Supplementary Material for: Fifteen years of neonatal Therapeutic Hypothermia: clinical trends show unchanged post-rewarming outcomes despite reduction in HIE severity

INTRODUCTION Hypoxic-ischemic encephalopathy (HIE) affects 1-2 per 1000 births and is associated with mortality and long-term neurodevelopmental challenges. At present, therapeutic hypothermia (TH) is the only neuroprotective intervention for these infants. This study examines whether HIE severity, clinical management during TH and post-rewarming outcomes have changed since its introduction 15 years ago.

METHODS Neonatal characteristics, HIE severity, management during TH and post-rewarming MRI of all infants with HIE undergoing TH between 2008-2023 were compared across three five-year epochs. Linear regression was used to estimate annual changes over time.

RESULTS In total, 252 infants underwent TH. Median GA (39.5 weeks), birth weight (3376 grams) and time to start TH (4.25 hours) remained stable over time. Apgar score at 5 minutes (p=0.031) and lowest pH <1h postpartum (p=0.020) increased over time. Thompson score at 1-3h decreased across epochs (p=0.046). There was an increase in percentage with normal-mild aEEG background patterns on admission (p=0.041) and a decrease in aEEG-confirmed seizures (p<0.001) and anti-seizure medication (p<0.001). Inotropic support decreased (p=0.007), and use of invasive mechanical ventilation decreased over the last five years. Mortality (28.6%) and post-rewarming composite adverse outcome (i.e. neonatal mortality and/or adverse MRI score) (37.9%) remained unchanged. Number of infants seen at 2-year follow-up increased (p<0.001).

CONCLUSION Over the last 15 years we treated more infants with milder HIE, as indicated by lower Thompson and milder aEEG scores, and the need for invasive cardiorespiratory support declined. However, there were no improvements in composite adverse outcome (mortality and/or adverse MRI score).

引言
缺氧缺血性脑病（Hypoxic-ischemic encephalopathy, HIE）的发病率为每1000名活产婴儿1至2例，常伴随死亡风险及长期神经发育障碍。目前，治疗性低温（therapeutic hypothermia, TH）是此类新生儿唯一获批的神经保护干预措施。本研究旨在探讨自15年前治疗性低温临床应用以来，HIE的严重程度、治疗期间的临床管理方案以及复温后的临床结局是否发生了变化。

方法
本研究回顾性分析2008年至2023年间所有接受治疗性低温的HIE患儿的临床数据，包括新生儿基线特征、HIE严重程度、治疗性低温实施期间的临床管理情况，以及复温后磁共振成像（Magnetic Resonance Imaging, MRI）结果，并按三个5年研究时间段进行分组比较。采用线性回归模型估算各指标随时间推移的年度变化趋势。

结果
本研究共纳入252名接受治疗性低温的HIE患儿。随研究时间推进，患儿的胎龄（中位数39.5周）、出生体重（中位数3376g）以及启动治疗性低温的时间间隔（中位数4.25小时）均无显著变化。5分钟Apgar评分（p=0.031）及产后1小时内的最低动脉pH值（p=0.020）呈显著上升趋势。患儿出生后1~3小时的Thompson评分随研究时间段推移显著下降（p=0.046）。入院时振幅整合脑电图（amplitude-integrated electroencephalogram, aEEG）背景模式为正常-轻度异常的患儿比例显著升高（p=0.041），而经aEEG确认的癫痫发作患儿占比（p<0.001）及抗癫痫药物使用比例（p<0.001）均显著降低。正性肌力药物支持的使用率显著下降（p=0.007），近5年来有创机械通气的使用也呈下降趋势。患儿总体死亡率（28.6%）及复温后复合不良结局（即新生儿死亡和/或MRI评分异常）的发生率（37.9%）均未发生明显改变。完成2年随访的患儿数量较前显著增加（p<0.001）。

结论
在过去15年间，本中心收治的HIE患儿病情整体更趋轻微，具体表现为更低的Thompson评分以及更轻度的aEEG背景模式异常，且有创心肺支持的临床需求显著下降。然而，复合不良结局（包括新生儿死亡和/或MRI评分异常）的发生率并未得到显著改善。