Comparison of the Efficiency of Fractional Anisotropy or Relaxation Time T2* in the Early Diagnosis of 6-Hydroxydopamine-Induced Rat Model of Parkinson's Disease.

Assessment of fractional anisotropy (FA) in diffusion tensor imaging (DTI) and measurement of transverse relaxation time T2* in T2*-weighted imaging (T2*WI) are the basic MRI methods used in the diagnosis of Parkinson's disease (PD). However, it remains unclear which of these methods is more effective. We compared the sensitivity and specificity of DTI and T2*WI methods to determine which one is more effective in the early diagnosis of PD. For PD modeling, male Sprague-Dawley rats (n = 20) were injected with 6-hydroxydopamine (6-OHDA) into the right substantia nigra (SN). MRI was performed and FA and T2* of each SN were measured and analyzed every week after surgery over 6 weeks. The ROC curves were used to evaluate sensitivity and specificity of these parameters in the early diagnosis of PD. Significant differences between the lesioned and control SN were revealed for FA and T2* values on weeks 1, 2, 5, and 6 and on weeks 5 and 6 after injection, respectively (p < 0.01). Six weeks after injection, the Youden's indices of FA and T2* were higher compared to the earlier terms of the experiment. The optimal cut-off values of FA and T2* were 0.395 (area under the ROC curve (AUC) = 0.9075, 100% sensitivity, and 70% specificity) and 9.48 msec (AUC = 0.9450, 90% sensitivity, and 95% specificity), respectively. In 6 weeks after 6-OHDA injection, the diagnostic effectiveness of FA and T2* was higher than that determined in previous weeks. The T2* was more effective than FA for early diagnosis of PD.