Corpus callosum morphology in bipolar disorder: Associations with illness duration

Bipolar disorder (BD) is associated with abnormalities in large-scale white matter connectivity, with the corpus callosum (CC) being one of the most consistently implicated structures. However, the relationship between proportional CC measures and clinical variables in BD has not yet been clearly established. This study investigated normalized CC measures in euthymic patients with BD and examined their associations with clinical characteristics.

A total of 108 euthymic patients with BD and 110 age- and sex-matched healthy controls were included. CC morphology was assessed on midsagittal magnetic resonance imaging using the corpus callosum index (CCI) and the intracranial volume (ICV)-adjusted CCI. Group comparisons were performed using appropriate parametric or non-parametric tests, supplemented by age- and sex-adjusted regression models. In the BD group, associations between clinical variables and imaging measures were examined using Spearman correlation and multiple regression analyses. Additional age-adjusted and exploratory stepwise regression models were conducted, and a full-sample interaction model was used to test whether the effects of age and sex differed between patients with BD and healthy controls.

Patients with BD showed lower CCI and CCI/ICV values than healthy controls, and these differences remained significant after adjustment for age and sex. In the BD group, illness duration was negatively correlated with CCI. In the illness-course enter regression model, illness duration was significantly associated with lower CCI/ICV. However, this association was attenuated after chronological age was added to the model. In the exploratory stepwise model, age and sex were retained as significant predictors of CCI/ICV. In the full-sample interaction model, age and sex were associated with CCI/ICV, but their effects did not significantly differ between patients with BD and healthy controls. BD diagnosis remained significantly associated with lower CCI/ICV after accounting for age, sex, and their interactions with group.

Proportional CC integrity appears to be reduced in BD beyond general age- and sex-related variation. Illness duration may represent a clinically meaningful course-related marker of callosal morphology, although its association with CCI/ICV partly overlaps with chronological age. These findings suggest that reduced callosal measures in BD may reflect both disorder-related structural vulnerability and cumulative illness-course effects.