Abstract
OBJECTIVE: To evaluate the association of caffeine with renal and other short-term clinical outcomes in neonates with moderate or severe hypoxic-ischaemic encephalopathy (HIE) undergoing therapeutic hypothermia (TH). STUDY DESIGN: This retrospective cohort study included neonates born at ≥36 weeks gestational age with moderate or severe HIE treated with TH at our centre between January 2013 and December 2022. In 2019, there was a practice change to administer a single intravenous 20 mg/kg caffeine citrate dose for moderate/severe HIE, forming two cohorts (non-caffeine vs caffeine). The primary outcome was acute kidney injury (AKI), defined by modified Kidney Disease: Improving Global Outcomes criteria. Secondary outcomes included daily serum creatinine, blood-urea nitrogen (BUN), urine output, HIE-pattern lesion on brain-MRI, early EEG background grade, seizures, duration of mechanical ventilation, length of stay, tube-feeding at discharge and in-hospital mortality. RESULTS: 85 neonates met the inclusion criteria; 38 received caffeine, and 47 did not. AKI incidence was lower in the caffeine group (29% vs 47%), although not statistically significant (p=0.09). BUN levels and serum creatinine levels were significantly lower in the caffeine group on days 3-5 (p<0.05). MRI abnormalities consistent with HIE were more frequent in the caffeine group (60% vs 36%, p=0.028), particularly in infants with moderate HIE (p=0.004). EEG background profiles differed, with caffeine-exposed neonates displaying fewer severe (34% vs 65%) and more mild-to-moderate patterns. No significant differences were noted in other secondary outcomes. CONCLUSION: Caffeine during TH lowered serum creatinine and BUN but did not significantly reduce AKI and was associated with higher rates of MRI lesions. Prospective, dose-controlled trials with pharmacokinetic monitoring and long-term neurodevelopmental follow-up are needed to clarify caffeine's renal benefits and neurological safety in neonatal HIE.