Network-guided Identification of the BNST as a Potential Neuromodulation Target in Temporal Lobe Epilepsy

Introduction: Temporal lobe epilepsy (TLE) is characterized by recurrent ictal discharges that progressively remodel critical arousal and respiratory circuitry, putting patients at risk for adverse respiratory events, autonomic instability, and sudden unexpected death in epilepsy (SUDEP). Recent evidence has implicated the bed nucleus of the stria terminalis (BNST) as a critical relay hub between brainstem respiratory nuclei and cortical arousal networks; however, it remains unclear how directional information flow within these circuits becomes compromised in TLE and whether such disruptions could inform neuromodulation targeting. We hypothesized that the BNST loses causal influence on important arousal and respiratory circuitry, indicating a loss in its ability to coordinate arousal responses to autonomic challenges.

Methods: Using simultaneous multislice imaging with multiband acceleration (TR=1.3s), we measured functional connectivity (FC) and effective connectivity using Granger causality (GC) in 37 TLE patients compared to 33 healthy controls. We applied network-based connectivity analyses using standard functional connectivity toolbox atlases and ascending arousal network atlases to characterize whole-brain connectivity patterns and effective connectivity with key arousal and respiratory control nuclei. This study was approved by the Vanderbilt University Medical Center IRB.

Results: TLE patients demonstrated bilateral reductions in BNST network integration capacity and causal influence on whole brain networks (p < 0.01). Patients exhibited profound loss of ascending BNST control over attention networks (p < 0.01), coupled with compensatory enhanced top-down cortical signaling. While standard connectivity measures showed preserved BNST-brainstem relationships, directional connectivity analyses revealed severe disruptions in BNST efferent control of key respiratory nuclei including ventral tegmental area, median raphe, and caudal solitary complex (p < 0.01), indicating compromised regulatory capacity despite maintained anatomical connections.

Conclusion: Our observations of BNST connectivity disruptions in TLE patients. Specifically, we documented the BNST's compromised ascending control over attention networks, profound directional signaling deficits to brainstem respiratory centers despite preserved standard connectivity, and reduced network integration capacity—all suggesting diminished ability to relay vital information between respiratory and arousal systems. Given its potential role in dysfunctional hypercapnia-induced arousal, the BNST represents a particularly promising target for neuromodulatory intervention in patients at elevated risk for SUDEP. If the BNST's function as a critical arousal integration center is compromised in humans with TLE, then targeted intervention could potentially restore this vital protective mechanism, with BNST deep brain stimulation already demonstrating safety and efficacy in other neuropsychiatric conditions, providing a translational pathway for clinical implementation in appropriately selected epilepsy patients.