Marieke van Rosmalen

Prognostic value of quantitative MRI in chronic inflammatory neuropathies 123 7 reflect other processes, for example differences in nerve changes that are part of the natural history of CIDP. Explorative studies of other techniques for biomarker potential, such as nerve conduction and nerve ultrasound, also failed to show changes after the start of immunomodulatory treatment 4,6,8 .This may suggest that effective treatment induces changes in nerves at other levels than those captured by imaging or conduction. Quantitative MRI correlates with histological findings 9–11 . RD indicates hindrance of diffusion for water perpendicular to the nervous tissue. An increased RD indicates an increased possibility of water diffusion and can be the result of demyelination or disturbance of the cytoskeleton 9–11 . The observed RD decreases could thus reflect remyelination following demyelination in CIDP that has been described in histological studies 32–34 . AD reflects the degree of diffusion along the axon and further decreases may correlate with swelling due to the breakdown or change in the permeability of the axolemma and axonal loss. The ongoing decrease of AD in CIDP might indicate progression of axonal loss. Fat fraction is an indicator for the presence of free water in (nervous) tissue and a decrease of fat fraction probably reflects active inflammation. The increase of fat fraction over time could be the result of remyelination or attenuation of inflammation. This study obviously has limitations. Although the sample size is bigger than in previous studies, it remains small, in particular for MMN. More inclusions would obviously have led to a more powered study. Patient characteristics nevertheless showed that the group of patients with MMN was clinically homogeneous and the sample size is in line with the exploratory character of this study. More information could have been obtained with a longer follow-up time, but treatment effects already occur in the course of weeks. We only evaluated the brachial plexus, but this is a site where abnormalities are often found, as shown by previous nerve ultrasound and qualitative MRI studies 35,36 . Based on this first longitudinal study on the prognostic value of quantitative MRI parameters in CIDP and MMN, we conclude that DTI, T2 mapping and fat fraction capture changes in the course of CIDP, but that their value as a prognostic instrument is probably low.