Although infectious complications are common in critically ill patients, the incidence of blood stream infection (BSI) and the impact on outcome in coronary care unit (CCU) patients receiving invasive mechanical ventilation (MV) have not been well described. The objective of the present study was to describe the incidence, pathogens and impact on outcome of BSI in CCU patients.
Records for all patients admitted to the CCU requiring MV for > 48 hours between August 1, 2000 and July 31, 2002 were examined. Exclusion criteria included: non-cardiac diagnosis, transfer from another ICU, chronic MV, CCU stay < 48 hours and lack of consent for research. BSI was defined as the isolation of potential pathogens in the blood. Two sets of positive blood cultures were required for coagulase-negative Staphylococcus infection.
The study included 202 patients with a mean age of 66.1 years; 61% male and 94% Caucasians. The most frequent admission cardiac diagnoses were acute myocardial infarction (34.7%) and cardiac arrest (24.6%). The most common indications for MV were cardiac arrest (25.8%) and cardiogenic pulmonary edema (23.7%). Twenty patients developed BSI (9.9%). The BSI occurred at a median of 2 (range, 0-35) days after hospital admission. A single pathogen was identified in 18 patients and two organisms in two patients. The most commonly identified bacterial organisms were: methicillin-sensitive Staphylococcus aureus (6) and coagulase-negative Staphylococcus aureus (3). The median length of hospital stay was 10 days for patients with and without BSI (range, 4-61 and 2-76, respectively)(p=0.9023). Hospital mortality was not significantly different between patients with and without BSI (45.0 % vs 34.6%, p=0.3574).
Although BSI occurs in about 10% of patients admitted to CCU and receiving MV, it does not increase hospital mortality or length of hospital stay. Methicillin-sensitive Staphylococcus aureus is the most common organism causing blood stream infection in these patients.
In CCU patients receiving invasive MV suspected of having infection, empiric antibiotic therapy should cover staphylococcal infections.
A. Ensminger, None.