of neo- 22. Ricci Z, Cruz D, Ronco C. The RIFLE criteria and
natal acute kidney injury (AWAKEN): a multicentre, mortality in acute kidney injury: a systematic review.
multinational, observational cohort study. Lancet Kidney Int. 2008;73(5):538–46.
Child Adolesc Health. 2017;1(3):184–94. 23. Wlodzimirow KA, Abu-Hanna A, Slabbekoorn M,
9. Jetton JG, Askenazi DJ. Acute kidney injury in the Chamuleau RA, Schultz MJ, Bouman CS. A compari-
neonate. Clin Perinatol. 2014;41(3):487–502. son of RIFLE with and without urine output criteria
10. Li S, Krawczeski CD, Zappitelli M, Devarajan P,
for acute kidney injury in critically ill patients. Criti
Thiessen-Philbrook H, Coca SG, et al. Incidence, risk Care. 2012;16(5):R200.
factors, and outcomes of acute kidney injury after 24. Macedo E, Malhotra R, Bouchard J, Wynn SK, Mehta
pediatric cardiac surgery: a prospective multicenter RL. Oliguria is an early predictor of higher mortality
study. Crit Care Med. 2011;39(6):1493–9. in critically ill patients. Kidney Int. 2011;80(7):760–7.
1 1. Morgan CJ, Zappitelli M, Robertson CM, Alton 25. Kellum JA, Sileanu FE, Murugan R, Lucko N, Shaw
GY, Sauve RS, Joffe AR, et al. Risk factors for AD, Clermont G. Classifying AKI by urine output
and outcomes of acute kidney injury in neonates versus serum creatinine level. J Am Soc Nephrol.
undergoing complex cardiac surgery. J Pediatr. 2015;26(9):2231–8.
2013;162(1):120–7.e1. 26. Akcan-Arikan A, Zappitelli M, Loftis LL, Washburn
12. Aydin SI, Seiden HS, Blaufox AD, Parnell VA,
KK, Jefferson LS, Goldstein SL. Modified RIFLE cri-
Choudhury T, Punnoose A, et al. Acute kidney injury teria in critically ill children with acute kidney injury.
after surgery for congenital heart disease. Ann Thorac Kidney Int. 2007;71(10):1028–35.
Surg. 2012;94(5):1589–95. 27.
Kidney Disease: Improving Global Outcomes
13. Andreoli SP. Acute kidney injury in children. Pediatr (KDIGO). Clinical practice guidelines for acute kid-
Nephrol. 2009;24(2):253–63. ney injury. Kidney Int. 2012;2(Suppl):19–36.
ERRNVPHGLFRVRUJ
190 T. N. Webb et al.
28. Sutherland SM, Byrnes JJ, Kothari M, Longhurst CA, 42. Han WK, Waikar SS, Johnson A, Betensky RA, Dent
Dutta S, Garcia P, et al. AKI in hospitalized children: CL, Devarajan P, et al. Urinary biomarkers in the
comparing the pRIFLE, AKIN, and KDIGO defini- early diagnosis of acute kidney injury. Kidney Int.
tions. Clin J Am Soc Nephrol. 2015;10(4):554–61. 2008;73(7):863–9.
29. Jefferies JL, Devarajan P. Early detection of acute kid- 43. Zwiers AJ, de Wildt SN, van Rosmalen J, de Rijke
ney injury after pediatric cardiac surgery. Prog Pediatr YB, Buijs EA, Tibboel D, et al. Urinary neutrophil
Cardiol. 2016;41:9–16. gelatinase-associated lipocalin identifies critically ill
30. Bennett M, Dent CL, Ma Q, Dastrala S, Grenier F, young children with acute kidney injury following
Workman R, et al. Urine NGAL predicts severity of intensive care admission: a prospective cohort study.
acute kidney injury after cardiac surgery: a prospec- Criti Care. 2015;19:181.
tive study. Clin J Am Soc Nephrol. 2008;3(3):665–73. 44. Melnikov VY, Faubel S, Siegmund B, Lucia MS,
31. Mori K, Lee HT, Rapoport D, Drexler IR, Foster Ljubanovic D, Edelstein CL. Neutrophil-independent
K, Yang J, et al. Endocytic delivery of lipocalin- mechanisms of caspase-1- and IL-18-mediated isch-
siderophore- iron complex rescues the kidney emic acute tubular necrosis in mice. J Clin Invest.
from ischemia-reperfusion injury. J Clin Invest. 2002;110(8):1083–91.
2005;115(3):610–21. 45. Klein SJ, Brandtner AK, Lehner GF, Ulmer H,
32. Mishra J, Ma Q, Prada A, Mitsnefes M, Zahedi K, Bagshaw SM, Wiedermann CJ, et al. Biomarkers for
Yang J, et al. Identification of neutrophil gelatinase- prediction of renal replacement therapy in acute kid-
associated lipocalin as a novel early urinary bio- ney injury: a systematic review and meta-analysis.
marker for ischemic renal injury. J Am Soc Nephrol. Intensive Care Med. 2018;44(3):323–36.
2003;14(10):2534–43. 46. Parikh CR, Devarajan P, Zappitelli M, Sint K,
33. Krawczeski CD, Woo JG, Wang Y, Bennett MR, Ma Thiessen-Philbrook H, Li S, et al. Postoperative
Q, Devarajan P. Neutrophil gelatinase-associated lipo- biomarkers predict acute kidney injury and poor
calin concentrations predict development of acute outcomes after pediatric cardiac surgery. J Am Soc
kidney injury in neonates and children after cardiopul- Nephrol. 2011;22(9):1737–47.
monary bypass. J Pediatr. 2011;158(6):1009–15.e1. 47. Washburn KK, Zappitelli M, Arikan AA, Loftis L,
34. Dong L, Ma Q, Bennett M, Devarajan P. Urinary bio- Yalavarthy R, Parikh CR, et al. Urinary interleukin-18
markers of cell cycle arrest are delayed predictors of is an acute kidney injury biomarker in critically ill chil-
acute kidney injury after pediatric cardiopulmonary dren. Nephrol Dial Transplant. 2008;23(2):566–72.
bypass. Pediatr Nephrol. 2017;32(12):2351–60. 48. Krawczeski CD, Goldstein SL, Woo JG, Wang Y,
35. Patel ML, Sachan R, Shyam R, Kumar S, Kamal Piyaphanee N, Ma Q, et al. Temporal relationship and
R, Misra A. Diagnostic accuracy of urinary neutro- predictive value of urinary acute kidney injury bio-
phil gelatinase-associated lipocalin in patients with markers after pediatric cardiopulmonary bypass. J Am
septic acute kidney injury. Int J Nephrol Renov Dis. Coll Cardiol. 2011;58(22):2301–9.
2016;9:161–9. 49. Parikh CR, Thiessen-Philbrook H, Garg AX, Kadiyala
36. Forster CS, Jackson E, Ma Q, Bennett M, Shah SS, D, Shlipak MG, Koyner JL, et al. Performance of kid-
Goldstein SL. Predictive ability of NGAL in identify- ney injury molecule-1 and liver fatty acid-binding
ing urinary tract infection in children with neurogenic protein and combined biomarkers of AKI after cardiac
bladders. Pediatr Nephrol. 2018;33(8):1365–74. surgery. Clin J Am Soc Nephrol. 2013;8(7):1079–88.
37. Varnell CD Jr, Goldstein SL, Devarajan P, Basu
50. Meersch M, Schmidt C, Van Aken H, Martens S,
RK. Impact of near real-time urine neutrophil Rossaint J, Singbartl K, et al. Urinary TIMP-2 and
gelatinase-associated lipocalin assessment on clinical IGFBP7 as early biomarkers of acute kidney injury
practice. Kidney Int Rep. 2017;2(6):1243–9. and renal recovery following cardiac surgery. PLoS
38. Lameire N, Vanholder R, Van Biesen W, Benoit
One. 2014;9(3):e93460.
D. Acute kidney injury in critically ill cancer patients: 51. Chawla LS, Davison DL, Brasha-Mitchell E, Koyner
an update. Criti Care. 2016;20(1):209. JL, Arthur JM, Shaw AD, et al. Development and
39. Zhang Z, Lu B, Sheng X, Jin N. Cystatin C in pre- standardization of a furosemide stress test to pre-
diction of acute kidney injury: a systemic review and dict the severity of acute kidney injury. Criti Care.
meta-analysis. Am J Kidney Dis. 2011;58(3):356–65. 2013;17(5):R207.
40. Zappitelli M, Krawczeski CD, Devarajan P, Wang Z, 52.
Koyner JL, Davison DL, Brasha-Mitchell E,
Sint K, Thiessen-Philbrook H, et al. Early postop- Chalikonda DM, Arthur JM, Shaw AD, et al.
erative serum cystatin C predicts severe acute kidney Furosemide Stress Test and Biomarkers for the
injury following pediatric cardiac surgery. Kidney Int. Prediction of AKI Severity. J Am Soc Nephrol.
2011;80(6):655–62. 2015;26(8):2023–31.
41. Basu RK, Wong HR, Krawczeski CD, Wheeler DS, 53. Borasino S, Wall KM, Crawford JH, Hock KM,
Manning PB, Chawla LS, et al. Combining functional Cleveland DC, Rahman F, et al. Furosemide response
and tubular damage biomarkers improves diagnostic predicts acute kidney injury after cardiac surgery
precision for acute kidney injury after cardiac surgery. in infants and neonates. Pediatr Crit Care Med.
J Am Coll Cardiol. 2014;64(25):2753–62. 2018;19(4):310–7.
ERRNVPHGLFRVRUJ
10 Diagnosis and Management of Acute Kidney Injury in Critical Illness 191
ERRNVPHGLFRVRUJ
Management of Fluid Overload
in the Pediatric ICU 11
Grace L. Ker and Sandeep Gangadharan
Introduction Pathophysiology
The clinical implications of fluid overload and its Some of the difficulties surrounding fluid man-
management currently remain controversial agement in critically ill patients arise from our
within the pediatric critical care literature. incomplete understanding of the movement and
Resuscitation with intravenous fluids is often distribution of fluid within the body, particularly
necessary in critical care for maintenance of per- in the setting of systemic disease states and thera-
fusion during shock or hypovolemic states. In peutic agents that impact endothelial permeabil-
recent years, a shift toward early recognition of ity and vascular autoregulation.
sepsis has encouraged rapid and early initiation Total body water accounts for approximately
of fluid at illness presentation with the hopes of 60% of body weight and is distributed between
improving outcomes. As a consequence, aggres- the intracellular space (ICS) and the extracellular
sive early fluid resuscitation can contribute to space (ECS). The percentage of water per body
fluid overload states in the setting of critical ill- weight fluctuates with age, such that premature
ness. Recent literature in both adult and pediatric infants and term newborns have higher total body
patients suggests that fluid overload in critical ill- water relative to adults, which rapidly decreases
ness can have harmful effects and contribute to to adult values during the first year of life [2]. The
morbidity and mortality [1]. However, the defini- intracellular space accounts for 55% of total body
tion and management of fluid overload remain water, while the remaining 45% exists within the
unclear, likely reflecting the complex relation- ECS. The ECS is further divided into three com-
ship between volume resuscitation, fluid distribu- partments: intravascular space or plasma (IVS),
tion, and critical illness. interstitial space (ISS), and transcellular space
(TCS), which account for 15%, 45%, and 40%,
respectively, of total extracellular fluid. The tran-
scellular space refers to digestive, cerebrospinal,
intraocular, pleural, peritoneal, and synovial flu-
ids [3]. The movement of water between the three
compartments of the extracellular space occurs
through a semipermeable membrane [4].
G. L. Ker (*) · S. Gangadharan The homeostatic state of fluid movement
Department of Pediatric Critical Care, Cohen within the body is determined by a balance
Children’s Medical Center, between the physical properties between the
New Hyde Park, NY, USA
ERRNVPHGLFRVRUJ
194 G. L. Ker and S. Gangadharan
Arterial End
Capillary
Capillary
Hydrostatic
Fig. 11.1 Diagram
Pressure (Pc)
demonstrating
relationship between
hydrostatic pressures
within the capillary bed Interstitial
at the arterial end. Hydrostatic
(Image created by Grace Pressure
Filtration
Ker)
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 195
Interstitial
Reabsorption
Oncotic
Pressure
In the precapillary arterioles, the hydrostatic In the postcapillary venules, the oncotic pressure
pressure is greater than the oncotic pressure, is greater than the hydrostatic pressure, favouring
favouring the movement of water into the the movement of water out of the interstitial fluid
interstitial fluid and back into the venules
Oncotic pressure
Relative pressure
Hydrostatic pressure
Movement of water
Venule
Arteriole
Fig. 11.3 Image demonstrating the Starling principle for oncotic pressure is greater than hydrostatic pressure
fluid movement. At the arterial end, hydrostatic pressure which then favors the movement of fluid into the vascula-
exceeds oncotic pressure with relative movement of water ture via absorption. (“Image courtesy of Alex Yartsev,
into the interstitial fluid via filtration. At the venule end, www.derangedphysiology.com”)
ERRNVPHGLFRVRUJ
196 G. L. Ker and S. Gangadharan
to overcome hydrostatic forces. Therefore, there endothelial glycocalyx layer serves as a semiper-
is no true reversal of filtration. Instead, filtered meable membrane separating plasma from inter-
fluid circulates throughout the interstitial space stitial space. Rather than gradients between the
and largely returns to the circulation as lymph interstitial oncotic pressure and intravascular
[5]. This principle for fluid movement is influ- oncotic pressure that serve as a driving force for
enced by substances that exist in the multiple lay- fluid movement, it is the subglycocalyx capillary
ers between the interstitial space and intravascular oncotic pressure (πs) that serves as the major
space [5]. These layers include a subglycocalyx determinant in transcapillary flow (Jv) [5]. This
space which is largely protein-free and an endo- model and the function of the endothelial glyco-
thelial glycocalyx space which consists of a calyx in fluid movement are illustrated in
matrix of glycoproteins and proteoglycans. The Fig. 11.4. Further research into the translation of
Part 1
Interstitial fluid
Endothelial cells
Subglycocalyx space:
Gaps between the
endothelial cells, and a
potential space under the
glycocalyx
Glycocalyx
Intravascular fluid
Fig. 11.4 Image demonstrating the role of the endothe- throughout the vessel does not exceed hydrostatic pres-
lial glycocalyx in fluid exchange (part 1) and revised sure with no true reversal of filtration occurring. (“Image
model of the Starling principle (part 2) involving the sub- courtesy of Alex Yartsev, www.derangedphysiology.
glycocalyx illustrating the increase in oncotic pressure com”)
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 197
Part 2
Movement of water
Arteriole
Venule
Fig. 11.4 (continued)
this model into clinical practice could lead to ultimately leads to hypervolemia and fluid
improvements in the management of fluid status overload.
in critically ill patients. Fluid overload states can impact organ func-
Fluid management strategies, particularly in tion influencing morbidity and mortality.
states of shock, are often complicated by the need Excessive fluid can alter the efficiency of the
to balance maintenance of oxygen delivery to heart by moving ventricular compliance far to the
match metabolic demand and maintain cardiac right on the Frank-Starling curve, potentially
output without avoiding fluid overload. Shock resulting in impaired cardiac function.
states and critical illness often have underlying Hypervolemia can cause pulmonary alveolar and
systemic inflammation resulting in anasarca and interstitial pulmonary edema, precipitating respi-
capillary leak, altering normal fluid movement ratory failure [7]. Specifically, worsening pulmo-
between compartments [6]. Dysregulation of nary edema will negatively affect lung
fluid distribution within intracellular and extra- compliance, potentiate V/Q mismatch, and, con-
cellular compartments, along with aggressive sequently, worsen both ventilation and oxygen-
fluid resuscitation and impaired renal clearance, ation. Finally, particularly during stress states,
ERRNVPHGLFRVRUJ
198 G. L. Ker and S. Gangadharan
renal perfusion may be impaired causing the 2001, Goldstein et al. developed definitions for
development of acute kidney injury (AKI), which fluid overload that relied upon measuring fluid
further complicates the ability of kidneys to input and output from PICU admission [9]. This
maintain euvolemia. method for calculation of fluid overload is dem-
An understanding of fluid distribution and onstrated below:
movement in the body therefore becomes critical
Sum of daily fluid
during periods of fluid resuscitation and over-
load. Management strategies targeting such ( fluid in - fluid out )
%FO = ´100.
pathophysiologic mechanisms are designed to ICU admission weighht
help reduce morbidity and mortality associated
The majority of pediatric studies since then
with fluid overload and multiorgan dysfunction.
have largely utilized this definition for the pur-
poses of research and clinical identification.
However, this method is potentially error prone
Measurement of Fluid Overload as it relies upon precise accounting of daily fluid
balance and accurate calculations. As an alterna-
tive method, fluctuations in weight have been
Case
used in some studies. For example, in a study of
A 12-year-old female presents to the emer- patients receiving continuous renal replacement
gency room with septic shock in the setting therapy (CRRT), measurements of weight at hos-
of acute influenza infection requiring intu- pital or ICU admission and initiation of CRRT
bation and aggressive fluid resuscitation. were utilized to determine fluid overload percent-
Four days after presentation, her hemody- age with formulas as demonstrated below [8]:
namics have improved, and she is able to
be weaned off inotropic support without CRRT initiation weight - ICU
difficulty. Her chest X-ray is significant for admission weight
%FO = ´100
pulmonary edema. A central venous cathe- ICU admission weight
ter within the internal jugular vein mea-
sures a central venous pressure of 5 OR
cmH2O. Her weight is 2 kg above her base- CRRT initiation weight - Hospital
line, but her intake/output flowsheet shows admission weight
%FO = ´100.
that she is overall in a negative fluid bal- Hospitaal admission weight
ance of 1 L for the duration of her hospital-
ization. Among the physician team caring These calculations were compared with the
for her, there is disagreement about her standard method of intake and output assessment
hydration status. Some physicians feel that and were found to have a high degree of correla-
she is euvolemic, while others feel that she tion with similar predictive ability for mortality
is grossly fluid overloaded. Possible man- [8]. Therefore, it is possible the two methods
agement pathways are discussed at length. could be utilized interchangeably in the clinical
A nurse asks if there is one clear best setting, though the method of utilizing daily
method to evaluate for fluid overload in a weights rather than summation of daily fluid
critically ill pediatric patient. intake is likely the less labor-intensive means of
determining fluid overload.
Despite the fact that both calculations have
been validated and are relatively easy to perform,
Despite an acknowledgement that fluid over- discrepancies between weight measurements and
load can be associated with poor outcomes in intake/output calculations often occur, thereby
critical care populations, the optimal clinical def- complicating management decisions. These dis-
inition of fluid overload remains unclear [8]. In crepancies likely result from the limitations of
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 199
these two methods. For instance, fluid-based cal- substituted as a reliable method to measure fluid
culations are unable to fully account for insensi- overload. Regardless of the method of fluid over-
ble losses, and weight-based calculations are load measurement, interpretation of fluid status
vulnerable to errors that can occur through the still requires clinician expertise, particularly in
use of different scales or weighing techniques. settings where physical exam and calculated val-
One relatively novel method devised to address ues are discrepant. The child in the scenario above
such issues involves the use of bioimpedance as a experienced a significant inflammatory response
measurement tool to determine fluid overload and capillary leak secondary to her infectious pro-
status. cess which predisposed her to pulmonary edema
Bioimpedance utilizes electrical properties of and fluid overload. It is not uncommon for criti-
biological tissues in evaluating their response to cally ill patients to have intravascular volume
electrical current with either active (in which depletion concurrent with total body fluid over-
electrostimulation triggers ionic activities inher- load. Therefore, despite signs of fluid overload on
ent within tissue cells) or passive response (i.e., clinical examination, if the patient were to remain
response only occurs with external stimulation) hypotensive or unable to tolerate diuresis, it would
[10]. The ability of tissues to impede frequency be suggestive of ongoing low intravascular vol-
signals is then extrapolated mathematically to ume despite overall fluid overload which would
calculate the composition of the tissue. While it alter clinical decision-making regarding the tim-
has a variety of medical applications, bioimped- ing of diuresis or fluid removal therapies.
ance has been proposed as a method to evaluate
for fluid status due to its ability to estimate total
body fluid and extracellular fluid. However, anaging Fluid Overload
M
although multifrequency bioimpedance has been with Renal Replacement Versus
reported as a reliable measure of estimating body Diuretic Therapy
water compartments in adults [10], the pediatric
literature to support its use remains sparse. Milani
et al. recently attempted to evaluate the use of Case
multifrequency bioimpedance measurements in A 16-year-old female presents with septic
pediatric patients requiring maintenance dialysis shock in the setting of Staphylococcus
and found that measurements of body water com- aureus infection requiring aggressive fluid
partments were imprecise when compared with resuscitation and vasoactive support in the
gold standard measurements of bromide or deu- first 24 h. The next day, her hemodynamics
terated water dilution, two methods which, while have improved, and she is able to be
accurate, remain impractical for daily clinical use weaned off inotropic support, though ana-
[10]. Although bioimpedance may serve as a sarca is noted on physical exam and pul-
potential noninvasive alternative to assess for monary edema is seen on chest radiography.
fluid status in the future, its reliability in critically An attending physician who is rounding
ill pediatric populations has yet to be realized. suggests initiation of diuresis with furose-
The determination of fluid overload in the criti- mide, but another attending is concerned
cal care setting depends upon the accurate assess- that aggressive medical diuresis may
ment of intake and output from admission and worsen the patient’s hemodynamic status
relating this information to each patient’s baseline and instead recommends initiation of renal
weight. Our recommendation is to use the replacement therapy, specifically continu-
Goldstein method of daily intake and output in ous veno-venous hemodiafiltration. A resi-
comparison from admission weight to evaluate dent asks if any method of fluid removal has
for percentage of fluid overload [9]. When fluid been proven to result better outcomes in
intake or output measurements are unavailable or critically ill children.
incomplete, daily weight measurements can be
ERRNVPHGLFRVRUJ
200 G. L. Ker and S. Gangadharan
Currently, options for managing fluid overload spective randomized controlled trial continuous
are limited to fluid restriction, medical manage- versus intermittent furosemide in postoperative
ment with diuretics, and/or renal replacement pediatric cardiac patients. Patients were random-
therapy (RRT). The decision regarding which to ized to either IV intermittent (1 mg/kg furose-
initiate in the setting of fluid overload is often mide every 4 h to be increased by 0.25 mg/kg
influenced by multiple extrinsic factors including every 4 h to a maximum of 1.5 mg/kg) or con-
physician preference, ease of access, size of the tinuous IV infusion (0.1 mg/kg/h of furosemide
patient, hemodynamic stability, and bleeding risk. doubled every 2 h to a maximum of 0.4 mg/kg/h)
to maintain urine output >1.0 mL/kg/h. Data
showed that those who received the continuous
Diuretic Therapy infusion had less variability in urine output and
required a lower cumulative daily dose of furose-
In fluid overload, diuretics are used to maximize mide, suggesting that continuous infusions may
renal salt and water excretion in the setting of be more advantageous in this patient population
volume overload. Diuretics are also used in acute [13]. Klinge et al. attempted to replicate this eval-
renal failure with the intention of converting oli- uation with 57 postoperative cardiac patients in a
guric to non-oliguric acute renal failure. Non- prospective, randomized study. Patients were
oliguric renal failure has lower associated given either intermittent IV furosemide or a con-
morbidity and may prevent the need for future tinuous infusion of furosemide when urine output
dialysis [11]. However, the use of diuretics is fell to less than 1 mL/kg/h and CVP was more
associated with certain risks including electrolyte than 5 cmH2O. Patients in the intermittent IV
derangements, ototoxicity among the neonatal furosemide group required less furosemide over
population, and AKI especially when used in the 3-day study period (1.2 mg/kg/day vs 1.8 mg/
conjunction with radiocontrast agents. kg/day) to achieve the same, targeted urine vol-
In patients with AKI, the effect of loop diuret- ume of >1 mL/kg/h [14]. Continuous infusions of
ics may be blunted secondary to reduced tubular furosemide result in more consistent hourly urine
secretion [12]. Therefore, it has been hypothe- output and may be preferred to intermittent dos-
sized that a lack of diuretic responsiveness may ing in hemodynamically unstable patients.
correlate with the degree of acute kidney injury. Alternative diuretic choices to furosemide are
Kakajiwala et al. retrospectively evaluated infants another area of ongoing study among the pediat-
after congenital heart surgery and found that ric population. Ethacrynic acid, which can be
responsiveness to furosemide was lower among used as an alternative to other loop diuretics in
patients with AKI. Their study defined lack of the setting of sulfa drug allergies, has been stud-
furosemide responsiveness as urine output of less ied dating as early as the 1960s. The mechanism
than 1.7 mL/kg/h at 2 h or 1.9 mL/kg/h at 6 h of action is identical to furosemide and other loop
after diuretic administration. After correcting for diuretics – direct reversible blockade of Na/K/2Cl
fluid balance, it was found that lack of furose- binding sites in the thick ascending loop of Henle.
mide responsiveness was predictive of Ethacrynic acid has been reported to be 30% less
AKI. While diuretic use may be beneficial in the potent than furosemide [15]. Ethacrynic acid was
management of fluid overload, a lack of respon- studied in 22 children aged 2–17 years of age
siveness to diuretic therapy may be reflective of with congestive heart failure. They reported 3.6%
concurrent renal injury which can complicate average weight loss in patients administered with
clinical course and management. ten intermittent doses of ethacrynic acid, with an
At the moment, diuretics remain a mainstay of increase in urinary volume by a factor of 2.5 over
therapy given that worsening fluid overload con- a 24-h period and with minimal electrolyte abnor-
tributes to increased mortality and morbidity, but malities [16]. In 2015, Ricci et al. performed a
the ideal method of administration remains prospective randomized double-blinded study on
unclear. In 1992, Singh et al. evaluated in a pro- 74 pediatric patients undergoing elective cardiac
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 201
surgery comparing furosemide versus ethacrynic glomerular arterioles and increasing glomerular
acid. Patients were included in the study if they filtration rate (GFR) [18]. Pediatric and neonatal
had clinical signs of fluid overload and random- studies confirm benefits of adjunctive therapy of
ized to either receive furosemide or ethacrynic aminophylline when used in combination with
acid, with primary outcome measuring mean furosemide therapy. Pretzlaff et al. administered a
urine output on postoperative day 0. Urine output bolus of 6 mg/kg of aminophylline to pediatric
was noted to be higher among patients who patients aged 2–46 months of age with fluid over-
received ethacrynic acid (6.9 mL/kg/h) compared load who were concurrently treated with furose-
with furosemide (4.6 ml/kg/h, p = 0.002) despite mide infusions of ≥6 mg/kg/day. Theophylline
a lower cumulative dose of ethacrynic acid levels were measured as peak levels 30 min after
(0.22 mg/kg/h vs 0.33 mg/kg/h, p < 0.0001), sug- administration. The mean peak level was 8.3 mcg/
gesting that less ethacrynic acid may be required mL, and urine output was positively correlated
to achieve the same urine output as furosemide in with theophylline level. After administration,
this patient population. The occurrence of AKI patients were noted to have a significant increase
was not statistically significant between groups, in urine output (>80%) during the first 2 h [19]. A
and no complications linked to electrolyte disor- neonatal case study described five infants,
ders were observed in either group. Based on 26–38 weeks of age, receiving furosemide infu-
these data, it can be concluded that ethacrynic sions at 0.2–0.6 mg/kg/h with ongoing oliguria
acid is at least as efficacious as furosemide in who were loaded with 4 mg/kg of aminophylline
children recovering from cardiac surgery and and treated with increasing doses until diuresis
may have a role in other critically ill patient pop- was achieved. All infants demonstrated weight
ulations such as those with sulfa allergies. It is loss, reduction of mean airway pressure, and FiO2
important to note however that ethacrynic acid is requirements, along with the resolution of pleural
currently markedly more expensive than furose- effusions. Aminophylline may therefore be help-
mide which, in the current health-care climate, ful for patients with ongoing fluid overload
will be a barrier to its use. despite loop diuretic therapy. For patients who
Additional diuretic options include bumetanide receive multiple doses of aminophylline in this
(a loop diuretic) and thiazide diuretics such as setting, monitoring of serum concentrations based
chlorothiazide and metolazone, which act to on traditional protocols to prevent toxicity would
inhibit Na/Cl co-transporter channels in the proxi- be prudent.
mal part of the distal convoluted tubule. This latter
class of diuretics, when used in conjunction with
a loop diuretic, may provide a synergistic effect, Renal Replacement Therapy
given that the mechanisms of action of the two
classes of drugs are different. Serum electrolytes, The utility and timing of RRT for fluid overload
especially sodium, potassium, and chloride, remain controversial. The majority of studies that
should be monitored frequently, as the risk of have evaluated the use of RRT as a therapeutic
electrolyte abnormalities increases when these modality in critically ill children have done so in
two classes of drugs are used concurrently [17]. the setting of an underlying diagnosis of acute
Aminophylline has been described as adjunc- kidney injury and other concurrent comorbidi-
tive therapy for diuresis. Aminophylline is a ties. Fluid overload frequently occurs in conjunc-
methylxanthine that functions as a diuretic by act- tion with renal disease associated with multiorgan
ing as an adenosine receptor antagonist, which dysfunction and is intimately intertwined physi-
serves to increase renal blood flow and inhibit sol- ologically with acute kidney injury. RRT for fluid
ute reabsorption. In cases of oliguria refractory to overload within the critical care environment is
traditional loop diuretics, there is some thought initiated in the setting of AKI or ARF in approxi-
that aminophylline may help increase efficacy of mately 4–8% of all intensive care patients
loop diuretics by promoting dilation of afferent [20–22].
ERRNVPHGLFRVRUJ
202 G. L. Ker and S. Gangadharan
Although there are multiple modalities of RRT In summary, fluid overload greater than 10%
available (e.g., peritoneal dialysis, intermittent has been associated with increased mortality
hemodialysis, or continuous hemofiltration), use and poor outcomes in pediatric and adult
of continuous veno-venous hemofiltration patients with critical illness. Diuretic medica-
(CVVH) or continuous veno-venous hemodiafil- tions, with furosemide generally accepted as the
tration (CVVHDF) has become increasingly pre- first-line agent, can be beneficial in improving
ferred by pediatric critical care providers [1]. urine output. Method of administration (contin-
Implementation of RRT for critically ill children uous vs intermittent IV) can be based on pro-
who become anuric or develop accepted criteria vider preference and patient factors (e.g.,
for dialysis (i.e., life-threatening hyperkalemia or hemodynamic stability, available intravenous
uremia) secondary to AKI is the standard of care access sites, etc.). Adjunctive treatment with
and, accordingly, not controversial. Use of RRT, aminophylline may be considered in refractory
however, in the setting of fluid overload in patients oliguria in patients already receiving traditional
without anuria or other definite criteria for dialy- diuretic therapies. While current literature does
sis is more debatable. Studies in pediatric patients not support RRT implementation based solely
have shown that the percentage of fluid overload on the percentage of fluid overload, it should be
prior to initiation of therapy is independently considered as therapeutic option for fluid
associated with survival in all patients receiving removal in patients with persistent or worsening
CVVH or CVVHDF [1, 20, 23]. For example, in fluid overload greater than 10% of pre-illness
one retrospective study of critically ill children weight despite maximal medical therapy. As the
with multiorgan dysfunction, fluid overload was use of CVVH and CVVHDF becomes more
independently associated with mortality; fluid commonplace, improved research will hope-
overload was 15.5% in non-survivors versus 9.2% fully allow for better developed algorithms to
in survivors (p < 0.01) [1]. Adult studies have dictate fluid removal therapies in the critical
shown similar findings. In one prospective study care setting.
of adult patients, 90-day mortality was doubled
among those who had fluid overload at the time of
RRT initiation (identified as greater than 10% luid Overload and Mechanical
F
fluid overload) [20]. Another adult study demon- Ventilation
strated that increased positive mean daily fluid
balance after RRT initiation was also associated
with increased mortality [23]. Despite these stud- Case
ies and others on the use of RRT with AKI and A 4-month-old child presents to the hospi-
fluid overload, the ideal timing for initiation of tal with respiratory distress in the setting of
RRT therapies remains unclear. Perhaps more enterovirus-induced pneumonitis. At the
importantly, it has not been clearly demonstrated time of initial presentation, he is noted to
that RRT utilization for fluid overload leads to be tachycardic with poor urine output for
improvement in survival outcomes. A meta-analy- which he receives aggressive fluid resusci-
sis in 2008 evaluating all randomized controlled tation. His respiratory status continues to
trials available in adult literature regarding the uti- worsen, and by day 4 of illness, he is endo-
lization of RRT demonstrated a mortality risk tracheally intubated and placed on
reduction that was inadequately powered to dem- mechanical ventilation. His chest X-ray
onstrate statistical significance [24]. A similar shows increased interstitial markings con-
systematic review of adult data performed in 2011 sistent with pulmonary edema. The resident
looked to compare early versus late initiation of asks if the fluid resuscitation contributed to
RRT and showed an association with reduced his decompensation and asks if this will
mortality, but results were complicated by signifi- significantly impact his recovery.
cant statistical heterogeneity [25].
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 203
Fluid overload and respiratory compromise that fluid overload was independently associated
are often connected, particularly in conditions with compromised respiratory function.
such as sepsis in which systemic inflammation Specifically, peak fluid overload correlated sig-
and endothelial injury lead to significant capil- nificantly with peak oxygenation index (OI) and
lary leak and extravasation of fluid into the inter- Pediatric Logistic Organ Dysfunction (PELOD)
stitial space. In recent years, an emphasis on score [26]. In a single-center retrospective review
increased early fluid resuscitation for septic of 636 mechanically ventilated children, percent
shock and the emergence of sepsis protocols with fluid overload correlated with number of invasive
early goal-directed therapy have created the ventilation days and OI at 48 h of mechanical
potential for patients to experience more clini- ventilation [30]. There are a small number of
cally significant long-term fluid overload. studies that did not identify a relationship
Stabilizing hemodynamics with aggressive between fluid overload and respiratory morbidity.
fluid resuscitation may come at a cost. Emerging One such example is a post hoc analysis of a
literature shows that prolonged positive fluid bal- study conducted by the Pediatric Acute Lung
ance after hemodynamic stability has been Injury and Sepsis Investigators (PALISI) that did
achieved can have a negative impact on organ not show an association between cumulative fluid
function and respiratory recovery. These findings balance and other respiratory measures such as
are present even when correcting for the underly- duration of ventilator weaning and extubation
ing severity of illness at the time of presentation. outcomes, though this analysis included all
For example, fluid overload ≥15% has been inde- mechanically ventilated patients rather than the
pendently associated with prolonged ventilation subset with ALI or ARDS [31].
and increased hospital length of stay [26]. While there may be some conflicting data
Fluid overload has been associated with on the relationship between fluid overload and
increased mortality and duration of mechanical respiratory morbidity, the overwhelming major-
ventilation in children identified with acute lung ity of published studies examining this relation-
injury (ALI), [27] a finding which was also ship has identified negative consequences of
observed in the pediatric arm of the Calfactant in this relationship. For this reason, some authors
Acute Respiratory Distress Syndrome trial [28]. have hypothesized that active avoidance of early
This trial was a multicenter prospective evalua- fluid overload may be an appropriate therapeu-
tion of intratracheally administered surfactant tic goal for patients with underlying critical
use in adult and pediatric patients diagnosed with respiratory illness to reduce time on mechanical
ARDS/ALI. The pediatric arm consisted of 24 ventilation [32]. In a prospective randomized
children’s hospitals across 6 countries and evalu- adult trial comparing conservative (i.e., main-
ated all-cause mortality at 90 days after study tenance of central venous pressure ≤ 4 mmHg)
entry, with secondary outcomes of oxygenation versus liberal (i.e., maintenance of central
changes after intervention, ventilator-free days at venous pressure 10–14 mmHg) fluid strategies
28 days, and total duration of ICU stay [29]. in patients with acute lung injury, oxygenation
Greater accumulated positive fluid balance from index and lung injury score were lower, and
day 1 to day 7 of illness was associated with in- the number of ventilator free days was higher
hospital mortality (p < 0.001), with non-survivors in patients who received the conservative fluid
averaging 8.7 ± 9.5 L/m2 compared with strategy. There was no statistical difference
1.2 ± 2.4 L/m2 in survivors. Increased cumulative in the primary endpoint of 60-day mortality:
fluid overload was also associated significantly 25.5% versus 28.4% in the conservative and
with less ventilator-free days and longer dura- liberal arms of the study, respectively [33]. To
tions of PICU and hospital stay [28]. our knowledge, similar pediatric studies have
In one of the few studies that has examined not yet been conducted.
fluid overload in the pediatric setting in the Outcomes for patients requiring mechanical
absence of RRT, Arikan and colleagues found ventilation are dependent on multiple factors,
ERRNVPHGLFRVRUJ
204 G. L. Ker and S. Gangadharan
and there are enough published data to con- with cyanotic heart lesions and/or worse kidney
clude that fluid overload likely contributes to function at baseline [35]. Many of the children
worse outcomes in these patients. Moreover, with these risk factors also have complex lesions
there are no studies that have demonstrated any requiring longer CPB durations. Prolonged expo-
benefit of fluid overload on respiratory function sure to CPB and the stress of surgery augments
or outcomes. Recent reviews of ALI and ARDS the postoperative inflammatory response and
management in pediatric patients have yet to consequent capillary leak, likelihood of low car-
make recommendations regarding concurrent diac output, and AKI, making fluid overload
fluid management strategies [34]. Despite the almost unavoidable in some cases. In a recent ret-
current lack of guidelines, it is reasonable to rospective review of 193 children who underwent
conclude that careful attention to fluid manage- cardiac surgery, the amount of fluid administered
ment and avoidance of fluid overload whenever in the first 6 h of ICU admission after surgery was
possible in patients with respiratory disease are the strongest independent risk factor for fluid
warranted. overload by postoperative day 2 [36]. In one of
the largest prospective studies to date focused on
this issue, which examined characteristics and
Fluid Overload and Cardiac Surgery outcomes of 1525 children who underwent car-
diac surgery over a 4-year period at a single insti-
tution, Lex et al. noted that patients who were
Case more likely to develop fluid overload greater than
A 4-month-old female undergoes surgical 5% were significantly younger (median 151 days
repair of tetralogy of Fallot. She returns versus 448 days) and smaller (4.3 kg versus
from the operating room endotracheally 8.4 kg) and had longer duration of CPB (131 min
intubated on mechanical ventilation. Her versus 78 min) and aortic cross clamp (69 min
first night of hospitalization is complicated versus 39 min) and lower urine output on postop-
by significant hemodynamic instability erative day 1 (1.7 mL/kg/h versus 2.6 mL/kg/h)
necessitating frequent fluid boluses. Over [37]. A summary of all potential risk factors for
the next day, she becomes progressively fluid overload is listed in Table 11.2.
more fluid overloaded with diffuse ana- Multiple studies have found associations of
sarca and worsening pulmonary edema. fluid overload with worse outcomes after pediat-
The resident asks if there are specific risks ric cardiac surgery [38]. For instance, a retro-
of fluid overload in the postoperative car-
diac patient when compared with other
patients. Table 11.2 Risk factors for fluid overload in cardiac
surgery
Preoperative
• Fluid overload secondary to heart failure
Pediatric patients undergoing cardiac surgery • Diminished baseline renal function
represent a unique population in which their • Younger age
pathophysiology predisposes them to fluid accu- • Lower body weight
mulation in the postoperative period. This predis- • Cyanotic congenital heart disease
position can vary based on many factors including Intraoperative
the age of the patient and the underlying comor- • Excessive exogenous fluid and blood products for
during surgery
bidities that may be present. Neonates are at par-
• Longer duration of cardiopulmonary bypass and
ticular risk to develop fluid overload due to their aortic cross clamp
increased total body water distribution and renal Postoperative
immaturity. Indeed, fluid overload can occur • Acute kidney injury
early in the postoperative period, tending to • Hemodynamic instability requiring aggressive
develop more readily in patients that are younger fluid resuscitation
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 205
spective cohort study evaluating 435 neonates infants with postoperative oliguria (defined as 4
undergoing cardiac surgery also recognized this total h of urine output <1 mL/kg/h during the first
relationship, with fluid overload ≥16% in the postoperative 24 h) were randomized to receive
postoperative period that was independently either PD or furosemide therapy. Patients who
associated with worse outcomes including longer were restricted to furosemide treatment were
hospital stay. These patients were also more more likely to develop 10% fluid overload, pro-
likely to experience cardiac arrest requiring CPR, longed mechanical ventilation, and prolonged
develop thrombosis, or require chest re-explora- CICU stay. Also, patients who received PD had
tion. Hazle et al. prospectively identified fluid fewer electrolyte abnormalities, though two
overload within the first 3 postoperative days fol- PD-related adverse outcomes were observed –
lowing cardiac surgery to be independently asso- peritoneal bleeding requiring a blood transfusion
ciated with poor outcomes in 49 infants and hydrocele formation [39].
<6 months of age who underwent surgery with In summary, postoperative cardiac patients are
CPB [38]. Specifically, they found that odds of predisposed to fluid overload due to a variety of
poor outcome (defined by a composite score that factors. Significant fluid overload in postopera-
included the need for RRT, time to first extuba- tive cardiac patients is common and associated
tion, intensive care length of stay, or death within with worse outcomes. While fluid overload may
30 days of surgery) increased by 7% with each be unavoidable in some patients, efforts to mini-
1% increase in fluid overload as calculated by mize the degree of fluid overload are appropriate.
maximum daily weight method, adjusted for For institutions that have the resources and expe-
peak serum creatinine [38]. rience to readily implement PD in the early post-
Fluid overload is therefore deemed to be haz- operative period, careful risk factor assessment
ardous to the postoperative pediatric cardiac could drive the creation of PD protocols for this
patient by current pediatric cardiac intensive care patient population. As experience with this
providers. As a result, many centers implement modality increases and more data become avail-
management strategies that are designed to pre- able, many of the risks associated with PD may
vent fluid overload. These measures include decrease, and centers may be more inclined to
some or all of the following: intraoperative and offer this modality more frequently. In the mean-
postoperative corticosteroids to reduce immuno- time, conservative fluid management, early initi-
genic-mediated capillary leak, conservative fluid ation of diuretic therapy, and careful attention to
management, early diuretic administration to daily fluid balance and weight measurements are
mitigate perioperative fluid/blood product admin- necessary to temper the risks of fluid overload in
istration, and dialysis. Emerging evidence sug- children recovering from cardiac surgery.
gests that early dialysis, particularly with
peritoneal dialysis (PD), may have benefits to
postoperative cardiac patients [39]. A single-cen- Fluid Overload and ECMO
ter retrospective study of 146 neonates published
in 2012 reported that children with AKI after car-
diac surgery in which PD was started early Case
(within the first day) had lower 30- and 90-day A 6-month-old infant with a history of pre-
mortality rates [40]. Of note, 52 complications maturity and chronic lung disease presents
related to PD were recorded in 44 patients in this to the hospital in respiratory distress from
study including dysfunction of the catheter (leak- respiratory syncytial virus infection. She is
age at the insertion site, displacement, or insuffi- endotracheally intubated and provided
cient drainage), hydrothorax, hemoperitoneum mechanical ventilation for respiratory fail-
related to insertion or withdrawal of the catheter, ure with hypoxemia. Despite multiple ven-
bowel perforation, and peritonitis. More recently, tilator maneuvers, her arterial blood gases
in an unblinded trial by Kwiatkowski et al., 73
ERRNVPHGLFRVRUJ
206 G. L. Ker and S. Gangadharan
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 207
Conclusion
• Options for management of fluid over-
Fluid overload remains an area of ongoing evolu- load include conservative fluid manage-
tion regarding its definition and management ment, diuretic therapy, and RRT
within the clinical setting of the critically ill child. including CVVH/CVVHDF and
Despite a large body of evidence to indicate that it PD. Furosemide therapy is generally
is associated with increased morbidity and mor- accepted as the first-line diuretic agent,
tality in specific disease states, management strat- but persistent or worsening fluid over-
egies are largely dictated by physician experience load despite conservative fluid manage-
and preference. Importantly, it can be stated with ment and furosemide therapy should
high confidence that there are no studies to date prompt adjunctive treatments or RRT.
that suggest that fluid overload is beneficial to • Further research of the effects of fluid
patients or has any positive effect on recovery or overload and its management in specific
survival. Definitions of fluid overload have largely patient populations and disease states is
relied upon calculated measurements, but these required to improve our practice and
may underrepresent fluid distribution during criti- decrease variations in care across centers.
cal illness and fail to completely capture the intri-
cacies of fluid extravasation that put certain organs
at higher risk of complications than others.
Clinicians should focus on judicious use of fluids References
including blood products along with careful
assessment of fluid status, thereby allowing for 1. Foland JA, Fortenberry JD, Warshaw BL, Pettignano
early interventions aimed at prevention of fluid R, Merrittv RK, Heard ML, Rogers K, Reid C,
Tanner AJ, Easley KA. Fluid overload before cson-
overload states and the associated cellular edema,
tinuous hemofiltration and survival in critically ill
organ dysfunction, morbidity, and mortality. children: a retrospective analysis. Crit Care Med.
2004;32:1771–6.
2. Jain A. Body fluid composition. Pediatr Rev.
Key Points 2015;36:141–52.
3. Bianchetti MG, Simonetti GD, Bettinelli A. Body
• The dynamics of fluid movement are fluids and salt metabolism - part I. Ital J Pediatr.
complex, and our understanding regard- 2009;35:36.
ing such is still evolving. 4. Agrò F.E. VM. Physiology of body fluid compartments
• Current methods of the assessment of and body fluid movements. Milano: Springer; 2013.
5. Woodcock TE, Woodcock TM. Revised Starling equa-
fluid overload are potentially error prone tion and the glycocalyx model of transvascular fluid
and can produce conflicting results. exchange: an improved paradigm for prescribing intra-
Daily assessment of fluid status how- venous fluid therapy. Br J Anaesth. 2012;108:384–94.
ever is crucial, and discrepancies 6. Rogers’ Textbook of Pediatric Intensive Care. 5th ed.
Philadelphia: Wolters Kluwer; 2008.
between weight and intake/output cal- 7. Holte K, Sharrock NE, Kehlet H. Pathophysiology and
culations should be reconciled by clini- clinical implications of perioperative fluid excess. Br J
cal exam and available physiologic Anaesth. 2002;89:622–32.
bedside data. 8. Selewski DT, Cornell TT, Lombel RM, Blatt NB, Han
YY, Mottes T, Kommareddi M, Kershaw DB, Shanley
• Fluid overload is associated with TP, Heung M. Weight-based determination of fluid
adverse outcomes in multiple disease overload status and mortality in pediatric intensive
states, and, to our knowledge, there are care unit patients requiring continuous renal replace-
no studies that demonstrate that fluid ment therapy. Intensive Care Med. 2011;37:1166–73.
