[Monitoring infection at the intensive care unit--a multicenter pilot study].

Abstract

During a period of 3 months an infection survey was carried out in 4 intensive care units (ICUs), 2 in Vienna, Austria, and one each in Ulm and Münster, Federal Republic of Germany, using a common protocol. A total of 329 patients was monitored prospectively. This pilot study was performed to evaluate the usefulness of parameters included in the monitoring form. It was attempted to characterize the patient populations of the four units. Mean duration of stay (1-12 days), mortality (8-26%), leading diagnosis upon admission, intubation rate (41-91%) and use of pulmonary artery catheter (12-35%) were distinctly different. The rate of patients admitted already with an infection was 9-43%, septicemia was diagnosed in up to 27% of the diseased. The rate of infection acquired in the unit was between 12 and 37%, the most frequent types were bronchopneumonia, septicemia and urinary tract infection. When septicemia patients were compared to non-septicemia patients who had been admitted for more than 3 days, it appeared that the latter stayed significantly shorter at the ICU and showed less frequently bronchopneumonia or urinary tract infection at the time of admission. Septicemia patients acquired more frequently additional infections like broncho-pneumonia or urinary tract infection while staying at the ICU. The median day of onset of septicemia was the fifth day and only in a quarter of cases diagnosis could be supported by a positive blood culture. The use of antibiotics in the 4 ICUs is compared and shows marked differences. Based upon experience with this type of infection survey a new modified protocol is introduced, which displays the time course of documented events.