Analysis of Influencing Factors of MACE in Adult Patients with First-Time Sepsis During Hospitalization
ZHAO Yifan
ZHANG Zhongqi
DING Lianshu
ZHANG Xiangyu
CAO Quan
Abstract：Objective To analyze the factors influencing the occurrence of major cardiovascular adverse events(MACE)in sepsis patients from the MIMIC-IV and eICU databases.Methods The study cohort included all adult patients with sepsis in the MIMIC-IV and eICU databases.Based on the occurrence of MACE during hospitalization,patients were divided into a MACE group(n=847)and a non-MACE group(n=7 116).After data extraction,demographic variables such as age,sex,and comorbidities(e.g.,hypertension,heart failure,diabetes,chronic obstructive pulmonary disease,chronic kidney disease,liver dysfunction),laboratory results,and Sequential Organ Failure Assessment(SOFA)scores at ICU admission were collected.Multivariate Cox regression analysis was conducted to identify independent risk factors for MACE.A decision tree model was developed using machine learning Methods and externally validated with data from the MIMIC-IV/eICU databases to assess its predictive value for MACE.Results Among the included patients,847(10.6%)experienced MACE during a median hospitalization duration of 6(1-15)days.Compared to the non-MACE group,the MACE group exhibited significantly higher age,ICU days,levels of blood urea nitrogen,thrombin,white blood cells,red blood cell distribution width,creatine kinase isoenzyme,Charlson Comorbidity Index scores,and proportions of comorbid conditions such as hypertension,peripheral vascular disease,chronic lung disease,rheumatism,diabetes,kidney disease,severe liver disease,vasopressin use,and mechanical ventilation(all P<0.05).Multivariate Cox regression analysis revealed that increased age(HR=1.010,95%CI:1.002-1.018),Charlson Comorbidity Index scores(HR=1.114,95%CI:1.071-1.158),blood urea nitrogen(HR=1.006,95%CI:1.002-1.009),thrombin(HR=1.006,95%CI:1.002-1.010),white blood cells(HR=1.005,95%CI:1.002-1.009),red blood cell distribution width(HR=1.066,95%CI:1.035-1.098),creatine kinase isoenzyme(HR=1.004,95%CI:1.002-1.006),SOFA scores(HR=1.137,95%CI:1.113-1.162),and mechanical ventilation(HR=1.377,95%CI:1.160-1.635)were associated with higher MACE risk(P<0.05).External validation using 23,479 sepsis cases from the eICU database demonstrated good model performance,with AUC=0.78,R2=0.78,and MSE=0.04.Conclusion In sepsis patients,pre-existing comorbidities and specific sepsis-related characteristics are associated with increased MACE risk.Understanding these factors provides valuable insights for developing strategies to enhance cardiovascular health management and care in this population.
Keywords：Adult Patients with First-Time SepsisDuring HospitalizationMIMIC-IV DatabaseMajor Cardiovascular Adverse EventsMachine Learning
Publication Date：2025-01-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 256-261 )