QIMR Berghofer Medical Research Institute, Brisbane, QLD, 4006, Australia.
School of Medicine and Dentistry, Griffith University, Gold Coast, QLD, 4215, Australia.
Pediatr Res. 2023 Jul;94(1):206-212. doi: 10.1038/s41390-022-02376-2. Epub 2022 Nov 14.
Preterm birth predisposes infants to adverse outcomes that, without early intervention, impacts their long-term health. To assist bedside monitoring, we developed a tool to track the autonomic maturation of the preterm by assessing heart rate variability (HRV) changes during intensive care.
Electrocardiogram (ECG) recordings were longitudinally recorded in 67 infants (26-38 weeks postmenstrual age (PMA)). Supervised machine learning was used to generate a functional autonomic age (FAA), by combining 50 computed HRV features from successive 5-minute ECG epochs (median of 23 epochs per infant). Performance of the FAA was assessed by correlation to PMA, clinical outcomes and the infant's functional brain age (FBA), an index of maturation derived from the electroencephalogram.
The FAA was strongly correlated to PMA (r = 0.86, 95% CI: 0.83-0.93) with a mean absolute error (MAE) of 1.66 weeks and also accurately estimated FBA (MAE = 1.58 weeks, n = 54 infants). The relationship between PMA and FAA was not confounded by neurodevelopmental outcome (p = 0.18, n = 45), sex (p = 0.88, n = 56), patent ductus arteriosus (p = 0.08, n = 56), IVH (p = 0.63, n = 56) or body weight at birth (p = 0.95, n = 56).
The FAA, an index derived from the ubiquitous ECG signal, offers direct avenues towards estimating autonomic maturation at the bedside during intensive care monitoring.
The development of a tool to track functional autonomic age in preterm infants based on heart rate variability features in the electrocardiogram provides a rapid and specialized view of autonomic maturation at the bedside. Functional autonomic age is linked closely to postmenstrual age and central nervous system function response, as determined by its relationship to functional brain age from the electroencephalogram. Tracking functional autonomic age during neonatal intensive care unit monitoring offers a unique insight into cardiovascular health in infants born extremely preterm and their maturational trajectories to term age.
早产使婴儿易发生不良后果,如果没有早期干预,会影响他们的长期健康。为了协助床边监测,我们开发了一种工具,通过评估重症监护期间心率变异性(HRV)的变化来跟踪早产儿的自主神经成熟。
对 67 名(胎龄 26-38 周)婴儿进行了心电图(ECG)的纵向记录。通过结合连续 5 分钟 ECG 时段的 50 个计算 HRV 特征(每个婴儿的中位数为 23 个时段),使用监督机器学习生成功能自主年龄(FAA)。通过与胎龄(PMA)、临床结局和婴儿的功能大脑年龄(FBA)(从脑电图中得出的成熟度指标)的相关性来评估 FAA 的性能。
FAA 与 PMA 高度相关(r=0.86,95%CI:0.83-0.93),平均绝对误差(MAE)为 1.66 周,并且还能准确估计 FBA(MAE=1.58 周,n=54 名婴儿)。PMA 和 FAA 之间的关系不受神经发育结局(p=0.18,n=45)、性别(p=0.88,n=56)、动脉导管未闭(p=0.08,n=56)、IVH(p=0.63,n=56)或出生体重(p=0.95,n=56)的影响。
FAA 是一种源自无处不在的 ECG 信号的指标,它为在重症监护监测期间在床边估计自主神经成熟提供了直接途径。
基于心电图中的心率变异性特征,开发了一种跟踪早产儿功能自主年龄的工具,为床边自主神经成熟提供了快速和专业的观察。功能自主年龄与胎龄和中枢神经系统功能反应密切相关,这与其来自脑电图的功能大脑年龄的关系决定。在新生儿重症监护病房监测期间跟踪功能自主年龄,可以深入了解极早产儿的心血管健康及其成熟轨迹到足月年龄。