PARIS—Intuitively, most physicians who work in the intensive care unit (ICU) would say that the onset of bacteremia adversely affects patient outcomes. Recent case-control studies have suggested that some types of bloodstream infections—specifically, primary and catheter-related bacteremias—do not increase mortality in the critically ill.[1,2] “But that just does not seem right,” says Christian Brun-Buisson, MD.

Dr. Brun-Buisson, a Professor of Intensive Care Medicine at the University of Paris, was a principal investigator in a new study that produced very different findings.[3] It found that primary, catheter-related, and secondary bacteremias were all associated with a longer ICU stay and with excess mortality, although the mortality increase was markedly higher in the patients with secondary bacteremia than it was in the other two groups.

MULTICENTER STUDY

Between February 15 and June 15, 1998, the investigators conducted a prospective cohort study in 15 adult medical and surgical/trauma ICUs at Paris-area hospitals. Because these units routinely culture all central venous catheters with quantitative techniques, it was easy for the investigators to determine the source of the bloodstream infection, said Dr. Brun-Buisson, who is also Director of the Medical Intensive Care and Infection Control Units at the Hôpital Henri Mondor in Créteil, France.

To be included in the study, patients had to be 16 years or older and to have stayed in the ICU for at least 48 hours. Patients who had been readmitted to the ICU were excluded.

To be diagnosed with an ICU-acquired bacteremia, patients had to have at least one positive blood culture obtained more than 48 hours after ICU admission. Also, the bacteremia had to be unrelated to an infection present on admission. Bacteremia was classified as primary if no source could be identified that was infected with the same pathogen found in the blood. It was considered catheter-related if there was inflammation at the catheter exit site and if at least 103 colony-forming units of the same pathogen could be recovered from the patient’s blood and from the catheter’s intravascular tip.

A secondary ICU-acquired bacteremia was diagnosed when the same pathogen could be cultured from the blood and another identified source.

IMPACT VARIES WITH TYPE

Of the 2,201 ICU patients who met the study’s inclusion criteria, 111 experienced an ICU-acquired bacteremia. Thirty-two of these were primary, 29 were catheter-related, and 50 were secondary bloodstream infections. Gram-positive and gram-negative organisms were each responsible for about 40% of the bacteremias; the remainder were polymicrobial. Staphylococcus aureus and Enterobacteriaceae were markedly more common, and coagulase-negative staphylococci were much less common, in the patients with secondary bacteremia than in the other groups; otherwise, the types of organisms found were similar. About 60% of patients received appropriate therapy within 48 hours of the first positive blood culture.

To determine the impact of each type of bacteremia on mortality and length of ICU stay, Dr. Brun-Buisson and his colleagues assessed the variables that were associated with an increased length of ICU stay among all 2,201 patients enrolled in their study: age, Simplified Acute Physiology Score II, patient location prior to ICU admission, admission category (medical vs surgical), fatality of underlying disease, and the onset of an ICU-acquired bacteremia. The investigators then performed a case-control analysis of bacteremic and non-bacteremic subjects who were matched for these variables (other than bacteremia). Adequate controls were available for 96 of the patients with bloodstream infections.

There were 50 deaths among the patients with bacteremia and 16 among controls (mortality rates: 52% and 17%, respectively), for an overall excess mortality rate from bacteremia of about 35%. The mortality rates were 50% for patients with primary bacteremias, 38.5% for those with catheter-related bloodstream infections, and 62% for those with secondary bacteremias. Dr. Brun-Buisson and colleagues calculate that the excess mortality rates for these three types of bacteremia were 29%, 12%, and 55%, respectively.

Overall, the onset of bacteremia lengthened the ICU stay by about 5.5 days—the median stay was 19 days for the controls and 24.5 days for the patients with bacteremia. However, the type of bacteremia had little influence on the excess length of stay.

—Timothy Begany

References
1. Digiovine B, Chenoweth C, Watts C, Higgins M. The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit Care Med. 1999;160:976-981.
2. Soufir L, Timsit JF, Mahe C, et al. Attributable morbidity and mortality of catheter-related septicemia in critically ill patients: a matched, risk-adjusted, cohort study. Infect Control Hosp Epidemiol. 1999;20:396-401.
3. Renaud B, Brun-Buisson C. Outcomes of primary and catheter-related bacteremia: a cohort and case-control study in critically ill patients. Am J Respir Crit Care Med. 2001;163:1584-1590.

Return to table of contents