Abstract
Carbapenemase-producing Enterobacteriaceae (CPE) are a significant threat to public health. In 2015, CDC revised the surveillance definition for CPE to include all Enterobacteriaceae resistant to any carbapenem tested. However, this definition is associated with poor specificity. We evaluated the performance of this definition, compared to carbapenemase PCR, for a collection of 125 Enterobacteriaceae We also investigated the impact of ancillary testing for carbapenemase of isolates that met the CDC CPE surveillance definition. The two ancillary tests evaluated were the Xpert Carba-R assay, a molecular test, and the carbapenem inactivation method (CIM). Two variables were evaluated for the CIM: suspension of organisms in double-distilled water (ddH(2)O) versus tryptic soy broth (TSB) to incubate disks, and incubation of plates for 6 h versus 18 to 20 h. The sensitivity and specificity of the Carba-R assay were 100% compared to the results of in-house PCR. The sensitivities of the CIM performed with TSB were 94.6% when read at 6 h and 97.7% when read at 18 to 20 h; the sensitivities with ddH(2)O were 88.0% when read at 6 h and 93.0% when incubated for 18 to 20 h. The specificity was 100% for all variables tested. Without ancillary testing, the sensitivity of the CDC definition was 98.9% for CPE, and the specificity was 6.1%. Testing isolates that screened positive by the CDC definition with the Xpert Carba-R did not change the sensitivity, and it improved the specificity to 100%. Similarly, the use of the CIM (TSB and 18 to 20 h of incubation) to confirm screen-positive isolates resulted in a sensitivity of 95.6% and specificity of 100%.