含血常规和凝血四项指标的风险模型对恶性血液病继发脓毒症的预测价值及临床效用

Predictive value and clinical utility of a risk model containing blood routine and four items of blood coagulation for sepsis secondary to hematological malignancies

目的构建含血常规和凝血四项指标预测恶性血液病继发脓毒症的风险模型,并分析其预测价值及临床效用,为临床工作提供有利参考。方法回顾性选取2020年6月至2023年10月平顶山市第一人民医院收治的326例恶性血液病患者的临床资料,根据是否继发脓毒症分为继发组150例和未继发组176例,比较两组患者的临床资料、血常规和凝血四项指标;采用Logistic回归分析恶性血液病继发脓毒症的影响因素,并构建风险模型;绘制受试者工作特征(ROC)曲线分析含与不含血常规和凝血四项指标的风险模型的预测价值,采用"rmda"软件包绘制决策(DCA)曲线分析含与不含血常规和凝血四项指标的风险模型对恶性血液病继发脓毒症的预测价值及临床效用。结果继发组患者的机械通气占比、总胆红素、肌酐、血清白介素-6(IL-6)、中性粒细胞表面分子CD64(CD64)、超敏C-反应蛋白(hs-CRP)水平明显高于未继发组,差异均有统计学意义(P<0.05);继发组患者的红细胞计数(RBC)、血红蛋白(Hb)、纤维蛋白原(FIB)水平分别为(88.06±18.40)×10^(9)/L、(2.82±0.83)g/L、(2.15±0.28)g/L,明显低于未继发组的(117.52±23.81)×10^(9)/L、(3.96±0.88)g/L、(3.73±0.35)g/L,差异均有统计学意义(P<0.05);继发组患者的PT、APTT、TT水平分别为(15.24±2.52)s、(41.47±9.42)s、(21.51±2.47)s,明显高于未继发组的(11.64±1.43)s、(28.84±7.81)s、(12.43±1.16)s,差异均有统计学意义(P<0.05);经Logistic回归分析结果显示,RBC、Hb、PT、APTT、TT、FIB、机械通气、IL-6、CD64、hs-CRP均为恶性血液病继发脓毒症的危险因素(P<0.05);经ROC分析结果显示,含与不含血常规和凝血四项指标的风险模型的AUC值分别为0.948(95%CI:0.911~0.973)、0.869(95%CI:0.818~0.910);经DCA曲线分析结果显示,与不含血常规和凝血四项指标曲线相比,含血常规和凝血四项指标的复合风险模型具有较高的净获益率。结论恶性血液病患者RBC、Hb、PT、APTT、TT、FIB与继发脓毒症有关,建立含上述血常规和凝血四项指标的风险模型对继发脓毒症有较高预测价值及临床效用。

Objective To construct a risk model containing blood routine and four items of blood coagulation for predicting sepsis secondary to hematological malignancies,and to analyze its predictive value and clinical utility,so as to provide a favorable reference for clinical work.Methods The clinical data of 326 patients with hematological malignancies admitted to Pingdingshan First People's Hospital from June 2020 to October 2023 were retrospectively selected.According to whether secondary sepsis occurred,they were divided into secondary group(150 cases)and non-secondary group(176 cases).The clinical data,blood routine,and four items of blood coagulation were compared between the two groups.Logistic regression was used to analyze the influencing factors of sepsis secondary to hematological malignancies,and a risk model was constructed.The receiver operating characteristic(ROC)curve was drawn to analyze the predictive value of the risk model with and without blood routine and four items of blood coagulation.The'rmda'software package was used for decision curve analysis(DCA)to analyze the predictive value and clinical utility of the risk model with and without blood routine and four items of blood coagulation for sepsis secondary to hematological malignancies.Results The proportion of mechanical ventilation,total bilirubin,creatinine,serum interleukin-6(IL-6),neutrophil surface molecule CD64(CD64),and high-sensitivity C-reactive protein(hs-CRP)levels in the secondary group were significantly higher than those in the non-secondary group(P<0.05).The levels of red blood cell count(RBC),hemoglobin(Hb),and fibrinogen(FIB)in the secondary group were(88.06±18.40)×10^(9)/L,(2.82±0.83)g/L,and(2.15±0.28)g/L,respectively,which were significantly lower than(117.52±23.81)×10^(9)/L,(3.96±0.88)g/L,and(3.73±0.35)g/L in the non-secondary group(P<0.05).The levels of PT,APTT,and TT in the secondary group were(15.24±2.52)s,(41.47±9.42)s,and(21.51±2.47)s,respectively,which were significantly higher than(11.64±1.43)s,(28.84±7.81)s,and(12.43±1.16)s in the non-secondary group(P<0.05).The results of logistic regression analysis showed that RBC,Hb,PT,APTT,TT,FIB,mechanical ventilation,IL-6,CD64,and hs-CRP were all risk factors for secondary sepsis in hematologic malignancies(P<0.05).The results of ROC analysis showed that the AUC values of the risk models with and without blood routine and coagulation indicators were 0.948(95%CI:0.911-0.973)and 0.869(95%CI:0.818-0.910),respectively.The results of DCA showed that the risk model with blood routine and four items of blood coagulation had a higher net benefit rate compared with that without blood routine and four items of blood coagulation.Conclusion RBC,Hb,PT,APTT,TT and FIB in patients with hematological malignancies are related to secondary sepsis.The establishment of a risk model containing blood routine and four items of blood coagulation has high predictive value and clinical utility for secondary sepsis.