本研究旨在对3种检测粪便样本中艰难梭菌的方法进行临床应用评估，为艰难梭菌的实验室检测提供参考。采用艰难梭菌毒素A/B(Clostridium difficile toxins A and B，CDAB)酶联免疫荧光检测法、环丝氨酸­-头孢西丁­-果糖琼脂(cycloserin­-cefoxitin-­fructose agar，CCFA)常规培养法和显色培养法(chromID™)同步检测粪便样本中的艰难梭菌，并对培养所得菌株进行tcdB基因扩增以验证其产毒性。以艰难梭菌培养联合tcdB基因扩增为参考方法，分别计算上述3种方法的灵敏度、特异度、阳性预测值和阴性预测值等参数，分析其与参考方法的一致性。研究共收集临床粪便样本164份，参考方法检测结果为58份阳性，106份阴性。经统计分析，艰难梭菌毒素A/B法的灵敏度、特异度、阳性预测值和阴性预测值分别为51.7%、95.3%、85.7%和78.3%，CCFA常规培养法分别为72.4%、98.1%、95.5%和86.7%，而艰难梭菌显色培养法分别为94.8%、92.5%、87.3%和97.0%。3种方法中，艰难梭菌显色培养法与参考方法的一致性最高(Kappa＝0.856)。采用该方法检测粪便样本中的艰难梭菌操作简便，成本经济，结果判断直观、准确，具有较好的临床应用价值。

This study aims to evaluate the clinical performance of three methods in the detection of Clostridium difficile (C. difficile) in stool samples. Clinical samples were subjected to enzyme­linked fluorescence assay for C. difficile toxins A and B (CDAB), conventional bacterial cultivation using cycloserine­-cefoxitin­-fructose agar (CCFA) and cultivation by chromogenic C. difficile identification medium (chromID™), respectively. A combination of bacterial culture and amplification of the tcdB gene was selected as the standard reference method for the evaluation. A total of 164 clinical stool samples were collected, of which 58 had positive results by the reference method whilst the rest 106 were negative. The CDAB method showed a sensitivity, specificity, positive predictive value and negative predictive value of 51.7%, 95.3%, 85.7% and 78.3%, respectively, while the corresponding values of the CCFA method turned out to be 72.4%, 98.1%, 95.5% and 86.7%, respectively. The chromID™ method gave corresponding values of 94.8%, 92.5%, 87.3% and 97.0%, respectively. Among the three methods, chromID™ method had the best consistency with the reference method (Kappa＝0.856). It is concluded that chromID™ is a simple and cost­effective method for the detection of C. difficile in stool samples, which can present easy­to­judge results and has a preferable value in clinical application.