BackgroundDiagnosis of idiopathic inflammatory myopathies (IIM) represents a clinical challenge and muscle biopsy itself is regarded as the cornerstone for confirming diagnosis of IIM. However, the significance of muscle biopsy as diagnostical tool remains inconclusiveObjectivesTo determine the diagnostic accuracy and clinical relevance of muscle biopsy in patients with suspected myositis.MethodsIn this retrospective cohort study, histopathological findings of muscle biopsy specimens of adult (≥ 18 years of age) patients with clinically suspected or differential diagnosis of myositis referred from all over Austria to the Medical University of Vienna through the period of 01.01.2007-31.10.2021 were analyzed. Following information were extracted from handwritten assignment sheets: referral department, suspected diagnosis (inflammatory and/or non-inflammatory myopathy), sampling location, demographic, clinical, laboratory and serologic data, and imaging results.Histological findings were grouped in: 1) inflammatory, 2) non-inflammatory (neurogenic, metabolic etc.), 3) inflammatory and non-inflammatory mixed, and 4) normal. Clinical information on start of immunosuppressive treatment after muscle biopsy could be partly retrieved form electronical medical records (EMR). Sensitivity, specificity, positive (PPV) and negative predictive values (NPV) of histological results were calculated using clinical diagnosis (defined as initiation of immunosuppressive therapy) as gold standard.Results778 muscle biopsy specimens of 762 patients with suspected myositis (49.6% female, mean age of 56 years) were analyzed. Cohort descriptives are displayed in Table 1.Overall, 373/778 (47.9%) muscle biopsies revealed inflammatory features only or had mixed results with non-inflammatory characteristics. Out of 541 samples with suspected diagnosis of myositis, diagnosis was confirmed histologically in 300 (55.5%) cases; 46 (8.5%) specimens additionally showed co-features of non-inflammatory myopathy, whereas 221 (40.3%) were non-inflammatory at all and 15 (2.8%) were normal. Proportion of myositis was more pronounced, when CK was elevated at time of biopsy (Figure 1).Figure 1.From suspicion to histological verification depending on CK-levelsOut of 237 samples with differential diagnosis of non-inflammatory myopathy in addition to myositis, 53 (22.4%) were histology. confirmed as inflammatory only and 20 (8.4%) showed mixed features. 153 (64.6%) biopsies were non-inflammatory only and 8 (3.4%) normal. Samples with no pathological features were observed in 24/778 (2.3%) of the cases, with 0.8% sampling error (no muscle tissue).Of 421/778 (54.1%) patients, clinical data could be extracted from EMR of which 209/421 (49.6%) had histological features of inflammation. In 287/421 (68%) cases, information regarding initiation of immunosuppressive treatment were available on EMR. 224/287 (78%) patients received immunosuppression (steroids and/or disease-modifying anti-rheumatic drug [DMARD]), thus fulfilled clinical diagnosis of myositis. In 155/224 (69.2%) cases, muscle biopsy confirmed histology. features of inflammation. Tests on diagnostic accuracy of muscle biopsy using initiation of any immunosuppressive treatment as gold standard were performed. The sensitivity of the test (biopsy) for diagnosing IIM was 71.4% with a specificity of 79.6%. PPV was high (92.8%), whereas NPV low (43.1%).ConclusionLess than 50% of biopsy samples with suspected diagnosis of myositis histologically revealed an inflammatory myopathy. The percentage increased when CK levels were elevated. Sensitivity and specificity of muscle biopsy was moderate, when using start of immunosuppressive therapy as gold standard for IIM diagnosis.