Predictors of mortality in patients with methicillin-resistant Staphylococcus aureus (MRSA) bacteraemia: the role of empiric antibiotic therapy

Abstract

The objective of this study was to evaluate prognostic factors and the influence of different empiric antibiotic therapies on outcome and mortality in a cohort of 100 inpatients with bacteraemia (84 cases nosocomial) caused by methicillin-resistant Staphylococcus aureus (MRSA). Patients were investigated by means of a standard protocol at a 944-bed hospital in the years 2000–2004. Empiric antibiotic therapies included vancomycin (n=49), teicoplanin (n=20), linezolid (n=17), other antibiotics active in vitro (n=7), and inactive antibiotics (n=7). Overall mortality was 40% (12% among linezolid-treated patients; 46.3% among glycopeptide-treated patients). In bivariate analyses, the following factors were statistically associated with higher mortality: rapidly fatal underlying disease, altered mental status, metabolic acidosis, and acute severe clinical condition at the onset of bacteraemia; development of complications (septic shock, renal failure, and disseminated intravascular coagulopathy); empiric monotherapy with glycopeptides (vs combination therapy with an aminoglycoside); and inadequate empiric treatment. Empiric therapy with linezolid was associated with lower mortality. In multivariate analysis, risk factors associated with higher mortality included acute severity of illness (OR 7.49; 95%CI 1.19–25.3) and altered mental status (OR 4.83; 95%CI 1.22–19.15) at onset, complications (OR 3.42; 95%CI 1.02–17.46), and inappropriate empiric treatment (OR 7.6; 95%CI 1.87–31.14). In multivariate analysis limited to patients who received empiric therapy with either linezolid (n=17) or glycopeptides (n=69), linezolid was associated with greater rates of survival (OR 7.7; 95%CI 1.1–53) and microbiological eradication (OR 11.76; 95%CI 1.46–90.9) but not with fewer complications (OR 0.71; 95%CI 0.16–3.25). In conclusion, the main prognostic factors associated with mortality in patients with MRSA bacteraemia are complications, acute severe clinical condition at onset, and inappropriate empiric treatment. Empiric therapy with linezolid was associated with greater survival and more successful microbiological eradication but did not reduce complications.