Creating a collaborative network for the study of bacteraemia in Denmark: frequency of recurrence with the same and different micro-organisms

Objectives: In most countries bacteraemia is a reportable infection only if the causative agent is subject to national surveillance. In a few countries all cases of bacteraemia are reportable but only one episode per microorganism is included per year. Therefore population-based data are sparse especially with regard to the epidemiology of recurrent bacteraemia. We present data from a newly established collaborative network in Denmark with prospective registration of bacteraemia in a population of approximately 1.7 mill. inhabitants.
Methods: Three departments of clinical microbiology (DCMs) participate in the network (Hvidovre Hospital and Herlev Hospital, The Capital Region and Aalborg Hospital, North Denmark Region). Bacteraemia is defined as a clinical episode with one or more positive blood cultures (BCs) given significance by a clinical microbiologist and the attending physicians. A recurrence was defined as a positive BC with the same microorganism/s as obtained ≥30 days or with another microorganism/s >48 hours after the first positive BC, respectively. We included all patients with bacteraemia during 2006-2007 and follow-up extended for 6 months or until death.
Results: In total, recurrent Escherichia coli bacteraemia was the most frequent with 23 to 31 episodes (table). A recurrent episode of either Staphyloccus aureus or Streptococcus pneumoniae bacteraemia was rare in the participating hospitals (2.0% to 3.9% and 0.6% to 1.0%, respectively). For all major blood culture pathogens the frequency range was 0% to 4.1% with the exception of Enterococcus faecalis (range 0% to 9.4%; table). A recurrent episode with any microorganism was most frequent following bacteraemia caused by E. faecalis (range 8.9% to 24.5%) and Klebsiella pneumoniae (range 6.7 to 11.3%). The frequency of recurrent candidaemia ranged from 6.9% to 13.6% (table).

Conclusion: The frequency of recurrent S. aureus and S. pneumoniae bacteraemia was lower than reported in previous studies. We found only small differences among major pathogens in the frequency of recurrence with the same microorganism. However, recurrent bacteraemia with any microorganism is common subsequent to K. pneumoniae and E. faecalis bacteraemia and to a lesser extent candidaemia. This suggests that host factors and therapeutic regimens in these patients should be targeted in further studies in order to optimise bacteraemia management.