Pediatric Anesthesia ISSN 1155-5645

REVIEW ARTICLE

Limitations and vulnerabilities of the neonatal

cardiovascular system: considerations for anesthetic
management
Andrew R. Wolf1,2 & Adrian T. Humphry1
1 Department of Paediatric Intensive Care, Bristol Royal Children’s Hospital, Bristol, UK
2 Department of Paediatric Anaesthesia, Bristol Royal Children’s Hospital, Bristol, UK

Keywords
neonate; cardiac physiology; cardiac
surgery; inotropes; myocardium
Correspondence
Andrew R. Wolf, Department of Paediatric
Intensive Care, Bristol Royal Children’s
Hospital, Upper Maudlin Street, Bristol BS2
8BJ, UK
Email: awolfbch@aol.com
Section Editor: Andrew R Wolf
Accepted 2 October 2013
doi:10.1111/pan.12290
Summary
Development of the cardiovascular system through the last trimester of preg-
nancy and the subsequent neonatal period is profound. Morphological
changes within the myocardium make the heart vulnerable to challenges such
as fluid shifts and anesthetic drugs. The sensitivity of the myocardium to meta-
bolic challenges and potential harm of drugs needed to maintain adequate blood
pressure and cardiac output are highlighted. Traditional monitoring under anes-
thesia has focussed on maintaining oxygenation and heart rate in the neonate
with less attention paid to blood pressure, cardiac output, and more importantly
organ well-being. There is now a better understanding of the limitations of
blood pressure homeostasis in the neonate and the potential consequences of
marginal hypoperfusion. This article highlights some of these vulnerabilities
particularly as they relate to anesthesia and surgery in the very young.

sets out to highlight some of these key issues that need

Introduction
The neonatal cardiovascular system does not represent a
small-scale model of the mature organism. The fetal
heart is morphologically immature and is functionally
adapted to a low-pressure, hypoxic environment where
there is minimal pulmonary blood flow (c. 8%) and low
systemic vascular resistance. Quite apart from the radi-
cal adjustment to an extra-uterine circulation at birth,
the neonatal heart has to rapidly adapt to a dual circula-
tion with significantly greater systemic vascular resis-
tance and increased metabolic requirements.
Traditionally, the neonatal heart and circulation has
been seen as a robust system in that it appears to
stand the traumas associated with birth and quite sub-
stantial periods of hypoxia. Much of the focus of anes-
thesia of the neonate has been centred on the
prevention of hypoxia and the maintenance of heart
rate with less attention to blood pressure and ade-
quacy organ flow. The challenge of optimizing cardio-
vascular support under anesthesia is now more
recognized and requires understanding of the limita-
tions of myocardial function, response to inotropes,
and immaturity of cardiovascular control. This article
consideration when anesthetizing the neonate.
The neonatal myocardium
The structure and function of the myocardium
changes markedly in the neonatal period, and during
this time, it is vulnerable and poorly adapted to the
challenges of anesthesia and surgery. Histologically,
the myocardial cells are disorganized compared with
the mature myocardium, less compacted and with
increased noncontractile tissue and water between
contractile units (1,2). This is reflected of the higher
proportion of extracellular fluid and total body water
compared with the older infant. As a result, the neo-
natal myocardium is inefficient as a filling and con-
tracting unit: ventricular compliance is considerably
reduced and wall tension/intraventricular pressure
rises rapidly with increased end diastolic volume (3).
There is also poor responsiveness to preload (a flat
Starling curve) and a vulnerability to overfilling, with
wall tension rising rapidly, coronary perfusion falling,
overdistention, and heart failure. However, within the
first month of life, diastolic relaxation improves

