Analysis of nosocomial blood-stream infections in Martin Teaching Hospital (1986-1999)

Nosocomial blood-stream infections (BSI) reported and epidemiologically investigated in teaching hospital in Martin over the period of 14 years (1986-1999) were analysed. The main attention was focused on nosocomial sepses of exogenous origin that can effectively be prevented. Over the studied period with 404 126 hospitalised patients, 822 cases of bacteremia were reported and epidemiologically investigated out of which 332 (40%) were concluded as a sepsis of nosocomial origin. The majority of nosocomial sepses occurred in non-pediatric departments (55%) compared to pediatric departments (45%). An overwhelming majority of nosocomial sepses were of exogenous origin (86%) out of which 147 (51%) occurred in non-pediatric and 140 (49%) in pediatric departments. We have not observed significant differences in incidence of internal or external risk factors in patients with exogenous sepses compared to endogenous ones which emphasizes the necessity of observing the principles of asepsis as well as antisepsis in all patients irrespective of the number of risk factors. The most common cause of exogenous nosocomial sepses were gramnegative bacteria that were also responsible for 16 out of 17 epidemic outbreaks in the given period. Gramnegative blood-stream infections were also associated with the highest case-fatality rate in pediatric patients, while in non-pediatric departments the highest lethality was observed among patients with mycotic sepses. However, the difference in lethality of mycotic versus gramnegative blood-stream infections was not statistically significant. It was also observed that the presence of more than 1 microorganism in the blood culture was not associated with significantly higher case-fatality rate compared to single microorganism sepses. It is necessary to establish an effective system of surveillance of BSI in close cooperation with laboratories of clinical microbiology in all health-care providing institutions in Slovak Republic using a standardised protocol for epidemiological investigation of nosocomial blood-stream infections proposed by authors of the paper.