Premium
The Role of High‐Frequency MRI Monitoring in the Detection of Brain Atrophy in Multiple Sclerosis
Author(s) -
Uher Tomas,
Krasensky Jan,
Sobisek Lukas,
Seidl Zdenek,
Bergsland Niels,
Dwyer Michael G.,
Kubala Havrdova Eva,
Zivadinov Robert,
Horakova Dana,
Vaneckova Manuela
Publication year - 2018
Publication title -
journal of neuroimaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.822
H-Index - 64
eISSN - 1552-6569
pISSN - 1051-2284
DOI - 10.1111/jon.12505
Subject(s) - medicine , multiple sclerosis , atrophy , pathology , radiology , neuroscience , psychiatry , biology
BACKGROUND AND PURPOSE A relatively high intraindividual variability of longitudinal magnetic resonance imaging (MRI) of brain volume loss (BVL) measurements over time renders challenging its application to individual multiple sclerosis (MS) patients. Objective of this study was to investigate if high‐frequency brain MRI monitoring affects identification of pathological BVL in an individual patient. METHODS One hundred fifty‐seven relapsing‐remitting MS patients had seven MRI scans over 12 months follow‐up. All 1,585 MRI scans were performed on the same 1.5T scanner using an identical scanning protocol. Volumetric analysis was performed by ScanView and SIENA software. Linear regression analysis was used for estimation of annualized BVL, with a cutoff greater than .4% defined as pathological. We compared proportions of patients with pathological BVL obtained by analysis of different number of MRI time‐points. RESULTS An analysis of seven MRI scans (months 0, 2, 4, 6, 8, 10, and 12) showed pathological BVL in 105 (65%) of patients. When three MRI scans were included (months 0, 6, and 12), we found 10 (6.4%) false negative and 9 (5.7%) false positive results compared with the analysis of seven MRI scans, used as a reference for assessment of pathological BVL. Analysis of two MRI time‐points (months 0 and 12) showed 10 (6.4%) false negative and 13 (8.3%) false positive results compared with analysis of seven MRI time‐points. Change in the accuracy of pathological BVL between results obtained by analysis of seven and two time‐points was 14.7%. CONCLUSIONS High‐frequency MRI monitoring may have a considerable effect on improving the precision of precisely identifying pathological BVL in individual patients. However, limitations in translation to clinical practice remain.