9. Goldstein SL, Currier H, Graf C, Cosio CC, Brewer
overload is beneficial to patients or their ED, Sachdeva R. Outcome in children receiving
outcomes. continuous venovenous hemofiltration. Pediatrics.
2001;107:1309–12.
ERRNVPHGLFRVRUJ
208 G. L. Ker and S. Gangadharan
10. Milani GP, Groothoff JW, Vianello FA, Fossali EF, tional study fluid balance and patient outcomes in
Paglialonga F, Edefonti A, Agostoni C, Consonni D, the randomized evaluation of normal vs. augmented
van Harskamp D, van Goudoever JB, Schierbeek H, level of replacement therapy trial. Crit Care Med.
Oosterveld MJS. Bioimpedance and fluid status in 2012;40:1753–60.
children and adolescents treated with Dialysis. Am J 24. Seabra VF, Balk EM, Liangos O, Sosa MA,
Kidney Dis. 2017;69:428–35. Cendoroglo M, Jaber BL. Timing of renal replace-
11. Mehta RL, Pascual MT, Soroko S, Chertow GM,
ment therapy initiation in acute renal failure: a meta-
for the PSG. Diuretics, mortality, and nonrecov- analysis. Am J Kidney Dis. 2008;52:272–84.
ery of renal function in acute renal failure. JAMA. 25. Karvellas CJ, Farhat MR, Sajjad I, Mogensen SS, Leung
2002;288:2547–53. AA, Wald R, Bagshaw SM. A comparison of early ver-
12. Kakajiwala A, Kim JY, Hughes JZ, Costarino A,
sus late initiation of renal replacement therapy in criti-
Ferguson J, Gaynor JW, Furth SL, Blinder JJ. Lack cally ill patients with acute kidney injury: a systematic
of furosemide responsiveness predicts acute kidney review and meta-analysis. Criti Care. 2011;15:R72.
injury in infants after cardiac surgery. Ann Thorac 26. Arikan AA, Zappitelli M, Goldstein SL, Naipaul A,
Surg. 2017;104:1388–94. Jefferson LS, Loftis LL. Fluid overload is associated
13. Singh NC, Kissoon N, al Mofada S, Bennett M, Bohn with impaired oxygenation and morbidity in critically
DJ. Comparison of continuous versus intermittent ill children. Pediatr Crit Care Med. 2012;13:253–8.
furosemide administration in postoperative pediatric 27. Flori HRCG, Liu KD, et al. Positive fluid balance
cardiac patients. Crit Care Med. 1992;20:17–21. is associated with higher mortality and prolonged
14. Klinge JM, Scharf J, Hofbeck M, Gerling S, Bonakdar mechanical ventilation in pediatric patients with acute
S, Singer H. Intermittent administration of furose- lung injury. Crit Care Res Pract. 2011;854142
mide versus continuous infusion in the postoperative 28. Willson DF, Thomas NJ, Tamburro R, Truemper
management of children following open heart surgery. E, Truwit J, Conaway M, Traul C, Egan EE. The
Intensive Care Med. 1997;23:693–7. relationship of fluid administration to outcome
15. Ricci Z, Haiberger R, Pezzella C, Garisto C, Favia I, in the pediatric calfactant in acute respiratory
Cogo P. Furosemide versus ethacrynic acid in pediat- distress syndrome trial. Pediatr Crit Care Med.
ric patients undergoing cardiac surgery: a randomized 2013;14:666–72.
controlled trial. Crit Care. 2015;19:2. 29. Willson DF, Thomas NJ, Tamburro R, Truemper E,
16. Sparrow AW, Friedberg DZ, Nadas AS. The use of Truwit J, Conaway M, Traul C, Egan EE. Pediatric
ethacrynic acid in infants and children with conges- calfactant in acute respiratory distress syndrome trial.
tive heart failure. Pediatrics. 1968;42:291–302. Pediatr Crit Care Med. 2013;14:657–65.
17. Roush GC, Kaur R, Ernst ME. Diuretics: a review and 30. Sinitsky L, Walls D, Nadel S, Inwald DP. Fluid over-
update. J Cardiovasc Pharmacol Ther. 2013;19:5–13. load at 48 hours is associated with respiratory morbid-
18. Ng GYT, Baker EH, Farrer KFM. Aminophylline
ity but not mortality in a general PICU: retrospective
as an adjunct diuretic for neonates—a case series. cohort study. Pediatr Crit Care Med. 2015;16:205–9.
Pediatr Nephrol. 2005;20:220–2. 31. Randolph AG, Forbes PW, Gedeit RG, Arnold JH,
19. Pretzlaff RK, Vardis RJ, Pollack MM. Aminophylline Wetzel RC, Luckett PM, O’Neil ME, Venkataraman
in the treatment of fluid overload. Crit Care Med. ST, Meert KL, Cheifetz IM, Cox PN, Hanson
1999;27:2782–5. JH. Cumulative fluid intake minus output is not
20. Vaara ST, Korhonen A-M, Kaukonen K-M, Nisula associated with ventilator weaning duration or extu-
S, Inkinen O, Hoppu S, Laurila JJ, Mildh L, bation outcomes in children. Pediatr Crit Care Med.
Reinikainen M, Lund V. Fluid overload is associ- 2005;6:642–7.
ated with an increased risk for 90-day mortality in 32. Ingelse SA, Wiegers HM, Calis JC, van Woensel JB,
critically ill patients with renal replacement therapy: Bem RA. Early fluid overload prolongs mechanical
data from the prospective FINNAKI study. Crit Care. ventilation in children with viral-lower respiratory
2012;16:R197. tract disease. Pediatr Crit Care Med. 2017;18:e106–11.
21. Uchino S, Kellum JA, Bellomo R, Doig GS,
33. Wiedemann HP, et al. Comparison of two fluid-man-
Morimatsu H, Morgera S, Schetz M, Tan I, Bouman agement strategies in acute lung injury. N Engl J Med.
C, Macedo E, Gibney N, Tolwani A, Ronco C. Acute 2006;354(24):2564–75.
renal failure in critically ill patients: a multinational, 34. Randolph AG. Management of acute lung injury and
multicenter study. JAMA. 2005;294:813–8. acute respiratory distress syndrome in children. Crit
22. Bagshaw SM, Laupland KB, Doig CJ, Mortis G, Fick Care Med. 2009;37:2448–54.
GH, Mucenski M, Godinez-Luna T, Svenson LW, 35. Hassinger AB, Wald EL, Goodman DM. Early post-
Rosenal T. Prognosis for long-term survival and renal operative fluid overload precedes acute kidney injury
recovery in critically ill patients with severe acute and is associated with higher morbidity in pediatric
renal failure: a population-based study. Criti Care. cardiac surgery patients. Pediatr Crit Care Med.
2005;9:R700–9. 2014;15:131–8.
23. Bellomo R, Cass A, Cole L, Finfer S, Gallagher M, 36. Seguin J, Albright B, Vertullo L, Lai P, Dancea A,
Lee J, Lo S, McArthur C, McGuiness S, Norton Bernier P-L, Tchervenkov CI, Calaritis C, Drullinsky
R, Myburgh J, Scheinkestel C, Su S. An observa- D, Gottesman R. Extent, risk factors, and outcome of
ERRNVPHGLFRVRUJ
11 Management of Fluid Overload in the Pediatric ICU 209
fluid overload after pediatric heart surgery. Crit Care fluid mobilization as determinants of the duration of
Med. 2014;42:2591–9. ECMO support. J Pediatr Surg. 1991;26:1016–22.
37. Lex DJ, Toth R, Czobor NR, Alexander SI, Breuer 44. Swaniker F, Kolla S, Moler F, Custer J, Grams R,
T, Sapi E, Szatmari A, Szekely E, Gal J, Szekely Barlett R, Hirschl R. Extracorporeal life support out-
A. Fluid overload is associated with higher mor- come for 128 pediatric patients with respiratory fail-
tality and morbidity in pediatric patients under- ure. J Pediatr Surg. 2000;35:197–202.
going cardiac surgery. Pediatr Crit Care Med. 45. Selewski DT, Askenazi DJ, Bridges BC, Cooper DS,
2016;17:307–14. Fleming GM, Paden ML, Verway M, Sahay R, King
38. Hazle MA, Gajarski RJ, Yu S, Donohue J, Blatt E, Zappitelli M. The impact of fluid overload on out-
NB. Fluid overload in infants following con- comes in children treated with extracorporeal mem-
genital heart surgery. Pediatr Crit Care Med. brane oxygenation: a multicenter retrospective cohort
2013;14:44–9. study. Pediatr Crit Care Med. 2017;18:1126–35.
39. Kwiatkowski DM, Goldstein SL, Cooper DS, Nelson 46. Hoover NG, Heard M, Reid C, Wagoner S, Rogers K,
DP, Morales DS, Krawczeski CD. Peritoneal dialy- Foland J, Paden ML, Fortenberry JD. Enhanced fluid
sis vs furosemide for prevention of fluid overload in management with continuous venovenous hemofiltra-
infants after cardiac surgery: a randomized clinical tion in pediatric respiratory failure patients receiv-
trial. JAMA Pediatr. 2017;171:357–64. ing extracorporeal membrane oxygenation support.
40. Bojan M, Gioanni S, Vouhe PR, Journois D, Pouard Intensive Care Med. 2008;34:2241–7.
P. Early initiation of peritoneal dialysis in neonates 47. Chen H, Yu RG, Yin NN, Zhou JX. Combination of
and infants with acute kidney injury following cardiac extracorporeal membrane oxygenation and continu-
surgery is associated with a significant decrease in ous renal replacement therapy in critically ill patients:
mortality. Kidney Int. 2012;82:474–81. a systematic review. Criti Care. 2014;18:675.
41. ECMO Registry of the Extracorporeal Life Support 48. Paden ML, Warshaw BL, Heard ML, Fortenberry
Organization (ELSO). 2017. JD. Recovery of renal function and survival after con-
42. Fleming GM, Sahay R, Zappitelli M, King E,
tinuous renal replacement therapy during extracorpo-
Askenazi DJ, Bridges BC, Paden ML, Selewski DT, real membrane oxygenation. Pediatr Crit Care Med.
Cooper DS. The incidence of acute kidney injury 2011;12:153–8.
and its effect on neonatal and pediatric extracorpo- 49. Askenazi DJ, Ambalavanan N, Hamilton K, Cutter
real membrane oxygenation outcomes: a multicenter G, Laney D, Kaslow R, Georgeson K, Barnhart DC,
report from the kidney intervention during extracor- Dimmitt RA. Acute kidney injury and renal replace-
poreal membrane oxygenation study group. Pediatr ment therapy independently predict mortality in
Crit Care Med. 2016;17:1157–69. neonatal and pediatric noncardiac patients on extracor-
43. Kelly RE Jr, Phillips JD, Foglia RP, Bjerke HS,
poreal membrane oxygenation. Pediatr Crit Care Med.
Barcliff LT, Petrus L, Hall TR. Pulmonary edema and 2011;12:e1–6.
ERRNVPHGLFRVRUJ
Part V
Hematologic Controversies
ERRNVPHGLFRVRUJ
Management of Cardiopulmonary
Bypass-Associated Coagulopathy 12
Rania K. Abbasi, Anne E. Cossu,
and Scott G. Walker
Coagulopathy in pediatric patients after cardio- protease cleaves and activates a subsequent pro-
pulmonary bypass (CPB) is highly complex and tease in the sequence, resulting in the produc-
multifactorial. Numerous controversies and prac- tion of thrombin followed by the formation of a
tice variations surround the management of fibrin clot [3]. The most common and prevailing
hemostasis after CPB, from the preoperative to illustration of the cascade model is a Y-shaped
the postoperative phases of the surgical process. paradigm with separate intrinsic and extrinsic
Though no true consensus exists for many of pathways that converge upon a common path-
these practices, there are data available in the lit- way (Fig. 12.1) [4, 5]. The intrinsic and extrin-
erature to provide a better of understanding of sic pathways are each initiated by factor XII
hemostasis in congenital heart disease and to (FXII) and tissue factor (TF)/activated factor
guide practice after CPB. VII (FVIIa), respectively. In this model, both
pathways lead to the formation of activated fac-
tor X (FXa) and activated factor V (FVa). In the
The Coagulation Cascade common pathway, FXa with FVa activates pro-
thrombin (FII) to thrombin (FIIa), which then
To understand the etiology of acquired coagu- cleaves fibrinogen to soluble fibrin. The extrin-
lopathy associated with cardiopulmonary bypass sic pathway functions to initiate and amplify
(CPB), it is imperative to have a thorough com- hemostasis, whereas the intrinsic pathway func-
prehension of in vivo hemostasis. tions to trigger a thrombin burst, leading to the
generation and stabilization of fibrin clot [4].
The cascade model has been greatly useful in
The Cascade Model of Hemostasis the elucidation of in vitro coagulation but was not
meant to be a model of in vivo hemostasis. Factor
The cascade or “waterfall” model of hemostasis deficiencies in the extrinsic or common pathways
was originally proposed in the 1960s [1, 2]. This can be reflected in the laboratory when determin-
model described the hemostatic process as a ing the prothrombin time (PT). Similarly, factor
sequence of enzymatic reactions wherein a deficiencies in the intrinsic or common pathways
may be reflected in the activated partial thrombo-
plastin time (aPTT). Deficiencies of specific fac-
R. K. Abbasi · A. E. Cossu · S. G. Walker (*) tors in each pathway however do not equally
Riley Hospital for Children, Indianapolis, IN, USA increase the risk of clinical bleeding or hemor-
e-mail: rabbasi@iupui.edu; acossu@iupui.edu; rhage [3, 4].
swalker1@iupui.edu
ERRNVPHGLFRVRUJ
214 R. K. Abbasi et al.
Key:
HMWK
Zymogen
XII XIIa
PK Enzyme
K
Cofactor
HMWK XIIa 2+
PL Ca
HMWK
XI XIa
PL Ca2+
EXTRINSIC
IX IXa 2+ VIIa
PL Ca PL Ca2+
INTRINSIC VIIIa TF
X Xa X
PL Ca2+
Va
II IIa
Pg Fb
COMMON
Fig. 12.1 The cascade model of coagulation HMWK, high molecular weight kininogen; PK, prekalli-
This model divides the cascade into separate intrinsic and krein; K, kallikrein; XII, factor XII; XIIa, activated factor
extrinsic pathways, both of which result in the formation XII; XI, factor XI; XIa, activated factor XI; IX, factor IX;
of FXa. The common pathway results in the formation of IXa, activated factor IX; VIIIa, activated factor VIII; X,
fibrin from fibrinogen through the action of Xa and Va. factor X; Xa, activated factor X; VIIa, activated factor VII;
Many of the coagulation factors require calcium (Ca2+) TF, tissue factor; Va, activated factor V; II, prothrombin;
and a phospholipid surface (PL). (Adapted from Smith [5] IIa, thrombin; Fg, fibrinogen; Fb, fibrin
with permission.)
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 215
TF Bearing Cell
Fibrinolysis
TF VIIa IX IXa
Fibrinolysis commences as soon as the hemo-
static process begins and is necessary to prevent
b AMPLIFICATION
IIa VIII unnecessary accumulation of intravascular fibrin
vWF
TF Bearing Cell
and mediate removal of thrombi. Plasmin is the
Va Xa II major fibrinolytic protease (6). Plasminogen is
VIIIa vWF cleaved to form active plasmin by two serine pro-
Platelet teases, urokinase-type plasminogen activator
IIa XI
IIa adhesion (uPA) and tissue plasminogen activator (tPA).
V Plasmin then cleaves fibrin at specific lysine and
V XIa
arginine amino acid residues, thereby dissolving
Activated fibrin clot and generating soluble fibrin degrada-
Va Platelet Platelet
tion products (Fig. 12.3).
Inhibitors are important to prevent excess
c PROPAGATION fibrinolysis. Serine protease inhibitors, or ser-
IIa pins, inhibit plasmin and plasminogen activators
through covalent binding and subsequent plasma
clearance [6]. Both tPA anduPA are inhibited by
plasminogen activator inhibitor-1 and activator
XIa IX VIIIa IXa X Va Xa II
inhibitor-2 (Fig. 12.3). Plasmin is inhibited pri-
marily by α2-antiplasmin and to a lesser extent by
Activated Platelet α2-macroglobulin [7]. Thrombin-activatable
fibrinolysis inhibitor (TAFI) is a non-serpin
inhibitor that removes C-terminal lysine and argi-
Figure 12.2 The cell-based model of hemostasis
nine residues on fibrin, which are the binding
The cell-based model incorporates TF-bearing cells and sites for plasminogen and tPA, thereby reducing
platelets into hemostasis. (a) Initiation phase: initiation plasminogen activation and plasmin production
occurs on the surface of TF-bearing cells. Small amounts (Fig. 12.3) [6, 7].
of thrombin and FIXa are produced and dissociate away
from the cell. (b) Amplification phase: thrombin gener-
ated on TF-bearing cells activates platelets and releases
vWF from vWF/FVIII complex. Thrombin also activates Coagulation in CPB
FV, FVIII, and FXI. (c) Propagation phase: many coagula-
tion proteins become localized to the surface of activated Postoperative bleeding, which occurs as a result
platelets. FIXa/FVIIIa (tenase complex) activates FXa,
leading to the formation of FXa/FVa (prothrombinase of widespread activation of the hemostatic sys-
complex) and a burst of thrombin production. (Adapted tem, is very common and often problematic in
from Smith [5] with permission.) pediatric patients recovering from cardiac sur-
gery with CPB. This hemostatic activation occurs
B. With the formation of a critical mass of solu- through several mechanisms including the con-
ble fibrin, there is spontaneous polymerization to tact system, inflammation, platelet activation,
fibrin strands, resulting in an insoluble fibrin and fibrinolysis [8, 9]. Upon commencement of
mesh. Furthermore, thrombin activates FXIII to CPB, there is a 5-fold increase in the production
ERRNVPHGLFRVRUJ
216 R. K. Abbasi et al.
α2-macroglobulin
Case
A 5-day-old 3-kg neonate is undergoing
FIBRIN FIBRIN DEGRADATION repair of truncus arteriosus type
PRODUCTS
I. Preoperative laboratory studies include
hematocrit 46%, platelets 214,000, and
Thrombin-activatable
THROMBIN
fibrinolysis inhibitor
fibrinogen 178 mg/dL. Coagulation studies
are otherwise normal. During the opera-
Fig. 12.3 Overview of the fibrinolytic system tion, the patient is cooled to 28 °C. Duration
Plasminogen is converted to the active protease, plas-
of CPB is 122 min. Various hemostatic
min, through the action of tPA and uPA. Plasmin, once
produced, can then degrade fibrin into soluble degrada- strategies are employed throughout the
tion products. Both tPA and uPA can be inhibited by operation.
plasminogen activator inhibitors 1 and 2. Plasmin is
inhibited by α2-antiplasmin and α2-macroglobulin.
Thrombin-activatable fibrinolysis inhibitor (TAFI) is a
potent inhibitor of fibrinolysis via indirect inhibition of
plasmin formation. TAFI binds to arginine and lysine Antifibrinolytic Therapy
residues on fibrin, which are also binding sites for plas-
minogen and tPA, thereby preventing conversion of plas- Prophylactic use of pharmacologic antifibrinolyt-
minogen to plasmin by tPA [7]. Blue arrows indicate
enzymatic activation and red arrows indicate inhibition ics to reduce intraoperative and postoperative
blood loss after cardiac surgery is a strategy uti-
of tPA and 20-fold increase in the production of lized by many institutions. These agents act by
thrombin and soluble fibrin [9, 10]. Soluble and inhibiting fibrinolysis and protecting platelets.
circuit-bound fibrin creates an enormous surface They are typically administered before the start
area for plasminogen activation [11–13]. of CPB and thus prior to stimulation of fibrinoly-
Consequently, after CPB begins, there is a 10- to sis by the cardiopulmonary bypass circuit.
100-fold increase in plasmin production [14]. There are three antifibrinolytic agents that are
This hyper-fibrinolytic state consumes fibrino- used in cardiac surgery to minimize bleeding and
gen, leaving little available for hemostasis after reduce the need for blood product transfusion:
CPB. Furthermore, excessive production of plas- aprotinin, ∑-aminocaproic acid (EACA), and
min can result in the damage of platelets through tranexamic acid (TXA). Aprotinin is the most
plasmin-mediated cleavage of the glycoprotein extensively studied antifibrinolytic agent. It is a
Ib receptor [15, 16] and partial activation, which serine protease inhibitor isolated from bovine
makes them less responsive to natural agonists lung tissue. Aprotinin forms reversible com-
such as adenosine diphosphate and arachidonic plexes which inhibit various types of proteases
acid [17–19]. such as plasmin and kallikrein, thereby prevent-
While hyper-fibrinolytic states are common ing fibrin clot degradation [8]. While aprotinin
after CPB, some patients can paradoxically has been shown to reduce the activation of plate-
develop a hypo-fibrinolytic state after lets in vitro [17, 20], various studies in adults
CPB. Plasminogen activator inhibitor-1 (PAI-1), have demonstrated the efficacy of aprotinin in
which prevents plasmin formation, is an acute decreasing hemostatic activation and in preserv-
phase reactant, and serum concentrations are ing platelet function [21–24]. Its efficacy in
often increased by 15-fold for several hours after reducing bleeding and transfusion requirements
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 217
after cardiac surgery in adults is also well- ance of the Food and Drug Administration (FDA)
established in many randomized controlled trials and other authorities, removed aprotinin from the
and meta-analyses [25–29]. Aprotinin has also worldwide market in November of 2007. Since
been studied extensively in pediatric patients, but its suspension from the markets in 2007, the data
many of these studies contain conflicting efficacy from three observational studies [41–43] have
and safety data. Several prospective, randomized, been reevaluated by the FDA and found not to
placebo-controlled, double-blind pediatric stud- provide sufficient evidence to show a negative
ies have found that aprotinin is associated with risk/benefit ratio for aprotinin when used for its
reduced amounts of bleeding and transfusions of licensed indication. Similarly, the regulatory
blood products after cardiac surgery, although the authority in Canada (Health Canada) and the
differences were not always statistically signifi- European regulatory authority have reanalyzed
cant [30–34]. A meta-analysis of 12 randomized data from the BART study. In 2011, Health
controlled studies measured the effect of apro- Canada subsequently allowed license of apro-
tinin on the proportion of children transfused, the tinin to be regranted for adults undergoing myo-
volume of blood transfused, and the volume of cardial revascularization only. The European
chest tube drainage [35]. The combination of the regulatory authority also revisited the aprotinin
data from these randomized controlled trials safety data and in 2012 recommended the resto-
clearly demonstrated reduction in the percentage ration of the license for use in Europe as well.
of patients requiring blood product transfusion. Aprotinin remains available in Canada, the
Conversely, several randomized, placebo-United Kingdom, Sweden, and the Netherlands.
controlled, double-blind studies have concluded In contrast, studies examining the safety of
that aprotinin is not efficacious in reducing blood aprotinin in pediatric patients have described dif-
loss or transfusion requirements when compared ferent results. For example, the safety of apro-
to placebo in pediatric patients after cardiac sur- tinin was evaluated in a large multi-institutional
gery [36–39]. cohort of pediatric patients undergoing cardiac
Regardless of these conflicting data, there surgery at 35 US children’s hospitals from 2003
have been safety concerns with the use of apro- to 2007 using the Pediatric Health Information
tinin due to reports of an increased incidence of Systems Database [46]. In a multivariable analy-
cardiovascular and cerebrovascular events, renal sis from this study, there was no difference in
failure, and mortality when compared to the mortality, dialysis, or hospital length of stay in
lysine analogues EACA and TXA [40–43]. The patients who received aprotinin versus those who
Blood Conservation Using Antifibrinolytics in a did not. Despite these data, aprotinin remains
Randomized Trial (BART) study compared the unavailable in the United States for use in pediat-
efficacy and safety of aprotinin, TXA, and EACA ric patients undergoing cardiac surgery.
in adult cardiac surgery patients at high risk for EACA and TXA are analogues of the amino
mortality. The trial was stopped at 80% enroll- acid lysine. EACA and TXA competitively bind
ment due to a 53% relative increase in all-cause to lysine residues on plasminogen, thereby
mortality and doubled risk of death from cardiac preventing the binding of plasminogen to fibrin
causes in the patients who received aprotinin [47]. Recall that plasminogen, tPA, and fibrin
compared to patients who received TXA or bind together to activate plasmin, which then
EACA [44]. A meta-analysis of randomized con- degrades fibrin into fibrin degradation products
trolled trials that analyzed the use of aprotinin, (Fig. 12.3). Plasminogen cleavage to plasmin is
EACA, and TXA in adult patients during non- thereby inhibited, which results in downstream
urgent cardiac surgery found that all three drugs inhibition of fibrinolysis. TXA may also improve
reduced the frequency of transfusion compared hemostasis by preventing plasmin-induced plate-
with placebo but that aprotinin increased the rela- let activation [48].
tive risk of death when compared to TXA and The role of EACA in reducing postoperative
EACA [45]. Bayer Healthcare, under the guid- bleeding has been studied extensively in adult
ERRNVPHGLFRVRUJ
218 R. K. Abbasi et al.
and pediatric patients after cardiac surgery with not conclusive, the published data from the adult
CPB. Published studies indicate that treatment population supports caution when prescribing
with EACA reduces bleeding after cardiac sur- TXA to patients who may be at increased risk for
gery with CPB when compared to placebo [49– seizures, regardless of age.
51]. A meta-analysis of five randomized clinical In a meta-analysis of all randomized con-
trials investigated the efficacy of EACA in reduc- trolled trials comparing aprotinin, TXA, and
ing bleeding and blood product transfusion and EACA in pediatric patients undergoing cardiac
maintenance of coagulation tests after pediatric surgery, both TXA and aprotinin were signifi-
cardiac surgery. The primary conclusions were cantly associated with reduced packed red blood
that, when compared to placebo, prophylactic cell (PRBC) transfusion compared to placebo,
EACA reduced postoperative blood transfusion, while the effect of EACA on red blood cell trans-
reduced rates of surgical re-exploration, and fusion could not be determined. Additionally,
helped preserve coagulation parameters after there was no evidence that EACA and TXA were
pediatric cardiac surgery [52]. less effective than aprotinin in reducing periop-
TXA, the other lysine analogue, is also com- erative blood loss. In a large multicenter study
monly used in patients undergoing cardiac sur- linking the Society of Thoracic Surgeons
gery with CPB. Several prospective, randomized, Congenital Heart Surgery Database (2004–2008)
double-blind, placebo-controlled studies have with medication data from the Pediatric Health
demonstrated TXA to be effective in reducing Information Systems Database, aprotinin was
postoperative blood loss in pediatric cardiac sur- compared to EACA and TXA in 22,258 patients
gical patients [53, 54]. One serious complication from 25 centers [62]. When comparing aprotinin
associated with TXA is the incidence of seizures. to EACA, there was no observed difference in
A number of studies have reported that TXA is efficacy and safety outcomes when all patients
associated with increased incidence of postopera- were considered, but EACA was associated with
tive seizures in adult patients after cardiac sur- significantly greater in-hospital mortality in the
gery [55–59]. A recent meta-analysis of subgroup of patients who underwent redo ster-
randomized controlled trials was performed with notomy and greater bleeding requiring reopera-
the aim of assessing the risk of seizures with tion in neonates. In contrast, TXA was associated
TXA use in adult patients after cardiac surgery. A with reduced in-hospital mortality, bleeding
pooled analysis of 16 studies enrolling 45,235 requiring surgical intervention, intensive care
adult patients demonstrated that TXA therapy unit length of stay, and dialysis when compared
was associated with a statistically significant to aprotinin in an analysis of all patients and in
increased (fourfold) occurrence rate of postoper- subgroup analysis of neonates. EACA and TXA
ative seizures. This same fourfold increase in the were not directly compared in this study. Studies
rate of seizures in patients receiving TXA versus that directly compare TXA versus EACA in pedi-
EACA was also demonstrated in a study of pedi- atric cardiac surgery found both agents to be
atric cardiac surgery patients, but the difference superior to placebo in reducing postoperative
did not reach statistical significance [63]. The bleeding and transfusion requirements, but there
occurrence of seizures with TXA may be dose- were no differences observed in these outcomes
related (i.e., increased risk with increased dos- between TXA and EACA [63–65].
age) [58, 60]. A prospective study comparing In summary, based on current data, antifibri-
TXA and EACA with regard to efficacy in reduc- nolytic therapy should be part of the management
ing postoperative bleeding and safety after pedi- of coagulopathy and bleeding after CPB. Despite
atric cardiac surgery found that there was no an impressive body of literature examining these
difference in postoperative bleeding in the first agents, it remains unclear which of the three
24 h after surgery and that there was no differ- available agents are optimal in this clinical set-
ence in the occurrence of seizures, renal injury, ting. For practitioners within the United States,
and renal failure [61]. Although pediatric data are aprotinin is not approved by the FDA and not
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 219
available for use, which dictates the use of either 9 months old undergoing biventricular repair
TXA or EACA. Recent literature provides evi- were randomized to either hematocrit of 20% or
dence that EACA and TXA have, at the very 30% while on CPB. Patients in the 20% hemato-
least, similar efficacy in reducing perioperative crit group had significantly reduced cardiac
bleeding and blood transfusion requirements index, higher serum lactate levels, and worse
when compared to aprotinin. Until additional 1-year psychomotor tests relative to the 30%
data are available, centers should be utilizing hematocrit group [74]. The second study ran-
antifibrinolytic therapy following CPB, the domized infants undergoing biventricular repair
choice of which can be guided by physician to either hematocrit of 25% or 35% while on
experience and preference. CPB, assessing the same clinical variables as the
earlier study. No significant differences in clini-
cal outcomes were found between groups. The
Case results of these two studies suggest that a hema-
Tranexamic acid (TXA) is utilized through- tocrit of 25% may be adequate for neurodevelop-
out the case. After induction of anesthesia mental outcomes [75, 76]. More recently, lower
and prior to initiation of CPB, a loading goal hematocrits have been reported in the
dose of 4 mg/kg is administered over Jehovah’s Witness population, where “bloodless”
10 min and followed by an infusion of surgery is requested. While some report hemato-
3 mg/kg/h throughout the case. crits as low as 21% while on CPB without adverse
events, the results of the aforementioned Boston
study and absence of long-term neurodevelop-
mental follow-up in these patients make this
ardiopulmonary Bypass Strategies
C practice controversial [77–80].
and Impact on Coagulopathy Techniques for ultrafiltration and hemocon-
centration are also variable between institutions.
Several practice variations and controversies sur- Ultrafiltration was introduced as a component of
rounding CPB management strategies affect CPB management to remove excess plasma water
coagulation [66]. The degree of allowable hemo- and improve hemodilution [81, 82]. It has also
dilution is one such controversy. Pediatric been shown to reduce the inflammatory mediator
patients, particularly neonates and infants, are burden induced by CPB. Three modes of ultrafil-
subject to hemodilution due to their small blood tration are currently in use: conventional (CUF),
volume in comparison to the volume of priming modified (MUF), and zero-balance (Z-BUF).
solution within the CPB circuit. Hemoglobin, Conventional ultrafiltration includes both MUF
platelets, and coagulation factors are all diluted, and Z-BUF. CUF is the original form of
which can adversely affect tissue perfusion, neu- ultrafiltration, occurring anytime during the CPB
rocognitive function, and coagulation. Smaller duration [82]. In the early 1990s, MUF was
pediatric circuits are used at some centers, as developed as a more efficient mode of ultrafiltra-
small as ~200 mL, to help minimize the degree of tion, utilizing different timing (immediately after
hemodilution [67–72], though even this volume CPB conclusion) and technique [81]. Z-BUF
is large relative to small infants (i.e., blood vol- takes place during the rewarming period, when
ume of a 3-kg neonate is ~240 mL). inflammatory mediator release is greatest. Blood
Dilution of blood components frequently volume is maintained at a constant during the
leads to increased transfusion requirements [73]. procedure utilizing a balanced electrolyte solu-
Transfusion targets for children requiring CPB tion [83]. Multiple studies have demonstrated
are also controversial. Two key randomized con- numerous benefits of both MUF and Z-BUF
trolled trials from Boston Children’s Hospital including reduced transfusion requirements,
suggested a minimum hematocrit of 25% while improved cardiac function, and reduced
on CPB. In the first study, infants less than duration of mechanical ventilation [82, 84–87].
ERRNVPHGLFRVRUJ
220 R. K. Abbasi et al.
Complications include possible hemodynamic crit, but patients receiving FFP had significantly
instability or technical difficulties during the higher post-CPB fibrinogen levels, fewer cryo-
ultrafiltration process. While there is no standard precipitate transfusions, and fewer overall
ultrafiltration practice between institutions, with donor exposures [94]. A prospective, random-
many using a combination of the various tech- ized study of 121 infants and children undergo-
niques, the data suggests that ultrafiltration ing CPB comparing PRBCs with FFP to PRBCs
should be implemented in pediatric CPB as a and 20% albumin solutions in the prime dem-
means of tempering the negative effects of onstrated no clear clinical benefit, with no dif-
hemodilution and, by association, post-CPB ference in total transfusions between groups.
coagulopathy. Patients receiving FFP did, however, have sig-
The composition of the CPB priming solution nificantly higher fibrinogen levels and clot
is an additional source of debate between practi- firmness on hemostatic assay parameters imme-
tioners. The need for blood in the prime is depen- diately post-CPB [95]. Miao et al. found similar
dent on several factors, including weight, results in 80 pediatric patients comparing
preoperative hematocrit, and circuit size [72]. PRBCs plus FFP to a PRBCs plus 4% succinyl-
Additionally, the target hematocrit of the prime ated gelatin prime [96]. Though not conclusive,
is variable between institutions. Some centers the literature suggests potential benefits of add-
utilize whole blood, others use PRBCs and FFP, ing FFP to the CPB prime in the neonatal popu-
and others utilize solely PRBCs—there is no lation; however, it may not be useful in older
standard practice. The chosen therapy may in children [97].
part be determined by the institution’s blood A particularly controversial addition to the
bank logistics [66, 88]. Several studies have pump prime of neonates in recent years is autol-
compared fresh whole blood versus stored com- ogous umbilical cord blood. Improvement in
ponent therapy in the CPB prime with conflict- prenatal diagnosis of congenital heart disease
ing results. Studies supporting its use have has allowed for adequate planning, harvesting of
demonstrated reduced donor exposures and cord blood, and surgical correction in the first
reduced bleeding, presumably due to preserved days of life [98–101]. Umbilical cord blood can
platelet and coagulation factor function in the be harvested from the neonate and used for up to
whole blood [89–92]. In contrast, a prospective, 3 weeks if preserved with citrate-phosphate-
randomized, double-blind study of 200 infants dextrose-adenine (CPDA) solution, the primary
undergoing cardiac surgery by Mou et al. dem- anticoagulant utilized in the preservation of
onstrated increased ICU length of stay and whole blood or PRBCs. Typically, 80–90 mL of
increased perioperative fluid overload in patients cord blood can be harvested [99]. While autolo-
receiving fresh whole-blood prime versus recon- gous cord blood had been previously used in
stituted blood [88]. other neonatal surgery, Fedevych and colleagues
The need for FFP in addition to PRBCs in first described its use in cardiac surgery in 2012
the prime is controversial. Advocates suggest [98, 99]. Of 61 neonates undergoing arterial
that dilutional coagulopathy can be avoided by switch operation, 21 had autologous umbilical
utilizing a 1:1 ratio of PRBCs to FFP, based on cord blood in place of reconstituted blood in the
literature on massive transfusion in trauma pump prime, with the volume of either type of
patients [93]. Data in congenital cardiac blood in the priming fluid based on the preopera-
patients are less robust but not without merit. A tive hematocrit. Outcomes in the cord blood
prospective, randomized study of 20 infants group were similar to clinical outcomes in the
less than 8 kg undergoing CPB compared control group [99]. While more studies are
PRBCs alone versus PRBCs with FFP in the needed, autologous umbilical cord blood may be
pump prime. There was no difference in 24-h an alternative to reduce homologous blood
chest tube drainage or postoperative hemato- transfusions.
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 221
ERRNVPHGLFRVRUJ
222 R. K. Abbasi et al.
Platelets also change during storage, becoming tion was less than 100 mg/dl after CPB. Post-
less functional over time [106, 107]. For these infusion fibrinogen concentrations, blood loss,
reasons, most centers reserve fresh blood prod- and transfusion requirements were not different
ucts for transfusion in pediatric heart surgery between groups [111]. Based on these promising
patients, generally less than 7–14 days old. preliminary data, fibrinogen concentrate may
In addition to component therapy from donor ultimately prove to be an important adjunctive
products, several plasma-derived procoagulants therapy for post-CPB coagulopathy.
are available to replace coagulation factors with- PCCs contain factors II, VII, IX, and X,
out allogenic transfusion. These include recom- though three-factor PCCs contain negligible
binant factor VIIa (rfVIIa), fibrinogen amounts of factor VII compared to the four-fac-
concentrate, and three- and four-factor prothrom- tor PCCs (4PCCs). The 4PCCs may contain
bin complex concentrates (PCCs). Theoretical either activated factor VII (FEIBA [Baxter
advantages of these agents over allogenic trans- International]) or inactive factor VII (Kcentra
fusion include: [CSL Behring, Marburg, Germany]). Depending
on the product, they may also contain the antico-
• Potential to increase patients’ clotting factors agulants protein C, protein S, and/or heparin
with less total volume [112]. While current FDA approval for these
• Less total volume, facilitating faster products is limited to congenital or acquired
administration coagulation defects such as hemophilia or the
• Lack of need for thawing (unlike FFP and need for warfarin reversal, there is accumulated
cryoprecipitate), allowing more rapid evidence of their efficacy in off-label use after
availability CPB in adults [113–115]. These adult studies
• Lack of need for crossmatching have led to the interest in using PCCs as an
• Lower infectious risks adjunctive therapy for bleeding after CPB in
children. Existing evidence in pediatrics is lim-
Among the available agents, rfVIIa has been ited to case reports and nonrandomized observa-
most widely used and studied in children after tional studies, with little guidance in the literature
CPB, despite its off-label indication in this set- regarding appropriate dosing and safety con-
ting. Furthermore, there is evidence in both adults cerns such as risk for thrombosis. In one single-
[108] and children [109] that rfVIIa elevates the center study comparing the use of Confidex
risk of thrombotic complications after CPB. In (Behring S.p.A., Milan, Italy), a 4FPCC, in 14
2012, a Congenital Cardiac Anesthesia Society infants undergoing cardiac surgery with CPB to
Task Force review was unable to provide 11 age-matched controls who received standard
evidence-based recommendations on its use but therapy [116], no complications were reported,
noted that observational evidence supports its and there was less chest tube drainage and fewer
utility as rescue therapy when conventional ther- transfusions in the Confidex group (although
apy has been maximized [110]. It is likely that these findings did not reach statistical signifi-
rfVIIa is most effective not as a sole hemostatic cance). Before routine use can be recommended,
agent but rather by increasing the effectiveness of further elucidation is also needed in terms of
procoagulants including platelets, fibrinogen, monitoring the effects of PCCs on coagulation,
and other clotting factors when they are adminis- which is only crudely derived from standard
tered prior to its use. tests such as prothrombin time or activated par-
Fibrinogen concentrate is derived from adult tial thromboplastin time.
human plasma and purified as a lyophilized pow- Donated PRBCs, which are often an impor-
der for injection. A randomized pilot trial com- tant component of the CPB priming solution in
pared 60 mg/kg of fibrinogen concentrate to children, are often not needed after CPB. This
10 ml/kg of cryoprecipitate in 63 children (mean may be the result of advanced surgical and per-
age 3.5 months) in whom fibrinogen concentra- fusion practices, use of antifibrinolytics, and
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 223
ERRNVPHGLFRVRUJ
224 R. K. Abbasi et al.
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 225
Numerous studies have demonstrated the util- lines exist. One recent 2-year retrospective study
ity of these hemostatic assays in reducing trans- detailing transfusion requirements before and
fusion requirements compared to standard after transfusion algorithm implementation in
transfusion practices. A prospective, double- pediatric cardiac surgery at a single center
blind randomized control trial comparing routine showed significant reductions in PRBC, FFP, and
management versus TEG-guided transfusion cryoprecipitate transfusions as well as a 75%
management in 105 adults undergoing cardiac reduction in mortality [131].
surgery demonstrated a reduction in overall
transfusion requirements from 60% to 42%, FFP
requirements from 31% to 8%, and platelet trans- Case
fusion requirements from 29% to 13%, with no After the initial resuscitation with platelets
overall increase in mediastinal tube drainage and cryoprecipitate, the patient continues
[128]. While data in pediatric patients are less to have persistent oozing from multiple
robust than in the adult population, several stud- surgical sites. A TEG is sent to further
ies have demonstrated significant reductions in delineate the cause of bleeding. The TEG
blood product transfusion when utilizing hemo- results show R time = 8.1 min, K
static assays [129–135]. A prospective proce- time = 4.8 min, α-angle = 48°, MA = 56 mm,
dure- and age-matched study comparing routine and LY30 = 0%.
management versus TEG-guided management in
100 children undergoing pediatric cardiac sur-
gery demonstrated reductions in overall
transfusion requirements from 92% to 64%,
reduction in PRBC transfusions from 78% to
58%, and reduction in FFP transfusions from
78% to 14%. Interestingly, platelets and fibrino-
gen were found to be underutilized, with platelet R
8.1 min
K
4.8 min
Angle
48 deg
MA
56.0 mm
LY30
0%
transfusions increasing from 12% to 38% and (4-8 min) (1-4 min) (47-74 deg) (55-73 mm) (0-8%)
ERRNVPHGLFRVRUJ
226 R. K. Abbasi et al.
learned during the current era to help guide clini- 8. Yeh T Jr, Kavarana MN. Cardiopulmonary bypass
cal practice. As more studies are published in the and the coagulation system. Prog Pediatr Cardiol.
2005;21:87–115.
literature, the management of this common and 9. Sniecinksi RM, Chandler WL. Activation of the
potentially life-threatening complication should hemostatic system during cardiopulmonary bypass.
continue to evolve and improve. Anesth Analg. 2011;113(6):1319–33.
10. Chandler WL, Velan T. Secretion of tissue plas-
minogen activator and plasminogen activator inhibi-
tor 1 during cardiopulmonary bypass. Thromb Res.
Take-Home Points
2003;112:185–92.
• The cell-based model of hemostasis 11. Zhao X, et al. Blood interactions with plasticized poly
conceptualizes the idea that the intrinsic (vinyl chloride): influence of surface modification. J
and extrinsic pathways function in Mater Sci Mater Med. 2008;19(2):713–9.
12. van den Goor JM, et al. Adhesion of thrombotic
parallel. components to the surface of a clinically used oxy-
• Antifibrinolytics, such as aprotinin, genator is not affected by Trillium coating. Perfusion.
aminocaproic acid, and tranexamic acid, 2006;21(3):165–72.
reduce intraoperative and postoperative 13. Nishida H, et al. Comparative study of biocompatibility
between the open circuit and closed circuit in cardio-
blood loss after cardiac surgery by pulmonary bypass. Artif Organs. 1999;23(6):547–51.
inhibiting fibrinolysis. 14. Chandler WL, Velan T. Plasmin generation and
• Practice variations exist in CPB manage- D-dimer formation during cardiopulmonary bypass.
ment strategies, including composition Blood Coagul Fibrinolysis. 2004;15(7):583–91.
15. de Haan J, van Oeveren W. Platelets and soluble fibrin
of the CPB pump prime, minimization promote plasminogen activation causing downregula-
of circuit priming volume, transfusion tion of platelet glycoprotein Ib/IX complexes: protec-
targets, and ultrafiltration. tion by aprotinin. Thromb Res. 1998;92:171–9.
• Platelets are first-line therapy for post- 16.
Michelson AD, Barnard MR. Plasmin-induced
redistribution of platelet glycoprotein Ib. Blood.
CPB bleeding. Cryoprecipitate is a rea- 1990;76:2005–10.
sonable next-line therapy for persistent 17. Rinder CS, et al. Platelet activation and aggregation
bleeding. during cardiopulmonary bypass. Anesthesiology.
• Point-of-care coagulation tests, such as 1991;75(3):388–93.
18. Slaughter TF, et al. Reversible shear-mediated plate-
TEG or ROTEM, can help identify the let dysfunction during cardiac surgery as assessed by
causes of persistent bleeding post-CPB the PFA-100 platelet function analyzer. Blood Coagul
and can assist with developing and Fibrinolysis. 2001;12(2):85–93.
implementing transfusion algorithms. 19. Velik-Salchner C, et al. An assessment of cardiopul-
monary bypass-induced changes in platelet function
using whole blood and classical light transmission
aggregometry: the results of a pilot study. Anesth
Analg. 2009;108(6):1747–54.
References 20. Day JRS, et al. Clinical inhibition of the seven-
transmembrane thrombin receptor (PAR1) by intra-
1. Macfarlane RG. An enzyme cascade in the blood venous aprotinin during cardiothoracic surgery.
clotting mechanism, and its function as a biological Circulation. 2004;110:2597–600.
amplifier. Nature. 1964;202:498–9. 21. Marx G, Pokar H, Reuter H, Doering V, Tilsner
2. Davie EW, Ratnoff OD. Waterfall sequence for intrin- V. The effects of aprotinin on hemostatic function
sic blood clotting. Science. 1964;145:1310–2. during cardiac surgery. J Cardiothorac Vasc Anesth.
3. Hoffman M, Monroe DM 3rd. A cell-based model of 1991;5:467–74.
hemostasis. Thromb Haemost. 2001;85:958–65. 22. Fuhrer G, Gallimore MI, Heller W, Hoffmeister
4. Hoffman M, Monroe DM. Coagulation 2006: a mod- HE. Aprotinin in cardiopulmonary bypass—effects
ern view of hemostasis. Hematol Oncol Clin N Am. on the Hageman factor (FXII)-Kallikrein system and
2007;21:1–11. blood loss. Blood Coagul Fibrinolys. 1992;3:99–104.
5. Smith SA. The cell-based model of coagulation. J Vet 23. Segal H, et al. Complement activation during major
Emerg Crit Care. 2009;19:3–10. surgery: the effect of extracorporeal circuits and
6. Cesarman-Maus G, Hajjar KA. Molecular mechanisms high-dose aprotinin. J Cardiothorac Vasc Anesth.
of fibrinolysis. Br J Haemotol. 2005;129:307–21. 1998;12:542–7.
