The neuroanatomical basis for depression, anxiety, rumination, and worry—a systematic review and meta-analysis

Rumination and worry are often linked to depression and anxiety respectively. Recently, rumination and worry have been proposed to have overlapping cognitive mechanisms and similar neuroanatomical irregularities. Within each of the two aforementioned syndromes, cognitive and behavioural commonalities have been reported between unipolar depression (UD) and bipolar depression (BD), and between general anxiety disorder (GAD) and social anxiety disorder (SAD) respectively. However, findings concerning the specific brain regions involved have been inconsistent. This study aims to comprehensively explore the neuroanatomical correlates of UD, BD, GAD, SAD, rumination, and worry.We will conduct six separate meta-analyses, focusing on 1) comparing grey matter volume (GMV) or grey matter density (GMD) differences between clinical groups (e.g., UD, BD, GAD, SAD) and healthy control (HC) groups, and 2) GMV and GMD correlates of rumination and worry. Three conjunction analyses will compare the meta-analytical findings between 1) UD and BD, 2) GAD and SAD, and 3) rumination and worry. Such conjunction analyses will inform us about the shared neuroanatomical correlates between the two types of depression, between the two types of anxiety, and between the two kinds of repetitive negative thinking. Finally, an additional conjunction analysis will be performed to compare the grey-matter-related reduction in internalising disorders (i.e., results from conjunction analysis between depressive and anxiety disorders) and repetitive negative thinking styles (i.e., results from conjunction analysis between rumination and worry). We will follow the PRISMA framework for systematic review and screening. Coordinates of statistically significant clusters of GMV and/or GMD will be extracted from whole-brain voxel-based morphometry (VBM) studies and synthesised in an activation likelihood estimate (ALE) meta-analysis. Effects of demographic, clinical diagnosis and medical history on the effect size will be tested in a meta-regression using effect-size seed-based d mapping (ES-SDM).