Abstract
BACKGROUND: Autonomic function is modulated by the respiratory network. We therefore hypothesized that in persons with epilepsy (PWE), generalized convulsive seizures (GCS) would induce greater heart rate variability (HRV) and parasympathetic activation in patients with low interictal central CO(2) chemosensitivity (measured with the hypercapnic ventilatory response, HCVR). We further hypothesized that postictal HRV would be associated with severity of postictal hypercapnia and hypoxemia. METHODS: We performed a retrospective analysis of PWE admitted to an epilepsy monitoring unit for video-EEG study and experienced GCS. Time synchronized video-EEG, ECG, respiratory effort, and airflow were continuously measured along with transcutaneous CO(2) and O(2) saturation. Interictal HCVR was measured using a modified hyperoxic rebreathing technique. Postictal respiration was analyzed and included the magnitude and duration of CO(2) rise and O(2) desaturation. HRV measures were derived from 5-minute artifact-free ECG recordings from interictal and postictal periods. Relationships between HRV and respiratory variables were analyzed using Spearman's correlation and multivariate models. RESULTS: Twenty-six patients had both a GCS and an interictal HCVR. Mean age was 36.6 (±11.8) years and mean duration of epilepsy 16.2 (±12.0) years. Interictal HCVR slope varied from -0.13 to 5.2 (median 2.1) L/min/mm Hg and was not related to postictal RMSSD or the change in RMSSD (interictal wake - postictal) induced by GCS (p > 0.11). In contrast, duration of postictal hypercapnia and oxygen desaturation were both significantly correlated with overall postictal HRV (SDNN) and with multiple HRV measures of parasympathetic tone (RMSSD, HF power, Cardiac Vagal Index). In multivariate analyses, duration of postictal oxygen desaturation was positively associated with increased postictal RMSSD (mean ratio 1.09, 95 % CI 1.04-1.14, p < 0.01). CONCLUSIONS: Postictal ventilation and oxygenation are tightly coupled to multiple measures of heart rate variability, consistent with respiratory modulation of the autonomic nervous system. Patients with more severe postictal respiratory depression exhibit greater parasympathetic activity after GCS even while the frequent occurrence of postictal tachycardia suggests a concomitant increase in sympathetic activity. These results have implications for the interpretation of postictal bradycardia and respiratory dysfunction and their relationships to sudden unexpected death in epilepsy.