A comprehensive systematic review and meta-analysis examining structural neuroimaging data across individuals with psychotic disorders and bipolar disorder indicates that childhood adversity does not show a statistically significant volumetric association with the hippocampus or the amygdala. Instead, synthesized findings across the literature point toward gray matter volume reductions situated more consistently in prefrontal cortical regions. These findings challenge the classical neurobiological model of stress, which has long posited that early-life adversity primarily targets subcortical limbic structures due to elevated glucocorticoid vulnerability. Conducted across dozens of clinical cohorts, the research highlights that structural traces of early-life maltreatment and trauma may be more robustly detected across higher-order cortical regions involved in emotional and cognitive regulation.
The investigation was pre-registered with the International Prospective Register of Systematic Reviews under protocol registration CRD42022351133 and conducted in strict compliance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The researchers performed systematic literature searches across the MEDLINE, EMBASE, and PsycINFO databases via the Ovid platform from inception through June 2024. In addition, manual searches of citation lists from eligible articles and previous reviews on related topics were undertaken to identify any missing candidate investigations. Applying the PICO framework, the team selected clinical observational studies, cohort studies, and case-control studies published in peer-reviewed English-language journals that evaluated adults aged 18 to 80 years diagnosed with psychotic disorders or bipolar disorder based on established diagnostic criteria such as the Diagnostic and Statistical Manual of Mental Disorders or the International Classification of Diseases.
Eligible investigations were required to compare gray matter volume between individuals with and without childhood adversity, or assess associations within patient cohorts using non-trauma comparisons, while quantitatively reporting brain volumetric data derived from structural magnetic resonance imaging. Across the identified literature, 29 studies met all qualitative synthesis inclusion criteria, capturing a combined clinical cohort of 3056 participants with a mean age of 36.6 years and an overall female representation of 47 percent. The individual sample sizes of the included studies spanned from 20 to 788 participants, and nearly half of the papers had been published between 2019 and 2024. Within the qualitative dataset, 28 percent of the studies examined individuals with schizophrenia across 8 cohorts, 17 percent evaluated first-episode psychosis across 5 cohorts, 34 percent focused on bipolar disorder across 10 cohorts, and 21 percent assessed transdiagnostic cohorts across 6 samples.
Study quality was formally appraised by two independent reviewers using a revised version of the Newcastle-Ottawa Scale for non-randomized cohort studies, with discrepancies resolved through consensus and input from a third reviewer. Quality scores across the 29 studies ranged from 4 to 6 on the scale’s star system, reflecting fair to good overall methodological quality, with 45 percent of studies achieving a rating of 6 and 52 percent achieving a score of 5. In terms of trauma measurement instruments, 65 percent of the analyzed investigations relied on the Childhood Trauma Questionnaire, most frequently utilizing the composite score to gauge overall exposure, while a smaller portion of studies implemented binary categorization criteria for trauma exposure. Predictors assessed across studies included general composite adversity alongside specific operationalized subtypes comprising physical abuse, sexual abuse, emotional abuse, physical neglect, and emotional neglect.
In the systematic qualitative synthesis of the 29 studies, 79 percent reported at least one statistically significant association between childhood adversity and regional gray matter alterations, whereas 21 percent observed only non-significant null outcomes. Among the studies reporting significant effects, 38 independent statistical analyses were identified, with 71 percent demonstrating negative associations where greater adversity corresponded to smaller brain volume, while 29 percent demonstrated positive associations. However, anatomical localization revealed stark divergences across brain compartments. For the hippocampus, 21 studies conducted evaluations, but 67 percent found no significant relationship with adversity. Of the 7 studies identifying significant hippocampal associations across 11 independent analyses, 7 supported volume reductions and 4 reported volume increases.
A parallel pattern of inconsistency emerged across investigations evaluating the amygdala. Among 14 studies that examined amygdala volume in relation to childhood trauma exposure, 50 percent documented null findings with no statistically significant volumetric associations, while the remaining 50 percent reported significant outcomes. Across the 7 independent analyses derived from those significant studies, 57 percent identified negative associations where trauma was linked to decreased amygdala volume, whereas 43 percent identified positive associations linking adversity to increased volume. The authors observed that this conflicting directionality precludes definitive conclusions regarding limbic atrophy and contradicts assumptions that stress-sensitive subcortical nuclei reliably undergo volumetric reduction following early trauma in these psychiatric populations.
