FDG-PET patterns associated with underlying pathology in corticobasal syndrome

Abstract

Objective: To evaluate brain ¹⁸Fluorodeoxyglucose PET (FDG-PET) differences among patients with a clinical diagnosis of corticobasal syndrome (CBS) and distinct underling primary pathologies.

Methods: We studied 29 patients with a diagnosis of CBS who underwent FDG-PET scan and postmortem neuropathologic examination. Patients were divided into subgroups on the basis of primary pathologic diagnosis: CBS-corticobasal degeneration (CBS-CBD) (14 patients), CBS-Alzheimer disease (CBS-AD) (10 patients), and CBS-progressive supranuclear palsy (CBS-PSP) (5 patients). Thirteen age-matched healthy patients who underwent FDG-PET were the control group (HC). FDG-PET scans were compared between the subgroups and the HC using SPM-12, with a threshold of p _FWE < 0.05.

Results: There were no differences in Mattis Dementia Rating Scale or finger tapping scores between CBS groups. Compared to HC, the patients with CBS presented significant hypometabolism in frontoparietal regions, including the perirolandic area, basal ganglia, and thalamus of the clinically more affected hemisphere. Patients with CBS-CBD showed a similar pattern with a more marked, bilateral involvement of the basal ganglia. Patients with CBS-AD presented with posterior, asymmetric hypometabolism, including the lateral parietal and temporal lobes and the posterior cingulate. Finally, patients with CBS-PSP disclosed a more anterior hypometabolic pattern, including the medial frontal regions and the anterior cingulate. A conjunction analysis revealed that the primary motor cortex was the only common area of hypometabolism in all groups, irrespective of pathologic diagnosis.

Discussion and conclusions: In patients with CBS, different underling pathologies are associated with different patterns of hypometabolism. Our data suggest that FDG-PET scans could help in the etiologic diagnosis of CBS.

Figure 1. Regional hypometabolism compared to controls in the whole corticobasal syndrome cohort

Statistical significance set at p < 0.05 family-wise error–corrected for multiple comparisons, minimum cluster size: 100 voxels. The right side of the image represents the hemisphere contralateral to the more severely affected limbs. Significant clusters overlaid on a volumetric brain MRI template.

Figure 2. Voxel-wise correlations

Voxel-wise correlations between ¹⁸Fluorodeoxyglucose–PET scans and Mattis Dementia Rating Scale–2 total score (A), finger tapping (B), Test of Oral and Limb Apraxia sum of standard scores (C), and Wechsler Adult Intelligence Scale III vocabulary score (D). Statistical significance set at p < 0.001 uncorrected for multiple comparisons, minimum cluster size: 100 voxels. The right side of the image represents the hemisphere contralateral to the more severely affected limbs. Significant clusters overlaid on a volumetric brain MRI template.

Figure 3. Regional hypometabolism

Regional hypometabolism compared to controls in the corticobasal syndrome (CBS)–corticobasal degeneration (CBD) (A), CBS–Alzheimer disease (B), and CBS–progressive supranuclear palsy (C) cohorts.(D) Conjunction analysis among the 3 groups. Statistical significance set at p < 0.05 family-wise error–corrected for multiple comparisons, except for C, where the threshold is p < 0.0001 uncorrected. Minimum cluster size: 100 voxels. The right side of the image represents the hemisphere contralateral to the more severely affected limbs. Significant clusters overlaid on a volumetric brain MRI template.