A Croxatto1, A T Coste2, T Pillonel2, C Bertelli2, G Greub2, G Prod'hom2. 1. Institute of Microbiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland. Electronic address: antony.croxatto@chuv.ch. 2. Institute of Microbiology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland.
Abstract
OBJECTIVES: Becton-Dickinson recently developed the Phoenix™ CPO (carbapenemase-producing organism) Detect Test, a growth-based test embedded in Gram-negative (GN) panels for the detection and confirmation of bacteria producing class A, B and D carbapenemases. This study aimed to (a) determine the performance of the CPO test, and (b) assess its added value in routine diagnostic workflows. METHODS: The performance of the BD Phoenix CPO test was analysed retrospectively on a collection of 185 molecularly characterized strains, including 92 CPOs, and prospectively on 135 and 160 routine isolates with and without CPO suspicion, respectively. RESULTS: In the retrospective study the CPO test exhibited 92.4% accuracy (95%CI 87.6-95.8), 97.8% sensitivity (95%CI 92.4-99.7) and 87.1% specificity (95%CI 78.6-93.2) for carbapenemase detection. The CPO test provided a classification to class A, B, and D for 81.3% of detected carbapenemases with 94.6% accuracy (95%CI 86.7-98.5). In the prospective study the CPO test detection performance showed 77.8% accuracy (95%CI 68.8-84.5), 100% sensitivity (95%CI 91.2-100) and 67.8% specificity (95%CI 57.3-77.1) with 135 CPO-suspicious isolates and 98.8% accuracy and specificity (95%CI 95.6-99.9) with 160 non-CPO-suspicious isolates. Compared to routine testing, the implementation of the CPO test allowed a mean reduction of 21.3 h (95%CI 17.6-25) in turnaround time, 16.8 min (95%CI 13.4-20.2) in hands-on time, and 20.6 CHF (95%CI 16.5-24.8) in costs. CONCLUSIONS: The CPO test is reliable for the detection of CPO with a high sensitivity. However, the relatively low detection specificity required the use of additional confirmatory methods. The carbapenemase classification accuracy is robust in providing preliminary results before molecular characterization. Finally, the implementation of the test in routine workflows allowed a significant reduction in turnaround time, hands-on time and cost compared to the conventional approach.
OBJECTIVES: Becton-Dickinson recently developed the Phoenix™ CPO (carbapenemase-producing organism) Detect Test, a growth-based test embedded in Gram-negative (GN) panels for the detection and confirmation of bacteria producing class A, B and D carbapenemases. This study aimed to (a) determine the performance of the CPO test, and (b) assess its added value in routine diagnostic workflows. METHODS: The performance of the BD Phoenix CPO test was analysed retrospectively on a collection of 185 molecularly characterized strains, including 92 CPOs, and prospectively on 135 and 160 routine isolates with and without CPO suspicion, respectively. RESULTS: In the retrospective study the CPO test exhibited 92.4% accuracy (95%CI 87.6-95.8), 97.8% sensitivity (95%CI 92.4-99.7) and 87.1% specificity (95%CI 78.6-93.2) for carbapenemase detection. The CPO test provided a classification to class A, B, and D for 81.3% of detected carbapenemases with 94.6% accuracy (95%CI 86.7-98.5). In the prospective study the CPO test detection performance showed 77.8% accuracy (95%CI 68.8-84.5), 100% sensitivity (95%CI 91.2-100) and 67.8% specificity (95%CI 57.3-77.1) with 135 CPO-suspicious isolates and 98.8% accuracy and specificity (95%CI 95.6-99.9) with 160 non-CPO-suspicious isolates. Compared to routine testing, the implementation of the CPO test allowed a mean reduction of 21.3 h (95%CI 17.6-25) in turnaround time, 16.8 min (95%CI 13.4-20.2) in hands-on time, and 20.6 CHF (95%CI 16.5-24.8) in costs. CONCLUSIONS: The CPO test is reliable for the detection of CPO with a high sensitivity. However, the relatively low detection specificity required the use of additional confirmatory methods. The carbapenemase classification accuracy is robust in providing preliminary results before molecular characterization. Finally, the implementation of the test in routine workflows allowed a significant reduction in turnaround time, hands-on time and cost compared to the conventional approach.
Authors: Ahmed S Keshta; Nazik Elamin; Mohammad Rubayet Hasan; Andrés Pérez-López; Diane Roscoe; Patrick Tang; Mohammed Suleiman Journal: Microbiol Spectr Date: 2021-12-08