TY - RPRT T1 - Surveillance of bloodstream infections in Belgian hospitals ('SEP')-Annual report 2014, Surveillance data 2000-2014 Y1 - 2015 A1 - N Hammami A1 - Marie-Laurence Lambert KW - Belgian KW - Bloodstream infection KW - Bloodstream infections KW - data KW - hospital KW - hospitals KW - INFECTION KW - infections KW - report KW - Surveillance AB -

Background
Bloodstream infections acquired in the hospital (HA-BSI) are an important cause of morbidity and mortality. Many HA-BSI, especially those associated to an invasive device, are preventable. The surveillance programme for BSI in the hospital (SEP) exists in Belgium since 1992. The surveillance protocol has been reviewed in 2012, with an emphasis on the usefulness of the data collection for guidance and evaluation of preventive actions. Since 2014, the participation for a minimum of 3 months per year became a legal obligation for all acute care hospitals in Belgium. This implies registration of standardized data for each BSI episode acquired in the hospital (HA-SEP: by definition, BSI occurring 2 days or more after admission of the patient in the hospital).
The report provides a summary of the surveillance data from Belgium for the period 2000-1014.

Participation of hospitals
The participation of hospitals has increased considerably in the recent years. In 2014, 130 hospitals registered their data for at least 3 months and 62 (48%) recorded data for the full year. This continuous participation over the year serves best the purpose of surveillance as a tool for prevention, and it also demonstrates the interest of hospitals for the problem of HA-BSI.

Annual trends
The mean incidence in 2014 was 7.9 HA-BSI / 10 000 patient-days and in intensive care unit (ICU) 39.8 / 10 000 patient-days. This incidence appears relatively stable since 2000. There was a great variability between hospitals (figure 1) and between the regions. The incidence was higher in teaching hospitals, and in Brussels (of 8 academic hospitals participating in 2014, 4 are in Brussels). The incidence was the lowest in Flanders. The BSI associated to a central venous catheter (probable or confirmed by a culture of catheter) seemed to increase these past 3 years. The distribution of the microorganisms involved in HA-BSI changed with an increase in the incidence of HA-BSI due to Gram-negative microorganisms (Escherichia coli, Klebsiella pneumoniae) and a decrease of HA-BSI due to Staphylococcus aureus.

Description of episodes and their probable origin.
In 2014, a total of 6 847 HA-BSI episodes were recorded; 86% met the definition 'at least one blood culture with a pathogen' (in 13% the definition 'a skin contaminant was isolated in at least 2 different blood cultures' was met). One episode out of 5 (21%) was acquired in the ICU.
For half of the episodes, the time to onset was 13 days or more after admission (median for BSI acquired in ICU: 10 days or more after admission). Half of the patients were more than 71 years old. One out of 5 (19%) patients died, but the causal link with HA-BSI is impossible to determine.
The most common origin of HA-BSI - confirmed or probable - is a central venous catheter (CVC, 26%), followed by a urinary tract infection (20%). These proportions differ according to whether the infection is acquired in the ICU or not (figure 4).
For 45% of the HA-BSI, the origin is confirmed (the same organism was isolated from blood cultures, and the site considered to be the source of infection). An invasive device is associated directly (vascular catheter) or indirectly (urinary catheter, intubation) in 43% of the HA-BSI.

Microorganisms and antimicrobial resistance markers.
The most common microorganisms in 2014 (HA-BSI) were E. coli (21%), coagulase-negative Staphylococcus (15%) and Staphylococcus aureus (11%). Half of hospitals reported no cases of HA-BSI caused by MRSA; the mean incidence was 0.16 MRSA-BSI / 10,000 patient-days. A comparison of resistance phenotypes between 2013 and 2014 is presented in table 1.

Key points

PB - WIV-ISP CY - Brussels U1 - 31582 ER -