Pathological cut-offs of global and regional brain volume loss in multiple sclerosis

Tomas Uher, Manuela Vaneckova, Jan Krasensky, Lukas Sobisek, Michaela Tyblova, Jana Volna, Zdenek Seidl, Niels Bergsland, Michael G Dwyer, Robert Zivadinov, Nicola De Stefano, Maria Pia Sormani, Eva Kubala Havrdova, Dana Horakova

Research output: Contribution to journalArticlepeer-review

Abstract

BACKGROUND: Volumetric MRI surrogate markers of disease progression are lacking.

OBJECTIVE: To establish cut-off values of brain volume loss able to discriminate between healthy controls and MS patients.

METHODS: In total, 386 patients after first demyelinating event suggestive of MS (CIS), 964 relapsing-remitting MS (RRMS) patients, 63 secondary-progressive MS (SPMS) patients and 58 healthy controls were included in this longitudinal study. A total of 11,438 MRI scans performed on the same MRI scanner with the same protocol were analysed. Annualised percentage changes of whole brain, grey matter, thalamus and corpus callosum volumes were estimated. We investigated cut-offs able to discriminate between healthy controls and MS patients.

RESULTS: At a predefined specificity of 90%, the annualised percentage change cut-off of corpus callosum volume (-0.57%) was able to distinguish between healthy controls and patients with the highest sensitivity (51% in CIS, 48% in RRMS and 42% in SPMS patients). Lower sensitivities (22%-49%) were found for cut-offs of whole brain, grey matter and thalamic volume loss. Among CIS and RRMS patients, cut-offs were associated with greater accumulation of disability.

CONCLUSION: We identified cut-offs of annualised global and regional brain volume loss rates able to discriminate between healthy controls and MS patients.

Original languageEnglish
Pages (from-to)1352458517742739
JournalMultiple Sclerosis
DOIs
Publication statusE-pub ahead of print - Nov 1 2017

Fingerprint

Dive into the research topics of 'Pathological cut-offs of global and regional brain volume loss in multiple sclerosis'. Together they form a unique fingerprint.

Cite this