Clinical Outcomes of Penicillin-Based Therapy Compared to Alternative Enterococcal Treatment in Patients With Vancomycin-Susceptible Enterococcal Bloodstream Infections: A Monocentric Cohort Study.

The aim of this retrospective cohort study was to evaluate clinical outcomes of penicillin-based therapies (ampicillin, ampicillin/sulbactam, or piperacillin/tazobactam) compared with alternative therapies (glycopeptides, linezolid, or daptomycin) in patients with vancomycin-susceptible enterococcal bloodstream infections (BSIs). Adult patients with a positive blood culture (BC) for a vancomycin-susceptible Enterococcus species were screened from January 2017 to December 2018. Patients who received adequate antibiotic therapy for at least 48 h were enrolled. The primary outcome, 30-day mortality, was compared between patients receiving penicillin-based and those receiving alternative therapies using the Fisher exact test. Secondary outcomes included 90-day mortality, relapse BSI, length of hospital stay (LOS), and potential antibiotic-related adverse events. To identify independent predictors of 30-day mortality, multivariable logistic regression analysis was weighted for ampicillin susceptibility using propensity scores. Of the 150 patients screened, 127 (84.6%) were included in the final analysis. Among them, 36 (28.3%) developed a septic shock. Aminopenicillin-susceptible Enterococcus species (mainly E. faecalis) accounted for 59.1% of bacteremia cases. Sixty-seven (52.1%) patients received a penicillin-based regimen (monotherapy n = 48, combination n = 19), and 60 (47.2%) patients received an alternative regimen. Patients receiving penicillin-based therapy exhibited a lower 30-day mortality rate of 13.4% compared to 40% (p = 0.001) in those receiving alternative treatments. The 90-day mortality rates were reduced (13.4% versus 41.7%), and LOS was shorter (median 21 versus 35 days) for patients with penicillin-based therapy. There was no significant difference in relapse BSI or potential adverse events between the groups (p  > 0.05). However, in the multivariable weighted regression model, only septic shock (aOR with 95% CI: 23.53 [1.66; 334.54]) was identified as an independent positive predictor of 30-day mortality. In conclusion, penicillin-based therapy was still an effective and safe treatment but was not associated with an improved survival in a weighted analysis of patients with enterococcal BSI.