D-二聚体对急诊脓毒症患者预后价值的前瞻性研究

Prognostic value of D-dimer in patients with sepsis in emergency department： a prospective study

目的探讨D-二聚体浓度能否预测急诊脓毒症患者28d病死率，以及D-二聚体浓度对于脓毒症患者预后的意义。方法采用前瞻性研究方法，选取2009年10月至2010年7月首都医科大学附属北京朝阳医院急诊科766例脓毒症患者，检测患者人院时的血中D-二聚体浓度（固相免疫层析法），并记录各项实验室检测指标和临床生理指标。以28d为终点比较存活组与死亡组患者的D-二聚体水平、急性生理学与慢性健康状况评分系统Ⅱ（APACHEII）评分和简化急性生理学评分系统II（SAPSⅡ）评分；分析D-二聚体水平与28d病死率的关系。结果766例脓毒症患者28d内共有233例死亡，死亡组血中D-二聚体浓度（μg/L）明显高于存活组[1220．0（789．0，1835．0）比323．0（158．0，642．0），P〈0．01]。D-二聚体水平与APACHEII评分和SAPS1I评分均呈显著正相关（r1=0．643，r2=0．632，均P〈0．01）。D-二聚体水平预测28d病死率的受试者工作特征曲线（ROC曲线）下面积为0．880，95％可信区间（95％CI）为0．855～0．904，P〈0．001，预测28d病死率的最佳界值为663．5μg/L。血中D-二聚体〉663．5μg/L、APACHEⅡ评分〉19．5分和SAPSII评分〉39．5分是预测患者28d病死率的独立危险因素[优势比（OR值）分别为17．5、15．7、19．6，均P〈0．001]。D-二聚体〈250．0、250．0～663．5、〉663．5μg/L3组间28d病死率（1．4％、12．3％、64．2％）、APACHE1I评分[分：11（9，13）、13（11，16）、19（15，22）]、SAPSⅡ评分[分：24（18，29）、31（24，36）、40（33，49）]比较差异均有统计学意义（均P〈0．01）。结论D-二聚体水平是预测急诊脓毒症患者28d病死率的独立指标。其预测能力与APACHE1I评分和SAPSII评分十分相近；检测D-二聚体水平有助于脓毒症患者的危险分层。

Objective To investigate the value of the concentration of D-dimer in predicting 28-day mortality of sepsis patients in emergency department in order to determine its prognostic values. Methods A prospective study was conducted. A total of 766 sepsis patients treated in the emergency department of Beijing Chaoyang Hospital of the Capital Medical University from October 2009 to July 2010 were enrolled, and admission blood samples were obtained for D-dimer measurement （solid phase immune chromatography）. A variety of clinical and laboratory variables were recorded. With 28 days as end point, the D-direct levels, acute physiology and chronic health evaluation Ⅱ（APACHEⅡ ） score, and simplify acute physiology score [[ （SAPS ]I ） were compared respectively between survivors and non-survivors. The status of each patient was ascertained, and the association between the concentration of D-dimer and 28-day mortality was assessed. Results Two hundred and thirty-three patients died during the 28-day observation period among 766 patients. D-dimer concentration （μg/L） among deceased were significantly higher than those of survivors [1220.0 （789.0, 1835.0） vs. 323.0 （158.0, 642.0）, P〈0.01]. The level of D-dimer showed positive correlation with APACHE Ⅱscore and SAPS Ⅱ （r1=0.643, r2=0.632, both P〈0.01 ）. D-dimer concentration had an area under the receiver operating characteristic （ROC） curve （AUC） of 0.880 for predicting 28-day mortality, with 95% confidence interval （95% CI） 0.855-0.904, P〈0.001; the optimal D-dimer cut point for predicting 28-day mortality was 663.5 μg/L. D-dimer concentration higher than 663.5 μg/L, APACHE Ⅱ score higher than 19.5 and SAPSⅡ higher than 39.5 were the independent predictors of 28-day mortality [odds ratio （OR）： 17.5, 15.7, 19.6, all P〈0.001].The 28-day mortality （1.4%, 12.3%, 64.2%）, APACHEⅡscore [11 （9, 13）, 13 （11, 16）, 19 （15, 22）] and SAPS ]I [24 （18, 29）, 31 （24, 36）, 40 （33, 49）] showed statistically significant among D-dimer 〈 250.0, 250.0-663.5, 〉663.5 μg/L three groups （all P 〈0.01 ）. Conclusions D-dimer concentration is an independent predictor of 28-day mortality in patients with sepsis. D-dimer level shows a high predictive value in patients with sepsis, similar to APACHE Ⅱscore and SAPS Ⅱ. It can be helpful in risk stratification in septic patients.