A diffusion MRI-derived perivascular metric related to glymphatic-associated processes in bipolar disorder vulnerability: Multimodal correlates across emotion dysregulation patients and offspring.

Résumé

Bipolar disorder (BD) is characterized by marked emotion dysregulation and high familial risk. Identifying early biological markers of vulnerability in BD, including unaffected offspring, is critical to improve risk stratification and intervention, and glymphatic-associated processes may contribute to this vulnerability. We examined 237 participants, including 97 patients with emotion dysregulation disorders (EDD; 34 BD, 33 borderline personality disorder [BPD], 30 attention-deficit/hyperactivity disorder [ADHD]), 67 offspring of EDD patients (EDDoff; including 23 BD offspring [BDoff]), and 73 healthy controls (CTRL). All participants underwent clinical assessments, diffusion and resting-state functional MRI, and serum immune and neurotrophic biomarker sampling. Perivascular diffusion was estimated using diffusion tensor imaging analysis along the perivascular space (DTI-ALPS), a diffusion-derived perivascular metric proposed to relate to glymphatic-associated processes, and compared across groups. To explore multimodal correlates of DTI-ALPS, principal component analyses (PCA) were conducted across clinical, biological, and neuroimaging domains. BD patients and BDoff showed significantly reduced DTI-ALPS compared with CTRL, ADHD, and BPD, supporting its potential role as a vulnerability-related imaging feature for BD. By contrast, ADHD and BPD showed comparable or higher DTI-ALPS relative to controls. Across participants, DTI-ALPS was associated with components reflecting white matter integrity and serum immune and neurotrophic markers. Within BD, lower DTI-ALPS correlated with more manic episodes and poorer working memory. Reduced DTI-ALPS may represent a BD-specific vulnerability-related feature, observable in both patients and at-risk offspring, and not shared by other emotion dysregulation disorders. Multimodal associations with white matter, inflammation, and symptoms underscore its relevance for risk stratification in high-risk populations.