7. Chapin JC, KA. Fibrinolysis and the control of blood 24. Longstaff C. Studies on the mechanisms of action
coagulation. Blood Rev. 2015;29:17–24. of aprotinin and tranexamic acid as plasmin inhibi-
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 227
tors and antifibrinolytic agents. Blood Coagul 41. Schneeweiss S, et al. Aprotinin during coronary-
Fibrinolysis. 1994;5:537–42. artery bypass grafting and risk of death. N Engl J
25. Munoz JJ, Birkmeyer NJ, Birkmeyer JD, O’Connor Med. 2008;358:771–83.
GT, Dacey LJ. Is epsilon-aminocaproic acid as effec- 42. Shaw AD, et al. The effect of aprotinin on outcome
tive as aprotinin in reducing bleeding with cardiac after coronary-artery bypass grafting. N Engl J Med.
surgery. Circulation. 1999;99:81–9. 2008;358:784–93.
26. Fremes SE, Wong BI, Lee E, Mai R, Christakis GT, 43. Mangano DT, et al. The risk associated with aprotinin
McLean RF, Goldman BS, Naylor CD. Metaanalysis of in cardiac surgery. N Engl J Med. 2006;354:353–65.
prophylactic drug treatment in the prevention of postop- 44. Fergusson DA, et al. A comparison of aprotinin and
erative bleeding. Ann Thorac Surg. 1994;58(6):1580–8. lysine analogues in high-risk cardiac surgery. N Engl
27. Laupacis A, Fergusson D. Drugs to minimize periop- J Med. 2008;358:2319–31.
erative blood loss in cardiac surgery: meta-analyses 45. Henry D, et al. The safety of aprotinin and lysine-
using perioperative blood transfusion as the out- derived antifibrinolytic drugs in cardiac surgery: a
come. The International Study of Perio-operative meta-analysis. CMAJ. 2009;180(2):183–93.
Transfusion (ISPOT) investigators. Anesth Analg. 46. Pasquali SK, et al. Safety of aprotinin in congenital
1997;85(6):1258–67. heart operations: results from a large multicenter data-
28. Sedrakyan A, Treasure T, Elefteriades JA. Effect of base. Ann Thorac Surg. 2010;90(1):14–21.
aprotinin on clinical outcomes in coronary artery 47. Gerstein NS, et al. Antifibrinolytic agents in car-
bypass graft surgery; a systematic review and meta- diac and noncardiac surgery: a comprehensive
analysis of randomized clinical trials. J Thorac overview and update. J Cardiothorac Vasc Anesth.
Cardiovasc Surg. 2004;128:442–8. 2017;31:2183–205.
29. Cochrane Database. Syst Rev. 2007;4 CD001886,
48. Soslau G, et al. Effect of tranexamic acid on plate-
CMAJ 2009; 180 (2): 183–93) let ADP during extracorporeal circulation. Am J
30. Herynkopf F, et al. Aprotinin in children under-
Hematol. 1991;38:113–9.
going correction of congenital heart defects. A 49. McClure PD, Izsak J. The use of epsilon-aminocaproic
double-blind pilot study. J Thorac Cardiovasc Surg. acid to reduce bleeding during cardiac bypass in chil-
1994;108:517–21. dren with congenital heart disease. Anesthesiology.
31. D’Errico CC, et al. The efficacy and cost of aprotinin 1974;40:604–8.
in children undergoing reoperative open heart surgery. 50. Williams GD, et al. Efficacy of epsilon-aminocaproic
Anesth Analg. 1996;83:1193–9. acid in children undergoing cardiac surgery. J
32. Miller BE, et al. Hematologic and economic impact of Cardiothorac Vasc Anesth. 1999;13(3):304–8.
aprotinin in reoperative pediatric cardiac operations. 51. Rao BH, et al. Epsilon aminocaproic acid in paediat-
Ann Thorac Surg. 1998;66:535–41. ric cardiac surgery to reduce postoperative blood loss.
33. Chauhan S, et al. Efficacy of aprotinin, epsilon amino- Indian J Med Res. 2000;111:57–61.
caproic acid, or combination in cyanotic heart disease. 52. Lu J, et al. Epsilon aminocaproic acid reduces blood
Ann Thorac Surg. 2000;70:1308–12. transfusion and improves the coagulation test after
34. Bulutcu FS, et al. Which may be effective to reduce pediatric open-heart surgery: a meta- analysis of 5
blood loss after cardiac operations in cyanotic chil- clinical trials. Int J Clin Exp Path. 2015;8(7):7978–87.
dren: tranexamic acid, aprotinin, or a combination? 53. Zonis Z, et al. The effect of preoperative tranexamic
Pediatr Anaesth. 2005;15:41–6. acid on blood loss after cardiac operations in children.
35. Arnold DM, et al. Avoiding transfusions in chil-
J Thorac Cardiovasc Surg. 1996;111:982–7.
dren undergoing cardiac surgery: a meta-analysis 54. Reid R, et al. The efficacy of tranexamic acid versus
of randomized trials of aprotinin. Anesth Analg. placebo in decreasing blood loss in pediatric patients
2006;102:731–7. undergoing repeat cardiac surgery. Anesth Analg.
36. Davies MJ, et al. Prospective, randomized, double- 1997;84(5):990–6.
blind study of high-dose aprotinin in pediatric car- 55. Sharma V, et al. The association between tranexamic
diac operations. Ann Thorac Surg. 1997;63:497–503. acid and convulsive seizures after cardiac surgery: a
37. Boldt J, et al. Aprotinin in pediatric cardiac opera- multivariate analysis in 11,529 patients. Anaesthesia.
tions: platelet function, blood loss, and use of homol- 2014;69:124–30.
ogous blood. Ann Thorac Surg. 1993a;55:1460–6. 56. Koster A, et al. Moderate dosage of tranexamic acid
38. Boldt J, et al. Comparison of two aprotinin dosage during cardiac surgery with cardiopulmonary bypass
regimens in pediatric patients having cardiac opera- and convulsive seizures: incidence and clinical out-
tions: influence on platelet function and blood loss. J come. Br J Anaesth. 2013;110:34–40.
Thorac Cardiovasc Surg. 1993b;105:705–11. 57. Manji RA, et al. Seizures following cardiac surgery:
39. Williams GD, et al. A randomized, controlled trial of the impact of tranexamic acid and other risk factors.
aprotinin in neonates undergoing open-heart surgery. Can J Anaesth. 2012;59:6–13.
Paediatr Anaesth. 2008;18:812–9. 58.
Kalavrouziotis D, et al. High-dose tranexamic
40. Mangano DT, et al. Mortality associated with apro- acid is an independent predictor of early seizure
tinin during 5 years following coronary artery bypass after cardiopulmonary bypass. Ann Thorac Surg.
graft surgery. JAMA. 2007;297:471–9. 2012;93(1):148–54.
ERRNVPHGLFRVRUJ
228 R. K. Abbasi et al.
5 9. Keyl C, et al. High-dose tranexamic acid is related results of a randomized trial in infants. J Thorac
to increased risk of generalized seizures after aor- Cardiovasc Surg. 2003;126:1765–74.
tic valve replacement. Eur J Cardiothorac Surg. 75. Newburger JW, et al. Randomized trial of hemato-
2011;39(5):e114–21. crit 25% versus 35% during hypothermic cardio-
60. Lin Z, Xiaoyi Z. Tranexamic acid-associated seizures: pulmonary bypass in infant heart surgery. J Thorac
a meta-analysis. Seizure. 2016;36:70–3. Cardiovasc Surg. 2008;135:347–54.
61. Martin K, Breuer T, Gerler R, Hapfelmeier A,
76. Wypij D, et al. The effect of hematocrit during hypo-
Schreiber C, Lange R, Hess J, Wiesner G. Transexamic thermic cardiopulmonary bypass in infant heart sur-
acid versus ε-aminocaproic acid: efficacy and safety gery: results from the combined Boston hematocrit
in paediatric cardiac surgery. Eur J Cardiothorac Surg. trials. J Thorac Cardiovasc Surg. 2008;135:355–60.
2011a;39(6):892–7. 77. Naguib AN, Winch PD, Tobias JD, et al. A single-
62. Chauhan S, et al. Comparison of epsilon aminocaproic center strategy to minimize blood transfusion in
acid and tranexamic acid in pediatric cardiac surgery. neonates and children under- going cardiac surgery.
J Cardiothorac Vasc Anesth. 2004;18(2):141–3. Paediatr Anaesth. 2015;25:477–86.
63. Martin K, et al. Tranexamic acid versus epsilon-
78. Budak AB, McCusker K, Gunaydin S. A structured
aminocaproic acid: efficacy and safety in paedi- blood conservation program in pediatric cardiac sur-
atric cardiac surgery. Eur J Cardiothorac Surg. gery. Eur Rev Med Pharmacol Sci. 2017;21:1074–9.
2011b;39(6):892–7. 79. Karimi M, Florentino-Pineda I, Weatherred T, et al.
64. Schouten ES, et al. The effect of aprotinin,
Blood conservation operations in pediatric cardiac
tranexamic acid, and aminocaproic acid on blood patients: a paradigm shift of blood use. Ann Thorac
loss and use of blood products in major pediatric Surg. 2013;95:962–7.
surgery: a meta-analysis. Pediatr Crit Care Med. 80. Olshove VF, et al. Perfusion techniques toward blood-
2009;10(2):182–90. less pediatric open heart surgery. JECT. 2010;42:122–7.
65. Pasquali SK, et al. Comparative analysis of antifi- 81. Naik SK, Knight A, Elliott M. A prospective random-
brinolytic medications in pediatric heart surgery. J ized study of a modified technique of ultrafiltration
Thorac Cardiovasc Surg. 2012;143(3):550–7. during pediatric open-heart surgery. Circulation.
66. Cholette JM et al. Patient blood management in pedi- 1991;84(5 Suppl III):422–31.
atric cardiac surgery: a review. Anesth Analg. 2017 82. Wang S, Palanzo D, Undar A. Current ultrafiltration
Oct 5: e-published ahead of print. techniques before, during, and after pediatric cardiopul-
67. De Somer F, et al. Low extracorporeal priming volumes monary bypass procedures. Perfusion. 2012;27:438–46.
for infants: a benefit? Perfusion. 1996;11:455–60. 83. Thompson LD, McElhinney DB, Findlay P, et al. A
68. Miyaji K, Kohira S, Miyamoto T, et al. Pediatric
prospective randomized study comparing volume-
cardiac surgery without homologous blood transfu- standardized modified and conventional ultrafiltra-
sion, using a miniaturized bypass system in infants tion in pediatric cardiac surgery. J Thorac Cardiovasc
with lower body weight. J Thorac Cardiovasc Surg. Surg. 2001;122:220–8.
2007;134:284–9. 84. Golab HD, Kissler J, de Jong PL, van de Woestijne
69. Redlin M, Huebler M, Boettcher W, et al. Minimizing PC, Takkenberg JJ, Bogers AJ. Clinical outcome and
intra-operative hemodilution by use of a very low blood transfusion after infant cardiac surgery with a
priming volume cardiopulmonary bypass in neo- routine use of conventional ultrafiltration. Perfusion.
nates with transposition of the great arteries. J Thorac 2015;30:323–31.
Cardiovasc Surg. 2011;142:875–81. 85. Kuranti N, Busangjaroen P, Srimueang T, et al.
70. Redlin M, Habazettl H, Boettcher W, et al. Effects Modified versus conventional ultrafiltration in pedi-
of a comprehensive blood-sparing approach using atric cardiac surgery: a meta-analysis of randomized
body weight-adjusted miniaturized cardiopulmo- controlled trials comparing clinical outcome param-
nary bypass circuits on transfusion requirements in eters. J Thorac Cardiovasc Surg. 2011;142:861–7.
pediatric cardiac surgery. J Thorac Cardiovasc Surg. 86. Bando K, et al. Effect of modified ultrafiltration in
2012;144:493–9. high-risk patients undergoing operations for congeni-
71. Richmond ME, Charette K, Chen JM, Quaegebeur tal heart disease. Ann Thorac Surg. 1998;66:821–8.
JM, Bacha E. The effect of cardiopulmonary bypass 87. Journois D, Israel-Biet D, Pouard P, et al. High-
prime volume on the need for blood transfusion after volume, zero-balanced hemofiltration to reduce
pediatric cardiac surgery. J Thorac Cardiovasc Surg. delayed inflammatory response to cardiopulmonary
2013;145:1058–64. bypass in children. Anesthesiology. 1996;85:965–76.
72. Durandy Y. Usefulness of low prime perfusion pediat- 88. Mou SS, Giroir BP, Molitor-Kirsch EA, et al. Fresh
ric circuit in decreasing blood transfusion. ASAIO J. whole blood versus reconstituted blood for pump
2007;53:659–61. priming in heart surgery in infants. N Engl J Med.
73. Draaisma AM, et al. Modified ultrafiltration after
2004;351:1635–44.
cardiopulmonary bypass in pediatric cardiac surgery. 89. Manno CS, Hedberg KW, Kim HC, et al. Comparison
Ann Thorac Surg. 1997;64:521–5. of the hemostatic effects of fresh whole blood, stored
74. Jonas RA, et al. The influence of hemodilution on whole blood, and components after open heart surgery
outcome after hypothermic cardiopulmonary bypass: in children. Blood. 1991;77:930–6.
ERRNVPHGLFRVRUJ
12 Management of Cardiopulmonary Bypass-Associated Coagulopathy 229
90. Valleley MS, Buckley KW, Hayes KM, Fortuna RR, 106. Fergusson DA, Hebert P, Hogan DL, et al. Effect
Geiss DM, Holt DW. Are there benefits to a fresh of fresh red blood cell transfusions on clinical out-
whole blood vs. packed red blood cell cardiopul- comes in premature, very low–birth-weight infants.
monary bypass prime on outcomes in neonatal and J Am Med Assoc. 2012;308:1443–51.
pediatric cardiac surgery? J Extra Corpor Technol. 107. Karam O, Tucci M, Bateman ST, et al. Association
2007;39:168–76. between length of storage of red blood cell units
91. Gruenwald CE, McCrindle BW, Crawford-Lean and outcome of critically ill children: a prospective
L, et al. Reconstituted fresh whole blood improves observational study. Crit Care. 2010;142:R57. http://
clinical outcomes compared with stored com- ccforum.com/content/14/2/R57
ponent blood therapy for neonates undergoing 108. Levi M, et al. Safety of recombinant activated fac-
cardiopulmonary bypass for cardiac surgery: a ran- tor VII in randomized clinical trials. N Engl J Med.
domized controlled trial. J Thorac Cardiovasc Surg. 2010;363:1791–800.
2008;136:1442–9. 109. Downey L, et al. Recombinant factor VIIa is asso-
92. Jobes DR, et al. Reduced transfusion require- ciated with increased thrombotic complications in
ment with use of fresh whole blood in pediatric pediatric cardiac surgery patients. Anesth Analg.
cardiac surgical procedures. Ann Thorac Surg. 2017;124:1431–6.
2015;99:1706–12. 110. Guzzetta NA, et al. Review of the off-label use of
93. Durandy Y. Use of blood products in pediatric car- recombinant activated factor VII in pediatric cardiac
diac surgery. Artif Organs. 2015;39(1):21–7. surgery patients. Anesth Analg. 2012;115:364–78.
94. McCall MM, et al. Fresh frozen plasma in the pedi- 111.
Galas FRBG, et al. Hemostatic effects of
atric pump prime: a prospective, randomized trial. fibrinogen concentrate compared with cryopre-
Ann Thorac Surg. 2004;77:983–7. cipitate in children after cardiac surgery: a ran-
95. Lee JW, et al. Fresh frozen plasma in pump priming domized pilot trial. J Thorac Cardiovasc Surg.
for congenital heart surgery: evaluation of effects on 2014;148:1647–55.
postoperative coagulation profiles using a fibrinogen 112. Ashikhmina E, et al. Prothrombin complex con-
assay and rotational thromboelastometry. Yonsei centrates in pediatric cardiac surgery: the cur-
Med J. 2013;54:752–6. rent state and the future. Ann Thorac Surg.
96. Miao X, et al. Evidence-based use of FFP: the influ- 2017;104:1423–31.
ence of a priming strategy without FFP during CPB 113. Song HK, et al. Safety and efficacy of prothrombin
on postoperative coagulation and recovery in pediat- complex concentrates for the treatment of coagulop-
ric patients. Perfusion. 2015;30(2):140–7. athy after cardiac surgery. J Thorac Cardiovasc Surg.
97. Faraoni D, Torres CS. No evidence to support a 2014;147:1036–40.
priming strategy with FFP in infants. Eur J Pediatr. 114. Ranucci M, et al. Randomized, double-blinded,
2014;173:1445–6. placebo-controlled trial of fibrinogen concentrate
98. Fedevych O, et al. Open cardiac surgery in the first supplementation after complex cardiac surgery. J
hours of life using autologous umbilical cord blood. Am Heart Assoc. 2015;4:1–10.
Eur J Cardiothorac Surg. 2011;40:985–9. 115. Ghadimi K, et al. Prothrombin complex concentrates
99. Chasovskyi K, et al. Arterial switch operation in the for bleeding in the perioperative setting. Anesth
first hours of life using autologous umbilical cord Analg. 2016;122:1287–300.
blood. Ann Thorac Surg. 2012;93:1571–6. 116. Giorni C, et al. Use of Confidex to control periopera-
100. Chasovskyi K, et al. Tissue perfusion in neonates tive bleeding in pediatric heart surgery: a prospective
undergoing open-heart surgery using autologous cohort study. Pediatr Cardiol. 2014;35:208–14.
umbilical cord blood or donor blood components. 117. Willems A, et al. Comparison of two red-cell
Perfusion. 2015;30(6):499–506. transfusion strategies after pediatric cardiac
101. Choi ES, et al. Cardiopulmonary bypass priming surgery: a subgroup analysis. Crit Care Med.
using autologous cord blood in neonatal congenital 2010;38:649–56.
cardiac surgery. Korean Circ J. 2016;46(5):714–8. 118. de Gast-Bakker DH, et al. Safety and effects of two
102. Andreasen JB, et al. Marked changes in platelet red blood cell transfusion strategies in pediatric car-
count and function following pediatric congenital diac surgery patients: a randomized control trial.
heart surgery. Pediatr Anesth. 2014;24:386–92. Intensive Care Med. 2013;39:2011–9.
103. Miller BE, et al. Predicting and treating coagu- 119. Cholette JM, et al. Children with single-ventricle
lopathies after cardiopulmonary bypass in children. physiology do not benefit from higher hemoglobin
Anesth Analg. 1997;85:1196–202. levels post cavopulmonary connection: results of a
104. Harker LA, et al. Mechanism of abnormal bleed- prospective, randomized, controlled trial of restric-
ing in patients undergoing cardiopulmonary bypass: tive versus liberal red-cell transfusion strategy.
acquired transient platelet dysfunction associ- Pediatr Crit Care Med. 2011;12:39–45.
ated with selective alpha-granule release. Blood. 120. Cholette JM, et al. Outcomes using a conservative
1980;56(5):824–34. versus liberal red blood cell transfusion strategy
105. Schoenfeld H, et al. Volume-reduced platelet con- in infants requiring cardiac operation. Ann Thorac
centrates. Curr Hematol Rep. 2006;5(1):82–8. Surg. 2017b;103:206–15.
ERRNVPHGLFRVRUJ
230 R. K. Abbasi et al.
121. Gaynor JW, et al. Neurodevelopmental out- 131. Whitney G, et al. Implementation of a transfu-
comes after cardiac surgery in infancy. Pediatrics. sion algorithm to reduce blood product utiliza-
2015;135(5):816–25. tion in pediatric cardiac surgery. Pediatr Anesth.
122. Morton PD, et al. Neurodevelopmental abnormali- 2013;23:639–46.
ties and congenital heart disease: insights into altered 132. Faraoni D, et al. Development of a specific algorithm
brain maturation. Circ Res. 2017;120:960–77. to guide haemostatic therapy in children undergoing
123. Moskowitz DM, et al. Predictors of transfusion cardiac surgery: a single-Centre retrospective study.
requirements for cardiac surgical procedures at Eur J Anaesthesiol. 2015;32:320–9.
a blood conservation center. Ann Thorac Surg. 133. Nakayama Y, Nakajima Y, Tanaka KA, et al.
2004;77:626–34. Thromboelastometry-guided intraoperative haemo-
124. Haas T, et al. Usefulness of standard plasma coag- static management reduces bleeding and red cell
ulation tests in the management of perioperative transfusion after paediatric cardiac surgery. Br J
coagulopathic bleeding: is there any evidence? Br J Anaesth. 2015;114:91–102.
Anaesth. 2015;114(2):217–24. 134. Kim E, et al. Predictive value of intraoperative
125. Segal JB, Dzik WH. Transfusion Medicine/ Thromboelastometry for the risk of perioperative
Hemostasis Clinical Trials Network Paucity of excessive blood loss in infants and children undergo-
studies to support that abnormal coagulation test ing congenital cardiac surgery: a retrospective analy-
results predict bleeding in the setting of invasive sis. J Cardiothorac Vasc Anesth. 2016;30(5):1172–8.
procedures: an evidence-based review. Transfusion. 135. Kane LC, et al. Thromboelastography—does it
2005;45:1413–25. impact blood component transfusion in pediatric
126. Whiting D, DiNardo JA. TEG and ROTEM: tech- heart surgery? J Surg Res. 2016;200:21–7.
nology and clinical applications. Am J Hematol. 136. Society of Thoracic Surgeons Blood Conservation
2014;89:228–32. Guideline Task Force et al. 2011 update to the
127. Thiruvenkatarajan V, Pruett A, Adhikary Society of Thoracic Surgeons and the Society of
SD. Coagulation testing in the perioperative period. Cardiovascular Anesthesiologists blood conserva-
Indian J Anaesth. 2014;58:565–72. tion clinical practice guidelines. Ann Thorac Surg.
128. Shore-Lesserson L, et al. Thromboelastography- 2011 Mar;91(3):944–82.
guided transfusion algorithm reduces transfu- 137. Karkouti K, McCluskey SA, Callum J, et al.
sions in complex cardiac surgery. Anesth Analg. Evaluation of a novel transfusion algorithm employing
1999;88:312–9. point-of-care coagulation assays in cardiac surgery: a
129. Miller BE, et al. Rapid evaluation of coagulopa- retrospective cohort study with interrupted time-series
thies after cardiopulmonary bypass in children analysis. Anesthesiology. 2015;122:560–70.
using modified thromboelastography. Anesth Analg. 138. American Society of Anesthesiologists Task Force
2000;90:1324–30. on Perioperative Blood Management. Practice
130. Romlin BS, et al. Intraoperative thromboelastometry guidelines for perioperative blood management:
is associated with reduced transfusion prevalence an updated report by the ASA task force on peri-
in pediatric cardiac surgery. Anesth Analg. 2011 operative blood management. Anesthesiology.
Jan;112(1):30–6. 2015;122:241–75.
ERRNVPHGLFRVRUJ
Anticoagulation for Extracorporeal
Life Support 13
Danny Eytan and Gail M. Annich
ERRNVPHGLFRVRUJ
232 D. Eytan and G. M. Annich
In this chapter we will present some of the con- balance between thrombosis and hemorrhage.
troversies surrounding current approaches to anti- Additional processes at work include protein
coagulation treatment and monitoring of patients adsorption to the circuit (especially but not lim-
who require ECLS. Some controversial questions ited to albumin and fibrinogen) and endothelial
that are still unanswered in this field include: cell function dysregulation. All these processes
lead to a robust activation of the common pathway
1. What is the best anticoagulant choice in
for coagulation, leading to generation of activated
ECLS? factor Xa and thrombin generation. Platelet adher-
2. How is anticoagulation monitored best in
ence and activation ensues, and multilateral
these patients? amplification loops lead to enhanced thrombi for-
3. What to do in cases of heparin resistance? mation throughout the circuit. As anticoagulant
4. What to do if heparin-induced thrombocyto- effector arms “kick-in,” a simultaneous mixture of
penia (HIT) is suspected? pro- and anticoagulant states exist, leading to con-
5. Do we need anticoagulation at all? current potential for circuit dysfunction due to
thrombi and bleeding tendency in the patient.
As conclusive data to answer these questions Building from experience accumulated from
is lacking, we will attempt to inform the reader use of cardiopulmonary bypass for cardiothoracic
on the current knowledge regarding some of procedures and from early experiments in ECLS,
these issues to allow for informed decision-mak- it was recognized that some form of antithrom-
ing in a case-by-case manner. We aim to describe botic therapy is essential to maintaining the cir-
the rationale for the need to anticoagulate patients cuit patency and function while minimizing risks
on ECLS, elaborate on heparin treatment and for bleeding. While heparin has traditionally been
monitoring of its effects, and discuss the concept the sole option for anticoagulation at most cen-
of heparin resistance. We will also proceed to ters, alternative agents exist and are already in
expose alternatives to systemic heparin antico- clinical use. Additionally, there are circuit modifi-
agulation such as direct thrombin inhibitors cations that may allow avoiding systemic antico-
(DTI) and shortly touch on issues related to agulation for short periods of time under specific
bleeding prior to ECLS and its implications. conditions, and the future holds promise of ECLS
mechanisms that may obviate the need for sys-
temic anticoagulation altogether [4].
Rationale for Anticoagulation
ERRNVPHGLFRVRUJ
13 Anticoagulation for Extracorporeal Life Support 233
ERRNVPHGLFRVRUJ
234 D. Eytan and G. M. Annich
arin effect, this test has come under significant and determined that the biggest cause of need
scrutiny, and yet it still remains the most fre- for red cell transfusion was blood sampling
quently used point of care (POC) test for antico- [14]. ACT remains the most readily bedside
agulation during ECLS [8]. POC test that can provide immediate results.
Activated partial thromboplastin time (aPTT) aPTT and anti-Xa measurements are plasma
and anti-factor Xa activity (anti-Xa) have become tests that require laboratory evaluation, with a
more commonly used to monitor heparinization turnaround time of 1–2 h. TEG or ROTEM
during ECLS. Activated partial thromboplastin requires a stable area in which to place the sam-
time was first described in 1953 but recognized ple and, because of the complexity of this equip-
as a viable test in 1961 [13]. In the adult popula- ment, must only be run by experienced
tion, most specifically in patients with uncompli- personnel, which limits their availability in most
cated cardiac or respiratory support, it is a very centers. Currently, while viscoelastic testing
reliable test for heparin effect. Unfortunately, POC is available, most experience with this
with developmental hemostasis, its reliability in monitoring has occurred in the cardiovascular
the pediatric patients, especially neonates, is very operating theatres.
poor because of globally low coagulation/antico- In 2014, the Extracorporeal Life Support
agulation factors in this population, most notably Organization published anticoagulation guide-
antithrombin (AT). lines after a task force reviewed center-specific
Anti-Xa testing was first described in 1973 by guidelines from many different ECLS centers
Denson and Bonnar but only recently has become worldwide. The result was essentially a review of
the standard of monitoring for heparin effect in each of the available anticoagulation/coagulation
the pediatric population as it has become more monitoring tests described above and a recom-
readily available in clinical laboratories [12]. mendation that each center should use this infor-
Specifically, it has helped to take into account the mation to determine which testing they will use
effect of developmental hemostasis and made to create their center-specific guidelines. Most
heparinization in the neonatal population more centers have developed ECLS anticoagulation
reliable and safe. protocols that include a POC whole blood test
The development of viscoelastic testing using and a plasma test, such as ACT and anti-Xa.
either thromboelastography (TEG) or rotational Without literature to provide guidance on the
thromboelastometry (ROTEM) is the more pre- superiority of one protocol over another, the
cise method of whole blood hemostasis testing combination of a plasma test with a POC whole
and has been adopted by many ECLS centers. It blood test, at the very least, has a physiologically
can aid in determining next management strate- rationale and represents a reasonable approach to
gies for those patients who are bleeding or having this complex clinical problem.
thrombotic complications during ECLS. To date,
however, there are only case reports or small Case Scenario An 8-month-old treated is cannu-
series published comparing its utility to conven- lated for veno-venous (VV) ECLS after develop-
tional testing [12]. ing severe ARDS and respiratory failure
The availability of these newer tests has led secondary to severe influenza infection. A hepa-
to a significant increase in the number of antico- rin infusion is initiated, with an ACT goal of 180–
agulation tests drawn per patient on daily basis, 220 s. In the past day, despite increased heparin
with many being repeated multiple times per infusion rates up to 60 units/kg/h, ACT and aPTT
day to ensure that they are maintained within the measurements are still subtherapeutic. No clots
appropriate range for what is deemed effective are seen on the circuit and membrane function is
anticoagulation during ECLS. In fact, Dalton excellent. Should this be defined as heparin resis-
et al. recently described a large data collection tance and if so what is the best course of action?
on anticoagulation management and outcomes What other alternatives for anticoagulation mon-
across several large ECLS centers in the USA itoring should be considered?
ERRNVPHGLFRVRUJ
13 Anticoagulation for Extracorporeal Life Support 235
ERRNVPHGLFRVRUJ
236 D. Eytan and G. M. Annich
this context with specific recommendations have spective case series, 20 adults on ECLS (both
been published [3, 10, 17]. In general, DTI are venoarterial and veno-venous) were reviewed; 10
not dependent on antithrombin for their effect but patients were treated with heparin, and 10 patients
work by direct inhibition of both circulating and with bivalirudin [27]. They found significantly
clot-bound thrombin – exposing the two major higher variations in aPTT in the heparin group
advantages over heparin. Their selective binding and no significant differences in bleeding or
to thrombin makes their pharmacodynamics thrombosis between groups even though both
more predictable. A significant downside, how- major bleeding and minor bleeding were more
ever, is that DTI do not have a specific reversal frequent in the heparin group. Both publications
agent, even though at least for bivalirudin the use concluded that DTIs are potential alternatives to
of activated factor VII has been proposed [23]. heparin in patients on ECLS, with one of the
Most of the literature on DTI use in ECLS relates groups reporting a switch from heparin to bivali-
to two commercially available products: bivaliru- rudin as the first choice for anticoagulation in
din and argatroban. In a recent survey of ECLS patients post cardiotomy. In the context of veno-
centers [8], most responders did not use any anti- venous ECLS, a case report by Jyoti et al. [28]
coagulation other than heparin in the months reports the successful switch of anticoagulation
prior to answering the survey, but over 50% from heparin to bivalirudin. A retrospective case
answered that they have used or can use DTI if series reporting on 12 children [29] showed that
indicated. Indications for DTI use include HIT, bivalirudin can be safely used in neonates and
heparin resistance not responsive to antithrombin children with maintenance dose ranges reflecting
or FFP administration, and development of considerable inter-patient variability. Such dos-
thrombosis while on heparin therapy. ing variability was observed also in adults with
Bivalirudin is a 20-amino acid synthetic poly- recommended doses to reach therapeutic targets
peptide analog of hirudin, which binds to the varying more than tenfold between different
active site of thrombin. It has a half-life of reports [10, 30, 31].
approximately 25 min and undergoes mostly pro- Argatroban is an l-arginine derivative with a
teolytic degradation. While metabolism is almost half-life similar to bivalirudin at 15 min; it under-
independent of liver and kidney function (~20% goes hepatic metabolism which might represent
renal elimination [24, 25]), infusion doses should an obstacle as patients treated with ECLS may
be adjusted according to renal function [26]. have hepatic functional impairment. Monitoring
Monitoring for bivalirudin’s anticoagulation and therapeutic targets are similar to bivalirudin
effect is most often accomplished by following as published in the ELSO anticoagulation guide-
aPTT with a goal 1.5–2× baseline or ACT >2.5× lines. Its use for ECLS has been published in case
baseline (reviewed in [3, 10]). We found no pro- reports and series for adult and pediatric patients
spective studies examining the use of bivalirudin with suspected or diagnosed HIT [32–38]. Of
in patients undergoing ECLS, but in the past sev- note, Cornell et al. reported successful use in five
eral years, two retrospective case series have adults without complications, and similarly
been published. Ranucci et al. [26] reviewed 21 Beiderlinden et al. reviewed nine patients, with
patients (9 children) on ECLS after cardiac oper- one critically ill patient suffering from major
ations, 8 of which were on heparin and 13 on bleeding attributed to hepatic impairment [32,
bivalirudin. They reported significantly more 39].
bleeding (as measured by chest drain output), Alternative potential targets to prevent circuit
FFP requirements and platelet transfusions in the clots can be found in antiplatelet agents. Inhibition
heparin group. Thromboembolic complications of platelet activation would decrease both activa-
did not differ between the groups, and when fac- tion of the coagulation cascade and thrombosis
toring in the complications, there was a signifi- within the circuit as seen in the laboratory with
cantly lower total cost of care in the pediatric nitric oxide (NO)-releasing surfaces. [10].
group treated with bivalirudin. In another retro- Prostacyclin, dipyridamole, and acetylsalicylic
ERRNVPHGLFRVRUJ
13 Anticoagulation for Extracorporeal Life Support 237
acid (ASA [aspirin]) have been used clinically in as ICU and hospital length of stay [46]. A subse-
single-center trials and within anticoagulation quent report reconfirmed only the effect on ICU
protocols for specific ventricular assist devices, length of stay and demonstrated a reduction in
but the data is scarce. Similarly, new oral antico- atrial fibrillation events [47]. A recent survey has
agulants such as rivaroxaban, highly specific found that heparin-bound circuit use is increasing
inhibitor of factor Xa, may be future potential tar- [48] as corroborated by Bembea et al. that
gets. Oral factor IIa inhibitors, such as dabigatran, reported that over 50% of 117 ELSO
are in use in adult patients for the prevention and (Extracorporeal Life Support Organization) cen-
treatment of thromboembolism, but data in pedi- ters used partial or complete heparin-bound cir-
atric and adult ECLS are lacking. cuits [8]. The effect of heparin-bound circuits on
the induction of HIT is more controversial [49–
51] and newer coating techniques use covalent
urrent Surface Modification
C bonding which might reduce the possibility of
Strategies heparin release [51].
The most common bio-passive coating, phos-
As systemic anticoagulation carries a set of risks, phorylcholine (PPC), has been shown to be
most importantly catastrophic bleeding, and as thromboresistant [52–55] and non-inferior to
there are controversies regarding the optimal heparin-bound circuits in pediatric patients [56].
agent to use for such purposes, a potential solu- It can also be utilized in cases of HIT where
tion might be to transition from a systemic solu- exclusion of all potential sources of heparin is
tion to a local one confined to the circuit. Regional needed. Furthermore, two studies by Ranucci
anticoagulation strategies such as citrate infusion et al. have shown that the use of phosphorylcho-
which are commonly used for extracorporeal line coating can allow a safe reduction in sys-
therapies such as continuous renal replacement temic heparinization during intraoperative ECLS
are impractical in the context of ECLS due to the [57, 58]. However, a prospective study on the
high blood flows required for the latter. However, combination of phosphorylcholine and heparin
modifications to the circuitry and oxygenator that coating has failed to demonstrate a significant
can locally prevent coagulation and thrombi have clinical effect [59]. Of note, most data on the use
the potential to obviate the need for systemic of coated circuits come from CPB. The ability to
anticoagulation or at least markedly reduce the extrapolate the conclusions to ECLS, in which
dosages needed. thrombosis mechanisms differ from those during
Current surface modifications available in CPB in several ways [47, 51], might be limited.
clinical practice can be divided into two major At this time, systemic anticoagulation remains a
groups: bio-passive and bioactive (heparin and necessary element to ECLS. Understanding the
nitric oxide). Most of the available data on sur- effects of each systemic anticoagulant and moni-
face-modified circuits come from cardiopulmo- toring for therapeutic goals remain the standard
nary bypass-related studies. Heparin-bound of care for management during ECLS.
circuits have been studied extensively and have
been shown to improve the biocompatibility,
reducing both activation of alternate complement ECLS Without Systemic
pathway and to some extent thrombus formation Anticoagulation
in the circuitry when blood is exposed to the arti-
ficial circuit, with some evidence to suggest With increasing frequency, the surface of the
improvement in clinical outcomes [40–45]. A new extracorporeal circuits is completely coated
meta-analysis by Mangoush et al. further con- with heparin. Moreover, a reduction of exposure
cluded that heparin-bound circuits reduce the to foreign surfaces is achieved by miniaturiza-
incidence of postoperative blood transfusions, re- tion of the ECMO system and combining spe-
sternotomy rates, duration of ventilation, as well cialized pumps and oxygenators with short
ERRNVPHGLFRVRUJ
238 D. Eytan and G. M. Annich
tubing systems [60]. These developments have the faculty argue that his intracranial injuries
led to the possibility of using ECLS for days preclude anticoagulation and therefore the use of
without any systemic anticoagulation; although ECLS in this setting.
there are no published reports to our knowledge
on this subject in infants and small children,
probably due to the lower flow rates which entail The Future of Surface Modification
a higher thrombotic risk. However, in adults
there have been several publications detailing Present research in surface modifications not yet
the successful use of ECLS without systemic transitioned to the clinical arena focus along
anticoagulation for several days, mainly in the three separate branches: passivation, biomimet-
context of multi-trauma and pulmonary hemor- ics, and endothelialization. Passivation using
rhage [61]. Bedeir et al. [62] conducted a sys- inert materials is in developmental phases. These
tematic review on the subject of ECLS use in products use low surface tension materials to pre-
trauma. They identified four reports from three vent protein adherence, most specifically fibrino-
centers that utilized initially systemic heparin- gen. Biomimetic surfaces such as heparin-bonded
free ECLS, whereas five reports accepted an circuits utilize a different strategy. Rather than
activated clotting time (ACT) target range lower target a specific portion of the anticoagulation
than 180 s. In general, they report a recent trend cascade (as with the abovementioned heparin-
toward extending anticoagulation-free periods bound circuits), this strategy tries to be more
followed by lower ACT ranges. None of the endothelial-like with elements of both platelet
studies identified reported complications related and thrombin inhibition. Experimental animal
to systemic or pulmonary thromboembolism or models have provided proof of principal; this is
morbidity related to unexpected circuit changes the next phase of circuits to come [64]. And
resulting from clotting. Even with intracranial finally the continuing work on customized endo-
bleeding, the risk of ECLS-related bleeding may thelialization of artificial devices for support,
be overestimated. Another retrospective report such as an artificial lung, is reaching levels where
[63] details three patients with significant intra- a seeded, endothelialized surface can sustain the
cranial hemorrhage undergoing heparin-free shear forces of flow through these systems. There
ECLS for up to 5 days and also received recom- is no doubt that in the next 20 years, a surface
binant factor VII and prothrombin complex. All obviating the need for systemic anticoagulation
three patients survived with no problems related will be used in the clinical arena of ECLS and
to clotting circuits or thromboembolism. implantable devices.
Survival ranging from 60% to 93% for patients
on ECLS with intracranial hemorrhages prior to
ECLS cannulation was reported. Thus, with References
appropriate patient selection and the new gener-
ation of ECLS setups, the benefits of an ECLS 1. Butt W, MacLaren G (2016) Extracorporeal mem-
brane oxygenation 2016: an update. F1000Res 5.
run without systemic anticoagulation might out- 2. Lasa JJ, Rogers RS, Localio R, Shults J, Raymond
weigh the risks. T, Gaies M, Thiagarajan R, Laussen PC, Kilbaugh
T, Berg RA, Nadkarni V, Topjian A. Extracorporeal
Case Scenario A 16-year-old male is admitted to cardiopulmonary resuscitation (E-CPR) during pedi-
atric in-hospital cardiopulmonary arrest is associated
the pediatric ICU after severe multiple trauma with improved survival to discharge: a report from
secondary to an explosion. Blast and penetrating the American Heart Association’s get with the guide-
injuries including bilateral lung contusions and lines-resuscitation (GWTG-R) registry. Circulation.
traumatic brain injury with several intracranial 2016;133:165–76.
3. Coughlin MA, Bartlett RH. Anticoagulation for extra-
hematomas are observed. Several hours into his corporeal life support: direct thrombin inhibitors and
admission, with deteriorating lung function and heparin. ASAIO J. 2015;61:652–5.
profound hypoxemia despite aggressive ventila- 4. Reynolds MM, Annich GM. The artificial endothe-
tion, option of VV ECLS is considered. Some of lium. Organogenesis. 2011;7:42–9.
ERRNVPHGLFRVRUJ
13 Anticoagulation for Extracorporeal Life Support 239
5. Annich GM. Extracorporeal life support: the pre- during extracorporeal membrane oxygenation. Artif
carious balance of hemostasis. J Thromb Haemost. Organs. 2011;35:1024–8.
2015;13(Suppl 1):S336–42. 21. Byrnes JW, Swearingen CJ, Prodhan P, Fiser R,
6. Vaquer S, de Haro C, Peruga P, Oliva JC, Artigas Dyamenahalli U. Antithrombin III supplementa-
A. Systematic review and meta-analysis of com- tion on extracorporeal membrane oxygenation:
plications and mortality of veno-venous extracor- impact on heparin dose and circuit life. ASAIO J.
poreal membrane oxygenation for refractory acute 2014;60:57–62.
respiratory distress syndrome. Ann Intensive Care. 22. Ryerson LM, Bruce AK, Lequier L, Kuhle S,
2017;7:51. Massicotte MP, Bauman ME. Administration of
7. Gorbet MB, Sefton MV. Biomaterial-associated Antithrombin Concentrate in infants and children on
thrombosis: roles of coagulation factors, comple- ECLS improves anticoagulation efficacy. ASAIO J.
ment, platelets and leukocytes. Biomaterials. 2014;60(5):559–63.
2004;25:5681–703. 23. Young G, Yonekawa KE, Nakagawa PA, Blain RC,
8. Bembea MM, Annich G, Rycus P, Oldenburg G, Lovejoy AE, Nugent DJ. Recombinant activated fac-
Berkowitz I, Pronovost P. Variability in anticoagula- tor VII effectively reverses the anticoagulant effects
tion management of patients on extracorporeal mem- of heparin, enoxaparin, fondaparinux, argatroban, and
brane oxygenation: an international survey. Pediatr bivalirudin ex vivo as measured using thromboelastog-
Crit Care Med. 2013;14:e77–84. raphy. Blood Coagul Fibrinolysis. 2007;18:547–53.
9. ECLS ELSO-Ea (2017) Extracorporeal Life Support 24. Gladwell TD. Bivalirudin: a direct thrombin inhibitor.
Organization - ECMO and ECLS > Publications > Clin Ther. 2002;24:38–58.
Red Book 5th Edition. 25. Hirsh J, O’Donnell M, Weitz JI. New anticoagulants.
10. Annich GM, Zaulan O, Neufeld M, Wagner D,
Blood. 2005;105:453–63.
Reynolds MM. Thromboprophylaxis in extracor- 26.
Ranucci M, Ballotta A, Kandil H, Isgrò G,
poreal circuits: current pharmacological strate- Carlucci C, Baryshnikova E, Pistuddi V, Group
gies and future directions. Am J Cardiovasc Drugs. SaCOR. Bivalirudin-based versus conventional hepa-
2017;17:425. rin anticoagulation for postcardiotomy extracorporeal
11. Bain J, Flannery AH, Flynn J, Dager W. Heparin membrane oxygenation. Crit Care. 2011;15:R275.
induced thrombocytopenia with mechanical circu- 27. Pieri M, Agracheva N, Bonaveglio E, Greco T, De
latory support devices: review of the literature and Bonis M, Covello RD, Zangrillo A, Pappalardo
management considerations. J Thromb Thrombolysis. F. Bivalirudin versus heparin as an anticoagulant
2017;44:76–87. during extracorporeal membrane oxygenation: a
12. Winkler AM. Managing the precarious hemostatic
case-control study. J Cardiothorac Vasc Anesth.
balance during extracorporeal life support: implica- 2013b;27:30–4.
tions for coagulation laboratories. Semin Thromb 28. Jyoti A, Maheshwari A, Daniel E, Motihar A,
Hemost. 2017;3:291–9. Bhathiwal RS, Sharma D. Bivalirudin in venove-
13. Rapaport SI, Vermylen J, Hoylaerts M. The multiple nous extracorporeal membrane oxygenation. J Extra
faces of the partial thromboplastin time APTT. J Corpor Technol. 2014;46:94–7.
Thromb Haemost. 2004;2(12):2250–9. 29. Nagle EL, Dager WE, Duby JJ, Roberts AJ, Kenny
14. Dalton HJ, Reeder R, Pamela G-F, et al. Factors
LE, Murthy MS, Pretzlaff RK. Bivalirudin in pediat-
associated with bleeding and thrombosis in children ric patients maintained on extracorporeal life support.
receiving extracorporeal membrane oxygenation. Am Pediatr Crit Care Med. 2013;14:e182–8.
J Respir Crit Care Med. 2017;196:762–71. 30. Shammas NW. Bivalirudin: pharmacology and clinical
15. Ranucci M. Antithrombin III. Key factor in extracorpo- applications. Cardiovasc Drug Rev. 2005;23:345–60.
real circulation. Minerva Anestesiol. 2002;68:454–7. 31. Van De Car DA, Rao SV, Ohman EM. Bivalirudin: a
16. Finley A, Greenberg C. Review article: heparin sensi- review of the pharmacology and clinical application.
tivity and resistance: management during cardiopul- Expert Rev Cardiovasc Ther. 2010;8:1673–81.
monary bypass. Anesth Analg. 2013;116:1210–22. 32. Beiderlinden M, Treschan T, Gorlinger K, Peters
17. Ryerson LM, Lequier LL. Anticoagulation manage- J. Argatroban in extracorporeal membrane oxygen-
ment and monitoring during pediatric extracorporeal ation. Artif Organs. 2007;31:461–5.
life support: a review of current issues. Front Pediatr. 33. Dolch ME, Frey L, Hatz R, Uberfuhr PA, Beiras-
2016;4:67. Fernandez A, Behr J, Irlbeck M, Lung Transplant
18. Beattie GW, Jeffrey RR. Is there evidence that fresh Group TM. Extracorporeal membrane oxygenation
frozen plasma is superior to antithrombin adminis- bridging to lung transplant complicated by heparin-
tration to treat heparin resistance in cardiac surgery? induced thrombocytopenia. Exp Clin Transplant.
Interact Cardiovasc Thorac Surg. 2014;18:117–20. 2010;8:329–32.
19. Wong TE, Huang YS, Weiser J, Brogan TV, Shah 34. Johnston N, Wait M, Huber L. Argatroban in adult
SS, Witmer CM. Antithrombin concentrate use extracorporeal membrane oxygenation. J Extra
in children: a multicenter cohort study. J Pediatr. Corpor Technol. 2002;34:281–4.