© 2013 John Wiley & Sons Ltd 5

Pediatric Anesthesia 24 (2014) 5–9
The neonatal cardiovascular system
significantly, and with it the ability to respond to
intravascular fluid shifts (4).
Cellular energy is derived primarily from glucose and
glycolysis in the hypoxic fetal environment, where oxy-
gen availability is limited (5). After birth, the neonate
progresses to a combination of carbohydrate and short-
chain fatty acids with a decrease in glycogen and lactate
concentrations and finally to all energy sources with the
primary source obtained from long-chain fatty acids in
the adult (6). Despite the ability of the neonatal myocar-
dium to use glycolysis as an oxygen-independent source
of high-energy phosphates, myocardial energy require-
ments come at their greatest in the neonate compared
with both fetus and adult, reflecting the age-related
requirements of cardiac index (7,8). Given the limitation
of external glycogen and other energy stores in the neo-
nate maintenance of energy input and avoiding hypogly-
cemia or excessive cardiac work load is essential to
avoid fatigue and substrate failure, which can occur
abruptly and catastrophically in this age group.
The excitation contraction coupling and later relaxa-
tion (diastolic phase) are points of particular vulnerabil-
ity in the neonatal heart. In the adult, ionized calcium
enters the cell with depolarization through L-type chan-
nels and passes rapidly through the t-tubular system.
The calcium influx triggers calcium release from within
the sarcoplasmic reticulum (SR) in response to interac-
tion with the ryanodine receptor, and this in turn acti-
vates the myosin/actin contractile process. During
diastole, calcium is rapidly sequestered within the SR to
allow relaxation of the contractile elements. The neona-
tal heart has reduced L-type channels on the myocardial
cell surface, and entry of calcium into the cell is primar-
ily by T-type channels, calcium-handling proteins, and
the reverse sodium/calcium exchange mechanism that is
more usually associated with calcium removal from the
cell (9,10). Propagation of the calcium flux into the cell
is poor due to the immaturity of the t-tubular system,
and at the SR, the paucity of ryanodine receptors is
associated with a limited release of calcium to activate
contraction (11). Finally, the reuptake into the SR is
limited at the end of systole, preventing effective relaxa-
tion during diastole. Cardiac contraction in the neonate
is therefore much more dependent extracellular, and
nuclear calcium for the initiation of contraction and
maintenance of adequate plasma calcium concentrations
may be a factor in cardiac performance (12,13). Devel-
opment of the SR, t-tubular system, and calcium-
handling proteins appears to be rapid, and it has been
suggested that they are relatively mature by 3 weeks in
the human neonatal heart (14).
In recent years, there has been considerable interest in
the potential ability of the myocardium to remodel its
6 © 2013 John Wiley & Sons Ltd
A.R. Wolf and A.T. Humphry
structure after cardiac repair and even to regenerate
myocardial tissue through stem cells. Recent data would
support the hypothesis that the human neonatal heart
has a greater regenerative ability than adult hearts that
may be due to greater numbers of stem cells and their
ability to differentiate (15). This bodes well for future
research in that this capacity may allow triggers for cell
regeneration to be enhanced in the infant heart that has
been affected by congenital, acquired, or even iatrogenic
derived injury.
Heart and vascular responses to inotropes
Given that the neonatal heart has a much reduced
capacity to increase its stroke volume in response to
increasing preload (16), beta-agonist catecholamines are
commonly used to increase cardiac output by increasing
myocardial contractility and heart rate, in conjunction
with drugs that optimize pulmonary or systemic vascu-
lar resistances. However, there is an increasing under-
standing that while catecholamines have a place in
management of the neonate both in terms of myocar-
dial performance and manipulation of the systemic vas-
cular resistance, the immature heart is very vulnerable
to the intrinsic effects of these drugs on the myocar-
dium. Irrespective of age, cardiac tissue requires ade-
quate myocardial perfusion via the coronary arteries to
deliver oxygen and substrates. Published values for
blood pressure are significantly higher than that usually
accepted in the anesthetized neonate (17,18), but within
the pediatric anesthetic community, there is consider-
able uncertainty of what constitutes hypotension requir-
ing intervention (18). There are limited modalities
available to increase blood pressure in the normal heart:
intravenous fluid administration to maintain vascular
volume, inotropes and chronotropes, which can
improve stroke volume/heart rate, and vasoconstriction,
which will raise blood pressure but at the expense of
increased heart work unless the systemic vascular resis-
tance is reduced (for example in conditions such as sep-
sis). While catecholamines may achieve the initial goal
of raising blood pressure to levels that will perfuse the
coronary arteries and increase flow to other vital
organs, the neonatal heart appears to be vulnerable to
the effects of catecholamines in terms of myocardial
damage and depletion of energy substrates. In an intact
neonatal piglet model (2–4 Kg), heart rate was
increased for a period of 2 h either with pacing or with
epinephrine infusion (19). While there was limited
change in the paced group, there was increased ventric-
ular wall stiffness in the catecholamine group, and
within the myocardial cell, evidence of areas of damage
with sarcolemmal rupture and mitochondrial damage
Pediatric Anesthesia 24 (2014) 5–9
A.R. Wolf and A.T. Humphry
