This study aimed to report long-term outcomes of 23 infants with perinatal hypoxic–ischaemic encephalopathy (HIE) who underwent therapeutic hypothermia (TH) at Chiang Mai University Hospital between February 2013 and February 2015. Informed consents were obtained at the follow-up clinic. The Bayley Scales of Infant and Toddler Development, Third Edition (BSID-III) was employed by developmental–behavioural paediatricians; visual impairment was evaluated by ophthalmologists and otoacoustic emissions (OAE)/auditory brainstem response (ABR) was used to identify any hearing impairment. Severe disability was defined by any of the following: Bayley-III (any domains)<70, Gross Motor Function Classification System≥3, severe visual impairment and profound hearing impairment. The results demonstrated that death, severe disabilities, and combined death and severe disabilities were 36.4%, 23.5% and 58.8%, respectively (table 1 and figure 1). If the analysis included lost to follow-up and non-participating patients, those outcomes would be 26.1%, 26.1% and 52.2%, which are closer to the figures published in other randomised controlled trial studies.1 Non-similar outcomes could be due to differences in TH indications, cooling devices and supportive care. During the study period, TH was indicated in infants fulfilling the following criteria: ≥35 weeks of gestational age (GA), >1800 g of birth weight (BW) and <12 hours of age; evidence of intrapartum asphyxia (2/3 of the following: Apgar score of ≤5 at 10 min, positive pressure ventilation (PPV) or continued resuscitation required at 10 min after birth and cord/blood gas within 1 hour after birth with pH<7.00 or base deficit (BD)>16); and diagnosis of moderate to severe HIE according to modified Sarnat criteria. Delayed TH up to 12 hours after birth was allowed because some patients were transferred from remote hospitals and TH was not available during transport. In addition, some patients developed moderate/severe HIE later on. Four patients with incomplete TH criteria died or had severe disabilities. As a result of delayed TH and prematurity, the efficacy of neuroprotective effects may be reduced.2 3 Semiautomatic cooling blankets with a rectal temperature probe were used for TH treatment. Core temperature was reduced to 33.5°C for 72 hours and subsequently increased by 0.3–0.5°C/hour until 36.5°C. Fluctuating temperatures >±0.5°C were found more frequently in a previous version of the cooling device than the one used currently. Respiratory and cardiovascular support was provided if indicated. Integrated electroencephalography was selectively performed; in cases involving a clinical seizure, treatment with intravenous phenobarbital and/or phenytoin was given as appropriate. The patients who died or had severe disabilities had a significantly higher level of blood lactate on the third day in comparison with those who survived without major disability (5.2 vs 2.4 mmol/L, p=0.03) This finding was also demonstrated in previous studies.