2013;163:1329–34.e1321 35. Mejak B, Giacomuzzi C, Heller E, You X, Ungerleider
20. Niebler RA, Christensen M, Berens R, Wellner H, R, Shen I, Boshkov L. Argatroban usage for antico-
Mikhailov T, Tweddell JS. Antithrombin replacement agulation for ECMO on a post-cardiac patient with
ERRNVPHGLFRVRUJ
240 D. Eytan and G. M. Annich
heparin-induced thrombocytopenia. J Extra Corpor 48. Sievert AN, Shackelford AG, McCall MM. Trends
Technol. 2004;36:178–81. and emerging technologies in extracorporeal life sup-
36. Phillips MR, Khoury AI, Ashton RF, Cairns BA,
port: results of the 2006 ECLS survey. J Extra Corpor
Charles AG. The dosing and monitoring of argatroban Technol. 2009;41:73–8.
for heparin-induced thrombocytopenia during extra- 49. Koster A, Sanger S, Hansen R, Sodian R, Mertzlufft
corporeal membrane oxygenation: a word of caution. F, Harke C, Kuppe H, Hetzer R, Loebe M. Prevalence
Anaesth Intensive Care. 2014;42:97–8. and persistence of heparin/platelet factor 4 antibodies
37. Scott LK, Grier LR, Conrad SA. Heparin-induced in patients with heparin coated and noncoated ven-
thrombocytopenia in a pediatric patient receiving tricular assist devices. ASAIO J. 2000;46:319–22.
extracorporeal membrane oxygenation managed with 50. Pappalardo F, Maj G, Scandroglio A, Sampietro
argatroban. Pediatr Crit Care Med. 2006;7:473–5. F, Zangrillo A, Koster A. Bioline heparin-coated
38. Young G, Yonekawa KE, Nakagawa P, Nugent
ECMO with bivalirudin anticoagulation in a patient
DJ. Argatroban as an alternative to heparin in extra- with acute heparin-induced thrombocytopenia: the
corporeal membrane oxygenation circuits. Perfusion. immune reaction appeared to continue unabated.
2004;19:283–8. Perfusion. 2009;24:135–7.
39. Cornell T, Wyrick P, Fleming G, Pasko D, Han Y, 51. Silvetti S, Koster A, Pappalardo F. Do we need hepa-
Custer J, Haft J, Annich G. A case series describing rin coating for extracorporeal membrane oxygenation?
the use of argatroban in patients on extracorporeal cir- New concepts and controversial positions about coat-
culation. ASAIO J. 2007;53:460–3. ing surfaces of extracorporeal circuits. Artif Organs.
40. Fosse E, Thelin S, Svennevig JL, Jansen P, Mollnes 2015;39:176–9.
TE, Hack E, Venge P, Moen O, Brockmeier V, 52. De Somer F, Francois K, van Oeveren W, Poelaert J,
Dregelid E, Halden E, Hagman L, Videm V, Pedersen De Wolf D, Ebels T, Van Nooten G. Phosphorylcholine
T, Mohr B. Duraflo II coating of cardiopulmonary coating of extracorporeal circuits provides natural
bypass circuits reduces complement activation, but protection against blood activation by the material
does not affect the release of granulocyte enzymes: a surface. Eur J Cardiothorac Surg. 2000;18:602–6.
European multicentre study. Eur J Cardiothorac Surg. 53. Pieri M, Turla OG, Calabro MG, Ruggeri L,
1997;11:320–7. Agracheva N, Zangrillo A, Pappalardo F. A new phos-
41. Gunaydin S, McCusker K, Sari T, Onur MA, Zorlutuna phorylcholine-coated polymethylpentene oxygenator
Y. Clinical performance and biocompatibility of hyal- for extracorporeal membrane oxygenation: a prelimi-
uronan-based heparin-bonded extracorporeal circuits nary experience. Perfusion. 2013a;28:132–7.
in different risk cohorts. Interact Cardiovasc Thorac 54. von Segesser LK, Tonz M, Leskosek B, Turina
Surg. 2010;10:371–6. M. Evaluation of phospholipidic surface coatings ex-
42. Jansen PG, te Velthuis H, Huybregts RA, Paulus R, vivo. Int J Artif Organs. 1994;17:294–9.
Bulder ER, van der Spoel HI, Bezemer PD, Slaats EH, 55. Yu J, Lamba NM, Courtney JM, Whateley TL, Gaylor
Eijsman L, Wildevuur CR. Reduced complement acti- JD, Lowe GD, Ishihara K, Nakabayashi N. Polymeric
vation and improved postoperative performance after biomaterials: influence of phosphorylcholine polar
cardiopulmonary bypass with heparin-coated circuits. groups on protein adsorption and complement activa-
J Thorac Cardiovasc Surg. 1995;110:829–34. tion. Int J Artif Organs. 1994;17:499–504.
43. Mahmood S, Bilal H, Zaman M, Tang A. Is a fully 56. Boning A, Scheewe J, Ivers T, Friedrich C, Stieh J,
heparin-bonded cardiopulmonary bypass circuit supe- Freitag S, Cremer JT. Phosphorylcholine or heparin
rior to a standard cardiopulmonary bypass circuit? coating for pediatric extracorporeal circulation causes
Interact Cardiovasc Thorac Surg. 2012;14:406–14. similar biologic effects in neonates and infants. J
44. McCarthy PM, Yared JP, Foster RC, Ogella DA, Borsh Thorac Cardiovasc Surg. 2004;127:1458–65.
JA, Cosgrove DM 3rd. A prospective randomized trial 57. Ranucci M, Isgro G, Soro G, Canziani A, Menicanti
of Duraflo II heparin-coated circuits in cardiac reop- L, Frigiola A. Reduced systemic heparin dose with
erations. Ann Thorac Surg. 1999;67:1268–73. phosphorylcholine coated closed circuit in coronary
45. Sohn N, Marcoux J, Mycyk T, Krahn J, Meng Q. The operations. Int J Artif Organs. 2004;27:311–9.
impact of different biocompatible coated cardiopul- 58. Ranucci M, Pazzaglia A, Isgro G, Cazzaniga A,
monary bypass circuits on inflammatory response and Ditta A, Boncilli A, Cotza M, Carboni G, Brozzi S,
oxidative stress. Perfusion. 2009;24:231–7. Bonifazi C. Closed, phosphorylcholine-coated circuit
46. Mangoush O, Purkayastha S, Haj-Yahia S, Kinross and reduction of systemic heparinization for cardio-
J, Hayward M, Bartolozzi F, Darzi A, Athanasiou pulmonary bypass: the intraoperative ECMO concept.
T. Heparin-bonded circuits versus nonheparin-bonded Int J Artif Organs. 2002;25:875–81.
circuits: an evaluation of their effect on clinical out- 59. Jacobs S, De Somer F, Vandenplas G, Van Belleghem
comes. Eur J Cardiothorac Surg. 2007;31:1058–69. Y, Taeymans Y, Van Nooten G. Active or passive bio-
47. Ranucci M, Balduini A, Ditta A, Boncilli A, Brozzi coating: does it matters in extracorporeal circulation?
S. A systematic review of biocompatible cardiopul- Perfusion. 2011;26:496–502.
monary bypass circuits and clinical outcome. Ann 60. Zonies D. ECLS in trauma: practical application and a
Thorac Surg. 2009;87:1311–9. review of current status. World J Surg. 2017;41:1159–64.
ERRNVPHGLFRVRUJ
13 Anticoagulation for Extracorporeal Life Support 241
61. Huang YK, Tsai FC, Tseng CN, Wang YC, Chang Roewer N, Wurmb T. Prolonged heparin-free extra-
YS, Chu JJ, Lin PJ. Versatile use of extra-corporeal corporeal membrane oxygenation in multiple injured
life support to resuscitate acute respiratory distress acute respiratory distress syndrome patients with
patients. Int J Clin Pract. 2007;61:589–93. traumatic brain injury. J Trauma Acute Care Surg.
62. Bedeir K, Seethala R, Kelly E. Extracorporeal life 2012;72:1444–7.
support in trauma: worth the risks? A systematic 64. Major TC, Brisbois EJ, Jones AM, et al. The effect
review of published series. J Trauma Acute Care Surg. of a polyurethane coating incorporating both a
2017;82:400–6. thrombin inhibitor and nitric oxide on hemocompat-
63. Muellenbach RM, Kredel M, Kunze E, Kranke
ibility in extracorporeal circulation. Biomaterials.
P, Kuestermann J, Brack A, Gorski A, Wunder C, 2014;35(26):7271–85.
ERRNVPHGLFRVRUJ
Part VI
Immunologic Controversies
ERRNVPHGLFRVRUJ
Secondary Hemophagocytic
Lymphohistiocytosis, 14
Macrophage Activation
Syndrome, and Hyperferritinemic
Sepsis-Induced Multiple-Organ
Dysfunction Syndrome
in the Pediatric ICU
Joseph A. Carcillo, Bita Shakoory,
and Leticia Castillo
ERRNVPHGLFRVRUJ
246 J. A. Carcillo et al.
Secondary HLH/MAS
Antibiotics/Supportive therapy
IVIG, Plasma Exchange,
Interleukin 1 receptor antagonist
Fig. 14.1 Five of the eight criteria may represent familial sepsis-induced multiple-organ dysfunction syndrome
hemophagocytic lymphohistiocytosis (HLH), macro- (MODS), each of which has a different therapeutic
phage activation syndrome (MAS), or hyperferritinemic approach
ERRNVPHGLFRVRUJ
14 Secondary Hemophagocytic Lymphohistiocytosis, Macrophage Activation Syndrome… 247
ERRNVPHGLFRVRUJ
248 J. A. Carcillo et al.
MODS. Steinberg and colleagues developed the conditions. Uncontrolled inflammation can be
sterile model of zymosan (the cell wall of the fun- due in part to ineffective NK cell cytolytic func-
gus saccharomyces A) plus mineral oil injected tion. Defects or deficiencies in the ability of the
intraperitoneally to induce MODS in rodents [8]. NK cell to kill viruses and cancer cells and to
This model results in initial hypovolemic shock turn off the host reticuloendothelial system,
followed by persistent macrophage activation. macrophage, dendritic cell, and lymphocyte
Injection of either zymosan or mineral oil alone activation can be related to one of the three
does not induce MODS, suggesting the need for a pathologic conditions: (1) absent NK cell cyto-
“two-hit” insult of both toll-like receptor (TLR) lytic activity unrelated to numbers of NK cells
stimulation by zymosan and unremitting particu- present (familial HLH), (2) reduced NK cell
late irritation by mineral oil to induce persistent cytolytic activity unrelated to numbers of NK
macrophage activation. In another rodent model, cells present (sHLH/MAS a ssociated with rheu-
Behrens and colleagues reported that repeated (i.e., matologic disease), and (3) normal NK cell
not single time) TLR9 stimulation with CpG oligo- cytolytic activity per cell but NK cell cytopenia
deoxynucleotides transformed an otherwise innoc- (sepsis-induced hyperferritinemic MODS).
uous endotoxin challenge (TLR4 stimulation) in These different mechanisms are further illus-
mice into a MODS model of macrophage activa- trated in Fig. 14.2.
tion with cytopenias, splenomegaly, hyperferri- Absent NK cell cytolytic activity is the basis
tinemia, and hepatitis [9]. Similarly, in another for familial HLH, which itself is comprised of a
murine model of cecal ligation and perforation- group of monogenic autosomal recessive or
induced sepsis, additional CpG injection induced X-linked primary immune deficiency diseases
cytokine production by macrophages and hepatic characterized by the absence of crucial compo-
mononuclear cells, followed by the development of nents of the perforin-granzyme pathway needed
liver injury and MODS-induced mortality [10]. for NK cells to kill viruses and induce apoptosis
Importantly, the sHLH phenotype elicited with in cancer and host inflammatory cells. Gene
repeated TLR9 stimulation is exacerbated in knockout models for this pathway result in
knockout mice deficient in native hepatic IL-1 murine HLH and death after infection with oth-
receptor antagonist protein production but can be erwise innocuous LCM (lymphocytic chorio-
ameliorated by interferon-α-induced production meningitis) virus infection. In the wild-type
of IL-1 receptor antagonist protein or with direct mouse exposed to LCM virus infection, host
administration of recombinant IL-1 receptor antag- NK and T cells respond to control the virus,
onist protein (anakinra) in vivo [11]. In short, liver whereas in the perforin knockout mouse, LCM
dysfunction in this model appears to be related in virus infection cannot be killed leading to T-cell
part to IL-1-mediated inflammation [12]. proliferation and activation and consequent
overproduction of interferon-γ (reflected by
very high levels of CXCL9, the monokine
Different Inflammation induced by interferon-γ). T lymphocyte-derived
Pathobiologies in sHLH interferon-γ induces macrophage activation and
organ injury. Etoposide, one of the traditional
The clinical criteria used to describe the con- treatments for familial HLH, destroys prolifer-
stellation of symptoms and signs indicative of ating activated T cells and reduces interferon-γ
these syndromes are considered to be biomark- production, preventing undue macrophage
ers for a state of uncontrolled macrophage and T activation.
lymphocyte inflammation. The NK cell, as the Reduced NK cytolytic activity is the basis of
most important “cellular” controller of macro- rheumatologic or drug-induced MAS. For
phage and T lymphocyte activation, is consid- instance, some patients with systemic juvenile
ered central to the pathobiology of these idiopathic arthritis (sJIA), a known trigger for
ERRNVPHGLFRVRUJ
14 Secondary Hemophagocytic Lymphohistiocytosis, Macrophage Activation Syndrome… 249
Fig. 14.2 Different natural killer (NK) cell pathobiolo- (light blue NK cells = reduced cytolytic activity); and
gies: familial hemophagocytic lymphohistiocytosis (HLH) hyperferritinemic sepsis-induced multiple-organ dysfunc-
(white NK cells = absent cytolytic activity); rheumatologic tion syndrome (MODS) (dark blue NK cells = low num-
disease-related macrophage activation syndrome (MAS) bers of NK cells with normal cytolytic activity)
sHLH, are hypomorphic or heterozygotes for the cancer cells as well as to induce apoptosis in
perforin-granzyme pathway gene variants. As het- activated macrophages. Because T cells and
erozygotes, these patients have some NK cytolytic interferon-
gamma production are already low
activity. For the most part, these patients respond to absent, etoposide is unlikely to be of benefit
nicely to anti-inflammatory therapies such as in reducing macrophage activation in these
anakinra (IL-1 receptor antagonist protein). The children. Indeed, etoposide may worsen out-
IL-1 receptor antagonist protein anakinra is FDA comes in sepsis patients by preventing the recov-
approved for inflammasome- driven conditions ery of lymphocyte counts needed to resolve
(e.g., sJIA). IL-18, the interferon-γ-inducible fac- infection. Compared to patients with other forms
tor, is also increased in these patients. There is of HLH, those with hyperferritinemic sepsis-
optimism that IL-18-binding protein (which neu- induced MODS have lower production of
tralizes IL-18) as well as interferon-γ monoclonal interferon-γ-inducing IL-18. The promise of
antibody could help these patients. interferon-γ monoclonal antibodies in treating
Reduced number of NK cells with normal NK familial HLH and MAS is therefore less likely to
cytolytic activity per cell is the basis of hyperfer- be realized in hyperferritinemic sepsis-induced
ritinemic sepsis-induced MODS. When NK cell MODS. Further, inability to produce interferon-γ
numbers recover, then inflammation and MODS is associated with increased mortality in experi-
resolve. During bacterial infection, NK cells mental models of sepsis.
switch from an overall low cytokine-producing, Two interferon-γ-independent pathways that
high cytolytic activity phenotype to a high induce macrophage activation-related hyperfer-
cytokine-producing, low cytolytic activity phe- ritinemia in patients with sepsis-induced MODS
notype [13]. In hyperferritinemic sepsis-induced are free hemoglobin and DNA viremia. These
MODS, NK cell cytopenia and T-cell cytopenia pathways are depicted in Fig. 14.3 [14].
occur with reticuloendothelial system activation Endotheliopathy in sepsis leads to hemolysis,
but decreased to absent interferon-γ production. particularly in patients with gene variants related
When the reduction in NK cell and T-cell num- to atypical hemolytic uremic syndrome (HUS)
bers is below 10% of normal, there is a signifi- and low to absent inhibitory complement produc-
cant decrease in host ability to kill viruses and tion. The released free hemoglobin complexes
ERRNVPHGLFRVRUJ
250 J. A. Carcillo et al.
Activated macrophage produces ferritin, which increases innate inflammation that induces IL-1
mediated liver injury while also suppressing adaptive immunity and host ability to quell viral infection
Fig. 14.3 Interferon-γ-independent pathways to macrophage activation during sepsis include free hemoglobin from
hemolysis or blood transfusions, as well TLR9 stimulation from DNA viral reactivation
ERRNVPHGLFRVRUJ
14 Secondary Hemophagocytic Lymphohistiocytosis, Macrophage Activation Syndrome… 251
c hildren (i.e., without these familial HLH risk fac- Because three of the eight clinical criteria
tors) were diagnosed with sHLH and had five of (e.g., soluble CD25 levels, NK cytotoxicity,
the aforementioned eight clinical criteria and had and hemophagocytosis) used to identify
five to six organ failures. Centers in one treatment patients with HLH are not easily accessible
cohort administered the familial HLH protocol of tests, rheumatologists have sought to redefine
dexamethasone and/or etoposide along with daily sHLH with other criteria sets that use more
plasma exchange to these children with sHLH and readily available laboratory tests. For example,
observed a 50% mortality rate, whereas centers in utilizing current literature, Ravelli and col-
the other treatment cohort administered a less leagues have provided a consensus statement
immune suppressive regimen of methylpredniso- defining sHLH in a child with known systemic
lone with or without intravenous immunoglobulin juvenile idiopathic arthritis (sJIA), a common
(IVIG) with daily plasma exchange and observed a trigger of macrophage activation, based on the
0% mortality rate. Based on the above data, though presence of fever, ferritin >684 ng/mL, and any
it may be appropriate to use the diagnostic criteria two of the following: platelet count <181 K,
for familial HLH to diagnose sHLH, the traditional ALT >48 IU/L, triglycerides >156 mg/dL, and
familial HLH protocol of dexamethasone and eto- fibrinogen <360 mg/dL [16]. Though recom-
poside should be replaced by less toxic treatment mended for patients with MAS secondary to
strategies for patients with sHLH. sJIA, these criteria can be applied to patients
with MAS secondary to other rheumatologic
diseases as well.
Clinical Vignette 2: sHLH Induced by Shakoory and colleagues have offered
Systemic Juvenile Arthritis another method for simplifying the diagnosis
A febrile 7-year-old child presents to an of and treatment of sHLH [17]. Specifically,
ICU with rash, leukocytosis, arthritis, and they suggested that the combination of hepato-
laboratory data concerning for sHLH. The biliary dysfunction and disseminated intravas-
intensivist calls the rheumatologist, who cular coagulation (DIC) can be representative
believes the presentation is consistent with of sHLH including hyperferritinemic sepsis-
systemic juvenile arthritis-related MAS. She induced MODS. Further, they hypothesized
recommends treatment with methylprednis- that if the combination of these two organ dys-
olone and the IL-1 receptor antagonist functions represents sHLH, then treatment
anakinra. Other laboratory testing is sent to with IL-1 receptor blockade should improve
rule out systemic lupus erythematosus, sar- sepsis-related macrophage activation as it
coidosis, scleroderma, Sjogren’s syndrome, does in the previously mentioned experimental
and Kawasaki’s disease. model [10] and in children with sJIA-related
Patients with autoimmune rheumatologic MAS [15–17]. In their secondary analysis of
disease have increased inflammasome acti- an adult with severe sepsis IL-1 receptor
vation and reduced NK activity without lym- blockade trial, Shakoory and colleagues com-
phoproliferation. Accordingly, rather than pared patients with combined hepatobiliary
etoposide and dexamethasone regimens, dysfunction (HBD) and DIC (HBD + DIC) to
methylprednisolone and anakinra are typi- those without this combination (non-
cally recommended to control inflammasome HBD + DIC) [17]. The investigators found
activation associated with rheumatologic the following: (1) 5.6% of severe sepsis
disease. Other biologics and chemotherapeu- patients had HBD + DIC; (2) patients with
tic regimens, however, such as cyclophospha- HBD + DIC had a higher incidence of shock
mide, methotrexate, tocilizumab, or etoposide (HBD + DIC = 95% versus non-
as well as plasma exchange, are considered if HBD + DIC = 79%) and acute kidney injury
the patient remains recalcitrant. (HBD + DIC = 61% versus non-
HBD + DIC = 29%), but not acute respiratory
ERRNVPHGLFRVRUJ
252 J. A. Carcillo et al.
Table 14.1 Simplified clinical criteria used to diagnose rheumatologic or drug-related secondary hemophagocytic
lymphohistiocytosis (HLH)/macrophage activation syndrome (MAS) and hyperferritinemic sepsis-induced multiple-
organ dysfunction syndrome (MODS)
Hyperferritinemic sepsis-induced
Familial HLH: five of eight criteria Rheumatologic/drug-induced MAS MODS
• Fever • Fever • Hepatobiliary dysfunction
• Ferritin > 500 ng/ml • Ferritin > 684 ng/mL • Disseminated intravascular
• Hypertriglyceridemia • With any two of the following: coagulation
• Hypofibrinogenemia Platelet count < 181,000 • +/- ferritin > 500 ng/mL
• Two-line cytopenia Alanine transferase > 48 IU/L
• Splenomegaly Triglyceride > 156 mg/dL
• Natural killer cell cytolytic Fibrinogen < 360 mg/dL
activity < 10%
• Hemophagocytosis
• Elevated soluble CD25
ERRNVPHGLFRVRUJ
14 Secondary Hemophagocytic Lymphohistiocytosis, Macrophage Activation Syndrome… 253
ERRNVPHGLFRVRUJ
254 J. A. Carcillo et al.
Undesired response
Consider new or
uneradicated infection
Ferritin Low Desired Response
Hold immunesuppression
and treat infection
Fig. 14.4 An approach to monitoring inflammation and sepsis-induced multiple-organ dysfunction syndrome
infection response during initiation of therapy for hemo- (MODS) by observing changes in C-reactive protein
phagocytic lymphohistiocytosis (HLH), macrophage (CRP) and ferritin. C-reactive protein reflects bacterial
activation syndrome (MAS), and hyperferritinemic
infection, whereas ferritin reflects macrophage activation
Suspect sepsis?
Obtain blood culture and give empiric antibiotics within one hour
Erythroderma observed?
Give IVIG 2 g/kg over 4 days + Anakinra 2.5 mg/kg q 6hrs for three days (maximum 100 mg per dose)
Consider Methylprednisone up to 30 mg/kg per day x 3 days
Thrombocytopenia + Acute Kidney injury + LDH > 200?
Five days of plasma exchange 1 1/2 volumes on day one and one volume daily until Ferritin < 500 ng/mL
ERRNVPHGLFRVRUJ
14 Secondary Hemophagocytic Lymphohistiocytosis, Macrophage Activation Syndrome… 255
ERRNVPHGLFRVRUJ
Diagnosis and Management
of Fungal Infections 15
in the Pediatric Intensive Care Unit
Christine L. Joyce, Christine M. Salvatore,
and James S. Killinger
ERRNVPHGLFRVRUJ
258 C. L. Joyce et al.
ERRNVPHGLFRVRUJ
15 Diagnosis and Management of Fungal Infections in the Pediatric Intensive Care Unit 259
ERRNVPHGLFRVRUJ
260 C. L. Joyce et al.
Given the potential for earlier diagnosis, options warranted; however, the most recent US guide-
for therapeutic monitoring, and decreased lines, revised in 2017, recommend consideration
morbidity and mortality, there is a significant
against routinely testing serum GM levels in
focus on better understanding the role of bio- patients with persistent febrile neutropenia [33,
markers in children [24]. 34]. This is based on its poor PPV, and fact that
the high NPV does not rule out other non-
aspergillus molds [34].
Galactomannan Assay The utility of GM as a screening tool is great-
est in high-risk patients with significant concern
GM, which is a component of the cell wall spe- for aspergillosis. It is possible that its use in
cific to Aspergillus spp. and released during the non-serum samples, particularly those from a
growth phase, is designed for the diagnosis of bronchoalveolar lavage, when used in addition
IA. It is the most widely studied fungal bio- to serum samples, may increase the overall sen-
marker in children [24]. When used in patients sitivity and specificity [35]. Adult studies have
meeting EORTC/MSG criteria for proven inva- also demonstrated that initial levels may be pre-
sive fungal disease, GM has a sensitivity of 76% dictive of outcome and that follow-up levels can
and specificity of 86% [25]. In the sole study be utilized to correlate for improvement in
performed on non-hematologic pediatric ICU extent of disease [36]. Thus while a positive
patients, it was found to have a sensitivity of marker coupled with risk factors and clinical
90% and specificity of 35.6% [26]. The sensitiv- condition may assist in diagnosis, significant
ity of GM is lacking in certain patient popula- caution needs to be taken when interpreting a
tions, however, including non-neutropenic solid negative result to ensure the clinician isn’t led
organ transplant recipients and those with cer- into excluding fungal disease as an etiology of a
tain primary immunodeficiencies [27, 28]. Its patient’s deterioration.
utility for aiding in the diagnosis of IA for
patients taking mold-active antifungal prophy-
laxis has also been questioned. Current recom- BDG Assay
mendations advise against use of GM in these
patients due to nullification of the GM signal and BDG is a major cell wall component of various
subsequent false negative testing [29]. medically important fungi. The BDG assay is
Additionally, its positive predictive value therefore more broadly applicable and used to
(PPV), when evaluated in a systematic review of detect Aspergillus spp., Candida spp., Fusarium
patients with cancer who underwent hematopoi- spp., Trichosporon spp., Saccharomyces cerevi-
etic stem cell transplantation (HSCT), was <75% siae, Acremonium spp., Coccidioides immitis,
in most studies and <50% in half of studies [30]. Histoplasma capsulatum, Sporothrix schenckii,
The other major concern surrounds high false- and Pneumocystis jirovecii. Sensitivity and speci-
positive rates, much higher in pediatric com- ficity in pediatric specific studies vary broadly.
pared to adult populations [31]. This is Reported sensitivities range from 50% to 82%
particularly true for premature infants [32]. Its and specificity from 46% to 82% [24]. In the one
strength appears to be in its consistently high study performed on PICU patients without hema-
negative predictive value (NPV) among high- tologic disease, sensitivity was reported at 53.3%
risk patients [26, 30]. and specificity at 63.2% [26].
Recommendations regarding use of GM as a One of the major issues surrounding the use of
screening tool are therefore controversial without BDG in pediatric patients is a lack of standard
a consensus. In a subset of high-risk patients, cutoff value in which to interpret a positive result.
those with certain cancers or undergoing HSCT, Studies suggest the adult cutoff of 80 pg/mL is
the European and US guidelines are conflicting. likely too low and that baseline BDG are higher
The European guidelines suggest screening is in children than adults [37]. Two pediatric studies
ERRNVPHGLFRVRUJ
15 Diagnosis and Management of Fungal Infections in the Pediatric Intensive Care Unit 261
demonstrated dramatically higher median BDG was negative in all 28 controls. Additionally,
levels in cases compared to controls (480 pg/mL PCR results were available within 24 h com-
vs. 61 pg/mL and 372 pg/mL compared to 57 pg/ pared to 48–96 h with respect to blood culture
mL); healthy pediatric controls were also found results. Detection limit was found to be 4 CFU/
to have levels as high as 473 pg/mL [26, 38]. mL as compared to 10 CFU/mL needed for
Using the data from one of these studies, receiver blood cultures [41].
operating characteristic (ROC) curves were con- The largest pediatric study evaluating the
structed and demonstrated a level of 177 pg/mL utility of PCR for aspergillosis was conducted
provided an optimal sensitivity of 78% and speci- in patients with hematologic or solid tumor
ficity of 91% [38]. malignancy. Using EORTC/MSG 2008 criteria,
The PPV of BDG is poor, with values ranging 125 patients were classified as either proven,
from 33.3% to 46.67% [26, 38]. A significant probable, or no IFD. When comparing GM,
number of medications and medical products BDG, and PCR, PCR had the highest sensitiv-
used in the critical care setting have been shown ity, specificity, and PPV at 82.7%, 54% and
to increase false-positive rates. Examples of these 72.9%, respectively. When various combina-
with frequent use include albumin infusions, tions were used, all parameters increased and
IVIG, piperacillin-tazobactam, ampicillin-were uniformly highest with the combination
clavulanate, alcohol swabs, and active antifungal of PCR and GM, with a sensitivity of 87.7%,
use [39]. specificity of 64%, PPV of 73.9%, and NPV of
The BDG assay is therefore not recommended 73% [42].
in any pediatric specific guidelines for routine Fungal PCR represents a promising avenue
use in screening or treatment management, and that may allow for more timely diagnosis, with
further studies are needed to determine optimal increased specificity and sensitivity. Currently,
cutoff values for children. Its value appears to be the lack of standardization or validation of clini-
in its consistently high NPV, with values as high cal technique prevents its recommendation in
as 93% [38], and thus the exclusion of invasive routine use [23]. It is likely that future recom-
fungal disease. Additionally, it has been shown to mendations will include a combination of bio-
be of value in guiding length of treatment, par- marker and PCR testing; currently there is not
ticularly in CSF, where a case series demon- enough known about pediatric specific values
strated reversion to negative levels in all nine and cutoffs to allow for accurate interpretation.
patients [40].
To date, fungal PCR has been evaluated most Mannans are also main cell wall components of
frequently in the diagnosis of IC and IA, with Candida spp. Studies evaluating Candida antigen
studies demonstrating sensitivities of 63–100% testing in non-neonatal pediatric patients are few
[24]. The largest pediatric study evaluating the [24]. The one study in PICU patients demon-
utility of PCR for candidemia looked at 54 strated sensitivities of 100% and 60% for the
children hospitalized in the ICU setting with mannan antigen and anti-mannan antibody assay,
suspected infection. The study utilized a multi- respectively. A high false-positive rate of 23%
plex PCR, designed to detect multiple targets was also reported [43]. While neonatal studies
simultaneously for the seven most causative have also demonstrated high sensitivities, the test
Candida strains. A total of 15% of patients had has demonstrated a consistent inability to detect
positive blood cultures, and 24% of patients C. parapsilosis [24], which, as discussed previ-
had positive PCR results (including all eight ously, composes the majority of non-albicans
patients with positive cultures). PCR testing candida infections [7].
ERRNVPHGLFRVRUJ
262 C. L. Joyce et al.
ERRNVPHGLFRVRUJ
15 Diagnosis and Management of Fungal Infections in the Pediatric Intensive Care Unit 263
ERRNVPHGLFRVRUJ
264 C. L. Joyce et al.
Table 15.3 (continued)
Mechanism of
Medication Class action Toxicity Indication Pediatric dosing
Micafungin Echinocandin Inhibits Hepatotoxicity -First-line therapy in <40 kg:
synthesis of the (less), leukopenia, neutropenic patients 4–5 mg/kg IV
glucose neutropenia, with IC every 24 h
homopolymer hemolytic anemia, -Salvage therapy in IA >40 kg:
BDG synthase GI intolerance, 50–150 mg IV
fever, headache, every 24 h
mucositis,
hypokalemia
Utilization is advised only in the neonatal population to avoid renal dysfunction; liposomal amphotericin is recom-
a
should be given to antifungal lock therapy, as mendations difficult [48, 52]. Therapeutic drug
an emerging body of literature demonstrates monitoring is currently recommended, though
feasibility and efficacy [50]. notably, it is unclear if higher levels correlate to
Treatment should continue for at least 2 weeks clinical efficacy. A rational approach suggests
following resolution of symptoms and negative aiming for levels above the MIC of the organism
culture results. In the case of hepatosplenic can- (>0.5 mcg/mL) and below toxic levels [48].
didiasis, treatment should continue through reso- Amphotericin B deoxycholate, and its lipid
lution of abdominal lesions [49]. derivatives, is recommended as first-line treat-
Though prophylaxis is recommended in cer- ment against IA when voriconazole cannot be
tain adult populations [49], there are currently no administered [29]. The clinician should be mind-
recommendations for prophylaxis in pediatric ful of the increased treatment response and sur-
patients. Utilization of validated clinical predic- vival benefits of voriconazole compared to
tion tools in adult ICUs has been helpful for iden- amphotericin B when making this decision [53].
tifying at-risk patients [51]. Though a pediatric Amphotericin B can also be used as salvage ther-
model was proposed, it failed validation in fur- apy, with a preference for the lipid derivatives
ther studies [11, 27]. given higher tissue concentrations and decrease
in nephrotoxicity [29, 48]. Echinocandins, indi-
vidually or in combination, are an alternative for
Treatment of Invasive Aspergillosis salvage therapy [29]. Combination therapy, cur-
rently without a sound evidence base but widely
During the past decade, there has been a notable used, remains a debated topic with need for fur-
increase in the development of antifungals effec- ther clinical trials, specifically in pediatrics.
tive against IA; however, only voriconazole, a For patients at high risk for IA with prolonged
second-generation triazole, and amphotericin B neutropenia, prophylaxis is strongly recom-
deoxycholate, a polyene and its lipid formula- mended, with continued prophylaxis, while
tions, are FDA-approved in the United States for patients remain immunosuppressed. Options
primary treatment. Though pediatric patients are include posaconazole, voriconazole, and mica-
treated with the same antifungals as adults, dos- fungin [29].
ing varies and in some cases is not yet known,
resulting in potential underdosing with conse-
quent clinical failure. For example, voriconazole, Empiric Therapy for PICU Patients
a first-line treatment against IA, has a linear phar-
macokinetic profile in children and nonlinear in In those PICU patients with underlying malig-
adults. Moreover, high inter-patient variability nancy or who have undergone HSCT, the recom-
has also been reported, making dosing recom- mendation for empiric antifungal therapy is clear,
ERRNVPHGLFRVRUJ
15 Diagnosis and Management of Fungal Infections in the Pediatric Intensive Care Unit 265
and it is to start empiric echinocandin or liposo- 4. Zaoutis TE, et al. The epidemiology and attributable
outcomes of candidemia in adults and children hos-
mal amphotericin B when these high-risk patients pitalized in the United States: a propensity analysis.
remain febrile and neutropenic past 96 h into Clin Infect Dis. 2005;41(9):1232–9.
their antibiotic course [34]. 5. International Pediatric Fungal Network. Available
It is the subset of patients without underlying from: http://www.ipfn.org/.
6. Brissaud O, et al. Invasive fungal disease in PICU:
malignancy but with significant risk factors that epidemiology and risk factors. Ann Intensive Care.
pose the greater diagnostic challenge. Given the 2012;2(1):6.
available evidence to date, the authors recom- 7. Steinbach WJ, et al. Results from a prospective,
mend restricting empiric antifungal therapy to international, epidemiologic study of invasive can-
didiasis in children and neonates. Pediatr Infect
those patients with known risk factors (i.e., Dis J. 2012;31(12):1252–7.
broad-spectrum antibiotics, prolonged neutrope- 8. Singhi S, Rao DS, Chakrabarti A. Candida coloniza-
nia, presence of indwelling central line) with tion and candidemia in a pediatric intensive care unit.
fever and SIRS/sepsis who fail to respond to Pediatr Crit Care Med. 2008;9(1):91–5.
9. Zaoutis T. Candidemia in children. Curr Med Res
broad-spectrum antibiotic therapy within 72 h. Opin. 2010;26(7):1761–8.
Given the high risk of Candida infection, an echi- 10. Vogiatzi L, et al. Invasive candidiasis in pediatric
nocandin would be the treatment of choice. intensive care in Greece: a nationwide study. Intensive
Care Med. 2013;39(12):2188–95.
11. Zaoutis TE, et al. Risk factors and predictors for
candidemia in pediatric intensive care unit patients:
Take-Home Points implications for prevention. Clin Infect Dis.
• Despite significant advances in diagnos- 2010;51(5):e38–45.
tic and therapeutic options, invasive fun- 12. Jordan I, et al. Per-species risk factors and predic-
tors of invasive Candida infections in patients admit-
gal disease remains a major contributor ted to pediatric intensive care units: development
to morbidity and mortality, particularly of ERICAP scoring systems. Pediatr Infect Dis J.
in PICU patients. 2014;33(8):e187–93.
• The diagnosis of invasive fungal disease 13. Hegazi M, et al. Characteristics and risk factors of
candidemia in pediatric intensive care unit of a ter-
requires a high index of suspicion, and tiary care children’s hospital in Egypt. J Infect Dev
testing should be limited to those Ctries. 2014;8(5):624–34.
patients with known risk factors. 14. Arslankoylu AE, et al. Symptomatic and asymptom-
Biomarkers, while promising, have yet atic candidiasis in a pediatric intensive care unit. Ital J
Pediatr. 2011;37:56.
to be widely validated in the pediatric 15. Wattier RL, et al. A prospective, international cohort
population. study of invasive mold infections in children. J
• Clinicians should be mindful of phar- Pediatric Infect Dis Soc. 2015;4(4):313–22.
macokinetic variation and dosing vari- 16. Burgos A, et al. Pediatric invasive aspergillosis: a
multicenter retrospective analysis of 139 contem-
ability when implementing treatment in porary cases. Pediatrics. 2008;121(5):e1286–94.
the pediatric population. 17. Zaoutis TE, et al. Epidemiology, outcomes, and
costs of invasive aspergillosis in immunocompro-
mised children in the United States, 2000. Pediatrics.
2006;117(4):e711–6.
18. Groll AH, et al. Five-year-survey of invasive asper-
References gillosis in a paediatric cancer centre. Epidemiology,
management and long-term survival. Mycoses.
1. Wisplinghoff H, et al. Nosocomial bloodstream infec- 1999;42(7–8):431–42.
tions in pediatric patients in United States hospitals: 19. Steinbach WJ. Invasive aspergillosis in pediatric
epidemiology, clinical features and susceptibilities. patients. Curr Med Res Opin. 2010;26(7):1779–87.
Pediatr Infect Dis J. 2003;22(8):686–91. 20. Francis JR, et al. Mucormycosis in children: review
2. Wilson LS, et al. The direct cost and incidence and recommendations for management. J Pediatric
of systemic fungal infections. Value Health. Infect Dis Soc. 2017;
2002;5(1):26–34. 21. Prasad PA, Vaughan AM, Zaoutis TE. Trends in zygo-
3. Pana ZD, et al. Epidemiology of invasive fun- mycosis in children. Mycoses. 2012;55(4):352–6.
gal disease in children. J Pediatric Infect Dis Soc. 22. Ascioglu S, et al. Defining opportunistic invasive
2017;6(suppl_1):S3–S11. fungal infections in immunocompromised patients
ERRNVPHGLFRVRUJ
266 C. L. Joyce et al.
with cancer and hematopoietic stem cell trans- 34. Lehrnbecher T, et al. Guideline for the management
plants: an international consensus. Clin Infect Dis. of fever and neutropenia in children with cancer and
2002;34(1):7–14. hematopoietic stem-cell transplantation recipients:
23. De Pauw B, et al. Revised definitions of invasive 2017 update. J Clin Oncol. 2017;35(18):2082–94.
fungal disease from the European Organization for 35. Desai R, Ross LA, Hoffman JA. The role of bron-
Research and Treatment of Cancer/invasive fun- choalveolar lavage galactomannan in the diagnosis of
gal infections cooperative group and the National pediatric invasive aspergillosis. Pediatr Infect Dis J.
Institute of Allergy and Infectious Diseases mycoses 2009;28(4):283–6.
study group (EORTC/MSG) consensus group. Clin 36. Bergeron A, et al. Prospective evaluation of clini-
Infect Dis. 2008;46(12):1813–21. cal and biological markers to predict the outcome of
24. Huppler AR, et al. Role of molecular biomarkers in invasive pulmonary aspergillosis in hematological
the diagnosis of invasive fungal diseases in children. patients. J Clin Microbiol. 2012;50(3):823–30.
J Pediatric Infect Dis Soc. 2017;6(suppl_1):S32–44. 37. Smith PB, et al. Quantification of 1,3-beta-D-glucan
25. Pfeiffer CD, Fine JP, Safdar N. Diagnosis of invasive levels in children: preliminary data for diagnostic use
aspergillosis using a galactomannan assay: a meta- of the beta-glucan assay in a pediatric setting. Clin
analysis. Clin Infect Dis. 2006;42(10):1417–27. Vaccine Immunol. 2007;14(7):924–5.
26. Zheng F, et al. Diagnostic values and limitations 38. Salvatore, C.M., Petraitiene R, Sitaras L, et al.
of (1,3)-beta-D-glucans and galactomannan assays Prospective study and analytical performance of
for invasive fungal infection in patients admitted serum (1-->3)-B-D-glucan in pediatric patients.
to pediatric intensive care unit. Mycopathologia. In: Program and Abstracts of ID Week. New
2017;182(3–4):331–8. Orleans, LA.
27. Fisher BT, et al. Failure to validate a multivariable 39. Wheat LJ. Approach to the diagnosis of invasive
clinical prediction model to identify pediatric inten- aspergillosis and candidiasis. Clin Chest Med.
sive care unit patients at high risk for Candidemia. J 2009;30(2):367–77. viii
Pediatric Infect Dis Soc. 2016;5(4):458–61. 40. Salvatore CM, et al. (1-->3)-beta-d-glucan in cere-
28. Falcone EL, Holland SM. Invasive fungal infec-
brospinal fluid as a biomarker for Candida and
tion in chronic granulomatous disease: insights into Aspergillus infections of the central nervous sys-
pathogenesis and management. Curr Opin Infect Dis. tem in pediatric patients. J Pediatric Infect Dis Soc.
2012;25(6):658–69. 2016;5(3):277–86.
29. Patterson TF, et al. Practice guidelines for the diag- 41. Taira CL, et al. A multiplex nested PCR for the detec-
nosis and management of aspergillosis: 2016 update tion and identification of Candida species in blood
by the Infectious Diseases Society of America. Clin samples of critically ill paediatric patients. BMC
Infect Dis. 2016;63(4):e1–e60. Infect Dis. 2014;14:406.
30. Lehrnbecher T, et al. Galactomannan, beta-D-glucan, 42. Gupta P, et al. Comparative evaluation of pan-
and polymerase chain reaction-based assays for the fungal real-time PCR, galactomannan and
diagnosis of invasive fungal disease in pediatric can- (1-3)-beta-D-glucan assay for invasive fungal
cer and hematopoietic stem cell transplantation: a infection in paediatric cancer patients. Mycoses.
systematic review and meta-analysis. Clin Infect Dis. 2017;60(4):234–40.
2016;63(10):1340–8. 43. Rao DS, et al. Mannan antigen detection in the diag-
31. Sulahian A, et al. Value of antigen detection using an nosis of patients with invasive candidiasis. Indian J
enzyme immunoassay in the diagnosis and prediction Med Res. 2002;116:13–20.
of invasive aspergillosis in two adult and pediatric 44. Katragkou A, et al. Diagnostic imaging and invasive
hematology units during a 4-year prospective study. fungal diseases in children. J Pediatric Infect Dis Soc.
Cancer. 2001;91(2):311–8. 2017;6(suppl_1):S22–31.
32. Siemann M, Koch-Dorfler M, Gaude M. False-
45. Lestner JM, et al. Antifungal agents and therapy for
positive results in premature infants with the Platelia infants and children with invasive fungal infections:
aspergillus sandwich enzyme-linked immunosorbent a pharmacological perspective. Br J Clin Pharmacol.
assay. Mycoses. 1998;41(9–10):373–7. 2013;75(6):1381–95.
33. Groll AH, et al. Fourth European conference on infec- 46. Cecinati V, et al. Antifungal therapy in children: an
tions in Leukaemia (ECIL-4): guidelines for diag- update. Eur J Pediatr. 2013;172(4):437–46.
nosis, prevention, and treatment of invasive fungal 47. Filioti J, Spiroglou K, Roilides E. Invasive candidiasis
diseases in paediatric patients with cancer or alloge- in pediatric intensive care patients: epidemiology, risk
neic haemopoietic stem-cell transplantation. Lancet factors, management, and outcome. Intensive Care
Oncol. 2014;15(8):e327–40. Med. 2007;33(7):1272–83.
ERRNVPHGLFRVRUJ
15 Diagnosis and Management of Fungal Infections in the Pediatric Intensive Care Unit 267
ERRNVPHGLFRVRUJ
Part VII
Endocrinologic Controversies
ERRNVPHGLFRVRUJ
Corticosteroid Therapy for Septic
Shock and Pediatric ARDS 16
Lauren Jacobs, Hector Wong, and Kusum Menon
ERRNVPHGLFRVRUJ
272 L. Jacobs et al.
ERRNVPHGLFRVRUJ
16 Corticosteroid Therapy for Septic Shock and Pediatric ARDS 273
oxide synthetase production leading to increased these endotype-defining genes were repressed in
vascular tone [23], stimulation of intercellular endotype A as compared to B. When contrasted
adhesion factor from vascular smooth muscle with endotype B, endotype A patients tended to
with a resulting decrease in capillary leak [23], be younger, had fewer PICU-free days, and had
and increasing the number of β-adrenergic recep- higher Pediatric Risk of Mortality (PRISM)
tors in the heart, thereby increasing myocardial scores, number of organ failures, and mortality
contractility [22]. Therefore, lack of sufficient [27]. After controlling for illness severity, comor-
cortisol at the cellular level may contribute to bidities, and age, odds of mortality and compli-
hemodynamic instability by limiting myocardial cated course (defined as continued failure of two
contractility while encouraging vasodilation and/ or more organs at day 7 of illness) were both 2.7
or capillary leak syndrome [24, 25]. times higher in endotype A patients (95% CI for
There are even further intricacies at the cellu- mortality, 1.2–6.0; 95% CI for complicated
lar level which likely influence a patient’s course, 1.5–4.8) [28]. Furthermore, corticoste-
response to corticosteroids. Endotypes are sub- roid administration was associated with increased
classes of diseases differentiated at the genomic mortality among the endotype A patients [28].
level or by other biological processes. Using Further work supports patient-to-patient vari-
genome-wide expression profiling in septic pedi- ability in GCR gene expression. Quax et al.
atric patients, Wong et al. identified two distinct established that response to corticosteroids
endotypes: A and B [26–28], which are displayed depends, in part, on functional polymorphisms in
as gene expression mosaics in Fig. 16.1. They the GCR gene [29]. Cvijanovich et al. studied
discovered 100 endotype-defining genes, corre- three functional GCR polymorphisms in pediat-
sponding to adaptive immunity and glucocorti- ric septic shock patients, finding that patients
coid receptor (GCR) signaling. The majority of who were homozygous for the wild-type allele
Endotype A
Endotype B
Fig. 16.1 Pediatric septic shock endotypes: The images Examples of three individual endotype A patients are
are gene expression mosaics of the 100 endotype-defining shown in the upper panel, and examples of three individ-
genes corresponding to adaptive immunity and glucocor- ual endotype B patients are shown in the lower panel. See
ticoid receptor signaling. The degree of red intensity cor- text for associations between endotype assignment and
responds to increased gene expression, and the degree of outcome and endotype assignment and response to
blue intensity corresponds to decreased gene expression. corticosteroids
ERRNVPHGLFRVRUJ
274 L. Jacobs et al.
had a higher risk of complicated course if exposed severely ill patients, as evidenced by a high
to corticosteroids, even after adjusting for PRISM median Sequential Organ Failure Assessment
score [30]. In a prospective cohort study in criti- (SOFA) score of 11, 70% mechanical ventilation
cally ill children evaluating the relationship rate, 33% renal replacement therapy rate, median
between GCR expression in peripheral white vasopressor (norepinephrine or equivalent) dose
blood cells and illness severity, patients with car- of 0.5 μg/kg/minute at the time of corticosteroid
diovascular failure, higher PRISM scores, and initiation, and 44% 28-day mortality. Just over
higher degree of organ failures all had signifi- half of the subjects underwent a low-dose cortisol
cantly lower GCR expression in both CD4+ and stimulation test. Within this subpopulation, 81%
CD8+ lymphocytes [31]. Thus, the heterogeneity failed to respond. The median time to corticoste-
of sepsis and the response to corticosteroids start roid administration was 8.5 h, with multivariate
at the genomic level. These intricacies make stan- analysis showing increased lag time to cortico-
dardizing treatment strategies challenging. steroid initiation associated with higher likeli-
hood of mortality.
