Prognostic risk factors for patients with surgical site infection in the intensive care unit
Objective To analyze the prognostic risk factors of surgical site infection in patients admitted to intensive care unit(ICU).Methods From July 1,2014 to December 31,2021,117 patients confirmed surgical site infection were enrolled and allocated into survival group and death group in the general ICU of the Second Hospital affiliated to Zhejiang University,Binjiang Campus.Patients'baseline information,laboratory tests,surgical and related treatments,and types of pathogenic bacteria after secretion cultures were recorded.Independent risk factors for poor patient prognosis were analyzed using multi-factor binary logistic regression,and receiver operating characteristic(ROC)were generated.And the artificial neural network(ANN)was used to analyze the importance of each variable,calculate the weights of the variables and the accuracy of the model.Results The results of univariate analysis revealed that sequential organ failure assessment(SOFA),acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ),prothrombin time(PT),activated partial thrombin time(APTT),N-terminal pro-brain nitric peptide(NT-pro-BNP),continuous renal replacement therapy(CRRT),days for CRRT treatment,days of vasoactive drug use,duration of mechanical ventilation,time in ICU stay,and multi-drug resistant organisms(MDROS)infection were associated with poor patients'prognosis.Multi-factor binary logistic regression analysis showed that APACHE Ⅱ score(odds ratio,OR=1.112,95％confidential interval,CI:1.009～1.225,P＜0.05),PT(OR=1.348,95％CI:1.078～1.686,P＜0.05),days for CRRT treatment(OR=1.358,95％CI:1.109～1.663,P＜0.05),and MDROS infection(OR=3.794,95％CI:1.084～13.281,P＜0.05)were independent risk factors for poor patients'prognosis with surgical site infections(P＜0.05).The ROC curve results showed that the area under the curve of APAPCHE Ⅱ score was the largest,and when the APACHE Ⅱ score was 16.5,its sensitivity and specificity were 71.5％and 65.9％,respectively,which could better predict the mortality outcome of patients with ICU surgical site infections.Higher APACHE Ⅱ scores,greater PT values,longer duration of CRRT treatment,and MDROS infection suggested a worse prognosis for patients,with a significantly increased risk of death in patients with incisional multi-drug resistant organisms infection when the APACHE Ⅱ score was＞16.5,the duration of CRRT treatment was＞7.5 days,and the PT was＞17.05 s.Artificial neural network(ANN)was used to analyze the significance of each variable,which showed 85.0％model accuracy for multi-drug resistant organisms infections(weight 0.389),APACHE Ⅱ scores(weight 0.228),PT values(weight 0.100),and CRRT treatment time(weight 0.283).Area under the ROC fitting curve 85％.Conclusions APACHE Ⅱ score,prolonged PT,days for CRRT treatment,and MDROS infection are independent risk factors for death in patients with surgical site infection,and hence early identification and treatment measures should be initiated.
Surgical site infectionIntensive care unitMortalityPrognosisRisk factorsMulti-drug resistant organisms
NETL
NSTL
万方数据
