The monitoring of infection by glycopeptide-resistant enterococci (GRE) is one of the main elements of hospital hygiene policy. It involves systematic rectal swabs in clinics at risk (asymptomatic carriage).We compare two GRE screening methods and evaluate a new kit associating multiplex PCR and hybridization (Génotype(®) Enterococcus, Hain Lifescience) on a panel of 448 samples collected over a 4-month period.The first method is based on direct inoculation of the sample; the second one involves a preliminary enrichment phase followed by molecular diagnosis allowing the identification of species of enterococci as well as glycopeptide resistance genes.All the resistant strains were isolated using the enrichment technique. The incidence of GRE (VanA) carriage was 0,55% (two out of 362 patients, two out of 448 isolates) with two Enterococcus faecium VanA. Six Enterococcus gallinarum VanC1 and two Enterococcus casseliflavus VanC2/C3 were also isolated and identified. The main clinics concerned are intensive care and hematology. The two patients with E. faecium VanA had been previously given glycopeptides for 10 days. For three strains, the molecular method allowed to correct prior erroneous results based on rapid identification (RapidID32Strep V2.0).The method using direct samples inoculation underestimates real incidence of GRE carriage. The performances of Génotype(®) Enterococcus molecular method, evaluated for other parameters using reference strains and DNA sequencing, offer new possibilities applicable to routine laboratory.