with swelling and calcium deposition. The pathological
changes appear to be age related with the neonatal
heart being the most vulnerable (20). Suggested causes
are reduced sequestration of calcium within the SR, fail-
ure of the ATP calcium relaxation mechanism, and free
radical generation from catecholamine oxidation. It
may also be related to the exhaustion of energy supply
with falling levels of ATP after catecholamine treatment
(19). This observation was also seen in a comparative
study of neonatal and mature sheep where heart work
was increased to similar levels of oxygen consumption
using an epinephrine infusion (20). In the mature sheep,
increased heart work was associated with stable myo-
cardial ATP/ADP concentrations, but in neonatal
lambs, there was a significant fall of high-energy phos-
phates indicative of altered control of energy provision.
These data support the clinical observations that while
catecholamines may be necessary to maintain the failing
neonatal heart, they can in themselves promote cellular
damage and failure of energy supply.
In recent years, milrinone has become a popular drug
in the management of the sick neonatal heart either on
its own or in combination with a catecholamine. How-
ever, its mode of action and whether it could be used
more effectively remain unclear. It is one of a group of
drugs that inhibit hydrolysis of cAMP within the myo-
cardium and vascular smooth muscle through blockade
of the phosphodiesterase enzyme (PDE). Milrinone is
often described as a PDE3 blocker, but it has actions on
other phosphodiesterase receptors and belongs to a fam-
ily of PDE enzymes, which have different effects within
vascular tissue. For example, sildenafil is primarily a
PDE4/5 blocker and has specific actions on dilation of
the pulmonary vascular smooth muscle and to a lesser
extent the peripheral vascular smooth muscle as well as
its most exploited property of treatment of male impo-
tence. Blockade of PDE3 within the myocardium
increases the availability of calcium within the sarco-
lemma during systole by enhancing calcium entry into
the cell and promoting release of calcium at the SR,
while increasing calcium reuptake by the SR in diastole
(21). Given the poor diastolic relaxation associated with
the neonatal heart and limitations of calcium flux as
described above, PDE blockade would be expected to
have beneficial effects throughout the cardiac cycle. Pre-
vious data have also shown that milrinone can restore
the efficiency of beta-agonists in the isolated myocyte
after a period of reperfusion, making it a drug that has
value in combination with a catecholamine not only for
its intrinsic effect but as a catecholamine sparing drug
(22). In the peripheral vascular smooth muscle, PDE3
blockade causes relaxation via both cAMP and cGMP
mechanisms.
© 2013 John Wiley & Sons Ltd
Pediatric Anesthesia 24 (2014) 5–9
The neonatal cardiovascular system
An early study with milrinone in 10 neonates after
cardiac surgery demonstrated an increased cardiac index
with falls in systemic and pulmonary vascular resistance
(23). This was followed by what has to date been the
definitive trial of milrinone in the prevention of low car-
diac output syndrome (LCOS) after cardiac bypass sur-
gery in neonates (24). The trial has been criticized for
not having an active comparator with a vasodilator, but
it has transformed management of the neonate having
bypass surgery away from the extensive use of potent
dilators at the outset to a more consistent use of
milrinone with minimal use of catecholamines. Never-
theless, there are some highly successful units who have
continued to publish excellent results using radical
alpha-blocking drugs such as phenoxybenzamine in the
most complex patients (25).
While there is little doubt as to the effectiveness of the
drug, the multiple actions of the drug on a variety of tar-
get sites leave the question of why it works and leave the
way open for more effective drugs in the treatment for
neonatal LCOS in the future. Experimentally, it appears
to have a preservative action on energy stores within the
mitochondria by blockade of specific ion channels (26),
and in adult cardiac surgery, the delivery of adequate
doses of milrinone prebypass is associated with the
maintenance of cAMP stores within the myocardium
after removal of the aortic cross-clamp (27). These data
would suggest a role for instituting milrinone not just as
a loading dose on bypass once the cross-clamp has been
applied, but well before.
One of the problems with the drug in intensive care is
that while it prevents vasoconstriction immediately after
bypass surgery (28), it cannot alone reliably modulate
vasodilation and neonates may then still require addi-
tional vasodilators. However, if high-dose infusions are
maintained over several days in the treatment for LCOS,
the drug accumulates, resulting in vasodilation and
hypotension that can be confused with pump failure or
sepsis. Understanding the pharmacokinetics of milri-
none in terms of adequate loading but reducing mainte-
nance infusion rates over time in the neonate with
limited renal function and low cardiac output has helped
to recognize and avoid this problem (29,30).
Down-regulation of beta-receptors in the sick neona-
tal heart results in catecholamine resistance and escala-
tion of dosage with less effect. Adult data from atrial
tissue suggest that this may be due to uncoupling of the
beta-receptors with reduced levels of cAMP in response
to beta-stimulation (31). There is current interest in the
use of low-dose vasopressin (AVP) in pediatric intensive
care to overcome the effects of catecholamine resistance
following Cardiopulmonary bypass in the neonate. This
seems counterintuitive as one of its major actions is to
7
The neonatal cardiovascular system A.R. Wolf and A.T. Humphry

