Vancomycin Minimum Inhibitory Concentration and Outcome in Patients With Staphylococcus aureus Bacteremia: Pearl or Pellet?

(See the article by Holmes et al, on pages 340–7.) For decades, vancomycin has been the mainstay of treatment for serious methicillin-resistant Staphylococcus aureus (MRSA) infections. Recently, a debate has arisen over its continued utility for this purpose [1], fueled by a steady stream of reports linking a worse clinical outcome of MRSA-infected patients and a higher vancomycin minimum inhibitory concentration (MIC) of the infecting pathogen (Table 1). In fact, 13 of the 16 peer-reviewed publications addressing the topic to date have found that patient outcomes are worse when the vancomycin MIC of the infecting MRSA strain is higher, even if it still falls in the susceptible range. This association between poor patient outcome and high vancomycin MIC has been attributed to difficulty achieving an optimal vancomycin area under curve (AUC)/MIC ratio [2, 3], which is widely regarded as the best predictor of successful vancomycin therapy [4]. Clinicians have responded to this association by either increasing the vancomycin dose or switching to an alternative antibiotic. However, recent practice guidelines from the Infectious Diseases Society of America (IDSA) for treatment of MRSA infections do not recommend using a vancomycin alternative based on an elevated MIC alone, provided that the isolate has an MIC in the susceptible range [5]. Summary of Published Studies Evaluating Association Between Vancomycin Minimum Inhibitory Concentration (MIC) of Infecting Pathogen and Patient Outcome NOTE. BMD, broth microdilution; BD, broth dilution; IE, infective endocarditis; MRSA, methicillin-resistant Staphylococcus aureus; SSTI, skin and skin structure infection. Summary of Published Studies Evaluating Association Between Vancomycin Minimum Inhibitory Concentration (MIC) of Infecting Pathogen and Patient Outcome NOTE. BMD, broth microdilution; BD, broth dilution; IE, infective endocarditis; MRSA, methicillin-resistant Staphylococcus aureus; SSTI, skin and skin structure infection. In this issue of the Journal, Holmes et al add another wrinkle to this story. In their prospective multinational cohort of 532 patients with S. aureus bacteremia, elevated vancomycin MIC was associated with increased 30-day mortality in patients with MRSA bacteremia [6]. The truly exciting aspect of the article, however, was unveiled by the investigators’ ingenious decision to consider the same association in patients infected with methicillin-susceptible S. aureus (MSSA)—most of whom never received vancomycin. When the investigators evaluated vancomycin MIC values of the isolates from patients with MSSA bacteremia who were treated exclusively with flucloxacillin, they found that the same association existed between higher vancomycin MIC and worse overall clinical outcomes. In these patients, of course, a low vancomycin AUC/MIC ratio cannot be invoked as the explanation for worse patient outcome. How, then, are we to explain this link between MIC and mortality? And what should clinicians do when faced with the increasingly common dilemma of how to treat patients with MRSA infection and a vancomycin MIC >2 μg/mL? This study provides insights into both of these key questions. Strengths of the study include its robust sample size and its prospective, multinational design. With well over 500 patients, it is the largest of the prospective cohorts yet assembled to tackle this issue. It provides further evidence to support the association between higher vancomycin MIC and worse patient outcome in MRSA-infected patients. For the first time, however, the investigators present novel, compelling evidence that this association may not be causal, by demonstrating it in the bloodstream isolates of MSSA-infected patients who were not treated with vancomycin. The authors speculate that the vancomycin MIC could be a marker of as yet unidentified host or organism factors that affect treatment outcome. For example, factors that alter the biology of the S. aureus cell wall or cell membrane could conceivably both influence vancomycin MIC and enhance bacterial virulence. Indeed, changes such as increased cell wall thickness, decreased autolysis, and metabolic changes characterize S. aureus isolates with reduced susceptibility to vancomycin [7]. Thus, it is possible that these or similar changes might also influence treatment outcomes with β-lactam antibiotics. Alternatively, it may simply be that the link is driven by patient comorbidities and healthcare contact. Because of limitations of their database, Holmes et al were unable to provide data on these comorbid conditions or clinical variables such as APACHE II scores. We are, therefore, left with the possibility that “sicker” patients (both the chronically unwell and the acutely ill, hospitalized patients) who acquire nosocomial staphylococcal infection are simply more likely to be infected with hospital-resident strains of higher-MIC S. aureus and are more likely to die because of—or in spite of—their resultant infections. With their discovery that high vancomycin MIC does not fully explain the higher mortality suffered by MRSA-infected patients, Holmes et al have made an important contribution to the field. A common inference from previous studies that focused solely on MRSA-infected, vancomycin-treated patients was that alternatives to vancomycin should be used when the vancomycin MIC was above some cutoff value (but still ≤ 2 μg/mL) [1, 8–10]. The findings reported in the Holmes et al article call such an assumption into question. At a minimum, the results of this study provide supporting evidence for the recommendations of the IDSA Treatment Guidelines for MRSA that vancomycin MIC alone should not guide the decision of whether to use vancomycin for isolates with MIC ≤2 μg/mL [5]. In summary, Holmes et al are to be commended for their work. Although there clearly appears to be a link between high vancomycin MIC and poor outcome in S. aureus–infected patients, Holmes et al show that this link is probably not causal. For this reason, the practice of systematically switching patients infected with MRSA exhibiting vancomycin MIC ≤2 μg/mL from vancomycin to an alternative antibiotic is probably often unnecessary. Vancomycin’s long reign as first-line therapy for serious MRSA infections may be in its twilight, but there is still no proven heir to the throne. Funding for this study was provided by by K24-AI093969 (VGF) and NO1-AI90023 (VGF).