Seven RCTs and a quasi-RCT were included in the review (n=650 in total). The efficacy analysis included only 3 RCTs and the quasi-RCT, but the analysis of safety included data from all 8 studies.
The overall methodological quality of the studies was described as 'acceptable'. Six studies used concealed allocation, none of the studies blinded the intervention, but three used one or more blinded assessments, and four reported the completeness of follow-up. One study had a 32% loss to follow-up.
Overall, there was a statistically significant reduction in death or moderate to severe neurodevelopmental disability associated with the use of hypothermia in comparison with the control group (RR 0.76, 95% CI: 0.65, 0.88). In addition, compared with the control, hypothermia showed a significant reduction in severe neurodevelopmental disability (RR 0.65, 95% CI: 0.48, 0.87), severe cerebral palsy (RR 0.64, 95% CI: 0.42, 0.98), life support withdrawn (RR 0.59, 95% CI: 0.36, 0.96) and the number of infants with a Mental Developmental Index less than 70 (RR 0.69, 95% CI: 0.50, 0.96). Significant benefits in favour of hypothermia were also seen in the overall death rate (RR 0.74, 95% CI: 0.58, 0.94), although the incidence of arrhythmia (RR 6.29, 95% CI: 1.43, 27.75) and thrombocytopenia (RR 1.51, 95% CI: 1.09, 2.10) were higher in the hypothermia group. There was no evidence of significant statistical heterogeneity for any of the above RRs.
Subgroup analyses were only conducted for the severity of encephalopathy due to the small number of included participants. Significant reductions in the combined outcome of death or moderate to severe neurodevelopmental disability were reported for patients with moderate HIE, but in patients with severe HIE the differences were not statistically significant.
The funnel plots suggested a lack of studies at the extremes of the point estimates and a clustering around point estimates, suggesting some minor heterogeneity.