Chungnam National University Yuseong-gu, Daejeon, South Korea
Min-joo Ahn1, Su-Jin Yoo2, Seung-Cheol Shim2, Seong Wook Kang2 and Jinhyun Kim3, 1Chungnam National University, Yuseong-gu, Daejeon, Republic of Korea, 2Chungnam National University Hospital, Daejeon, Republic of Korea, 3Chungnam National University Hospital, Jung-gu, Daejeon, South Korea
Background/Purpose: Anti-synthetase syndrome (ASS) is a distinct subset of idiopathic inflammatory myopathies (IIM), characterized by antibodies targeting an aminoacyl tRNA synthetase (ARS) along with clinical features including myositis, interstitial lung disease (ILD), non-erosive arthritis, Raynaud's phenomenon, unexplained fever, and mechanic's hands. Autoantibodies are thought to play a key role in the pathogenesis of IIM, but 40% of IIM patients, even those with clinical features of ASS, test negative to all known myositis-specific antibodies (MSAs). This study aimed to compare the diagnostic performance of the Connors', Solomon's, and Lega's diagnostic criteria, and to propose how to classify ASS in patients without a known ARS.
Methods: Data from 154 consecutive patients with suspected IIM presenting at our institute between May 2016 and March 2022 were retrospectively reviewed. We checked the fulfillment of Connors', Solomon's, and Lega's criteria sets in all patients. MSAs were tested with Euroline myositis profile or ELISA. The sensitivity, specificity, and accuracy of the three criteria sets were compared with the treating physician's diagnosis as the gold standard. The area under the receiver-operating-characteristic curve (AUROC) was used.
Results: Twenty patients were diagnosed with ASS. Another 4 were suspected to have ASS by their treating physicians but were negative for an ARS (possible ASS). Seventy-two patients were diagnosed with IIM other than ASS while 58 patients had diagnoses other than IIM. The sensitivities were 100% for Connors' criteria set, 75% for Solomon's, and 100% for Lega's when possible ASS patients were excluded. Their specificities were 97.7, 99.2, and 98.5%, respectively. When the possible ASS patients were included, the sensitivities dropped to 83.3, 62.5, and 83.3%, respectively. Connors' and Lega's criteria sets were more sensitive than Solomon's while specificities did not differ between the 3 criteria sets. Of the patients who did not have a known ARS, 15 patients tested positive to cytoplasmic ANA. In these patients, the presence of 1 or more clinical features included in Connors' criteria except ARS was highly associated with the diagnosis of ASS (area under the curve [AUC]=0.909, 95% CI=0.739, 1.000), with 3 features an optimal cut-off value (sensitivity 100%, specificity 90.9%). When cytoplasmic ANA positivity plus 3 or more features of ASS were added to the 3 criteria sets except ARS, sensitivities improved to 100, 79.2, and 100% for Connors', Solomon's, and Lega's, respectively, with only a slight decrease in specificities (97.0%, 98.5%, and 97.7%, respectively).
Conclusion: Connors' and Lega's criteria sets showed higher sensitivity and a comparable specificity compared with Solomon's criteria. The addition of positive cytoplasmic ANA and 3 or more features of ASS to the criteria sets could improve diagnostic accuracy by enabling suspected patients without ARS to be diagnosed with ASS.
Disclosures: M. Ahn, None; S. Yoo, None; S. Shim, None; S. Kang, None; J. Kim, None.