Methods: We conducted a retrospective chart review of adult inpatients with CoNS from blood cultures in a university hospital and long-term acute care facility, which had batched Staphylococcus PNA-FISH for S. aureus versus CoNS in the 27 months prior (pre-microarray) to a real-time microarray for 18 months (post-microarray) after implementation. A post-microarray comment was also added: “Possible contaminant. Clinical correlation recommended.” Anti-CoNS exposure was recorded as either ≤1 dose or >1 dose.
Results: 1081 patients were identified and 470 excluded for administration of empiric anti-CoNS before BCx, anti-CoNS for other infection, death/discharge before results, polymicrobial BCx, or requested identification of CoNS result. Significantly more patients post-microarray received ≤1 dose (58% (245/425) pre-, 68% (127/186) p=0.013) and fewer received >1 dose (42% (180/425) pre-, 32% (59/186) post-microarray p=0.013). Pre- and post-microarray groups (n=425, n=186) significantly differed in the latter’s higher incidence of valve disease or cardiac device (11% v. 18%, p=0.04), mortality (7 v. 12%, p=0.04), longer length of stay (9.5 v. 14.5 days, p=0.016) and older age (60 v. 63 years, p=0.04).
Conclusion: Implementing a rapid microarray assay for gram-positive blood cultures with a descriptive comment resulted in significantly less exposure to anti-CoNS antibiotics in patients with CoNS from BCx, despite this group’s older age and higher incidence of valve disease or cardiac devices. Since many represent nonpathogenic contaminants, such assays may aid in decreasing unnecessary antimicrobial exposure.
M. Poulter, None
J. Eby, None
A. Mathers, None