Heart rate variability and electroencephalography in Infants with hypoxic ischaemic encephalopathy

Abstract

Hypoxic-ischaemic encephalopathy (HIE) remains a significant cause of neurological injury in the newborn. HIE is associated with altered autonomic function. Heart rate variability (HRV) is a direct measure of autonomic control and this thesis aims to investigate the use of HRV as a reliable, routinely recorded physiological marker to identify HIE severity, seizure activity, and predict long-term outcome. Measures of HRV were calculated on a cohort of full-term infants with HIE, recruited from 2003-2012, prior to and following the introduction of therapeutic hypothermia (TH) in the neonatal intensive care unit. Simultaneous, 1h periods of EEG and ECG activity were recorded from 6h-90h after birth. EEG activity was used to grade epochs as mild, moderate, or severe. Normalised-RR intervals were used to calculate seven HRV features from the ECG signal: Mean (mean NN), standard deviation (SDNN), triangular interpolation (TINN), power in high frequency (HF), low frequency (LF), very low frequency bands (VLF), and LF/HF ratio. There were significant correlations between HRV and EEG grade in normothermic (r = -0.45 to -0.33; n=44) and hypothermic groups (r = -0.39 to -0.22 ; n=74). In infants with moderate HIE, there were significant differences between normothermic and hypothermic groups (HF: p=0.016, LF/HF ratio: p=0.006, mean NN: p<0.001). HRV was significantly associated with outcome in both normothermic and hypothermic groups (normothermic at 24h, 48h and hypothermic 12h, 24h, and 48h after birth). Infants with neonatal seizures and severe HIE had an increase in HRV during electrographic seizures (TINN: p= 0.034, VLF: p=0.028, LF: p=0.006, HF: p=0.015). The measurement of autonomic function with HRV has potential as a diagnostic and prognostic tool in infants with HIE.