Medizinische Universität Graz Austria/Österreich - Forschungsportal - Medical University of Graz
Gewählte Publikation:
Damulina, A.
Comparison of brain volume changes and lesion metrics in multiple sclerosis between two MS centers in Graz and Linz
Studium für die Gleichwertigkeit; Humanmedizin; [ Diplomarbeit ] Graz Medical University; 2018. pp. 70
[OPEN ACCESS]
FullText
- Autor*innen der Med Uni Graz:
- Betreuer*innen:
- Enzinger Christian
- Pichler Alexander
- Altmetrics:
- Abstract:
- Background and objective: Cerebral Magnetic Resonance Imaging (MRI) has been proposed for disease monitoring in patients with Multiple Sclerosis (MS). However, outside standardized trials, considerable interscanner and intrascanner variability has to be expected for such measures. So far, the reliability of such brain measures in clinical practice has received relatively little systematic investigation. Our aim thus was to determine the effect of using different MRI protocols at different scanners on the measurement of lesion load and brain volumes metrics in clinical similar cohorts of MS patients of two different centers. Methods: 174 MS patients from two MS centers (3D MPRAGE MR scans of 111 patients from Graz and 2D SE MR scans of 63 patients from Linz) at baseline and follow-up (after a mean of 3.42+1.58 years) were analyzed by using SIENAX and SIENA. Patients were comparable regarding age, disease duration, Expanded Disability Status Scale, and the proportion of patients on disease-modifying therapy. Variability between 2D and 3D data segmentation was estimated by the difference in brain metrics, percentage brain volume change (PBVC), coefficients of variation and percentage differences in volumes of brain metrics. Results: Lesion load estimation showed similar T2 hyperintense lesion load but significantly different T1 hypointense (“black hole”) lesion load between the two cohorts (0.35 cm3 in Graz and 0 cm3 in Linz). Results from SIENAX varied significantly for all brain metrics except the global brain volume between the Graz and Linz cohorts (white matter volume 768 cm3 and 854 cm3, grey matter volume 797 cm3 and 706 cm3, respectively). Mean percentage differences showed that despite similar results concerning the global brain volume, the ratios of white and grey matter volumes between the two centers were completely different (-12% in grey and 10% in white matter between the cohorts). SIENA analyses demonstrated that PBVC was significantly lower in Linz cohort than in Graz (-0.19% and -0.39%, respectively). Conclusions: Even in cohorts of MS patients with similar clinical characteristics, obtaining MRI data at different scanners (despite locally standardized protocols) yields significant differences in several lesion and brain volume metrics. Given the suggested role of brain volume measurements as a surrogate of atrophy in the concept of “no evidence of disease activity” mirroring the neurodegenerative component of the disease, this study emphasizes the need for staying cautious when interpreting such values both within as well as across cohorts of different centers.