Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial

Article Outline

• Copyright

Gluckman PD, Wyatt JS, Azzopardi D, Ballard R, Edwards AD, Ferriero DM, et al. on behalf of the CoolCap Study Group. Lancet 2005;365:663-70.

Context No current therapies exist to improve the outcome from neonatal encephalopathy. Experimental data have demonstrated benefit from cerebral hypothermia following hypoxia-ischemia.

Objective To determine the efficacy of selective head cooling with mild systemic hypothermia on survival free of severe disability at 18 months of age in newborn infants with moderate or severe hypoxic-ischemic encephalopathy.

Design Multi-center, international, unmasked, randomized controlled trial.

Setting Twenty-five perinatal centers in the United States, United Kingdom, New Zealand, and Canada.

Participants Term newborn infants of at least 36 weeks gestation (n=234) with clinical evidence of moderate to severe neonatal encephalopathy (based on modified Sarnat criteria) and/or seizures, and clinical evidence of perinatal hypoxia-ischemia (Apgar score of 5 or less at 10 minutes, continued resuscitation or respiratory support at 10 minutes, or severe acidosis within one hour of birth) and with moderately or severely abnormal background activity or seizures on amplitude integrated electroencephalography (aEEG).

Interventions Infants were randomly assigned within 6 hours of birth to either head cooling (cooling cap with water circulated at 8-12°C [Olympic Medical Cool Care System]) with concomitant mild systemic hypothermia (rectal temperature 34-35°C) for 72 hours, or to conventional care (rectal temperature 36.8-37.2°C).

Main Outcome Measures The primary outcome was death or severe disability at 18 months of age. Secondary outcomes consisted of potential adverse effects of cooling and complications of hypoxia-ischemia including death, arrhythmia, coagulopathy, hypotension, abnormal renal function, electrolyte disturbance, bone marrow depression, raised liver enzymes, and metabolic acidosis.

Results Baseline clinical and aEEG characteristics were similar in the two groups. In the 218 (93%) infants followed to 18 months, there was no significant difference in the primary outcome of death or severe disability in the infants treated with hypothermia (unadjusted: 55% vs 66%, P=0.10, OR 0.61 [95% CI 0.34-1.09]; adjusted [for baseline aEEG amplitude, presence of seizures, and age at randomization]: P=0.05, OR 0.57 [95% CI 0.32-1.01]), or on any secondary outcome measures.

Two predefined subgroup analyses based on pre-randomization background aEEG amplitude abnormalities demonstrated: a) no apparent effect of delayed cerebral hypothermia on outcome in infants with severe aEEG abnormalities (n=46: 79% vs 68%, P=0.51, OR 1.8 [95% CI 0.49-6.4]), and b) benefit in infants with intermediate (moderate) aEEG abnormalities (n=172: 48% vs 58%, P=0.021, OR 0.47 [95% CI 0.26-0.87]; adjusted P=0.009, OR 0.42 [95% CI 0.22-0.88]) and the number needed to treat was six infants (95% CI 3-27).

Conclusions Selective head cooling with mild systemic hypothermia is a feasible therapeutic maneuver without clear-cut evidence of benefit in selected infants with moderate or severe hypoxic-ischemic encephalopathy. It may, however, improve the outcome for encephalopathic newborn infants with intermediate (moderate) abnormality on aEEG background amplitude.

Comment This important, well-designed randomized controlled trial investigated the biologically plausible hypothesis that hypothermia will improve outcome in newborn infants who have sustained a peripartum hypoxic-ischemic insult and who are at significant risk of death and/or neurological sequelae. In the study overall there was no definite benefit of delayed selective head cooling with mild systemic hypothermia on adverse outcome at 18 months of age.

One rationale for using the aEEG was to select infants most likely to benefit from therapeutic hypothermia, although there was no pre-randomization stratification according to aEEG severity. It is not evident why severely affected infants (based on aEEG criteria) were included as participants in this trial, when Gluckman et al. hypothesized that based on evidence from experimental models they were unlikely to benefit from this intervention. It may be one explanation for the overall negative result of this trial. More importantly, this study was not adequately powered to assess subgroup effects definitively.

The practical aspects of the selection of infants and the application of therapeutic hypothermia remain challenging. It is clear from experimental data that early initiation of therapeutic hypothermia is critical to its success, and yet the mean age at randomization in this trial was 4.8 hours. In addition, the participants are not easily recognizable, and the methods difficult to apply in everyday clinical practice. The majority of newborn infants around the world with hypoxic-ischemic encephalopathy are not born in tertiary care perinatal centers with aEEG monitors, or in centers with the expertise to apply or interpret the aEEGs. Further, many centers do not have access to the complicated type of cooling device used in this study, and probably never will. Results of current randomized controlled trials of therapeutic hypothermia of a more pragmatic design with straightforward clinical eligibility criteria, a simple method of cooling, and without the need for complicated equipment will be important.

As clinicians caring for infants with hypoxic-ischemic encephalopathy and its devastating sequelae, we want to embrace therapies that may help these infants and their families. Does this trial by Gluckman et al. provide the necessary evidence to initiate therapeutic hypothermia in all asphyxiated, encephalopathic newborn infants? Clearly, the answer is no. It does, however, give us a reason to continue and to participate in current trials, to await their results as well as their synthesis into systematic reviews, and, above all, to remain patiently optimistic.