increase peripheral resistance. However, AVP does have
some intrinsic inotropic effect, increases myocardial sen-
sitivity to catecholamines, and causes both coronary and
pulmonary vasodilatation (32). A small retrospective
study showed that, during the first postoperative day,
patients who had been started on AVP immediately after
surgery required less fluid resuscitation and catecholam-
ines (33). There were also trends toward shorter periods
of mechanical ventilation and intensive care stay.
Cardiovascular control in the neonate
Autonomic control of the cardiovascular system is sig-
nificantly reduced in the neonate leaving it prone to
blood pressure variations. From comparative anatomic
studies, the autonomic innervation of the heart appears
to be present from birth (34), but immature. Studies of
the autonomic function using frequency analysis of
heart-rate variability indicate that at birth all autonomic
responses are limited but that the sympathetic compo-
nent is the more dominant (35). In infancy, this reverses
and parasympathetic component becomes more domi-
nant until later childhood. Additionally, the preterm
and ex-preterm infant have enhanced vulnerability from
lack of cardiovascular autoregulation, which fails to
catch up with the responsiveness of the term infant for
up to 2 years (36,37). One of the consequences of the
reduced autonomic responses is the flattening of barore-
ceptor sensitivity (38). The result is that the responsive
changes in heart rate and vascular resistance from
reduced blood pressure are more limited. These limita-
tions in cardiovascular homeostasis make the neonate
vulnerable to maintaining blood pressure within tight
limits, particularly under anesthesia when autonomic
reflexes are further suppressed (39). Data are now begin-
ning to emerge that the consequences of hypotension
from drug administration may be significant in terms of
both short- and long-term neurological outcomes (40).
This issue is dealt with in more detail elsewhere in this
neonatal edition of Pediatric Anesthesia.
In addition to global cardiovascular control, individ-
ual-organ-specific autoregulation helps to maintain
blood flow independently of blood pressure within the
normal operating range of blood pressure. Cerebral
blood flow is also greatly affected by changes in arterial
CO2 tension and hypoxia. Hypocarbia causes vasocon-
striction and has been implicated in the development of
periventricular leukomalacia in premature infants (41).
Hypoxia has a vasodilatory effect. Healthy neonates do
show autoregulation in respect of cerebral blood flow
(CBF), and this may be present from an early gesta-
tional age (42). However, evidence is increasing that in
sick neonates, cerebral autoregulation is lost. In the
newborn lamb, 20 min of hypoxia results in the loss of
autoregulation for several hours (43). Healthy term neo-
nates are able to normalize CBF after a rise in blood
pressure, but not sick term infants and all preterms;
there is correlation between the elevation of blood pres-
sure and CBF, indicating some failure of cerebral auto-
regulation (44). Given that anesthetic agents impair
cerebral autoregulation even at low concentrations, the
necessity to control blood pressure, oxygenation, and
carbon dioxide tensions during anesthesia in the vulner-
able neonate that has very limited compensations is of
key importance. Interest is now growing in monitors
such as tissue and cerebral oximetry that can measure
organ perfusion as well as the simpler cardiovascular
parameters.
Conflict of interest
No conflicts of interest declared.

