BackgroundResistance to carbapenem antibiotics is emerging worldwide among Enterobacteriaceae. To prevent hospital transmission due to unnoticed carriage of carbapenemase producing micro-organisms in newly admitted patients, or follow-up of patients in an outbreak setting, a molecular screening method was developed for detection of the most prevalent carbapenemase genes; blaOXA-48, blaVIM, blaIMP, blaNDM and blaKPC.MethodsA real-time multiplex PCR assay was evaluated using a collection of 86 Gram negative isolates, including 62 carbapenemase producers. Seven different laboratories carried out this method and used the assay for detection of the carbapenemase genes on a selection of 20 isolates.ResultsBoth sensitivity and specificity of the multiplex PCR assay was 100%, as established by results on the strain collection and the inter-laboratory comparisons.ConclusionsIn this study, we present a multiplex real-time PCR that is a robust, reliable and rapid method for the detection of the most prevalent carbapenemases blaOXA-48, blaVIM, blaIMP, blaNDM and blaKPC, and is suitable for screening of broth cultured rectal swabs and for identification of carbapenemase genes in cultures.

BackgroundThe objective of our study was the development of a semi-quantitative real-time PCR to detect uropathogens. Two multiplex PCR reactions were designed to detect Escherichia coli, Klebsiella spp., Enterobacter spp., Citrobacter spp., Proteus mirabilis, Enterococcus faecalis, and Pseudomonas aeruginosa. 16S based PCR was performed in parallel to detect Gram-positive and Gram-negative bacteria. Firstly to identify non-targeted agents of infection in the same urine specimen, and secondly to quantify background flora. The method was evaluated in comparison with standard bacterial culture, and a commercial PCR kit for detection of uropathogens.FindingsAnalysis with a known panel of 116 clinical isolates yielded a PCR specificity of 100%. Analysis of urine specimens from 211 patients revealed a high correlation of PCR Cq values with both culture positivity and quantity. Concordance between PCR and culture was 98% when both methods yielded results. PCR was found to be more sensitive than culture. With a cut-off Cq value of 33, the negative predictive value of PCR was 94%. The 16S PCR confirmed most results. One specimen was positive by 16S PCR suggesting another cause of infection not detected by the specific PCR assays.ConclusionWe conclude that it is feasible to detect and identify uropathogens by multiplex real-time PCR assay.

bIn an outbreak setting, we screened 16,296 samples from 3,644 patients by PCR for the presence of bla OXA-48 , bla VIM , bla IMP , bla NDM , and bla KPC . The bla OXA-48 gene was found in samples from 43 patients infected with 9 different species of Enterobacteriaceae. Five patients had Pseudomonas aeruginosa isolates containing bla VIM . The negative predictive value of screening was 100%, and the positive predictive value was 86%. W e describe here the results of screening for the most prevalent carbapenemase genes coding for VIM, IMP, NDM, KPC, and OXA-48 by a real-time multiplex PCR in the aftermath of an outbreak of infection with Klebsiella pneumoniae isolates containing bla (1). The mean number of samples per patient was 4.5 (range, 1 to 20). Swabs were collected from discharged contact patients (n ϭ 3,200), from weekly intensive care unit (ICU) surveillance (n ϭ 351), and from patients admitted or readmitted and/or considered otherwise at risk (n ϭ 361). Rectal swabs were incubated overnight in brain heart infusion broth containing 0.25 mg/liter ertapenem and, if positive or inhibited in the PCR screen, plated on MacConkey no. 3 agar (Oxoid) supplemented with ertapenem (0.25 mg/liter). The PCR protocols, primer and probe sequences, and their validation have been described previously (2). Positive controls in the PCR screen consisted of P. aeruginosa containing bla VIM2 or bla IMP18 and K. pneumoniae containing bla OXA-48 , bla NDM1 , or bla KPC1 . Antibiotic susceptibility testing was performed using the Vitek 2 system, and MICs for ertapenem and meropenem were determined by Etest (both testing systems from bioMérieux, Marcy l'Etoile, France). Extended-spectrum ␤-lactamase (ESBL) confirmatory tests were performed by disk diffusion (3).In total, 186 rectal swabs from 87 patients were positive by PCR screen, 17 showed inhibition, and 16,093 samples were negative. Screen-positive and inhibited samples were cultured on MacConkey agar plates supplemented with ertapenem, in addition to 167 screen-negative samples as a control group, totaling 370 samples. Of these, 318 cultures became positive, growing, respectively, Escherichia coli (n ϭ 170), K. pneumoniae (n ϭ 81), Morganella morganii (n ϭ 16), P. aeruginosa (n ϭ 15), Proteus mirabilis (n ϭ 14), Shewanella spp. (n ϭ 5), Enterobacter cloacae (n ϭ 5), Citrobacter spp. (n ϭ 4), Klebsiella oxytoca (n ϭ 4), Serratia marcescens (n ϭ 2), and 1 culture each of Providencia stuartii and Kluyvera cryocrescens. Fifty-one cultures remained negative, including 17 of inhibited samples and 34 of screen-positive samples.PCR identified bla For 158/167 bla OXA-48 -negative isolates, antibiotic susceptibilities ranged from Յ0.25 to Ն16 mg/liter for imipenem and Յ0.25 to 1 mg/liter for meropenem, while 93/133 bla OXA-48 -positive isolates, including all different species, showed ranges of 0.5 to Ն16 mg/liter for imipenem and Յ0.25 to 4 mg/liter for meropenem.Using the Vitek 2 system and applying CLSI recommended clinical breakpoints for resistance (3), 27 E. coli and 5 K. pn...