Molecular epidemiology of coagulase-negative...

Molecular epidemiology of coagulase-negative staphylococci in

hospitals and in the community

Micael Widerström

Department of Clinical Microbiology, Clinical Bacteriology and Infectious Diseases

Umeå University Umeå 2010

1

Copyright © Micael Widerström ISBN: 978-91-7459-020-3 ISSN: 0346-6612 New Series No 1354 Cover: PFGE patterns of MRSE genotypes edited by Bertil Nordström Printed by: Print & Media, Umeå University, Umeå, Sweden, 2010

In memory of my father Wisdom is not a product of schooling but of the lifelong attempt to acquire it Albert Einstein

Table of Contents ABSTRACT..................................................................................................... 5 SAMMANFATTNING AV AVHANDLINGEN (Abstract in Swedish).…………… 7 ORIGINAL PAPERS................................................................................ 9 ABBREVIATIONS................................................................................. 10 DEFINITIONS...................................................................................... 11 INTRODUCTION

Background..................................................................................... 12 Molecular typing methods for CoNS ............................................. 15 Pathogenesis...................................................................................... 24 Clinical presentation of CoNS infection......................................... 25 Antimicrobial resistance in CoNS.................................................. 29 Molecular epidemiology of CoNS...................................................... 33

AIM.................................................................................................... 34 STUDY DESIGN AND BACTERIAL ISOLATES........................................... 35 METHOD............................................................................................ 36 RESULTS AND DISCUSSION.................................................................. 39

Distribution of different species and genotypes of methicillin- resistant CoNS among patients treated at the Östersund Hospital (Study I)................................................................................ 39 Comparison of multidrug-resistant genotypes of S. epidermidis among patients treated in the Umeå University Hospital and the Östersund Hospital (Study I)......................................................... 42 Investigation of multidrug-resistant genotypes of S. epidermidis from eleven hospitals in northern Europe (Study II)........................ 43 Evaluation of the molecular epidemiology and the frequency of antibiotic resistance among community-associated S. epidermidis (Study III)............................................................................................ 45 Study of the molecular epidemiology of S. saprophyticus associated with UTI in women (Study IV).......................................... 48

GENERAL DISCUSSION ........................................................................ 51 REFLECTION AND FURTHER STUDIES.................................................. 55 CONCLUSIONS.................................................................................... 56 ACKNOWLEDGEMENTS....................................................................... 57 REFERENCE....................................................................................... 59

Organisation Umeå University Department of Clinical Microbiology SE-901 87 Umeå Author Micael Widerström Title Molecular epidemiology of coagulase-negative staphylococci with focus on hospital-associated transmission of multidrug resistant S. epidermidis Background Coagulase-negative staphylococci (CoNS) and in particular Staphylococcus epidermidis have emerged as major pathogens primarily causing nosocomial infections in patients with indwelling medical devices. These infections are often caused by multidrug-resistant strains of S. epidermidis (MDRSE). Other clinical entities due to CoNS are lower urinary tract infections (UTI) in women and native valve endocarditis. The purpose of this work was to investigate the frequency of antibiotic resistance and the molecular epidemiology of both hospital and community-associated isolates of S. epidermidis in order to examine if certain clones are related to MDRSE infections. Furthermore, we aimed to explore if specific clones of S. saprophyticus are associated with UTI in women. Methods A total of 359 hospital-associated methicillin-resistant isolates of CoNS obtained from 11 hospitals in northern Europe and 223 community-associated staphylococcal isolates were examined. Furthermore, 126 isolates of S. saprophyticus isolated from women with uncomplicated UTI from five different locations in northern Europe were analyzed. Pulsed-field gel electrophoresis (PFGE) was used for genotyping. Additionally, some of the S. epidermidis isolates were analyzed with multilocus sequence typing (MLST). Antibiotic susceptibility was determined for all isolates by the disc diffusion test. Results 293 of the 359 (82%) hospital-associated and 124 of the 223 (56%) community-associated isolates belonged to the species S. epidermidis. Among the hospital-associated S. epidermidis isolates, two dominating PFGE types (type A and B) were distinguished, comprising 78 (27%) and 51 (17%) isolates, respectively. Type A, which was detected in a Norwegian and eight Swedish hospitals, corresponded with a novel sequence type (ST215). Type B was discovered in a German, a Danish and seven Swedish hospitals and

5

corresponded with ST2. In contrast, community-associated isolates of S. epidermidis were genetically extremely diverse with no predominating genotype, and showed a low rate of antibiotic resistance; only two (1.6%) methicillin-resistant strains were detected.

