Performance of polymyxin B agar-based tests among carbapenem-resistant Enterobacterales

Therapeutic options for infections caused by Carbapenem-resistant Enterobacterales (CRE) are restricted and include polymyxins-centred schemes. Evaluation of in vitro susceptibility is difficult and time consuming. Agar-based methodologies are an alternative to broth microdilution (BMD) and we aimed to evaluate the accuracy of those methods among Enterobacterales. A total of 137 non-duplicated CRE were subjected to polymyxin B BMD, agar screening test (Mueller Hinton plates containing 3 µg ml^-1 of polymyxin B) and agar dilution (antibiotic serially diluted 0·25-64 µg ml^-1 ). CRE of 42·3% were resistant to polymyxin B (MICs range: 0·25->64 µg ml^-1 ) and 16·8% presented borderline MICs. Sensitivity, specificity, PPV and NPV were 86·2, 98·7, 98 and 90·7% for screening test and 86·2, 97·5, 96·1 and 90·6% for agar dilution. ME was 0·73 and 1·5% for screening and agar dilution respectively; VME was 5·8% for both techniques. In general, agar-based methods had a good performance. As far as we know, this is the first study to propose an agar screening test using polymyxin B instead of colistin.