Blanket use of intranasal mupirocin for outbreak control and long-term prophylaxis of endemic methicillin-resistant Staphylococcus aureus in an open ward.

In December 1992, a thoracic ward in a Melbourne teaching hospital experienced an increase in patients infected with methicillin-resistant Staphylococcus aureus (MRSA). It was decided to attempt to control the outbreak by cohorting positive patients (infected and colonized), as well as nurse cohorting, emphasis on handwashing, and use of intranasal mupirocin initially three times a day for three days, then thrice weekly, for all patients in the ward (with or without MRSA). The campaign comprised for phases of 53, 45, 92 and 365 days, respectively. Patient and nurse cohorting stopped at the end of phase I. In phases I and II, surveillance nose swabs were taken on admission, then twice weekly; in phase III, on admission and weekly and in phase IV, on admission until the end of 1993. In phases I and II (98 days), only one patient acquired MRSA. When the frequency of mupirocin prophylaxis was decreased to once weekly (phase III), two patients acquired MRSA in 92 days (no significant difference): thrice weekly administration resumed (phase IV), during which there were three acquisitions in 365 days. The rates of nose colonization of admissions were 6.4%, 6.3%, 9.7% and 3.1% in phase I-IV, respectively. Only three patients were treated with vancomycin between July 1993 and June 1994 (significantly lower than historical rates, P = 0.0086). No mupirocin resistance was seen in MRSA isolates from this ward during phases I, II and III. In areas of low-level endemic MRSA, the blanket use of thrice-weekly intranasal mupirocin may be effective in decreasing serious infections with MRSA, and does not necessarily elicit mupirocin resistance.