Postoperative wound infections remain a major source of morbidity and, less frequently, a source of mortality in the surgical patient.' Their occurrence nationally is estimated to be more than 500,000 per year, or about 2.8 per 100 operations performed.2 This incidence varies from surgeon to surgeon, from hospital to hospital, from surgical procedure to surgical procedure and, most importantly, from patient to patient. The increased cost attributable to these wound infections depends primarily on whether prolonged hospitalization or rehospitalization is necessary.3 Major complications, such as deep sternal wound infections, have a grave impact, increasing the duration of hospitalization as much as 20-fold and the cost of hospitalization five-fold.4 Traditional surveillance of the surgical wound, practiced widely in the 1970s, depended primarily on infection control personnel searching for positive cultures from the microbiology laboratory. Finding a positive culture of wound drainage or exudate triggered a review of the patient's chart and of the patient, if still hospitalized. Errors in this approach were caused by inadequate and widely varying definitions of surgical wound infection, in addition to missing clinical infections when cultures were not done or were falsely negative. Using a representative sample of U.S. general hospitals (Study of the Efficacy of Nosocomial Infection Control [SENIC] Project), the efficacy of infection surveillance and control in preventing nosocomial infections was established by the Centers for Disease Control (CDC) in 1985.5 A 32% reduction in nosocomial infections was noted from 1970 through 1976 in the participating hospitals where the essential components of the intensive