Among 126 analyzed isolates of S. saprophyticus, 47 different PFGE profiles were identified. Several clusters of genetically highly related isolates were detected among isolates obtained from different locations and periods of time. Conclusion We have demonstrated the occurrence, persistence and potential dissemination of two multidrug-resistant S. epidermidis (MDRSE) genotypes, including a novel sequence type (ST215), within hospitals in northern Europe. Community-associated isolates of S. epidermidis showed a low rate of methicillin-resistance and were genetically heterogeneous. These results indicate that MDRSE by large are confined to the hospital setting in our region. Moreover, although the S. saprophyticus population was quite heterogeneous, indistinguishable isolates of S. saprophyticus causing lower UTI in women were identified in different countries 11 years apart, indicating the persistence and geographical spread of some clones of S. saprophyticus.

Key words: Staphylococcus epidermidis; Staphylococcus saprophyticus; Electrophoresis, Gel, Pulsed-Field; genetic diversity; Methicillin-Resistance; UTI; Drug Resistance, Multiple, Bacterial; Epidemiology, Molecular; Cross Infection; Molecular Sequence Data.

6

POPULÄRVETENSKAPLIG SAMMANFATTNING (ABSTRACT IN SWEDISH)

Bakgrund: Infektioner som uppkommer i samband med sjukhusvård, så kallade vårdrelaterade infektioner, orsakar betydande lidande för de drabbade och stora kostnader för sjukvården. Koagulasnegativa stafylokocker (CoNS) är bakterier som tillhör den skyddande hudfloran och som sällan orsakar infektioner hos friska människor. CoNS, och speciellt den vanligaste förekommande arten Staphylococcus epidermidis, har dock en förmåga att vidhäfta och infektera inopererat främmande material och är därför en vanlig orsak till vårdrelaterade infektioner. Trots att S. epidermidis är en viktig orsak till infektioner inom sjukvården saknas idag grundläggande kunskap om dess förekomst, spridningsvägar och om det finns speciella stammar som har större förmåga till spridning, resistensutveckling och benägenhet att förorsaka infektioner.

S. epidermidis har liksom andra CoNS en förmåga att bilda en så kallad biofilm kring främmande material, såsom katetrar, ledproteser, hjärtklaffsproteser och kärlproteser. Biofilmen försvårar och omöjliggör ofta utläkning av infektionen om inte det främmande materialet byts ut. Dessutom har S. epidermidis en förmåga att bli motståndskraftiga mot flertalet antibiotika. Dessa så kallade multiresistenta S. epidermidis kan spridas mellan patienter och personal samt har påvisats i sjukhusmiljön där förekomsten av resistenta S. epidermidis ofta avspeglar antibiotikaförbrukningen på den enskilda avdelningen.

En annan art av CoNS, Staphylococcus saprophyticus, är en vanlig orsak till urinvägsinfektion företrädesvis hos yngre och medelålders kvinnor. Det är oklart hur dessa sprids och okänt om det är vissa stammar av S. saprophyticus som ger upphov till dessa urinvägsinfektioner.

Genetisk typning av bakterier är avgörande för att kunna påvisa förekomst och spridning av vissa stammar. Pulsfält-gel-elektrofores (PFGE) är en användbar metod för genotypning, dvs. att genetisk särskilja bakterier inom samma art. PFGE avbildar en bakteries arvsmassa som ett fingeravtryck (”fingerprint”) och genom att jämföra dessa avtryck från flera bakterier kan genetiskt identiska eller närbesläktade isolat, så kallade genotyper, påvisas.

Målsättning: – att kartlägga om genetiskt identiska eller närbesläktade multiresistenta S. epidermidis (genotyper) förekommer och sprids bland patienter vid Östersunds sjukhus.

