1. Seminars in Fetal & Neonatal Medicine (2008) 13, 30e34
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/siny
Current controversies in hypothermic
neuroprotection
John D.E. Barks*
C. S. Mott Children’s Hospital, University of Michigan Health System, Ann Arbor, MI 48109, USA
KEYWORDS Summary In 2005, three randomised controlled trials (RCTs) showed that treating infants
Hypoxiceischaemic with hypoxiceischaemic encephalopathy (HIE) with hypothermia decreased the combined out-
encephalopathy; come of death or disability at 12e18 months, although treatment effects were modest. More
Neonatal recently, the US Food and Drug Administration (FDA) approved a device for selective head cool-
encephalopathy; ing. In addition, the protocol from another of the three trials, using equipment available in
Newborn infant; many hospitals, has been in the public domain for over a year. Why has this not led to a con-
Perinatal asphyxia; sensus that hypothermia is the standard of care for HIE? This is explored. Important questions
Therapeutic for future research will focus on ways to improve on initial results with cooling, such as drug
hypothermia plus hypothermia combination therapy and refining duration and depth of cooling or duration
of rewarming. Although the latter are important questions for future clinical trials, those who
are convinced by the evidence to date should focus on safe implementation of cooling using
protocols with established safety and efficacy and should consider ways to increase access
to cooling for eligible babies.
ª 2007 Published by Elsevier Ltd.
Introduction Why is everyone not convinced?
More than 2 years have passed since the results of three Until recently, the most definite sign that the neonatal
randomised controlled trials (RCTs), which individually and community was not universally convinced of the efficacy
collectively demonstrated the efficacy of hypothermia for of hypothermia was that an RCT (the ICE trial), with a
the treatment of perinatal hypoxiceischaemic encephalop- non-hypothermic control group, continued to enroll pa-
athy (HIE) in term or late preterm infants, were publish- tients in several countries. It would be logical to suppose
ed.1e4 Yet, to date, there is still no general consensus that that investigators who participated in the published or
hypothermia is the standard of care for such infants. Why is completed trials are most likely to be convinced of the
this the case? This review aims to address some of the cur- safety and efficacy of cooling, and that this conviction
rent controversies surrounding hypothermia as a treatment would be demonstrated by their centres now offering
for HIE. hypothermia to infants meeting their original trial criteria.
This appears to be true for the National Institute of Child
Health and Human Development (NICHD) and Neonatal
* Tel.: þ1 734 763 4109; fax: þ1 734 763 7728. Research Network trial centres. Furthermore, many of the
E-mail address: jbarks@med.umich.edu centres in the Brain Cooling RCT also participated in the
1744-165X/$ - see front matter ª 2007 Published by Elsevier Ltd.
doi:10.1016/j.siny.2007.09.004

2. Current controversies in hypothermic neuroprotection 31
subsequent FDA-approved continued access protocol and trial device, has never been ‘tested’ on a human infant
are now transitioning to the commercial Cool Capâ device. prior to its commercial release. The Cool Capâ System
One commonly expressed reservation about the early now features a graphical, touch-screen, user interface
completed trials is that none of them was powered to that walks the user through the entire cooling process. Al-
address the death versus disability question, i.e. clinicians though the Cool Capâ system is quite expensive compared
may still need to be convinced that cooling is not simply to the Blanketrol cooling mattress, economic modelling in-
permitting more babies to survive with severe handicaps, dicates that, from a societal perspective, the Cool Capâ
without increasing the number of non-handicapped survi- system would be cost-effective under several different re-
vors. It is likely that the results of the three pending RCTs gional implementation strategies in the state of Massachu-
(see below) will provide enough patients to evaluate sepa- setts.8 Despite its cost-effectiveness at a regional/societal
rately the effects of hypothermia on death and on disability level, individual hospitals might be deterred from cooling
in survivors by using meta-analysis. A related concern is that by price concerns.
outcome in later childhood, e.g. cognitive outcome at school The NICHD body cooling protocol is publicly available on
age, is unknown; interestingly, other new therapies in the Internet as a written protocol at https://neonatal.
