Performance of distinct phenotypic methods for carbapenemase detection: The influence of culture media

We evaluated the performance of five phenotypic tests [Modified Hodge Test (MHT); combined-disk test (CDT) using phenylboronic acid, EDTA, and cloxacillin; CarbaNP and CarbAcinetoNP; Blue-Carba, Carbapenembac™ and Carbapenembac Metallo™] for carbapenemase detection in Gram-negative bacilli (GNB). A total of 73 carbapenemase producers and 27 non-carbapenemase producers were tested. All GNB were subcultured onto Müeller-Hinton agar (MHA), MacConkey agar (MAC), and sheep blood agar (SBA). High sensitivity (100%) and specificity (100%) was observed for MHA using CarbaNP, Blue-Carba, and Carbapenembac™. The sensitivity and specificity of CarbaNP (98.6%/100%), Blue-Carba (97.1%/91.0%), and Carbapenembac™ (100%/96.5%) were slightly lower for SBA. In contrast, unacceptable sensitivity rates of CarbaNP (71.1%) and Blue-Carba (66.6%), but not Carbapenembac™ (97.3%), were observed for MAC. The colorimetric methods showed high sensitivity and specificity to detect carbapenemase production from isolates grown on MHA or SBA. However, colonies obtained from MAC must not be tested for carbapenemase detection by colorimetric methods. •A total of 100 Gram-negative bacilli (GNB) were simultaneously subcultured onto Müeller-Hinton agar (MHA), MacConkey agar (MAC), and sheep blood agar (SBA), and tested against five distinct phenotypic tests for carbapenemase detection;•Carbapenembac™ was the only phenotypic test able to detect 100% of carbapenemase-producing isolates regardless of the media where colonies were grown.•The colorimetric methods CarbaNP and Blue-Carba showed excellent sensitivity and specificity to detect carbapenemase production from isolates grown on MHA or SBA;•MAC must be avoid for carbapenemase detection by colorimetric methods (CarbaNP and Blue-Carba).