Evaluation of ambulance blood cultures in patients with suspected sepsis. A rural prospective cohort study.

purposeTo describe the epidemiology of blood cultures (BCs) drawn from patients with suspected sepsis by a rural ambulance service.

methodsPatients were included if they had clinically suspected sepsis and at least one BC was drawn. Variables associated with positive BCs were identified with logistic regression, culture results before and after ambulance antimicrobial therapy were compared and the susceptibility to empirical antimicrobial therapy in bacterial isolates was determined.

resultsAmong 392 included patients, sepsis severity score were higher when BCs were positive than negative, but there was no significant difference in 30-day all-cause mortality between BC negative and positive patients. In 47 patients ambulance blood cultures were positive (47/347 (13,5%)). Fever (OR 1.54, 95% CI 1.11-2.15) was associated with higher odds of positive cultures, while lower respiratory tract infection was associated with lower odds of positive cultures (OR 0.10, 95% CI 0.04-0.30). In 205 patients who had both ambulance and hospital BCs drawn and received ambulance antimicrobial therapy, only the ambulance BCs were positive in 16 patients, whereas in 4 patients only hospital cultures were positive. Bacterial isolates from positive BCs were susceptible to the empirical antimicrobial regimen in 80% of cases.

conclusionPositive ambulance BCs were associated with higher sepsis severity, and ambulance cultures were most useful in febrile patients with a suspected focus of infection other than the lower airways. Repeated BCs after hospital admission added only limited diagnostic value if an adequate ambulance BC had already been drawn.