RESUMO
OBJECTIVE: To identify risk factors associated with nosocomial bloodstream infections caused by multiple clones of the staphylococcal cassette chromosome mec (SCCmec) type IV strain of methicillin-resistant Staphylococcus aureus (MRSA). DESIGN: An unmatched case-control study (at a ratio of 1:2) performed during the period from October 2002 through September 2003. SETTING: A 2,000-bed tertiary care teaching hospital affiliated with the University of São Paulo in São Paulo, Brazil. METHODS: Case patients (n=30) were defined either as patients who had a bloodstream infection due to SCCmec type IV MRSA diagnosed at least 48 hours after hospital admission or as neonates with the infection who were born in the hospital. Control patients (n=60) were defined as patients with SCCmec type III MRSA infection diagnosed at least 48 hours after hospital admission. Genes encoding virulence factors were studied in the isolates recovered from case patients, and molecular typing of the SCCmec type IV MRSA isolates was also done by pulsed-field gel electrophoresis and multilocus sequence typing. RESULTS: In multivariate analysis, the following 3 variables were significantly associated with having a nosocomial bloodstream infection caused by SCCmec type IV strains of MRSA: an age of less than 1 year, less frequent use of a central venous catheter (odds ratio [OR], 0.07 [95% confidence interval {CI}, 0.02-0.28]; p= .025), and female sex. A second analysis was performed that excluded the case and control patients from the neonatal unit, and, in multivariate analysis, the following variables were significantly associated with having a nosocomial bloodstream infection caused by SCCmec type IV strains of MRSA: less frequent use of a central venous catheter (OR, 0.12 [95% CI, 0.03-0.55]; p= .007), lower Acute Physiology and Chronic Health Evaluation II score on admission (OR, 0.14 [95% CI, 0.03-0.61]; p= .009), less frequent surgery (OR, 0.21 [95% CI, 0.06-0.83]; p= .025), and female sex (OR, 5.70 [95% CI, 1.32-24.66]; p= .020). Of the 29 SCCmec type IV MRSA isolates recovered from case patients, none contained the Panton-Valentine leukocidin, gamma-hemolysin, enterotoxin B or C, or toxic shock syndrome toxin-1. All of the isolates contained genes for the LukE-LukD leukocidin and alpha-hemolysin. Genes for enterotoxin A were present in 1 isolate, and genes for beta-hemolysin were present in 3 isolates. CONCLUSIONS: "Classical" risk factors do not apply to patients infected with the SCCmec type IV strain of MRSA, which is an important cause of nosocomial bacteremia. This strain infects a patient population that is less ill and has had less frequent invasive procedures than a patient population infected with the multidrug-resistant strain of SCCmec type III MRSA. We found that virulence factors were rare and that Panton-Valentine leukocidin was absent. There were multiple clones of the SCCmec type IV strain in our hospital. Children under 1 year of age were at a higher risk. There was a predominant clone (sequence type 5) in this patient population.
