National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD, USA.
Oral Dis. 2018 Mar;24(1-2):184-190. doi: 10.1111/odi.12772.
The introduction of new classification criteria for Sjögren's syndrome, known as the 2016 American College of Rheumatology/European League against Rheumatism Classification Criteria (ACR-EULAR), created a need for the evaluation of its performance in an external cohort. The purpose of this study was to compare the performance of the 2016 ACR-EULAR classification set with the widely used American-European Consensus Group Classification criteria (AECG) in the cohort at the National Institutes of Health, USA, and to compare the performance of the sets in classifying both primary and secondary Sjögren's syndrome (pSS and sSS).
The study cohort at the NIH (N = 1,303) was enrolled for clinical suspicion of SS. Participants were classified as SS, pSS, and sSS according to both classification sets. Performance of 2016 ACR-EULAR and AECG sets was compared holding each as gold standard to the other. Statistical analysis of test diagnostics and agreement between the two sets were undertaken.
By the AECG set, 701 were classified as having SS (627 pSS, 74 sSS) and 714 were classified with SS (647 pSS, 67 sSS) by the 2016 ACR-EULAR set. Sensitivity and specificity of the two sets were comparable in classifying SS, pSS, and sSS. There was high agreement between the two sets for classifying SS (κ = 0.79), pSS (κ = 0.81), and sSS (κ = 0.87). The specificity of the 2016 ACR-EULAR set was significantly higher for classifying sSS than pSS, while the sensitivity was similar for the two disease groups. However, this pattern was also exhibited by the AECG set.
There was high agreement between the two classification sets with comparable performance diagnostics. There was no evidence of superior performance value by the new 2016 ACR-EULAR set over the AECG set, and the two sets were found to be equivalent. Findings from our cohort indicate that 2016 ACR-EULAR classification could be extended to classification of sSS.
为了评估新的干燥综合征分类标准（即 2016 年美国风湿病学会/欧洲抗风湿病联盟分类标准，ACR-EULAR）在外部队列中的表现，引入了该标准。本研究的目的是比较该标准与广泛使用的美国-欧洲共识组分类标准（AECG）在国立卫生研究院（美国）队列中的表现，并比较这两种标准在原发性和继发性干燥综合征（pSS 和 sSS）分类中的表现。
该研究队列纳入了国立卫生研究院具有干燥综合征临床可疑症状的患者（N=1303）。参与者根据这两种分类标准被分类为干燥综合征、原发性干燥综合征和继发性干燥综合征。以另一种标准为金标准，比较 2016 年 ACR-EULAR 和 AECG 标准的表现。对两种标准的诊断测试的统计学分析和一致性进行了评估。
根据 AECG 标准，701 人被分类为患有干燥综合征（627 例为原发性干燥综合征，74 例为继发性干燥综合征），714 人被分类为患有干燥综合征（647 例为原发性干燥综合征，67 例为继发性干燥综合征）。两种标准在分类干燥综合征、原发性干燥综合征和继发性干燥综合征时，其敏感性和特异性相当。两种标准在分类干燥综合征（κ=0.79）、原发性干燥综合征（κ=0.81）和继发性干燥综合征（κ=0.87）时具有高度一致性。2016 年 ACR-EULAR 标准对分类继发性干燥综合征的特异性明显高于原发性干燥综合征，而敏感性在两组疾病中相似。然而，AECG 标准也表现出了这种模式。
两种分类标准之间具有高度一致性，且具有相当的诊断性能。新的 2016 年 ACR-EULAR 标准在性能价值方面没有优于 AECG 标准的证据，且两种标准被认为是等效的。我们队列的研究结果表明，2016 年 ACR-EULAR 分类标准可扩展到继发性干燥综合征的分类。
