Determinants of Outcomes After Head Cooling for Neonatal Encephalopathy
OBJECTIVE. The goal of this study was to evaluate the role of factors that may determine the efficacy of treatment with delayed head cooling and mild systemic hypothermia for neonatal encephalopathy.
METHODS. A total of 218 term infants with moderate to severe neonatal encephalopathy plus abnormal amplitude-integrated electroencephalographic recordings, assigned randomly to head cooling for 72 hours, starting within 6 hours after birth (with the rectal temperature maintained at 34.5 ± 0.5°C), or conventional care, were studied. Death or severe disability at 18 months of age was assessed in a multicenter, randomized, controlled study (the CoolCap trial).
RESULTS. Treatment, lower encephalopathy grade, lower birth weight, greater amplitude-integrated electroencephalographic amplitude, absence of seizures, and higher Apgar score, but not gender or gestational age, were associated significantly with better outcomes. In a multivariate analysis, each of the individually predictive factors except for Apgar score remained predictive. There was a significant interaction between treatment and birth weight, categorized as ≥25th or <25th percentile for term, such that larger infants showed a lower frequency of favorable outcomes in the control group but greater improvement with cooling. For larger infants, the number needed to treat was 3.8. Pyrexia (≥38°C) in control infants was associated with adverse outcomes. Although there was a small correlation with birth weight, the adverse effect of greater birth weight in control infants remained significant after adjustment for pyrexia and severity of encephalopathy.
CONCLUSIONS. Outcomes after hypothermic treatment were strongly influenced by the severity of neonatal encephalopathy. The protective effect of hypothermia was greater in larger infants.
Neonatal encephalopathy associated with exposure to hypoxia/ischemia continues to be an important cause of acute neurologic injury, occurring in ∼2 to 3 cases per 1000 term live births in developed countries, with a higher incidence in less developed countries.1,2 Recently, we reported in a multicenter, randomized trial (the CoolCap trial) that head cooling with mild systemic hypothermia was associated with an improved rate of survival, without severe disability at 17 to 22 months of age, for all infants except those defined a priori as having the most severe changes on the baseline amplitude-integrated electroencephalographic (aEEG) recordings.3,4 There was no increase in rates of significant adverse events for cooled infants. These findings are consistent with those for whole-body cooling.5,6 A recent meta-analysis of those studies confirmed a significant overall effect of postpartum hypothermia but noted that there was considerable potential for additional improvement,7 and it is not certain that the 2 approaches are comparable in either their long-term or short-term outcomes.
It is of great importance for future clinical and experimental studies to determine the clinical factors that might have influenced the effectiveness of treatment. We reported previously, for example, that baseline aEEG recordings could stratify patients as those who might respond and those for whom therapy would have no effect.3 The Apgar scores during resuscitation and the clinical severity of encephalopathy after birth are also predictive of adverse neurodevelopmental outcomes,8 but it is not known whether they modulate responses to treatment. Similarly, growth-restricted infants and boys have higher rates of encephalopathy,9 but it is not known whether these factors affect the severity of outcomes or influence therapeutic efficacy. It is possible, for example, that being born small is associated with prenatal insults that might compromise the brain before the peripartum period.10 In the CoolCap study,3 clinical grading of encephalopathy and aEEG recordings were obtained before randomization, to improve the specificity of case selection and to provide objective pretreatment data on the severity of brain injury.11 We now present the results of an exploratory analysis to examine a range of possible clinical factors that might influence outcomes after selective head cooling for treatment of neonatal encephalopathy, including the encephalopathy grade, 5-minute Apgar score, gender, birth weight, aEEG findings, and pyrexia in control infants during the 76-hour interval after randomization, to generate hypotheses for future studies.
