Axonal water fraction as marker of white matter injury in primary-progressive multiple sclerosis: a longitudinal study.

BACKGROUND AND PURPOSE

Diffuse white matter (WM) injury is prominent in primary-progressive multiple sclerosis (PP-MS) pathology and is a potential biomarker of disease progression. Diffusion kurtosis imaging allows the quantification of non-Gaussian water diffusion, providing metrics with high WM pathological specificity. The aim of this study was to characterize the pathological changes occurring in the normal-appearing WM of patients with PP-MS at baseline and at 1-year follow-up and to assess their impact on disability and short-term disease progression.

METHODS

A total of 26 patients with PP-MS and 20 healthy controls were prospectively enrolled. Diffusion kurtosis imaging single-shot echo-planar imaging (EPI) was acquired on a 3-T scanner (Philips Achieva, Best, The Netherlands) (voxel size, 2 × 2 × 2 mm , 30 directions for each b-value = 1000, 2000 s/mm and one b = 0 s/mm ). A two-compartment biophysical model of WM tract integrity was used to derive spatial maps of axonal water fraction (AWF), intra-axonal diffusivity, extra-axonal axial and radial diffusivities (D , D ) and tortuosity from the following WM tracts: corpus callosum (CC), corticospinal tract (CST) and posterior thalamic radiation (PTR).

RESULTS

At baseline, patients with PP-MS showed a widespread decrease of AWF, tortuosity and D and an increase of D in CC, CST and PTR (P ranging from 0.001 to 0.036). At 1-year follow-up, a significant AWF decrease was detected in the body of CC (P = 0.048), PTR (P = 0.008) and CST (P = 0.044). Baseline AWF values in CST significantly discriminated progressed from non-progressed patients (P = 0.021; area under the curve, 0.854).

CONCLUSION

Based on its change over time and its relationship with disease progression, among the analyzed metrics, AWF seems the most sensitive metric of WM tissue damage in PP-MS and therefore it could be considered as a marker for monitoring disease progression.