Study on the value of different diagnostic/classification criteria for the diagnosis of Takayasu's arteritis / 中华风湿病学杂志

ABSTRACT

Objective:To compare the diagnostic efficacy of Chinese diagnostic model, the 1990 American College of Rheumatology (ACR) classification criteria and the 2018 ACR new classification criteria (draft) for Takayasu arteritis (TA).Methods:A total of 196 TA patients who came to our hospital from January 1, 2009 to May 31, 2019 in the TA database of the department of rheumatology and immunology, Zhongshan Hospital, Fudan University and 131 patients with other vascular diseases visited during the same period were selected. General characteristics, clinical data, laboratory tests and imaging tests of all patients were collected. Categorical variables were presented as numbers and percentages, between-group differences were analyzed using the χ2 test. Continuous variables were presented as the Mean± SD for a normal distribution or median and interquartile range (IQR) for a non-normal distribution. Between-group differences were analyzed using the Student's t-test or Mann- Whitney test, as appropriate. The sensitivity, specificity, positive predictive value, negative predictive value, accuracy and receiver operating characteristics of the above diagnostic/classification criteria area under the curve were analyzed. P<0.05 was considered significant. Results:In terms of sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and area under receiver operating characteristic curve (ROC), Chinese diagnostic models was 85.7%, 96.2%, 89.9%, 97.1%, 81.5%, 0.909, 1990 ACR criteria was 47.4%, 97.7%, 67.6%, 96.9%, 55.4%, 0.726, 2018 ACR classification criteria was 79.1%, 98.5%, 86.5%, 98.7%, 75.9%, 0.888. The difference between the Chinese diagnostic model and the 2018 ACR criteria in AUC was not statistically significant ( Z=1.186 , P>0.05). The sensitivity, accuracy and diagnostic efficiency of Chinese diagnostic model was the best, that of the 1990 ACR classification criteria was the worst, and the specificity of the 2018 ACR classification criteria was the highest. The Kappa value of the 2018 ACR classification criteria and the Chinese diagnostic model was 0.719, which had good consistency, and the Kappa value of the consistency between the 1990 ACR classification criteria and the Chinese TA diagnostic model was 0.516. Conclusion:The Chinese diagnostic model, which is based on the clinical characteristics of the Chinese TA population, has a good diagnostic efficacy for the Chinese population. The 2018 ACR classification criteria (draft) is highly consistent with the Chinese TA diagnostic model, and can be promoted and applied in practice.