Most recently, Annane et al. published a fol-
I n Support of Corticosteroids low-up trial evaluating the impact of hydrocorti-
for Sepsis sone plus fludrocortisone therapy on mortality in
over 1200 adults with septic shock [34]. Subjects
The majority of evidence to support corticoste- in the treatment arm had lower mortality at
roid administration stems from the adult litera- 90 days (49% versus 43%, p-value 0.03) and
ture. In 2002, Annane et al. performed a 180 days (53% versus 47%, p-value 0.04), as
placebo-controlled randomized double-blind well as at the time of ICU and hospital discharge.
study involving almost 300 adult patients with Subjects receiving study drugs also had a greater
fluid-refractory septic shock [32]. Enrollment number of vasopressor-free (17 versus 15 days,
occurred within 8 h of diagnosis, excluding p-value <0.001) and organ failure-free days (14
patients receiving etomidate in the prior 6 h. versus 12 days, p-value 0.003) within the first
Subjects were randomized to either the placebo 28 days of enrollment. The latter findings are
arm or the treatment arm, consisting of hydrocor- supported by a recent large randomized trial of
tisone 50 mg every 6 h and fludrocortisone 50 μg 3800 adult subjects [35]. Subjects in the treat-
daily for 7 days. Seventy-seven percent of sub- ment arm exhibited shorter time to resolution
jects had corticosteroid insufficiency, as diag- of shock [median 3 days (IQR, 2–5 days) versus
nosed by a low-dose ACTH stimulation test. 4 days (IQR, 2–9 days); HR, 1.32 (95% CI, 1.23–
Compared to the placebo group, the corticoste- 1.41)] and duration of mechanical ventilation
roid group had a trend toward decreased 28-day [median 6 days (IQR, 3–18 days) versus 7 days
mortality (OR, 0.65; 95% CI, 0.39–1.07) and (IQR, 3–24 days); HR, 1.13 (95% CI,
ICU mortality (OR, 0.61; 95% CI, 0.37–1.02) 1.05–1.22)].
and significantly faster cessation of vasopressor There is a relative paucity of support for corti-
therapy (HR, 1.54; 95% CI, 1.10–2.16). Within costeroid use in pediatric septic shock within the
the treatment arm, hydrocortisone conferred the pediatric literature. A recent meta-analysis evalu-
greatest survival benefit among patients who ated the findings of eight randomized controlled
failed to respond to an ACTH stimulation test. trials (RCT) of corticosteroids in pediatric shock
Additionally, there was no difference in adverse [36]. The analysis revealed differing corticoste-
events between the groups. roid protocols, small sample sizes (n = 22–98),
Park et al. performed a retrospective analysis and a lack of generalizability to tertiary care pop-
of adult patients receiving low-dose corticoste- ulations as many studies were conducted in
roids for septic shock to evaluate the relationship patients with dengue hemorrhagic shock. Two
between survival and time to corticosteroid initi- studies showed statistically significant mortality
ation [33]. The cohort was comprised of 178 reduction with corticosteroid administration, but
ERRNVPHGLFRVRUJ
16 Corticosteroid Therapy for Septic Shock and Pediatric ARDS 275
the pooled meta-analysis results did not support low cortisol levels were present in 47% of all
that finding. No studies to date have defined a subjects and 60% of patients exposed to etomi-
subpopulation of septic shock patients who may date. Regardless of baseline adrenal status, there
benefit from corticosteroid administration. was no survival benefit with corticosteroids. The
Enrichment strategies may better identify corticosteroid group, and especially those with
which children with septic shock are most likely no response to an ACTH stimulation test, did
to benefit from corticosteroids. Enrichment refers have a significantly quicker resolution of shock
to the selection of patients in whom an interven- (cessation of vasopressor requirement); although
tion is more likely to be beneficial, compared to the proportion of patients achieving shock rever-
unselected patients. Although initially promising, sal was no different. Additionally there were
the corticotropin stimulation test, regardless of higher rates of superinfections among subjects in
dose used, does not appear to be a sufficiently the treatment arm, raising the question of safety.
robust enrichment strategy to select patients most The most recent 3800 patient trial [35] also did
likely to benefit from corticosteroids [32, 37]. not demonstrate a survival benefit with hydro-
A recent post hoc analysis combined prognos- cortisone administration (28% mortality treat-
tic and predictive enrichment to identify a sub- ment arm versus 29% placebo) (OR, 0.95; 95%
group of children with septic shock more likely CI, 0.82–1.1). Notably, both of these trials
to benefit from corticosteroids [38]. The pediatric enrolled patients up to 72 h post-diagnosis,
sepsis biomarker risk model (PERSEVERE) was whereas Annane’s studies required enrollment
used for prognostic enrichment by assigning a within 24 h, suggesting that time to initiation of
baseline mortality probability [39–41]. Predictive corticosteroids may play a role in corticosteroid
enrichment was carried out by assigning patients responsiveness.
to endotype A or B, as described above. Pediatric-specific data regarding the role of
Corticosteroids were associated with a more than corticosteroids in septic shock are substantially
tenfold reduction in the rate of complicated less robust. A meta-analysis of randomized con-
course among endotype B subjects with an inter- trolled pediatric studies conducted in developing
mediate to high PERSEVERE-based mortality countries showed no difference in mortality
risk [38]. Endotype B subjects with low between placebo and corticosteroid groups [36].
PERSEVERE-based mortality risk and all endo- Atkinson et al. retrospectively evaluated the rela-
type A subjects, regardless of baseline mortality tionship between corticosteroids and mortality in
risk, received no such benefit [38]. almost 500 pediatric patients with septic shock
[42]. Initial analysis revealed an association
between corticosteroid use and increased mortal-
Against Corticosteroids for Sepsis ity (OR, 2.3; 95% CI, 1.3–4), but patients who
received corticosteroids had higher PRISM
The oft-cited counterargument to the initial find- scores, more organ dysfunction, and increased
ings of Annane et al. [32] is the 2008 work of vasoactive support needs. The authors then
Sprung et al. [37]. This randomized, double- employed prognostic enrichment, stratifying
blind placebo-controlled study included nearly patients by PRISM score and PERSEVERE-
500 adult patients in septic shock. Subjects were derived mortality risk to elucidate which, if any,
eligible for enrollment up to 72 h after diagnosis groups would benefit from corticosteroids.
and then randomized to either placebo or treat- Neither PERSEVERE- nor PRISM-based strati-
ment. Of note, exposure to etomidate was not an fication yielded any evidence of survival benefit
exclusion criterion, and nearly 20% of subjects associated with corticosteroid usage, regardless
had received etomidate prior to enrollment. The of degree of risk. Additionally, an observational
treatment group received hydrocortisone 50 mg study using data from the RESOLVE trial (dou-
every 6 h for 5 days, followed by a taper over an ble-blind placebo-controlled trial evaluating acti-
additional 6 days. Following a stimulation test, vated protein C for pediatric sepsis) found a 41%
ERRNVPHGLFRVRUJ
276 L. Jacobs et al.
ERRNVPHGLFRVRUJ
16 Corticosteroid Therapy for Septic Shock and Pediatric ARDS 277
Table 16.1 Comparison of diagnostic criteria for pediatric and adult ARDS [48–50]
Adult ARDS Pediatric ARDS
Timing Within 7 days of known insult Within 7 days of known insult
Radiographic New bilateral opacities on chest New infiltrate on chest imaging
findings x-ray or CT scan consistent with Consistent with pulmonary parenchymal illness
bilateral pulmonary edema
Oxygenation Mild Moderate Severe Noninvasive Invasive MV Invasive MV Invasive
MV Mild Moderate MV
Severe
PaO2/FiO2 PaO2/FiO2 ≤ 100 CPAP ≥5 4 ≤ OI ≤ 8 8 ≤ OI ≤ 16 OI ≥ 16
>200 > 100 With PaO2/FiO2 5 ≤ OSI ≤ 7.5 7.5 ≤ OSI ≤ 12.3 OSI ≥ 12.3
≤ 300 ≤ 200 PEEP ≤ 300
With PEEP With ≥5 SpO2/
≥5 PEEP ≥5 FiO2 ≤ 24
Exclusion Respiratory failure explained by Perinatal-related lung injury
criteria heart failure or fluid overload
ERRNVPHGLFRVRUJ
278 L. Jacobs et al.
ERRNVPHGLFRVRUJ
16 Corticosteroid Therapy for Septic Shock and Pediatric ARDS 279
ERRNVPHGLFRVRUJ
280 L. Jacobs et al.
Against Corticosteroids for ARDS ticosteroid use and mortality, but this relationship
did not persist in multiple regression analysis.
A 2006 prospective double-blind study random- Interestingly, the authors evaluated subgroups of
ized 180 adult patients to either placebo or treat- patients they postulated would have greater ben-
ment, consisting of a 2 mg/kg methylprednisolone efit from corticosteroids: immunocompromised
load followed by a corticosteroid taper of at least patients (with likely prior corticosteroid expo-
21 days [94]. Patients could be enrolled between sure), patients with reactive airway disease or
7 and 28 days from diagnosis of ARDS. Similar chronic lung disease, patients with shock requir-
to other studies, the treatment group had signifi- ing vasoactive medications, and patients with
cantly higher ventilator-free and ICU-free days at multi-organ dysfunction (≥3 non-pulmonary
day 28 and shorter time to extubation. But there organ failures). Corticosteroids provided no sur-
were two novel observations of concern. First, vival benefit in any of these groups. Of note, this
patients in the corticosteroid arm were signifi- study differs from the randomized adult trials in
cantly more likely to require re-intubation. that more patients received hydrocortisone than
Second, there was no survival benefit seen with methylprednisolone, and some patients were pre-
corticosteroids, and for those subjects enrolled scribed dexamethasone or a combination of more
beyond day fourteen of ARDS onset, there was a than one corticosteroid.
significantly higher mortality rate among the Drago et al. published a pediatric randomized
treatment group. These results – a lack of trend placebo-controlled pilot study in 2015 that mir-
toward survival benefit, and possibly even harm- rors prior adult studies [97]. Thirty-five patients
ful effects if instituted later in the clinical course – were randomized into the placebo or treatment
are in opposition to other studies. arm within 72 h of PARDS diagnosis. The thera-
A retrospective study analyzing a large peutic course consisted of a 2 mg/kg methylpred-
(n = 607) cohort of adults with ARDS related to nisolone load followed by a tapered continuous
H1N1 conferred similar findings [95]. Forty-six infusion over an additional 14 days. Despite
percent of patients received corticosteroids, for a higher illness severity in the placebo group, there
median of 7 days, with initiation at a median of 0 was no difference in the duration of mechanical
days from the onset of critical illness and 1 day ventilation, ventilator-free days, oxygenation
from intubation. Univariate analysis revealed that index, and PICU or hospital length of stay. As
subjects who received corticosteroids had higher this was only a pilot study to evaluate feasibility
hospital mortality and fewer ventilator-free and of future larger studies, the authors did not dis-
ICU-free days at day 28 of illness. After propen- cuss mortality risk, nor did they perform multi-
sity matching, corticosteroid exposure was no variate analyses.
longer independently associated with mortality
risk (OR, 1.52; CI, 0.9–2.58), but they incurred
no survival benefit. Summary of Corticosteroids
Results from pediatric studies are relatively for ARDS
similar. Yehya et al. conducted a prospective
observational study examining outcomes in 283 Due to a paucity of pediatric-specific evidence
pediatric patients with ARDS and either ≤24 h proving corticosteroids are beneficial in PARDS,
of corticosteroid exposure (typically a peri-extu- clinicians are often forced to apply adult data and
bation dose) or >24 h of corticosteroid exposure protocols to children. Despite the disparate
[96]. Both multivariate and propensity-matched results on the utility of corticosteroids in even
analyses revealed that longer corticosteroid expo- large adult trials, pediatric intensivists are ready
sure and total dose were independently associ- prescribers, with upward of 60% of patients
ated with fewer ventilator-free days and longer receiving corticosteroids for PARDS [52, 61, 98].
duration of mechanical ventilation. The unad- Some would argue the benefit of corticoste-
justed analysis found an association between cor- roids hinges on the timing of initiation. Early ini-
ERRNVPHGLFRVRUJ
16 Corticosteroid Therapy for Septic Shock and Pediatric ARDS 281
tiation could temper the inflammatory response, benefit from corticosteroids. Secondly, large, ran-
but if fibroproliferative changes have ensued, the domized clinical trials must be conducted in
patient has potentially reached the “point of no pediatric populations, with an emphasis on the
return.” The results of the H1N1 adult study and subgroups with purported greater benefit. Until
the available pediatric studies are antithetical to further proof of both benefit and a lack of harm
that notion, as even early institution of corticoste- exists, pediatric intensivists should practice cau-
roids failed to decrease mortality risk [95–97]. tion when prescribing corticosteroids for sepsis
Others might contend that given the robust het- or ARDS.
erogeneity of this disease process, it is imperative
to identify patients most likely to benefit from Take-Home Bullet Points
corticosteroids. Even though Yehya et al. [96]
showed no improvement in certain higher-risk • Pediatric sepsis and pediatric acute respiratory
subgroups, perhaps the identification and there- distress syndrome (PARDS) are heteroge-
fore stratification of patients must occur at the neous disease processes.
genomic level. Recently, the PERSEVERE bio- • Sepsis is increasingly prevalent with a rela-
markers were adapted to estimate the risk of mor- tively high mortality rate among children in
tality among children with PARDS [99], offering ICUs.
a new approach to better select high-risk patients • PARDS has a high rate of ICU mortality, up to
and target research toward these populations. 70% in immunocompromised patients.
In summary, there is a lack of evidence that • There is a paucity of pediatric research evalu-
corticosteroids improve clinical course or pro- ating the role of corticosteroids in sepsis and
vides survival benefit for pediatric acute respira- ARDS.
tory distress syndrome. Early initiation of • Adult studies on corticosteroid use in both
corticosteroids appears unassociated with an sepsis and ARDS often report discrepant
increased likelihood of adverse events, while late results, and cannot be extrapolated to pediatric
initiation is likely futile and, at the very worst, populations.
harmful. But to suggest that clinicians should • Based on the available literature, there is a
routinely employ corticosteroids for early lack of evidence to support the routine use of
PARDS because it is non-injurious and has theo- corticosteroids for pediatric sepsis or ARDS.
retical benefits overlooks the role that genotypic • Further research is needed to identify sub-
and phenotypic variations play in this disease phenotypes which would benefit from
process – an avenue which researchers are just corticosteroids.
starting to explore. • Large prospective, randomized, double-blind
placebo-controlled pediatric trials are needed
to evaluate the efficacy and safety of
Conclusions and Future Directions corticosteroids.
ERRNVPHGLFRVRUJ
282 L. Jacobs et al.
4. Hartman ME, et al. Trends in the epidemiology of 23. Sasidharan P. Role of corticosteroids in neona-
pediatric severe sepsis*. Pediatr Crit Care Med. tal blood pressure homeostasis. Clin Perinatol.
2013;14(7):686–93. 1998;25(3):723–40. xi
5. Ruth A, et al. Pediatric severe sepsis: current trends 24. Hinshaw LB, et al. Corticosteroid/antibiotic treatment
and outcomes from the pediatric health informa- of adrenalectomized dogs challenged with lethal E.
tion systems database. Pediatr Crit Care Med. coli. Circ Shock. 1985;16(3):265–77.
2014;15(9):828–38. 25. Ledingham IM, Watt I. Influence of sedation on mor-
6. Weiss SL, et al. The epidemiology of hospital death tality in critically ill multiple trauma patients. Lancet.
following pediatric severe Sepsis: when, why, and 1983;1(8336):1270.
how children with Sepsis die. Pediatr Crit Care Med. 26. Wong HR, et al. Identification of pediatric septic
2017;18(9):823–30. shock subclasses based on genome-wide expression
7. Rhodes A, et al. Surviving Sepsis campaign: interna- profiling. BMC Med. 2009;7:34.
tional guidelines for Management of Sepsis and Septic 27. Wong HR, et al. Validation of a gene expression-based
Shock: 2016. Intensive Care Med. 2017;43(3):304–77. subclassification strategy for pediatric septic shock.
8. Angus DC, et al. Epidemiology of severe sepsis Crit Care Med. 2011;39(11):2511–7.
in the United States: analysis of incidence, out- 28. Wong HR, et al. Developing a clinically feasible per-
come, and associated costs of care. Crit Care Med. sonalized medicine approach to pediatric septic shock.
2001;29(7):1303–10. Am J Respir Crit Care Med. 2015;191(3):309–15.
9. Oberholzer A, Oberholzer C, Moldawer LL. Sepsis 29. Quax RA, et al. Glucocorticoid sensitivity in health
syndromes: understanding the role of innate and and disease. Nat Rev Endocrinol. 2013;9(11):670–86.
acquired immunity. Shock. 2001;16(2):83–96. 30. Cvijanovich NZ, et al. Glucocorticoid receptor poly-
10. Aziz M, et al. Current trends in inflammatory and morphisms and outcomes in pediatric septic shock.
immunomodulatory mediators in sepsis. J Leukoc Pediatr Crit Care Med. 2017;18(4):299–303.
Biol. 2013;93(3):329–42. 31. Shibata AR, Troster EJ, Wong HR. Glucocorticoid
11. Rittirsch D, Flierl MA, Ward PA. Harmful molecu- receptor expression in peripheral WBCs of critically ill
lar mechanisms in sepsis. Nat Rev Immunol. children. Pediatr Crit Care Med. 2015;16(5):e132–40.
2008;8(10):776–87. 32. Annane D, et al. Effect of treatment with low
12. Opal SM, DePalo VA. Anti-inflammatory cytokines. doses of hydrocortisone and fludrocortisone on
Chest. 2000;117(4):1162–72. mortality in patients with septic shock. JAMA.
13. Hoesel LM, et al. Harmful and protective roles of neu- 2002;288(7):862–71.
trophils in sepsis. Shock. 2005;24(1):40–7. 33. Park HY, et al. Early initiation of low-dose cortico-
14. Hotchkiss RS, Karl IE. The pathophysiology and treat- steroid therapy in the management of septic shock:
ment of sepsis. N Engl J Med. 2003;348(2):138–50. a retrospective observational study. Crit Care.
15. Rose SR, et al. Diagnosis of ACTH deficiency.
2012;16(1):R3.
Comparison of overnight metyrapone test to either 34. Annane D, et al. Hydrocortisone plus fludrocorti-
low-dose or high-dose ACTH test. Horm Res. sone for adults with septic shock. N Engl J Med.
1999;52(2):73–9. 2018;378(9):809–18.
16. Bone M, et al. Assessment of adrenal function in the 35. Venkatesh B, et al. Adjunctive glucocorticoid ther-
initial phase of meningococcal disease. Pediatrics. apy in patients with septic shock. N Engl J Med.
2002;110(3):563–9. 2018;378(9): 797–808.
17. Menon K, et al. A prospective multicenter study of 36. Menon K, et al. A systematic review and meta-
adrenal function in critically ill children. Am J Respir analysis on the effect of steroids in pediatric shock.
Crit Care Med. 2010;182(2):246–51. Pediatr Crit Care Med. 2013;14(5):474–80.
18. Annane D, et al. Guidelines for the diagnosis and 37. Sprung CL, et al. Hydrocortisone therapy for
management of critical illness-related corticosteroid patients with septic shock. N Engl J Med.
insufficiency (CIRCI) in critically ill patients (part 2008;358(2):111–24.
I): Society of Critical Care Medicine (SCCM) and 38. Wong HR, et al. Combining prognostic and predictive
European Society of Intensive Care Medicine (ESICM) enrichment strategies to identify children with septic
2017. Intensive Care Med. 2017;43(12):1751–63. shock responsive to corticosteroids. Crit Care Med.
19. Sarthi M, et al. Adrenal status in children with septic 2016;44(10):e1000–3.
shock using low-dose stimulation test. Pediatr Crit 39. Wong HR, et al. The pediatric sepsis biomarker risk
Care Med. 2007;8(1):23–8. model. Crit Care. 2012;16(5):R174.
20. Wehling M. Specific, nongenomic actions of steroid 40. Wong HR, et al. Testing the prognostic accuracy of
hormones. Annu Rev Physiol. 1997;59:365–93. the updated pediatric sepsis biomarker risk model.
21. Seri I, Evans J. Controversies in the diagnosis and PLoS One. 2014;9(1):e86242.
management of hypotension in the newborn infant. 41. Wong HR, et al. Pediatric Sepsis biomarker risk
Curr Opin Pediatr. 2001;13(2):116–23. model-II: redefining the pediatric Sepsis biomarker
22. Munck A, et al. Glucocorticoid receptors and actions. risk model with septic shock phenotype. Crit Care
Am Rev Respir Dis. 1990;141(2 Pt 2):S2–10. Med. 2016;44(11):2010–7.
ERRNVPHGLFRVRUJ
16 Corticosteroid Therapy for Septic Shock and Pediatric ARDS 283
42. Atkinson SJ, et al. Corticosteroids and pediatric septic 59. Li Y, et al. Epidemiological features and risk factor
shock outcomes: a risk stratified analysis. PLoS One. analysis of children with acute lung injury. World J
2014;9(11):e112702. Pediatr. 2012;8(1):43–6.
43. Zimmerman JJ, Williams MD. Adjunctive corticoste- 60. Dahlem P, et al. Incidence and short-term outcome of
roid therapy in pediatric severe sepsis: observations acute lung injury in mechanically ventilated children.
from the RESOLVE study. Pediatr Crit Care Med. Eur Respir J. 2003;22(6):980–5.
2011;12(1):2–8. 61. Yehya N, Servaes S, Thomas NJ. Characterizing
44. Wong HR, et al. Corticosteroids are associated with degree of lung injury in pediatric acute respiratory dis-
repression of adaptive immunity gene programs in tress syndrome. Crit Care Med. 2015;43(5):937–46.
pediatric septic shock. Am J Respir Crit Care Med. 62. De Luca D, et al. The use of the Berlin defini-
2014;189(8):940–6. tion for acute respiratory distress syndrome dur-
45. Markovitz BP, et al. A retrospective cohort study of ing infancy and early childhood: multicenter
prognostic factors associated with outcome in pediat- evaluation and expert consensus. Intensive Care Med.
ric severe sepsis: what is the role of steroids? Pediatr 2013;39(12):2083–91.
Crit Care Med. 2005;6(3):270–4. 63. Khemani RG, et al. Pulse oximetry vs. PaO2 metrics
46. Nichols B, et al. Hydrocortisone therapy in catechol- in mechanically ventilated children: Berlin definition
amine-resistant pediatric septic shock: a pragmatic of ARDS and mortality risk. Intensive Care Med.
analysis of clinician practice and association with out- 2015;41(1):94–102.
comes. Pediatr Crit Care Med. 2017;18(9):e406–14. 64. Rettig JS, et al. High-frequency oscillatory ven-
47. Menon K, et al. A randomized controlled trial of corti- tilation in pediatric acute lung injury: a multi-
costeroids in pediatric septic shock: a pilot feasibility center international experience. Crit Care Med.
study. Pediatr Crit Care Med. 2017;18(6):505–12. 2015;43(12):2660–7.
48. Force ADT, et al. Acute respiratory distress syndrome: 65. Yehya N, Thomas NJ. Relevant outcomes in pediat-
the Berlin definition. JAMA. 2012;307(23):2526–33. ric acute respiratory distress syndrome studies. Front
49. Khemani RG, et al. Pediatric acute respiratory distress Pediatr. 2016;4:51.
syndrome: definition, incidence, and epidemiology: 66. DeBruin W, et al. Acute hypoxemic respiratory failure
proceedings from the pediatric acute lung injury con- in infants and children: clinical and pathologic char-
sensus conference. Pediatr Crit Care Med. 2015;16(5 acteristics. Crit Care Med. 1992;20(9):1223–34.
Suppl 1):S23–40. 67. Rowan CM, et al. Invasive mechanical ventilation and
50. Pediatric Acute Lung Injury Consensus Conference mortality in pediatric hematopoietic stem cell trans-
G. Pediatric acute respiratory distress syndrome: con- plantation: a multicenter study. Pediatr Crit Care Med.
sensus recommendations from the Pediatric Acute 2016;17(4):294–302.
Lung Injury Consensus Conference. Pediatr Crit Care 68. Khemani RG, et al. Effect of tidal volume in children
Med. 2015;16(5):428–39. with acute hypoxemic respiratory failure. Intensive
51. Quasney MW, et al. The outcomes of children with Care Med. 2009;35(8):1428–37.
pediatric acute respiratory distress syndrome: pro- 69. Costil J, et al. Acute respiratory distress syndrome
ceedings from the pediatric acute lung injury con- (ARDS) in children: multicenter collaborative study
sensus conference. Pediatr Crit Care Med. 2015;16(5 of the French Group of Pediatric Intensive Care.
Suppl 1):S118–31. Pediatr Pulmonol Suppl. 1995;11:106–7.
52. Erickson S, et al. Acute lung injury in pediatric inten- 70. Albuali WH, et al. Have changes in ventilation prac-
sive care in Australia and New Zealand: a prospective, tice improved outcome in children with acute lung
multicenter, observational study. Pediatr Crit Care injury? Pediatr Crit Care Med. 2007;8(4):324–30.
Med. 2007;8(4):317–23. 71. Peters MJ, et al. Acute hypoxemic respiratory
53. Kneyber MC, et al. Acute respiratory distress syn- failure in children: case mix and the utility of
drome: is it underrecognized in the pediatric intensive respiratory severity indices. Intensive Care Med.
care unit? Intensive Care Med. 2008;34(4):751–4. 1998;24(7):699–705.
54. Lopez-Fernandez Y, et al. Pediatric acute lung injury 72. Hough CL. Steroids for acute respiratory distress syn-
epidemiology and natural history study: incidence and drome? Clin Chest Med. 2014;35(4):781–95.
outcome of the acute respiratory distress syndrome in 73. Ware LB, et al. Prognostic and pathogenetic value of
children. Crit Care Med. 2012;40(12):3238–45. combining clinical and biochemical indices in patients
55. Zimmerman JJ, et al. Incidence and outcomes of pedi- with acute lung injury. Chest. 2010;137(2):288–96.
atric acute lung injury. Pediatrics. 2009;124(1):87–95. 74. Barnett N, Ware LB. Biomarkers in acute lung
56. Flori HR, et al. Pediatric acute lung injury: prospective injury—marking forward progress. Crit Care Clin.
evaluation of risk factors associated with mortality. 2011;27(3):661–83.
Am J Respir Crit Care Med. 2005;171(9):995–1001. 75. Katzenstein AL, Bloor CM, Leibow AA. Diffuse alve-
57. Matthay MA, Ware LB, Zimmerman GA. The
olar damage—the role of oxygen, shock, and related
acute respiratory distress syndrome. J Clin Invest. factors. A review. Am J Pathol. 1976;85(1):209–28.
2012;122(8):2731–40. 76. Clark JG, et al. Type III procollagen peptide in the
58. Ware LB, Matthay MA. The acute respiratory distress adult respiratory distress syndrome. Association of
syndrome. N Engl J Med. 2000;342(18):1334–49. increased peptide levels in bronchoalveolar lavage
ERRNVPHGLFRVRUJ
284 L. Jacobs et al.
fluid with increased risk for death. Ann Intern Med. 88. Meduri GU, et al. Plasma and BAL cytokine
1995;122(1):17–23. response to corticosteroid rescue treatment in late
77. Thompson BT, Chambers RC, Liu KD. Acute
ARDS. Chest. 1995;108(5):1315–25.
respiratory distress syndrome. N Engl J Med. 89. Hooper RG, Kearl RA. Established adult respiratory
2017;377(6):562–72. distress syndrome successfully treated with cortico-
78. Smith LS, Zimmerman JJ, Martin TR. Mechanisms steroids. South Med J. 1996;89(4):359–64.
of acute respiratory distress syndrome in children and 90. Meduri GU, et al. Effect of prolonged methylpred-
adults: a review and suggestions for future research. nisolone therapy in unresolving acute respiratory
Pediatr Crit Care Med. 2013;14(6):631–43. distress syndrome: a randomized controlled trial.
79. Rubenfeld GD, et al. Incidence and outcomes of acute JAMA 1998;280(2):159–65.
lung injury. N Engl J Med. 2005;353(16):1685–93. 91. Meduri GU, et al. Methylprednisolone infusion in
80. Brun-Buisson C, et al. Epidemiology and outcome early severe ARDS: results of a randomized con-
of acute lung injury in European intensive care units. trolled trial. Chest. 2007;131(4):954–63.
Results from the ALIVE study. Intensive Care Med. 92. Meduri GU, et al. Prolonged glucocorticoid treatment
2004;30(1):51–61. is associated with improved ARDS outcomes: analy-
81. Meduri GU, et al. Nuclear factor-kappaB- and gluco- sis of individual patients’ data from four randomized
corticoid receptor alpha- mediated mechanisms in the trials and trial-level meta-analysis of the updated lit-
regulation of systemic and pulmonary inflammation erature. Intensive Care Med. 2016;42(5):829–40.
during sepsis and acute respiratory distress syndrome. 93. Kimura D, et al. Plasma biomarker analysis in pediat-
Evidence for inflammation-induced target tissue resis- ric ARDS: generating future framework from a pilot
tance to glucocorticoids. Neuroimmunomodulation. randomized control trial of methylprednisolone: a
2005;12(6):321–38. framework for identifying plasma biomarkers related
82. Bhargava M, et al. Bronchoalveolar lavage fluid pro- to clinical outcomes in pediatric ARDS. Front Pediatr.
tein expression in acute respiratory distress syndrome 2016;4:31.
provides insights into pathways activated in subjects 94. Steinberg KP, et al. Efficacy and safety of cortico-
with different outcomes. Sci Rep. 2017;7(1):7464. steroids for persistent acute respiratory distress syn-
83. Rogers AJ, et al. Profiling of ARDS pulmonary edema drome. N Engl J Med. 2006;354(16):1671–84.
fluid identifies a metabolically distinct subset. Am J 95. Delaney JW, et al. The influence of corticosteroid treat-
Physiol Lung Cell Mol Physiol. 2017;312(5):L703–9. ment on the outcome of influenza A(H1N1pdm09)-
84. Calfee CS, et al. Subphenotypes in acute respiratory related critical illness. Crit Care. 2016;20:75.
distress syndrome: latent class analysis of data from 96. Yehya N, et al. Corticosteroid exposure in pediat-
two randomised controlled trials. Lancet Respir Med. ric acute respiratory distress syndrome. Intensive
2014;2(8):611–20. Care Med. 2015;41(9):1658–66.
85. Calfee CS, et al. Distinct molecular phenotypes of 97. Drago BB, et al. Double-blind, placebo-controlled
direct vs indirect ARDS in single-center and multi- pilot randomized trial of methylprednisolone infu-
center studies. Chest. 2015;147(6):1539–48. sion in pediatric acute respiratory distress syndrome.
86. Meduri GU, et al. Fibroproliferative phase of
Pediatr Crit Care Med. 2015;16(3):e74–81.
ARDS. Clinical findings and effects of corticoste- 98. Santschi M, et al. Acute lung injury in children: thera-
roids. Chest. 1991;100(4):943–52. peutic practice and feasibility of international clinical
87. Meduri GU, et al. Corticosteroid rescue treatment of trials. Pediatr Crit Care Med. 2010;11(6):681–9.
progressive fibroproliferation in late ARDS. Patterns 99. Yehya N, Wong HR. Adaptation of a biomarker-based
of response and predictors of outcome. Chest. Sepsis mortality risk stratification tool for pediatric
1994;105(5):1516–27. acute respiratory distress syndrome. Crit Care Med.
2018;46(1):e9–e16.
ERRNVPHGLFRVRUJ
Management of Diabetic
Ketoacidosis 17
Laura Kitzmiller, Courtney Frye, and Jeff Clark
ERRNVPHGLFRVRUJ
286 L. Kitzmiller et al.
admissions each year [6]. Patients with both ney). This diuresis causes both dehydration and
type 1 and type 2 diabetes are at risk for devel- electrolyte abnormalities. Counter-regulatory
oping DKA [7]; and an estimated 15–70% of hormones including glucagon, catecholamines,
patients with diabetes will develop DKA at and cortisol are produced, which can worsen aci-
some time in their life [8]. dosis and dehydration. These hormones contrib-
Risk factors for developing DKA include ute to the metabolic dysfunction by increasing
those with newly diagnosed diabetes, very young hyperglycemia by augmenting hepatic glucose
children, lower socioeconomic background, poor production and decreasing peripheral glucose
compliance, and adolescent girls [8]. Timely and uptake from the serum. The catecholamines also
appropriate diagnosis and management of DKA promote lipolysis and the production of ketones
is critical, as each episode can be life-threatening. by the activation of hepatic beta-oxidation of free
The highest mortality occurs in association with fatty acids to form ketone bodies. This accumula-
cerebral edema (CE), which occurs in only about tion of ketones leads to the anion gap metabolic
1% of DKA episodes but can account for 50–60% acidosis seen in DKA [8].
of diabetes-related deaths in children [5]. In addi- There are several significant electrolyte abnor-
tion, life-threatening metabolic and electrolyte malities that occur in DKA. Patients with DKA
abnormalities can occur in DKA. However, little present with a total body deficiency of both
data exist as to optimum management of these potassium and phosphate as both intracellular
complications, and significant controversy exists ions are transferred from inside the cell into the
as to the exact mechanisms and ideal manage- serum in exchange for a hydrogen ion; the
ment of each. In part due to this lack of data, it is exchanged potassium and phosphate are then
important to have a strong working knowledge on excreted in the urine. It is therefore recommended
the pathophysiology and management of DKA. to replace both of these electrolytes during treat-
Her bedside blood glucose test is above the ment for DKA [9].
upper limit of detection. She has a capillary Hyperglycemia increases serum osmolarity
blood gas that shows the following: pH 6.9, PCO2 causing flux of water into the extracellular space
13 torr, and bicarbonate 7 mEq/L. Her urinalysis and leading to dilutional hyponatremia [10].
shows a specific gravity of greater than 1.030, Calculation of undiluted sodium concentrations
ketones of “high,” and glucose of “high.” Her can help in the estimation of dehydration and
serum glucose is found to be 797 mg/dL. metabolic dysfunction. The relationship that
sodium is lowered by 1.6 mEq/L for each 100 mg/
dL rise in glucose greater than 100 mg/dL helps
Pathophysiology and Clinical estimate undiluted serum sodium concentrations
Manifestations that should result once hyperglycemia resolves
[11]. If a normal or high sodium is observed, this
DKA is defined as a blood glucose concentration is suggestive of more severe dehydration [8].
greater than 200 mg/dL (11 mmol/L), ketonuria/ Sodium replacement via intravenous fluids
ketonemia, and acidosis with a pH less than 7.3 should be initiated at the onset of DKA therapy.
[2, 8]. The primary abnormality is caused by a The serum sodium should increase as the serum
relative or absolute insulin deficiency. glucose decreases. Of note, if the serum sodium
Hyperglycemia results from this as well as levels do not increase with therapy, this may be a
increased gluconeogenesis, glycogenolysis, and potential sign of imminent cerebral edema [9].
poor peripheral glucose utilization. This lack of The clinical presentation of diabetes and DKA
adequate intracellular glucose leads to ineffective is well-described; symptoms include polyuria,
energy production and cellular metabolic dys- polydipsia, and weight loss. DKA often presents
function. An osmotic diuresis ensues as serum with a history of these symptoms (in a child who
glucose levels exceed the renal threshold for glu- has not been previously diagnosed with diabetes)
cose reabsorption (180 mg/dL in the healthy kid- in addition to symptoms of abdominal pain,
ERRNVPHGLFRVRUJ
17 Management of Diabetic Ketoacidosis 287
Table 17.1 Risk factors for the development of cerebral Table 17.2 Initial lab values for clinical scenario
edema in diabetic ketoacidosis
Serum sodium 132 mEq/L
New-onset diabetes mellitus Serum potassium 6.1 mEq/L
Age younger than 5 years Serum chloride 101 mEq/L
Lower pH at presentation Serum bicarbonate 8 mEq/L
Lower bicarbonate concentration at presentation Blood urea nitrogen 22 mg/dL
Higher blood urea concentration at presentation Plasma creatinine 0.8 mg/dL
Serum calcium 10.6 mg/dL
Serum magnesium 1.5 mg/dL
dehydration, hyperventilation, and altered mental Serum phosphorus 5.1 mg/dL
status. Physical exam findings often include Β-Hydroxybutyrate 4 mmol/L
tachycardia, poor capillary refill, dry mucus
membranes, poor skin turgor, tachypnea with In addition to cerebral edema, additional treat-
Kussmaul breathing, and altered mental status. able causes of encephalopathy need to be evalu-
Cerebral edema in the setting of pediatric ated. Stores of water-soluble vitamins can be
DKA is a rare but life-threatening complication, quickly depleted in the setting of anorexia and
and risk factors associated with development of osmotic diuresis such as occurs with
edema are listed in Table 17.1. Although its inci- DKA. Thiamine deficiency has been described in
dence is about 1%, it is responsible for the the setting of diabetes and DKA in children and
majority of deaths and poor neurologic outcomes can present with acute encephalopathy [15, 16].
of patients with DKA [4]. Signs and symptoms Thiamine is a critical cofactor in carbohydrate
include headache, worsening mental status, metabolism. The mechanism for encephalopathy
incontinence, relative bradycardia, and focal with acute depletion is not as well understood as
neurologic signs such as pupillary asymmetry. that with chronic thiamine deficiency (e.g.,
Signs of cerebral edema must be identified Wernicke’s encephalopathy) but may be related
quickly, and management initiated to prevent to loss of blood-brain barrier function, cerebral
neurologic injury or death. The mechanism of edema, or decreased in production of excitatory
edema is thought to be multifactorial. Cytotoxic neurotransmitters [17]. Although testing for thia-
or cellular edema may exist, and mechanisms mine deficiency is not routine and blood levels
that lead to this may include metabolic failure of that correlate with clinical symptoms are not well
neurons or blood-brain barrier (BBB) cells, isch- established, supplementation is relatively easy
emia due to underperfusion of the brain from and may prove beneficial when other causes of
dehydration, severe acidosis, severe hyperventi- encephalopathy have been ruled out.
lation with cerebral vasoconstriction, and possi- The remainder of her lab work has returned
bly neuroinflammation [12]. However, some and is shown in Table 17.2. The patient receives
studies suggest that vasogenic edema exists, pos- a bolus of normal saline followed by the initia-
sibly due to loss of cerebrovascular autoregula- tion of an insulin infusion at 0.1 U/kg/h. She is
tion or failure of blood-brain barrier cells [13, placed on IV fluids using 0.45% NS at a rate to
14]. The distinction potentially has clinical sig- include maintenance plus correct for 10% dehy-
nificance in that preventative therapies that allow dration over 24 h. Three hours later, she is still
manipulation of cerebral blood flow and volume tachycardic with HR 140. Her perfusion is
may be beneficial in the setting of one, but not improved but her respiratory effort is unchanged.
the other. As an example, does osmotherapy, She is more somnolent and is difficult to arouse.
which is a common therapy for cerebral edema, She will awaken and answer yes and no appro-
worsen vasogenic edema by increasing circulat- priately but is otherwise unresponsive to voice
ing volume or increasing delivery of osmotic or tactile stimulation. She does withdraw appro-
substances across a permeable blood-brain bar- priately to pain. Her pupils are now 7 mm and
rier? In the setting of DKA, the answer is not reactive. Her bedside glucose testing is now
known. 280 mg/dL.
ERRNVPHGLFRVRUJ
288 L. Kitzmiller et al.
ERRNVPHGLFRVRUJ
17 Management of Diabetic Ketoacidosis 289
neous regimen once corrected. The 2011 Joint same dose of glargine less than 2 h prior to insu-
British Diabetes Societies guidelines for the lin drip termination. The study separately ana-
management of adult DKA concluded that con- lyzed new (50) and pre-existing (99) type 1
tinuation of home basal insulin in the acute period diabetic patients. There were no differences in
is unlikely to be detrimental and could facilitate the rates of hypoglycemia. Two cases of cerebral
easier transition to subcutaneous insulin while edema occurred in the control group, but none in
preventing rebound hyperglycemia and ketogen- the intervention group. A statistically significant
esis [22]. Several adult and pediatric studies eval- increase in hypokalemia occurred in the treat-
uate the use of long-acting insulin analogs in ment group of new-onset diabetics (minimum
acute diabetes management. potassium 3.1 mmol/L) with no adverse out-
A randomized, mixed cohort prospective comes in either group. Interestingly, the treat-
study of 61 adults evaluated the rate of rebound ment group had a longer length of insulin infusion
hyperglycemia after insulin drip discontinuation. and duration of acidosis, but no change in hospi-
A variety of conditions requiring insulin infusion tal LOS [26].
were studied (surgery, type 1 and type 2 diabetes) These studies evaluate long-acting insulin
including treatment of DKA. The intervention analog use in acute diabetes management of both
group received 0.25 units/kg of glargine within pediatric and adult patients focusing mostly on
12 h of initiation of insulin infusion compared to DKA. As a whole, they show no consistent
standard insulin drip therapy. There was signifi- improvement in DKA resolution or
cantly less rebound hyperglycemia in the inter- LOS. Although they do confirm less rebound
vention group (p < 0.001) with no episodes of hyperglycemia after discontinuation of continu-
hypoglycemia [23]. Another prospective, ran- ous insulin therapy, they do not inform optimal
domized trial evaluated 40 adult DKA patients time frame for administration of subcutaneous,
over a 6-month period. The intervention group long-acting insulin. In general, the studies are
received 0.3 units/kg of glargine within 2 h of limited by small sample size and study designs
DKA diagnosis in addition to standard insulin introducing bias indicating that a larger, prospec-
drip therapy. The time to closure of anion gap and tive study may be necessary to thoroughly answer
hospital LOS was less in the intervention group, these questions. There were no adverse outcomes
but not statistically significant. There were no noted except clinically insignificant hypokalemia
differences in incidents of hypoglycemia [24]. in new-onset type 1 diabetic patients reinforcing
In 2007, a retrospective review of 71 pediatric the safety of this intervention.
patients treated for DKA compared standard
DKA management to standard DKA manage-
ment plus the administration of 0.3 units/kg of luid Management and Prevention
F
glargine in the first 6 h of diagnosis. There were of Cerebral Edema
no incidents of hypoglycemia or adverse neuro-
logic outcomes in either group. The intervention The degree to which fluid therapy contributes to
group had statistically significant improvement the development of (CE) in the setting of DKA is
in time to correction of acidosis, length of insulin undetermined [5, 27]. However, therapies that are
infusion, and total IV insulin required. PICU known or suspected to increase CE in other clini-
length of stay (LOS) decreased with a trend cal settings are usually avoided in the treatment
toward shorter hospital LOS [25]. of DKA. Preventing a low or rapidly declining
In 2017, Harrison et al. published a retrospec- serum osmolarity is a mainstay of safe therapy.
tive chart review of 149 episodes of pediatric As two of the primary determinants of serum
DKA in patients greater than 2 years old. The osmolarity, this can be accomplished by control-
intervention group received 0.3 to 0.5 units/kg of ling the rate of fall of serum glucose and sodium
glargine more than 4 h prior to discontinuation of concentration. This is the rationale for adminis-
the insulin drip. The control group received the tering isotonic fluids during the early stages of
ERRNVPHGLFRVRUJ
290 L. Kitzmiller et al.
DKA resuscitation and treatment. As discussed, tonic saline boluses have been advocated [9].
serum sodium should increase during the treat- Mannitol is a typical osmotic agent that increases
ment of DKA due to the fall in serum glucose. serum osmolarity, draws water out of the brain,
Failure of rise or even decline in serum sodium and is typically followed by an osmotic diuresis
during treatment is associated with an increased as mannitol is filtered and excreted in the kidney.
risk for CE formation; however causality is In addition, it may have additional beneficial rhe-
unproven [28, 29]. This, along with our current ologic and regional blood flow effects. Hypertonic
understanding of fluid shifts during serum osmo- saline also increases serum osmolarity and
larity changes, has led to the general recommen- decreases brain water content but may have less
dation that isotonic fluids be administered during osmotic diuretic effect, thus maintaining intra-
the early stages of treatment of severe DKA [9]. vascular volume and cardiac output better than
Whether this decreases the incidence or severity mannitol. This is the rationale for preference of
of edema is unknown. However, considering the hypertonic saline over mannitol in the treatment
morbidity of CE relative to hyperchloremic of CE in the setting of DKA and intravascular
metabolic acidosis, prudence supports the use of volume depletion. However, a retrospective
isotonic fluid until such time as the risk for cere- review of osmolar therapy and outcome demon-
bral edema is past, or additional data regarding strated an association of 3% saline with increased
the type of edema may also help clarify optimum mortality for the treatment of cerebral edema in
fluid management in DKA. DKA [30]. Although a physiologic rationale
Utilization of isotonic fluid during rehydration exists for the use of hypertonic saline over man-
may help promote the rise in sodium and decrease nitol in this setting, the ideal osmotic agent is
the likelihood of edema formation but also unclear. Therefore, suspicion and recognition of
includes the administration of chloride. This cerebral edema and prompt treatment, using
increased chloride load eventually needs to be osmotic agents as necessary, are key. In addition,
eliminated via the kidneys and can result in attention to other factors that can influence CE
hyperchloremic metabolic acidosis. Although its and increase intracranial pressure is vital to opti-
significance is debated, this new- and late-onset mizing outcome, such as maintenance of ade-
acidosis may increase the length of hospitaliza- quate ventilation, temperature control, and
tion and need for additional therapy [9]. possible surgical intervention. Surgical options
The next hour, she is difficult to arouse with can include decompressive craniotomy, intracra-
painful stimuli. Her HR is now 117 bpm, BP is nial pressure monitor placement, and external
138/82 mmHg, RR is 23 per minute, and SpO2 is ventricular drain placement in severe cases.
