Clinical characteristics of postoperative patients with suspected infections in the intensive care unit:a multi-center cohort study of China
10.3969/j.issn.1008-9691.2025.03.005
- VernacularTitle:中国成人ICU外科术后可疑感染患者的现况分析
- Author:
Shuguang YANG
1
;
Yao SUN
;
Ting WANG
;
Hua ZHANG
;
Fengxue ZHU
;
Youzhong AN
;
Huiying ZHAO
Author Information
1. 北京大学人民医院重症医学科,北京 100044
- Publication Type:Journal Article
- Keywords:
Intensive care unit;
Postoperative infection;
Drug resistant
- From:
Chinese Journal of Integrated Traditional and Western Medicine in Intensive and Critical Care
2025;32(3):279-285
- CountryChina
- Language:Chinese
-
Abstract:
Objective To investigate mortality risk factors and characterize pathogen distribution and antimicrobial resistance patterns in intensive care unit(ICU)patients with suspected infection following surgery.Methods A total of 65 hospitals in 16 provinces in China from July 1,2021,to December 31,2022.Clinical data were collected for surgical patients transferred to the ICU with suspected infection.Data included demographics[sex,age,underlying conditions(hypertension,diabetes,cardiovascular/cerebrovascular disease,hematologic disease)],surgical site,infection site,microbiological results with susceptibility testing,drug resistance and acute physiology and chronic health evaluationⅡ(APACHEⅡ)scores and the length of hospital stays.Patients were stratified by prognosis into death group and survial group,by drug resistance status into resistant and non-resistant groups.Univariate and multivariate Logistic regression identified risk factors for mortality and antimicrobial resistance.Draw the receiver operator characteristic curve(ROC curve)to evaluate the predictive value of each risk factor for patient prognosis and drug resistance occurrence.Results A total of 677 patients with suspected postoperative infection in the ICU were enrolled.There were 96 deaths and 591 survivors.① Analysis of risk factors affecting prognosis:univariate analysis showed that compared with the survival group,the patients in the had a higher APACHEⅡscore,the proportion of patients with previous cerebrovascular disease,surgery sites in the abdomen,chest,brain,pelvis,limbs,other areas,as well as those with pulmonary infection,bloodstream infection,urinary tract infection,Gram-positive bacterial infection(Candida),fungal infection,multi-drug resistant bacterial infection was higher,and the length of hospital stay was shorter(all P<0.05).Multivariate Logistic regression analysis identified higher APACHEⅡscore[odds ratio(OR)=1.15,95%confidence interval(95%CI)was 1.11-1.20],pulmonary infection(OR=4.07,95%CI was 2.05-8.11),bloodstream infection(OR=2.61,95%CI was 1.52-4.51),and urinary tract infection(OR=2.20,95%CI was 1.01-4.42)were independent risk factors for prognosis(all P<0.05).The ROC curve analysis showed that the death risk prediction model established based on the above independent risk factors had certain predictive value for the prognosis of ICU postoperative patients with suspected infection,area under the curve(AUC)=0.820,95%CI was 0.770-0.860,P<0.05.②Regarding antimicrobial resistance:250 patients developed resistance and 427 did not.Univariate analysis showed compared with the non-resistant group,the APACHEⅡscore,the proportion of patients with cerebrovascular diseases,hematological diseases,surgeries at chest,brain,limbs,other sites,as well as those with pulmonary infection,bloodstream infection,urinary tract infection,intracranial infection,Gram-negative bacillus infection,and Gram-positive cocci infection(Staphylococcus epidermidis,Staphylococcus aureus,Enterococcus faecalis),and the mortality rate in the resistant group were significantly higher,the proportion of patients with surgeries at abdominal cavity,pelvic cavity and abdominal cavity infection were significantly lower,and the length of hospital stay was significantly longer(all P<0.05).Multivariate Logistic regression analysis bloodstream infection(OR=4.00,95%CI was 2.22-7.19),urinary tract infection(OR=3.25,95%CI was 1.47-7.17),Klebsiella pneumoniae infection(OR=2.23,95%CI was 11.22-44.02),Acinetobacter baumannii infection(OR=48.12,95%CI was 20.10-115.17),Pseudomonas aeruginosa infection(OR=34.06,95%CI was 13.00-89.25),Escherichia coli infection(OR=24.97,95%CI was 10.55-59.13),Stenotrophomonas maltophilia infection(OR=19.04,95%CI was 3.30-109.96),and Staphylococcus aureus infection(OR=13.48,95%CI was 4.57-39.78)were independent risk factors for resistance(all P<0.01).The ROC curve analysis showed that the predictive model for drug resistance established based on the above independent risk factors had certain predictive value for drug resistance in adult patients with suspected infections after surgery in the ICU.The AUC=0.920,95%CI was 0.890-0.940,P<0.05.Conclusion Higher APACHEⅡscores and the presence of pulmonary,bloodstream,or urinary tract infections were associated with increased mortality in ICU patients with suspected postoperative infection.Patients with bloodstream or urinary tract infections,or infections caused by Klebsiella pneumoniae,Acinetobacter baumannii,Pseudomonas aeruginosa,Escherichia coli,Stenotrophomonas maltophilia,or Staphylococcus aureus,had significantly higher odds of developing antimicrobial resistance.
