Effect of changes in surgical practice on the rate and detection of nosocomial infections: a prospective analysis.

The practice of surgery is being performed increasingly on an outpatient basis. How these changes have influenced the nosocomial infection rate and the ability of standard, Center for Disease Control (CDC)-designed surveillance techniques to detect these infections is unknown. The goal of this study was to determine whether recent changes in surgical care have led to an increased nosocomial infection rate based on number of discharges and whether current surveillance techniques are adequate to detect these complications. Data were collected prospectively on all nosocomial infections over a 1-year period on the general surgery, trauma, and transplant units at a university hospital, as independently observed by both the study team [surgical auditors (SA)] and CDC-trained infection control practitioners (ICP). The patient study group had a high acuity of illness (for 516 episodes of infection, mean APACHE II score of 15.4, 45% intensive care unit-bound, mortality of 16%). The overall infection rate per 100 discharges was 23.8 for SA and 12.2 for ICP (P < 0.001 by chi2), higher than historical reports. SA detected significantly more surgical site infections, pneumonias, and non-Clostridium difficile-related gastrointestinal infections. These relative rates of detection, however, were similar to those described previously in prior studies using similar methodologies. The nosocomial infection rate in surgical patients, based on number of discharges, appears to be increasing, perhaps due to increased inpatient acuity of illness. Current epidemiological methods provide estimates of infection rates with effectiveness similar to that reported in previous epidemiological studies but fail to recognize many infections otherwise identified by surgeons dedicated to infection control.