凝血功能联合APACHEⅡ及SOFA评分对血流感染患者预后的预测价值

Prognostic value of coagulation function combined with acute physiology and chronic health evaluationⅡand sequential organ failure assessment scores for patients with bloodstream infection

目的探讨血小板计数(PLT)、凝血4项、急性生理学与慢性健康状况评分Ⅱ(APACHEⅡ)和序贯器官衰竭评分(SOFA)对血流感染患者预后的评估价值。方法采用回顾性方法,选择2016年1月至2020年10月在宁夏医科大学总医院重症监护病房(ICU)住院治疗至少1次血培养阳性的血流感染患者作为研究对象,收集患者的基础资料、病原菌分布、凝血功能和28 d预后等指标,根据血培养24 h内的实验室检查结果计算24 h内APACHEⅡ评分、SOFA评分。按28 d预后将患者分为存活组和死亡组,比较两组患者上述指标的差异。采用多因素Logistic回归分析筛选出影响血流感染患者28 d死亡的危险因素;绘制受试者工作特征曲线(ROC曲线),分析各危险因素对血流感染患者28 d预后的预测效能。结果共纳入215例血流感染患者,28 d存活117例,死亡98例。检出病原菌256株,其中革兰阴性菌(G^(-)菌)161株(占62.89%),革兰阳性菌(G^(+)菌)76株(占29.69%),真菌17株(占6.64%),其他2株(占0.78%);主要病原菌依次为大肠埃希菌(53株,20.70%)、肠球菌(37株,14.45%)、肺炎克雷伯菌(34株,13.28%)。与存活组比较,死亡组患者年龄更大(岁:60.98±16.08比55.64±16.35),体温、SOFA评分、APACHEⅡ评分、合并恶性肿瘤和肺部感染的比例及血乳酸(Lac)、肌酐(Cr)水平均更高〔体温(℃):39.12±1.10比38.67±1.09,SOFA评分(分):13.05±4.40比7.85±3.74,APACHEⅡ评分(分):24.01±8.18比15.38±6.59,合并恶性肿瘤比例:15.31%(15/98)比12.82%(15/117),合并肺部感染比例:84.69%(83/98)比72.65%(85/117),Lac(mmol/L):7.13±6.04比4.31±2.98,Cr(μmol/L):189.73±141.81比124.55±106.17,均P<0.05〕,凝血酶原时间(PT)、活化部分凝血活酶时间(APTT)、凝血酶时间(TT)均明显延长〔PT(s):19.51±15.16比14.94±2.86,APTT(s):52.74±26.82比40.77±15.30,TT(s):21.59±18.46比17.38±2.59,均P<0.05〕,PLT明显降低〔×10^(9)/L:43.50(18.75,92.75)比86.00(36.00,154.50),P<0.05〕。Logistic回归分析结果显示,体温、年龄、SOFA评分、APACHEⅡ评分是影响血流感染患者预后的独立危险因素〔优势比(OR)分别为1.388、1.023、0.817、0.916,95%可信区间(95%CI)分别为1.001~1.926、1.001~1.046、0.730~0.913、0.867~0.968,P值分别为0.046、0.043、0.000、0.002〕。ROC曲线分析显示,SOFA评分、APACHEⅡ评分、体温、年龄对血流感染患者预后均有一定的预测价值,ROC曲线下面积(AUC)分别为0.815、0.795、0.625、0.594(均P<0.05);且4者联合预测的AUC高达0.851,特异度为79.3%,敏感度为74.2%,提示联合变量预测血流感染患者死亡的准确性更高。结论PLT、凝血4项有助于评估ICU血流感染患者预后,APACHEⅡ评分和SOFA评分与血流感染患者预后直接相关。

Objective To evaluate the prognostic value of platelet count(PLT),coagulation indexes,acute physiology and chronic health evaluationⅡ(APACHEⅡ),and sequential organ failure assessment(SOFA)in patients with bloodstream infection.Methods A retrospective single center cohort study was conducted,patients with at least one positive blood culture bloodstream infection hospitalized in the intensive care unit(ICU)of Ningxia Medical University General Hospital from January 2016 to October 2020 were selected as the research objects,basic data and pathogen distribution,coagulation function,and prognosis at 28 days were collected,the APACHEⅡscore,SOFA score based on the results of laboratory examination within 24 hours of blood culture were calculated.Patients were divided into the survival group and the death group according to the 28-day prognosis,and the differences of the above indicators were compared.Multivariate Logistic regression analysis was used to screen out the risk factors for 28-day death of patients with bloodstream infection.Receiver operator characteristic curve(ROC curve)was drawn to analyze the predictive efficacy of various risk factors for 28-day prognosis of patients with bloodstream infection.Results A total of 215 patients with bloodstream infection were enrolled,of which 117 survived and 98 died within 28 days.The 256 strains of pathogenic bacteria were detected,including 161(62.89%)Gram-negative bacteria(G^(-)),76(29.69%)Gram-positive bacteria(G^(+)),17 fungi(6.64%),and 2 other strains(0.78%).The main pathogenic bacteria were Escherichia coli(53 strains,20.70%),Enterococcus(37 strains,14.45%),and Klebsiella pneumoniae(34 strains,13.28%).Compared with the survival group,patients in the death group were older(years old:60.98±16.08 vs.55.64±16.35),had higher levels of body temperature,SOFA score,APACHEⅡscore,proportion of malignant tumor and pulmonary infection,blood lactic acid(Lac),and creatinine[Cr;body temperature(℃):39.12±1.10 vs.38.67±1.09,SOFA score:13.05±4.40 vs.7.85±3.74,APACHEⅡscore:24.01±8.18 vs.15.38±6.59,proportion of malignant tumor:15.31%(15/98)vs.12.82%(15/117),proportion of patients with pulmonary infection:84.69%(83/98)vs.72.65%(85/117),Lac(mmol/L):7.13±6.04 vs.4.31±2.98,Cr(μmol/L):189.73±141.81 vs.124.55±106.17,all P<0.05].The prothrombin time(PT),activated partial thrombin time(APTT),and thrombin time(TT)were significantly longer[PT(s):19.51±15.16 vs.14.94±2.86,APTT(s):52.74±26.82 vs.40.77±15.30,TT(s):21.59±18.46 vs.17.38±2.59,all P<0.05],PLT was significantly decreased[×10^(9)/L:43.50(18.75,92.75)vs.86.00(36.00,154.50),P<0.05].Logistic regression analysis showed that body temperature,age,SOFA score and APACHEⅡscore were independent risk factors[odds ratio(OR)were 1.388,1.023,0.817 and 0.916,respectively,95%confidence intervals(95%CI)were 1.001-1.926,1.001-1.046,0.730-0.913,0.867-0.968,with respective P values of 0.046,0.043,0.000,0.002].ROC curve analysis showed that SOFA score,APACHEⅡscore,temperature,age had certain predictive values for the prognosis of patients with bloodstream infection,and area under ROC curve(AUC)was 0.815,0.795,0.625 and 0.594,respectively(all P<0.05).The AUC predicted by the combination of the 4 variables was as high as 0.851,the specificity was 79.3%,and the sensitivity was 74.2%,suggesting that the combination variables could predict the death of patients with bloodstream infection with higher accuracy.Conclusions PLT and coagulation indexes are helpful to evaluate the prognosis of patients with bloodstream infection in ICU.APACHEⅡscore and SOFA score are directly related to the prognosis of patients with bloodstream infection.