Comparison of outcomes from daptomycin or linezolid treatment for vancomycin-resistant enterococcal bloodstream infection: A retrospective, multicenter, cohort study

Abstract

Background: The optimal treatment for bloodstream infections (BSIs) with vancomycin-resistant enterococci (VRE) is unknown. Objective: This study examined outcomes in patients treated with daptomycin or linezolid for VRE BSI. Methods: A retrospective, multicenter, cohort study was performed via chart review. Hospitalized patients treated for VRE BSI with daptomycin or linezolid from September 1, 2003, to June 30, 2007, were identified via pharmacy and microbiology reports at each institution. Patients aged <18 years or with polymicrobial bacteremia were excluded from analysis. Linezolid and daptomycin were included because the participating institutions used either of the 2 agents as first-line treatment for VRE BSI. Univariate and multivariate analyses were performed to determine the effect of drug selection on mortality and duration of BSI. Duration of BSI was defined as the amount of time from the draw date of the first positive blood culture to the draw date of the first finalized negative blood culture. Adverse events were not assessed. Results: One-hundred one patients from 3 participating US hospitals experiencing VRE BSI were identified. Sixty-seven patients were treated with daptomycin and 34 with linezolid. Baseline characteristics appeared comparable between the daptomycin- and linezolidtreated groups, with the exception of shock ( P = 0.049), prior vancomycin treatment ( P = 0.002), and prior linezolid treatment ( P < 0.001), all of which occurred significantly more often in daptomycin-treated patients. Inpatient mortality occurred in 31 daptomycin- and 10 linezolid-treated patients (46.3% vs 29.4%; P = NS). Linear regression found that shock (P = 0.015), infective endocarditis ( P = 0.021), and concurrent rifampin or gentamicin treatment ( P = 0.01) were associated with prolonged duration of positive cultures. Logistic regression revealed that shock (odds ratio [OR] = 14.24; P = 0.008), infection with Enterococcus faecium (OR = 53.10; P = 0.024), previous linezolid treatment (OR = 6.63; P = 0.031), concurrent rifampin or gentamicin treatment (OR = 6.48; P = 0.046), and a nonline source of infection (OR = 6.67; P = 0.019) were associated with increased mortality. Conclusions: In this retrospective cohort analysis, there were no significant differences in mortality of VRE BSI between patients receiving daptomycin or linezolid. Underlying comorbidities appeared to best predict outcome; however, given the retrospective nature of this study, larger, prospective, randomized, comparative studies are needed to control for potential biases and determine definitive outcome differences between these 2 antimicrobials.