BACKGROUND: The most frequent phenotypes of dysferlin myopathy are limb-girdle muscular dystrophy 2B (LGMD2B) and Miyoshi myopathy (MM). Our objective was to find clinical or MRI markers to differentiate phenotypes of dysferlin myopathy regardless of initial symptoms. METHODS: This retrospective study included 29 patients with confirmed mutations in the DYSF gene (14 MM, 12 LGMD2B, 1 asymptomatic hyperCKemia, and 2 symptomatic carriers). All underwent an annual clinical examination (Medical Research Council scale), functional status assessment, and creatine kinase, pulmonary, and cardiac testing. For research purposes, we performed lower limb MRI studies in all 29 patients to identify the pattern of muscle impairment and to quantify involvement. Statistical correlations between MRI findings and phenotype, disease duration, and functional status were determined. RESULTS: The mean clinical follow-up was 6.4 ± 5.7 years. No significant differences were found in the rate of progression, functional prognosis, or mutations between patients with MM and patients with LGMD2B. The MRI pattern of muscle involvement was the same for patients with MM and patients with LGMD2B. The adductor magnus and gastrocnemius medialis were the first to be impaired in both phenotypes. The progression of muscle involvement correlated with clinical status. CONCLUSIONS: Splitting dysferlin myopathy into separate phenotypes does not reveal significant differences in terms of rate of progression, prognosis, genotype, or MRI pattern. The finding that proximal and distal muscles are already impaired in the MRI at onset in both MM and LGMD2B favors grouping all phenotypes under the term dysferlin myopathy. © 2010 by AAN Enterprises, Inc.