This study was performed in 25 perinatal centers with a trial design registered with the US Food and Drug Administration under the investigational device exemption/premarket approval program. The institutional review board of each center approved the protocol, and written informed consent was obtained from parents before randomization. From July 1999 to January 2002, infants at ≥36 weeks of gestation with acute encephalopathy were recruited by using a stepwise protocol including clinical evidence of exposure to perinatal hypoxia/ischemia, an abnormal neurologic examination, and an abnormal aEEG recording. The entry criteria, method of randomization, exclusion criteria, defined adverse events, data collection, independent data safety and monitoring board, general management, primary outcome of death or severe disability at 18 months of age, safety outcomes, and interaction between severity of aEEG changes and response to hypothermia were reported elsewhere.3,4
Study Entry Criteria
The required clinical criteria were an Apgar score of ≤5 at 10 minutes after birth, continued need for resuscitation (including endotracheal or mask ventilation) at 10 minutes after birth, or severe acidosis (defined as either pH of <7.00 or base deficit of ≥16 mmol/L in an umbilical cord blood sample or an arterial or venous sample obtained within 60 minutes after birth). The infants were then assessed, by a certified examiner, for evidence of moderate (grade 2) or severe (grade 3) encephalopathy according to criteria modified from those described by Sarnat and Sarnat,8 including lethargy, stupor, or coma with ≥1 of hypotonia, abnormal reflexes including oculomotor or pupillary abnormalities, absent or weak suck, or clinical evidence of seizures.
Infants who satisfied the aforementioned criteria then underwent ≥20 minutes of aEEG recording (Lectromed, Letchworth, United Kingdom) by investigators. The aEEG recording could be performed at any time after 1 hour of age, except within 30 minutes after intravenous administration of anticonvulsant therapy, provided the results were available before 5.5 hours. Infants were selected for randomization if they had a moderately or severely abnormal background aEEG voltage (moderate: upper margin of aEEG activity above 10 μV and lower margin below 5 μV; severe: upper margin below 10 μV) and/or electroencephalographically determined seizures (identified by a sudden increase in voltage accompanied by narrowing of the band of aEEG activity and followed by a brief period of suppression).11 Infants with aEEG seizures could be recruited despite normal or mild aEEG voltage changes, because seizures were considered to be a significant adverse prognostic factor.12
For infants assigned randomly to head cooling, a cooling cap (Olympic Medical Cool Care System; Olympic Medical, Seattle, WA) was fitted around the head for 72 hours.13–15 The system consisted of a small, thermostatically controlled, cooling unit and a pump that circulated water through the cooling cap. The initial water temperature was set to 8°C to 12°C. All infants were nursed under a radiant overhead heater, which was servo-controlled to the abdominal skin of the infant and adjusted to maintain the rectal temperature at 34.5 ± 0.5°C. Adjustments were made to the cooling cap water temperature to stay within these limits. At the time of initiation of hypothermia, the overhead heater was turned off for 20 to 30 minutes to accelerate cooling; the heater was turned back on once the rectal temperature had decreased to 35.5°C.
At the end of the 72-hour cooling period, the cooled infants were rewarmed slowly, at no more than 0.5°C per hour, until their temperature was within the normal temperature range. Cooling was discontinued earlier if the parents withdrew consent or if discontinuation was required for clinical reasons (eg, extracorporeal membrane oxygenation), in the opinion of the attending neonatologist.
Infants assigned randomly to the control noncooled group were cared for under an overhead radiant heater, which was servo-controlled to the infants' abdominal skin temperature to maintain the rectal temperature at 37.0 ± 0.2°C. The infants received standard clinical care for their center, under the care of the attending neonatologist.
The primary outcome was the combined incidence of death and severe neurodevelopmental disability in survivors at 18 months of age. Severe neurodevelopmental disability was defined as (1) gross motor function classification levels 3 through 5 (nonambulatory, sits with support applied to lower back, or limited or no self-mobility),16 (2) Bayley Mental Developmental Index17 of <70, or (3) bilateral cortical visual impairment. The primary outcome was analyzed according to the intent-to-treat principle. Two-sided P values of <.05 were considered statistically significant.
