A systematic review and meta-analysis of brain volume abnormalities in disruptive behaviour disorders, antisocial personality disorder and psychopathy

Individuals with disruptive behaviour disorders in youth and antisocial personality disorder and psychopathy as adults share some clinical characteristics, but also diverge in important ways. Existing meta-analyses of structural imaging studies suggest abnormalities within these disorders; however, so far none has examined the role of variability. Here we performed a systematic review and meta-analysis to examine both variability (coefficient of variation ratio) and magnitude of brain volume differences between antisocial groups and healthy controls (quantified using Hedges’ g ). A comprehensive search was conducted of PubMed, EMBASE, Web of Science, Scopus and PsycINFO from inception to 31 January 2022 (pre-registered with PROSPERO, ID number CRD42021250980 , registered 25 June 2021). We included studies which included individuals with disruptive behaviour disorder (± callous–unemotional traits) or antisocial personality disorder (± psychopathy), defined using standardized classificatory tools (Diagnostic and Statistical Manual of Mental Disorders or International Classification of Diseases criteria for disruptive behaviour disorders and antisocial personality disorder, Psychopathy Checklist: Revised or Psychopathy Checklist: Screening Version for psychopathy) and a healthy control group, and which had sufficient data to extract mean and standard deviations, or t or P values, for both groups. We measured the relative variability of brain regions in antisocial individuals compared with controls, by using the log coefficient of variability ratio. Between-group differences in mean volumes were quantified using standardized mean difference. Risk of bias was assessed using modified version of the Newcastle–Ottawa Scale for case–control studies. Twenty-three studies met inclusion criteria. In antisocial individuals, there was significantly increased variability for total grey matter ( Z = −2.6581, P = 0.0079) and overall decreases in mean volume for total whole brain ( g = −0.41; 95% confidence interval (CI) −0.67 to −0.15, P = 0.0016), total grey matter ( g = −0.6; 95% CI −0.93 to −0.26, P = 0.004) and amygdala ( g = −0.89; 95% CI −1.55 to −0.22, P = 0.009), compared with healthy controls. This suggests a key role for structural variability in clinical divergence within these disorders. The key limitations were lack of studies for some brain regions of interest, including insula, and inconsistent clinical phenotyping. Further studies should seek to specify how this neurobiological variability maps to clinical variability and whether this holds potential value as a biomarker to guide prognosis or treatment selection.