Abstract Introduction The ePlex blood culture identification (BCID) gram-positive panel is a PCR-based approach used for the detection of common pathogens associated with bloodstream infections and corresponding antibiotic-resistance genes. One of the genes, mecA, is associated with methicillin, penicillin, and other penicillin-like-antibiotic-resistant gram-positive bacteria (GPB). It is unclear whether the BCID panel has predictive utility and accuracy in treating hospital infections due to GPB. This retrospective study attempts to determine the consistency and accuracy of BCID gene testing compared to more traditional susceptibility methods at a large public hospital. Method A search query was performed for all BCID results at one teaching hospital in a large metropolitan area for one year (2022). The query included demographic information, BCID results, culture results, and combined susceptibility data. The results were filtered to include only cases in which the mecA gene was detected. The filtered results were then analyzed to determine if the susceptibility results from the susceptibility platform (VITEK) were consistent with the expected susceptibility of the target gene, specifically for oxacillin and vancomycin. The susceptibility cutoff was set as >0.5 mcg/mL and <1 mcg/mL respectively, the standard threshold used at the institution. Results A total of 1211 culture results were obtained, of which the mecA gene was identified by BCID in 321 cases, along with the following GPB: Staphylococcus aureus, Staphylococcus epidermidis, and Staphylococcus sp. (not aureus). The corresponding cultures grew the following isolates: S. aureus, S. epidermidis, S. hominis, and S. haemolyticus. Of the 321 BCID results, 75 were identified as S. aureus. The isolates grew methicillin-resistant Staph aureus (MRSA) (69/75) 92%, while the typical S. aureus grew (6/75) in 8% of the isolates. In these cases, we compared susceptibility testing for both isolates as either sensitive or resistant, with a MIC of >0.5 mcg/mL for oxacillin to be designated as resistant, and <1 mcg/mL to be designated as sensitive to vancomycin. Of the 75 specimens, the concordance rate was 97% (73/75) for oxacillin and 100% (75/75) for vancomycin. The average receipt to verification turnaround time (TAT) was 36 hours for BCID results and 121 hours (about 5 days) for culture sensitivity results. Conclusion The results show that oxacillin resistance can be predicted by detecting S. aureus and the mecA gene using the BCID panel. However, the mecA gene can be identified in other GPB such as S. epidermidis, in which it should not be used to infer oxacillin or vancomycin susceptibilities. In the case of mecA and S. aureus detection by BCID, treating the patient as a MRSA infection may be appropriate before the susceptibilities have been performed. This approach can reduce the turnaround time for antimicrobial resistance reporting and improve antimicrobial stewardship efforts.
Read full abstract