Risk Factors and Prognosis of Polymyxin- and Carbapenem-Resistant Enterobacteriaceae Infections: A Propensity-Matched Real-World Study

Abstract

The risk factors and prognosis of polymyxin- and carbapenem-resistant Enterobacteriaceae (PR-CRE) infections were analyzed to reduce their incidence and concurrently improve patient prognosis. This retrospective study analyzed patients with CRE infections admitted to West China Hospital of Sichuan University between 1 September 2019 and 30 September 2023. Based on polymyxin susceptibility, the cases were categorized into PR-CRE and PS-CRE (polymyxin-susceptible CRE) groups, with 1:1 propensity score matching performed between the two cohorts. Comprehensive data, including demographic characteristics, laboratory findings, antibiotic regimens, and clinical outcomes, were collected and analyzed to identify risk factors for PR-CRE infections and evaluate treatment efficacy. This study aims to provide evidence-based references for infection control strategies and antimicrobial stewardship in managing PR-CRE infections. A total of 254 patients were included in this study, with 127 patients in the PR-CRE group. The sensitivity rates of isolates in the PR-CRE group to tigecycline and ceftazidime-avibactam were 94.4% and 88.9%, respectively. Multivariate analysis identified chronic organic disease (OR 2.747, 95% CI 1.303-5.789; p = 0.008) and the use of polymyxin ≥ 3 days (OR 19.203, 95% CI 7.126-51.752; p < 0.001) as independent risk factors for PR-CRE infection. Moreover, ceftazidime-avibactam-based regimens were superior to tigecycline-based regimens for the treatment of PR-CRE infections (71.43% vs. 58.46%), especially in critically ill patients (33.33% vs. 58.82%). Finally, a SOFA score ≥ 5.5 (HR 6.718, 95% CI 2.526-17.866; p < 0.001) was identified as an independent risk factor for 28-day mortality in patients with PR-CRE infection. The presence of chronic organic diseases and the use of polymyxin for ≥3 days were identified as independent risk factors associated with PR-CRE infections in hospitalized patients, highlighting the need to optimize polymyxin use. Furthermore, the efficacy of ceftazidime-avibactam-based regimens may be superior to tigecycline-based regimens for the treatment of PR-CRE infections.

Keywords: carbapenem-resistant Enterobacteriaceae; polymyxin; prognosis; risk factors.

Figure 1

Receiver operating characteristic curve of SOFA score for predicting treatment outcomes in patients with polymyxin- and carbapenem-resistant Enterobacteriaceae infections.

Figure 2

Analysis of 28-day survival in patients with polymyxin- and carbapenem-resistant Enterobacteriaceae infection at different sites. Kaplan–Meier survival analysis demonstrated that the survival rates of patients with different infection sites varied significantly (p = 0.039). Upon pairwise comparisons, it was found that patients with urinary tract infections had a higher survival rate compared to those with bloodstream (p = 0.007) or lung infections (p = 0.008), and there was no statistical difference among other groups (p > 0.05).

Figure 3

Analysis of 28-day survival in patients with polymyxin- and carbapenem-resistant Enterobacteriaceae (PR-CRE) infection treated with different regimens. (a) Patients with SOFA score < 5.5 were divided into three groups according to different antimicrobial regimens: ceftazidime–avibactam (CAZ-AVI)-based regimen (n = 11), tigecycline (TGC)-based regimen (n = 38), and CAZ-AVI combined with TGC group (n = 6). Survival analysis showed no statistically significant difference in the 28-day survival rate among PR-CRE-infected patients (p = 0.306). (b) Patients with SOFA score ≥ 5.5 were divided into three groups according to different antimicrobial regimens: ceftazidime–avibactam (CAZ-AVI)-based regimen (n = 17), tigecycline (TGC)-based regimen (n = 27), and CAZ-AVI combined with TGC group (n = 7). Survival analysis showed no statistically significant difference in the 28-day survival rate among PR-CRE-infected patients (p = 0.609).