Exploratory analysis was performed by using logistic regression analyses for the primary outcome for treatment plus selected potential covariates, including grade of encephalopathy, 5-minute Apgar score, gender, birth weight, time of randomization, aEEG background, and presence of seizures determined with aEEG recording at the time of randomization, first individually and then with all covariates included in a multivariate model. We were unable to examine the specificity of changes in the arterial cord blood samples because of limited sample size for this parameter. In view of the significant effect of birth weight reported below, gestational age was also tested individually. To explore more thoroughly the effect of birth weight on outcomes, a logistic regression model with treatment/subgroup interaction terms in which birth weight was dichotomized as <25th percentile for gender at term gestation18 versus ≥25th percentile was used. Among the 218 patients, there were 72 infants (39 cooled and 33 control) in the <25th percentile group and 146 (69 cooled and 77 control) in the ≥25th percentile group. Using lower percentile cutoff points would have resulted in too few infants in the lower-weight group even for exploratory purposes. We then examined separately the hypothesis that the effect of birth weight on outcomes in the control group might be related to susceptibility to pyrexia, defined as ≥38°C in 4 hourly measurements during the observation period. Other factors, including race/ethnicity, location or number of patients treated at the sites, and maternal complications were tested separately in the main multivariate model. Outcome incidences were tested by using Fisher's exact test.
Baseline and Maternal Characteristics
Outcome data at 18 months were available for 218 of 234 patients enrolled originally (93%3); 8 infants from each group were lost to follow-up monitoring. Baseline patient characteristics for the groups were generally well balanced (Table 1). Because of stratification of treatment randomization according to participating site only and the generally small number of patients enrolled at each site, the Apgar score at 5 minutes after birth and aEEG background activity showed trends toward more-severe injury for infants assigned randomly to cooling. Maternal characteristics for the 218 patients with 18-month outcome data were similar to those of the total 234 patients and were generally balanced, with the exception that more control mothers experienced complications during pregnancy and labor (Table 2). These factors were tested separately in the multivariate logistic regression model (see below).
Staged Logistic Regression Analysis
In a logistic regression analysis with treatment group, the grade of encephalopathy at randomization showed the greatest predictive value. In order, greater severity of aEEG background changes, presence of aEEG seizures, lower continuous 5-minute Apgar score, and greater birth weight (in 100-g steps) were also associated with adverse outcomes (Table 3). The dichotomized Apgar score did not reach significance, although the direction of change was similar to that for the continuous Apgar score. There was no effect of gender, gestational age, or age at randomization (Table 3). The apparent effect of treatment with cooling was similar across each of these analyses, and P was <.05 after adjustment for encephalopathy grade, aEEG background, and continuous Apgar score (Table 3).
Full Logistic Regression Model
When all of the factors were examined together in a logistic regression model (Table 4), encephalopathy grade, aEEG background, aEEG seizures, and birth weight remained independently associated with unfavorable outcomes. In contrast, Apgar score was not significantly predictive when corrected; there was again no effect of gender. In this combined model, cooling was associated significantly with reduced risk of unfavorable outcomes. When this prespecified secondary analysis was repeated with only the significant variables (treatment group, encephalopathy grade, aEEG background, aEEG seizures, and birth weight), the outcome of the analysis was similar and each variable remained significantly predictive (data not shown).
Effect of Severity of Encephalopathy
The encephalopathy grade at the time of randomization was the single most predictive covariate, both alone and in combination. There was no significant interaction between encephalopathy grade and hypothermia treatment. The incidence of unfavorable outcomes was reduced similarly after cooling for infants with either moderate (grade 2) encephalopathy (28 [45%] of 62 cooled infants and 39 [57%] of 69 control infants) or severe (grade 3) encephalopathy (28 [70%] of 40 cooled infants and 32 [91%] of 35 control infants). There were few infants with mild (grade 1) encephalopathy (2 of 5 cooled infants and 0 of 3 control infants); data were missing for 1 cooled infant and 3 control infants.
