目的探讨不同病因发热患者血清降钙素原（PCT）的诊断截断值（cut-off value）。方法回顾性分析2006年4月至2012年10月以“发热待查”收治，年龄16岁以上，检测PCT前2 d内未用过抗生素且最终病因明确的细菌或病毒感染住院病例，应用法国生物梅里埃公司miniVIDAS免疫分析仪及配套试剂采用酶联荧光分析技术检测患者血清PCT水平，应用ROC曲线判断细菌感染和病毒感染患者PCT水平的最佳cut-off值。结果符合入选标准共150例，其中病毒感染性发热104例和细菌感染性发热46例；对细菌感染性发热和病毒感染性发热患者确立的PCT最佳cut-off值为0.5750 ng/ml，AUC为0.844，其敏感性、特异性、阳性预测值、阴性预测值和总预测准确率分别为67.4％、97.1％、85.71％、87.07％和79.33％；对全身性细菌感染与局部细菌感染患者确立的PCT最佳cut-off值为2.1450 ng/ml，其AUC为0.739，其敏感性、特异性、阳性预测值、阴性预测值和总预测准确率分别为76.9％、69.7％、50％、88.46％和71.74％。结论血清PCT检测对诊断细菌感染性发热及全身性细菌感染具有较好诊断价值，分别以0.5750 ng/ml及2.1450 ng/ml作为cut-off值具有较好的敏感性和特异性，可提高发热患者的鉴别诊断效率。
Objective To explore the value of serum procalcitonin ( PCT ) in different causes of fever patients.Methods We collected the cases with final diagnosis with bacterial or viral infections from April 2006 to October 2012,whose ages were no less than 16 years,and the PCT level was tested without antibiotics treatment .The serum PCT was determined by Enzyme linked fluorescence analysis ( ELFA) using miniVIDAS.The optimal cut-off values of PCT were established by using receiver operator characteristic curve (ROC curve).Results 150 cases were retrospectively enrolled in this study ,in which there were 46 cases with bacterial infection and 104 cases with virus infection.The ROC analysis showed that the optimal cut-off value was 0.5750 ng/ml between bacterial infection and viral infection.The area under ROC curve was 0.844, and the specificity, sensitivity, positive predictive value , negative predictive value and the total accuracy were 67.4%,97.1%,85.71%,87.07%and 79.33%,respectively . For systemic bacteria infection and local bacteria infection ,the ROC analysis showed the optimal cut-off value was 2.1450 ng/ml.The area under ROC curve was 0.739.The specificity,sensitivity,positive predictive value,negative predictive value and the total accuracy were 76.9%, 69.7%, 50%, 88.46% and 71.74%, respectively . Conclusions Serum PCT could serve as a good risk predictor for bacterial infection and systemic bacterial infection . The cut-off value was 0.5750 and 2.1450,respectively,which had the highest diagnostic efficiency with specificity and sensitivity .It might be helpful to differential diagnosis of fever .