Prevalence, risk factors and molecular epidemiology of highly resistant gram negative rods in hospitalized patients in the Dutch region Kennemerland

Dennis Souverein, Sjoerd M. Euser, Bjorn L. Herpers, Bram Diederen, Patricia Houtman, Marina van Seventer, Ingeborg van Ess, Jan Kluytmans, John W. A. Rossen, Jeroen W. Den Boer

This paper describes (1) the Highly Resistant Gram Negative Rod (HR-GNR) prevalence rate, (2) their genotypes, acquired resistance genes and (3) associated risk factors of HR-GNR colonization among the hospitalized population in the Dutch region Kennemerland. Between 1 October 2013 and 31 March 2014, cross-sectional prevalence measurements were performed in three regional hospitals as part of each hospitals infection control program. Rectal swabs were analyzed at the Regional Public Health Laboratory Kennemerland by direct culturing. Genotypes and acquired resistance genes of positive isolates were determined using Whole Genome Sequencing with the MiSeq instrument (Illumina). Association between several independent variables and HR-GNR positivity was examined using logistic regression models. Out of 427 patients, 24 HR-GNR positive isolates were recovered from 22 patients, resulting in a regional HR-GNR colonization prevalence (95 % CI) of 5.2 % (3.6-7.9). Of these 22 positive patients, 15 were Extended Spectrum Beta-Lactamase (ESBL) positive (3.5 % (2.1-5.7)), 7 patients were positive for a Fluoroquinolones and Aminoglycosides (Q&A) resistant Enterobacteriaceae (1.6 % (0.8-3.3)) and from one patient (0.2 % (0-1.3)) a Stenotrophomonas maltophilia resistant towards co-trimoxazole was isolated. No carbapenemase producing Enterobacteriaceae (CPE), multi-resistant Acinetobacter species or multi-resistant Pseudomonas aeruginosa were isolated. The ESBL genes found were bla CTX-M-1 (n = 4, 25.0 %), bla CTX-M-15 (n = 3, 18.8 %), bla CTX-M-27 (n = 2, 12.5 %), bla CTX-M-14b (n = 2, 12.5 %), bla CTX-M-9 (n = 2, 12.5 %), bla CTX-M-14 (n = 1, 6.3 %), bla CTX-M-3 (n = 1, 6.3 %), bla SHV-11 (n = 1, 6.3 %) and bla SHV-12 (n = 1, 6.3 %). Being known HR-GNR positive in the past was the only significant associated risk factor for HR-GNR positivity, odds ratio (95 % CI): 7.32 (1.82-29.35), p-value = 0.005. Similar ESBL prevalence rates and genotypes (3.5 %) were found in comparison to other Dutch studies. When previously HR-GNR positive patients are readmitted, they should be screened for HR-GNR colonization since colonization with GR-GNRs could be prolonged. We recommend for future studies to include all defined HR-GNRs in addition to ESBLs in prevalence studies, in order to obtain a more comprehensive overview of colonization with HR-GNRs.