Magnetic Resonance Imaging Biomarkers Distinguish Normal Pressure Hydrocephalus From Progressive Supranuclear Palsy

• Published in: Movement disorders Vol. 35; no. 8; pp. 1406 - 1415
• Main Authors: Quattrone, Andrea, Sarica, Alessia, La Torre, Domenico, Morelli, Maurizio, Vescio, Basilio, Nigro, Salvatore, Barbagallo, Gaetano, Nisticò, Rita, Salsone, Maria, Arcuri, Pier Paolo, Novellino, Fabiana, Bianco, Maria Giovanna, Arabia, Gennarina, Cascini, Giuseppe, Quattrone, Aldo
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.08.2020; Wiley Subscription Services, Inc
• Subjects: automated ventricular volumetry; Automation; Biomarkers; callosal angle; Differential diagnosis; Humans; Hydrocephalus; Hydrocephalus, Normal Pressure - diagnostic imaging; idiopathic normal pressure hydrocephalus; Magnetic Resonance Imaging; Movement disorders; MR Hydrocephalic Index; Paralysis; Parkinson's disease; Pressure; Progressive supranuclear palsy; Reproducibility of Results; Supranuclear Palsy, Progressive - diagnostic imaging; Ventricle; MR Hydrocephalic Index; automated ventricular volumetry; idiopathic normal pressure hydrocephalus; callosal angle; progressive supranuclear palsy
• ISSN: 0885-3185; 1531-8257; 1531-8257
• DOI: 10.1002/mds.28087

ABSTRACT Background Idiopathic normal pressure hydrocephalus and PSP share several clinical and radiological features, making differential diagnosis, at times, challenging. Objectives To differentiate idiopathic normal pressure hydrocephalus from PSP using MR volumetric and linear measurements. Methods Twenty‐seven idiopathic normal pressure hydrocephalus patients, 103 probable PSP patients, and 43 control subjects were consecutively enrolled. Automated ventricular volumetry was performed using Freesurfer 6 on MR T1‐weighted images. Linear measurements, such as callosal angle and a new measure, termed MR Hydrocephalic Index, were calculated on MR T1‐weighted images. Receiver operating characteristic analyses were used for differentiating between patient groups. Generalizability and reproducibility of the results were validated, dividing each participant group in two cohorts used as training and testing subsets. Results Ventricular volumes and linear measurements (callosal angle and Magnetic Resonance Hydrocephalic Index) revealed greater ventricular enlargement in patients with idiopathic normal pressure hydrocephalus than in PSP patients and controls. PSP patients had ventricular volume larger than controls. Automated ventricular volumetry and Magnetic Resonance Hydrocephalic Index were the most accurate measures (98.5%) in differentiating patients with idiopathic normal pressure hydrocephalus from PSP patients, whereas callosal angle misclassified several PSP patients and showed low positive predictive value (70.0%) in differentiating between these two diseases. All measurements accurately differentiated idiopathic normal pressure hydrocephalus patients from controls. Accuracy values obtained in the training set (automated ventricular volumetry, 98.4%; Magnetic Resonance Hydrocephalic Index, 98.4%; callosal angle, 87.5%) were confirmed in the testing set. Conclusions Our study demonstrates that AVV and Magnetic Resonance Hydrocephalic Index were the most accurate measures for differentiation between idiopathic normal pressure hydrocephalus and PSP patients. Magnetic Resonance Hydrocephalic Index is easy to measure and can be used in clinical practice to prevent misdiagnosis and ineffective shunt procedures in idiopathic normal pressure hydrocephalus mimics. © 2020 International Parkinson and Movement Disorder Society