95% without supplemental oxygen. Her pupils As her mental status has continued to
are 8 mm and less briskly reactive. Her with- decline, a blood gas is obtained that shows a
drawal to pain is now non-specific in all extremi- severe metabolic acidosis with a pH of 6.7 and
ties. Her respiratory pattern now shows less a bicarb of 9 mEq/L. She is not withdrawing to
hyperpnea and is less regular. Her bedside blood pain. Her pupils remain 8 mm and are slug-
glucose test shows 155 mg/dL, and serum labs gishly reactive. Her blood pressure has also
are shown in Table 17.2. decreased to 86/42.
In the setting of suspected or documented edema Acidosis is one of the hallmarks of DKA. This
and increased intracranial pressure (ICP), tradi- acidosis slowly corrects as fluid resuscitation and
tional osmolar therapies are often employed. insulin therapy result in improving perfusion and
Mannitol, which has been used for decades in the cessation of ketogenesis [31]. Because the level
treatment of CE, and, more recently, 3% hyper- of acidosis can be severe, bicarbonate was once a
ERRNVPHGLFRVRUJ
17 Management of Diabetic Ketoacidosis 291
mainstay of therapy and thought to improve out- in a subset of patients. However, a study in adults
comes, reduce mortality, and decrease length of show that administration of bicarbonate to
stay in the hospital. However, the administration patients in DKA with an initial pH less than 7.0
of bicarbonate therapy to treat acidosis in the set- did not decrease time to improvement of acidosis
ting of DKA remains controversial as multiple or decrease hospital length of stay [34], and cur-
studies have failed to show any significant bene- rent recommendations are that sodium bicarbon-
fit, and some have shown an association with ate not be given routinely in pediatric patients
increased risk. A multicenter study sought to with DKA except in the case of extremely severe
define risk factors for the development of CE acidosis and refractory hemodynamic instability
[29]; while studying a patient cohort, it was found [9, 35]. As the data remains unclear and inconclu-
that the only therapeutic intervention that was sive, the only known indication and recommen-
associated with increased risk of the development dation for bicarbonate administration in the
of CE was administration of bicarbonate therapy. clinical setting of DKA is life-threatening
Additional studies showed no difference in the hyperkalemia.
rate of complications or time to metabolic
recovery between patients given bicarbonate and
those who were not [32]; in fact, the data sug- Conclusion
gested that bicarbonate may provide more risk
than benefit as those patients were found to have Diabetes mellitus is the most common endocrine
prolonged hospital stays. disorder among pediatric patients. Patients with
There are several physiologic reasons for why both type 1 and type 2 diabetes are at significant
bicarbonate therapy should not be used. First, risk of developing DKA at least one time in their
increasing serum bicarbonate leads to increased lives. Patients with DKA develop dehydration
carbon dioxide formation, which easily crosses and electrolyte abnormalities and are at risk of
the blood-brain barrier and can lead to paradoxi- developing cerebral edema. Cerebral edema
cal CNS acidosis [8]. Sodium bicarbonate admin- remains the most common cause of morbidity
istration has also been shown to be associated and mortality among those patients with
with hypokalemia due to the correction of acido- DKA. The mainstays of DKA therapy are insulin,
sis and intracellular shift of potassium. In a study fluid resuscitation, and electrolyte replacement.
by Viallon et al. [33], patients with DKA were There remain significant controversies regarding
divided into two groups: those that received the specifics of fluid resuscitation, the prevention
bicarbonate and those that did not. Not only was and treatment of cerebral edema, and the efficacy
there no difference in the time to resolution of of bicarbonate administration.
laboratory parameters, but the patients that
received bicarbonate required more frequent
potassium replacement than those patients who
did not. An additional concern of bicarbonate
References
administration is it may cause impaired tissue 1. Skitch SA, Valani R. Treatment of pediatric diabetic
perfusion and oxygenation due to a leftward shift ketoacidosis in Canada: a review of treatment proto-
of the oxyhemoglobin dissociation curve. cols from Canadian pediatric emergency departments.
Because of these factors, bicarbonate administra- CJEM. 2015;17(6):656–61.
2. Stockwell JA, Preissig CM, Society of Critical Care
tion is no longer recommended on a routine basis; Medicine. Comprehensive critical care: pediatric.
however, the controversy remains as there may Mount Prospect: Society of Critical Care Medicine;
be utility for its use in patients with severe acido- 2012. xvi, 972 pages.
sis (pH < 6.9). It is known that severe acidosis 3. Veverka M, et al. A pediatric diabetic ketoacido-
sis management protocol incorporating a two-bag
affects myocardial function and impairs tissue intravenous fluid system decreases duration of intra-
perfusion; bicarbonate may help attenuate these venous insulin therapy. J Pediatr Pharmacol Ther.
effects by correcting the pH more toward normal 2016;21(6):512–7.
ERRNVPHGLFRVRUJ
292 L. Kitzmiller et al.
4. Barrot A, Huisman TA, Poretti A. Neuroimaging 22. Savage MW, Dhatariya KK, Kilvert A, Rayman
findings in acute pediatric diabetic ketoacidosis. G, Rees JAE, Courtney CH, Hilton L, Dyer PH,
Neuroradiol J. 2016;29(5):317–22. Hamersley MS. Joint British Diabetes Societies
5. Hsia DS, et al. Fluid management in pediatric patients guideline for the management of diabetic ketoacido-
with DKA and rates of suspected clinical cerebral sis. Diabet Med. 2011;28:508–15.
edema. Pediatr Diabetes. 2015;16(5):338–44. 23. Hsia E, Seggelke S, Gibbs J, Hawkins RM,
6. Umpierrez GE, Kitabchi AE. Diabetic ketoacido- Cohlmia E, Rasouli N, Wang C, Kam I, Draznin
sis: risk factors and management strategies. Treat B. Subcutaneous administration of glargine to dia-
Endocrinol. 2003;2(2):95–108. betic patients receiving insulin infusion prevents
7. Fasanmade OA, Odeniyi IA, Ogbera AO. Diabetic rebound hyperglycemia. J Clin Endocrinol Metab.
ketoacidosis: diagnosis and management. Afr J Med 2012;97:3132–7.
Med Sci. 2008;37(2):99–105. 24. Doshi P, Potter A, De Los SD, Banuelos R, Darger
8. Nichols DG, Rogers MC. Rogers’ textbook of pedi- BF, Chathampally Y. Prospective randomized trial
atric intensive care. 4th ed. Philadelphia: Lippincott of insulin glargine in acute management of diabetic
Williams & Wilkins; 2008. xxxi, 1839 p. ketoacidosis in the emergency department: a pilot
9. Wolfsdorf JI, et al. ISPAD Clinical Practice Consensus study. Acad Emerg Med. 2015;22:658–62.
Guidelines 2014. Diabetic ketoacidosis and hypergly- 25. Shankar V, Haque A, Churchwell KB, Russell
cemic hyperosmolar state. Pediatr Diabetes. 2014;(15 W. Insulin glargine supplementation during early
Suppl 20):154–79. management phase of diabetic ketoacidosis in chil-
10. Orlowski JP, Cramer CL, Fiallos MR. Diabetic keto- dren. Intensive Care Med. 2007;33:1173–8.
acidosis in the pediatric ICU. Pediatr Clin N Am. 26. Harrison VS, Rustico S, Palladion AA, Ferrara C,
2008;55(3):577–87, x. Hawkes CP. Glargine co-administration with intrave-
11. Duck SC, Wyatt DT. Factors associated with brain nous insulin in pediatric diabetic ketoacidosis is safe
herniation in the treatment of diabetic ketoacidosis. J and facilitates transition to a subcutaneous regimen.
Pediatr. 1988;113(1 Pt 1):10–4. Pediatr Diabetes. 2017;18:742–8.
12. Glaser N, et al. Brain cell swelling during hypocap- 27.
Long B, Koyfman A. Emergency medicine
nia increases with hyperglycemia or ketosis. Pediatr myths: cerebral edema in pediatric diabetic keto-
Diabetes. 2014;15(7):484–93. acidosis and intravenous fluids. J Emerg Med.
13. Glaser NS, et al. Mechanism of cerebral edema
2017;53(2):212–21.
in children with diabetic ketoacidosis. J Pediatr. 28.
Hale PM, et al. Factors predicting cerebral
2004;145(2):164–71. edema in young children with diabetic ketoacido-
14. Tasker RC, Acerini CL. Cerebral edema in children sis and new onset type I diabetes. Acta Paediatr.
with diabetic ketoacidosis: vasogenic rather than cel- 1997;86(6):626–31.
lular? Pediatr Diabetes. 2014;15(4):261–70. 29. Glaser N, et al. Risk factors for cerebral edema in
15. Clark JA, et al. Acute thiamine deficiency in diabetic children with diabetic ketoacidosis. The Pediatric
ketoacidosis: diagnosis and management. Pediatr Crit Emergency Medicine Collaborative Research
Care Med. 2006;7(6):595–9. Committee of the American Academy of Pediatrics.
16. Rosner EA, et al. Low thiamine levels in children N Engl J Med. 2001;344(4):264–9.
with type 1 diabetes and diabetic ketoacidosis: a pilot 30. Decourcey DD, et al. Increasing use of hypertonic
study. Pediatr Crit Care Med. 2015;16(2):114–8. saline over mannitol in the treatment of symptomatic
17. Butterworth RF. Effects of thiamine deficiency on cerebral edema in pediatric diabetic ketoacidosis: an
brain metabolism: implications for the pathogen- 11-year retrospective analysis of mortality*. Pediatr
esis of the Wernicke-Korsakoff syndrome. Alcohol Crit Care Med. 2013;14(7):694–700.
Alcohol. 1989;24(4):271–9. 31. Dunger DB, et al. European Society for Paediatric
18. Ugale J, et al. Measured degree of dehydration in chil- Endocrinology/Lawson Wilkins Pediatric Endocrine
dren and adolescents with type 1 diabetic ketoacido- Society consensus statement on diabetic keto-
sis. Pediatr Crit Care Med. 2012;13(2):e103–7. acidosis in children and adolescents. Pediatrics.
19. Wolfsdorf JI, Glaser N, Sperling MA. Diabetic
2004;113(2):e133–40.
ketoacidosis in infants, children, and adolescents: 32. Green SM, et al. Failure of adjunctive bicarbonate to
a consensus statement from the American Diabetes improve outcome in severe pediatric diabetic ketoaci-
Association. Diabetes Care. 2006;29:1150–9. dosis. Ann Emerg Med. 1998;31(1):41–8.
20. Cooke DW, Plotnick L. Management of diabetic keto- 33. Viallon A, et al. Does bicarbonate therapy improve
acidosis in children and adolescents. Pediatr Rev. the management of severe diabetic ketoacidosis? Crit
2008;29:431–6. Care Med. 1999;27(12):2690–3.
21. Wolfsdorf JI, Allgrove J, Craig ME, et al. Diabetic 34. Duhon B, et al. Intravenous sodium bicarbonate ther-
ketoacidosis and hyperglycemic hyperosmolar state: a apy in severely acidotic diabetic ketoacidosis. Ann
consensus statement from the International Society for Pharmacother. 2013;47(7–8):970–5.
Pediatric and Adolescent Diabetes. Pediatr Diabetes. 35. Kamalakannan D, et al. Diabetic ketoacidosis in preg-
2014;15:154–79. nancy. Postgrad Med J. 2003;79(934):454–7.
ERRNVPHGLFRVRUJ
Part VIII
Neurologic Controversies
ERRNVPHGLFRVRUJ
Optimizing Sedation
in the Pediatric ICU 18
Rita V. Alvarez and Chani Traube
ERRNVPHGLFRVRUJ
296 R. V. Alvarez and C. Traube
ERRNVPHGLFRVRUJ
18 Optimizing Sedation in the Pediatric ICU 297
a beneficial side effect in mechanically ventilated Another possible option is low-dose ketamine,
patients). In addition to pain control, neurologic an NMDA receptor antagonist. As NMDA recep-
effects include sedation, meiosis, nausea, and pruri- tor activation mediates opioid tolerance, a low-
tus. Opioids increase smooth muscle tone which dose ketamine infusion (when given alongside an
can lead to constipation and urinary retention. opioid) may delay the development of tolerance.
Despite these well-known side effects, opioids are These drugs also have sedative properties, which
highly effective and useful analgesics [3]. can be a beneficial side effect in mechanically
A multimodal approach to treatment of pain ventilated children [3, 5]. However, there is evi-
appears to be most effective. An underappreci- dence (in vitro and in young animal models) that
ated analgesic is acetaminophen. By inhibiting has suggested a link between NMDA receptor
central cyclooxygenase, it has potent analgesic antagonism and a toxic influx of intraneuronal
and antipyretic effects. It works synergistically calcium, leading to increased reactive oxygen
with opioids, causing an “opioid-sparing” species and neuronal cell death. This may lead to
effect. Availability of an intravenous form of alterations in brain development, with disturbed
acetaminophen allows for effective use in nearly motor function, learning, and memory [58–60].
all critically ill patients (although it must be Research is needed to determine the applicability
avoided in those with significant hepatic insuf- of these studies in young children.
ficiency) [3]. Lastly, regional anesthesia (such as nerve
Another highly effective class of non-opioid blocks and epidural catheters) is highly effective
analgesics are the nonsteroidal anti-inflammatory at optimizing pain control with minimal systemic
drugs, including ibuprofen and ketorolac. These effects. The increased availability of regional
inhibit peripheral cyclooxygenase, with strong anesthesia in the PICU over recent years has been
analgesic, antipyretic, and anti-inflammatory a huge asset for improving pain control [61–69].
effects. The potential hemostatic (antiplatelet) Nonpharmacologic analgesia may be espe-
effects are infrequently seen clinically. NSAIDs cially useful in children. Studies suggest that
must be avoided with significant renal impair- massage, acupuncture, and distraction (e.g.,
ment [3]. Anecdotally, we have found that music therapy) can be beneficial [70–79]. Further
mechanically ventilated children who are treated research is needed to demonstrate the effective-
with acetaminophen and ketorolac routinely ness of adjunctive nonpharmacologic analgesia
achieve greater comfort on lower doses of in critically ill children.
opioids.
Other non-opioid analgesics include the alpha
agonists, such as clonidine and dexmedetomi- Sedation
dine. Postsynaptic alpha receptors are activated
in the CNS leading to decreased sympathetic As trainees we are taught that sedation is
activity via inhibition of norepinephrine release. employed to facilitate care, to prevent dislodge-
Sedation occurs with decreased firing at the locus ment of catheters and monitors, as well as to
coeruleus in the brainstem; analgesia results from decrease the stress response triggered by the per-
binding in the spinal cord [46–50]. In adults, the ception of anxiety and pain. In the past, sedation
use of alpha agonists has been associated with was considered an absolute necessity when car-
reduced need for opioids, benzodiazepines, and ing for critically ill patients, especially when
propofol [50, 51]. In children, limited literature invasive therapies such as mechanical ventilation
indicates alpha agonists may decrease amount of were employed. However, decades of research
benzodiazepines and opioids needed in those have suggested significant consequences associ-
mechanically ventilated [53]. Regarding clini- ated with overuse of sedative agents [4]. Animal
cally significant adverse effects, bradycardia is studies have shown widespread neuroapoptosis
reported with dexmedetomidine and hypotension and decreased neurogenesis with administration
with clonidine [46–48, 50–57]. of sedatives and anesthetics [80–88], including
ERRNVPHGLFRVRUJ
298 R. V. Alvarez and C. Traube
benzodiazepines. There is growing evidence that associated with development and duration of
this neurotoxicity occurs in premature infants delirium, increased length of mechanical venti-
[89–94]. In adults and children, there is mounting lation, and increased length of ICU stay [4, 17,
literature showing an association between benzo- 105, 113–117].
diazepines, delirium, and poor outcomes [10, In children, recent studies have demonstrated
12–14, 52, 95–108]. similar outcomes in association with benzodiaze-
Since sedation is potentially associated with pine-based sedation. Benzodiazepines have been
poor outcomes, it is important to re-examine the independently associated with pediatric delirium,
reasons why we sedate children. There are sev- with odds ratios ranging from 2.2 (in a multina-
eral common misconceptions regarding seda- tional study including 994 children and 25 differ-
tion of critically ill children [4]. There is the ent PICUs) to OR 3.8 in a cardiac ICU (n = 99
belief that sedated children are more comfort- patients) and OR 5.2 in a single-center ICU study
able; in fact, as noted previously, oversedation is including more than 1500 children [11–13, 95].
a major reason for inadequate analgesia [5]. Benzodiazepines have not merely been indepen-
There is also the myth that sedatives facilitate dently associated with delirium; benzodiazepines
sleep. Although a deeply sedated child may have also been linked to outcome measures.
appear to be sleeping, sedatives (particularly Smith et al. showed that greater benzodiazepine
benzodiazepines) completely disrupt sleep exposure was associated with lower likelihood of
cycles and interfere with the body’s ability to ICU discharge (HR 0.65, P = 0.01) and longer
consolidate restorative sleep [1, 109, 110]. delirium duration (IRR 2.5, P = 0.005) in 300
There is also misplaced compassion – those at children ages 6 months to 5 years old. This study
the bedside may want the child to “rest and not showed that dexmedetomidine increased likeli-
remember” the traumatic hospitalization. hood of ICU discharge (p = 0.008) [118]. Modi
Unfortunately, children who are sedated are not et al. established a causal inference between ben-
completely amnestic; they often lay down delu- zodiazepines and pediatric delirium. In 580 sub-
sional memories which paradoxically increase jects, a temporal relationship was demonstrated,
the chance for psychological morbidity in survi- where benzodiazepines were highly and indepen-
vors [111]. Finally, bedside providers may pre- dently associated with transition from normal
sume that a sedated child will be easier to mental status to delirium. Data showed a dose-
manage. In fact, oversedation often leads to response effect, with probability of transitioning
delirium, and delirious children are notoriously to delirium increasing with dose of benzodiaze-
difficult to care for [4, 112]. pine given during the previous day. Sophisticated
In 2011, the Society of Critical Care Medicine marginal structural modeling was used to care-
(SCCM) released Guidelines for the Management fully control for time-dependent confounders
of Pain, Agitation, and Delirium in Adult including cognitive status, opioids, and mechani-
Patients. The SCCM guidelines recommend cal ventilation. This pseudo-randomized sample
light sedation for ICU patients, with a preference revealed that benzodiazepines independently
for propofol or dexmedetomidine rather than increased subsequent delirium risk by 333%.
benzodiazepines [10]. In the adult literature, These data suggest that alternatives to benzodiaz-
oversedation is associated with poor outcomes epine-based sedation (prioritizing dexmedetomi-
including prolonged mechanical ventilation and dine) may be beneficial in critically ill children.
ICU length of stay, greater incidence of delirium, Although we have suggested that much of the
and death. On the contrary, minimizing sedation sedation currently provided in the PICU is unnec-
is associated with decreased respiratory depres- essary, there are clearly clinical situations where
sion, diaphragmatic atrophy, ventilator-associ- sedation is absolutely warranted. For example, it
ated pneumonia, and prevalence and duration of is reasonable to sedate children for procedures,
coma. Benzodiazepine use in particular has been facilitate patient-ventilator synchrony, allow for
ERRNVPHGLFRVRUJ
18 Optimizing Sedation in the Pediatric ICU 299
tolerance of monitors and catheters, and decrease accidental extubations with this sedation-sparing
metabolic demand (i.e., in the setting of shock or approach [122]. Another groundbreaking study
intracranial hypertension). Just as systematic was published from Advocate Children’s
assessments and reassessments are necessary for Hospital, where 70 children after cardiothoracic
optimizing management of pain, the same is true surgery were randomized to either continuous
for sedation [30]. We recommend the use of the sedation (morphine and midazolam infusions) or
Richmond Agitation-Sedation Scale (RASS), an placebo (a saline infusion). Both groups were
intuitive tool that allows for assignment of seda- allowed as-needed open-label doses of morphine
tion targets in ventilated and non-ventilated chil- and midazolam. Significantly, there was no dif-
dren of all ages, and titration of medication to ference in the number of intermittent doses
achieve sedation goal [119]. Please see case study required between the two groups. Children with-
for an example. out background sedation required no more
In contrast to the literature in critically ill “breakthrough” doses than children on continu-
adults, daily sedation interruption (DSI) with ous infusions. This decrease in sedation exposure
spontaneous breathing trials (SBT) has not been led to a significantly shorter hospital length of
proven effective in pediatrics. Vet et al. com- stay in the placebo group (4.9 days versus
pared DSI to a standard sedation protocol in a 8.4 days, p = 0.04) [123].
randomized controlled trial including 129 chil-
dren in three PICUs in the Netherlands. In the
DSI group, children were on continuous proto- Delirium
colized sedative infusions; sedation was held
once daily to allow the child to emerge for an Delirium is an acute neuropsychiatric syndrome
attempt at spontaneous breathing; the child was with altered cognition and consciousness. It is an
then sedated again. In the control group, children acute encephalopathy that always represents a
were maintained on protocolized sedation. In change from baseline [124]. Its neuropathophysi-
this study cohort, DSI was not beneficial; there ology is complex and incompletely understood.
was no change in ventilator-free days, length of Dopaminergic, serotoninergic, glutaminergic,
stay, or cumulative amount of sedation received. and cholinergic pathways in the cerebral cortex,
There were higher rates of agitation reported in striatum, substantia nigra, and thalamus have
the DSI group and an overall increased mortality been implicated. Imbalance in the synthesis,
[120]. Many PICUs are adopting a different release, and inactivation of neurotransmitters can
approach: rather than continuous sedation with result in altered cognitive function, behavior, and
DSI, keep children as lightly sedated as possible mood [108]. It is likely that delirium in the ICU
throughout the day, and proactively wean the is a result of multiple factors inherent to the
ventilator [121]. patient, the presenting illness, and the ICU care
Pediatric evidence shows that an analgoseda- [125]. Associated risk factors (Table 18.1) as well
tion approach is feasible in critically ill children as short-term outcomes (Table 18.2) associated
(Fig. 18.2). In 2011, Seattle Children’s Hospital with delirium in children have been described in
employed a morphine-based “comfort” protocol the growing literature [11–14, 95, 106, 112, 121,
in mechanically ventilated patients, with sedation 125–140]. Pediatric delirium has been associated
added only if needed. In a cohort of 166 children, with increased length of stay, costs, time to extu-
they demonstrated a significant decrease in seda- bation, and excess mortality.
tion days (7 versus 5 days, p = 0.026) and a trend Delirium is a common occurrence in criti-
toward shorter time on mechanical ventilation cally ill children. High prevalence rates have
(HR 0.81, p = 0.06). There was also a trend been reported in the general PICU (17–38%),
toward decreased ICU length of stay (HR 0.81, the pediatric cardiac ICU (49–57%), and the
p = 0.058). Importantly, there was no increase in postoperative PICU population (66%) [11, 13,
ERRNVPHGLFRVRUJ
300 R. V. Alvarez and C. Traube
YES NO
Continue opioid scheduled or Give opioid prns.
infusion and prn. If scheduled opioid, consider adjusting dose or starting an infusion.
If on infusion, adjust rate if requiring multiple prn boluses.
Consider adding α-2 agonists prn q1hr or infusion.
YES NO
Continue opioid infusion and prn pain. Address analgesia first.
If on α-2 agonists prn or infusion, continue. Give opioid prns and adjust infusion rate if requiring multiple prn boluses.
Treat anxiety:
Give α-2 agonists prn q1hr and adjust infusion rate if multiple prns.
± Give a benzodiazepine prn q1-6hr.
YES NO
Continue opioid infusion and prn pain. Address analgesia first.
If on α-2 agonists prn or infusion, continue. Give opioid prns and adjust infusion rate if requiring
multiple prn boluses.
Fig. 18.2 Sample analgosedation protocol for mechanically ventilated patients. Prn, as needed
95, 106, 127, 140]. Delirium can be classified mixed type in which patients may have a nor-
by subtypes as defined by the level of psycho- mal level of psychomotor activity or fluctuat-
motor activity. The pediatric literature sup- ing level of activity. Lastly, hyperactive
ports that the hypoactive subtype is most delirium is seen as mood lability, agitation,
common (46–56%) and may present as a slug- and restlessness and is the least common sub-
gish, lethargic, apathetic, and even stuporous type described (5–8%) [11, 12, 14]. Adult
child. Following in frequency (43–45%) is literature supports that of all the subtypes, the
ERRNVPHGLFRVRUJ
18 Optimizing Sedation in the Pediatric ICU 301
Table 18.2 Outcomes associated with delirium of delirium and targeting light levels of sedation
Increased PICU length of stay [12, 144]. Another exciting study showed a step-
Increased hospital length of stay wise decrease in delirium rates with a bundled
Increased duration of mechanical ventilation approach: institution of unit-wide delirium
Higher cost of care screening, followed by protocolized sedation,
Increased mortality and then an early mobilization initiative.
Delirium rates decreased 39% over the course of
hypoactive patient is associated with the worst the study [139].
outcomes [141, 142]. This has yet to be
explored within pediatrics.
SCCM has released clinical practice guide- Conclusion: A Changing Paradigm
lines recommending widespread delirium screen-
ing in adults, as well as treatment to decrease Progress is impossible without change, and those
duration of delirium and ameliorate its long-term who cannot change their minds cannot change
anything.
effects [10]. There are now validated tools avail- —George Bernard Shaw
able for use at the bedside to screen for delirium
in children of all ages [106, 129, 143]. The Traditional sedation protocols have used a
European Society of Paediatric and Neonatal sedative-first approach with added opioids as
Intensive Care (ESPNIC) has recommended that needed for pain. These protocols are more likely
all children in the PICU be monitored for delir- to be associated with oversedation, which is in
ium routinely using the Cornell Assessment of turn linked to masking of pain and delirium,
Pediatric Delirium [30]. increased length of mechanical ventilation, ICU
Importantly, we have evidence that delirium length of stay, and health-care costs. An anal-
is amenable to intervention [10]. Simply by gosedation approach in critically ill adults has
implementing routine screening for delirium, we led to decreased delirium, shorter lengths of
can decrease delirium burden in at-risk children. mechanical ventilation, and ICU length of stay
Widespread screening results in earlier recogni- [15]. Emerging data in pediatrics suggests simi-
tion of delirium and allows for intervention and lar effects. The pediatric critical care commu-
decrease in delirium duration. Widespread delir- nity is on the cusp of a culture shift. In the next
ium screening also improves staff knowledge decade, we may see the pendulum swing toward
about delirium and creates opportunities for an analgesic-first approach to sedation, with
unit-wide implementation of preventive strate- incorporation of delirium prevention into our
gies. A single-center experience describes a daily PICU practice. This is an opportunity to
decrease in delirium rates from 46% to 35% in potentially improve outcomes in our pediatric
children on invasive mechanical ventilation over patients.
a 2-year period, simply by increasing awareness
ERRNVPHGLFRVRUJ
302 R. V. Alvarez and C. Traube
ERRNVPHGLFRVRUJ
18 Optimizing Sedation in the Pediatric ICU 303
titrated to effect. Alex is able to sleep for several an analgosedation approach allowed for prioriti-
hours but then awakens and communicates pain zation of adequate pain control first, followed by
when his nurse suctions his endotracheal tube. A added anxiety control with dexmedetomidine as
plan is made for preemptive morphine prior to needed. The benefits of verbal communication
suctioning. and environmental comforts described here are
Despite moderate ventilator settings, he is often overlooked; it may be time to employ vari-
awake and reasonably cooperative. This allows ous therapeutic methods in real time instead of
for early mobilization, and although he refuses to waiting for PTSD to develop [7–9]. Pediatric
ambulate, he can be moved to his bedside chair intensivists have reported success with active
for a 30-min period. He is interactive with par- parental involvement and incorporation of child
ents and staff and watches a movie with his mom. life, physical, occupational, and speech therapists
He is exhausted from the exercise and is able to into routine PICU care. This remains an area for
sleep for 5 h at night (especially once his parents much needed research.
provide his special blanket and stuffed animal
from home). Twice daily delirium screening
occurs; when his delirium score (using the References
Cornell Assessment of Pediatric Delirium) begins
to rise, the medical team reviews his medication 1. Kudchadkar SR, Yaster M, Punjabi NM. Sedation,
sleep promotion, and delirium screening practices in
list and discontinues unnecessary anticholinergic the care of mechanically ventilated children: a wake-
medication; they also remove his Foley catheter. up call for the pediatric critical care community*.
His subsequent delirium scores improve. Crit Care Med. 2014;42(7):1592–600.
Over the first 3 days, his A-a gradient 2. Curley MA, Wypij D, Watson RS, Grant MJ, Asaro
LA, Cheifetz IM, et al. Protocolized sedation vs
increases. His RASS target is changed to −1 usual care in pediatric patients mechanically ven-
(drowsy) to facilitate patient-ventilator syn- tilated for acute respiratory failure: a randomized
chrony; he receives several boluses of dexme- clinical trial. JAMA. 2015;313(4):379–89.
detomidine and an increase in his 3. Tobias JD. Acute pain management in infants and
children-Part 2: intravenous opioids, intravenous
dexmedetomidine infusion to achieve target nonsteroidal anti-inflammatory drugs, and managing
sedation. By day 4 of mechanical ventilation, his adverse effects. Pediatr Ann. 2014;43(7):e169–75.
chest radiograph and peak inspiratory pressures 4. Peitz GJ, Balas MC, Olsen KM, Pun BT, Ely
have improved. His sedation target is changed EW. Top 10 myths regarding sedation and delir-
ium in the ICU. Crit Care Med. 2013;41(9 Suppl
back to RASS 0 (awake and calm) and his medi- 1):S46–56.
cation adjusted accordingly. He begins to tolerate 5. Sigakis MJ, Bittner EA. Ten myths and misconcep-
a slow ventilator wean. He participates in physi- tions regarding pain management in the ICU. Crit
cal therapy twice daily and is kept cognitively Care Med. 2015;43(11):2468–78.
6. Myhren H, Ekeberg O, Toien K, Karlsson S,
stimulated during the day to facilitate nighttime Stokland O. Posttraumatic stress, anxiety and
sleep. As he has been spontaneously breathing depression symptoms in patients during the first year
and exercising for most of his ICU stay, he has post intensive care unit discharge. Crit Care (Lond).
not experienced significant deconditioning. He 2010;14(1):R14.
7. Caldas JC, Pais-Ribeiro JL, Carneiro SR. General
weans quickly and is successfully extubated after anesthesia, surgery and hospitalization in children
6 total days of invasive mechanical ventilation. and their effects upon cognitive, academic, emo-
tional and sociobehavioral development – a review.
Paediatr Anaesth. 2004;14(11):910–5.
8. Rennick JE, Rashotte J. Psychological outcomes
Discussion in children following pediatric intensive care
unit hospitalization: a systematic review of the
Our “traditional” approach to sedation allowed research. J Child Health Care. 2009;13(2):128–49.
for a still and quiet patient, which may be sooth- 9. Lerwick JL. Psychosocial implications of pediat-
ric surgical hospitalization. Semin Pediatr Surg.
ing to the observer – however, the sedated patient 2013;22(3):129–33.
may still be experiencing pain, fear, and anxiety 10. Barr J, Fraser GL, Puntillo K, Ely EW, Gelinas C,
[145]. Alternatively, managing this patient with Dasta JF, et al. Clinical practice guidelines for the
ERRNVPHGLFRVRUJ
304 R. V. Alvarez and C. Traube
management of pain, agitation, and delirium in adult 25. Anwar K. Pathophysiology of pain. Dis Mon.
patients in the intensive care unit. Crit Care Med. 2016;62(9):324–9.
2013;41(1):263–306. 26. Cross SA. Pathophysiology of pain. Mayo Clin Proc.
11. Alvarez RV, Palmer C, Czaja AS, Peyton C, Silver 1994;69(4):375–83.
G, Traube C, et al. Delirium is a common and early 27. Wiatrowski R, Norton C, Giffen D. Analgosedation:
finding in patients in the pediatric cardiac intensive improving patient outcomes in ICU seda-
care unit. J Pediatr. 2018;195:206. tion and pain management. Pain Manag Nurs.
12. Traube C, Silver G, Gerber LM, Kaur S, Mauer 2016;17(3):204–17.
EA, Kerson A, et al. Delirium and mortality in 28. Joshi GP, Ogunnaike BO. Consequences of inad-
critically ill children: epidemiology and out- equate postoperative pain relief and chronic persis-
comes of pediatric delirium. Crit Care Med. tent postoperative pain. Anesthesiol Clin North Am.
2017;45(5):891–8. 2005;23(1):21–36.
13. Patel AK, Biagas KV, Clarke EC, Gerber LM, 29. Batton DG, Barrington KJ, Wallman C. Prevention
Mauer E, Silver G, et al. Delirium in children after and management of pain in the neonate: an update.
cardiac bypass surgery. Pediatr Crit Care Med. Pediatrics. 2006;118(5):2231–41.
2017;18(2):165–71. 30. Harris J, Ramelet AS, van Dijk M, Pokorna P,
14. Silver G, Traube C, Gerber LM, Sun X, Kearney J, Wielenga J, Tume L, et al. Clinical recommenda-
Patel A, et al. Pediatric delirium and associated risk tions for pain, sedation, withdrawal and delirium
factors: a single-center prospective observational assessment in critically ill infants and children: an
study. Pediatr Crit Care Med. 2015;16(4):303–9. ESPNIC position statement for healthcare profes-
15. Devabhakthuni S, Armahizer MJ, Dasta JF, Kane- sionals. Intensive Care Med. 2016;42(6):972–86.
Gill SL. Analgosedation: a paradigm shift in inten- 31. Scott J, Huskisson EC. Graphic representation of
sive care unit sedation practice. Ann Pharmacother. pain. Pain. 1976;2(2):175–84.
2012;46(4):530–40. 32. Manworren RC, Stinson J. Pediatric pain measure-
16. Bartel B. New sedation practices in the adult ment, assessment, and evaluation. Semin Pediatr
intensive care unit: analgosedation. S D Med. Neurol. 2016;23(3):189–200.
2012;65(6):234–5. 33. Chanques G, Viel E, Constantin JM, Jung B, de
17. Gradwohl-Matis I, Mehta S, Dunser MW. What’s Lattre S, Carr J, et al. The measurement of pain
new in sedation strategies? Intensive Care Med. in intensive care unit: comparison of 5 self-report
2015;41(9):1696–9. intensity scales. Pain. 2010;151(3):711–21.
18. Chanques G, Conseil M, Roger C, Constantin JM, 34. Hjermstad MJ, Fayers PM, Haugen DF, Caraceni
Prades A, Carr J, et al. Immediate interruption of A, Hanks GW, Loge JH, et al. Studies comparing
sedation compared with usual sedation care in criti- numerical rating scales, verbal rating scales, and
cally ill postoperative patients (SOS-ventilation): visual analogue scales for assessment of pain inten-
a randomised, parallel-group clinical trial. Lancet sity in adults: a systematic literature review. J Pain
Respir Med. 2017;5(10):795–805. Symptom Manag. 2011;41(6):1073–93.
19. Alderson SM, McKechnie SR. Unrecognised, under- 35. Buttes P, Keal G, Cronin SN, Stocks L, Stout
treated, pain in ICU––Causes, effects, and how to do C. Validation of the critical-care pain observation
better. Open J Nurs. 2013;3:108–13. tool in adult critically ill patients. Dimens Crit Care
20. Grosclaude C, Asehnoune K, Demeure D, Millet S, Nurs. 2014;33(2):78–81.
Champin P, Naux E, et al. Opinion of different pro- 36. de Jong A, Baartmans M, Bremer M, van Komen
fessional categories about the intensity of procedural R, Middelkoop E, Tuinebreijer W, et al. Reliability,
pain in adult intensive care units. Ann Fr Reanim. validity and clinical utility of three types of pain
2010;29(12):884–8. behavioural observation scales for young children
21. Pasero C, Puntillo K, Li D, Mularski RA, Grap MJ, with burns aged 0–5 years. Pain. 2010;150(3):561–7.
Erstad BL, et al. Structured approaches to pain man- 37. van Dijk M, de Boer JB, Koot HM, Tibboel D,
agement in the ICU. Chest. 2009;135(6):1665–72. Passchier J, Duivenvoorden HJ. The reliability and
22. Chanques G, Sebbane M, Barbotte E, Viel E, Eledjam validity of the COMFORT scale as a postoperative
JJ, Jaber S. A prospective study of pain at rest: inci- pain instrument in 0 to 3-year-old infants. Pain.
dence and characteristics of an unrecognized symp- 2000;84(2):367–77.
tom in surgical and trauma versus medical intensive 38. van Dijk M, Peters JW, van Deventer P, Tibboel
care unit patients. Anesthesiology. 2007;107(5): D. The COMFORT Behavior Scale: a tool for
858–60. assessing pain and sedation in infants. Am J Nurs.
23. Novaes MA, Knobel E, Bork AM, Pavao OF, 2005;105(1):33–6.
Nogueira-Martins LA, Ferraz MB. Stressors in ICU: 39. Ahn Y, Jun Y. Measurement of pain-like response to
perception of the patient, relatives and health care various NICU stimulants for high-risk infants. Early
team. Intensive Care Med. 1999;25(12):1421–6. Hum Dev. 2007;83(4):255–62.
24. Kyranou M, Puntillo K. The transition from acute to 40. Manworren RC, Hynan LS. Clinical validation of
chronic pain: might intensive care unit patients be at FLACC: preverbal patient pain scale. Pediatr Nurs.
risk? Ann Intensive Care. 2012;2(1):36. 2003;29(2):140–6.
ERRNVPHGLFRVRUJ
18 Optimizing Sedation in the Pediatric ICU 305
41. Merkel SI, Voepel-Lewis T, Shayevitz JR, Malviya for sedation of children hospitalized in the intensive
S. The FLACC: a behavioral scale for scoring post- care unit. J Hosp Med. 2008;3(2):142–7.
operative pain in young children. Pediatr Nurs. 56. Chen K, Lu Z, Xin YC, Cai Y, Chen Y, Pan
1997;23(3):293–7. SM. Alpha-2 agonists for long-term sedation dur-
42. Voepel-Lewis T, Merkel S, Tait AR, Trzcinka A, ing mechanical ventilation in critically ill patients.
Malviya S. The reliability and validity of the face, Cochrane Database Syst Rev. 2015;1:CD010269.
legs, activity, cry, consolability observational tool as 57. Lam F, Ransom C, Gossett JM, Kelkhoff A, Seib
a measure of pain in children with cognitive impair- PM, Schmitz ML, et al. Safety and efficacy of dex-
ment. Anesth Analg. 2002;95(5):1224–9. medetomidine in children with heart failure. Pediatr
43. Voepel-Lewis T, Zanotti J, Dammeyer JA, Merkel Cardiol. 2013;34(4):835–41.
S. Reliability and validity of the face, legs, activ- 58. Lecointre M, Vezier C, Benard M, Ramdani Y, Dupre
ity, cry, consolability behavioral tool in assessing N, Brasse-Lagnel C, et al. Age-dependent alterations
acute pain in critically ill patients. Am J Crit Care. of the NMDA receptor developmental profile and
2010;19(1):55–61. adult behavior in postnatally ketamine-treated mice.
44. Willis MH, Merkel SI, Voepel-Lewis T, Malviya Dev Neurobiol. 2015;75(3):315–33.
S. FLACC Behavioral Pain Assessment Scale: a 59. Liu F, Patterson TA, Sadovova N, Zhang X, Liu S,
comparison with the child’s self-report. Pediatr Zou X, et al. Ketamine-induced neuronal damage
Nurs. 2003;29(3):195–8. and altered N-methyl-D-aspartate receptor func-
45. Patel AK, Bell MJ, Traube C. Delirium in tion in rat primary forebrain culture. Toxicol Sci.
pediatric critical care. Pediatr Clin N Am. 2013;131(2):548–57.
2017;64(5):1117–32. 60. Li X, Li Y, Zhao J, Li L, Wang Y, Zhang Y, et al.
46. Wang JG, Belley-Cote E, Burry L, Duffett M, Administration of ketamine causes autophagy and
Karachi T, Perri D, et al. Clonidine for sedation in apoptosis in the rat fetal hippocampus and in PC12
the critically ill: a systematic review and meta-anal- cells. Front Cell Neurosci. 2018;12:21.
ysis. Crit Care (Lond). 2017;21(1):75. 61. Lonnqvist PA, Ecoffey C, Bosenberg A, Suresh S,
47. Hayden JC, Breatnach C, Doherty DR, Healy Ivani G. The European society of regional anesthe-
M, Howlett MM, Gallagher PJ, et al. Efficacy of sia and pain therapy and the American society of
alpha2-agonists for sedation in pediatric critical regional anesthesia and pain medicine joint commit-
care: a systematic review. Pediatr Crit Care Med. tee practice advisory on controversial topics in pedi-
2016;17(2):e66–75. atric regional anesthesia I and II: what do they tell
48. Pichot C, Ghignone M, Quintin L. Dexmedetomidine us? Curr Opin Anaesthesiol. 2017;30(5):613–20.
and clonidine: from second- to first-line sedative 62. Hutchins J, Castro C, Wang Q, Chinnakotla
agents in the critical care setting? J Intensive Care S. Postoperative pain control with paravertebral
Med. 2012;27(4):219–37. catheters after pediatric total pancreatectomy and
49. Tobias JD, Berkenbosch JW. Initial experience islet autotransplantation: a retrospective cohort
with dexmedetomidine in paediatric-aged patients. study. Paediatr Anaesth. 2016;26(3):315–20.
Paediatr Anaesth. 2002;12(2):171–5. 63. Chalmers DJ, Bielsky A, Wild TT, Siparsky GL,
50. Venn RM, Bradshaw CJ, Spencer R, Brealey D, Wilcox DT. Continuous local anesthetic infusion for
Caudwell E, Naughton C, et al. Preliminary UK children with spina bifida undergoing major recon-
experience of dexmedetomidine, a novel agent for struction of the lower urinary tract. J Pediatr Urol.
postoperative sedation in the intensive care unit. 2015;11(2):72.e1–5.
Anaesthesia. 1999;54(12):1136–42. 64. Bairdain S, Dodson B, Zurakowski D, Waisel DB,
51. Jakob SM, Ruokonen E, Grounds RM, Sarapohja Jennings RW, Boretsky KR. Paravertebral nerve
T, Garratt C, Pocock SJ, et al. Dexmedetomidine block catheters using chloroprocaine in infants with
vs midazolam or propofol for sedation during pro- prolonged mechanical ventilation for treatment
longed mechanical ventilation: two randomized of long-gap esophageal atresia. Paediatr Anaesth.
controlled trials. JAMA. 2012;307(11):1151–60. 2015;25(11):1151–7.
52. Riker RR, Shehabi Y, Bokesch PM, Ceraso D, 65. Di Pede A, Morini F, Lombardi MH, Sgro S, Laviani
Wisemandle W, Koura F, et al. Dexmedetomidine R, Dotta A, et al. Comparison of regional vs. sys-
vs midazolam for sedation of critically ill patients: a temic analgesia for post-thoracotomy care in infants.
randomized trial. JAMA. 2009;301(5):489–99. Paediatr Anaesth. 2014;24(6):569–73.
53. Czaja AS, Zimmerman JJ. The use of dexmedetomi- 66. Wu Y, Liu F, Tang H, Wang Q, Chen L, Wu H,
dine in critically ill children. Pediatr Crit Care Med. et al. The analgesic efficacy of subcostal transver-
2009;10(3):381–6. sus abdominis plane block compared with tho-
54. Arenas-Lopez S, Riphagen S, Tibby SM, Durward racic epidural analgesia and intravenous opioid
A, Tomlin S, Davies G, et al. Use of oral clonidine analgesia after radical gastrectomy. Anesth Analg.
for sedation in ventilated paediatric intensive care 2013;117(2):507–13.
patients. Intensive Care Med. 2004;30(8):1625–9. 67. Lukosiene L, Rugyte DC, Macas A, Kalibatiene L,
55. Carroll CL, Krieger D, Campbell M, Fisher DG, Malcius D, Barauskas V. Postoperative pain man-
Comeau LL, Zucker AR. Use of dexmedetomidine agement in pediatric patients undergoing minimally
ERRNVPHGLFRVRUJ
306 R. V. Alvarez and C. Traube
invasive repair of pectus excavatum: the role of inter- tional neuroscience approach. J Vis Exp: JoVE.
costal block. J Pediatr Surg. 2013;48(12):2425–30. 2017;124:55193. https://doi.org/10.3791/55193.
68. Bosenberg AT, Johr M, Wolf AR. Pro con debate: the 82. Walters JL, Paule MG. Review of preclinical stud-
use of regional vs systemic analgesia for neonatal ies on pediatric general anesthesia-induced devel-
surgery. Paediatr Anaesth. 2011;21(12):1247–58. opmental neurotoxicity. Neurotoxicol Teratol.
69. Zanaboni S, Krauss B, Buscaglia R, Montagnini C, 2017;60:2–23.
Gratarola A, Gualino J, et al. Changes in respiratory 83. Vutskits L, Davidson A. Update on developmental
and hemodynamic parameters during low-dose pro- anesthesia neurotoxicity. Curr Opin Anaesthesiol.
pofol sedation in combination with regional anesthe- 2017;30(3):337–42.
sia for herniorrhaphy and genitourinary surgery in 84. Lin EP, Lee JR, Lee CS, Deng M, Loepke AW. Do
children. Paediatr Anaesth. 2007;17(10):934–41. anesthetics harm the developing human brain? An
70. van der Heijden MJE, Jeekel J, Rode H, Cox S, van integrative analysis of animal and human studies.
Rosmalen J, Hunink MGM, et al. Can live music Neurotoxicol Teratol. 2017;60:117–28.
therapy reduce distress and pain in children with 85. Karnwal A, Lippmann M. Neurotoxicity of anes-
burns after wound care procedures? A randomized thetic drugs on developing brain. Anesth Analg.
controlled trial. Burns. 2018;44:823. 2017;124(4):1377.
71. Mofredj A, Alaya S, Tassaioust K, Bahloul H, 86. Jackson WM, Gray CD, Jiang D, Schaefer ML,
Mrabet A. Music therapy, a review of the potential Connor C, Mintz CD. Molecular mechanisms of
therapeutic benefits for the critically ill. J Crit Care. anesthetic neurotoxicity: a review of the current liter-
2016;35:195–9. ature. J Neurosurg Anesthesiol. 2016;28(4):361–72.
