Study on the impact of meditation intervention on the white matter structure in schizophrenia

Abstract

Schizophrenia is associated with abnormal white matter structural connectivity, and meditation has been investigated as an adjunctive intervention for clinical symptoms. This study examined whether an 8-month meditation intervention was associated with longitudinal changes in clinical symptoms, white matter structural connectivity, and graph-theoretical network properties in participants with chronic schizophrenia. Sixty-four participants diagnosed with chronic schizophrenia were enrolled and randomly assigned to either the meditation (Med) group or the general rehabilitation programs (GRP) group, and were followed for 8 months. Diffusion MRI data and clinical symptoms were collected at baseline and at 3 and 8 months after the start of the intervention. A total of 28 Med and 26 GRP participants were included for analysis. The Group × Time interaction was significant for PANSS positive score (F = 6.94, P FDR = 0.028, η 2 = 0.118) and PANSS total score (F = 6.49, P FDR = 0.028, η 2 = 0.111), with the Med group showing greater longitudinal symptom reduction. Across 4005 structural connectivity edges, no Group × Time interaction survived FDR correction (17 edges showed nominal uncorrected P < 0.05; minimum P FDR = 1.000). Node-level graph metric analyses (540 tests across six metrics × 90 nodes) yielded no FDR-significant Group × Time effects. Among 816 exploratory Spearman correlations between edge-level FA changes and PANSS symptom changes, eight survived all-correlation FDR correction (all involving the right anterior cingulum-right orbital middle frontal gyrus connection). A covariate-adjusted sensitivity analysis controlling for age, education years, and baseline chlorpromazine-equivalent dose confirmed that no edge survived whole-family FDR correction. The 8-month meditation intervention was associated with greater reductions in PANSS positive and total scores than the active control condition. White matter edge-level and node-level findings did not survive stringent multiple-comparison correction and should be interpreted as exploratory; the within-group changes over time cannot be attributed specifically to meditation. The brain-clinical associations are likewise exploratory and require confirmation in larger studies.