Abstract
Background
Sepsis is a leading cause of morbidity, mortality, and health care costs worldwide.
Methods
We conducted a multi-center, prospective cohort study evaluating the yield of blood cultures drawn before and after empiric antimicrobial administration among adults presenting to the emergency department with severe manifestations of sepsis (ClinicalTrials.gov: NCT01867905). Enrolled patients who had the requisite blood cultures drawn were followed for 90 days. We explored the independent association between blood culture positivity and its time to positivity in relation to 90-day mortality.
Findings
325 participants were enrolled; 90-day mortality among the 315 subjects followed-up was 25·4% (80/315). Mortality was associated with age (mean age in those who died was 72·5 ±15·8 vs. 62·9 ±17·7 years among survivors, p<0·0001), greater Charlson Comorbidity Index (2 (IQR 1,3) vs. 1 (IQR 0,3), p=0·008), dementia (13/80 (16·2%) vs. 18/235 (7·7%), p=0·03), cancer (27/80 (33·8%) vs. 47/235 (20·0%), p=0·015), positive qSOFA score (57/80 (71·2%) vs. 129/235 (54·9%), p=0·009), and normal white blood cell counts (25/80 (31·2%) vs. 42/235 (17·9%), p=0·02). The presence of bacteremia, persistent bacteremia after antimicrobial infusion, and shorter time to blood culture positivity were not associated with mortality. Neither the source of infection nor pathogen affected mortality.
Interpretation
Although severe sepsis is an inflammatory condition triggered by infection, its 90-day survival is not influenced by blood culture positivity nor its time to positivity.
Funding
Vancouver Coastal Health; St-Paul’s Hospital Foundation Emergency Department Support Fund; the Fonds de Recherche Santé – Québec (CPY); Intramural Research Program of the NIH, Clinical Center (AL); the Maricopa Medical Foundation
Seasonal variations in blood culture numbers and time to positivity and potential impact of reducing incubation periods
