Preoperative autonomic failure in neonates with critical congenital heart disease

Abstract

Background: Neonates with critical congenital heart disease (cCHD) undergo a complicated transition to ex-utero life. However, continuous monitoring of autonomic tone using heart rate variability is currently lacking.

Materials and methods: We retrieved continuous electrocardiograms from the time of admission or from 10 days prior to surgery for neonates with dextro-transposition of the great arteries (d-TGA) and hypoplastic left heart syndrome (HLHS). Beat-to-beat intervals (RRi) were calculated and divided into 10-min epochs. Spectral metrics, including low-frequency (LF) and high-frequency (HF) powers, as well as detrended fluctuation analysis metrics (α_S, α_L, RMS_S, and RMS_L), were computed for RRi within each epoch and averaged over 24 h. The relationship between RRi metrics and time to surgery was analyzed using linear mixed-effects models, adjusting for prenatal and postnatal factors.

Results: The study included 10 neonates with HLHS and 23 with d-TGA. RRi metrics were available for 110 days. In the unadjusted models, LF power (Estimate: -4.4×10^-3, P = 0.02), HF power (-4.1 x 10^-2, 5 x 10^-5), RMS_S(-3.7 x 10^-4, 8.7 x 10^-3), and RMS_L(-1.4× 10^-3, 0.02) were all negatively associated with time to surgery, with HF power showing the strongest association. After adjusting for covariates, HF power retained its significant negative association with time (-0.04, P = 0.03). The cCHD diagnosis did not significantly influence RRi metrics.

Conclusions: In neonates with cCHD, there is a progressive decline in autonomic function leading up to surgery. RRi metrics may serve as valuable indicators of deteriorating physiology in these patients.

Keywords: Autonomic dysfunction; Critical congenital heart disease; Heart rate variablity; Hypoplastic left heart syndrome; Preoperative monitoring.