Staphylococcal Bacteremia: Methicillin Resistance and Mortality Revisited

Abstract & Commentary

Synopsis: A meta-analysis of 31 published studies of Staphylococcus aureus bacteremia found that MRSA bacteremia was associated with a significantly higher mortality than MSSA bacteremia.

Source: Cosgrove SE, et al. Comparison of mortality associated with methicillin-resistant and methicillin-susceptible Staphylococcus aureus bacteremia: A meta-analysis. Clin Infect Dis. 2003;36:53-59.

Cosgrove and colleagues reviewed studies comparing mortality rates for MSSA and MRSA bacteremia published in the English language literature from 1980-2000. They excluded studies in which the odds ratio of mortality for MSSA or MRSA bacteremia could not be determined, fewer than 2 cases of MRSA bacteremia were reported, the subjects were children, or the data had been previously published. They identified 31 studies meeting criteria, with a total of 3963 cases. Of these, 2603 (65.7%) had MSSA bacteremia and 1360 (34.3%) had MRSA bacteremia. Twenty-four studies (77.4%) found no significant difference in mortality and (22.6%) found significantly higher mortality associated with MRSA bacteremia. No study found a significantly higher mortality associated with MSSA bacteremia. Combining the results of all studies demonstrated a significant increase in mortality associated with MRSA bacteremia compared with MSSA bacteremia (pooled OR 1.93, 95% CI 1.54-2.42, P < .001). There was, however, significant heterogeneity among the studies’ results (P = .03 by the Q statistic), indicating that the different studies may be estimating multiple effects on the association between methicillin resistance and bacteremia mortality.

Cosgrove et al then analyzed the subgroup of 11 studies in which the association between methicillin resistance and mortality was adjusted for potential confounding factors. The pooled OR for this subgroup of studies was 1.88 (95% CI 1.32-2.18; P < .001), and there was no significant heterogeneity among results (P = .16). They conducted further subgroup analyses, including studies that reported unadjusted mortality, predominantly nosocomial bacteremias, a high proportion of catheter-related infections, endocarditis, and outbreaks of MRSA infection. In each of the additional subgroup analyses, there was a significant association between methicillin resistance and mortality and minimal or no evidence of heterogeneity of results.

Comment by Robert Muder, MD

The question of whether methicillin resistance is an independent contributor to mortality in S aureus bacteremia has been the subject of debate for more than 20 years. Cosgrove et al attempt to answer this contentious issue by use of meta-analysis, a technique that has not been widely used in the field of infectious diseases. This study has several strengths, which include explicit and reasonable inclusion and exclusion criteria for the studies examined, an attempt to identify confounding factors and heterogeneity among studies, and explicit a prior hypotheses. They identified a consistent association between methicillin resistance and mortality whether considering the total number of studies, the studies that were adjusted for potential confounders, the studies that were unadjusted, and in each of several additional subgroup analyses.

One potential weakness in this analysis is the possibility that the studies that attempted to adjust for potential confounders did not completely identify and adjust for all confounding factors. However, the relative consistency of the increase in mortality associated with MRSA bacteremia is striking and suggests that the contribution of methicillin resistance to mortality is a real one.

MRSA strains are not more virulent than susceptible strains. A likely cause for the increased mortality associated with MRSA bacteremia is that vancomycin is intrinsically less active against S aureus than are beta-lactam antibiotics.1,2 The newer antistaphylococcal agents active against MRSA, linezolid and quinupristin/dalfopristin, appear to be clinically equivalent but not superior to vancomycin, when used to treat a variety of S aureus infections.3-5 At present, there are no large randomized trials comparing either agent with vancomycin for the treatment of staphylococcal bacteremia.

Despite the limitations of meta-analyses, Cosgrove et al provide strong evidence for an association between methicillin resistance and mortality in S aureus bacteremia. One likely reason is the relatively poor antistaphylococcal activity of antimicrobial agents available for the treatment of MRSA infection. Although new agents with greater activity are clearly needed, they are not likely to be available for the foreseeable future. Improved control of the spread of MRSA within hospitals currently offers the best opportunity for reduction in morbidity and mortality due to MRSA infection.

Dr. Muder, Hospital Epidemiologist, Pittsburgh VA Medical Center, Pittsburgh, is Associate Editor of Infectious Disease Alert.


1. Small PM, Chambers HF. Vancomycin for Staphylococcus aureus endocarditis in intravenous drug users. Antimicrob Agents Chemother. 1990;34:1227-1231.

2. Levine DP, et al. Slow response to vancomycin or vancomycin plus rifampin in methicillin-resistant Staphylococcus aureus endocarditis. Ann Intern Med. 1991; 115:674-680.

3. Rubenstein E, et al. Linezolid (PNU-100766) versus vancomycin in the treatment of hospitalized patients with nosocomial pneumonia: A randomized, double-blind, multi-center study. Clin Infect Dis. 2001;32: 402-412.

4. Stevens DL, et al. Linezolid versus vancomycin for the treatment of methicillin-resistant Staphylococcus aureus infections. Clin Infect Dis. 2002;34:1481-1490.

5. Fagon J, et al. Treatment of Gram-positive nosocomial pneumonia. Prospective randomized comparison of quinupristin/dalfopristin versus vancomycin. Am J Resp Crit Care Med. 2001;163:1759-1760.