Exploratory Analysis of Birth Weight
The analyses in Tables 3 and 4 suggested that there was a detrimental effect toward unfavorable primary outcomes with increasing weight. Infants in the control group with favorable outcomes were smaller than infants in the control group with unfavorable outcomes, weighing 3229 ± 633 vs 3630 ± 597 g (P = .001, Mann-Whitney test), and had less-severe encephalopathy grades (2.0 ± 0.4 vs 2.5 ± 0.5; P = .0001). To understand this effect, additional explorations were conducted in which birth weight was dichotomized. The birth weight of the infants of <25th percentile was 2785 ± 290 g, compared with 3785 ± 498 g. As shown in Table 5, there was a marked interaction between cooling and birth weight (P = .003), as well as a significant effect of birth weight (P = .005), controlled for severity of encephalopathy. Consistent with the previous analysis, infants in the control group with birth weights of ≥25th percentile had a significantly higher rate of unfavorable primary outcomes than did those with weights of <25th percentile (59 [77%] of 77 infants vs 14 [42%] of 33 infants; P < .001). In contrast, there was a highly significant cooling effect in the ≥25th percentile group (unfavorable outcomes for 35 [51%] of 69 cooled infants vs 59 [77%] of 77 control infants; P < .002) but not in the <25th percentile group (24 [62%] of 39 cooled infants vs 14 [42%] of 33 control infants; P = .16). Acute complications in labor were reported for 124 (85%) of 146 infants of ≥25th percentile and 51 (71%) of 72 infants of <25th percentile (odds ratio [OR]: 2.32; 95% confidence interval [CI]: 1.18–4.56; P = .018, Fisher's exact test). The number needed to treat for benefit in the ≥25th percentile group was 3.8 (95% CI: 2.5–9.6).
Rectal temperature measurements confirmed that the great majority of infants in both groups achieved and maintained their target temperature ranges (Fig 1). Thirty-four control patients had rectal temperatures of ≥38°C at any time during the 76-hour monitoring period, of whom 28 had unfavorable outcomes; of the remaining 76 patients without pyrexia at any time, 45 had unfavorable outcomes (OR: 3.2; 95% CI: 1.2–8.4; P = .028, Fisher's exact test). Only 11 patients in the hypothermia group (including 1 patient who was not cooled) experienced temperatures of ≥38°C at randomization or at rewarming, of whom 9 had unfavorable outcomes; of the 97 patients in the hypothermia group who did not experience pyrexia, 50 had unfavorable outcomes (OR: 4.2; 95% CI: 0.97–18; P = .11).
There was a significant but modest correlation between greater birth weight and the frequency of pyrexia of ≥38°C in 4 hourly measurements during the 76-hour monitoring period in the control group (r2 = 0.05; P = .011). Both the incidence of pyrexia and birth weight (in 100-g steps) were significant when tested separately in a multivariate logistic regression model incorporating severity of encephalopathy, background aEEG changes, and presence of seizures (pyrexia: OR: 2.06; 95% CI: 1.04–4.09; P = .038; birth weight: OR: 1.1; 95% CI: 1.02–1.2; P = .012). When the incidence of pyrexia and weight were included together in the regression model, there was no significant effect of pyrexia (OR: 1.9; 95% CI: 0.9–3.7; P = .086) but birth weight remained significant (OR: 1.09; 95% CI: 1.00–1.19; P = .045).
Logistic regression indicated that there was no significant interaction effect for gender (P = .16) or for Apgar score at 5 minutes (n = 213; P = .28). Maternal ethnicity/race (main effect, P = .85), delivery method (vaginal versus cesarean section; main effect, P = .9), and maternal complications during pregnancy or labor (yes versus no; main effect, P = .73) had no significant main effects or interactions with cooling treatment. Similarly, there was no effect on outcomes or interaction with treatment for the number of infants enrolled at each site (<10 vs 10–19 infants; main effect, P = .5; ≥20 vs 10–19 infants; main effect, P = .61) or non–United States versus United States sites (main effect, P = .42).
The present secondary analysis of the CoolCap study extends recent findings that hypothermia can reduce rates of death or disability at 18 months of age after severe neonatal encephalopathy3,5 to show that outcomes after hypothermic treatment are affected greatly not only by the severity of encephalopathy but also by birth size. There was a striking interaction between treatment and weight, such that larger infants had higher rates of adverse outcomes in the control group but also a substantial apparent reduction in adverse outcomes after delayed hypothermia, compared with no significant effect of hypothermia for smaller infants. Furthermore, spontaneous pyrexia was associated with a marked increase in adverse outcomes for control infants, controlled for severity of encephalopathy. Although there was a small intriguing association between pyrexia and larger birth size, it seemed that this could not fully explain the adverse outcomes associated with greater birth weight for control infants.
