Temperature control and short-term outcomes of phase-changing material versus servo-controlled whole-body cooling in neonates with hypoxic-ischemic encephalopathy: a prospective comparative cohort study

Background: Therapeutic hypothermia improves outcomes in neonates with moderate-to-severe hypoxic-ischaemic encephalopathy (HIE), but access to servo-controlled systems may be limited. Phase-changing material (PCM) devices are a lower-complexity alternative requiring continuous core-temperature monitoring. Methods: This prospective, single-centre, observational comparative cohort study included 60 neonates with moderate-to-severe HIE treated for 72 hours with either servo-controlled whole-body cooling (n=30) or PCM cooling (n=30). Allocation was non-randomised and determined by device availability at admission. The primary outcomes were time to target temperature (33.5-34.5°C) and proportion of cooling time within the target range. Secondary outcomes included rewarming duration, respiratory and haemodynamic support, complications, treatment completion, seizures, neuroimaging and neurological status at discharge. Results: Servo-controlled cooling reached target temperature faster than PCM cooling (1.81±0.37 vs 2.53±0.77 hours; mean difference -0.72 hours, 95% CI -1.03 to -0.41; p<0.001) and maintained the target range for a greater proportion of cooling time (90.94±4.38% vs 79.52±6.86%; mean difference 11.42 percentage points, 95% CI 8.45 to 14.39; p<0.001). Rewarming duration did not differ significantly (7.10±0.99 vs 7.48±1.16 hours; p=0.175). No statistically significant between-group differences were detected for ventilation, inotropic support, thrombocytopenia, coagulopathy, electrolyte abnormalities, skin complications, completion of 72-hour cooling, seizures, neuroimaging or discharge neurological outcome. In exploratory multivariable analysis, HIE severity, timing of cooling and occurrence of a major adverse event were associated with unfavourable discharge outcome; cooling modality was not (adjusted OR 0.71, 95% CI 0.21-2.38; p=0.581). Conclusions: In this small, non-randomised cohort, servo-controlled cooling provided faster and more precise temperature control than PCM cooling. PCM cooling was feasible when continuous core-temperature monitoring and intensive neonatal care were available. The study did not establish equivalence of PCM and servo-controlled cooling for clinical or neurological outcomes.