Characteristics and outcomes of methicillin-resistant Staphylococcus aureus surgical-site infections in patients with cancer: a case-control study.

BACKGROUND: Methicillin-resistant Staphylococcus aureus (MRSA) infections remain a significant cause of morbidity and mortality. We experienced an increased incidence of MRSA surgical-site infections (MRSA SSIs) at our institution. However, to our knowledge, no studies have evaluated the risk factors and outcomes of MRSA SSIs in cancer patients. METHODS: We conducted a case-control study and identified all patients who had developed MRSA SSIs at our institution from July 1, 2002 to July 30, 2003, and all patients who had undergone surgery by the same surgical team during the same time period but who had not developed MRSA SSIs. Cases and controls were age-matched at 1:2 ratio. RESULTS: The study included 29 cases and 58 controls. Mean interval between surgery and MRSA SSI onset was 17.8 days (range 3-75 days). Cases were more likely than controls to have progressive cancer (72 versus 38%), have received antibiotics (mainly quinolones) within 24 h of surgery (17 versus 2%), have had ongoing infection (10 versus 0%), and have had longer hospital and intensive care unit stays (11.0 versus 7.8 days and 3.4 versus 1.5 days) (all P < 0.05). In a multivariate logistic regression analysis, significant predictors of MRSA SSI in cancer patients were antibiotics use <24 h of surgery and progressive cancer. No surgical factors (i.e., procedure time or timing of perioperative antibiotics) were associated with increased risk of MRSA SSI. CONCLUSIONS: Several clinical and postoperative factors were associated with increased risk of MRSA SSI in cancer patients, but antibiotic use before surgery (especially quinolones) and progressive cancer were the only independent predictors.

Central venous catheter and Stenotrophomonas maltophilia bacteremia in cancer patients.

BACKGROUND: Stenotrophomonas maltophilia bacteremia is frequently found in cancer patients. This study attempted to determine how often the catheters were the source of this infection and the risk factors associated with catheter-related bacteremias. METHODS: The microbiology records were retrospectively reviewed of all cancer patients having S. maltophilia bacteremia and indwelling central venous catheters seen between January 1998 and January 2004. In a multivariate analysis the patients' clinical characteristics, antimicrobial therapy, outcome, and source of bacteremia that were significantly associated with definite catheter-related S. maltophilia bacteremia as opposed to secondary bacteremia were identified. RESULTS: A total of 217 bacteremias were identified in 207 patients: 159 (73%) were primary catheter-related (53 definite, 89 probable, and 17 possible), 11 (5%) were primary noncatheter-related, and 47 (22%) were secondary. Multivariate analysis showed the following factors to be independently associated with definite catheter-related bacteremias: 1) polymicrobial bacteremia (odds ratio [OR], 7.6; 95% confidence interval [95% CI], 1.3-45.5); 2) no prior intensive care unit admission (OR, 0.06; 95% CI, 0.005-0.578); and 3) nonneutropenic status at onset (OR, 0.07; 95% CI, 0.013-0.419). The response rate to appropriate antibiotics and catheter removal was 95% in the patients with definite catheter-related bloodstream infections, compared with only 56% in the patients with secondary bacteremias (P = .001). CONCLUSIONS: The majority of the S. maltophilia bacteremias occurring in cancer patients with indwelling central venous catheters appear to be catheter-related and are often polymicrobial. Catheter-related S. maltophilia bacteremias occurred more frequently in noncritically ill, nonneutropenic patients, and prompt removal of the catheter was found to be associated with a better prognosis.