More evidence of the benefits of rational antimicrobial use in clinical practice.

Abstract

Many of us work in overcrowded hospitals with overworked nursing staff and overstressed physicians. It is easy to dream about moving to a brand-new hospital with wonderfully clean wards and wonderfully inspired staff (with perfect hand hygiene). Surely then all of our infection control problems would be over forever, or would they? Unless physicians within the hospital modified their patterns of antibiotic use, it is doubtful that our problems with antibiotic-resistant bacteria would be over. Selection of antibiotic-resistant organisms would almost certainly occur. The extent of the subsequent problem with antibiotic resistance would be determined in part by the level of misuse of antimicrobial agents. Antibiotic management or, as it is sometimes termed, antibiotic stewardship clearly has an integral role in infection control. It has long been appreciated that the use of narrower spectrum, more targeted antimicrobials for as short a time as possible is associated with a lower likelihood of the development of resistance. It can be difficult to promote the importance of this message to prescribers. They often consider the immediate interests of their patients as being of paramount importance and do not appreciate or observe the subsequent resistance problems that may arise. Four articles in this issue of Infection Control and Hospital Epidemiology provide us with further data that will help us to promote rational antimicrobial prescribing.1-4 The article by Lodise et al. examines risk factors for bacteremia with methicillin-resistant versus methicillinsusceptible Staphylococcus aureus (MRSA vs MSSA).3 Importantly, these authors take the local perspective in their assessment of risk factors for MRSA. Undoubtedly, they have read the numerous previous articles examining risk factors for MRSA infection. However, they realize that if they are going to reduce vancomycin use by promoting empiric use of antistaphylococcal beta-lactam antibiotics in preference to vancomycin in selected patients in their hospital, they are going to need local data to have confidence in their recommendations. Risk factors for methicillin resistance in Detroit may be different from risk factors for methicillin resistance in Des Moines, Dublin, or Darwin. Their case–control study involved 494 patients with S. aureus bacteremia during 21⁄2 years. Almost half of the infections (45.5%) were caused by MRSA. Onset of bacteremia while in the hospital, a history of hospitalization, and the presence of decubitus ulcers were associated with an increased likelihood of MRSA versus MSSA infection. However, the most significant predictive factor, with an odds ratio of 9.2, was prior antibiotic use. This study reinforces an association that has been established for many other organisms. Importantly, it also allows those advising antibiotic choice prior to susceptibilities becoming available to make recommendations with close to 90% certainty as to the likelihood of an organism being MRSA or MSSA. The next step is for these authors to show that with the use of the information from this study, vancomycin use can be decreased while maintaining favorable outcomes from staphylococcal bacteremias. The long-term use of intravenous glycopeptides is sometimes clinically indicated and essential. There has been much concern over the long-term use of vancomycin leading to the development of colonization and infection with vancomycin-intermediate S. aureus (VISA). The article by Bernard et al. in this issue is encouraging in this regard.2 For 34 patients with MRSA osteomyelitis, intravenous vancomycin was given in a standard dose (20 mg/kg/d) for a mean of 34 days or a high dose (40 mg/kg/d) for a mean of 37 days. Trough vancomycin levels of 10 to 15 mg/L were targeted in the patients given the standard dose and 20 to 25 mg/L in the patients given the high dose. Swabs from wounds, the anterior nares, and the groin were surveyed during therapy and 2 months after the