Abstract
Introduction
D
Enterococci have intrinsic resistance to several antibiotics, such as beta-lactams, clindamycins, fluoroquinolones, trimethoprim–sulfamethoxazoles, aminoglycosides, 5 and species with an extended resistance to normally effective antibiotics have been found. 28 The bacteria can transfer antibiotic resistance to other enterococci as well as to other species of bacteria. As an example, transfer of vancomycin resistance to the clinically much more feared staphylococci has been confirmed with the first case described in 2002 in the United States. 24
Resistance to vancomycin is caused by a series of genes (classified by the prefix van) that encode an enzyme important for the cell wall. Until now, eight different kinds of resistance genes—vanA, B, C, D, E, G, L, and M—are known. 9 Some enterococci, such as Enterococcus gallinarum, Enterococcus casseliflavus, and Enterococcus flavescens, are intrinsically resistant to vancomycin due to the chromosomal presence of vanC. VanA, B, L, and M can be transferred between bacteria, while vanC, D, and E are nontransferable. Due to the high transferability and high clinical importance of vanA and vanB genes, we have focused on the presence of these genes.
Enterococci that have acquired resistance to vancomycin (vancomycin-resistant enterococci [VRE]) are responsible for nosocomial infections and are associated with a high morbidity and mortality in humans.6,20 By definition, VRE include all enterococci that exhibit resistance to vancomycin, but the expression is in clinical practice usually limited to the two genotypes vanA and vanB of Enterococcus faecium and Enterococcus faecalis. These strains possess acquired and transferable vancomycin resistance biomarkers, which account for 80–90% (E. faecium) and 5–10% (E. faecalis) of the clinically relevant infections in humans. 5
The prevalence of VRE varies across the world, but VRE are found in hospitals and communities in most countries today. For many years now, a low prevalence of VRE has been reported in Swedish humans. On the other hand, the occurrence of VRE in Swedish broilers has been reported to be 40% 17 and 79% in sludge samples from wastewater sewage plants. 22 Between 2007 and 2010, a number of VRE outbreaks were reported in hospitals in Stockholm and in the two counties Västmanland and Halland.
In the United States, colonization and infection with VRE mainly affect patients in hospital settings, while the reservoir in European countries seems to be in the community and only to a lesser extent in hospitals.5,26 Previous studies on healthy community-dwelling individuals have demonstrated an asymptomatic carriage rate of vanA or vanB ranging between 2% and 28%3,7,23 and between 0.3% and 2.9% among hospitalized patients. 23 In 2009, VRE were detected in 11.6% of fecal samples from healthy children in Portugal. 11 VRE can spread between individuals in the same household or in shared hospital rooms.4,31
The aim of this study was to determine the prevalence of VRE among young, healthy, Swedish preschool children and to investigate whether this group constitutes a reservoir.
Materials and Methods
Sample collection
Fecal samples from 313 individual diapers were collected during September and October 2010.
Subjects
All 63 municipal preschools within the central parts of Uppsala, a city with ∼140,000 inhabitants, were invited to participate in the study. Approximately 58% of the children between 1–5 years in Uppsala were thereby covered. The remaining children attend private preschools, other child care providers, or stay at home. A total of 46 out of 63 preschools agreed to take part in the study (the maximum number of diapers collected from a single preschool was 16, the minimum 2, and the average ∼7). Information about the study was sent to the directors of the participating preschools for further distribution to the staff and parents. According to the Regional Ethics Committee, no ethical consent was needed since the samples were classified as biological waste and none of the participants could be identified.
Sample handling
In preschools where staff and parents gave their consent, one diaper per child, marked with the child's age, was collected immediately after defecation. All diapers were transported to the Department of Clinical Microbiology, Uppsala University. Fecal samples from the diapers were collected by sterile swabs. The swabs were immediately inoculated with the sample into a 1 ml bile azide esculin broth and stored at −70° for analysis.
