Statistical modeling of heart rate variability to unravel the factors affecting autonomic regulation in preterm infants

\u3cp\u3eAnalyzing heart rate variability (HRV) in preterm infants can help track maturational changes and subclinical signatures of disease. We conducted an observational study to characterize the effect of demographic and cardiorespiratory factors on three features of HRV using a linear mixed-effects model. HRV-features were tailored to capture the unique physiology of preterm infants, including the contribution of transient pathophysiological heart rate (HR) decelerations. Infants were analyzed during stable periods in the incubator and subsequent sessions of Kangaroo care (KC) - an intervention that increases comfort. In total, 957 periods in the incubator and during KC were analyzed from 66 preterm infants. Our primary finding was that gestational age (GA) and postmenstrual age (PMA) have the largest influence on HRV while the HR and breathing rate have a considerably smaller effect. Birth weight and gender do not affect HRV. We identified that with increasing GA and PMA, overall HRV decreased and increased respectively. Potentially these differences can be attributed to distinct trajectories of intra- and extrauterine development. With increasing GA, the propensity towards severe HR decelerations decreases, thereby reducing overall variability, while with increasing PMA, the ratio of decelerations and accelerations approaches unity, increasing overall HRV.\u3c/p\u3