2026 ECNP Congress

PS04-4188

Comparative efficacy and safety of non-invasive brain stimulation therapy for schizophrenia: a systematic review and network meta-analysis

Yaohui Wei — K. Huber, S. Siafis, J. Schneider‐Thoma, W. Strube, U. Vogelmann, S. Shi, S. Schulte-Frohlinde, F. Liczbinski, C. Li, J. M. Davis, R. C. Smith, F. Padberg, S. Leucht · Technical University of Munich, Department of Psychiatry and Psychotherapy, München, Germany

2026 ECNP Congress

Introduction

Guideline recommendations on the use of non-invasive brain stimulation (NIBS) for schizophrenia remain inconsistent, largely due to the lack of comprehensive comparative evidence across different stimulation modalities. While individual techniques such as repetitive transcranial magnetic stimulation (rTMS) and transcranial electrical stimulation (tES) have shown promise, their relative efficacy and safety compared with other approaches, including electroconvulsive therapy (ECT), magnetic seizure therapy (MST), and transcutaneous vagus nerve stimulation (tVNS), have not been systematically evaluated within a unified analytical framework. This gap limits the ability of clinicians and guideline developers to make evidence-based decisions regarding optimal NIBS strategies.

Aims

This study aimed to provide the first comprehensive comparison of the efficacy and safety of multiple NIBS modalities for schizophrenia using a network meta-analytic approach, and to examine their differential effects across symptom domains.

Methods

We conducted a systematic review and network meta-analysis of rater-blinded randomised controlled trials (RCTs) investigating NIBS in individuals with schizophrenia. Eligible interventions included rTMS, tES, ECT, MST, and tVNS. We searched the Cochrane Schizophrenia Group’s specialised register from inception to January 13, 2026. Study selection was conducted in two stages (title/abstract and full text) in duplicate by two independent authors. Data extraction and risk of bias assessment (Cochrane RoB 2) were performed independently by two authors using a Microsoft Access platform, with disagreements resolved by a third reviewer. Corresponding authors of all eligible studies were contacted to confirm methodology, provide missing data, or clarify trial details. The primary outcome was overall symptom severity, expressed as standardized mean differences (SMDs). Secondary outcomes included positive, negative, and depressive symptoms, auditory hallucinations, and adverse events. Random-effects network meta-analyses were performed to integrate direct and indirect evidence. Sensitivity analyses (e.g., excluding studies with extreme effect sizes) and subgroup analyses (e.g., stimulation parameters) were conducted to assess robustness. The study protocol was prospectively registered (PROSPERO: CRD42024589088).

Results

We included 93 trials with 3910 participants. Compared with sham stimulation, rTMS (SMD: -0·40, 95% CI -0·57 to -0·23) and tES (-0·39, -0·62 to -0·17) were associated with greater improvement in overall symptoms. ECT (-0·60, -1·29 to 0·09) and MST (-0·61, -1·77 to 0·55) hold a promise but their evidence was uncertain due to sparse data. These results were robust in sensitivity analyses. rTMS showed greater efficacy for positive symptoms (-0·33, -0·46 to -0·19) and auditory hallucinations (-0·33, -0·52 to -0·15), whereas tES appeared more beneficial for negative (-0·51, -0·76 to -0·25) and depressive symptoms (-0·65, -0·98 to -0·33). Tolerability was comparable across stimulation strategies.

Conclusions

NIBS are generally safe and show efficacy across symptom domains, with rTMS more efficacious for positive symptoms and hallucinations, and tES for negative and depressive symptoms. Treatment guidelines should be updated to reflect their differences and consider NIBS as a treatment option added to antipsychotics. Large-scale RCTs or individual participant data meta-analyses are needed to further confirm these findings.

References

No references