References
1 Marijianowski MM, van der Loos CM, Mo-
hrschladt MF et al. The neonatal heart has a
relatively high content of total collagen and
type I collagen, a condition that may explain
the less compliant state. J Am Coll Cardiol
1994; 23: 1204–1208.
2 Pelouch V, Kolar F, Milerova M et al.
Effect of the preweaning nutritional state on
the cardiac protein profile and functional
performance of the rat heart. Mol Cell Bio-
chem 1997; 177: 221–228.
3 Romero T, Covell J, Friedman WF. A com-
parison of pressure-volume relations of the
fetal, newborn, and adult heart. Am J Phys-
iol 1972; 222: 1285–1290.
4 Kozak-Barany A, Jokinen E, Rantonen T
et al. Efficiency of left ventricular diastolic
function increases in healthy full-term
infants during the first months of life: a pro-
spective follow-up study. Early Hum Dev
2000; 57: 49–59.
5 Chen F, De Diego C, Xie L-H et al.
Effects of metabolic inhibition on conduc-
tion, Ca transients, and arrhythmia vulner-
ability in embryonic mouse hearts. Am J
Physiol Heart Circ Physiol 2007; 293:
H2472–H2478.
6 Ascuitto RJ, Ross-Ascuitto NT. Substrate
metabolism in the developing heart. Semin
Perinatol 1996; 20: 542–563.
7 Fisher DJ, Heymann MA, Rudolph AM.
Myocardial consumption of oxygen and car-
bohydrates in newborn sheep. Pediatr Res
1981; 15: 843–846.
8 Modi P, Suleiman MS, Reeves B et al. Myo-
cardial metabolic changes during pediatric
cardiac surgery: a randomized study of 3 car-
dioplegic techniques. J Thorac Cardiovasc
Surg 2004; 128: 67–75.
9 Qu Y, Boutjdir M. Gene expression of SER-
CA2a and L- and T-type Ca channels during
human heart development. Pediatr Res 2001;
50: 569–574.
10 Qu Y, Ghatpande A, El-Sherif N et al. Gene
expression of Na+/Ca2+ exchanger during
development in human heart. Cardiovasc Res
2000; 45: 866–873.
11 Seki S, Nagashima M, Yamada Y et al. Fetal
and postnatal development of Ca2+ tran-
sients and Ca2+ sparks in rat cardiomyo-
cytes. Cardiovasc Res 2003; 58: 535–548.

