In search of the optimal MRI marker for progressive supranuclear palsy: a large, single-center, retrospective study on the effect of phenotype, diagnostic certainty and disease duration

Le résumé fourni par la source

INTRODUCTION: Multiple MRI markers have been introduced as surrogate markers of progressive supranuclear palsy (PSP). Midbrain surface, midbrain/pons surface ratio (M/P) and the Magnetic Resonance Parkinsonism Index (MRPI) have produced high diagnostic accuracy in differentiating PSP from other parkinsonian disorders. A systematic comparison of the diagnostic accuracy of available MRI markers, and the effect of disease duration, clinical presentation, level of clinical certainty on the performance of these markers is lacking. MATERIALS AND METHODS: In this single-center, retrospective study, 244 subjects were included (80 PSP, 38 corticobasal degeneration, 45 multiple system atrophy, 36 Parkinson's disease patients, 45 control subjects). All patients underwent a standardized MRI acquisition protocol and automated MRI data preprocessing through Freesurfer for volumetric data. Midbrain distance, surface and volume, superior cerebellar peduncle (SCP) width and volume, and composite markers including the M/P and M/P 2.0 ratios and the MRPI and MRPI 2.0 were measured. The diagnostic accuracy of these markers was calculated, and the effect of clinical phenotype, level of disease certainty and disease duration were investigated. RESULTS: Surface-based MRI markers were superior to distance- and volume-based markers. Midbrain surface was the optimal MRI marker for PSP (AUC = 0.937; sensitivity 88.8%; specificity 84.2%). MRI markers were more accurate in diagnosing probable PSP vs. possible/suggestive PSP and PSP-Richardson syndrome vs. PSP variants. MRI markers exhibited comparable diagnostic accuracy in early (≤ 24 months) vs. late (≤ 48 months) disease stages. Midbrain atrophy correlated with PSP disease severity as well as ocular motor, gait and bulbar deficit severity. DISCUSSION: Planimetric MRI markers are optimal for PSP diagnosis. Level of disease certainty (i.e. probable vs. possible/suggestive) and clinical presentation (PSP-RS vs. PSP variants) affect the diagnostic accuracy of MRI markers. MRI markers are useful even in early (≤ 24 months) stages of PSP.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.