Associate Professor McMaster University Hamilton, Ontario, Canada
Background: Coagulase negative staphylococci (CONS) can cause late onset sepsis in the NICU but can also be considered a contaminant in blood cultures resulting in unnecessary antibiotic use. Objective: As part of a quality improvement initiative in our NICU, we aimed to improve our treatment of CONS positive blood cultures. We implemented a standard approach for CONS positive cultures to improve contaminant diagnosis and reduce antibiotic use in Oct 2019. Design/Methods: A before-after comparison of CONS positive cases (April-Sept 2019 compared to Oct 2019-Feb 2022) was completed using retrospective chart review. T-test, Fisher’s exact test and Chi- squared tests were used to compare treatment outcomes pre and post algorithm and characteristics of cases deemed contaminant versus true infection as appropriate. Results: Ninety-five CONS positive cases were reviewed, 82% of all isolates were cloxacillin resistant. Sixty-two cases were considered contaminants and 33 infections. After applying the algorithm, more cases of contamination post algorithm (n=57/78) compared to pre algorithm (n=7/17) were diagnosed (p=0.02). The mean time to positivity reduced significantly in the contaminant group after the algorithm was implemented (21.6 to 17.7h, p=0.03). We achieved 92% compliance with repeat cultures being collected before vancomycin was initiated, 100% compliance with stopping antibiotics when considered a contaminant and 75.9% compliance with treatment duration recommendations. The empiric antibiotic chosen was vancomycin in 24 cases (25.3%). Time to positivity, clinical status and repeat cultures can help to differentiate true infection versus contamination. In our analysis the mean time to positivity among infections was 15 hours, compared to a time to positivity of more than 24 hours in most studies being considered contamination. Providers diagnosed more contaminant cases after applying the algorithm . Compliance with all elements of the algorithm was more than 75%.
Conclusion(s): Our QI initiative reduced the number of CONS positive cultures treated as infection. Repeating cultures prior to starting vancomycin can assist in determining contaminated cultures. Future directions include further optimization of the algorithm (e.g., reducing the time to positivity cut-off to be deemed contaminant at 18h and reduce treatment durations to 5 to 7days if no line in situ ) and ongoing improvements in blood culture collection and measurement of compliance to our algorithm to further reduce unnecessary antibiotic use in the NICU.
