Development of differential diagnostic models for distinguishing between limb-girdle muscular dystrophy and idiopathic inflammatory myopathy

We developed two practical differential diagnosis models for LGMD and IIM based on the analysis of four accessible indicators, including the age of onset, cervical flexor weakness, the ratio of synchronous Mb/CK values and diffuse MHC-I expression. Further studies with larger samples are needed to refine the predictive efficiency of the differential diagnostic models.

A total of 71 IIM patients, 24 LGMDR2 patients and 22 LGMDR1 patients diagnosed at our neuromuscular center were enrolled. Differences in clinical, laboratory and histopathological characteristics were comprehensively compared. A nomogram and a decision tree were developed to distinguish between LGMD and IIM patients.

Limb-girdle muscular dystrophy (LGMD) is usually confused with idiopathic inflammatory myopathy (IIM) in clinical practice. Our study aimed to establish convenient and reliable diagnostic models for distinguishing between LGMD and IIM.

Compared to patients with LGMD, IIM patients exhibited a significantly older age of onset, a higher prevalence of cervical flexor weakness and a more commonly diffuse MHC-I expression on muscle pathology. The ratio of synchronous serum myoglobin (Mb, ng/ml) to creatine kinase (CK, U/L) before immunotherapy was significantly higher in IIM patients than in LGMD patients. Receiver operating characteristic analysis indicated a high differential diagnostic efficiency of synchronous Mb/CK with a cutoff value of 0.18. A nomogram prediction model and a decision tree were developed based on four independent indicators (age of onset, cervical flexor weakness, synchronous Mb/CK and diffuse MHC-I expression). Five-fold cross-validation and bootstrapping techniques substantiated the discriminate efficacy of the nomograph and decision tree.