Five-year prospective surveillance of nosocomial bloodstream infections in an Estonian paediatric intensive care unit.

BACKGROUND: Few studies provide rates of nosocomial bloodstream infections (BSIs) in mixed neonatal and paediatric intensive care units (PICUs). AIM: To determine the rate, pathogens and outcome of BSIs in an Estonian PICU. METHODS: Data were collected prospectively from 1st January 2004 to 31st December 2008 in the PICU of Tartu University Hospital. The definition criteria of the US Centers for Disease Control and Prevention were applied for the diagnosis of laboratory-confirmed BSI. FINDINGS: A total of 126 episodes of BSI were identified in 89 patients (74 neonates, eight infants, seven patients aged >1 year). Among neonates 42 (57%) had birth weight <1000 g. The overall incidence of BSI was 9.2 per 100 admissions, incidence density 12.8 per 1000 patient-days. Primary BSI was diagnosed in 92 episodes. Central line (CL)-associated BSI incidence density for neonates was 8.6 per 1000 CL-days with the highest incidence (27.4) among neonates with extremely low birth weight. The most common pathogens were coagulase-negative staphylococci (43%) and Serratia marcescens (14%). Resistance to meticillin was detected in four out of seven S. aureus isolates (all were part of an outbreak) and 23% of Enterobacteriaceae were extended spectrum beta-lactamase (ESBL)-producing strains. Overall case-fatality rate was 10%. CONCLUSION: We observed higher rates of BSIs in our mixed PICU than reported previously. High levels of antimicrobial resistance were detected. Future research should focus on the effects of infection control measures to prevent outbreaks and to decrease incidence of CL-associated BSI.

Epidemiology of nosocomial bloodstream infections in Estonia.

A prospective multicentre hospital-wide surveillance study was performed to investigate nosocomial bloodstream infections (BSIs) and to promote BSI surveillance in Estonia in 2004-2005. All patients from the acute care departments of two referral centres and one central hospital were included. A total of 549 episodes of BSI occurred in 507 patients (0.6 cases per 1000 patient-days). Of those, 55% occurred in intensive care units and 47% were catheter-associated infections. Of BSI cases, 24% occurred in patients with haematological malignancy. The in-hospital case-fatality rate was 31%. Of causative micro-organisms, 315 (53%) were Gram-positive aerobes, 232 (39%) were Gram-negative aerobes and 35 (6%) were fungi. Anaerobic bacteria accounted for 2%. The most common pathogens were coagulase-negative staphylococci (26%), Enterobacteriaceae (24%), enterococci (13%) and pseudomonas (10%). Eight percent of BSI were polymicrobial. Seven percent of Staphylococcus aureus isolates were meticillin resistant. Of pseudomonas isolates, 19%, 25%, 30% and 44% were resistant to ceftazidime, meropenem, piperacillin/tazobactam and imipenem, respectively. The incidence of BSI did not differ significantly from other reported studies. With the exception of relatively high antimicrobial resistance among pseudomonas, the overall resistance patterns of Estonian nosocomial bloodstream pathogens were similar to those seen in Nordic countries and lower than in Central and Southern Europe. This study contributes to the development and implementation of surveillance in Estonian hospitals.