In contrast to the inconsistent limbic results, studies evaluating additional gray matter regions, particularly the prefrontal cortex, demonstrated far more coherent patterns. Across 17 studies evaluating regions such as the prefrontal cortex, anterior cingulate cortex, caudate, and nucleus accumbens using region-of-interest or whole-brain approaches, 71 percent observed significant associations and only 29 percent reported null findings. Among the 20 independent analyses conducted within these investigations, 80 percent revealed negative associations indicating volumetric reductions linked to childhood adversity, while 20 percent reported positive associations. Five studies utilizing whole-brain methods consistently identified significant volume reductions across prefrontal areas, including the dorsolateral prefrontal cortex, ventromedial prefrontal cortex, anterior cingulate cortex, and frontal gyrus, where effects survived stringent statistical corrections for multiple comparisons.
To provide quantitative verification, the authors conducted formal random-effects meta-analyses using Comprehensive Meta-Analysis software, calculating effect sizes as Pearson’s r converted through Fisher’s r-to-z transformations. Quantitative synthesis was restricted to patient samples reporting hippocampal and amygdala volumes with at least four eligible samples per analysis. Primary hemisphere-specific meta-analyses revealed no significant association between childhood adversity and hippocampal volume in either the right hemisphere across five studies (r = 0.010, p = 0.893) or the left hemisphere (r = -0.036, p = 0.669). Secondary pooling of combined bilateral hippocampal volume across studies likewise showed no statistically significant effect (r = -0.118, p = 0.144). Similarly, lateralized meta-analyses for the amygdala identified no significant correlation with childhood adversity in the right hemisphere (r = -0.017, p = 0.878) or the left hemisphere (r = -0.022, p = 0.883), and combined amygdala volume remained non-significant (r = -0.042, p = 0.756).
Heterogeneity across the quantitative analyses varied substantially, ranging from low in the right hippocampus with an I-squared value of 41 percent to substantial in the left amygdala with an I-squared of 79 percent. Funnel plot evaluations and Egger’s regression tests indicated no significant publication bias for hippocampal analyses or the left amygdala, but Egger’s test reached statistical significance for the right amygdala with a p-value of 0.016, suggesting that small-study effects or publication bias could influence right amygdala literature. Meta-regression models were performed to examine whether age, sex, illness stage, segmentation method, or Newcastle-Ottawa Scale quality scores acted as moderating factors. Methodological quality emerged as a significant moderator for the right hippocampus (beta = 0.242, p = 0.030), and sex served as a significant moderator in the total hippocampal volume model (beta = 0.020, p = 0.007). In the amygdala analyses, illness stage, segmentation approach, and study quality all emerged as significant moderators across both hemispheres and combined volume, with p-values spanning 0.007 to 0.014.
The study authors emphasize several biological and methodological mechanisms that likely explain these findings. The prefrontal cortex undergoes extended maturational processes, including prolonged synaptic pruning and myelination extending well into early adulthood, making its developmental window exceptionally sensitive to chronic stress-induced disruptions. Furthermore, clinical variables such as illness chronicity and pharmacotherapy introduce complex biological interactions; for example, lithium treatment is documented to promote neuroplasticity and increase gray matter volume in the hippocampus and amygdala, potentially counteracting or masking structural consequences of early trauma. The authors conclude that future neuroimaging investigations must prioritize longitudinal designs, nuanced characterization of trauma timing and subtypes, rigorous sex stratification, and data-driven multivariate network methods to accurately map the widespread structural consequences of early-life adversity.
Subject of Research: Structural neuroimaging meta-analysis of childhood adversity and brain volumes in psychotic and bipolar disorders
Article Title: Revisiting the link between childhood adversity and stress-sensitive brain regions in psychosis and bipolar disorder: a systematic review and meta-analysis
Article References: Petrova, T., Tennfjord, A. Ø., Cavero, D., Holohan, A., Kizilkaya, M., Ebrahimian-Roodbari, A., Lepreux, I., Reimer, M., Sideli, L., Gadelrab, R., Trotta, G., Rodriguez, V., Andreassen, O. A., Klauser, P., Alameda, L., & Aas, M. (2026). Revisiting the link between childhood adversity and stress-sensitive brain regions in psychosis and bipolar disorder: a systematic review and meta-analysis. Translational Psychiatry, 16(1), Article 539. https://doi.org/10.1038/s41398-026-04457-w
Image Credits: AI Generated
DOI: 10.1038/s41398-026-04457-w
Keywords: childhood adversity, psychosis, bipolar disorder, structural MRI, gray matter volume, hippocampus, amygdala, prefrontal cortex, systematic review, meta-analysis, transdiagnostic, neuroimaging, Childhood Trauma Questionnaire, cortical thinning, clinical neurobiology, early life stress
News Source: Glenn Wilkins. (November 16, 2026). Study Finds Childhood Adversity Affects Prefrontal Cortex Rather Than Limbic Structures. Scienmag.



