Winning the battle but losing the war: methicillin-resistant Staphylococcus aureus (MRSA) infection at a teaching hospital.

A methicillin-resistant Staphylococcus aureus (MRSA) control policy, aimed at eradication, was established at a 1000-bed hospital in 1985, applied consistently for 10.5 years, and then relaxed. Its components included screening of high-risk patients, transfer of carriers to exhaust-ventilated isolation rooms, closure of wards to new admissions when local transmission was detected, MRSA screening during outbreaks, and prospective collection of clinical and epidemiological information. During the eradication policy period, every 6 months, a mean of 5.1 patients (range 1-12) already carrying MRSA were admitted, and a mean of 3.6 (range 0-16) acquired carriage in the hospital. The largest outbreak comprised 11 patients despite epidemic MRSA strain EMRSA-16 being introduced six times, and MRSA did not become endemic. MRSA-positive admissions increased progressively from 1993; nursing staff workload increased, areas available for alternative patient accommodation were reduced, the resulting ward closures interfered with clinical services, and hence the control policy was relaxed in mid-1995. Isolation facilities were overwhelmed with 622 new patient-isolates in the next 18 months, and there were 67 clinical infections in 1996. The proportion of blood cultures positive for MRSA rose nearly sevenfold by 1996 and 27-fold by 1997. Thus, repeated eradication of MRSA, even epidemic strains, by use of a stringent policy, is possible given sufficient resources, whereas flexible national guidelines designed to control, but not eradicate, epidemic staphylococci, are currently unlikely to be successful. The costs of eradication policies need to be weighed against those of endemicity.