Professional Documents
Culture Documents
doi: 10.1093/tropej/fmx061
Advance Access Publication Date: 7 August 2017
Original paper
ABSTRACT
The cornerstone of emergency management of sepsis is early, goal-directed therapy. The purpose of
this study was to evaluate the effect of intraosseous (IO) vs. intravenous (IV) access for resuscitation
of patients with septic shock admitted to pediatric intensive care unit. This prospective interven-
tional randomized clinical trial study was conducted on 60 patients with septic shock who need
rapid administration of fluids and drugs; 30 cases were randomly chosen for IO vascular access,
while the other 30 were selected for IV access. The IO route was successfully secured in all cases
with a significant shorter time of vascular access insertion, shorter length of stay and reduction in
mortality in IO group vs. IV group (p ¼ 0.001, 0.045, 0.002, respectively). In pediatric emergencies,
as in case of shock, the use of IO route is recommended to get rapid vascular access as soon as pos-
sible, as it revealed better outcome.
C The Author [2017]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
V 132
Intraosseous Vs. Intravenous Access in Pediatric Septic 133
use of IO access as first-line therapy in septic shock imperfecta, cellulitis, osteomyelitis or fractured bone
management in pediatrics especially in Egypt. The aim in the lower limbs.
of this work was to evaluate the effect of initial IO ac-
Primary outcomes: Primary outcomes were the
cess vs. initial IV access on resuscitation of patients
time for establishing a successful access, time for re-
with septic shock admitted to PICU.
turn of spontaneous circulation (ROSC) parameters
and mortality.
SUBJECTS AND METHODS
Secondary outcomes: Secondary outcomes were
This prospective interventional randomized clin-
Table 1. Comparison of clinical characteristics, success of first trial, characteristics of access and
fate between IV and IO groups
Patients and access characteristics IV group IO group p value
(n ¼ 30) (n ¼ 30)
Age (months)
Min-Max 1.0–36.0 1.5–30.0 0.772
Median 5.50 4.00
Deceased patients had significantly longer time of LOS among IO group, whereas there were increased
all ROSC parameters including normalization of CRT, cumulative hazards of mortality with prolonged LOS
equalization of central and peripheral pulsations, get- among IV group. The two survival curves for IV
ting warm extremities, getting UOP >1 ml/kg/h and group and IO group are statistically significant
attaining normal MAP (p ¼ 0.000, 0.002, 0.000, 0.037, (v2 (log-rank) ¼ 4.385, p ¼ 0.036).
0.017, respectively). The percent of patients success-
fully reaching normal CVP and ScvO2 within the first
6 h were significantly higher in the discharged group DISCUSSION
(p ¼ 0.002, 0.002, respectively) (Table 3). Our study showed that the IO route was successfully
Figure 2 shows the Kaplan–Meier survival curve secured in all cases but the IV line could not be
of the cumulative survival with a prolonged hospital secured in 50% of patients. There was a significantly
stay (mortality in sense to time). It showed no in- shorter time of vascular access insertion, shorter LOS
crease in cumulative hazards of mortality with the and reduction in mortality in IO group vs. IV group.
136 Intraosseous Vs. Intravenous Access in Pediatric Septic
Table 2. Comparison of time taken for ROSC parameters after full resuscitation between IV and
IO group
Time taken for ROSC to be restored IV group IO group p value
(n ¼ 30) (n ¼ 30)
Nonreversal of shock in sepsis often progresses to study reported that success of the first attempt of vas-
multiple organ failures (MOF) with a mortality rate cular access was significantly higher in the IO group
of 54% [18]. Patients treated rapidly had a signifi- vs. the IV group (100% vs. 50%). Furthermore,
cantly low mortality rate [19]. In the present study, Reasdes et al. [22] showed that individuals random-
the time needed to reverse shock and reach the first ized to tibial IO access had a successful first attempt
6 h goal and therapeutic end points was significantly at vascular access (91%) compared with either hu-
longer in deceased group vs. discharged. meral IO access (51%) or peripheral IV access
The failure rates of IV access in emergency situ- (43%). Moreover, Fiorito et al. [23] reported that
ations are between 10% and 40% and the average first attempt success was achieved in 78% of IO ac-
time necessary for peripheral IV catheterization was cess and Myers et al. [24] reported it between 81%
reported to be 2.5–13 min, and sometimes even up and 100% in IO access.
