Objective: To gain insight into the incidence of nosocomial infections and associated risk factors in Intensive Care Units (ICUs).
Design: Prospective.
Method: From July 1997 to December 1999, standardised surveillance of nosocomial infections was implemented in ICUs in 16 hospitals in the Netherlands. Surveillance was performed in patients with an ICU stay of > or = 48 hrs; data were collected from admission until discharge from ICU. Data-collection included demographic data and patient- and treatment-related risk factors. The data were aggregated in a national database.
Results: In the research period, hospitals sent good quality data for aggregation in the national database on 2795 patients (61% male) and 27,922 ICU patient days. The median length of stay was six days, the median 'Acute physiology and chronic health evaluation' (APACHE) II score was 17 and the median age was 67 years. A total number of 749 infected patients were found with 1,177 nosocomial infections (27% of patients, 42 infections/1000 patient days), consisting of 43% pneumonia, 20% sepsis, 21% urinary tract infections, 16% other types of infections. Out of all the patients, 62% was on mechanical ventilation, 64% had a central venous line and 89% had a urinary catheter in situ. Selective decontamination of the gastrointestinal tract was used for 12% of the patients, and systemic antibiotics for 68%. Micro-organisms most frequently isolated were Pseudomonas aeruginosa in patients with pneumonia, Staphylococcus epidermidis in catheter-related bloodstream infections and Escherichia coli in patients with urinary tract infections. Large differences in device use and incidence of infections were observed between the ICUs.
Conclusion: The aggregated data gave insight into the incidence of nosocomial infections and associated risk factors in ICUs. The data are meant as references to support decision- and policy-making in local infection control programs.