External validation of the 2017 ACR/EULAR classification criteria for inflammatory myopathies in a Mexican cohort: Role of autoantibodies in the diagnosis and classification of patients with inflammatory myopathies

Abstract

Objective

This retrospective study aimed to perform the first external validation of the ACR/EULAR classification criteria for inflammatory myopathy (IIM) in a Mexican dynamic cohort where the patients were evaluated with clinical and laboratory values. As secondary objectives, we presented the clinical characteristics of the patients and included antibodies other than anti Jo1 to evaluate their impact on our population.

Methodology

This study included 70 patients with IIM and 70 patients with differential diagnoses of IIM, according to the absolute score of the classification criteria. We obtained sensitivity and specificity in the modality without biopsy, and as an exploratory analysis, we added other antibodies from the myositis extended panel. We analyzed the area under the curve (AUC) of three models: score without antibodies, with anti Jo1 and with any antibody.

Results

The ACR/EULAR criteria showed increased specificity and at least similar sensitivity to that of the original cohort (85% sensitivity and 92% specificity), with a cohort point of >55%. When we classified patients into definite, probable, possible, and no IIM categories, by adding the extended myopathy panel, 6 of the 10 patients initially classified as “no IIM” changed their classification to “Probable IIM” and 4 to “Definite IIM”; of the 16 patients classified as “probable IIM,” 15 changed their classification to “Definite IIM.”

Conclusion

Considering the limitations of this study, we concluded that the 2017 EULAR/ACR criteria for IIM classification are sensitive and specific for classifying patients with IIM in the Mexican population. Additionally, the addition of antibodies other than anti-Jo1 may improve performance in certain populations.