Selective vagus nerve stimulation to target cardiac efferent activity in humans for treatment of heart disease

Introduction and Objective: Spatially selective vagus nerve stimulation (sVNS) offers a promising solution to mitigate the off-target effects associated with traditional VNS in which the whole nerve is stimulated. This could serve as a precise method to address chronic heart failure by specifically targeting efferent cardiac fibres and thereby enhance therapeutic outcomes by avoiding off-target effects and eliminating the need for titration. However, the spatial organisation of afferent and efferent cardiac-related fibres within the vagus nerve remains unknown. To address this, we have determined the functional organisation of the cardiac afferent and efferent projections within the cervical vagus nerve of pigs and have initiated translational work in humans. Methods: A multi-electrode sVNS cuff (28-electrode array: 2 rings x 14 electrodes) was placed at the mid-cervical level of the exposed left vagus nerve in pigs and prior to surgical implantation of an FDA approved VNS device in consenting epilepsy patients (NCT05664854, REC 22/LO/0463). Current was applied sequentially to each pair of electrodes in the same radial (o’clock) position on the 2 rings while recording evoked cardiac, pulmonary and laryngeal responses. This was repeated at different parameters optimised for the respective functional responses and repeated after vagotomy in pigs. Results: In pigs, there were 2.0±1.7 (mean±SD) most effective electrode pairs eliciting heart rate (HR) decrease of -7.8±3.4% associated with cardiac efferent activation. A HR increase of 10.2±5.3% (n=6) was induced and sustained post-vagotomy, associated with cardiac afferent activation. There was significant spatial separation of all four regions but specifically, cardiac afferent and efferent CoMs (200±137°, p<0.05, mean±1SD) at the mid-cervical level. In preliminary studies in one human, there was 5% tachycardia with full VNS. During sVNS, this was sustained with 3-4% tachycardia for 11/14 electrode pairs (cardiac afferent). On 3 electrodes, a 4, 5, and 7.5% bradycardia was achieved, respectively. This displayed selective activation of cardiac efferent projections. Conclusions: Significant separation between afferent and efferent cardiac-related regions occurred in all pigs, localised to roughly the opposite sides of the nerve, in close proximity to the pulmonary and laryngeal regions, respectively. For the first time, we have shown this within the vagus nerve in one human; work is in progress on further human studies. This would allow the ability to selectively elicit therapeutic-related efferent effects without affecting afferent-related reflexes. This paves the way for more targeted neuromodulation to enhance precision and efficacy of VNS therapy in treating heart failure. Medical Research Council Grant MR/Z504555/1 National Institutes of Health grant 3OT2OD026545 This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.