72. Brittner M, Le Pertel N, Gold MA. Acupuncture in 87. Rappaport BA, Suresh S, Hertz S, Evers AS, Orser
pediatrics. Curr Probl Pediatr Adolesc Health Care. BA. Anesthetic neurotoxicity – clinical implications
2016;46(6):179–83. of animal models. N Engl J Med. 2015;372(9):796–7.
73. Bradt J, Dileo C, Magill L, Teague A. Music inter- 88. Sanders RD, Hassell J, Davidson AJ, Robertson
ventions for improving psychological and physical NJ, Ma D. Impact of anaesthetics and surgery
outcomes in cancer patients. Cochrane Database on neurodevelopment: an update. Br J Anaesth.
Syst Rev. 2016;(8):CD006911. 2013;110(Suppl 1):i53–72.
74. Yang C, Hao Z, Zhang LL, Guo Q. Efficacy and 89. Ng E, Taddio A, Ohlsson A. Intravenous mid-
safety of acupuncture in children: an overview of azolam infusion for sedation of infants in the neo-
systematic reviews. Pediatr Res. 2015;78(2):112–9. natal intensive care unit. Cochrane Database Syst
75. Tsao GJ, Messner AH, Seybold J, Sayyid ZN, Rev. 2017;1:CD002052.
Cheng AG, Golianu B. Intraoperative acupuncture 90. Andropoulos DB. Effect of anesthesia on the
for posttonsillectomy pain: a randomized, double- developing brain: infant and fetus. Fetal Diagn
blind, placebo-controlled trial. Laryngoscope. Ther. 2018;43(1):1–11.
2015;125(8):1972–8. 91. Andropoulos DB, Greene MF. Anesthesia and devel-
76. Gilbey P, Bretler S, Avraham Y, Sharabi-Nov A, oping brains – implications of the FDA warning. N
Ibrgimov S, Luder A. Acupuncture for posttonsil- Engl J Med. 2017;376(10):905–7.
lectomy pain in children: a randomized, controlled 92. Sun LS, Li G, Miller TL, Salorio C, Byrne MW,
study. Paediatr Anaesth. 2015;25(6):603–9. Bellinger DC, et al. Association between a single
77. Matsota P, Christodoulopoulou T, Smyrnioti general anesthesia exposure before age 36 months
ME, Pandazi A, Kanellopoulos I, Koursoumi and neurocognitive outcomes in later childhood.
E, et al. Music’s use for anesthesia and analge- JAMA. 2016;315(21):2312–20.
sia. J Altern Complement Med (New York, NY). 93. Duerden EG, Guo T, Dodbiba L, Chakravarty MM,
2013;19(4):298–307. Chau V, Poskitt KJ, et al. Midazolam dose correlates
78. Lin YC, Tassone RF, Jahng S, Rahbar R, Holzman with abnormal hippocampal growth and neurodevel-
RS, Zurakowski D, et al. Acupuncture man- opmental outcome in preterm infants. Ann Neurol.
agement of pain and emergence agitation in 2016;79(4):548–59.
children after bilateral myringotomy and tym- 94. Davidson AJ, Disma N, de Graaff JC, Withington
panostomy tube insertion. Paediatr Anaesth. DE, Dorris L, Bell G, et al. Neurodevelopmental
2009;19(11):1096–101. outcome at 2 years of age after general anaesthesia
79. Evans S, Tsao JC, Zeltzer LK. Paediatric pain man- and awake-regional anaesthesia in infancy (GAS):
agement: using complementary and alternative med- an international multicentre, randomised controlled
icine. Rev Pain. 2008;2(1):14–20. trial. Lancet (Lond). 2016;387(10015):239–50.
80. Loepke AW. Developmental neurotoxicity of seda- 95. Traube C, Silver G, Reeder RW, Doyle H, Hegel E,
tives and anesthetics: a concern for neonatal and Wolfe HA, et al. Delirium in critically ill children: an
pediatric critical care medicine? Pediatr Crit Care international point prevalence study. Crit Care Med.
Med. 2010;11(2):217–26. 2017;45(4):584–90.
81. Whitaker EE, Zheng CZ, Bissonnette B, et al. Use 96. O’Neal JB, Shaw AD. Predicting, preventing,
of a piglet model for the study of anesthetic-induced and identifying delirium after cardiac surgery.
developmental neurotoxicity (AIDN): a transla- Perioperative Medicine (Lond). 2016;5:7.
ERRNVPHGLFRVRUJ
18 Optimizing Sedation in the Pediatric ICU 307
97. Pandharipande P, Shintani A, Peterson J, Pun BT, 112. Traube C, Mauer EA, Gerber LM, Kaur S,
Wilkinson GR, Dittus RS, et al. Lorazepam is an Joyce C, Kerson A, et al. Cost associated with
independent risk factor for transitioning to delir- pediatric delirium in the ICU. Crit Care Med.
ium in intensive care unit patients. Anesthesiology. 2016;44(12):e1175–e9.
2006;104(1):21–6. 113. Treggiari MM, Romand JA, Yanez ND, Deem SA,
98. Pandharipande PP, Pun BT, Herr DL, Maze M, Goldberg J, Hudson L, et al. Randomized trial of
Girard TD, Miller RR, et al. Effect of sedation with light versus deep sedation on mental health after
dexmedetomidine vs lorazepam on acute brain critical illness. Crit Care Med. 2009;37(9):2527–34.
dysfunction in mechanically ventilated patients: 114. Shehabi Y, Bellomo R, Reade MC, Bailey M,
the MENDS randomized controlled trial. JAMA. Bass F, Howe B, et al. Early intensive care seda-
2007;298(22):2644–53. tion predicts long-term mortality in ventilated
99. Wilson JE, Brummel NE, Stollings critically ill patients. Am J Respir Crit Care Med.
JL. Benzodiazepine-associated delirium dosing 2012;186(8):724–31.
strategy or cumulative dose? Intensive Care Med. 115. Chlan LL, Weinert CR, Heiderscheit A, Tracy MF,
2015;41(12):2245–6. Skaar DJ, Guttormson JL, et al. Effects of patient-
100. Zaal IJ, Devlin JW, Hazelbag M, Klein directed music intervention on anxiety and sedative
Klouwenberg PM, van der Kooi AW, Ong DS, et al. exposure in critically ill patients receiving mechani-
Benzodiazepine-associated delirium in critically ill cal ventilatory support: a randomized clinical trial.
adults. Intensive Care Med. 2015;41(12):2130–7. JAMA. 2013;309(22):2335–44.
101. Devlin JW, Peelen LM, Slooter A. Benzodiazepine- 116. Shehabi Y, Chan L, Kadiman S, Alias A, Ismail WN,
associated delirium: further considerations. Intensive Tan MA, et al. Sedation depth and long-term mor-
Care Med. 2016;42(9):1517–8. tality in mechanically ventilated critically ill adults:
102. Zaal IJ, Devlin JW, Peelen LM, Slooter AJ. A sys- a prospective longitudinal multicentre cohort study.
tematic review of risk factors for delirium in the Intensive Care Med. 2013;39(5):910–8.
ICU. Crit Care Med. 2015;43(1):40–7. 117. Tanaka LM, Azevedo LC, Park M, Schettino G,
103. MacLaren R, Preslaski CR, Mueller SW, Kiser TH, Nassar AP, Rea-Neto A, et al. Early sedation and
Fish DN, Lavelle JC, et al. A randomized, double- clinical outcomes of mechanically ventilated
blind pilot study of dexmedetomidine versus mid- patients: a prospective multicenter cohort study.
azolam for intensive care unit sedation: patient recall Criti Care (Lond). 2014;18(4):R156.
of their experiences and short-term psychological 118. Smith HAB, Gangopadhyay M, Goben CM,
outcomes. J Intensive Care Med. 2015;30(3):167–75. Jacobowski NL, Chestnut MH, Thompson JL, et al.
104. Kamdar BB, Niessen T, Colantuoni E, King LM, Delirium and benzodiazepines associated with pro-
Neufeld KJ, Bienvenu OJ, et al. Delirium transitions longed ICU stay in critically ill infants and young
in the medical ICU: exploring the role of sleep qual- children. Crit Care Med. 2017;45(9):1427–35.
ity and other factors*. Crit Care Med. 2015;43:135. 119. Kerson A. Validity of the Richmond Agitation-
105. Reade MC, Finfer S. Sedation and delir- Sedation Scale (RASS) in critically ill children. J
ium in the intensive care unit. N Engl J Med. Intensive Care. 2016;4:65.
2014;370(5):444–54. 120. Vet NJ, de Wildt SN, Verlaat CW, Knibbe CA, Mooij
106. Traube C, Silver G, Kearney J, Patel A, Atkinson MG, van Woensel JB, et al. A randomized controlled
TM, Yoon MJ, et al. Cornell assessment of pediat- trial of daily sedation interruption in critically ill
ric delirium: a valid, rapid, observational tool for children. Intensive Care Med. 2016;42(2):233–44.
screening delirium in the PICU*. Crit Care Med. 121. Traube C, Greenwald BM. “The Times They
2014;42(3):656–63. Are A-Changin”: universal delirium screening
107. Turkel SB, Hanft A. The pharmacologic man- in pediatric critical care. Pediatr Crit Care Med.
agement of delirium in children and adolescents. 2017;18(6):594–5.
Paediatr Drugs. 2014;16(4):267–74. 122. Deeter KH, King MA, Ridling D, Irby GL, Lynn
108. Maldonado JR. Neuropathogenesis of delir- AM, Zimmerman JJ. Successful implementation of
ium: review of current etiologic theories and a pediatric sedation protocol for mechanically venti-
common pathways. Am J Geriatr Psychiatry. lated patients. Crit Care Med. 2011;39(4):683–8.
2013;21(12):1190–222. 123. Penk JS, Lefaiver CA, Brady CM, Steffensen CM,
109. Kudchadkar SR, Aljohani OA, Punjabi NM. Sleep Wittmayer K. Intermittent versus continuous and
of critically ill children in the pediatric intensive intermittent medications for pain and sedation after
care unit: a systematic review. Sleep Med Rev. pediatric cardiothoracic surgery; a randomized con-
2014;18(2):103–10. trolled trial. Crit Care Med. 2018;46(1):123–9.
110. Barnes SS, Kudchadkar SR. Sedative choice and 124. American Psychiatric Association: Diagnostic and
ventilator-associated patient outcomes: don’t sleep statistical manual of mental disorders: DSM-V.
on delirium. Ann Transl Med. 2016;4(2):34. 5th ed. Arlington, VA: American Psychiatric
111. Colville G, Kerry S, Pierce C. Children’s factual and Association; 2013.
delusional memories of intensive care. Am J Respir 125. Smith HA, Fuchs DC, Pandharipande PP, Barr
Crit Care Med. 2008;177(9):976–82. FE, Ely EW. Delirium: an emerging frontier in the
ERRNVPHGLFRVRUJ
308 R. V. Alvarez and C. Traube
ERRNVPHGLFRVRUJ
Diagnosis of Brain Death
and Organ Donation After 19
Circulatory Death
Anthony A. Sochet, Alexandra K. Glazier,
and Thomas A. Nakagawa
Abbreviations Introduction
AAP American Academy of Pediatrics Understanding how death is determined and the
CPR Cardiopulmonary resuscitation associated ethical conflicts is paramount to the
DCD Donation after circulatory determina- pediatric critical care provider charged with car-
tion of death ing for critically ill children, determining death,
DND Donation after neurologic determina- and providing continued, humanistic support to
tion of death families that want their child to be an organ
DNR Do not resuscitate donor. Death conceptually is the permanent ces-
OPO Organ procurement organization sation of biological function. While this basic
PICU Pediatric intensive care unit notion has endured over time, the methods by
SCCM Society of Critical Care Medicine which death is determined have evolved and
UAGA Uniform Anatomical Gift Act adapted to changing social values and advances
UDDA Uniform Determination of Death Act in biomedical technology. Historically, death was
WLST Withdrawal of life-sustaining therapies determined by the cessation of vital functions
such as respiration and heartbeat. Advances in
modern medical care allow certain physiologic
functions to be artificially maintained for pro-
longed periods of time, such as mechanical venti-
lators that can breathe for patients and mechanical
assist devices that can provide artificial circula-
tion. With these advances, application of tradi-
tional means to determine death by the absence
A. A. Sochet · T. A. Nakagawa (*) of breathing and circulation was no longer clear
Anesthesiology and Critical Care Medicine, The nor fully satisfactory. Medical innovations such
Johns Hopkins University School of Medicine,
as organ transplantation also necessitated re-
Baltimore, MD, USA
examination of how death could be determined.
Division of Pediatric Critical Care Medicine, Johns
The recovery of vital organs for transplantation
Hopkins All Children’s Hospital,
St. Petersburg, FL, USA mandated legal, ethical, and clinical precision to
e-mail: thomas.nakagawa@jhmi.edu determine when and how death could be appro-
A. K. Glazier priately established in circumstances where
New England Donor Services, Waltham, MA, USA breathing and circulation were artificially
ERRNVPHGLFRVRUJ
310 A. A. Sochet et al.
ERRNVPHGLFRVRUJ
19 Diagnosis of Brain Death and Organ Donation After Circulatory Death 311
The President’s Commission unfortunately failed presented in Figs. 19.1 and 19.2. This informa-
to adequately address unique considerations tion helps frame further discussion about contro-
regarding determination of neurologic death in versies related to determination of death and
the pediatric population resulting in the 1987 task organ donation in this chapter.
force guidelines for the determination of brain
death in children. These guidelines were revised
by the Society of Critical Care Medicine (SCCM), Case Presentation: Part 1
American Academy of Pediatrics (AAP), and
Child Neurology Society in 2011 [9, 10]. The A 7-year-old child is admitted to your pediatric
revised multi-society guidelines continued to intensive care unit (PICU) following a motor
emphasize that neurologic death for infants and vehicle collision. The child was an unrestrained,
children is a clinical diagnosis based on the front seat passenger, ejected from the vehicle and
absence of neurologic function with a known found pulseless by the emergency medical per-
irreversible cause of coma [10]. sonnel. The child was intubated on the scene and
Transplantation medicine has continued to received cardiopulmonary resuscitation (CPR) for
evolve since the highly publicized 1967 heart >30 min before return of spontaneous circulation.
transplant. The growing need for transplantable After initial stabilization in the emergency depart-
organs is evident by the increasing gap between ment, the child was noted to have minor orthope-
donated organs and transplant recipients despite dic injuries, dermal injuries, and fixed, dilated
significant increases in the number of organs pupils. Neuroimaging reveals the loss of gray-
donated and transplanted [11]. In 2018, wait- white matter differentiation, effacement of cere-
listed individuals for organ transplantation in the bral and cerebellar cortical tissue, and grossly
United States exceeded 115,000, with nearly restricted diffusion consistent with global anoxic
75,000 active candidates. Children account for injury and intracranial hypertension. After 2 days
more than 1.5% of those waitlisted individuals. of medical management in the PICU, the patient
In 2017, 16,488 organ donors provided 34,772 remains mechanically ventilated without observed
transplanted organs of which 82% were deceased spontaneous respiratory effort, perfusion is main-
donors (with pediatric donors accounting for tained with minimal dose of vasoactive agents, no
0.7% of all donors). This represents the highest electrical activity is noted on the electroencepha-
number of donors and organs transplanted ever logram, no reaction to painful stimuli is elicited,
achieved in the United States and a 20% increase and pupils remain fixed and dilated. Since admis-
over the previous 5 years. Nonetheless, approxi- sion, the child has not received any sedative or
mately 7000 individuals on the waiting list neuromuscular blocking agents.
(~6%) die annually waiting for an organ trans- Your colleague has been the primary provider
plant [12]. Additionally, candidates are removed for this patient and is concerned the child’s exam-
from the waiting list and die because they ination is consistent with neurologic death. The
become too sick to transplant while waiting for a designated organ procurement organization
lifesaving organ. The AAP supports all efforts to (OPO) has been contacted as required by federal
facilitate and increase organ donation. The regulation [15]. The family is informed about
opportunity for organ donation is rare in the proceeding with testing to determine neurologic
pediatric population, but the need is significant death. They express their desire to have their
[13]. The sheer volume of individuals, including child’s organs donated, and your colleague pro-
children, awaiting transplantation and their asso- vides this information to the OPO. The first
ciated waitlist mortality stress the importance of examination is consistent with neurologic death.
advocating for and preserving the option for Your colleague expresses concern that continuing
organ donation as a means to save lives. A time- any medical therapies for this child who is likely
line of important historical events related to brain dead is unethical if only for the purposes of
death determination and organ transplantation is organ donation and asks if you are willing to
ERRNVPHGLFRVRUJ
312 A. A. Sochet et al.
Resuscitation and Critical Care Coma and Consciousness Ethics and End-of-Life Care Organ Transplantation
1947 - Beck performs first 1938 – Sugar demonstrates carotid 1954 - Fletcher, Harvard Theologist, 1954 - Murray performs first
successful cardiac defibrillation occlusion results in isoelectric EEG publishes Morals and Medicine, organ transplant
1947- Moerch develops piston 1949 - Moruzzi and Magoun argues for euthanasia based on 1960 - First organ transplant
ventilator for use in OR describe ascending reticular patient autonomy managed with
1950 - Bower and Bennett develop activating system 1957 - Pope Pius XII issues The immunosuppression
positive pressure ventilation 1953 - Rishede reports absent Prolongation of Life, physicians 1962 - Murray performs first
1952 - Ibsen advances respiratory cerebral blood flow in patients with not obligated to offer cadaveric organ transplant
support and establishes first ICU herniation and apnea “extraordinary” measures 1963 - Starzl performs first
1954 – Engstrom introduces volume- 1954 - Fessard publishes 1962 - Ayd publishes The Hopeless liver transplant
cycled ventilator Mechanisms of Nervous Integration Case: Medical and Moral 1963 - Hardy performs first
1955 - Bird develops pressure- and Consciousness Considerations argues for lung transplant
cycled ventilator 1958 - Magoun publishes The withdrawal of care 1963 – Reversal of rejection
1956 - Zoll develops external Waking Brain 1965 - AMA organizes First shown possible using
defibrillator (AC) 1959 - Fischgold describes four National Congress on Medical prednisone and azothiaprine
1958 - Safar develops mouth-to- stages of coma. Stage IV (coma Ethics and Professionalism 1966 - Kelly performs first
mouth respiration carus) has absence of reflexes, 1966 -Williamson publishes Life or pancreas transplant
1959 – First modern ICUs breathing and isoelectric EEG and Death - Whose Decision? 1966 - CIBA Symposium and
established (University of 100% mortality 1966 - Beecher publishes Ethics and Alexandre criteria for declaring
Pittsburgh, Peter Safar, UCLA, Max 1959 - Werthemier and Jouvet Clinical Research death
WeiI) describe “death of the nervous 1967 - Fletcher publishes Moral 1967 - Barnard performs first
1960 - Kouwenhoven develops system” Responsibility: Situation Ethics at heart transplant
closed chest massage 1959 - Mollaret and Gollon describe Work, argues that euthanasia 1968 – Cooley performs heart-
1962 - Lown develops portable “coma depasse” would “harmonize civil law with lung transplantation
external defibrillator (DC) 1963 - Schwab establishes triad of medical morals.”
1966 - First cardiopulmonary criteria for establishing death 1968 - Beecher publishes Ethical
resuscitation guidelines developed 1966 - Plum and Posner publish The Problems Created by the Hopelessly
Diagnosis of Stupor and Coma Unconscious Patient
Fig. 19.1 Parallel developments that converged in the formulation of the concept of neurologic death. (Used with
permission from De Georgia [14] and Elsevier)
assume care of this patient. A do not resuscitate practices present an unacceptable conflict of
(DNR) initiative is established for this child. interest. Closer examination of the issue reveals
that the goals of caring for the child and the facil-
itation of organ donation are often congruent.
Case Discussion: Part 1 A conflict of interest is generally defined as
having three elements: (1) two incompatible
The critical care provider expressed internal con- interests (2) where the possibility of benefiting
flict providing medical care to a child with one’s interest could influence the course of action
impending neurologic death and familial request (3) to the detriment of the other interest. While
to proceed with donation after neurologic deter- there are a number of interests involved in this
mination of death (DND). Pediatric critical care case scenario, the two primary interests are the
is fraught with parallel and, at times, competing patient and the potential organ transplantation
interests that may result in the feeling of conflict recipients. Continued critical care of this patient
for the provider [15–17]. First and foremost, the will serve two separate goals by potentially ben-
physician is obligated to provide medical care to efiting different patients. Continued critical care
the critically ill child. In the case of the dying benefiting donation may also medically benefit
child, ensuring palliative or end-of-life care and the patient by saving their life or at a minimum
supporting a family’s request for donation must not cause harm [15]. Further, the maintenance of
be balanced. Continuing to treat the patient in the a patient in these circumstances to preserve the
face of unclear medical benefit in part for pur- opportunity for donation is required by the law
poses of preserving the potential for organ dona- [4, 18]. The two interests can thus be viewed as in
tion raises the ethical concern of whether such alignment rather than incompatible.
ERRNVPHGLFRVRUJ
1967 - 1968 1975 1978 1981 1994 2007 2011 2013-Present
• First Heart • Karen Ann • Uniform • “Defining Death: • Wijdicks and • President’s • Nakagawa TA et • Several high-
Transplantation Quinlan, a young Determination of Medical, Legal, American Council on al. along with the profile cases,
performed by woman in a Death Act (UDDA) and Ethical Issues Academy of Bioethics American including Aden
Bernard in South persistent provided legal in the Neurology publish publishes Academy of Hailu, Israel
Africa. vegetative state, basis for defining Determination of the first “Controversies in Pediatrics, Society Stinson, Isaac
receives public both neurologic Death” reported standardized the Determination of Critical Care Lopez, and Jahi
• First Heart-Lung
spotlight and and circulatory by the President’s criteria for of Death” Medicine and McMath, continue
transplant
performed by prompts the death, but not Commission determination of addressing Child Neurology to challenge
creation of the specific clinical or neurologic death concerns Society publish existing
Cooley.
President’s diagnositic in adults to regarding the pediatric definitions of
• A definition of Commission for requirements. include: (1) coma terminology guidelines for death.
Irreversible Coma: the Study of or “brain death” and determination of
Report of the Ad Ethical Problems unresponsiveness, revised the neurologic death.
Hoc Committee of in Medicine and (2) absence of concept of death
the Harvard Biomedical and brainstem as the “cessation
Medical School to Behavioral reflexes, and (3) of the
Examine the Research. apnea fundamental vital
ERRNVPHGLFRVRUJ
Definition of Brain • Confirmatory tests work of a living
Death. JAMA. were not organism–the
work of self-
19 Diagnosis of Brain Death and Organ Donation After Circulatory Death
Fig. 19.2 Timeline of events following the 1968 Ad Hoc Harvard Medical School Committee to examine the definition of brain death
313
314 A. A. Sochet et al.
The second element of a conflict focuses on tion rates for organ donation are improved when
the possibility of personal benefit influencing a trained requestors are the primary interface with
course of action. In the case scenario, the physi- families [18]. This does not preclude a collabora-
cian’s ethical duty to the patient is primary but tive approach for authorization with the OPO
not exclusive from other legitimate purposes representative and medical team as may be desir-
(such as supporting the patient’s family or pro- able and effective particularly for pediatric
viding donor management) to the extent both donors [19–21]. The clinician who has built a
interests can be served in a compatible manner. unique bond with a family during the process of
The responsibility for healthcare professionals to providing care to their critically ill and dying
deliver care that might not medically benefit the child is potentially the most ideal individual to
patient is not unique to donation. Examples of discuss this sensitive topic with a grieving family
this include maintaining a patient on artificial [22]. A coordinated plan with the OPO for the
support at the family’s request to allow for family most appropriate approach under the circum-
members to gather and say “goodbye.” This is stances of the case should be agreed upon in
viewed as consistent with and not incompatible advance. Discussions regarding donation should
with good patient care. Physicians manage mul- always occur in a private and respectful manner
tiple interests frequently including personal [23, 24]. If a provider believes a conflict of inter-
financial interest (payor arrangements) and hos- est exists or is uncomfortable caring for a child
pital resource allocation (“right of the last bed”). who will become an organ donor, the institution
There is nothing unique with balancing multiple should provide an alternative, willing provider to
interests when considering a course of patient assume care of the patient. In extreme circum-
care. The treating physician who continues medi- stances, although not ideal, the institution can
cal care of patients with catastrophic brain inju- arrange for patient transfer to a facility where the
ries does not have the potential for personal or family’s wishes can be honored. Pediatric dona-
financial gain. The potential benefit from organ tion is unique because parental authorization is
donation accrues to the donor’s family, recipi- required since the child cannot provide their con-
ents, and society, not the treating physician. sent to be an organ donor. Importantly, an adult
The third element of a conflict is in regard to a (18 years of age or older) can make his or her
disadvantage to other interests. Yet, there is no own legally binding donation prior to death
clear detriment to the patient in these cases. To the through a donor registry or other signed docu-
contrary, there are known benefits to the donor ment (such as a living will) [4]. If the patient is
families’ experience of grief. For families who do 18 years of age or older and has registered as a
not want to pursue donation, providing the oppor- donor, the parents will be approached to discuss
tunity to make that decision does not represent the donation process, but additional authorization
harm. Accordingly, in most cases, the continued will not be requested, and the law does not allow
treatment of a patient to preserve the opportunity for family to override the patient’s donation deci-
for donation does not present an actual conflict, sion [4].
and the treating physician should feel comfortable
that the goals for the patient, family, and potential
transplant recipients are compatible. Therefore, Case Presentation: Part 2
the physician’s ethical duty to the patient is pri-
mary but not exclusive from care that facilitates or Twelve hours after your colleague completes the
optimizes organ donation. first neurologic death examination, you perform
Best practices and federal regulation require the second examination. Your findings are consis-
that only trained requestors approach families for tent with neurologic death. As you update the
donation authorization [15]. In most instances family, they state that they do not believe in brain
OPO staff fill that role, but it may also be hospital death and fully expect medical treatment will
staff trained by the OPO. Data indicate authoriza- continue until their child recovers. The family
ERRNVPHGLFRVRUJ
19 Diagnosis of Brain Death and Organ Donation After Circulatory Death 315
indicates that any attempt to stop medical care at Individual state laws recognize an individual
this time would result in legal ramifications. whose entire brain function has ceased is dead,
but how the cessation of total brain function is
determined is a matter of medical practice, con-
Case Discussion: Part 2 sistent with the UDDA [6]. Once death is
declared, there is no further legal or ethical duty
A recent high-profile case from California exem- to continue medical treatments including
plifies an increasingly common, controversial, mechanical support on a decedent. The law has
and unfortunate circumstance for both a grieving never generally required healthcare providers to
family and the medical community [25]. The provide treatment to deceased patients.
case involved a young girl who suffered a cardiac Exceptions exist in the circumstance of authori-
arrest after a tonsillectomy for obstructive sleep zation for organ donation or in states permitting
apnea. Severe neurologic injury occurred, and religious and moral exceptions to neurologic
she was declared dead by neurologic criteria con- death. Three states have such exceptions, with
sistent with well-established medical and legal New Jersey and New York accepting religious or
standards [10]. The family petitioned the court to moral exceptions and California requiring a “rea-
prevent withdrawal of medical support which the sonably brief period of accommodation” to fami-
court granted. The patient was deceased under lies after declaration of brain death from any
the law, and the hospital was not obligated to per- objection, moral, religious, or otherwise [3, 25].
form any further medical interventions such as a If there are disagreements regarding the determi-
tracheostomy or gastrostomy as requested by the nation or definition of death, the court system
family. Artificial support continued for the may consider a temporary restraining order to
deceased child for 22 days until an agreement allow for a second opinion on the medical deter-
was reached allowing the family to arrange trans- mination of death based on neurologic criteria
fer of this deceased child to a facility in New and assessment of harms.
Jersey. Despite a legal death certificate issued Medical providers work with families pre-
from the state of California, artificial support for paring them for the potential or eventual death
this deceased child continued until rcent cardiac of their child in situations where medical thera-
arrest and asytole. In a similar case, a mother in pies can no longer cure or restore health.
Kentucky petitioned the court to prevent the hos- Physicians are obligated to provide support and
pital from removing the ventilator and feeding guidance to families as they attempt to under-
tube from her child who had been declared dead stand what has happened to their child and face
by neurologic criteria [26]. The mother argued difficult end-of-life decisions. Appropriate
that a child’s parents, not the hospital, has the emotional support for the family should be
right to make decisions about the medical treat- offered, including adequate time to grieve with
ment of the child including withdrawal of artifi- their child after death has occurred. There are
cial support after death determination. The judge no existing guidelines to prescribe a reasonable
ruled because the child was dead in accordance or optimal duration for continuation of care to
with medical standards and the law, no parental accommodate a grieving family after determi-
right survives, and medical therapies were dis- nation of neurologic death. Qualitative studies
continued. These two controversial, high-profile have shown that parental presence at the time of
cases are examples where families disagree with death, provision of adequate information from
the medical team regarding continued medical the healthcare team, and a sympathetic environ-
treatment following death. Notably, these cases ment cultivate the provider-family relationships
open an important discussion about the rights of and end-of-life decision-making and enable
parents regarding continued medical support for healthy grieving [27, 28]. Special attention to
a deceased child when there are no specified legal decoupling or separating the brain death exami-
state exceptions. nations from discussions regarding donation
ERRNVPHGLFRVRUJ
316 A. A. Sochet et al.
could assist the provider in determining the travel for out of town family members.
needs, beliefs, or accommodations for individ- Compassionate support continues for this griev-
ual families. Although grieving families should ing family. At the end of the 24-h period, the
be provided adequate, respectful time and space family realizes their child has died and wishes to
to grieve after determination of neurologic proceed with donation after neurologic death
death, intensive care services are a limited (DND). As the process of donation proceeds, the
resource, and, therefore, the clinician and insti- donor develops hemodynamic instability.
tution must triage beneficence with non-malefi- Telemetry is consistent with ventricular tachy-
cence in postmortem accommodations. cardia, and chest compressions with advanced
These cases demonstrate the importance of a life support are initiated. Some members of the
careful, humanistic, and guided response to medical team are concerned that CPR is being
scenarios where a family requests exception or performed on a child that has an established
accommodation after neurologic determination DNR order.
of death. Ideally, disagreements should be iden-
tified prior to brain death examinations and are
best suited for discussion in a multidisciplinary Case Discussion: Part 3
care conference with healthcare team members
who are trained in end-of-life care. The option Preserving the option of donation following
for a second opinion could be arranged by the neurologic death is essential once authorization
institution to allow for a third-party evaluation for donation has occurred. Treatment of cardiac
or confirmation of determination of death. In arrest for the donor should be viewed as a part
the case of highly confrontational interactions, of active donor management. While ethical con-
risk management, hospital administration, and cerns of violating a previous DNR order may
medical ethics committees can provide insight surface, the donor is no longer a patient once
and guidance with a goal of removing the medi- death is declared; therefore issues related to
cal team from the center of controversy. CPR become irrelevant [29, 30]. Continued sup-
Guidance should be provided to the critical care port of the donor including the use of CPR pre-
team for ongoing medical care such as docu- serves the option of donation helping to fulfill
mentation of vital signs, suctioning, position- the family’s request for their child to be an organ
ing, and other routine ICU procedures for the donor [21]. The use of extracorporeal support
decedent. A well-thought-out plan should be for the brain-dead donor has been considered
established to communicate a consistent mes- and used in extreme cases including dialysis for
sage with the family that their child has died fluid removal and correction of severe electro-
while assisting them with the death of their lyte disturbances [31]. The use of extracorporeal
child in a timely and efficient manner. Finally, support to limit warm ischemic time for dona-
it must be reiterated to a patient’s family that tion after circulatory determination of death
respectful, compassionate care will continue (DCD) donors should be avoided as re-estab-
for their child regardless of the medical out- lishing anterograde circulation in this specific
come or trajectory including palliative medical patient is in direct conflict with circulatory
therapies, withdrawal of life-sustaining medical determination of death [32].
therapies, or authorization for organ donation.
ERRNVPHGLFRVRUJ
19 Diagnosis of Brain Death and Organ Donation After Circulatory Death 317
hold her child in her arms until the heart stops Case Discussion: Part 5
beating. The father asks you if it is still possible
for their child to be a donor under these In this particular scenario, the patient has already
circumstances. died with death determined by neurologic crite-
ria. The team members should be reminded the
DCD process is being facilitated to meet the par-
Case Discussion: Part 4 ents’ request and all procedures performed in
preparation for donation in this case are on a
Organ donation continues to remain a possibility deceased person. However, in other DCD cir-
to fulfill both wishes of this family. Although the cumstances, where the patient has in fact not
more controlled organ procurement process been declared dead by neurologic criteria, the cri-
achieved by DND allows for optimal organ per- teria for determination of circulatory death are
fusion and decreased ischemic time, organ dona- met when examination is consistent with the
tion from a neurologically deceased patient can absence of circulatory function observed for a
be accomplished through a DCD process follow- minimum of 2–5 min depending on the medical
ing circulatory arrest in this scenario. protocol and hospital DCD policy. However,
In 2017, 1879 (18.3%) of deceased donors some question if a patient is truly deceased after
represented cases of DCD of which 138 (7.3%) 2 min of circulatory arrest. Electrical activity of
were children [11]. While the total number of the brain measured by electroencephalography
organs available for recovery from a DCD donor (EEG) has been used as a surrogate to determine
might decrease, the ability to recover and trans- brain function. Potential DCD donors have been
plant DCD organs such as the kidney, liver, and compared to patients suffering from acute cardiac
lungs helps provide the most needed organs for arrest where cessation of cerebral activity mea-
children awaiting a transplant. Transplantation sured by EEG occurs within 15–30 s following
outcomes data appears comparable for organs loss of circulation [35–38]. Withdrawal of medi-
recovered from both types of donors. A recent cal therapies for a potential DCD donor rarely
systematic review for renal transplant outcomes results in immediate circulatory arrest. Rather,
found there was delayed initial graft function the donor experiences a gradual reduction in
from DCD donors compared to DND, but no dif- blood pressure ultimately reaching a point of
ferences in long-term graft function or patient inadequate cerebral perfusion pressure prior to
survival [33]. Similar data for hepatic, renal, and complete circulatory arrest. During the process of
visceral transplant would suggest comparable dying, cerebral blood flow falls below a threshold
graft function and survival in some cases [34]. of adequate perfusion ultimately affecting cere-
bral cellular activity before complete circulatory
arrest occurs. Importantly, when withdrawal of
Case Presentation: Part 5 life-sustaining medical therapies occurs, suitable
comfort measures including administration of
As the OPO prepares for a DCD donation process analgesic and sedative agents are provided to the
by requesting laboratory-antigen testing, organ patient as medical therapies cease.
function and size calculations, and heparinization The UDDA states that death occurs when
of the patient, your team members express concern there is irreversible cessation of circulatory and
that the DCD donor may feel discomfort during respiratory functions, or irreversible cessation of
these assessments or during recovery of organs. all functions of the entire brain, including the
Additionally, they are troubled by performing pro- brain stem [6]. Function describes the fundamen-
cedures and medical therapies that provide no tal purpose of what an organ is supposed to do.
direct benefit to the patient prior to determination Function is therefore different from activities as
of circulatory death despite the fact this child has defined by physiologic properties of cells or
been pronounced dead by neurologic criteria. groups of cells that can be measured by l aboratory
ERRNVPHGLFRVRUJ
318 A. A. Sochet et al.
means [39]. The presence of electrical activity DCD, each component follows different legal
indicating activity of cells does not necessarily frameworks – informed consent for the with-
correlate with function of the organ. In circula- drawal and authorization for the donation
tory death, death is declared following mechani- under the UAGA [43–45].
cal asystole defined as the loss of perfusion After permission to proceed with DCD, ante-
pressure. Perfusion pressure is an indicator of mortem medical therapies are employed that have
contractility of the heart muscle to provide no direct patient benefit and may pose potential
anterograde circulation to the body. Residual harm such as anticoagulation and immunomodu-
electrical activity of the heart may continue after lation therapy. Anticoagulation is administered
mechanical asystole has occurred. In sum, elec- antemortem to reduce the risk of thrombosis from
trical activity is not a measure of organ function altered perfusion during the dying process. Non-
or, in the case of DCD, contractility [32]. beneficence and potential maleficence in ante-
Although two standards are stated separately mortem treatment are outweighed in DCD by the
in the UDDA, the loss of circulation and respira- altruistic goals of the family wishing to donate on
tion and the loss of entire brain function are their child’s behalf [45]. Prohibiting parents from
essentially a single standard based on cessation carrying out donation of organs on their child’s
of brain function. Cessation of brain functions behalf would be an unethical restriction of auton-
follows quickly after the loss of circulatory and omy and to the detriment of society in general and
respiratory function in the absence of medical those awaiting transplantation specifically. Some
intervention. If resuscitative measures such as OPOs may utilize regional perfusion techniques
CPR or other mechanical or pharmacologic sup- using extracorporeal membrane oxygenation after
port restore circulation and respiration before all circulatory death to the thorax or abdomen [39].
brain functions are eliminated by the lack of cir- While perfusion techniques may continue local-
culating oxygenated blood, the patient could not ized circulation and oxygenation to create more
be determined dead. Cessation of circulation and viable organs for procurement, in the context of
respiration can be viewed as a valid means for the DCD donor, the restoration of circulation,
establishing cessation of brain function and even if regional, could be viewed as contrary to
therefore death [32]. circulatory death determination and therefore
The American Academy of Pediatrics and unethical [31, 32, 46].
other major organizations have adopted policy
statements for the DCD process including
identifying and providing guidance to bioethi- Supplemental Discussion
cal concerns [40, 41]. The UAGA provides a
legal framework for gifting of organs after Anencephalic infants have historically not been
death following authorization by either the accepted as donors because of continued brain
individual him- or herself if an adult or by sur- stem activity preventing determination of neuro-
rogate decision-makers in a specified priority logic death. The anencephalic infant donor gained
[42]. Parents are the prioritized decision-mak- national attention in the late 1980s after a success-
ers to authorize donation from a pediatric ful heart transplant occurred following recovery
donor with the option for other surrogate deci- of organs from an anencephalic infant. A case
sion-makers if there are no parents or they are series assessing feasibility of organ donation in
unavailable [4]. Authorization for DCD occurs this population concluded anencephalic infants
prior to the withdrawal of life-sustaining thera- supported with intensive care measures could not
pies (WLST) and declaration of death does not be used as organ donors because they did not
alter the legal requirements. The UAGA per- progress to neurologic death [47]. The authors
mits surrogates to make the donation decision suggested modification to neurologic death, and
at the same time as the decision to WLST [4]. organ donation criteria were required if anenceph-
Although both are required to coordinate a alic infants were to be considered organ donation
ERRNVPHGLFRVRUJ
19 Diagnosis of Brain Death and Organ Donation After Circulatory Death 319
ERRNVPHGLFRVRUJ
320 A. A. Sochet et al.
to Examine the Definition of Brain Death. JAMA. 18. 45 Code of Federal Regulation (CFR) Part 486.
1968;205(6):337–40. Subpart G. Requirements for certification and des-
3. Sarbey B. Definitions of death: brain death and what ignation and conditions for coverage: organ procure-
matters in person. J Law Biosci. 2016;3(3):743–52. ment organizations. Available at: https://www.law.
4. National Conference of Commissioners on Uniform cornell.edu/cfr/text/42/part-486/subpart-G. Accessed
State Laws. Revised Uniform Anatomical Gift Act 8 Mar 2018.
(2006). Available at: http://www.uniformlaws.org/ 19. Rodrique JR, Cornell DL, Howard RJ. Pediatric
shared/docs/anatomical_gift/uaga_final_aug09.pdf. organ donation: what factors most influence par-
Accessed 19 Jan 2018. ents’ donation decisions? Pediatr Crit Care Med.
5. Defining death: medical, legal, and ethical issues in 2008;9(2):180–5.
the determination of death. Presidents Commission 20. Martin DE, Nakagawa TA, Siebelink MJ, et al.
for the Study of Ethical Problems in Medicine and Pediatric deceased donation – a report of the
Biomedical and Behavioral Research 1981. Transplantation Society meeting in Geneva.
6. National Conference of Commissioners on Uniform Transplantation. 2015;99:1403–9.
State Laws. Uniform Determination of Death Act 21. Nakagawa TA, Shemie SD, Dreyden-Palmer K, et al.
(1981). Available at: http://www.uniformlaws.org/ Donation following neurologic and circulatory determi-
shared/docs/determination%20of%20death/udda80. nation of death. Ped Crit Care Med. 2018;(8S Suppl 2):
pdf. Accessed 18 Jan 2018. S26–S32.
7. Wijdicks EF. Determining brain death in adults. 22. ACRE Trial Collaborators. Effect of “collabora-
Neurology. 1995;45(5):1003–11. tive requesting” on consent rate for organ donation:
8. Wijdicks EF, Varelas PN, Gronseth GS, et al. randomized controlled trial (ACRE trial). BMJ.
American Academy of Neurology. Evidence based 2009;8:b3911.
guideline update: determining brain death in adults: 23. Gortmaker SL, Beasley CL, Sheehy E, et al.
report of the Quality Standards Subcommittee of Improving the request process to increase family con-
the American Academy of Neurology. Neurology. sent for donation. J Transpl Coord. 1998;8(4):210–7.
2010;74(23):1911–8. 24. Chandler JA, Connors M, Holland G, Shemie SD.
9. Task force for the determination of brain death in chil- “Effective” requesting: a scoping review of the literature
dren. Guidelines for the determination of brain death on asking families to consent to organ and tissue dona-
in children. Pediatr Neurol. 1987;3(4):242–3. tion. Transplantation. 2017;101(5S Suppl 1):S1–16.
10. Nakagawa TA, Ashwal S, Marthur M, et al. Guidelines 25. Burkle CM, Sharp RR, Wijdicks EF. Why brain death
for the determination of brain death in infants and is considered death and why there should be no confu-
children: an update of the 1987 task force recommen- sion. Neurology. 2014;83(16):1464–9.
dations. Crit Care Med. 2011;39(9):2139–55. 26. Galofaro C. Brain-dead baby taken off life sup-
11. Organ procurement and transplantation network.
port. Cour J. 2014; Available at: https://www.
Available at: http://optn.transplant.hrsa.gov. Accessed courier-journal.com/story/news/crime/2014/07/23/
17 Jan 2018. baby-taken-life-support-judge-rules-mother-
12. Workman JK, Myrick CW, Meyers RL, et al. Pediatric force-hospital-treat-legally-dead-child/13032423/.
organ donation and transplantation. Pediatrics. Accessed 19 Jan 2018.
2013;131(6):e1723–30. 27. Meert KL, Thurston CS, Sarnaik AP. End-of-life deci-
13. Committee on Hospital Care, Section on Surgery, and sion-making and satisfaction with care: parental per-
Section on Critical Care. Policy statement – pediat- spectives. Pediatr Crit Care Med. 2000;1(2):179–85.
ric organ donation and transplantation. Pediatrics. 28. Hoover SM, Bratton SL, Roach E, Olson LM. Parental
2010;125(4):822–8. experiences and recommendations in donation after
14. De Georgia MA. History of brain death as death: 1968 circulatory determination of death. Pediatr Crit Care
to the present. J Crit Care. 2014;29:673–8. Med. 2014;15(2):105–11.
15.
Souter MJ, Blissitt PA, Blosser S, et al. 29. Dalle Ave AL, Gardiner D, Shaw DM. Cardio-
Recommendations for the critical care management of pulmonary resuscitation of brain-dead organ donors: a
the devastating brain injury: prognostication, psycho- literature review and suggestions for practice. Transpl
social and ethical management: a position statement Int. 2016;29(1):12–9.
for healthcare professionals from the Neurocritical 30. Shafer TJ, Cosio C. Cardiopulmonary resuscitation of
Care Society. Neurocrit Care. 2015;23(1):4–13. organ donors. Prog Transplant. 2011;21(4):351–2.
16. Kentish-Barnes N, Duranteau J, Montlahuc C, et al. 31. Dalle Ave AL, Shaw DM, Bernat JL. Ethical issues in
Clinician’s perception and experience of organ the use of extracorporeal membrane oxygenation in
donation from brain dead patients. Crit Care Med. controlled donation after circulatory determination of
2017;45:1489–99. death. Am J Transplant. 2016;16:2293–9.
17. MacDonald SI, Shemie SD. Ethical challenges and 32. Bernat JL, Capron AM, Bleck T, et al. The circula-
the donation physician specialist: a scoping review. tory-respiratory determination of death in organ dona-
Transplantation. 2017;101:S27–40. tion. Crit Care Med. 2010;38:972–9.
ERRNVPHGLFRVRUJ
19 Diagnosis of Brain Death and Organ Donation After Circulatory Death 321
33.
Weiss MJ, Hornby L, Witteman W, Shemie c irculatory determination of death – summary report.
SD. Pediatric donation after circulatory determination Pediatr Crit Care Med. 2017;18(11):1035–46.
of death: a scoping review. Pediatr Crit Care Med. 42. Glazier AK. Principles of gift law and the regulation
2016;17(3):e87–108. of donation. Transpl Int. 2011;24(4):368–72.
34. Morrissey PE, Monaco AP. Donation after circulatory 43. Overby KJ, Weinstein MS, Fiester A. Addressing con-
death: current practices, ongoing challenges, and poten- sent issues in donation after circulatory determination
tial improvements. Transplantation. 2014;97(3):258–64. of death. Am J Bioeth. 2015;15(8):3–9.
35. Pana R, Hornby L, Shemie SD, et al. Time to loss of 44. Marquis D. The impossibility of obtaining informed
brain function and activity during circulatory arrest. J consent to donation after circulatory determination of
Crit Care. 2016;34:77–83. death. Am J Bioeth. 2015;15(8):25–7.
36. Rady MY, Verheijde JL. Neuroscience and aware- 45. Brierley J, Shad D. Premortem interventions in dying
ness in the dying human brain: implications for organ children to optimize organ donation: an ethical analy-
donation. J Crit Care. 2016;34:121–3. sis. J Med Ethics. 2016;42(7):424–8.
37.
Norton L, Gibson RM, Gofton T, et al. 46. Dalle Ave AL, Shaw DM, Gardiner D. Extracorporeal
Electroencephalographic recording during withdrawal membrane oxygenation (ECMO) assisted cardiopul-
of life-sustaining therapy until 30 minutes after decla- monary resuscitation or uncontrolled donation after
ration of death. Can J Neurol Sci. 2017;44:139–45. circulatory determination of death following out-of-
38. Shapey IM, Mulesan P. Regional perfusion by extra- hospital refractory cardiac arrest – an ethical analy-
corporeal membrane oxygenation of abdominal sis of an unresolved clinical dilemma. Resuscitation.
organs from donors after circulatory death: a system- 2015;108:87–94.
atic review. Liver Transpl. 2013;19:1292–303. 47. Peabody JL, Emery JR, Ashwal S. Experience with
39. Shemie SD, Hornby L, Baker A, et al. International anencephalic infants as prospective organ donors. N
guideline development for the determination of death. Engl J Med. 1989;321:344–50.
