Mortality predictors of Staphylococcus aureus bacteremia: a prospective multicenter study

Abstract

Background: Staphylococcus aureus is one of the causes of both community and healthcare-associated bacteremia. The attributable mortality of S. aureus bacteremia (SAB) is still higher and predictors for mortality and clinical outcomes of this condition are need to be clarified. In this prospective observational study, we aimed to examine the predictive factors for mortality in patients with SAB in eight Turkish tertiary care hospitals.

Methods: Adult patients with signs and symptoms of bacteremia with positive blood cultures for S. aureus were included. All data for episodes of SAB including demographics, clinical and laboratory findings, antibiotics, and outcome were recorded for a 3-year (2010-2012) period. Cox proportional hazard model with forward selection was used to assess the independent effect of risk factors on mortality. A 28-day mortality was the dependent variable in the Cox regression analysis.

Results: A total of 255 episodes of SAB were enrolled. The median age of the patients was 59 years. Fifty-five percent of the episodes were considered as primary SAB and vascular catheter was the source of 42.1 %. Healthcare associated SAB was defined in 55.7 %. Blood cultures yielded methicillin-resistant S. aureus (MRSA) as a cause of SAB in 39.2 %. Initial empirical therapy was inappropriate in 28.2 %. Although overall mortality was observed in 52 (20.4 %), 28-day mortality rate was 15.3 %. Both the numbers of initial inappropriate empirical antibiotic treatment and the median hours to start an appropriate antibiotic between the cases of fatal outcome and survivors after fever onset were found to be similar (12/39 vs 60/216 and 6 vs 12 h, respectively; p > 0.05). High Charlson comorbidity index (CCI) score (p = 0.002), MRSA (p = 0.017), intensive care unit (ICU) admission (p < 0.001) and prior exposure to antibiotics (p = 0.002) all were significantly associated with mortality. The Cox analysis defined age [Hazard Ratio (HR) 1.03; p = 0.023], ICU admission (HR 6.9; p = 0.002), and high CCI score (HR 1.32; p = 0.002) as the independent predictive factors mortality.

Conclusions: The results of this prospective study showed that age, ICU stay and high CCI score of a patient were the independent predictors of mortality and MRSA was also significantly associated with mortality in SAB.

Fig. 1

Kaplan–Meier survival estimates of patients regarding intensive care unit (ICU) stay