This analysis was designed to explore more thoroughly patient data obtained before randomization, to identify variables that might affect outcomes in both treatment and control groups. This posthoc approach has 2 well-known limitations. First, the subset sample sizes are much smaller than the overall treatment group, which leads to an increased risk of false-negative results. Second, multiple comparisons increase the risk of false-positive findings. Nonetheless, this analysis can generate hypotheses that can be tested in ongoing clinical trials. For the purposes of this analysis, P < .05 was taken to indicate variables of potential future interest.
As expected, outcomes were influenced greatly by the neurologic status before randomization, with independent effects of both the severity of encephalopathy and the aEEG variables. There was a significant treatment effect, reflecting an imbalance in randomization between the groups that was reported previously.4 Although the 5-minute Apgar score was weakly predictive of outcomes when used by itself, the effect was not significant when corrected for encephalopathy grade, consistent with previous data,9,19 which emphasizes that the Apgar score is effectively a surrogate measure for encephalopathy. It is well established that the grade of encephalopathy determined at ≥24 hours of age is closely associated with outcomes.8,20–22 The present findings strongly support the value of this assessment even in the first few hours of life.14 The incidence of adverse outcomes in the control group was similar to previous reports, although there was a slightly higher incidence of death or disability for infants with moderate encephalopathy,8,20,21 which probably reflects selection of more-severe cases arising from the requirement for abnormal aEEG findings. Therefore, these findings are not necessarily representative of the predictive value of encephalopathy assessments in isolation.
It is notable that our analysis suggested that hypothermia exerted neuroprotective effects in infants with both moderate and severe encephalopathy. This seems to conflict with experimental evidence suggesting that hypothermia is less effective after the most severe insults.23–25 However, a recent multicenter trial of whole-body cooling also suggested protective effects in both the moderate and severe encephalopathy groups.5 We speculate that some infants with severe encephalopathy had slowly evolving injuries (as reported in some experimental paradigms26) and therefore retained the potential to respond to hypothermia treatment initiated within 6 hours after birth. Conversely, the presence of severe abnormalities in the aEEG recording (especially the combination of suppressed background activity and seizures) was associated with failure to respond to hypothermia,3 which likely reflects a subset of infants with the most-profound or most-advanced injuries.4 These factors are closely interrelated, because experimentally more-severe injury leads to more-rapid secondary evolution.27
Unexpectedly, the present analysis indicated that birth weight was an important determinant of outcomes. In the control group, lighter infants tended to have improved outcomes, even when controlling for the severity of clinical encephalopathy in the multivariate analysis. It is important to note that this group represents the lower end of the reference range (weight below the 25th percentile for term) and not growth-restricted infants. This finding has not been reported previously, although there is some evidence that exposure to preeclampsia may reduce the risk of cerebral palsy in premature infants.28,29 In experimental studies, previous exposure to hypoxia, under specific circumstances, protected against injury.30–32 It can be speculated that this so-called preconditioning might provide a mechanism through which some smaller infants might have less-severe injuries.
An alternative speculation is that, because of their greater body mass, greater thermal inertia, and lesser relative surface area, larger infants might have a greater tendency toward postasphyxial pyrexia, compared with smaller infants. It is known that even minor, delayed, induced increases in brain temperature after exposure to hypoxia/ischemia can exacerbate subsequent injury.33–36 Consistent with this possibility, there was a modest but significant correlation between birth weight and the incidence of pyrexia in 4 hourly measurements for control infants; in turn, pyrexia was associated with worse outcomes. Although more-frequent pyrexia might have contributed in part to the effect of birth weight, larger birth size was still independently associated with adverse outcomes after multivariate adjustment for severity of encephalopathy and pyrexia, which suggests that additional mechanisms are likely to be involved.
