早期感染性休克相关性血小板减少症的危险因素及对预后的影响

Risk factors of early septic shock-related thrombocytopenia and its impact on prognosis

目的探讨早期感染性休克相性血小板减少症的危险因素及与预后的关系。方法回顾性分析2016年6月至2020年11月江苏省苏北人民医院重症监护病房(ICU)感染性休克患者的临床资料。根据感染性休克早期(使用血管活性药物24 h内)的最低血小板计数(PLT)将患者分为轻度血小板减少组〔PLT(50~100)×10^(9)/L〕、重度血小板减少组(PLT<50×10^(9)/L)和血小板正常组(PLT>100×10^(9)/L)。分析3组患者间一般资料、实验室指标、机械通气时间、ICU住院时间、总住院时间及28 d病死率的差异;用多因素Logistic回归分析血小板减少症的影响因素,并绘制不同PLT水平患者28 d Kaplan-Meier生存曲线。结果共纳入486例感染性休克患者,其中轻度血小板减少组123例,重度血小板减少组75例,血小板正常组288例。合并糖尿病(χ^(2)=30.460、P<0.001)、腹腔感染(χ^(2)=15.024、P=0.001)、尿路感染(χ^(2)=36.633、P<0.001)、血流感染(χ^(2)=7.755、P=0.022)、革兰阴性(G-)杆菌感染(χ^(2)=19.569、P<0.001)、高乳酸血症(H=23.404、P<0.001)、降钙素原(PCT)升高(H=43.368、P<0.001)、高急性生理学与慢性健康状况评分Ⅱ(APACHEⅡ,F=11.122、P<0.001)、高序贯器官衰竭评分(SOFA,F=84.328、P<0.001)的患者在感染性休克24 h内更容易合并血小板减少症。早期感染性休克相关性血小板减少症的多因素Logistic回归分析显示,糖尿病〔优势比(OR)=0.19,95%可信区间(95%CI)为0.08~0.42,P<0.001〕、尿路感染(OR=0.33,95%CI为0.13~0.83,P=0.018)、G-杆菌感染(OR=0.20,95%CI为0.07~0.58,P=0.003)、高乳酸血症(OR=1.25,95%CI为1.07~1.46,P=0.005)和高APACHEⅡ评分(OR=0.85,95%CI为0.78~0.92,P<0.001)是PLT<50×10^(9)/L的独立危险因素;腹腔感染是PLT(50~100)×10^(9)/L的独立危险因素(OR=0.56,95%CI为0.34~0.95,P=0.03);高SOFA评分是PLT≤100×10^(9)/L的独立危险因素〔PLT<50×10^(9)/L:OR=2.03,95%CI为1.65~2.52,P<0.001;PLT(50~100)×10^(9)/L:OR=1.31,95%CI为1.16~1.48,P<0.001〕。而3组间机械通气时间、ICU住院时间、总住院时间差异均无统计学意义(H值分别为0.142、2.134、3.990,均P>0.05)。感染性休克患者28 d病死率随血小板减少症的严重程度增加而升高(χ^(2)=40.406,P<0.001);其中重度血小板减少组和轻度血小板减少组28 d病死率显著高于血小板正常组〔66.7%(50/75)、43.1%(53/123)比27.8%(80/288),均P<0.05〕。Kaplan-Meier生存曲线分析显示,28 d生存率随PLT降低逐步下降,血小板正常组28 d生存率更高(Log-Rank检验:χ^(2)=80.667,P<0.001)。结论糖尿病、腹腔感染、尿路感染、G-杆菌感染、高乳酸血症、高APACHEⅡ评分和高SOFA评分是发生早期感染性休克相关血小板减少症的独立危险因素;感染性休克患者早期出现PLT降低预示28 d死亡风险增加。

Objective To investigate the risk factors and prognosis of early septic shock-related thrombocytopenia.Methods Retrospective analysis of clinical data of patients with septic shock admitted to the department of intensive care unit(ICU)of Northern Jiangsu People's Hospital from June 2016 to November 2020 was conducted.According to the lowest platelet count(PLT)in the early stage of septic shock(within 24 hours of using vasoactive drugs),the patients were divided into mild thrombocytopenia group[PLT(50-100)×10^(9)/L],severe thrombocytopenia group(PLT<50×10^(9)/L)and normal platelet group(PLT>100×10^(9)/L).The differences in general information,laboratory indicators,mechanical ventilation time,length of ICU stay,in-hospital stay,and 28-day mortality among the three groups were analyzed.Multivariate Logistic regression was used to analyze the influencing factors of thrombocytopenia,and the 28-day Kaplan-Meier survival curve of patients with different PLT levels was drawn.Results A total of 486 patients with septic shock were enrolled,including 123 patients with mild thrombocytopenia,75 patients with severe thrombocytopenia and 288 patients with normal platelets.Patients with diabetes(χ^(2)=30.460,P<0.001),abdominal infection(χ^(2)=15.024,P=0.001),urinary tract infection(χ^(2)=36.633,P<0.001),bloodstream infection(χ^(2)=7.755,P=0.022),Gram negative(G-)bacilli infection(χ^(2)=19.569,P<0.001),hyperlactic acidemia(H=23.404,P<0.001),elevated procalcitonin(PCT,H=43.368,P<0.001),high acute physiology and chronic health evaluationⅡ(APACHEⅡ,F=11.122,P<0.001)and high sequential organ failure assessment(SOFA,F=84.328,P<0.001)were more likely to have thrombocytopenia within 24 hours of septic shock.Multivariate Logistic regression analysis of early septic shock-related thrombocytopenia showed that,diabetes[odds ratio(OR)=0.19,95%confidence interval(95%CI)was 0.08-0.42,P<0.001],urinary tract infection(OR=0.33,95%CI was 0.13-0.83,P=0.018),G-bacilli infection(OR=0.20,95%CI was 0.07-0.58,P=0.003),hyperlactic acidemia(OR=1.25,95%CI was 1.07-1.46,P=0.005)and high APACHEⅡscore(OR=0.85,95%CI was 0.78-0.92,P<0.001)were independent risk factors for platelets<50×10^(9)/L.Abdominal infection was an independent risk factor for PLT(50-100)×10^(9)/L(OR=0.56,95%CI was 0.34-0.95,P=0.03).High SOFA score was an independent risk factor for PLT≤100×10^(9)/L[PLT<50×10^(9)/L:OR=2.03,95%CI was 1.65-2.52,P<0.001;PLT(50-100)×10^(9)/L:OR=1.31,95%CI was 1.16-1.48,P<0.001].There were no significant differences in mechanical ventilation time,length of ICU stay,and in-hospital stay among the three groups(H values were 0.142,2.134,and 3.990,respectively,all P>0.05).The 28-day mortality of septic shock patients increased with the severity of thrombocytopenia(χ^(2)=40.406,P<0.001),and the 28-day mortality of severe thrombocytopenia group and mild thrombocytopenia group was significantly higher than those of the normal platelet group[66.7%(50/75),43.1%(53/123)vs.27.8%(80/288),both P<0.05].Kaplan-Meier survival curve analysis showed that the 28-day survival rate gradually decreased with the decrease of PLT,and the 28-day survival rate was higher in the normal platelet group(Log-Rank test:χ^(2)=80.667,P<0.001).Conclusions Diabetes,abdominal infection,urinary tract infection,G-bacilli infection,hyperlactic acidemia,high APACHEⅡscore,and high SOFA score are independent risk factors for early septic shock-related thrombocytopenia.Early thrombocytopenia in patients with septic shock indicates a high risk of 28-day death.