8 © 2013 John Wiley & Sons Ltd

Pediatric Anesthesia 24 (2014) 5–9
A.R. Wolf and A.T. Humphry
12 Schroder EA, Wei Y, Satin J. The developing
cardiac myocyte: maturation of excitability
and excitation-contraction coupling. Ann N
Y Acad Sci 2006; 1080: 63–75.
13 Poindexter BJ, Smith JR, Buja LM et al.
Calcium signaling mechanisms in dedifferen-
tiated cardiac myocytes: comparison with
neonatal and adult cardiomyocytes. Cell Cal-
cium 2001; 30: 373–382.
14 Wiegerinck RF, Cojoc A, Zeidenweber CM
et al. Force frequency relationship of the
human ventricle increases during early post-
natal development. Pediatr Res 2009; 65:
414–419.
15 Simpson DL, Mishra R, Sharma S et al. A
strong regenerative ability of stem cells
derived from neonatal hearts. Circulation
2012; 126(Suppl 1): S46–S53.
16 Rudolph A. Distribution and regulation of
blood flow in the fetal and neonatal lamb.
Circ Res 1985; 57: 811–821.
17 Jones JE, Jose PA. Neonatal blood pressure
regulation. Semin Perinatol 2004; 28:
141–148.
18 Naifu OO, Voepel-Lewis T, Morris M. How
do pediatric anaesthesiologists define intra-
operative hypotension. Pediatr Anesth 2009;
19: 1048–1053.
19 Caspi J, Coles JG, Benson LN et al. Heart
rate independence of catecholamine-induced
myocardial damage in the newborn pig. Pedi-
atr Res 1994; 36: 49–54.
20 Caspi J, Coles JG, Benson LN et al. Age-
related responses to epinephrine induced
myocardial stress: a functional and ultra-
structural study. Circulation 1991; 84(Suppl
111): 394–399.
21 Portman MA, Heineman FW, Balaban RS.
Developmental changes in the relation
between phosphate metabolites and oxygen
consumption. J Clin Invest 1989; 83: 456–
464.
22 Skoyles JR, Sherry KM. Pharmacology,
mechanism of action and uses of selective
phosphodiesterase inhibitors. Br J Anaesth
1992; 68: 293–302.
23 Chang AC, Atz AM, Wernovsky G et al.
Milrinone: systemic and pulmonary hemody-
namic effects in neonates after cardiac sur-
gery. Crit Care Med 1995; 23: 1907–1914.
© 2013 John Wiley & Sons Ltd
Pediatric Anesthesia 24 (2014) 5–9
24 Hoffman TM, Wernovsky G, Atz AM et al.
Efficacy and safety of milrinone in preventing
low cardiac output syndrome in infants and
children after corrective surgery for congeni-
tal heart disease. Circulation 2003; 107: 996–
1002.
25 Tweddell JS, Hoffman GM, Mussatto KA
et al. Improved survival of patients undergo-
ing palliation of hypoplastic left heart syn-
drome: lessons learned from 115 consecutive
patients. Circulation 2002; 106(Suppl 1): I
82–I 89.
26 Chanoit G, Zhou J, Lee S et al. Inhibition of
phosphodiesterases leads to prevention of the
mitochondrial permeability transition pore
opening and reperfusion injury in cardiac
H9c2 cells. Cardiovasc Drugs Ther 2011; 25:
299–306.
27 Janelle GM, Urdaneta F, Blas ML et al.
Inhibition of phosphodiesterase type III
before aortic cross-clamping preserves intra-
myocardial cyclic adenosine monophosphate
during cardiopulmonary bypass. Anesth
Analg 2001; 92: 1377–1383.
28 Stocker CF, Shekerdemian LS, Nørgaard
MA et al. Mechanisms of a reduced cardiac
output and the effects of milrinone and
levosimendan in a model of infant cardiopul-
monary bypass. Crit Care Med 2007; 35:
252–259.
29 Zuppa AF, Nicolson SC, Adamson PC
et al. Population pharmacokinetics of mil-
rinone in neonates with hypoplastic left
heart syndrome undergoing stage 1 recon-
struction. Anesth Analg 2006; 102: 1062–
1069.
30 Paradisis M, Jiang X, McLachlan AJ et al.
Population pharmacokinetics and dosing reg-
imen design of milrinone in preterm infants.
Arch Dis Child Fetal Neonatal Ed 2007; 92:
F204–F209.
31 Schranz D, Droege A, Broede A et al.
Uncoupling of human cardiac beta-adreno-
ceptors during cardiopulmonary bypass with
cardioplegic cardiac arrest. Circulation 1993;
87: 422–426.
32 Holmes C, Landry D, Granton J. Science
review: vasopressin and the cardiovascular
system part 2 – clinical physiology. Crit Care
2004; 8: 15–23.
The neonatal cardiovascular system
33 Alten J, Borasino S, Toms R et al. Early ini-
tiation of arginine vasopressin infusion in
neonates after complex cardiac surgery. Pedi-
atr Crit Care Med 2012; 13: 300–304.
34 Hopkins D, Gootman P, Gootman N et al.
Anatomy medullary and peripheral auto-
nomic neurons innervating the neonatal
porcine heart. J Auton Nerv Syst 1997; 64:
74–84.
35 Yiallourou SR, Sands SA, Walker AM et al.
Maturation of heart rate and blood pressure
variability during sleep in term-born infants.
Sleep 2012; 35: 177–186.
36 Patural H, Pichot V, Jaziri F et al. Auto-
matic control of very preterm newborns: a
prolonged dysfunction. Early Human Dev
2008; 84: 681–687.
37 De Rogalski Landrot I, Roche F, Pichot V
et al. Autonomic nervous system activity in
premature and full-term infants from theo-
retical term to 7 years. Auton Neurosci 2007;
136: 105–109.
38 Gournay V, Drouin E, Roze J. Development
of baroreflex control of heart rate in preterm
and full term infants. Arch Dis Child Fetal
Neonatal Ed 2002; 86: F151–F154.
39 Murat I, Levron J, Berg A et al. Effects of
fentanyl on baroreceptor reflex control of
heart rate in newborn infants. Anesthesiology
1998; 68: 717–722.
40 Anand KJ, Hall RW, Desai N et al. Effects
of morphine analgesia in ventilated preterm
neonates: primary outcomes from the NEO-
PAIN randomised trial. Lancet 2004; 363:
1673–1682.
41 Shankaran S, Langer JC, Kazzi SN et al.
Cumulative index of exposure to hypocarbia
and hyperoxia as risk factors for periventric-
ular leukomalacia in low birth weight infants.
Pediatrics 2006; 118: 1654–1659.
42 Greisen G. Autoregulation of cerebral blood
flow in newborn babies. Early Hum Dev
2005; 81: 423–428.
43 Tweed A, Cote J, Lou H et al. Impairment
of cerebral blood flow autoregulation in the
newborn lamb by hypoxia. Pediatr Res 1986;
20: 516–519.
44 Boylan G, Young K, Panerai R et al.
Dynamic cerebral autoregulation in sick new-
born infants. Pediatr Res 2000; 48: 12–17.
9