Long-term risk of dementia, stroke and epilepsy after electroconvulsive therapy: a systematic review and meta-analysis of cohort studies

Introduction

Electroconvulsive therapy is an established and highly effective treatment for severe mood and psychotic disorders, particularly in cases resistant to pharmacological interventions. Despite its well-documented clinical efficacy, concerns persist regarding potential long-term neurological adverse outcomes, especially the risk of dementia, stroke and epilepsy. These concerns continue to influence public perception and clinical decision-making, and may contribute to stigma and hesitancy surrounding the use of electroconvulsive therapy. Existing evidence on long-term neurological safety remains fragmented, methodologically heterogeneous and occasionally inconsistent, warranting a comprehensive and systematic synthesis of available longitudinal data.

Aims

The aim of this study was to systematically evaluate the long-term risk of dementia, stroke and epilepsy following exposure to electroconvulsive therapy compared with psychiatric control populations not treated with electroconvulsive therapy, using data derived from large-scale cohort studies.

Methods

A systematic review and meta-analysis were conducted in accordance with PRISMA guidelines. PubMed, Scopus and Web of Science were searched from inception to 30 September 2025. Eligible studies were observational cohort studies with a mean follow-up of at least 12 months that compared the incidence of dementia, stroke or epilepsy in patients treated with electroconvulsive therapy versus psychiatric controls. Study quality was assessed using standard methodological criteria, including adjustment for relevant confounders. Hazard ratios (HR) were extracted as the primary effect size. Random-effects meta-analyses were performed due to anticipated clinical and methodological heterogeneity across studies. Statistical heterogeneity was quantified using the I² statistic. Sensitivity analyses and subgroup analyses, including analyses restricted to elderly populations, were conducted where data permitted.

Results

Eight cohort studies comprising a total of 656,775 participants met inclusion criteria. Seven studies were assessed as high methodological quality and were included in quantitative synthesis. Pooled analysis showed no significant increase in the long-term risk of dementia following electroconvulsive therapy (HR 0.99, 95% confidence interval (CI) 0.79–1.23, Z=0.12, p=0.90, I²=84%). After exclusion of one outlier study, results remained non-significant with reduced heterogeneity (HR 1.07, 95% CI 0.91–1.25, Z=0.80, p=0.42, I²=57%). Subgroup analysis focusing on elderly patients also demonstrated no significant association between electroconvulsive therapy and dementia risk (HR 1.02, 95% CI 0.53–1.98, Z=0.06, p=0.95, I²=81%). For stroke outcomes, pooled data indicated a statistically significant reduction in the risk of both first-time and recurrent events following electroconvulsive therapy (HR 0.80, 95% CI 0.73–0.88, Z=4.84, p<0.00001, I²=0%). Only two studies reported epilepsy outcomes, precluding meta-analysis for this endpoint.

Conclusions

Available evidence from cohort studies does not indicate an increased long-term risk of dementia or stroke following electroconvulsive therapy. While findings regarding stroke suggest a potential protective association, these results should be interpreted cautiously due to the observational nature of the data and possible residual confounding. Evidence regarding epilepsy risk remains insufficient and requires further investigation. Overall, the findings support the long-term neurological safety of electroconvulsive therapy, while highlighting the need for additional high-quality longitudinal research, particularly concerning seizure-related outcomes.

References

No references