The adverse effects of pyrexia in adults with acute brain injuries have been accepted for some years, and active antipyretic therapy is widely used in adult neurointensive care.37 However, the CoolCap trial seems to be the first study with prospective, detailed, core temperature measurements demonstrating evidence of this association in newborn infants. The mechanisms of such spontaneous pyrexia in the noncooled infants are unknown but likely include heat production related to intense seizures,38 induction of inflammatory cytokines,39,40 and possibly overshoot heating related to operation of the servo-controlled overhead heater.36 Because only 3 control infants had sepsis,3 it is likely that the adverse association was related directly to the changes in temperature. Because there has never been a therapeutic reason to make sick neonates pyrexial, our findings reinforce the importance of measuring core temperatures and preventing hyperthermia in encephalopathic infants,35 even if hypothermia is not used as therapy.
In the treatment group, our analysis suggested that heavier infants showed a very marked beneficial response to hypothermia, whereas lighter infants did not. The unexpected influence of birth weight on the response to treatment needs to be explored in future studies. At present, there is no support from experimental studies for this concept, and our findings may represent a statistical artifact. Because the larger infants in the control group had worse outcomes, there might simply have been more scope for improvement after hypothermia. Alternatively, the higher rate of labor complications observed for larger infants raises the possibility that the infants had a higher rate of acute peripartum insults and that the injury process was at a less-advanced stage at the time of initiation of treatment, when it was still amenable to neuroprotective intervention.41 In the present study, however, there was no effect of labor complications per se on the outcome of treatment.
Alternatively, the effect of infant weight might be related to cap temperatures. A recent study with 7 healthy piglets that had not received an hypoxic/ischemic insult reported that, for a given cap temperature, underlying cortical temperatures were significantly lower in smaller piglets,42 which raises the possibility that the brain might be overcooled in small infants. However, others showed that head cooling in newborn piglets was highly neuroprotective after hypoxia/ischemia, despite a similar level of cortical cooling.43 Moreover, piglets have much smaller brains than term infants, and it is unclear whether these data apply to human infants. Finally, Iwata et al42 examined the impact of fixed cap temperatures. However, although brain temperatures could not be measured in the CoolCap trial, the investigators' experience is that smaller infants consistently required substantially higher (less cold) cap temperatures than larger infants. It is possible, therefore, that smaller infants might have received less-effective head cooling, with a reduced brain to rectal temperature gradient.
Male infants are known to have small increases in the incidence of encephalopathy at term,9 and limited data in neonatal rats raise the possibility that hypothermia might be more protective for female infants.24 In contrast, the present analysis suggested that there was no independent effect of gender on outcomes and no evidence of different effects of treatment for boys and girls. Furthermore, we found no effect of race, presence of maternal and labor complications, or study site on therapeutic outcomes.
Although this analysis adds to the evidence that head cooling with mild hypothermia is therapeutically valuable in neonatal encephalopathy, it is important to note that the outcomes studied to date are only from 18 months of age. Given the experimental data showing that the optimal temperatures for neuroprotection differed between the cortex and basal ganglia44 and the preliminary clinical finding that head cooling was associated with a greater reduction in the incidence of cortical injury,45 long-term follow-up monitoring is clearly needed.
The present results confirm that early encephalopathy grading is a strong independent predictor of long-term outcomes of neonatal encephalopathy, that aEEG recording provides additional predictive value, and that there is no independent effect of Apgar scores. Infants with severe and moderate encephalopathy demonstrated similar trends in responding to hypothermia treatment. Unexpectedly, in this study larger control infants had a greater incidence of death or disability even after adjustment for the effects of encephalopathy. Strikingly, however, the larger infants showed a greater therapeutic response to hypothermia. Although there was an association between greater birth weight and the incidence of pyrexia, there was an apparent independent effect of birth weight. These unexpected findings indicate important hypotheses that should be examined in current and future studies. The adverse outcomes associated with pyrexia in the present study strongly support previous recommendations that pyrexia should be rigorously prevented.46
The study was funded by Olympic Medical (Seattle, WA). Drug and Device Development Co (Redmond, WA) entered and held the data and performed all analyses according to the instructions of the scientific advisory committee. The sponsor supported the study financially, provided administrative support to the sites, supplied the aEEG monitors and the cooling devices, and monitored initial data recording and accuracy but had no input into the manuscript.
