Abstract
Background
Blood cultures have high sensitivity to detect bacteremia in septic neonates when >=1 ml of blood is collected. Neonatologists often cite low confidence in microbiologic sampling as rationale for continuing antibiotics without a focus of infection despite negative blood cultures, resulting in prolonged antimicrobial therapy. We aim to describe the blood culture sample volumes in NICU patients, to identify factors associated with sample volumes < 1ml, and to compare the sample volumes of patients treated for culture-negative sepsis with those with bloodstream infections and those treated for a ≤72-hour sepsis rule-out
Methods
Data from this observational cohort study were collected retrospectively and prospectively from NICU patients with blood cultures obtained from September 2018 to February 2019. Clinical data were collected through chart review. All inoculated culture bottles were weighed for volume calculation. We determined the association of age, weight, sample source, and time of collection with volume < 1mL. Continuous variables were analyzed using Wilcoxon-Mann-Whitney, and categorical variables using chi-squared test. For aim 3, the volumes of the groups were compared using analysis of variance.
Results
A total of 310 blood cultures were identified, corresponding to 159 patients. Of these, 49 (16%) were positive. Among the negative blood cultures, 86% were collected in patients who subsequently received antibiotics (Figure 1). Median inoculated volume was 0.6 ml (IQR: 0.1-2.4). Weight and age at time of culture collection, source of sample, and time of collection were not significantly associated with the inoculation of < 1ml of blood. Median volume of blood was 0.6ml (0.3-0.6) for sepsis rule-out, 0.6ml (0.2-0.6) for bloodstream infection, and 0.6ml (0.6-1.4) for culture-negative sepsis. No difference was found among the three groups (p=0.54)
Figure 1. Classification of blood cultures identified during study period
Conclusion
The blood volume collected for cultures in the NICU is lower than recommended. Clinical and environmental characteristics are not significantly associated with the inoculated volume. The volume of blood sampled does not differ in patients with culture-negative sepsis, bloodstream infection and sepsis rule-out, and should not be a justification for longer duration of antibiotic therapy
Disclosures
All Authors: No reported disclosures
Bacteremic sepsis leads to higher mortality when adjusting for confounders with propensity score matching