neonatology, like surfactant or inhaled nitric oxide, have rti.org/studies_hypothermia.cfm. Unlike the FDA-approved
not been held to such a high standard before becoming head cooling system, the equipment used in the NICHD pro-
accepted therapies. Both the NICHD trial and the Brain tocol, the Blanketrol IIâ Hypo/hyperthermia system (Cin-
Cooling trial are planning school-age follow-up evaluations. cinnati Sub-Zero), was not designed or FDA-approved for
For others, an important question is, ‘have enough neonates. Thus, its use for treatment of HIE in neonates
patients been studied to be reasonably certain that hypo- constitutes an ‘off-label’ use, and for some hospitals in par-
thermia is really effective?’ In simple terms, the concern is ticularly litigious settings, this may be a deterrent to its use
that combining early ‘smaller’ trials of a new therapy in in the treatment of HIE. Yet, it might be argued that many
a meta-analysis is similar to taking multiple ‘peeks’ at data medications commonly used in neonatology are used ‘off-
and prematurely stopping a prospective trial e it increases label’ without hesitation. Although many neonatologists
the chance of a type I error. One way of addressing this has may not be familiar with the Blanketrolâ, it is likely to be
recently been proposed, using the ‘optimal information size’ found at least in limited numbers in the operating rooms
concept to calculate the number of infants needed in pooled and/or patient equipment departments of many North
trials to confirm that a treatment effect exists and that American hospitals and it is sold in other parts of the world,
estimates of the treatment effect size are reasonable.5 Inves- including Europe and the Middle East. Thus, some neonatal
tigators in the Brain Cooling and NICHD trials would probably intensive care units (NICUs) would be able to implement the
dispute their characterisation as ‘small’ given the time and NICHD cooling protocol after investing the time necessary
effort involved in accruing over 200 patients each, but in to find and connect the equipment and supplies within their
fact compared to some adult clinical trials in conditions of own institution, and adapting the NICHD protocol for their
greater prevalence, they are relatively small trials. The re- own use (e.g. creating a graphic presentation of the proto-
sults of three additional large RCTs that have closed enroll- col), as we have done at the University of Michigan Health
ment (the Chinese selective head cooling trial, the TOBY System. Others would need to purchase at least two of the
trial and the ICE trial) should provide sufficient numbers of pa- devices, one for the patient and one for back-up. It must be
tients to address the concern about optimal information size.5 emphasised that safe use of the Blanketrolâ device for hy-
pothermia in neonates depends on strict adherence to the
NICHD protocol, including use of a second, adult-sized mat-
What are the practical barriers to tress suspended from a pole to damp out temperature fluc-
implementation? tuations that would otherwise occur with use of a system
overpowered for neonates. The equipment setup and the
At the time of writing, there is no consensus that hypo- protocol are sufficiently complicated that body cooling
thermia should be the standard of care for infants with should not be undertaken on the spur of the moment, but
suspected HIE. The advice from authoritative sources that only after advance planning.
hypothermia should only be undertaken using established In Europe, the TOBY trial hypothermia protocol is in the
protocols with known safety and efficacy is still sound.6,7 public domain on the Internet, but the hypothermia device
For those convinced by the published evidence, cost and/ used was not available for purchase at the time of writing,
or limited equipment availability may be barriers to imple- and safety and efficacy data have not yet been reported.
