2172. True Positivity of Common Blood Culture Contaminants among Pediatric Hospitalizations in the United States, 2009–2016

Abstract

Abstract Background Distinguishing blood culture (BC) results between common contaminants (CC) and truly pathogenic organisms can be challenging, especially among pediatric patients, but is important for effective clinical care. However, no recent studies have analyzed the true positivity of common BC contaminants in pediatric patients using linked laboratory data from a large national sample of United States hospitals. Methods We conducted a retrospective cohort study among patients ages < 19 using the Premier Healthcare Database (2009–2016), limiting to hospitals reporting ≥ 4 years of BC data and encounters with one of the five most frequent CC among laboratory-confirmed BC. True positivity was defined for each CC as a second positive BC within 48 hours among all BCs. A multivariable logistic regression model including all variables significant in univariate analyses was created comparing encounters: (1) with and without a second BC; and (2) second BC positive vs. negative, with corresponding adjusted odds ratios (aOR) and 95% confidence intervals (CI) reported. Results A total of 5056 isolates corresponding to 4915 encounters with a CC were included in this analysis; 3075 (61%) isolates had a second BC within 48 hours. Adjusted odds of a second BC were higher for encounters from urban (aOR: 1.73, 95% CI: 1.31, 2.29) and ≥ 500 bed hospitals (aOR 1.40, 95% CI: 1.20,1.63). True positivity was 20.2% for coagulase-negative staphylococci (CoNS), 5.9% for Bacillus spp., 5.2% for Viridans group streptococci, 5.0% for Diphtheroids spp., and 3.1% for Micrococcus spp. True positivity for CoNS was higher among neonates but all other organisms were higher for non-neonates (figure). Adjusted odds of true positivity were higher for encounters with chronic conditions (OR 1.44, 95% CI: 1.13, 1.82), a central line in place (OR: 1.65, 95% CI: 1.30, 2.10), per length of stay day (OR: 1.01 (1.01, 1.01), and with an intensive care unit admission (OR: 1.39, 95% CI: 1.08, 1.77). Conclusion True positivity varied substantially by organism, and in most cases was higher among non-neonates. Regional variations for conducting a second BC within 48 hours were found, and more seriously ill patient encounters were more likely to have a common contaminant be pathogenic. Disclosures All authors: No reported disclosures.