Meta-analysis of brain structural changes after electroconvulsive therapy in depression

Abstract

BackgroundIncreases in the volume of the amygdala and hippocampus after electroconvulsive therapy (ECT) are among the most robust effects known to the brain-imaging field. Recent advances in the segmentation of substructures of these regions allow for novel insights on the relationship between brain structure and clinical outcomes of ECT. ObjectiveWe aimed to provide a comprehensive synthesis of evidence available on changes in brain structure after ECT, including recently published data on hippocampal subfields. MethodsA meta-analysis of published studies was carried out using random-effects models of standardized mean change of regional brain volumes measured with longitudinal magnetic resonance imaging of depressive patients before and after a series of ECT. ResultsData from 21 studies (543 depressed patients) were analysed, including 6 studies (118 patients) on hippocampal subfields. Meta-analyses could be carried out for seven brain regions for which data from at least three published studies was available. We observed increases in left and right hippocampi, amygdalae, cornua ammonis (CA) 1, CA 2/3, dentate gyri (DG) and subicula with standardized mean change scores ranging between 0.34 and 1.15. The model did not reveal significant volume increases in the caudate. Meta-regression indicated a negative relationship between the reported increases in the DG and relative symptom improvement (−0.27 (SE: 0.09) per 10%). ConclusionsECT is accompanied by significant volume increases in the bilateral hippocampus and amygdala that are not associated with treatment outcome. Among hippocampal subfields, the most robust volume increases after ECT were measured in the dentate gyrus. The indicated negative correlation of this effect with antidepressant efficacy warrants replication in data of individual patients.