不基于甲状腺核素扫描的SAT诊断评分量表的建立

Establishment of diagnostic scale for subacute thyroiditis not based on radioisotope scanning

目的建立一个不基于甲状腺核素扫描的亚急性甲状腺炎(SAT)诊断评分量表。方法疑诊SAT患者309例,180例最终确诊为SAT(SAT组),129例为非SAT(对照组)。收集两组相关临床指标,采用二元logistic回归分析建立SAT诊断评分量表,绘制ROC曲线获取最佳截断值并验证量表的有效性。结果经单因素分析,以两组间有统计学差异的17个变量为自变量,以诊断SAT为因变量进行二分类logistic回归分析得到甲状腺压痛、甲状腺质地坚硬、血沉加快和甲状腺激素水平升高4个独立预测因子,回归系数分别为5.801、2.586、2.520和2.073。以上述4个变量建立SAT诊断评分量表,ROC曲线下面积为0.991(95%CI=0.982~0.999),确立7分作为最佳截断值,诊断SAT的灵敏度为92.78%,特异度为98.45%。结论首次建立了不基于甲状腺核素扫描的SAT诊断评分量表,具有较高的灵敏度和特异度。

Objective To establish a diagnostic scale for subacute thyroiditis(SAT)not based on radioisotope scanning.Methods According to final diagnosis,309 patients suspected to be SAT were divided into two groups of A(with SAT,180 cases)and B(without SAT,129 cases).Clinical indicators of two groups were collected.Diagnostic scale for SAT was established through binary logistic regression analysis.The ROC curve was drawn to obtain the cut-off value and verify the validity of the scale.Results Eighteen differences between two the groups(P<0.05)were treated as the independent variables through single factor analysis,coming with diagnosis of SAT as dependent variable for binary logistic regression analysis to get four independent predictors of thyroid tenderness,hard thyroid texture,rapid erythrocyte sedimentation rate and elevated thyroid hormone level.The regression coefficients were 5.801,2.586,2.520 and 2.073,respectively.The above four variables were used to establish the SAT diagnostic scoring scale.The area under the ROC curve was 0.991(95%CI=0.982~0.999).Taking 7 points as the best cut-off value for the diagnosis of SAT,the sensitivity was 92.78%and the specificity was 98.45%.Conclusion A SAT diagnostic scoring scale not based on radioisotope scanning is established with higher sensitivity and specificity for the diagnosis of SAT.