Screening for VRE was performed with a selective bile azide esculin broth, supplemented with 4 mg/L vancomycin and 60 mg/L aztreonam (ICN Biomedicals, Inc. Aurora, OH), and polymerase chain reaction (PCR) for vanA and vanB according to the Swedish Institute for Infectious Disease Control 9 and the methods described previously. 19 Briefly, this real-time PCR protocol was used to detect the vanA and vanB gene. Amplification was conducted in a total volume of 20 μl containing 12.1 μl of PCR-water, 4 μl of LC480 M-Mix (2×), 0.4 μl of VanA F (10 μM) (5′-CGGCAAGACAATATGAC AGCAA-3′), 0.4 μl of VanA R (10 μM) (5′-TCAGTACAAT GCGGCCGTTA-3′), 0.4 μl of VanB F (10 μM) (5′-GGGAGG ATGGTGCGATACA-3′), 0.4 μl of VanB R (10 μM) (5′-CCG AAATCGCTTGCTCAA-3′), 0.15 μl of VanA prob (10 μM) (5′-HEX-CAGTTATAACCGTTCCCGCAGACCTT-BHQ1-3′), 0.15 μl of VanB prob (10 μM) (5′-FAM-CTTTGTGAAG CCGGCACGGTCAGGTT-BHQ1-3′), and 2 μl of samples. The following cycling parameters were used in the PCR run: 95° for 5 min (95° for 10 sec, 60° for 30 sec) for 45 cycles. Finally, the amplified PCR product was analyzed.
Results, Discussion, and Conclusion
All together 313 fecal samples were analyzed. None of them demonstrated a color change after testing in the bile azide esculin broth and, somewhat surprisingly, neither vanA nor vanB could be demonstrated with PCR screening. False-positive vanB results with PCR assays on stool samples are not uncommon1,10,27 due to nonviable enterococci or, more commonly, intestinal anaerobic commensal bacteria, such as Clostridium sp, carrying vanB genes. These bacteria coexist in the normal intestinal flora and may contribute to VRE emergence by transfer of resistance to enterococci. 10 Unlike the results from healthy preschool children in Australia in 2006, where Graham et al. demonstrated that 27% of the children carried vanB genes in nonenterococcal bacteria, 10 we did not detect any vanB-positive samples in our PCR assay. Hence, no culture confirmation of samples was performed.
VRE were first described in 1988 in the United Kingdom and France.15,29 Since then, our knowledge about these vancomycin-resistant strains has increased steadily as the resistance has dispersed over the world. Regardless of this, the prevalence of VRE has decreased in Europe during recent years, especially in Greece, Slovenia, Sweden, and the United Kingdom. 21 Germany constitutes an exception, with a rising trend. Ireland has the highest occurrence in Europe, being the only European country with a prevalence of more than 25% of VRE among invasive strains. 21 The Scandinavian countries have for a long time experienced a low VRE prevalence among humans. Still, outbreaks have occurred in Finland as well as Sweden in the last years,9,18 whereas as yet, no major outbreak has taken place in Norway or Denmark.25,30 No VRE were reported in healthy Swedish individuals in 1996–97 18 and only 0.4% of patients with diarrhea carried VRE after travelling abroad. VRE are mandatorily notifiable in Sweden since 2000. In a recent study, no VRE were found among patients in nursing homes in Sweden. 2 Today, 1–2% of invasive enterococci reported in Sweden are VRE, according to data from EARS-net. 32 A relatively high prevalence of VRE has been demonstrated in Swedish waste water treatment plants,12,22 and VRE have also been found to be common among livestock—particularly in broiler farms—in Sweden and in the rest of Europe.13,16,22 In a previous study, we demonstrated that 2.9% of healthy preschool children carried extended-spectrum beta-lactamases (ESBL) and that identical strains probably spread among children attending the same preschool. 14 The presence of ESBL-producing bacteria among healthy preschool children increased our suspicion about the possible carriage and spread of VRE as well, like in Portugal. 11 We can conclude that this was not the case: no VRE were found among the same population of children, and hence, there were no signs of spread of resistance in the preschools. Our findings are consistent with the decline in VRE prevalence in Sweden during recent years and do not support the suspicion that preschool children are a reservoir for VRE in Sweden.
Footnotes
Acknowledgments
We thank the staff at the participating preschools for collecting the diapers. Without their help this study would not have been possible. We send a cordial thank you to Dr. Tryggve Nevéus, whose never failing sensible and perceptive remarks have made this article both more distinct and comprehensible.
Disclosure Statement
Drs. Kaarme and Olsen, Mr. Hasan, and Mr. Rashid have no conflicts of interest and no financial relationships relevant to this article to disclose.
This study was financially supported by the Medical Faculty, Uppsala University, the Gillbergska, Olle Engkvist, Karin Korsner, Marcus Borgström, Lundells PO, and Bergmark Foundations.