Do white matter microstructural alterations associated with childhood maltreatment explain psychopathology in early adulthood?-a longitudinal neuroimaging study.

The current study aims to identify tracts in which white-matter changes at age 19 interweave with different experiences of childhood abuse and neglect and whether these changes can explain psychopathology at ages 19 and 22. IMAGEN is a longitudinal, genetic, neuroimaging database in which community adolescents were followed from 14 to 22 years-old across eight sites in Europe. This study includes participants (n = 769, 55% females; Mage = 18.41 and 21.85) with high-quality DTI data who completed the Childhood Maltreatment Questionnaire and self-reported psychopathology symptoms at ages 19 and 22. White-matter changes-assessed via fractional anisotropy (FA)-were computed from Tract-Based Spatial Statistics. Radial diffusion (RD) was further calculated as a complementary indicator. We first isolated WM projections associated with childhood maltreatment to then test their specific impact on clinical problems. Emotional neglect was associated with lower FA, but not RD, in 5 clusters, most of them representing thalamic projections to temporal and prefrontal regions. Only C5 of emotional neglect, comprising several tracts previously reported in the literature, was linked to psychopathology at ages 19 and 22. The effect spans across internalizing and externalizing dimensions at age 19 (hyperactivity and emotional problems), and covers the entire spectrum of psychopathology at age 22. The moderation role of sex revealed that results at age 22 are restricted to female participants. An unexpected, but weak, effect was found in physical neglect in its association with higher FA. Still, this positive association correlated with lower levels of clinical problems at age 19 and did not explain psychopathological outcomes at 22. In conclusion, emotional neglect is associated with developmental brain changes that explain clinical problems at both ages 19 and 22. Since there are no specific-symptom effects, our results possibly unveil a neurobiological risk factor operating at the transdiagnostic level of psychopathology.