Abstract
Repeat blood cultures are common in children after an initial positive culture. However, in contrast to adults, there are little data to help guide clinicians when a repeat culture is necessary to assess for persistent bacteremia. This study identifies factors associated with persistent bloodstream infections (BSI) in children to inform diagnostic stewardship. This cross-sectional study of children less than 18 years with at least one positive blood culture over a 5-year period utilized a generalized linear equation model to predict patient and microbial factors associated with persistent BSI defined as a positive blood culture with the same organism >48 hours after the index culture. Four hundred and five patients had 502 positive blood cultures yielding 556 organisms. Sixty-seven (13.2%) cultures were persistently positive. Anaerobic organisms (0/37) and Streptococcus species (0/104) were never recovered from repeat cultures. Staphylococcus aureus (OR 9.45, CI 5.15-17.35) and yeast (OR 78.18, CI 9.45-646.6) were statistically associated with persistent BSI. Patients with prior positive cultures (OR 1.44, CI 1.12-1.84) or a central venous catheter (OR 2.20, 95% CI 1.04-3.92) were also at risk for persistence. Immune dysfunction and elevated inflammatory markers at the time of the index blood culture were not significantly associated with persistence. Yeast or S. aureus were associated with persistent BSI, while anaerobes and Streptococcus species were never persistent. Patient characteristics at the time of blood draw did not predict persistence other than having previous positive blood cultures or a central venous catheter. These data can inform when repeat blood cultures have clinical value and reduce the risk of unnecessary blood draws in children. IMPORTANCE: We identify factors associated with bloodstream infection persistence in children. Our findings can help guide blood culture stewardship efforts in pediatric patients, especially in light of blood culture supply shortages.