@article{mbs:/content/journal/jmm/10.1099/jmm.0.000931, author = "Toner, Geoff and Russell, Clark D. and Hamilton, Fiona and Templeton, Kate and Laurenson, Ian F.", title = "Phenotypic and molecular detection methods for carbapenemase-producing organisms and their clinical significance at two Scottish tertiary care hospitals", journal= "Journal of Medical Microbiology", year = "2019", volume = "68", number = "4", pages = "560-565", doi = "https://doi.org/10.1099/jmm.0.000931", url = "https://www.microbiologyresearch.org/content/journal/jmm/10.1099/jmm.0.000931", publisher = "Microbiology Society", issn = "1473-5644", type = "Journal Article", keywords = "molecular diagnostics", keywords = "carbapenemase-producing organisms", keywords = "antimicrobial resistance", keywords = "carbapenem-producing Enterobacteriaceae", keywords = "carbapenem resistance", abstract = " Purpose . This study evaluated in-house PCR testing for local identification of bacteria carrying the major carbapenemase genes (bla OXA-48-like, bla VIM, bla NDM, bla KPC and bla IMP). Methodology . Carbapenemase-producing organisms (CPOs) isolated from patients managed in two tertiary care hospitals in Scotland from September 2014–January 2017 were investigated. A combination of chromogenic screening agar (ChromID CARBA SMART), a carbapenem hydrolysis test (Rapidec Carba NP) and in-house real-time PCR for the bla OXA-48-like, bla VIM, bla NDM, bla KPC and bla IMP genes were utilized. All isolates were sent to the AMRHAI reference unit for confirmatory testing. Results . During the 29-month study period 39 CPO were isolated from 34 patients. The average turnaround time for a workflow involving phenotypic and molecular testing was 4.2 days. PCR had a sensitivity and specificity of 100 %. The most common carbapenemase genes were bla OXA-48-like (31%), bla VIM (23%) and bla NDM (20%). Resistance to antimicrobials other than beta-lactams was common; the most active agents were colistin, amikacin and fosfomycin. Twenty-seven patients were considered to be colonized (although CPO detection influenced empiric antimicrobials in five) and a CPO was implicated in infection in seven patients (bacteraemia in immunocompromised patients, n=2; surgical site infections, n=2; osteomyelitis in a patient with diabetes mellitus, n=1; and urinary tract infections, n=2). All patients survived infection. Conclusion . In a lowincidence setting we demonstrate the efficacy of a combined local laboratory workflow for rapid detection of CPOs, incorporating phenotypic and molecular testing. In 7/34 patients the CPO was implicated as a pathogen and detection influenced antimicrobial decision-making in five colonized patients.", }