Purpose: To examine the usefulness of diffusion kurtosis imaging for the diagnosis of Parkinson disease (PD).
Materials and methods: Examinations were performed with the understanding and written consent of each subject, with local ethics committee approval, and in compliance with national legislation and Declaration of Helsinki guidelines. Diffusion-weighted magnetic resonance imaging was performed in 30 patients with idiopathic PD (mean age, 64.5 years ± 3.4 [standard deviation]) and 30 healthy subjects (mean age, 65.0 years ± 5.1). Mean kurtosis, fractional anisotropy, and mean, axial, and radial diffusivity of the basal ganglia were compared between the groups. Disease severity was assessed by using Hoehn and Yahr staging and the motor section of the Unified Parkinson's Disease Rating Scale (mean scores, 2.0 and 33.6, respectively). Receiver operating characteristic (ROC) analysis was used to compare the diagnostic accuracies of the indexes of interest. Pearson correlation coefficient analysis was used to correlate imaging findings with disease severity.
Results: Mean kurtosis in the putamen was higher in the PD group (0.93 ± 0.15) than in the control group (0.71 ± 0.09) (P < .000416). The area under the ROC curve (AUC) was 0.95 for both the ipsilateral putamen and the ipsilateral substantia nigra. The mean kurtosis for the ipsilateral substantia nigra had the best diagnostic performance (mean cutoff, 1.10; sensitivity, 0.92; specificity, 0.87). In contrast, AUCs for the tensor-derived indexes ranged between 0.43 (axial and radial diffusivity in substantia nigra) and 0.65 (fractional anisotropy in substantia nigra).
Conclusion: Diffusion kurtosis imaging in the basal ganglia, as compared with conventional diffusion-tensor imaging, can improve the diagnosis of PD.
© RSNA, 2011.