The Value of Quantitative Susceptibility Mapping and Morphometry in the Differential Diagnosis of Parkinsonism

BACKGROUND AND PURPOSE: Differentiating Parkinson disease (PD) from atypical parkinsonism syndrome (APS), including multiple system atrophy (MSA) and progressive supranuclear palsy (PSP), is challenging, and there is no reference standard. Integrating quantitative susceptibility mapping (QSM) and morphometry can help differentiate PD from APS and improve the internal diagnosis of APS.

MATERIALS AND METHODS: In this retrospective study, we enrolled 55 patients with PD, 17 with MSA-parkinsonian type (MSA-P), 15 with MSA-cerebellar type (MSA-C), and 14 with PSP. Thirty-three age-matched healthy subjects served as controls. All subjects underwent QSM imaging and 3D T1WI with manual quantification of ROI and morphometry. ROIs were selected in the basal ganglia and brainstem nuclei, such as the putamen (Pu), globus pallidus (GP), and red nucleus (RN). Morphometry included MR Parkinson disease index (MRPI), the midbrain area-pons area ratio (M/P), and the ratio of the vertical line of the long axis of the midbrain and pons (Ratio). Differential variables between groups were extracted and a binary logistic regression was established to differentiate the differential diagnoses of PD and APS and diseases within APS. The diagnostic value was assessed using the area under the curve (AUC), sensitivity, and specificity.

RESULTS: The combination of Pu and GP performed best when used to distinguish PD from MSA-P, with an AUC of 0.800 (95% CI: 0.664–0.936). The AUC was optimal when MRPI and M/P were combined to distinguish PD from MSA-C at 0.823 (95% CI: 0.686–0.960). Ratio alone performed best in differentiating PD from PSP, with an AUC of 0.848 (95% CI: 0.711–0.985). The AUC for Ratio alone in distinguishing MSA-P from PSP was 0.871 (95% CI: 0.738–1.0). The AUC when using only M/P to distinguish MSA-C from PSP was 0.931 (95% CI: 0.845–1.0). QSM and morphometry each offer distinct advantages in the differential diagnosis among the aforementioned groups. The combination of QSM and morphometry provided the highest diagnostic value in differentiating PD from APS, highlighting the significance of integrating these 2 imaging techniques for enhanced diagnostic precision in clinical practice. The best indicators described above showed equally high differential diagnostic values in patients with a disease duration of ≤3 years.

CONCLUSIONS: QSM and morphometry will improve the differential diagnosis between PD and APS, as well as improve the internal diagnosis of APS.