to 30 min [17, 20]. Rosetti et al. [21] stated that in In the present study, there was a significantly
pediatric patients, in a state of shock or cardiac ar- shorter median access time in the IO group vs. the
rest, vascular access may not be attainable in 6% IV group (52.5 s vs. 90 s). This is in agreement with
cases; Banerjee et al. [10] reported that the IV can- Banerjee et al. [10] who reported that average time
nulation was successful in 66% children with severe taken to obtain IO access was 67 s vs. 129 s for IV ac-
dehydration, while the IO route achieved in all the cess. However, some studies [17, 25] reported IO
cases within the first 5 min. Similarly, the present time >1 min; this depends on how the timing was
Intraosseous Vs. Intravenous Access in Pediatric Septic 137
Table 3. Comparison of the clinical variables between discharged and deceased patients
Clinical variables Discharged Deceased p value
(n ¼ 46) (n ¼ 14)
PIM2
Min-Max 20.4–90.0 49.0–98.00 0.000*
Median 48.10 90.50
Length of stay (days)
calculated. The adults were studied in the emergency significantly shorter for individuals assigned to the
department; Leidel et al. [17] reported that mean tibial IO access group (4.6 min) compared with
time required for IO access (2.3 6 0.8 min) was sig- those assigned to the humeral IO access group
nificantly lower vs. central venous catheterization (7.0 min) and neither time was significantly different
(CVC) (9.9 6 3.7 min). Similarly, Leidel B.A. [25] from that of the IV access group (5.8 min). This
reported that the median time for IO access was sig- difference could be explained that the study was
nificantly lower vs. CVC (2 vs. 8 min). done on adult patients experiencing a nontraumatic
In contrast to the present study and the above- out-of-hospital cardiac arrest and the longer timing is
mentioned results of various articles, Reasdes et al. because time to successful vascular access was meas-
[22] reported that the time to initial success was ured as the interval between paramedic arrival on
138 Intraosseous Vs. Intravenous Access in Pediatric Septic
scene and documented successful vascular access In the present study, there was a significantly lon-
placement. ger time to improve UOP >1 ml/kg/h, normalize
IO blood aspirate may serve as a reliable alterna- warm extremities and equalize central and peripheral
tive to venous access, especially for hemoglobin and pulse difference in the IV group vs. IO group. This
hematocrit levels. Exceptions are CO2 levels and delay, besides the other studied factors, could be an
platelet counts, which may be lower, and white blood important factor for higher mortality in the IV
cell counts, which may appear elevated [26]. In the group.
present study, blood sampling was successful only in In the present study, the median LOS was signifi-
43.3% of IO needle insertion while it was signifi- cantly shorter in the IO group vs. IV (6 vs. 7.5 days).
cantly higher (100%) after IV cannulation. Blood The Kaplan–Meier survival curve using cumulative
sampling was possible later on after the first success- hazards demonstrated that there is an increase in
ful fluid infusion shot. mortality risk with prolonged LOS in IV group vs.
There were no complications concerning the IO IO group where no cumulative hazards were seen.
needle insertion in our cases; this is in concordance The mortality in IV group was 40% vs. 6.7% in the
with other reports [27, 28]. Fiorito et al. [23] re- IO group. Both groups did not differ statistically in
ported complications in 12% of IO insertion but all median age and weight. Moreover, their condition
limited to infiltration and local edema. on admission as revealed by the median PIM2 score
In the present study, the median time of return of showed no statistical significant difference. The fac-
normal CRT and MAP was significantly shorter in tor of early start of resuscitation seems to be crucial.
IO group vs. the IV group. Each hour delay in rever- To summarize, the longer access insertion time
sal of hypotension or reaching CRT <2 s is associ- and LOS in the IV group vs. IO group could be ex-
ated with two-fold increase in the odds ratio of death plained by decreased vasodilatation in IV group. The
from MOF [29]. result would be the delay of ROSC parameter and
Intraosseous Vs. Intravenous Access in Pediatric Septic 139
ultimately linked with a significantly higher mortality 8. de Caen AR, Berg MD, Chameides L, et al. Part 12:
rate in the IV group. Pediatric advanced life support: 2015 American Heart
The limitation of the present study is that it is a Association Guidelines Update for Cardiopulmonary
Resuscitation and Emergency Cardiovascular Care.
single-center study; multicentered randomized clin-
Circulation 2015;132(18 Suppl 2):S526–42.
ical trials are needed to confirm the present study 9. Kissoon N, Orr RA, Carcillo JA. Updated American
results. College of Critical Care Medicine–pediatric advanced life
support guidelines for management of pediatric and neo-
natal septic shock: relevance to the emergency care clin-
cardiac arrest: a randomized controlled trial. Ann Emerg 26. Miller LJ, Philbeck TE, Montez D, Spadaccini CJ. A new
Med 2011;58:509–16. study of intraosseous blood for laboratory analysis. Arch
23. Fiorito BA, Mirza F, Doran TM, et al. Intraosseous access Pathol Lab Med 2010;134:1253–60.
in the setting of pediatric critical care transport. Pediatr 27. Ong ME, Chan YH, Oh JJ, et al. An observational, pro-
Crit Care Med 2005;6:50–3. spective study comparing tibial and humeral intraosseous
24. Myers LA, Russi CS, Arteaga GM. Semiautomatic intraoss- access using the EZ-IO. Am J Emerg Med 2009;27:8–15.
eous devices in pediatric prehospital care. Prehosp Emerg 28. Cooper BR, Mahoney PF, Hodgetts TJ, et al. Intra-osseous
Care 2011;15:473–6. access (EZ-IO) for resuscitation: UK military combat ex-
25. Leidel BA, Kirchhoff C, Bogner V, et al. Comparison of perience. J R Army Med Corps 2007;153:314–6.