Whole-body quantitative muscle MRI in myotonic dystrophy type 2: a sensitive tool for pattern recognition, disease stratification and objective assessment of clinical severity

Abstract

Background Myotonic dystrophy type 2 (DM2) is a frequent yet underdiagnosed muscle disorder. Although muscle MRI is increasingly used in neuromuscular disorders, there is an overt lack of evidence regarding its role in DM2. This study aimed to characterise whole-body muscle involvement using semiquantitative and quantitative MRI (qMRI) and to determine its relationship with clinical disease severity. Methods In this prospective study, 27 genetically confirmed DM2 patients and 26 age-matched and sex-matched healthy controls underwent whole-body MRI. Semiquantitative muscle involvement was assessed using Fischer grading, while qMRI included fat fraction, T2 and diffusion tensor imaging metrics. Imaging findings were correlated with clinical measures including 6-minute walk test (6MWT), Medical Research Council (MRC) sum score, hip flexion strength, pulmonary function, laboratory parameters and CCTG repeat length. Results DM2 patients demonstrated significantly higher muscular degeneration than controls, predominantly involving proximal muscles. Semiquantitative whole-body MRI correlated with MRC sum score, 6MWT and CCTG repeat length. Importantly, the average involvement of the most severely affected muscles closely reflected overall disease burden, indicating that a focused set of muscles adequately captures global muscle involvement. QMRI confirmed these findings, demonstrating increased fat fraction and T2 values, both of which correlated with clinical performance. Conclusions Whole-body qMRI provides a robust, objective biomarker for DM2 severity. The close agreement between focused and whole-body assessment enables efficient disease stratification, with potential applications in clinical trials and personalised patient management. QMRI may provide a standardised imaging outcome measure for future clinical studies in DM2, warranting longitudinal validation.