The following investigators participated in the CoolCap Study Group: executive committee: P. D. Gluckman (chair, co-principal investigator), J. S. Wyatt (co-principal investigator), A. J. Gunn (scientific officer); scientific advisory committee: J. S. Wyatt (chair), R. Ballard, A. D. Edwards, D. M. Ferriero, P. D. Gluckman, A. J. Gunn, R. Polin, C. Robertson, A. Whitelaw; data safety committee: R. Soll (chair), M. Bracken, C. Palmer, M. Heymann, A. Wilkinson; hospital investigators: J. Kaiser (Arkansas Children's Hospital; 11 patients), M. Battin, D. Armstrong (University of Auckland-National Women's Hospital, New Zealand; 11 patients), J. Khan (Children's Memorial Hospital and Prentice Women's Hospital of Northwestern Memorial Hospital; 3 patients), T. Raju (University of Illinois at Chicago Medical Center; 1 patient), R. Polin, R. Sahni, U. Sanocka (Children's Hospital of New York-Presbyterian, Columbia University; 18 patients), A. Rosenberg, J. Paisley (Children's Hospital of Denver; 23 patients), R. Goldberg, M. Cotton (Duke University Medical Center; 14 patients), A. Peliowski, E. Phillipos (Royal Alexandra Hospital/University of Alberta Hospital; 20 patients), D. Azzopardi, A. D. Edwards (Hammersmith Hospital, London, United Kingdom; 1 patient), F. Northington (Johns Hopkins University; 2 patients), J. Barks, S. Donn (University of Michigan Medical Center-Mott Children's Hospital; 12 patients), B. Couser (Children's Hospital and Clinics of Minneapolis; 16 patients), D. Durand (Children's Hospital and Research Center at Oakland; 8 patients), K. Sekar (Children's Hospital of Oklahoma; 4 patients), D. Davis, M. Blayney (Children's Hospital of Eastern Ontario/The Ottawa Hospital; 1 patient), S. Adeniyi-Jones (AI Dupont Children's Hospital at Thomas Jefferson University Medical Center; 6 patients), T. Yanowitz (Magee Women's Hospital/Children's Hospital of Pittsburgh; 10 patients), R. Guillet, N. Laroia (Golisano Children's Hospital at Strong; 10 patients), N. Finer, F. Mannino (University of California, San Diego, Medical Center Hillcrest; 8 patients), J. Partridge (University of California, San Francisco, Children's Hospital; 2 patients), D. Davidson (Schneider Children's Hospital; 14 patients), A. Whitelaw (Southmead Hospital, Bristol, United Kingdom; 13 patients), M. Thoresen (St Michael's Hospital, Bristol, United Kingdom; 8 patients), J. S. Wyatt, F. O'Brien (University College Hospital, London, United Kingdom; 4 patients), B. Walsh (Vanderbilt Children's Hospital; 13 patients), J. Perciaccante, M. O'Shea (Wake Forest University Baptist Medical Center; 1 patient); manufacturer's representatives: J. Jones, T. Weiler, J. Mullane, D. Hammond, J. Parnell (Olympic Medical, Seattle, WA).
We thank the many technicians, nurses, physicians, and scientists at the participating sites who contributed to the development and implementation of this project and the parents who consented to enrollment of their infants in this trial; we appreciate their trust in us under conditions of great stress and anxiety. We thank the many charities and research funding agencies that supported the preliminary research necessary for this study.
- Accepted December 29, 2006.
- Address correspondence to Alistair Jan Gunn, MBChB, PhD, Department of Physiology, Faculty of Medicine and Health Science, University of Auckland, Private Bag 92019, Auckland, New Zealand. E-mail:
Financial Disclosure: Olympic Medical provided financial grants to the University of Auckland (Dr Gluckman) and University College London (Dr Wyatt) to cover the costs of administering the CoolCap trial. Each participating trial site (not the individual site investigators) received fixed partial reimbursement for each infant enrolled, covering the additional costs of performing the trial, including clinical time, laboratory tests, and neurodevelopmental assessments, and local trial administration. Dr Azzopardi authored a manual of aEEG interpretation that was distributed by Olympic Medical for use in the trial. Olympic Medical loaned equipment to Drs Gunn, Thoresen, Whitelaw, and Wyatt for pilot studies preceding the trial. The University of Auckland has applied for a related patent that names Dr Gunn; however, Drs Gunn and Gluckman have no financial interest.
