Improved Outcomes with Linezolid for Methicillin-Resistant Staphylococcus aureus Infections: Better Drug or Reduced Vancomycin Susceptibility?

In a recent issue of this journal, Weigelt et al. (7) reported the results of a randomized trial of linezolid versus vancomycin for complicated skin and soft tissue infections. Overall, the outcomes for patients treated with linezolid and vancomycin were equivalent; however, patients with proven methicillin-resistant Staphylococcus aureus (MRSA) infections had significantly improved outcomes if they were treated with linezolid (7). Notably, there was no difference in outcome for patients with proven methicillin-susceptible S. aureus (MSSA) infection. These results are intriguing and are similar to the results of a post hoc subgroup analysis of two prospective trials of linezolid versus vancomycin treatment for nosocomial pneumonia, which suggested improved outcome for linezolid-treated patients with proven MRSA infection but equivalent outcomes for all S. aureus infections (8). Why does linezolid appear superior to vancomycin for MRSA infection but equivalent for infections caused by MSSA?

Weigelt et al. suggest the difference may be because the severity of illness was greater in the MRSA group than the MSSA group (7), while Wunderink et al. did not really address this issue, although the APACHE II scores were no different between the MSSA and MRSA groups in the post hoc analysis (8). Both of these reports are notable because they neglect an important consideration, namely, the potential role of reduced vancomycin susceptibility in a subset of MRSA isolates leading to differences in outcome for linezolid and vancomycin.

Low-level vancomycin resistance in S. aureus is increasingly recognized as an important clinical entity (1, 4, 5). In particular, heterogenous vancomycin-intermediate S. aureus has been reported from many countries (6) but is difficult to detect using routine susceptibility testing methods, with population analysis profile testing considered necessary for accurate identification (2, 3). Although these strains have a vancomycin MIC within the susceptible range (often 2 or 4 mg/liter), they have been associated with vancomycin treatment failure (1), and therapy with agents such as linezolid has been effective in these patients (4). Reduced vancomycin susceptibility appears to be far more common in MRSA than in MSSA strains (3, 6).

The recent study by Weigelt et al. (7) does not report vancomycin susceptibility data for any S. aureus isolates; they do not appear to have considered this important issue. Thus, the apparent clinical superiority of linezolid against MRSA may only be relevant to those strains that exhibit reduced vancomycin susceptibility. Without knowing what proportion of the MRSA strains in the study (7) exhibited heterogenous resistance to vancomycin, doubt remains about the relative efficacies of linezolid and vancomycin for fully vancomycin-susceptible MRSA isolates, which in our institution still constitute over 90% of our MRSA bacteremia isolates (1).