Background Although Magnetic Resonance Imaging has been widely used as a diagnostic tool in neuromuscular disorders, the analysis of images has remained essentially qualitative.
Objectives Even if visual scores of deterioration have been proposed and achieve a good sensitivity their specificity remain low. The calculation and mapping of T2 relaxation time, a quantitative index of fat and muscle physical characteristics, can provide reliable information regarding fatty infiltration and inflammatory signs in neuromuscular disorders.
Methods In addition to routine images acquired using T1 and STIR sequences, T2 mapping has been obtained from the dominant thigh of 55 patients with various muscle diseases (myositis, myopathies, rhabdomyolysis) and 36 control subjects. Based on the T2 values, we have developed an automatic segmentation method in order to distinguish muscle from fat. Two variables have been defined i.e. the most frequent T2 values (T2mf) and the numbers of T2 values ranging from 60 ms (the upper limit of normal muscle) to 75 ms (the lower limit of fat) expressed as a surface (NT2 60-75ms).
Results We observed large T2 variations in patients and controls with a clear positive effect of age (0.23 p<0.001) and no relation to sex. T2mf and NT260–75ms values reduced this variation and were significantly affected by muscle diseases (Table 1). They increased from rabdomyolysis, myositis, to myopathies (Table 2). A diagnostic strategy obtained from a CHAID analysis shows that if STIR images analyses were not convincing to conclude, T2mf and NT260–75ms variable allowed to distinguish patients from controls.
Conclusions T2 mapping is a quantitative approach more robust than the sole visual inspection to characterize changes in the muscles of patients with neuromuscular diseases. Additional indices such as T2mf and NT260–75ms provide alternative measures of disease activity which could be used for diagnosis, investigation of the natural history of muscle diseases and the effects of therapeutic strategies.
Disclosure of Interest None declared
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.