Early Maturation of Heart Rate Variability in Very Preterm Infants Depends on Neonatal Factors and Is Associated With Neurodevelopmental Risk

ABSTRACT Aim This study investigated central autonomic network maturation deviations using a previously defined machine learning model set to estimate a functional maturation age (FMA) from heart rate variability (HRV) analysis. We investigated whether these deviations were associated with maternal, fetal, perinatal, or postnatal complications and with neurodevelopment at 2 years of age. Methods Differences (ΔHRV) between post‐menstrual age (PMA) and FMA were measured in a multicenter prospective cohort of 132 preterm infants born < 30 weeks gestational age (GA). The relationship between ΔHRV and clinical factors, occurrence of complications or alterations in neurodevelopment assessed by Ages and Stages Questionnaire (ASQ) at 2 years of age was studied. Results ΔHRV expressed in weeks delay at 34 weeks of PMA was greater when GA was lower (3 IQR 2.3–3.9) at 24–26 versus 1.3 (IQR 0.8–2.4) at 28–30 weeks GA ( p < 0.0001), in cases of bronchopulmonary dysplasia ( p < 0.0001) or patent ductus arteriosus ( p < 0.0001). ΔHRV was also associated with altered social skills at 2 years of age (OR 2.05, 95% CI 1.02; 4.14 for each week, p < 0.05) but not with ASQ total score ( p = 0.06). Conclusion ΔHRV, early and non‐invasively estimated, depends on GA, perinatal and postnatal complications. Its assessment could contribute to evaluation of neurodevelopmental risk.