COLONIZATION AND INFECTION WITH ANTIBIOTIC-RESISTANT BACTERIA IN A LONG-TERM-CARE FACILITY

OBJECTIVE: To assess colonization and infection with methicillin-resis tant Staphylococcus aureus (MRSA), high-level gentamicin-resistant ent erococci (R-ENT) and gentamicin and/or ceftriaxone-resistant Gram-nega tive bacilli (R-GNB) and the factors that are associated with coloniza tion and infection with these organisms. DESIGN: Monthly surveillance for colonization and infection over a period of 2 years. In the second year, an intervention to decrease MRSA colonization by the use of mup irocin ointment was carried out. SETTING: Long-term care facility atta ched to an acute care Veterans Affairs Medical Center. PATIENTS: A tot al of 551 patients in the facility were followed for a period of 2 yea rs. MEASUREMENTS: Colonization and infection rates with MRSA, R-ENT, a nd R-GNB. Analysis of risk factors associated with colonization and in fection with these three groups of organisms. MAIN RESULTS: In the fir st year, colonization rates were highest for MRSA (22.7+/-1% patients colonized each month) and R-ENT (20.2+/-l%) and lower for R-GNB (12.6/-1%). After introduction of decolonization of nares and wounds with m upirocin, the rate of MRSA colonization fell significantly to 11.5 +/- 1.8%, but rates remained unchanged for R-ENT and R-GNB. Risk factors for MRSA colonization included the presence of wounds and decubitus ul cers. For R-ENT, the presence of wounds, renal failure, intermittent u rethral catheterization, low serum albumin, and poor functional level were significant. For R-GNB, intermittent urethral catheterization, ch ronic renal disease, inflammatory bowel disease, presence of wounds, a nd prior pneumonia were significantly associated with colonization. Ov erall, of infections caused by known organisms, 49.6% were due to MRSA , R-ENT, or R-GNB, and 50.4% were due to susceptible organisms. Infect ions were more commonly due to R-GNB (21.1% of all infections) than to R-ENT (8.3%) or MRSA (4.6%). The most common infections were urinary tract infections (42.9% of all infections) and skin and soft tissue in fections (31.9% of all infections). Risk factors for MRSA infections w ere diabetes mellitus and peripheral vascular disease for R-GNB infect ions were intermittent urethral catheterization and indwelling urethra l catheters, and no one factor was associated with R-ENT infection. CO NCLUSIONS: In our long-term care facility, colonization with resistant MRSA and R-ENT was more common than R-GNB, but infections were more o ften due to R-GNB than R-ENT and MRSA. Several host factors, which pot entially could be modified in order to prevent infections, emerged as important in colonization and infection with these antibiotic-resistan t organisms.