The relationship between thromboelastography indexes and in-hospital mortality among sepsis patients in ICU

Jiang Zixin; Yang Qilin; Pan Zelin; Chen Jieru; Xiong Xuming; Zhang Zhenhui

doi:10.3969/j.issn.1002-1949.2025.01.003

Abstract：Objective To investigate the correlation between thromboelastography(TEG)indexes and in-hospital mortality among sepsis patients in intensive care unit(ICU).Methods A retrospective observational cohort study was conducted,the clinical data of 185 sepsis patients underwent TEG on admission in Intensive Care Unit of the Second Affiliated Hospital of Guangzhou Medical University between June 2019 and June 2023 were analyzed.According to the hospital outcomes(survival or death),the patients were divided into survival group(n=112)and death group(n=73)and their baseline characteristics and TEG indexes were analyzed.Multivariate Cox regression analysis was employed to assess the relationship between TEG indexes and in-hospital mortality in sepsis patients.The predictive value of TEG indexes for in-hospital mortality rate in sepsis patients was analyzed by using receiver operating characteristic(ROC)curve,and Kaplan-Meier survival curves of in-hospital mortality were drawn based on the optimal cut-off values for grouping.Additionally,sensitivity analysis stratified each categorical variable.Results The study included 185 sepsis patients in the ICU,and found that 73 cases(39.5％)died in hospital.Significant differences were found in various factors between two groups,including the value of reaction time(R value),blood coagulation time(K value),α angle,oxygenation index,total bilirubin(TBIL),lactic acid,activated partial thromboplastin time(APTT),acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ)and sequential organ failure assessment(SOFA)(P＜0.05).Multivariate Cox regression analysis revealed for every 1-minute increase in K value,the in-hospital mortality rate of sepsis patients increased by 21％(HR=1.21,95％CI 1.1-1.33,P＜0.001),while for every 1-mm increase in maximum amplitude(MA)value,the in-hospital mortality rate of sepsis patients decreased by 4％(HR=0.96,95％CI 0.93-0.98,P＜0.001).Additionally,ROC curve analysis demonstrated that the AUC of K value and MA value for predicting in-hospital mortality in sepsis patients was 0.834(95％CI 0.760-0.887)and 0.648(95％CI 0.552-0.719),respectively.With an optimal cut-off value of 2.15 min for K value,the sensitivity was 0.685,the specificity was 0.857,the positive predictive value was 0.758,and the negative predictive value was 0.807.Furthermore,at an optimal cut-off value of 55.45 mm for MA,the sensitivity was 0.397,the specificity was 0.866,the positive predictive value was 0.660,and the negative predictive value was 0.688.The patients were categorized into groups based on the optimal cut-off values of K value and MA value:low K value group(＜2.15 min)and high K value group(≥2.15 min),low MA value group(＜55.45 mm)and high MA value group(≥55.45 mm),respectively.According to Kaplan-Meier survival curves,the cumulative hospitalization survival rates in high K value group were significantly lower than those in low K value group(Log-rank test:P＜0.0001).The cumulative hospitalization survival rates of high MA value group were significantly higher than those of low MA value group(P＜0.0001).Multivariate Cox regression analysis revealed that the high K value group had a 3.56 times higher risk of in-hospital mortality compared to those in the low K value group after adjusting for the variables such as age,sex,main diagnosis,comorbidities,lactic acid,D-dimer,α angle,MA value,APACHEⅡ score and SOFA score(HR=3.56,95％CI 1.83-6.92,P＜0.001).Stratified analysis of categorical variables indicated the relationship of K value and MA value with in-hospital mortality rate remained consistent.Conclusions In the patients with sepsis in ICU,there is a strong correlation of K value and MA value in TEG with in-hospital mortality rate.Sepsis patients with K value exceeding 2.15 min and MA value less than 55.45 mm had a significantly higher risk of mortality.