OBJECTIVES: To identify the attributed mortality rate of bloodstream hospital infection by Staphylococcus aureus resistant to methicillin (MRSA) and its effect on length of hospital stay. DESIGN:Case-control study SETTING: Hospital São Paulo da Universidade Federal de São Paulo, a 660-bed, tertiary-care teaching hospital in São Paulo, Brazil. PATIENTS: Seventy one adults patients with hospital-acquired MRSA bacteremia diagnosed between January 1, 1991, and September 30, 1992, and 71 MRSA-free controls were matched by the following criteria: age, sex, underlying disease, surgical procedure, same risk time and admissiom date. RESULTS: The incidence of patients with hospital sepsis by MRSA accounted for 73.22% of the patients with hospital bloodstream infection by Staphylococcus aureus . The mortality rate of the cases was 56.33 (40/71) and 11.26 (8/71) of the controls. The attributable mortality rate was 45.07% (OR=17.0; IC 95%=3.58 -- 202.26; p=0.000001). The length of hospital stay median time was of 32.55 days for the cases and 29.75 for the controls (p = 0.32). CONCLUSION: A high level of sepsis by MRSA was observed in all the Staphylococcus aureus bacteremia. The bloodstream hospital infection by MRSA itself does provide a high level of mortality independently from the patients base disease, without however, increasing their hospital length of stay.
Sepsis; Mortality; Staphylococcus aureus; Cross infection