Autonomic Network Signatures of Interictal Cardiorespiratory Dysfunction in Patients with Temporal Lobe Epilepsy.

BACKGROUND AND OBJECTIVES:Temporal lobe epilepsy (TLE) co-occurs with interictal impairments in autonomic regulation and respiratory chemoreception, which have been related to higher risk of sudden unexpected death in epilepsy (SUDEP). Studies indicate that seizures lead to reorganization of central autonomic circuits extending beyond the temporal lobe. However, the relationship between brain network disruptions and interictal cardiorespiratory dysfunction remains largely unknown. The goal of this work was to test whether central autonomic network alterations underlie interictal cardiorespiratory dysfunction and higher SUDEP risk. METHODS:We included simultaneous fMRI and peripheral physiology from patients with drug-resistant unilateral TLE and healthy controls. We used mediation and correlation analyses to relate atypical functional connectivity strength and structural volume of central autonomic regions in TLE to alterations of high-frequency and low-frequency heart rate variability (HF-HRV/LF-HRV), breathing rate (BR), and respiratory rate variability (RRV). Multivariate models were then used to investigate associations with disease variables and risk factors of SUDEP. Finally, we analyzed a subcohort of patients with postsurgical data and examined whether autonomic connectivity changes following surgery impact cardiorespiratory function. RESULTS:Patients with TLE (n = 72, 49% female, 40.0 ± 12.8 years) exhibited widespread reductions in autonomic connectivity and volume compared with controls (n = 105, 53% female, 36.7 ± 12.8 years) (pFDR < 0.05). Reduced connectivity mediated higher BR, lower RRV, and lower LF-HRV in TLE relative to controls (|z| = 2.14 to 5.28, pFDR < 0.05). Left seizure laterality was independently related to reduced LF-HRV and RRV (Cohen's f = 0.42, 0.45; pFDR = 0.0055, 0.0046), and canonical correlation analysis linked atrophy in autonomic regions to generalized seizure occurrence and earlier age at onset (r = 0.61, pFWER = 0.018). In the postsurgical subcohort (n = 34), autonomic connectivity alterations after surgery were positively correlated with changes in LF-HRV and RRV (ρ = 0.52, 0.41; pFDR = 0.0076, 0.029). DISCUSSION:Our results suggest that reorganization of the central autonomic network may contribute to chronic autonomic and interoceptive deficits in TLE. Mapping these brain network disruptions may help guide novel strategies to restore cardiorespiratory function. Validation in SUDEP cases is needed to establish clinical utility as biomarkers of SUDEP.