mentation of cooling within an individual centre, while ge- Geographical barriers to cooling are relevant in some
ography may pose an additional barrier to implementation regions, because there is a consensus amongst all published
of neonatal cooling at a regional level. However, lack of protocols that cooling needs to be initiated within 6 h of
availability of detailed cooling protocols should no longer birth. This 6-h limit in clinical trials was based on animal
be an issue. data, which suggest that the effectiveness of cooling dimin-
The Olympic Cool Capâ System for selective head cooling ishes as time increases from the hypoxiceischaemic insult to
is the only FDA-approved device for treatment of perinatal initiation of cooling, with the closing of the ‘window’ at
HIE with hypothermia. It is relatively expensive compared a time between 5.5 and 8 h after the insult.9 In many larger
with other cooling systems. Its availability is limited at the countries, babies may be born at great distances from
present time due to the manufacturer’s phased roll-out, regional NICUs, so that they could not possibly reach a
which has been made necessary by the fact that the ‘cooling centre’ within 6 h of birth. If such infants are to ben-
commercial device, a substantial update of the clinical efit from therapeutic hypothermia, practical strategies will

3. 32 J.D.E. Barks
be needed to evaluate and initiate cooling in the birth hos- not influence the effect of cooling on outcome. In a post
pital, in collaboration with a regional referral NICU and prior hoc analysis, data from the NICHD body cooling trial were ex-
to neonatal transport. Such strategies were implemented in amined to determine whether there were factors identifi-
the published trial of Eicher et al., which demonstrated able prior to 6 h of life that could be incorporated into
overall efficacy1,10; cooling, prior to and/or during trans- either a scoring scheme or a decision tree scheme to deter-
port, was also incorporated into the TOBY and ICE trial pro- mine which infants were likely to benefit from cooling and
tocols. Until the results of the latter two trials are which would not.13 Although both approaches yielded means
published, individual centres are left to consider whether of identifying infants in the NICHD study population that
the results of the Eicher trial and the experimental evidence would not benefit from cooling, very few infants (2e3%)
are compelling enough to support implementation of a trans- had such scores, and neither scheme has yet been validated
port cooling protocol and, as we have done, they may prospectively in a new patient population. Thus, these scor-
choose to adapt Eicher’s transport cooling protocol to fit ing systems are not currently suitable as a tool for excluding
their centre’s cooling strategy. Given the limited evidence, infants from hypothermic intervention. Similarly, the Brain
safety concerns should be paramount in developing proto- Cooling investigators have not advocated use of aEEG criteria
cols for initiation of cooling at birth hospitals without NICUs to exclude infants from cooling. To date, the most that can
and/or during neonatal transport. In particular, there must be said, based on common sense and consistent with clinical
be a method of safely measuring core (oesophageal or rec- trial protocols, is that infants judged to be in extremis are
tal) temperature; standard bedside digital thermometers not cooling candidates.
may not register temperatures in the target range for the All currently published cooling protocols limit initiation
established cooling protocols. of treatment to within 6 h of birth. There is no human data
An often-overlooked barrier to cooling is denial of, or to support efficacy of cooling initiated >6 h after birth. This
failure to recognise, an infant’s birth depression and/or may be an important question for future study. However,
encephalopathy. Some practitioners delivering babies per- the question could become largely irrelevant if recognition
sist in paradoxically responding to a litigious environment of eligibility improves as evidence for, and acceptance of,
by not drawing umbilical cord blood gases, in a misguided cooling increase, and if protocols for evaluation and initia-
effort to ‘protect’ themselves. This omission eliminates tion of cooling at birth hospitals become more widely
valuable information that could identify an infant requiring disseminated.
further evaluation for cooling, or that might suggest a non-
hypoxiceischaemic explanation for neonatal depression.
Although not every infant 36 weeks’ gestation with a cord How might we improve upon our results with
pH 7 is a candidate for cooling, all need evaluation for cooling?
signs of encephalopathy. Sometimes, evaluation for enceph-
alopathy and consideration for cooling may be overlooked in For those convinced by the current evidence, this is an
the flurry of managing other critical complications associ- important question, which will be the focus for future re-
ated with birth depression, such as hypotension, blood search. Several possible strategies include combinations of
loss, meconium aspiration syndrome and persistent pulmo- potentially neuroprotective drugs with hypothermia; earlier
nary hypertension. initiation of cooling; increased depth or duration of cooling;
Even if the need to evaluate for encephalopathy is and prolonged rewarming. There is some experimental
recognised, clinicians vary in their level of experience evidence supporting additive or synergistic effects of drugs
and confidence in performing the neonatal neurological with hypothermia, including some FDA-approved drugs (top-
examination, and the signs of encephalopathy can be iramate, N-acetylcysteine).14,15 Many other FDA-approved
subtle, may not always fit neatly into published grading drugs, e.g. anticonvulsants and anti-inflammatories, still
schemes, can evolve over time, or may be mistaken for the need to be evaluated experimentally in combination with
side-effects of sedatives or of severe systemic illness.11 Am- cooling, to provide a foundation for future human studies.
