BACKGROUND AND PURPOSE: DTI in cerebral X-linked adrenoleukodystrophy may demonstrate abnormalities in both affected and nonaffected WM; these values have not been studied serially after hematopoietic stem cell transplantation. The purpose of this study was to study pretransplant and posttransplant DTI parameters serially and ultimately to determine the ability of pretransplant DTI parameters to predict clinical outcome after HSCT in children with ALD. MATERIALS AND METHODS: Eight patients with posterior-pattern cerebral ALD underwent DTI at 3T before HSCT (T0), at 30-60 days (T1), 90-120 days (T2), 180 days (T3), and 1 year (T4) after HSCT. FA and MD were serially measured in 19 regions, and these measurements were compared with those in control patients. MR imaging severity (Loes) scores were recorded. Correlations were performed between DTI parameters and Loes scores, neurologic function scores, and several neuropsychologic scores. RESULTS: Both FA and MD in subjects differed significantly from that in controls at nearly every time point within cerebellar WM, callosal splenium, and parieto-occipital WM; FA alone was significantly different at each time point within the optic radiations, lateral geniculate, and the Meyer loop (P <.05). Loes scores at T0 correlated strongly with each clinical score at T4 (r = 0.771-0.986, P <.05). The only significant DTI correlation at T0 with a clinical score at T4 was callosal body FA with adaptive function (r = 0.976, P <.001). Correlating the change in DTI values with change in NFS (change between T0 and T4) showed that only AMD within the optic radiations correlated strongly with ANFS (r = 0.903, P <.05). CONCLUSIONS: DTI values at T0 were generally poor predictors of outcome at 1 year, whereas Loes scores were generally good predictors. AMD within the optic radiations strongly correlates with ANFS over that year. In addition, certain normal-appearing regions, such as cerebellar WM, may have DTI abnormalities before HSCT that persist after HSCT.