Background: National and international institutions, as CDC or ECDC, use different surveillance definitions for central line-associated bloodstream infection (CLABSI). This hinders comparison of findings. In Belgium, registration of CLABSI in hospitals is legally required since 2014 for a minimum of one quarter a year, as part of the national surveillance of bloodstream infections. For Belgian data collected since 2013, we assessed the proportion of cases corresponding to three different CLABSI classifications and its impact on CLABSI incidence and trends.
Materials/methods: To be considered as CLABSI in the Belgian bloodstream infection surveillance, a CLABSI must first meet the surveillance definition for hospital-associated bloodstream infection: being a laboratory-confirmed bloodstream infection (LCBI) occurring two days or more after admission at the hospital. Depending on surveillance information we then define three CLABSI classifications:
(1) Confirmed CLABSI: LCBI with clinical suspicion that a central line (CL) is the cause of LCBI and the association between LCBI and CL is microbiologically confirmed (same microorganism in blood culture and on CL),
(2) Probable CLABSI: LCBI with clinical suspicion that a CL is the cause of LCBI but no microbiological confirmation,
(3) Possible CLABSI: LCBI not secondary to an infection at another body site – origin recorded in surveillance form as ‘unknown’ – but CL present within the two days prior to the LCBI.
Results: The proportion of confirmed, probable and possible CLABSI was more or less stable since 2013 and the mean CLABSI incidence per 10,000 bed days varied in the same way (see table). Most CLABSI reported in the Belgian bloodstream infections surveillance were confirmed cases (around 40%), followed by probable (around 35%), and possible (around 25%) CLABSI.
Conclusions: Because proportion of cases meeting each of the three CLABSI classifications remained the same throughout the surveillance years, the CLABSI surveillance definition (depending on the classification(s) included in this definition) didn’t matter to follow-up CLABSI trend. The three different CLABSI classifications showed similar incidence trends.
Table: Central line-associated bloodstream infection (CLABSI) proportion, number and mean incidence per 10,000 bed-days in Belgian hospitals according to classification, 2013-2016