plitude-integrated EEG (aEEG), which was originally used in Although experimental evidence indicates that cooling is
the Cool Cap trial to exclude infants anticipated to do well more effective when initiated earlier after the insult,9 the
without intervention, might become a useful way of decid- therapeutic time window for the benefit of cooling appears
ing whether an infant with equivocal findings on examina- to decrease as the severity of the initial insult increases.16
tion is ‘encephalopathic enough’ to benefit from cooling. The two largest trials to date were unable to detect an ef-
fect of time of initiation on outcome, perhaps because
cooling of most infants was started relatively late in the
Who should be cooled? Who should not be time window in both trials, between 4 and 5 h after birth.2,3
cooled? In the Eicher trial, cooling was initiated with ice packs prior
to or during transport, and enrollment time was thus earlier
The results of the Cool Cap trial indicated that the subgroup than in the two larger trials1; it is tempting to speculate
of infants with the most severe aEEG tracings (severe that earlier time of initiation facilitated detection of effi-
suppression plus seizures) did not improve with cooling, cacy, despite a smaller sample size than either the Brain
whereas the majority of infants, who had less severe aEEG Cooling or the NICHD trial. Perhaps the results of the re-
abnormalities, showed improved outcome with cooling.2 In maining RCTs or analysis of data from the VermonteOxford
contrast, a post hoc analysis from the Cool Cap trial,12 consis- Neonatal Encephalopathy registry will shed additional light
tent with the results of the NICHD body cooling trial,3 showed on the importance of initiation time. There is no evidence
that clinical encephalopathy grade (Sarnat II versus III) did to indicate that longer durations of cooling or rewarming,

4. Current controversies in hypothermic neuroprotection 33
or greater depth of cooling will result in better outcomes safety should be a priority, and data on cooled infants
(nor that they are safe), so the use of these approaches should be collected systematically, e.g. via the Vermonte
should be limited to evaluation in RCTs. Oxford Network Neonatal Encephalopathy registry or the
TOBY Register.
Neurodevelopmental outcome
As indicated earlier, it is not yet known whether the early Practice points
promising results with hypothermia based on evaluation at
12e24 months will translate into improved performance at Therapeutic hypothermia safely decreases the
school age. However, given the lack of any other therapies combined outcome of death or disability at 12e
for HIE on the immediate horizon, this limitation is not likely 24 months, in infants with apparent HIE.
to present a barrier to implementation for many neonatol- Published protocols and the necessary equipment
ogists, nor is it likely to deter parents from accepting the are available.
therapy. Use of therapeutic hypothermia for HIE is not yet
Of more immediate concern, neonatologists and pediat- the standard of care.
ric neurologists should remember that what we think we Only protocols with established safety and efficacy
know about the prognostic implications of early neurolog- should be used.
ical findings and early neuroimaging, such as magnetic
resonance imaging (MRI) is based almost entirely on data
from the pre-cooling era. As a neonatologist, who has
participated in the cooling of over 50 infants, using both
selective head cooling and the NICHD body cooling pro-
tocol, the author has noted that even infants that ulti- Research directions
mately do well at 18e24 months do not always show
a dramatic improvement in neurological status after re- Does hypothermia reduce disability in survivors,
warming, but rather many seem to show a gradual recovery independent of its effect on mortality?
over several days. Does hypothermia change the prognostic implica-
Regarding the prognostic implications of early MRI, e.g. tions of post-hypothermic neurological examina-
at 7e10 days after birth, there is tantalising (but inconsis- tion or MR imaging?
tent) preliminary data to suggest that cooling is associated Does earlier initiation of cooling, within the first
with a change in the pattern and/or extent of injury,17,18 6 h of life, improve outcome?
but this has so far not been evaluated in large numbers of Can drug plus hypothermia combinations safely
patients nor are there any reports so far of the correlation improve outcome?
of early MRI findings in cooled infants with 18e24 months’ Could longer duration or lower target temperature
neurodevelopmental outcome. It is possible that MRI find- safely improve outcome?
ings at 7e10 days in infants with encephalopathy treated
with cooling may represent reversible changes in evolution.
The latter question is likely to be addressed by the NICHD
trial investigators in a secondary analysis. In the meantime,
neonatologists and neurologists should exercise some cau-
References
tion in assuming that what they knew about prognostication
1. Eicher DJ, Wagner CL, Katikaneni LP, et al. Moderate hypother-
in the pre-cooling era will still be true in infants that have
mia in neonatal encephalopathy: efficacy outcomes. Pediatr
recently undergone therapeutic hypothermia, particularly Neurol 2005;32(1):11e7.
regarding infants with moderate encephalopathy or inter- 2. Gluckman PD, Wyatt JS, Azzopardi D, et al. Selective head cool-
mediate degrees of injury on MRI. ing with mild systemic hypothermia after neonatal encephalop-
athy: multicentre randomised trial. Lancet 2005;365(9460):
663e70.
