Longitudinal quantitative MRI assessment of cortical damage in multiple sclerosis: A pilot study
2017
PURPOSE: Quantitative MRI (qMRI) allows assessing cortical pathology in multiple sclerosis (MS) on a microstructural level, where cortical damage has been shown to prolong T1 -relaxation time and increase proton density (PD) compared to controls. However, the evolution of these changes in MS over time has not been investigated so far. In this pilot study we used an advanced method for the longitudinal assessment of cortical tissue change in MS patients with qMRI in comparison to cortical atrophy, as derived from conventional MRI. MATERIALS AND METHODS: Twelve patients with relapsing-remitting MS underwent 3T T1 /PD-mapping at two timepoints with a mean interval of 12 months. The respective cortical T1 /PD-values were extracted from the middle of the cortical layer and the cortical thickness was measured for surface-based identification of clusters with increasing/decreasing values. RESULTS: Statistical analysis showed clusters with increasing PD- and T1 -values over time (annualized rate for T1 /PD increase in these clusters: 3.4 ± 2.56% for T1 , P = 0.0007; 2.3 ± 2.59% for PD, P = 0.01). Changes are heterogeneous across the cortex and different patterns of longitudinal PD and T1 increase emerged. Analysis of the cortical thickness yielded only one small cluster indicating a decrease of cortical thickness. CONCLUSION: Changes of cortical tissue composition in MS seem to be reflected by a spatially inhomogeneous, multifocal increase of the PD values, indicating replacement of neural tissue by water, and of the T1 -relaxation time, a surrogate of demyelination, axonal loss, and gliosis. qMRI changes were more prominent than cortical atrophy, showing the potential of qMRI techniques to quantify microstructural alterations that remain undetected by conventional MRI. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2017;46:1485-1490.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
32
References
13
Citations
NaN
KQI