- ↵Vannucci RC, Perlman JM. Interventions for perinatal hypoxic-ischemic encephalopathy. Pediatrics.1997;100 :1004– 1014
- ↵Edwards AD, Azzopardi DV. Therapeutic hypothermia following perinatal asphyxia. Arch Dis Child Fetal Neonatal Ed.2006;91 :F127– F131
- ↵Badawi N, Kurinczuk JJ, Keogh JM, et al. Antepartum risk factors for newborn encephalopathy: the Western Australian case-control study. BMJ.1998;317 :1549– 1553
- ↵al Naqeeb N, Edwards AD, Cowan FM, Azzopardi D. Assessment of neonatal encephalopathy by amplitude-integrated electroencephalography. Pediatrics.1999;103 :1263– 1271
- ↵McBride MC, Laroia N, Guillet R. Electrographic seizures in neonates correlate with poor neurodevelopmental outcome. Neurology.2000;55 :506– 513
- ↵Gunn AJ, Gluckman PD, Gunn TR. Selective head cooling in newborn infants after perinatal asphyxia: a safety study. Pediatrics.1998;102 :885– 892
- ↵Battin MR, Dezoete JA, Gunn TR, Gluckman PD, Gunn AJ. Neurodevelopmental outcome of infants treated with head cooling and mild hypothermia after perinatal asphyxia. Pediatrics.2001;107 :480– 484
- ↵Battin MR, Penrice J, Gunn TR, Gunn AJ. Treatment of term infants with head cooling and mild systemic hypothermia (35.0°C and 34.5°C) after perinatal asphyxia. Pediatrics.2003;111 :244– 251
- ↵Bayley N. Bayley Scales of Infant Development. 2nd ed. San Antonio, TX: Psychological Corp; 1993
- ↵National Center for Health Statistics. 2000 CDC growth charts: United States. Available at: www.cdc.gov/growthcharts. Accessed March 2, 2007
- ↵Perlman JM, Risser R. Can asphyxiated infants at risk for neonatal seizures be rapidly identified by current high-risk markers? Pediatrics.1996;97 :456– 462
- ↵Robertson CM. Long-term follow-up of term infants with perinatal asphyxia. In: Stevenson DK, Sunshine P, eds. Fetal and Neonatal Brain Injury. Cambridge, United Kingdom: Cambridge University Press; 2002:829–858
- ↵O'Shea TM, Klinepeter KL, Dillard RG. Prenatal events and the risk of cerebral palsy in very low birth weight infants. Am J Epidemiol.1998;147 :362– 369
- ↵Baena RC, Busto R, Dietrich WD, Globus MY, Ginsberg MD. Hyperthermia delayed by 24 hours aggravates neuronal damage in rat hippocampus following global ischemia. Neurology.1997;48 :768– 773
- ↵Gunn AJ, Gunn TR. Effect of radiant heat on head temperature gradient in term infants. Arch Dis Child Fetal Neonatal Ed.1996;74 :F200– F203
- ↵Martin-Ancel A, Garcia-Alix A, Pascual-Salcedo D, Cabanas F, Valcarce M, Quero J. Interleukin-6 in the cerebrospinal fluid after perinatal asphyxia is related to early and late neurological manifestations. Pediatrics.1997;100 :789– 794
- ↵Rutherford MA, Azzopardi D, Whitelaw A, et al. Mild hypothermia and the distribution of cerebral lesions in neonates with hypoxic-ischemic encephalopathy. Pediatrics.2005;116 :1001– 1006
- ↵Niermeyer S, Kattwinkel J, Van Reempts P, et al. International Guidelines for Neonatal Resuscitation: an excerpt from the Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: International Consensus on Science. Pediatrics.2000;106(3) . Available at: www.pediatrics.org/cgi/content/full/106/3/e29
- Copyright © 2007 by the American Academy of Pediatrics