Comparison of four low-cost carbapenemase detection tests and a proposal of an algorithm for early detection of carbapenemase-producing Enterobacteriaceae in resource-limited settings

Abstract

Rapidly progressing antibiotic resistance is a great challenge in therapy. In particular, the infections caused by carbapenem-resistant Enterobacteriaceae (CRE) are exceedingly difficult to treat. Carbapenemase production is the predominant mechanism of resistance in CRE. Early and accurate identification of carbapenemase-producing carbapenem-resistant Enterobacteriaceae (CP-CRE) is extremely important for the treatment and prevention of such infections. In the present study, four phenotypic carbapenemase detection tests were compared and an algorithm was developed for rapid and cost-effective identification of CP-CRE. A total of 117 Enterobacteriaceae (54 CP-CRE, 3 non-CP-CRE, and 60 non-CRE) isolates were tested for carbapenemase production using modified Hodge test (MHT), modified carbapenem inactivation method (mCIM), Carba NP test (CNPt), and CNPt-direct test. The overall sensitivity/specificity values were 90.7%/92.1% for MHT, 100%/100% for mCIM, 75.9%/100% for CNPt, and 83.3%/100% for CNPt-direct. OXA-48-like enzymes were detected with 93.2% sensitivity by MHT and >77.3% sensitivity by two Carba NP tests. MHT could only detect half of the NDM carbapenemase producers. CNPt-direct exhibited enhanced sensitivity compared to CNPt (100% vs 25%) for detection of NDM producers. Considering these findings we propose CNPt-direct as the first test followed by mCIM for rapid detection of CP-CRE. With this algorithm >80% of the CP-CRE could be detected within 24 hours from the time the sample is received and 100% CP-CRE could be detected in day two. In conclusion, mCIM was the most sensitive assay for the identification of CP-CRE. CNPt-direct performed better than CNPt. An algorithm consisting CNPt-direct and mCIM allows rapid and reliable detection of carbapenemase production in resource-limited settings.