16 September 2019

Vancouver, BC – Sepsis is a leading cause of death in hospitals, requiring prompt and efficient action after it is diagnosed. However, a new study by Dr. David D. Sweet, Dr. Matthew Cheng and Dr. Rob Stenstrom from the University of British Columbia has found that administering treatment before blood cultures have been taken can significantly diminish opportunities to identify infections and treat patients effectively. Findings from the study, which was published today in the Annals of Internal Medicine, suggest that delivering antibiotic treatment prior to drawing blood cultures to test for infections makes it significantly less likely to identify the cause of illness, reducing the potential to find the best course of treatment. This research is the first to examine the risk of obtaining blood cultures shortly after antimicrobial treatment, compared to cultures drawn prior to treatment.

The prospective study, which ran across the United States and Canada, found that testing after treatment resulted in a loss of almost 50% of available clinical information. These findings confirm and provide further evidence for the clinical guidelines for the care of patients with sepsis, which affects more than 27 million people globally each year and leads to at least eight million deaths annually.

For British Columbians, these findings provide evidence that bolsters efforts currently underway in the province. In particular, the BC Emergency Sepsis Guidelines, created through a provincial Sepsis Clinical Expert Group led by the BC Patient Safety & Quality Council (BCPSQC), advise emergency departments to draw blood cultures prior to administering antibiotics. This advice is now corroborated by the evidence. From 2012 – 13, when the guidelines were first published, to 2016 – 17, the in-hospital sepsis rate for British Columbia has improved, going from above to below the national average (4.6 to 3.5 per 1,000).

“These findings have the potential to save lives,” said Sweet, who is also the Provincial Clinical Lead for Sepsis at BCPSQC. “The more we know about how to treat sepsis and septic shock, the better our chances at reducing mortality and morbidity. Now we need to ensure these guidelines are being followed in emergency rooms and hospital wards around the world.”

Sweet leads the BCPSQC’s efforts to support physicians and nurses across the province to champion sepsis improvement and reduce morbidity and mortality associated with sepsis. To do so, BCPSQC connects health care providers to share sepsis resources, improve consistency of care, and spread innovation and improvement ideas by administering the BC Sepsis Network. With the support of BCPSQC and the Ministry of Health, the Network’s 200 leaders from across the province encourage the use of the BC Emergency Department Sepsis Guidelines.

For more information on this media release, please contact:

Courtney Chu
Communications Specialist, BC Patient Safety & Quality Council


Background: Administering antimicrobials without waiting for blood culture collection could potentially decrease time to treatment and improve outcomes, but it is unclear how this strategy impacts diagnostic sensitivity.

Objective: To determine the sensitivity of blood cultures obtained shortly after the initiation of antimicrobial therapy in patients with severe manifestations of sepsis.

Design: Patient-level, single-arm, diagnostic study.

Setting: Seven emergency departments across North America.

Participants: Adults with severe manifestations of sepsis, including a systolic blood pressure < 90 mmHg and/or a serum lactate concentration ≥ 4 mmol/L.

Diagnostic procedure: Additional sets of blood cultures were obtained within 120 minutes following antimicrobial treatment initiation.

Measurements: The sensitivity of blood cultures obtained after initiation of antimicrobial therapy.

Results: Of 3164 participants screened, 325 were included (mean age 65.6 years; 63.0% male) and had repeat blood cultures drawn after the initiation of antimicrobial therapy (median time of 70 minutes, interquartile ranges 50 to 110). Pre-antimicrobial blood cultures were positive for one or more microbial pathogens in 102/325 (31.4%) patients; post-antimicrobial blood cultures were positive in 63/325 (19.4%). The absolute difference in the proportion of positive blood cultures was 12.0% ([95% CI 5.4 to 18.6%]; p<0.0001) between pre and post-antimicrobial blood cultures. The sensitivity of post-antimicrobial blood cultures was 52.9% [95% CI 42.8 to 62.9%]. When the results of other microbiological cultures were included, microbial pathogens were recovered in 69 of 102 [67.7%, 95% CI 57.7 to 76.6%].

Limitations: Only a proportion of screened patients were recruited.

Conclusions: Among patients with severe manifestations of sepsis, the administration of empiric antimicrobial therapy significantly reduces the sensitivity of blood cultures drawn shortly after treatment initiation.

ClinicalTrials.gov number: NCT01867905