Conclusion
3. Shankaran S, Laptook AR, Ehrenkranz RA, et al. Whole-body hy-
pothermia for neonates with hypoxiceischemic encephalopa-
Until the results of several, now-closed, RCTs are pub- thy. N Engl J Med 2005;353(15):1574e84.
lished, the lack of consensus for therapeutic hypothermia 4. Edwards AD, Azzopardi DV. Therapeutic hypothermia following
as the standard of care for infants with apparent HIE is perinatal asphyxia. Arch Dis Child Fetal Neonatal Ed 2006;
likely to continue. There is no evidence that one method of 91(2):F127e31.
cooling is superior. Although devices and protocols to 5. Kirpalani H, Barks J, Thorlund K, Guyatt G. Cooling for neona-
perform hypothermia according to already-published pro- tal hypoxiceischemic encephalopathy e is the answer in? Pedi-
tocols are now available, it is unclear how rapidly this atrics 2007;120(5):1131e2.
6. Higgins RD, Raju TN, Perlman J, et al. Hypothermia and perina-
technology will be disseminated while the results of the
tal asphyxia: executive summary of the National Institute of
remaining trials are pending. It is reasonable for some Child Health and Human Development workshop. J Pediatr
centres, particularly those that participated in the pub- 2006;148(2):170e5.
lished trials, to offer cooling according to the published 7. Blackmon LR, Stark AR. Hypothermia: a neuroprotective ther-
protocols. It is also reasonable for centres that are not yet apy for neonatal hypoxiceischemic encephalopathy. Pediatrics
convinced to await further evidence. In this interim period, 2006;117(3):942e8.

5. 34 J.D.E. Barks
8. Gray J, Geva A, Zheng Z, Zupancic J. CoolSim: using industrial 14. Liu Y, Barks JD, Xu G, Silverstein FS. Topiramate extends the
modeling techniques to examine the impact of selective head therapeutic window for hypothermia-mediated neuroprotec-
cooling in a model of perinatal regionalization. Pediatrics, in tion after stroke in neonatal rats. Stroke 2004;35(6):1460e5.
press. 15. Jatana M, Singh I, Singh AK, Jenkins D. Combination of systemic
9. Gunn AJ, Gunn TR. The ‘pharmacology’ of neuronal rescue hypothermia and N-acetylcysteine attenuates hypoxiceische-
with cerebral hypothermia. Early Hum Dev 1998;53(1):19e35. mic brain injury in neonatal rats. Pediatr Res 2006;59(5):
10. Eicher DJ, Wagner CL, Katikaneni LP, et al. Moderate hypother- 684e9.
mia in neonatal encephalopathy: safety outcomes. Pediatr 16. Iwata O, Iwata S, Thornton JS, et al. ‘Therapeutic time win-
Neurol 2005;32(1):18e24. dow’ duration decreases with increasing severity of cerebral
11. Leviton A, Nelson KB. Problems with definitions and classifications hypoxiaeischaemia under normothermia and delayed hypo-
of newborn encephalopathy. Pediatr Neurol 1992;8(2):85e90. thermia in newborn piglets. Brain Res 2007;1154:173e80.
12. Wyatt JS, Gluckman PD, Liu PY, et al. Determinants of out- 17. Inder TE, Hunt RW, Morley CJ, et al. Randomized trial of systemic
comes after head cooling for neonatal encephalopathy. Pediat- hypothermia selectively protects the cortex on MRI in term hypo-
rics 2007;119(5):912e21. xiceischemic encephalopathy. J Pediatr 2004;145(6):835e7.
13. Ambalavanan N, Carlo WA, Shankaran S, et al. Predicting out- 18. Rutherford MA, Azzopardi D, Whitelaw A, et al. Mild hypothermia
comes of neonates diagnosed with hypoxemiceischemic en- and the distribution of cerebral lesions in neonates with hypo-
cephalopathy. Pediatrics 2006;118(5):2084e93. xiceischemic encephalopathy. Pediatrics 2005;116(4):1001e6.