Kalyanam Shivkumar

Objective. Vagus nerve stimulation (VNS) is an established clinical therapy for drug-resistant epilepsy and other inflammatory conditions. However, off-target stimulation can produce unwanted side effects that limit therapeutic stimulation and hinder the development of new neuromodulation therapies. Selective VNS (sVNS) offers a strategy to reduce off-target organ activation; however, this approach is not available in humans, and no implantable or portable devices exist to trial sVNS in the clinical setup. This work aimed to design, manufacture, and validate an implantable wireless, battery-free stimulator with a selectively addressable output stage for targeted current delivery to specific regions of the human vagus nerve. Approach. We developed a near-field communication-controlled, wirelessly powered, battery-free, temporary implantable multichannel stimulation device, compatible with a 15-channel sVNS cuff electrode (14 selective electrode pairs and one circumferential whole-nerve channel). The device was encapsulated for short-term implantation and evaluated through benchtop characterisation, accelerated ageing, and validation in an acute porcine and a pilot human study. Main result. The sVNS device was evaluated in a porcine ( n = 4) trial and a first-in-human pilot study ( n = 1). Selective bradycardia of 23.28 ± 12.91% was observed in pigs and 7.5% in the human participant. In the human, a clear separation of bradycardic and tachycardic effects was observed, with additional selectivity in laryngeal activity. Cardiac and laryngeal responses were separated by 231° around the circumference of the nerve. Significance . This work demonstrates the feasibility of wireless battery-free sVNS for cardiac applications using a temporary implantable device. Geometrically selective stimulation has the potential to improve therapeutic efficacy while reducing stimulation-related side effects, and may facilitate future therapies for heart failure and other autonomic disorders.

BACKGROUND: The inferior pyramidal space (IPS), the epicardial adipose tissue extending from the inferior crux of the heart, contains the atrioventricular node (AVN) at its apex. However, innervation of the human AVN via the IPS remains unknown. METHODS: Seven whole-mount samples of the human IPS involving the AVN were obtained from hearts rejected for transplantation (85.7% male, 49.4±12.8 years old). The IPS was further divided into the anterior (behind the floor of the triangle of Koch) and posterior (within the Eustachian ridge) fat pads relative to the coronary sinus. A whole-mount tissue-clearing method with immunohistochemical staining was applied to each fat pad to count and characterize the nerve fascicles ≥10 µm and ganglia ≥6000 µm ² . RESULTS: Within the anterior fat pad, sympathetic nerve fascicles ascended towards the AVN along the AVN artery, whereas nonsympathetic nerve fascicles approached the AVN from the superior and posterior aspects without a consistent relationship to the AVN artery. Within the IPS, sympathetic nerve fascicles were thicker than nonsympathetic nerve fascicles (51.7±32.8 versus 22.1±10.6 µm; P P <0.01). The compact AVN regions were devoid of thick nerve fascicles and ganglionated plexuses but filled with a fine meshwork of both sympathetic and nonsympathetic nerve fibers. CONCLUSIONS: The human IPS exhibits a distinct pattern of autonomic innervation and distribution of ganglia. These neural structures are likely to contribute to the physiological control of the human AVN and could be a novel target for neuromodulation to restore normal AVN function.

Ambulatory electrocardiograms (ECG) provides continuous monitoring of the heart's electrical activity. However, many existing machine learning and artificial intelligence models for analyzing ambulatory ECG traces are often unimodal and do not incorporate patient clinical context. In this study, we propose a multimodal framework integrating ambulatory ECG-derived representations with clinical text embeddings to predict two cardiac outcomes: sudden cardiac death and pump failure death. Ambulatory ECG traces are preprocessed, segmented, and encoded via a multiple instance learning and temporal convolutional neural network framework. In parallel, patient clinical features are parsed into structured prompts, which are passed through a large language model to generate clinical reasoning; this reasoning passes through a biomedical language encoder to generate a text embedding. With the ECG and text embeddings, we systematically evaluate multiple fusion strategies, including concatenation- and gating-based approaches, to integrate these two data modalities. Our results demonstrate that multimodal models consistently outperform unimodal baselines, with adaptive fusion mechanisms providing the greatest improvements in predictive performance. Decision curve analysis highlights the potential clinical utility of the proposed framework for risk stratification. Finally, we visualize model attention across modalities, including ECG attention patterns, segment-level saliency, heart rate variability features, and clinical reasoning, to contextualize patient-specific predictions.

Vagus nerve stimulation (VNS) is an established clinical therapy for drug-resistant epilepsy and shows potential for treating other conditions, including depression, rheumatoid arthritis, diabetes, and heart failure. However, stimulation often produces unwanted side effects such as hoarseness, coughing, and paraesthesia. In some cases, these effects limit the delivery of therapeutic stimulation levels and hinder the development of new neuromodulation therapies. Selective VNS (sVNS) offers a strategy to reduce off-target organ activation. Methods: This work presents an NFC-controlled, wirelessly powered, battery-free, temporary implantable multichannel stimulation device, made with off-the-shelf components, enabling selective stimulation of specific regions of the vagus nerve (VN). The encapsulated device is suitable for short-term implantation in animals. Main result: The sVNS device was investigated in a porcine (n = 4) trial and an n = 1 pilot human experiment. Selective bradycardia of 23.28 ± 12.91% was achieved in pigs and 7.5% in the human participant. In the human, a clear separation of cardiac efferent and afferent stimulation was observed, with additional selectivity in laryngeal activity. Physiological effects of laryngeal and cardiac fibre separation were measured to be 231°. Significance: Geometrically selective stimulation of VN fascicles has the potential to improve clinical outcomes, enhance therapeutic efficacy, and reduce stimulation-related side effects. This strategy may enhance neuromodulation approaches for the treatment of heart failure using VNS.

BACKGROUND: The mechanisms behind endocardial entanglement and resultant delamination during left bundle branch area pacing, as well as potential solutions, have not been fully demonstrated in the human heart. METHODS: Active fixation of a lumenless screw-in lead into the right ventricular septum was attempted with or without delivery system support using donor human hearts rejected for transplant. Delivery sheath support was ensured through firm catheter support maintained by a fixed clip. Endocardial entanglement, endocardial delamination after unscrewing, and lead penetration were evaluated using both macroscopic and microscopic analyses. RESULTS: Each of 10 lead deployments with or without support was attempted using 4 donor human hearts. There was a higher incidence of endocardial entanglement and resultant helix-only penetration in the nonsupport group than the support group (70% versus 0%). Unscrewing after the endocardial entanglement consistently revealed extensively delaminated endocardium entwined around the helix. The median endocardial delamination area was larger in the entanglement group than in the nonentanglement group (25.0 mm ² versus 0.9 mm ² , P <0.001). In the regions with endocardial entanglement, histological analysis confirmed extensive endocardial delamination with minimal myocardial injury. On the other hand, localized endocardial delamination and deep intramural penetration were observed in the regions without entanglement. Screw-in procedures reattempted into the regions with extensive endocardial delamination following entanglement achieved deep penetration even without delivery system support. CONCLUSIONS: Endocardial entanglement and delamination were confirmed in donor human hearts. Adequate delivery system support, particularly its coaxiality and counterforce effect, is likely to prevent endocardial entanglement and facilitate effective penetration of a lumenless screw-in lead into the human ventricular septum.