Abstract
OBJECTIVE: Verigene blood culture panels comprise rapid diagnostic testing, which aids in early bacteremia species identification. This study determined the concordance of Verigene rapid diagnostic results compared with the Vitek reference standard in patients admitted to a children's hospital. METHODS: This was a 3-year retrospective observational study of neonatal and pediatric patients ≤18 years admitted to a children's hospital with confirmed bacteremia for whom Verigene testing was performed. Verigene testing was conducted on cultures with reported growth on Gram stain and final organism speciation confirmed via Vitek. Percent concordance and positive percent agreement with 95% CIs were calculated for Verigene panel-identifiable organisms. Negative percent agreement with 95% CIs was calculated for non-panel organisms. Time-to-result was calculated from Gram stain reporting to both Verigene and Vitek final organism susceptibility. RESULTS: One hundred thirty-five Gram-positive (GP) and 51 Gram-negative (GN) isolates were identified through Vitek. Verigene GP panel-detectable organisms were correctly identified 96.9% (125/129) at the genus level and 95.3% (123/129) at the species level. Overall positive percent agreement was 95.3 (CI: 90.2-98.3). Negative percent agreement was 83.3 (CI: 35.9-99.6) for the 6 non-panel GP organisms. All GN isolates were correctly identified on Verigene. Median time-to-result was 2.9 hours (IQR 2.6, 3.2) and 44.4 hours (IQR: 35.4, 52.5) for Verigene and final susceptibilities, respectively. There was a statistically significant time savings of 41.5 hours (CI: 29.8-53.2) for identification and detection of resistance markers (p < 0.0001). CONCLUSION: Verigene concordance at our institution aligns with results from previously published studies and can be considered a reliable clinical decision-support tool.