Course and outcome of bacteremia due to Staphylococcus aureus: evaluation of different clinical case definitions

In a retrospective survey of patients hospitalized in the University Hospital of Basel, Switzerland, the course and outcome of 281 cases of true bacteremia due to Staphylococcus aureus over a 7-year period were analyzed. The main purpose was to evaluate different case definitions. In 78% of cases the source of bacteremia was obvious; vascular access sites (27%) and wounds (10%) were the most common sources. Metastasizing foci were more common in cases of primary vs. secondary bacteremia (P < .001). The incidence of endocarditis was higher in cases in which no portal of entry was defined (P < .03). The overall mortality rate was high at 34% partly because of inappropriate initial antibiotic therapy. With the introduction of an infectious disease service at the hospital, the fraction of misjudged results of blood culture diminished 2.5-fold. Among the differently defined cases, the mortality rate was significantly higher for cases of complicated vs. uncomplicated bacteremia (P < .01), for cases of primary vs. secondary bacteremia (P = .05), and for patients with endocarditis or other secondary foci (P < .001). Since only one methicillin-resistant strain was isolated, multiresistant staphylococci were not a problem in the hospital. Different case definitions allowed the detection of patients at increased risk for complications and death. In the treatment of sepsis with no evident focus, initial antimicrobial therapy should include the use of agents with antistaphylococcal activity.