Intensive Care Med. 2014;40(6):788–97. 48. American Academy of Pediatrics, Committee on
40. American Academy of Pediatrics – Committee on Bioethics. Infants with anencephaly as organ sources:
Bioethics. Ethical controversies in organ donation ethical considerations. Pediatrics. 1992;89:1116–9.
after circulatory death. Pediatrics. 2010;125:822–8. 49.
Nakagawa TA, Zollinger C, Chao J, et al.
41. Weiss MJ, Hornby L, Rochwerg B, et al. Canadian Anencephalic infants as organ donors after circulatory
guidelines for controlled pediatric donation after death. Transplantation. 2017;101:S60.
ERRNVPHGLFRVRUJ
Therapeutic Hypothermia
in the Pediatric ICU 20
Jessica S. Wallisch and Ericka L. Fink
ERRNVPHGLFRVRUJ
324 J. S. Wallisch and E. L. Fink
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 325
Table 20.1 Key pediatric TBI therapeutic hypothermia randomized controlled trials
TH
target Control Timing
temp temp of TH Duration of Rate of
RCT (°C) (°C) initiation TH rewarming Outcome results Comments
Adelson 32–33 36.5– <6 h 48 h 1.0 °C every Trend toward First multicenter trial
(2005) 37.5 from 3–4 h improved survival in pediatric TBI
injury with TH (mortality Phase II trial – not
8% vs 16%, powered to detect
p = 0.44) differences in
Significantly lower survival or outcome
ICP decreased and Low multicenter
less time (%) enrollment rate (22%
greater than of children with
20 mmHg within severe TBI)
the first 24 h with 65–70% of patients
TH had a temperature
No difference in protocol deviation
rates of (average: >2 °C per
coagulopathy, patient)
arrhythmia, or Rebound intracranial
infection hypertension after
Lower rate of rewarming in TH
post-traumatic group
seizures with TH Post hoc analysis
found no difference
in functional or
neurocognitive
outcomes by
treatment group
Hutchinson 32.5 37.0 < 8 h 24 h 0.5 °C every Higher mortality in Mean time to cooling
(2008) from 2 h TH group (21% vs 6.3 ± 2.3 h after
injury 12%, p = 0.06); injury and time to
adjusted HR, 2.36 target temp
(95% CI, 3.9 ± 2.6 h
1.04–5.37) 97% of norm patients
Trend toward higher adhered to target
rate of unfavorable temp
outcome (PCPC) at Significantly more
6 months with TH normothermia
(31% vs 22%, patients received
p = 0.14); adjusted hypertonic saline
OR, 2.33 (95% CI, (46% vs 31%)
0.92–5.93) More hypotension
Lower ICP during and lower CPP in TH
cooling but higher patients during
during rewarming rewarming
with TH (p < 0.001)
(continued)
ERRNVPHGLFRVRUJ
326 J. S. Wallisch and E. L. Fink
Table 20.1 (continued)
TH
target Control Timing
temp temp of TH Duration of Rate of
RCT (°C) (°C) initiation TH rewarming Outcome results Comments
Adelson 32–33 36.5– < 6 h 48–72 h 0.5–1.0 °C Trend to increased Early termination for
(2013) 37.5 from (additional every 12–24 h mortality at futility after 77
injury 24 h 3 months (15% TH patients enrolled
maintained if vs 5% norm, 75% of patients
ICP was p = 0.15) enrolled at three sites
elevated at No difference in (out of 15
48 h) functional outcome participating centers)
(GOS and GOS-E Mean time to cooling
Peds) at 3 months 5.1 ± 0.9 h after
by treatment group injury and time to
target temp 3.2 h)
Higher rate of
decompressive
craniectomy in norm
(45%) vs TH (18%),
p = 0.02
Beca (2015) 32–33 36–37 <6 h Minimum of ≤0.5 °C every No difference in 30% of norm patients
from 72 h 3 h with mortality (13% TH had temperature
injury control of vs 4% norm, elevations >38 °C
hypotension p = 0.34) No differences in
and ICP No difference in MAP and CPP
functional outcome during cooling or
MAP, ICP, or CPP
during rewarming
but more episodes of
hypotension in TH
group during
rewarming (17% vs
0%, p = 0.05)
Abbreviations: CPP, cerebral perfusion pressure; GOS, Glasgow Outcome Scale; GOS-E, Glasgow Outcome
Scale-Extended; ICP, intracranial pressure; PCPC, pediatric cerebral performance category; RCT, randomized con-
trolled trial; TH, therapeutic hypothermia; TBI, traumatic brain injury
studies discussed above due to delays in care and However after establishing that there has been no
challenges with consent. These infants and tod- expansion of the intracranial hemorrhage with
dler-aged children, like the patient described in the repeat imaging, we would suggest consideration
first case study, frequently have injuries com- of TH as a third tiered therapy if refractory intra-
pounded by hypoxic-ischemic injury [42]. cranial hypertension persisted despite maximal
Ongoing comparative effectiveness trials may help medical management (e.g., sedation, neuromuscu-
to bring more clarity on these topics [43], and new lar blockade, hyperosmolar therapy, mild hyper-
pediatric TBI guidelines are expected in 2018. ventilation, treatment of seizures, and barbiturate
coma) in the absence of the option for or while
awaiting surgical intervention such as drainage or
Case Scenario 1 Considerations diversion of cerebrospinal fluid or decompressive
craniectomy (Fig. 20.1). While there is no evi-
As the literature suggests, we would not advocate dence to support the tempo at which to add thera-
for prophylactic initiation of therapeutic hypother- pies, we advocate for aggressive escalation to
mia for this patient to affect a favorable outcome. target ICP < 20 mmHg.
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 327
Fig. 20.1 Targeted
temperature
management in pediatric SevereTBI
severe TBI. Controlled
normothermia is
recommended as part of
targeted temperature
management in all
patients who sustain a Targeted Temperature
severe TBI (GCS ≤ 8). Management
Patients with refractory
intracranial hypertension
(unresponsive to
maximal medical Controlled Normothermia
management) may • Target: 36.5-37.0ºC
benefit from improved • Initiation: immediately
ICP control with • Duration: up to 5 days*
therapeutic hypothermia,
while consideration is
given for decompressive
craniectomy or in cases
deemed not to be Refractory
surgical candidates. intracranial
hypertension?
Abbreviations: TBI, No Yes
traumatic brain injury;
GCS, Glasgow Coma Continue Controlled
Scale; ICP, intracranial Normothermia Consider Therapeutic Hypothermia
pressure • Target: 32-34ºC*
• Duration: up to 48 hours*
• Rewarming: < 0.5ºC per hour*
* Suggested, determine clinically
Case Scenario 2
Key Points
• TTM should be used to prevent fever in
the early post-TBI period.
• Prophylactic TH has not been shown to A 15-year-old male who collapsed while
improve outcomes after severe pediatric playing basketball received bystander car-
TBI. diopulmonary resuscitation (CPR). An
• TH has a role in treating refractory automated external defibrillator (AED)
intracranial hypertension in children placed by paramedics noted ventricular
with severe TBI, with a cautious tachycardia (VT) and provided two defibril-
rewarming protocol. lation shocks, resulting in return of sponta-
neous circulation (ROSC) after 15 min of
CPR. He presented to your ICU comatose
(withdrawal to pain without localization).
What is the role if any of TH for this patient?
ERRNVPHGLFRVRUJ
328 J. S. Wallisch and E. L. Fink
se of TTM for Hypoxic-Ischemic
U respectively, p = 0.04), but this difference did not
Injury meet the prespecified secondary outcome thresh-
old for statistical significance. Subgroup analyses
TTM aimed at controlled normothermia has been were underpowered to make robust conclusions,
proven as an effective neuroprotective strategy including for arrests associated with shockable vs
and should be incorporated into clinical practice non-shockable rhythms.
[31]. Therefore, the use of TTM for this indica- Next, the THAPCA in-hospital cardiac arrest
tion is also not controversial and will not be dis- (IHCA) trial evaluated 257 children utilizing the
cussed further in this chapter. same treatment protocol as the OHCA trial. The
TH has had great success and implementation IHCA study was terminated early prior to reach-
into clinical care for neonates with perinatal ing full enrollment based on interim analysis that
hypoxic-ischemic encephalopathy (HIE) [44]. revealed lack of efficacy in demonstrating sur-
TH also provides a 30% survival benefit and vival with favorable functional outcome at 1 year
increased likelihood of favorable neurological (36% for TH vs 39% for normothermia, p = 0.63)
outcome in the most recent Cochrane review [51]. Critiques of this study noted a relatively
meta-analysis for adults surviving out-of-hospital short median “no-flow” time (period from car-
cardiac arrest (OHCA) due to ventricular fibrilla- diac arrest to initiation of CPR) paired with a
tion, but not for non-cardiac etiologies or in-hos- long duration of cardiopulmonary resuscitation,
pital cardiac arrest events [45]. which may result in a less severe neurologic
Evidence has been less supportive of TH in injury that is less modifiable by TH [52, 53]. A
children after cardiac arrest. First, observational single-center pilot RCT just published compared
non-randomized studies did not detect differ- serum brain biomarker profiles to 24 vs 72 h of
ences in outcome with the use of TH [46–50]. TH (33 °C) in children comatose after resuscita-
While these reports did not collectively find a tion from cardiac arrest [54]. Serum neuron-spe-
benefit, they showed that the general use of TH cific enolase (NSE) and S100b concentrations
was unprotocolized and was more often provided had previously been shown to be associated with
for children with the worst prognoses. These lim- patient outcome after cardiac arrest [55, 56].
itations supported the need for an RCT in this Here, serum S100b and NSE were increased
population. (worse) at specific time points within the first
To date, three RCTs of TH after cardiac arrest week post-arrest in patients only receiving 24 h
have been completed in the pediatric population of TH. In multivariate analysis, there was an
(Table 20.2). The Therapeutic Hypothermia After association of shorter TH duration and higher
Pediatric Cardiac Arrest (THAPCA) OHCA (worse) serum S100b at 1 week post-arrest. This
study was a multicenter RCT of 260 infants and study highlighted the need for additional investi-
children who remained comatose after cardiac gation into the optimal duration of TH as well as
arrest [49]. Patients were randomized to TTM the potential utility for biomarkers in measuring
(36.8 °C for 120 h) vs TH (33 °C for 48 h fol- outcome and patient response to therapy.
lowed by slow rewarming over 16 h to 36.8 which The 2015 International Consensus on
was maintained for the duration of 120 h). The Cardiopulmonary Resuscitation and Emergency
primary outcome of survival with good func- Cardiovascular Care (ILCOR) Pediatric Basic
tional outcome at 1 year was not statistically dif- Life Support and Pediatric Advanced Life
ferent among treatment groups, 20% for patients Support (PALS) Treatment Recommendations
receiving TH versus 12% for patients receiving suggest that TTM be used as part of the post-
controlled normothermia (p = 0.14). Twelve- cardiac arrest care for infants and children fol-
month survival was 38% vs 29%, respectively lowing out-of-hospital cardiac arrest (OHCA)
(p = 0.13). Intriguingly, survival duration was [57]. The guidelines parallel findings in
longer for the TH vs normothermia group (mean THAPCA, citing a range of 32–34 °C for TH
survival from time of injury 149 vs 119 days, for 48 h or 36–37.5 °C for controlled normo-
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 329
Table 20.2 Key pediatric cardiac arrest therapeutic hypothermia randomized controlled trials
Norm
TH target control Timing of Duration Rate of
RCT temp (°C) (°C) initiation of TH rewarming Results Comments
THAPCA- 32–34 °C 36– <6 h after 48 h ≥16 h No difference in Median time to cooling
OHCA 37.5 °C ROSC followed survivors with good 5.9 h and time to target
Moler by Norm neurobehavioral temp 2.6 h
(2015) for outcome at 12 months TH group more likely to
remainder with TH (20% vs develop hypokalemia
of 120 h 12%, p = 0.14) (23% vs 14%, p = 0.04)
Better survival and thrombocytopenia
analysis trend for TH (10% vs 1%, p = 0.001)
group (149 ± 14 vs but less likely to receive
119 ± 14 days, RRT (2% vs 7%,
p = 0.04); 12-month p = 0.03)
survival not
statistically different
between groups (38%
vs 29%, p = 0.13)
THAPCA- 32–34 °C 36– <6 h after 48 h ≥16 h No difference in Early termination for
IHCA 37.5 °C ROSC followed survival with good futility after 329
Moler by Norm neurobehavioral patients enrolled
(2017) for outcome at 12 Median time to cooling
remainder months (TH 36% vs 4.9 h and time to target
of 120 h Norm 39%) temp 2.1 h
No difference in Median time from
12-month survival cardiac arrest to CPR
(TH 49% vs Norm 0 min; median duration
46%) of CPR 22 min
Fink 33 ± 1 °C N/A At 24 h vs 0.5 °C Nonsignificant Included patients with
(2018) clinician 72 h every 4 h decreased mortality both IHCA and OHCA
discretion followed in 72 h group (24% that had TH initiated by
by Norm vs 47%, p = 0.3) clinical team
for No difference in rate Partly concurrent with
remainder of unfavorable THAPCA (prioritized
of 120 h outcome (65% for enrollment) included
24 h vs 59% for 72 h) children that did not
Serum NSE, MBP qualify for THAPCA
and S100b all First study to include
elevated after CA biomarkers
NSE and S100b 24 h group had
significantly temperature elevations
increased at day 7 in >38 °C during Norm
24 h as compared to period (after rewarming)
72 h group 72 h group had more
Shorter duration of females, more patients
TH remained with cardiac etiology for
independently arrest (18% vs 6%), and
associated with trend to longer duration
elevated S100b at of CPR (25.5 min vs
7 days in multivariate 17.0 min, p = 0.10)
analysis
Abbreviations: IHCA, in-hospital cardiac arrest; MBP, myelin basic protein; NSE, neuron-specific enolase; Norm,
normothermia; OHCA, out-of-hospital cardiac arrest; ROSC, return of spontaneous circulation; RCT, randomized con-
trolled trial; RRT, renal replacement therapy; S100b, S100 calcium-binding protein B; TH, therapeutic hypothermia
ERRNVPHGLFRVRUJ
330 J. S. Wallisch and E. L. Fink
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 331
Fig. 20.2 Targeted
temperature CPR
management in pediatric
hypoxic-ischemic injury.
Fever avoidance is
recommended in all
patients following ROSC
hypoxic-ischemic injury.
Targeted temperature
management should be Prevent
employed for patients hyperthermia
who remain comatose
following hypoxic- Coma?
ischemic injury with
consideration given to Yes
both controlled
normothermia and Targeted Temperature
therapeutic hypothermia. Management
Abbreviations: CPR,
cardiopulmonary
resuscitation; ROSC,
return of spontaneous
circulation; TTM, Controlled Normothermia Therapeutic Hypothermia
targeted temperature • Target: 36.0-37.5ºC • Target: 32-34ºC
management; ICP, • Duration: 48 h followed by
• Duration: up to 5 days
intracranial pressure controlled normothermia for 72 h
• Rewarming: Slowly, over ≥ 16 h
ERRNVPHGLFRVRUJ
332 J. S. Wallisch and E. L. Fink
estimated using the rectal or esophageal route; ous EEG monitoring [87]. Furthermore, there are
peripheral (e.g., oral, axillary, or temporal) moni- clear changes in cardiovascular physiology during
toring has been shown to be inferior [71]. The rewarming that result in shifts in blood pressure
neurocritical care guidelines recommend contin- and heart rate resulting from redistribution of
uous central temperature monitoring, preferably blood flow secondary to peripheral vasodilation
with an esophageal or bladder temperature probe [88]. Patients are at increased risk of hyperkale-
[72]. In practice, patients may require two central mia during rewarming due to cellular shifts and
probes, one each for monitoring and feedback of iatrogenic replacement of hypokalemia during
the temperature-modulating device. TH [89]. Altogether, the rewarming phase is one
Clinicians have a menu of options for tech- that necessitates close monitoring with gradual
niques to achieve TTM, but limited data exist com- and controlled temperature modulation to prevent
paring their performance characteristics. Among overshooting or rapid temperature correction.
the most readily available are surface application
of ice, fans, air-cooling blankets, and intravascular
infusions of cold saline (4 °C). However, advanced Complications
temperature-modulating devices have been found
to achieve target temperature more rapidly and Shivering
reliably [73–75]. Several studies which have
directly compared modalities of cooling support Shivering is a frequent complication that may
the use of endovascular cooling and surface gel impede efficacy of TTM application as shivering
pads for rapid induction and more reliable mainte- increases metabolic demand and cerebral energy
nance of TH [74–77]. The neurocritical care consumption [90]. A first line of prevention or treat-
guidelines advocate the use of a servo-controlled ment includes the non-pharmacologic intervention
body wrap surface-cooling device in neonates of surface counter warming with an air-circulating
with hypoxic-ischemic encephalopathy and the blanket; however, in practice, acetaminophen, seda-
use of intravascular catheters or surface-cooling tion, and neuromuscular blockade are routinely
gel pads in older populations [72]. Endovascular added in a stepwise approach [72, 91–94].
cooling devices are only FDA-approved for use in
adult patients at this time. Novel intranasal and
esophageal cooling devices are currently being Cardiovascular Complications
studied but have not yet reached the same level of
support in the literature [78, 79]. TH can have a number of effects on cardiovascu-
The rewarming period is critical for the emer- lar physiology, including bradycardia and arrhyth-
gence of complications and modulation of the mias. These effects can be particularly detrimental
beneficial effect of TH. Preclinical studies have to patients who may have compromised function
found that a rewarming rate of 0.5–1 °C/h did not post-CA [95–97]. Due to higher systemic vascu-
affect the benefit of TH, but faster rewarming lar resistance index in patients who are cooled to
negated this effect [80, 81]. Rapid rewarming is 33 °C vs 36 °C, cardiac index may be adversely
tied to increases in ICP and changes in cerebro- affected and, combined with the associated brady-
vascular autoregulation [82–84]. The rewarming cardia, can result in decreased organ and tissue
period has also been associated with the emer- perfusion, hyperlactemia, and increased require-
gence of interictal epileptiform discharges, sei- ments for vasoactive support [95, 98].
zures, and status epilepticus [85, 86]. Indeed, the
American Clinical Neurophysiology Society con-
sensus statement on continuous EEG in critically Metabolic Derangements
ill adults and children cites acute supratentorial
brain injury with altered mental status including In addition to hyperlactemia and metabolic aci-
during and after TH as an indication for continu- dosis, serum potassium levels decrease during
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 333
ERRNVPHGLFRVRUJ
334 J. S. Wallisch and E. L. Fink
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 335
20. Kinoshita K, Chatzipanteli iK, Vitarbo E, Truettner 32. Puccio AM, Fischer MR, Jankowitz BT, Yonas H,
JS, Alonso OF, Dietrich WD. Interleukin-1beta Darby JM, Okonkwo DO. Induced normothermia
messenger ribonucleic acid and protein levels after attenuates intracranial hypertension and reduces
fluid-percussion brain injury in rats: importance of fever burden after severe traumatic brain injury.
injury severity and brain temperature. Neurosurgery. Neurocrit Care. 2009;11(1):82–7.
2002;51(1):195–203; discussion 203. 33. Rincon F, Hunter K, Schorr C, Dellinger RP, Zanotti-
21. Lu XC, Shear DA, Deng-Bryant Y, Leung LY, Wei Cavazzoni S. The epidemiology of spontaneous
G, Chen Z, Tortella FC. Comprehensive evaluation fever and hypothermia on admission of brain injury
of neuroprotection achieved by extended selective patients to intensive care units: a multicenter cohort
brain cooling therapy in a rat model of penetrating study. J Neurosurg. 2014;121(4):950–60.
ballistic-like brain injury. Ther Hypothermia Temp 34. Suz P, Vavilala MS, Souter M, Muangman S, Lam
Manag. 2016;6(1):30–9. AM. Clinical features of fever associated with poor
22. Rocha-Ferreira E, Kelen D, Faulkner S, Broad KD, outcome in severe pediatric traumatic brain injury. J
Chandrasekaran M, Kerenyi A, Kato T, Bainbridge Neurosurg Anesthesiol. 2006;18(1):5–10.
A, Golay X, Sullivan M, et al. Systemic pro- 35. Marion DW, Obrist WD, Carlier PM, Penrod LE,
inflammatory cytokine status following therapeutic Darby JM. The use of moderate therapeutic hypo-
hypothermia in a piglet hypoxia-ischemia model. J thermia for patients with severe head injuries: a pre-
Neuroinflammation. 2017;14(1):44. liminary report. J Neurosurg. 1993;79(3):354–62.
23. Sterz F, Safar P, Tisherman S, Radovsky A, 36. Marion DW, Penrod LE, Kelsey SF, Obrist WD,
Kuboyama K, Oku K. Mild hypothermic cardio- Kochanek PM, Palmer AM, Wisniewski SR,
pulmonary resuscitation improves outcome after DeKosky ST. Treatment of traumatic brain injury
prolonged cardiac arrest in dogs. Crit Care Med. with moderate hypothermia. N Engl J Med.
1991;19(3):379–89. 1997;336(8):540–6.
24. Vitarbo EA, Chatzipanteli K, Kinoshita K, Truettner 37. Adelson PD, Ragheb J, Kanev P, Brockmeyer
JS, Alonso OF, Dietrich WD. Tumor necrosis factor D, Beers SR, Brown SD, Cassidy LD, Chang Y,
alpha expression and protein levels after fluid per- Levin H. Phase II clinical trial of moderate hypo-
cussion injury in rats: the effect of injury severity and thermia after severe traumatic brain injury in chil-
brain temperature. Neurosurgery. 2004;55(2):416– dren. Neurosurgery. 2005;56(4):740–54; discussion
24; discussion 424–415. 740–754.
25. Dietrich WD, Bramlett HM. Therapeutic hypother- 38. Adelson PD, Wisniewski SR, Beca J, Brown
mia and targeted temperature management in trau- SD, Bell M, Muizelaar JP, Okada P, Beers SR,
matic brain injury: clinical challenges for successful Balasubramani GK, Hirtz D, et al. Comparison
translation. Brain Res. 2016;1640(Pt A):94–103. of hypothermia and normothermia after severe
26. Bao L, Chen D, Ding L, Ling W, Xu F. Fever burden traumatic brain injury in children (Cool Kids):
is an independent predictor for prognosis of trau- a phase 3, randomised controlled trial. Lancet
matic brain injury. PLoS One. 2014;9(3):e90956. Neurol. 2013;12(6):546–53.
27. Grossestreuer AV, Gaieski DF, Donnino MW, Wiebe 39. Beca J, McSharry B, Erickson S, Yung M, Schibler
DJ, Abella BS. Magnitude of temperature eleva- A, Slater A, Wilkins B, Singhal A, Williams G,
tion is associated with neurologic and survival out- Sherring C, et al. Hypothermia for traumatic brain
comes in resuscitated cardiac arrest patients with injury in children-a phase II randomized controlled
postrewarming pyrexia. J Crit Care. 2017;38:78–83. trial. Crit Care Med. 2015;43(7):1458–66.
28. Heindl UT, Laub MC. Outcome of persistent veg- 40. Hutchison JS, Ward RE, Lacroix J, Hebert PC,
etative state following hypoxic or traumatic brain Barnes MA, Bohn DJ, Dirks PB, Doucette S,
injury in children and adolescents. Neuropediatrics. Fergusson D, Gottesman R, et al. Hypothermia
1996;27(2):94–100. therapy after traumatic brain injury in children.
29. Hifumi T, Kuroda Y, Kawakita K, Yamashita S, Oda N Engl J Med. 2008;358(23):2447–56.
Y, Dohi K, Maekawa T. Fever control management 41. Kochanek PM, Carney N, Adelson PD, Ashwal S,
is preferable to mild therapeutic hypothermia in trau- Bell MJ, Bratton S, Carson S, Chesnut RM, Ghajar
matic brain injury patients with abbreviated injury J, Goldstein B, et al. Guidelines for the acute medi-
scale 3–4: a multi-center, randomized controlled cal management of severe traumatic brain injury in
trial. J Neurotrauma. 2016;33(11):1047–53. infants, children, and adolescents. Pediatr Crit Care
30. Hinson HE, Rowell S, Morris C, Lin AL, Schreiber Med. 2012;13(Suppl 1):S1–82. Second edition.
MA. Early fever after trauma: does it matter? J 42. Berger RP, Adelson PD, Richichi R, Kochanek
Trauma Acute Care Surg. 2018;84(1):19–24. PM. Serum biomarkers after traumatic and hypox-
31. Nielsen N, Wetterslev J, Cronberg T, Erlinge emic brain injuries: insight into the biochemical
D, Gasche Y, Hassager C, Horn J, Hovdenes J, response of the pediatric brain to inflicted brain
Kjaergaard J, Kuiper M, et al. Targeted tem- injury. Dev Neurosci. 2006;28(4–5):327–35.
perature management at 33 degrees C versus 36 43. Bell MJ, Adelson PD, Hutchison JS, Kochanek PM,
degrees C after cardiac arrest. N Engl J Med. Tasker RC, Vavilala MS, Beers SR, Fabio A, Kelsey
2013;369(23):2197–206. SF, Wisniewski SR, et al. Differences in medical
ERRNVPHGLFRVRUJ
336 J. S. Wallisch and E. L. Fink
therapy goals for children with severe traumatic outcome after pediatric cardiac arrest. Pediatr Crit
brain injury-an international study. Pediatr Crit Care Care Med. 2009;10(4):479–90.
Med. 2013;14(8):811–8. 57. Maconochie IK, de Caen AR, Aickin R, Atkins DL,
44. Jacobs SE, Berg M, Hunt R, Tarnow-Mordi WO, Biarent D, Guerguerian AM, Kleinman ME, Kloeck
Inder TE, Davis PG. Cooling for newborns with DA, Meaney PA, Nadkarni VM, et al. Part 6: pediatric
hypoxic ischaemic encephalopathy. Cochrane basic life support and pediatric advanced life support:
Database Syst Rev. 2013;1:CD003311. 2015 International Consensus on Cardiopulmonary
45. Arrich J, Holzer M, Havel C, Mullner M, Herkner Resuscitation and Emergency Cardiovascular
H. Hypothermia for neuroprotection in adults after Care Science with Treatment Recommendations.
cardiopulmonary resuscitation. Cochrane Database Resuscitation. 2015;95:e147–68.
Syst Rev. 2016;2:CD004128. 58. Geocadin RG, Wijdicks E, Armstrong MJ, Damian
46. Doherty DR, Parshuram CS, Gaboury I, Hoskote A, M, Mayer SA, Ornato JP, Rabinstein A, Suarez JI,
Lacroix J, Tucci M, Joffe A, Choong K, Farrell R, Torbey MT, Dubinsky RM, et al. Practice guideline
Bohn DJ, et al. Hypothermia therapy after pediatric summary: reducing brain injury following cardio-
cardiac arrest. Circulation. 2009;119(11):1492–500. pulmonary resuscitation: report of the Guideline
47. Fink EL, Clark RS, Kochanek PM, Bell MJ, Watson Development, Dissemination, and Implementation
RS. A tertiary care center’s experience with thera- Subcommittee of the American Academy of
peutic hypothermia after pediatric cardiac arrest. Neurology. Neurology. 2017;88(22):2141–9.
Pediatr Crit Care Med. 2010;11(1):66–74. 59. Guilliams K, Rosen M, Buttram S, Zempel J,
48. Lin JJ, Hsia SH, Wang HS, Chiang MC, Lin Pineda J, Miller B, Shoykhet M. Hypothermia for
KL. Therapeutic hypothermia associated with pediatric refractory status epilepticus. Epilepsia.
increased survival after resuscitation in children. 2013;54(9):1586–94.
Pediatr Neurol. 2013;48(4):285–90. 60. Lin JJ, Lin KL, Hsia SH, Wang HS, Group
49. Moler FW, Silverstein FS, Holubkov R, Slomine CS. Therapeutic hypothermia for febrile infection-
BS, Christensen JR, Nadkarni VM, Meert KL, related epilepsy syndrome in two patients. Pediatr
Clark AE, Browning B, Pemberton VL, et al. Neurol. 2012;47(6):448–50.
Therapeutic hypothermia after out-of-hospi- 61. Orlowski JP, Erenberg G, Lueders H, Cruse
tal cardiac arrest in children. N Engl J Med. RP. Hypothermia and barbiturate coma for
2015;372(20):1898–908. refractory status epilepticus. Crit Care Med.
50. Scholefield BR, Morris KP, Duncan HP, Perkins GD, 1984;12(4):367–72.
Gosney J, Skone R, Sanders V, Gao F. Evolution, 62. Buttram SD, Au AK, Koch J, Lidsky K, McBain K,
safety and efficacy of targeted temperature manage- O'Brien N, Zielinski BA, Bell MJ. Feasibility study
ment after pediatric cardiac arrest. Resuscitation. evaluating therapeutic hypothermia for refractory
2015;92:19–25. status epilepticus in children. Ther Hypothermia
51. Moler FW, Silverstein FS, Holubkov R, Slomine Temp Manag. 2015;5(4):198–202.
BS, Christensen JR, Nadkarni VM, Meert 63. Legriel S, Lemiale V, Schenck M, Chelly J,
KL, Browning B, Pemberton VL, Page K, Laurent V, Daviaud F, Srairi M, Hamdi A, Geri G,
et al. Therapeutic hypothermia after in-hospi- Rossignol T, et al. Hypothermia for neuroprotec-
tal cardiac arrest in children. N Engl J Med. tion in convulsive status epilepticus. N Engl J Med.
2017;376(4):318–29. 2016;375(25):2457–67.
52. Arrich J, Herkner H. Hypothermia after in-hos- 64. Montaldo P, Oliveira V, Lally PJ, Chaban B, Atreja
pital cardiac arrest in children. N Engl J Med. G, Kirmi O, Thayyil S. Therapeutic hypothermia in
2017;376(17):1695. neonatal cervical spine injury. Arch Dis Child Fetal
53. Wu MJ, Guo J, Yu H. Hypothermia after in-hos- Neonatal Ed. 2016;101(5):F468.
pital cardiac arrest in children. N Engl J Med. 65. Tracy B, Armola R, Micham J. The “cold cord”: a
2017;376(17):1695. review of therapeutic hypothermia for traumatic spi-
54. Fink EL, Clark RSB, Berger RP, Fabio A, Angus nal cord injuries. Am J Crit Care. 2015;24(6):540–3.
DC, Watson RS, Gianakas JJ, Panigrahy A, Callaway 66. Systemic Hypothermia in Acute Cervical Spinal
CW, Bell MJ, et al. 24 vs. 72 hours of hypothermia Cord Injury. https://ClinicalTrials.gov/show/
for pediatric cardiac arrest: A pilot, randomized con- NCT02991690
trolled trial. Resuscitation. 2018;126:14–20. 67. Cariou A, Payen JF, Asehnoune K, Audibert G,
55. Fink EL, Berger RP, Clark RS, Watson RS, Angus Botte A, Brissaud O, Debaty G, Deltour S, Deye N,
DC, Richichi R, Panigrahy A, Callaway CW, Bell Engrand N, et al. Targeted temperature management
MJ, Kochanek PM. Serum biomarkers of brain in the ICU: guidelines from a French expert panel.
injury to classify outcome after pediatric cardiac Anaesth Crit Care Pain Med. 2018;80:55–61.
arrest*. Crit Care Med. 2014;42(3):664–74. 68. Hong JM, Lee JS, Song HJ, Jeong HS, Choi HA,
56. Topjian AA, Lin R, Morris MC, Ichord R, Drott H, Lee K. Therapeutic hypothermia after recanaliza-
Bayer CR, Helfaer MA, Nadkarni V. Neuron-specific tion in patients with acute ischemic stroke. Stroke.
enolase and S-100B are associated with neurologic 2014;45(1):134–40.
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 337
69. Geurts M, Petersson J, Brizzi M, Olsson-Hau S, monary resuscitation in a rat model. Crit Care Med.
Luijckx GJ, Algra A, Dippel DW, Kappelle LJ, van 2014;42(2):e106–13.
der Worp HB. COOLIST (Cooling for Ischemic 81. Wang B, Armstrong JS, Lee JH, Bhalala U,
Stroke Trial): a multicenter, open, randomized, Kulikowicz E, Zhang H, Reyes M, Moy N, Spicer
phase II, clinical trial. Stroke. 2017;48(1):219–21. D, Zhu J, et al. Rewarming from therapeutic
70. Lyden P, Hemmen T, Grotta J, Rapp K, Ernstrom hypothermia induces cortical neuron apoptosis
K, Rzesiewicz T, Parker S, Concha M, Hussain in a swine model of neonatal hypoxic-ischemic
S, Agarwal S, et al. Results of the ICTuS 2 Trial encephalopathy. J Cereb Blood Flow Metab.
(Intravascular Cooling in the Treatment of Stroke 2). 2015;35(5):781–93.
Stroke. 2016;47(12):2888–95. 82. Koizumi H, Suehiro E, Fujiyama Y, Yoneda H,
71. Niven DJ, Gaudet JE, Laupland KB, Mrklas KJ, Ishihara H, Nomura S, Fujii M, Suzuki M. Effects
Roberts DJ, Stelfox HT. Accuracy of peripheral of brain temperature on cerebrovascular auto-
thermometers for estimating temperature: a sys- regulation during the acute stage of severe trau-
tematic review and meta-analysis. Ann Intern Med. matic brain injury. Acta Neurochir Suppl.
2015;163(10):768–77. 2016;122:193–7.
72. Madden LK, Hill M, May TL, Human T, Guanci 83. McIntyre LA, Fergusson DA, Hebert PC, Moher D,
MM, Jacobi J, Moreda MV, Badjatia N. The imple- Hutchison JS. Prolonged therapeutic hypothermia
mentation of targeted temperature management: an after traumatic brain injury in adults: a systematic
evidence-based guideline from the Neurocritical review. JAMA. 2003;289(22):2992–9.
Care Society. Neurocrit Care. 2017;27(3):468–87. 84. Naito H, Isotani E, Callaway CW, Hagioka S,
73. Akula VP, Joe P, Thusu K, Davis AS, Tamaresis Morimoto N. Intracranial pressure increases during
JS, Kim S, Shimotake TK, Butler S, Honold J, rewarming period after mild therapeutic hypother-
Kuzniewicz M, et al. A randomized clinical trial mia in postcardiac arrest patients. Ther Hypothermia
of therapeutic hypothermia mode during trans- Temp Manag. 2016;6(4):189–93.
port for neonatal encephalopathy. J Pediatr. 85. Abend NS, Topjian A, Ichord R, Herman ST,
2015;166(4):856–61. Helfaer M, Donnelly M, Nadkarni V, Dlugos DJ,
74. Heard KJ, Peberdy MA, Sayre MR, Sanders A, Clancy RR. Electroencephalographic monitoring
Geocadin RG, Dixon SR, Larabee TM, Hiller K, during hypothermia after pediatric cardiac arrest.
Fiorello A, Paradis NA, et al. A randomized con- Neurology. 2009;72(22):1931–40.
trolled trial comparing the Arctic Sun to standard 86. Kim KW, Pargeon KL, Labar AS, Friedman O,
cooling for induction of hypothermia after cardiac Kandula PN, Labar DR. EEG characteristics in
arrest. Resuscitation. 2010;81(1):9–14. cooled and rewarmed periods in post-cardiac
75. Hoedemaekers CW, Ezzahti M, Gerritsen A, van arrest therapeutic hypothermia patients. J Clin
der Hoeven JG. Comparison of cooling methods Neurophysiol. 2017;34(4):381–90.
to induce and maintain normo- and hypothermia in 87. Herman ST, Abend NS, Bleck TP, Chapman KE,
intensive care unit patients: a prospective interven- Drislane FW, Emerson RG, Gerard EE, Hahn CD,
tion study. Crit Care. 2007;11(4):R91. Husain AM, Kaplan PW, et al. Consensus state-
76. Glover GW, Thomas RM, Vamvakas G, Al-Subaie ment on continuous EEG in critically ill adults and
N, Cranshaw J, Walden A, Wise MP, Ostermann M, children, part I: indications. J Clin Neurophysiol.
Thomas-Jones E, Cronberg T, et al. Intravascular ver- 2015;32(2):87–95.
sus surface cooling for targeted temperature manage- 88. Kozar M, Javorka K, Javorka M, Matasova K,
ment after out-of-hospital cardiac arrest - an analysis Zibolen M. Changes of cardiovascular regula-
of the TTM trial data. Crit Care. 2016;20(1):381. tion during rewarming in newborns undergoing
77. Sonder P, Janssens GN, Beishuizen A, Henry whole-body hypothermia. Neuro Endocrinol Lett.
CL, Rittenberger JC, Callaway CW, Dezfulian C, 2015;36(5):434–8.
Polderman KH. Efficacy of different cooling tech- 89. Soeholm H, Kirkegaard H. Serum potassium
nologies for therapeutic temperature management: changes during therapeutic hypothermia after out-
a prospective intervention study. Resuscitation. of-hospital cardiac arrest-should it be treated? Ther
2017;124:14–20. Hypothermia Temp Manag. 2012;2(1):30–6.
78. Hegazy AF, Lapierre DM, Butler R, Martin J, 90. Badjatia N, Kowalski RG, Schmidt JM, Voorhees
Althenayan E. The esophageal cooling device: a new ME, Claassen J, Ostapkovich ND, Presciutti
temperature control tool in the intensivist's arsenal. M, Connolly ES, Palestrant D, Parra A, et al.
Heart Lung. 2017;46(3):143–8. Predictors and clinical implications of shivering
79. Springborg JB, Springborg KK, Romner B. First during therapeutic normothermia. Neurocrit Care.
clinical experience with intranasal cooling for 2007;6(3):186–91.
hyperthermia in brain-injured patients. Neurocrit 91. Badjatia N, Strongilis E, Prescutti M, Fernandez
Care. 2013;18(3):400–5. L, Fernandez A, Buitrago M, Schmidt JM, Mayer
80. Lu X, Ma L, Sun S, Xu J, Zhu C, Tang W. The SA. Metabolic benefits of surface counter warm-
effects of the rate of postresuscitation rewarming ing during therapeutic temperature modulation. Crit
following hypothermia on outcomes of cardiopul- Care Med. 2009;37(6):1893–7.
ERRNVPHGLFRVRUJ
338 J. S. Wallisch and E. L. Fink
92. Choi HA, Ko SB, Presciutti M, Fernandez L, 103. Polderman KH. Mechanisms of action, physiologi-
Carpenter AM, Lesch C, Gilmore E, Malhotra R, cal effects, and complications of hypothermia. Crit
Mayer SA, Lee K, et al. Prevention of shivering Care Med. 2009;37(7 Suppl):S186–202.
during therapeutic temperature modulation: the 104. Jeppesen AN, Kirkegaard H, Ilkjaer S, Hvas
Columbia anti-shivering protocol. Neurocrit Care. AM. Influence of temperature on thromboelastome-
2011;14(3):389–94. try and platelet aggregation in cardiac arrest patients
93. Logan A, Sangkachand P, Funk M. Optimal man- undergoing targeted temperature management. Crit
agement of shivering during therapeutic hypo- Care. 2016;20(1):118.
thermia after cardiac arrest. Crit Care Nurse. 105. Trabka-Zawicki A, Tomala M, Zelias A, Paszek
2011;31(6):e18–30. E, Zajdel W, Stepien E, Zmudka K. Adaptation of
94. Park B, Lee T, Berger K, Park SM, Choi KE, global hemostasis to therapeutic hypothermia in
Goodsell TM, Rosengart A. Efficacy of nonpharma- patients with out-of-hospital cardiac arrest: throm-
cological antishivering interventions: a systematic boelastography study. Cardiol J. 2017;
analysis. Crit Care Med. 2015;43(8):1757–66. 106. Jacob M, Hassager C, Bro-Jeppesen J, Ostrowski SR,
95. Bro-Jeppesen J, Annborn M, Hassager C, Wise MP, Thomsen JH, Wanscher M, Johansson PI, Winther-
Pelosi P, Nielsen N, Erlinge D, Wanscher M, Friberg Jensen M, Kjaergaard J. The effect of targeted tem-
H, Kjaergaard J. Hemodynamics and vasopressor perature management on coagulation parameters
support during targeted temperature management and bleeding events after out-of-hospital cardiac
at 33 degrees C versus 36 degrees C after out-of- arrest of presumed cardiac cause. Resuscitation.
hospital cardiac arrest: a post hoc study of the tar- 2015;96:260–7.
get temperature management trial*. Crit Care Med. 107. Carrick MM, Tyroch AH, Youens CA, Handley
2015;43(2):318–27. T. Subsequent development of thrombocytopenia
96. Thomsen JH, Nielsen N, Hassager C, Wanscher and coagulopathy in moderate and severe head
M, Pehrson S, Kober L, Bro-Jeppesen J, Soholm injury: support for serial laboratory examina-
H, Winther-Jensen M, Pellis T, et al. Bradycardia tion. J Trauma. 2005;58(4):725–9; discussion
during targeted temperature management: an early 729–730.
marker of lower mortality and favorable neuro- 108. Clifton GL, Valadka A, Zygun D, Coffey CS,
logic outcome in comatose out-of-hospital car- Drever P, Fourwinds S, Janis LS, Wilde E, Taylor P,
diac arrest patients. Crit Care Med. 2016;44(2): Harshman K, et al. Very early hypothermia induc-
308–18. tion in patients with severe brain injury (the National
97. Zhang W, Lu M, Zhang C, Zhang R, Ou X, Zhou J, Acute Brain Injury Study: Hypothermia II): a ran-
Li Y, Kang Y. Therapeutic hypothermia increases the domised trial. Lancet Neurol. 2011;10(2):131–9.
risk of cardiac arrhythmia for perinatal hypoxic isch- 109. Stockmann H, Krannich A, Schroeder T, Storm C.
aemic encephalopathy: a meta-analysis. PLoS One. Therapeutic temperature management after cardiac
2017;12(3):e0173006. arrest and the risk of bleeding: systematic review
98. Bro-Jeppesen J, Hassager C, Wanscher M, and meta-analysis. Resuscitation. 2014;85(11):
Ostergaard M, Nielsen N, Erlinge D, Friberg H, 1494–503.
Kober L, Kjaergaard J. Targeted temperature man- 110. Wang CH, Chen NC, Tsai MS, Yu PH, Wang AY,
agement at 33 degrees C versus 36 degrees C and Chang WT, Huang CH, Chen WJ. Therapeutic hypo-
impact on systemic vascular resistance and myo- thermia and the risk of hemorrhage: a systematic
cardial function after out-of-hospital cardiac arrest: review and meta-analysis of randomized controlled
a sub-study of the target temperature management trials. Medicine. 2015;94(47):e2152.
trial. Circ Cardiovasc Interv. 2014;7(5):663–72. 111. Ishikawa K, Tanaka H, Shiozaki T, Takaoka M, Ogura
99. Polderman KH, Peerdeman SM, Girbes AR. H, Kishi M, Shimazu T, Sugimoto H. Characteristics
Hypophosphatemia and hypomagnesemia induced of infection and leukocyte count in severely head-
by cooling in patients with severe head injury. J injured patients treated with mild hypothermia. J
Neurosurg. 2001;94(5):697–705. Trauma. 2000;49(5):912–22.
100. Cueni-Villoz N, Devigili A, Delodder F, Cianferoni 112. Geurts M, Macleod MR, Kollmar R, Kremer PH,
S, Feihl F, Rossetti AO, Eggimann P, Vincent JL, van der Worp HB. Therapeutic hypothermia and the
Taccone FS, Oddo M. Increased blood glucose vari- risk of infection: a systematic review and meta-anal-
ability during therapeutic hypothermia and outcome ysis. Crit Care Med. 2014;42(2):231–42.
after cardiac arrest. Crit Care Med. 2011;39(10): 113. Tortorici MA, Kochanek PM, Poloyac SM. Effects
2225–31. of hypothermia on drug disposition, metabolism,
101. Raper JD, Wang HE. Urine output changes during and response: a focus of hypothermia-mediated
postcardiac arrest therapeutic hypothermia. Ther alterations on the cytochrome P450 enzyme system.
Hypothermia Temp Manag. 2013;3(4):173–7. Crit Care Med. 2007;35(9):2196–204.
102. Morgan ML, Anderson RJ, Ellis MA, Berl T. 114. Bjelland TW, Klepstad P, Haugen BO, Nilsen T,
Mechanism of cold diuresis in the rat. Am J Phys. Salvesen O, Dale O. Concentrations of remifen-
1983;244(2):F210–6. tanil, propofol, fentanyl, and midazolam during
ERRNVPHGLFRVRUJ
20 Therapeutic Hypothermia in the Pediatric ICU 339
rewarming from therapeutic hypothermia. Acta dren with traumatic brain injury. Crit Care Med.
Anaesthesiol Scand. 2014;58(6):709–15. 2013;41(10):2379–87.
115. Caldwell JE, Heier T, Wright PM, Lin S, McCarthy 117. Fukuoka N, Aibiki M, Tsukamoto T, Seki K, Morita
G, Szenohradszky J, Sharma ML, Hing JP, Schroeder S. Biphasic concentration change during continuous
M, Sessler DI. Temperature-dependent pharmaco- midazolam administration in brain-injured patients
kinetics and pharmacodynamics of vecuronium. undergoing therapeutic moderate hypothermia.
Anesthesiology. 2000;92(1):84–93. Resuscitation. 2004;60(2):225–30.
116. Empey PE, Velez de Mendizabal N, Bell MJ, Bies 118. Schaible DH, Cupit GC, Swedlow DB, Rocci ML
RR, Anderson KB, Kochanek PM, Adelson PD, Jr. High-dose pentobarbital pharmacokinetics in
Poloyac SM, Pediatric TBICHI. Therapeutic hypo- hypothermic brain-injured children. J Pediatr.
thermia decreases phenytoin elimination in chil- 1982;100(4):655–60.
ERRNVPHGLFRVRUJ
Index
ERRNVPHGLFRVRUJ
342 Index
ERRNVPHGLFRVRUJ
Index 343
ERRNVPHGLFRVRUJ
344 Index
ERRNVPHGLFRVRUJ
Index 345
ERRNVPHGLFRVRUJ
346 Index
ERRNVPHGLFRVRUJ
Index 347
ERRNVPHGLFRVRUJ
348 Index
ERRNVPHGLFRVRUJ
Index 349
ERRNVPHGLFRVRUJ
350 Index
W Z
Withdrawal of life-sustaining therapies (WLST), 318 Zero-balance ultrafiltration (ZUF), 219
ERRNVPHGLFRVRUJ