Determinants and outcomes of bloodstream infection in adults associated with one versus two sets of positive index blood cultures

Abstract

ObjectivesTo investigate whether positivity in one or both index sets of blood cultures influences clinical determinants and mortality when diagnosing bloodstream infections (BSI). MethodsRetrospective population-based surveillance of all mono-microbial BSI was conducted among residents of the western interior of British Columbia. Clinical details were obtained by chart review and all-cause case-fatality was established at 30 days. Index cultures were defined as the first two sets of cultures initially drawn to diagnose incident BSI. ResultsA total of 2500 incident BSI were identified of which 945 (37.8%) and 1555 (62.2%) were based on one and two positive index cultures, respectively. There was an overall difference in the distribution of pathogens, with both Staphylococcus aureus and Streptococcus pneumoniae more likely to have two positive index cultures. Different foci of infection were associated with one versus two positive index cultures. Overall, 409 patients died within 30 days of index BSI for an all-cause case-fatality of 16.4%; with no difference between two positive (250/1555; 16.1%) and one positive (159/945; 16.8%; p 0.3) index blood culture. The number of positive index blood cultures was not associated with 30-day case-fatality after adjustment for confounding variables using logistic regression analysis. ConclusionsAlthough approximately one-third of BSI are diagnosed on the basis of a single positive blood culture and are associated with different clinical determinants, whether one or both index blood cultures are positive is not associated with lethal outcome.