Li Jian-jun, Li Yan-li, Wang Tian-fang
Department of Diagnostics of Traditional Chinese Medicine, School of Preclinical Medicine, Beijing University of Chinese Medicine, Beijing.
Zhongguo Zhong Xi Yi Jie He Za Zhi. 2011 Dec;31(12):1632-4.
To study the method for establishing a quantization diagnostic standard for immunoglobulin A (IgA) nephropathy of qi-yin deficiency syndrome (QYDS).
1,016 patients with primary IgA nephropathy were recruited in this study. They were randomly assigned to the training sample group (344 cases of QYDS and 456 cases of non-QYDS) and the testing sample group (77 cases of QYDS and 139 cases of non-QYDS) using SPSS software. On the basis of epidemiological survey, the typing standards for QYDS and common clinical symptoms were taken as candidate correlated factors. The correlated factors were selected using binary Logistic stepwise regression. The correlated factors were scored using conditional probability conversion method. The threshold value of the quantization diagnostics was determined using maximum likelihood method. The receiver operating characteristic (ROC) curve was drawn to calculate the area under curve (AUC), sensitivity, specificity, and accuracy rating. The retrospective and prospective tests were performed on the established quantization diagnostic standard for QYDS.
The quantization diagnosis threshold value of IgA nephropathy of QYDS was 12 points. The sensitivity, specificity, and accuracy rating of the quantization diagnostic standard were 75.3%, 68.0%, and 71.1% in the retrospective test, as well as 60.4%, 84.4%, and 69.0% in the prospective tests. The AUC was 0.80 and 0.78 respectively.
It was a feasible method to set up a quantization diagnostic standard for IgA nephropathy of QYDS by taking the occurrence frequency of symptoms. But this method failed to cover the strength information of symptoms.
研究建立气阴两虚证IgA肾病量化诊断标准的方法。
本研究纳入1016例原发性IgA肾病患者，运用SPSS软件将其随机分为训练样本组（气阴两虚证344例，非气阴两虚证456例）和测试样本组（气阴两虚证77例，非气阴两虚证139例）。在流行病学调查的基础上，将气阴两虚证分型标准及常见临床症状作为候选相关因素，采用二元Logistic逐步回归法筛选相关因素，运用条件概率转换法对相关因素进行评分，采用最大似然法确定量化诊断阈值，绘制受试者工作特征（ROC）曲线计算曲线下面积（AUC）、敏感度、特异度及准确率。对建立的气阴两虚证量化诊断标准进行回顾性和前瞻性检验。
气阴两虚证IgA肾病的量化诊断阈值为12分。回顾性检验中量化诊断标准的敏感度、特异度及准确率分别为75.3%、68.0%、71.1%，前瞻性检验中分别为60.4%、84.4%、69.0%。AUC分别为0.80和0.78。
以症状出现频率制定气阴两虚证IgA肾病量化诊断标准是可行的方法，但该方法未涵盖症状的强度信息。
