Systematic review for VNS vs. pharmaceutical modulations for multifaceted neurological disorder management through cross-case, network meta-analysis

Abstract

Background

As an adjunct or alternative to conventional pharmacotherapy, vagus nerve stimulation (VNS) which is FDA-approved has arisen as a novel means for various neurological disorders.

Method

We searched multiple databases (through 2024) for randomised trials and observational studies of VNS (invasive and transcutaneous) and pharmacological treatments (e.g. cholinergic agents, antiepileptics, antidepressants) across several neurological disorders. Prior to comparing between VNS and pharmacological treatments, subgroup analyses of VNS studies were performed for disorder type, patient demographics, VNS stimulation parameters, and treatment duration to illustrate whether VNS itself can be effective to a satisfactory extent to be compared against the conventional method. Efficacy and adverse effects were evaluated, based on the proportion of patients achieving more than 50 % symptom reduction or equivalent clinical improvement, or all-cause mortality where applicable. Evaluation between VNS and pharmacological treatments was performed through network meta-analysis, followed by assessment of heterogeneity (I²) and meta-regression. Risk of bias was evaluated with Cochrane criteria, and all studies (including those with high risk of bias) were included in the primary analysis (with sensitivity analyses excluding high-bias studies).

Results

We included 56 VNS-related studies (n = 5773 participants) and 29 pharmacological drug-based studies (n = 14827 participants) from spanning epilepsy, depression, migraine/headache, Alzheimer's disease, inflammatory disorders, and heart failure. A network meta-analysis directly comparing VNS to pharmacological drugs yielded an overall advantage for VNS (summary SMD = 0.27 favouring VNS, 95 % CI 0.19–0.35). However, the high heterogeneity and risk of bias have been assessed, indicating potential issues with the VNS studies.

Conclusion

Overall, VNS was shown to be a viable therapeutic modality across diverse neurological disorders, superior to standard pharmacological treatments with a distinct adverse effect profile. It appears particularly beneficial in conditions where conventional drugs have limited success (e.g. refractory epilepsy, depression), although patient-specific factors influence outcomes. Further high-quality trials are anticipated to optimise stimulation parameters, confirm long-term benefits, and manage patient selection for VNS.