Abstract
BACKGROUND: White matter alterations observed using diffusion weighted imaging have become a hallmark of chronic schizophrenia, but it is unclear when these changes arise over the course of the disease. Nearly all studies reported to date have been cross-sectional, so despite their large sample sizes, they cannot determine whether changes accumulate as a degenerative process or patients with preexisting white matter damage are predisposed to more chronic forms of schizophrenia. METHODS: We examined 160 scans comprising 2 years of annual follow-up data from 42 control participants and 28 patients with schizophrenia recruited in the first 2 years since their diagnosis, totaling 2 to 3 scans per participant. We also examined 6-month follow-up data obtained from an ultra-high field (7T) scanner (68 scans; n = 19 patients with first-episode schizophrenia, n = 15 control participants) as a validation dataset. A longitudinal model was used to compare the trajectory of diffusion tensor parameters in patients and control participants. RESULTS: Positive and negative symptom scores were correlated with diffusion parameters using region of interest-based approaches. No longitudinal differences between patients and control participants were observed for any diffusion tensor imaging parameter in either dataset. However, we did observe consistent associations between white matter alterations and negative symptoms in both datasets. CONCLUSIONS: White matter does not appear to be susceptible to schizophrenia-linked degeneration in the early stages of disease, but preexisting pathology may be linked to disease severity.