Abstract: Background　With the enhancement of diagnosis consciousness of clinicians in pulmonary embolism（PE）,many suspected patients accepted the radiological examination.The latest guidelines for diagnosis and treatment of PE proposed that low or intermediate potential PE patients with D-dimer level below the cut-off point could be basically ruled out PE.Infections could increase the level of D-dimer,therefore the cut-off point of D-dimer need be adjusted in infection patients.Objective　To investigate the optimal cut-off point of D-dimer for excluding PE,in order to reduce unnecessary radiological examinations.Methods　All hospitalized patients with low or intermediate potential PE who presented to the Department of Respiratory of the First Affiliated Hospital of Xinjiang Medical University from October 2013 to January 2015 were enrolled.All patients were assessed by Wells score and underwent multi-slice spiral CT pulmonary angiography（CTPA） examination.The plasma levels of D-dimer and PCT were measured.According to the clinical manifestations and PCT level,all patients were divided into infection group and non-infection group.According to the results of CTPA to confirm the diagnosis of PE.ROC curve of D-dimer levels in the diagnosis of PE were drawed.The optimal cut-off point was selected at the maximal Youden’s index.The sensitivity,specificity,positive predictive value and negative predictive value were calculated.Results　A total of 245 patients with low or intermediate potential PE were enrolled in the study.According to the clinical manifestations and PCT level,32 patients could not be grouped and excluded from the study.213 patients were included in the study finally,including 110 cases of infection group and 103 cases of non-infected group.There were no statistically significant differences in gender,age,Wells score and detection rate of PE between infection group and non-infection group（P>0.05）.There were statistically significant differences in D-dimer and PCT levels between the two groups（P<0.05）.There were no statistically significant differences in Wells score and PCT level between PE and non-PE patients in infected and non-infected groups（P>0.05）.There were statistically significant difference in the level of D-dimer between PE and non-PE patients in the infected and non-infected groups（P<0.05）.The area under the ROC curve of the level of D-dimer in the diagnosis of PE was 0.679 in the infection group and 0.705 in the non-infection group.There was no statistically significant difference between the two groups（Z=0.34,P=0.37）.The optimal cut-off value of D-dimer level in the diagnosis of PE was 1 100 μg/L in the infection group.The sensitivity was 47.7%,the specificity was 89.4%,the positive predictive value was 47.7%,the negative predictive value was 87.9%.The optimal cut-off value of D-dimer level in the diagnosis of PE was 140 μg/L in the non-infection group.The sensitivity was 86.8%,the specificity was 47.7%,the positive predictive value was 86.8%,the negative predictive value was 46.2%.ConclusionFor the infection patients with low or intermediate potential PE,1 100 μg/L as the optimal cut-off point for D-dimer is more specific.More patients would avoid unnecessary radiological examinations and the medical resources would be saved.

Key words: Pulmonary embolism, Infection, D-dimer, Sensitivity, Specificity

摘要： 背景　随着临床医生对肺栓塞诊断意识的增强，许多疑似患者进行放射性检查。最新肺栓塞诊疗指南提出中低度可能的疑似患者，D-二聚体水平低于临界值可基本排除肺栓塞。感染可引起D-二聚体水平升高，因此合并感染时，D-二聚体临界值需要修正。目的　探讨合并感染时D-二聚体排除肺栓塞的最佳临界值，减少不必要的放射性检查。方法　收集2013年10月—2015年1月新疆医科大学第一附属医院呼吸科就诊的中低度可能的疑似肺栓塞住院患者。患者均评定Wells评分，测定血浆D-二聚体、降钙素原（PCT）水平，行多层螺旋CT肺动脉成像（CTPA）检查。根据临床表现及PCT水平将入选患者分为感染组与非感染组，根据CTPA检查结果明确诊断肺栓塞。绘制D-二聚体水平诊断肺栓塞的ROC曲线，选择Youden’s指数最大的切点作为最佳临界值，计算其灵敏度、特异度、阳性预测值、阴性预测值。结果　共纳入中低度可能的疑似肺栓塞患者245例，剔除根据临床表现及PCT水平不能明确分组者32例，最终213例患者纳入研究，其中感染组110例，非感染组103例。感染组与非感染组患者性别、年龄、Wells评分及肺栓塞检出率比较，差异无统计学意义（P>0.05）；两组患者D-二聚体、PCT水平比较，差异有统计学意义（P<0.05）。感染组和非感染组肺栓塞与非肺栓塞患者Wells评分、PCT水平比较，差异无统计学意义（P>0.05）；感染组和非感染组肺栓塞与非肺栓塞患者D-二聚体水平比较，差异有统计学意义（P<0.05）。感染组D-二聚体水平诊断肺栓塞的ROC曲线下面积为0.679，非感染组为0.705，两组比较，差异无统计学意义（Z=0.34，P=0.37）。感染组D-二聚体水平诊断肺栓塞的最佳临界值为1 100 μg/L，灵敏度为47.7%，特异度为89.4%，阳性预测值为47.7%，阴性预测值为87.9%；非感染组D-二聚体水平诊断肺栓塞的最佳临界值为140 μg/L，灵敏度为86.8%，特异度为47.7%，阳性预测值为86.8%，阴性预测值为46.2%。结论　对于合并感染的中低度可能的疑似肺栓塞患者，取1 100 μg/L作为 D-二聚体的最佳临界值特异度较高，可为更多患者减少不必要的放射性检查，节约医疗资源。

关键词: 肺栓塞, 感染, D-二聚体, 灵敏度, 特异度