2-Year Study of Vagus Nerve Stimulation for Higher-Grade Treatment-Resistant Depression: Clinical Outcomes and Policy Recommendation

This prospective, multi-center observational study aimed to evaluate the long-term real-world effectiveness, safety, and clinical utility of vagus nerve stimulation (VNS) in individuals diagnosed with higher-grade treatment-resistant depression (HG-TRD), defined by failure to respond to at least four adequate therapeutic interventions. The study was conducted in three Israeli psychiatric centers between 2020 and 2025. Vagus nerve stimulation (VNS) is a neuromodulation therapy involving the surgical implantation of a device that delivers electrical stimulation to the vagus nerve, targeting brain regions associated with mood regulation. While VNS is approved in several countries for treatment-resistant depression (TRD), real-world data outside of structured clinical trials for TRD remain limited. This study sought to fill that gap, particularly in the Israeli healthcare context. Study Population and Procedures Eligible participants were adults with chronic or recurrent major depressive episodes, who had previously failed at least four depression treatments (e.g., pharmacotherapy, Electro-Convulsive Treatment (ECT), esketamine, psychotherapy). Participants underwent VNS implantation and were followed prospectively for 24 months. Follow-up assessments were conducted at 6, 12, 18, and 24 months, evaluating depression severity, suicidality, and adverse effects using standardized clinical tool, the Montgomery-Åsberg Depression Rating Scale (MADRS). Device programming and stimulation adjustments were conducted biweekly in the early phases post-implantation and subsequently as clinically indicated. Stimulation parameters were titrated based on clinical response and tolerability. Data were collected in dedicated TRD specialty clinics and recorded by trained raters using structured clinical forms. All assessments were performed in-person during scheduled follow-up visits. The study was approved by the institutional Helsinki committees at all participating sites. Written informed consent was obtained from all participants. Statistical Plan and Handling of Data: * Sample Size: The final sample included 16 participants who completed 24-month follow-up and were eligible for efficacy analyses. All 19 implanted patients were included in safety analyses. * Primary analysis involved repeated measures ANOVA to examine changes in MADRS total scores over time. * Categorical outcomes (response, remission, partial response) were computed at each time-point and cumulatively (best outcome achieved at any time). * Suicidality was analyzed via MADRS item 10, both as a continuous and categorical outcome (≥50% or ≥30% reduction). * Hospitalization burden was analyzed as an exploratory outcome, comparing inpatient days during three time intervals (pre-implantation, year 1 post-implantation, year 2 post-implantation). * Missing data: Participants who permanently deactivated the device within the first 6 months were excluded from efficacy analyses but retained for safety reporting. Missing values in the MADRS and item 10 of this tool (suicidality) were imputed using the last observation carried forward method. Imputation was performed in the case of a missing value between two measurement points, but was not done after the last measurement point (for example, if a participant had values up to month 12, imputing was not done for month 18). Safety Assessment: Adverse events were monitored throughout the study and including voice alteration, throat discomfort, and coughing, as well as rarer complications such as dysphagia or transient vocal cord paresis. All adverse events were documented in accordance with clinical research guidelines.