Surveillance of nosocomial infections: Prospective study in a pediatric intensive care unit

Background, Patients and Methods: From November 1997 through May 1998, the incidence of nosocomial infections was studied prospectively in a 10-bed mu ltidisciplinary pediatric intensive care unit in Germany. A standardized su rveillance [SEKI] system based on the National Nosocomial Infection Surveil lance [NNIS] System of the Centers for Disease Control and Prevention [CDC] was used. The CDC definitions for nosocomial infections were adapted to th e current practice of pediatric intensive care in Germany. Infection rates were calculated as infections per 100 patients, per 1000 patient-days, and per 1000 device-days (central venous catheters, urinary-catheters, and mech anical ventilation). Results: Fifteen nosocomial infections were recorded i n 201 patients during 1035 patient-days. The overall nosocomial infection r ates were 7.5/100 patients and 14.5/1000 patient-days. Device-associated no socomial infection rates for urinary-catheters and mechanical ventilation w ere 7.2/1000 utilization-days and thus below the 75(th) percentile of the l ast NNIS report. Central line infection rates were 10.7/1000 utilization da ys and therefore above the 75(th) percentile of the NNIS data (10.2/1000). The median length-of-stay was 5.1 days. Conclusions: Surveillance data are indispensable for internal and external quality control, and prospective su rveillance of nosocomial infections should become an essential component of hospital infection control programs in pediatric intensive care in Germany . The standardized calculation of (device utilization ratios and) device-sp ecific infection rates yields results which can be compared with national a nd international surveillance data. SEKI meets the criteria of a practice o riented, prospective and standardized surveillance system. Considerable eff orts for collecting and interpreting the required data should be balanced a gainst the benefit of prevention of nosocomial infections in this populatio n of critically ill persons.