7

– att undersöka om samma genotyper även kan påvisas bland patienter vid tio andra sjukhus i norra Europa och hos friska personer. – att utvärdera vilka genotyper av S. saprophyticus som orsakar urinvägs-infektion hos kvinnor.

Metod: PFGE användes för att undersökta 359 CoNS stammar resistenta mot stafylokockpenicilliner från patienter vid 11 sjukhus i norra Europa och jämfördes med CoNS stammar från 214 friska personer. Dessutom undersöktes 126 S. saprophyticus stammar som orsakat urinvägsinfektion hos kvinnor från fem olika områden i norra Europa.

Resultat: Majoriteten av CoNS stammar från de sjukhusvårdade patienterna och ungefär hälften av CoNS stammarna från de friska tillhörde arten S. epidermidis. Bland de S. epidermidis som återfanns hos sjukhusvårdade patienter påvisades två dominerande genotyper, vilka tillsammans utgjorde nästan hälften (44%) av alla undersökta S. epidermidis stammar. Dessa två genotyper, som alltid var resistenta mot flera olika antibiotikaklasser, kunde påvisas hos patienter vid Östersunds sjukhus samt även vid 8 av 10 undersökta sjukhus. Däremot uppvisade S. epidermidis stammar från friska personer i norra Sverige en hög grad av genetisk olikhet och låg förekomst av antibiotikaresistens.

Den genetiska analysen av 126 S. saprophyticus stammar visade att flera av kvinnorna med urinvägsinfektion hade stammar med mycket likartad PFGE mönster. Sammanfattning: Vid undersökning av S. epidermidis från patienter som vårdats vid 11 sjukhus i norra Europa kunde vi dokumentera förekomst och potentiell spridning av två multiresistenta genotyper. Dessa genotyper kunde inte påvisas bland friska personer i samhället. Våra resultat pekar mot att multiresistenta S. epidermidis i norra Sverige nästan uteslutande förekommer på sjukhus, och att hittillsvarande antibiotikapolicy och hygienrutiner inte tycks förhindra att dessa stammar etableras och sprids på sjukhus.

Vidare påvisades S. saprophyticus stammar tillhörande samma genotyp hos kvinnor från olika länder med flera års mellanrum. Detta kan tyda på att vissa genotyper av S. saprophyticus som förorsakar urinvägsinfektion är etablerade och sprids inom och mellan länder.

8

Original papers

This thesis is based on the following papers, which will be referred to in the text by the corresponding Roman numerals.

I. Widerström M, Monsen T, Karlsson C, Wiström J. Molecular epidemiology of meticillin-resistant coagulase-negative staphylococci in a Swedish county hospital: evidence of intra- and interhospital clonal spread. 2006. J Hosp Infect. 64:177-83.

II. Widerström M, Monsen T, Karlsson C, Edebro H, Johansson A,

Wiström J. Clonality among multidrug-resistant hospital-associated Staphylococcus epidermidis in northern Europe. 2009. Scand J Infect Dis. 26:1-8.

III. Widerström M, Wiström J, Ek E, Edebro H, Monsen T. Near absence of

methicillin-resistance and pronounced genetic diversity among Staphylococcus epidermidis isolated from healthy persons in northern Sweden. Manuscript.

IV. Widerström M, Wiström J, Ferry S, Karlsson C, Monsen T. Molecular

epidemiology of Staphylococcus saprophyticus isolated from women with uncomplicated community-acquired urinary tract infection. 2007. J Clin Microbiol. 45:1561-4.

Reprints were made with permission of the publishers.

9

Abbreviations

CoNS coagulase-negative staphylococci DNA deoxyribonucleic acid HCW health care worker ICU intensive care unit IS insertion sequence MDRSE multidrug-resistant Staphylococcus epidermidis MIC minimum inhibitory concentration MLST multilocus sequence typing MLVA multilocus variable number tandem repeats analysis MR-CoNS methicillin-resistant coagulase-negative staphylococci MRSA methicillin-resistant Staphylococcus aureus MRSE methicillin-resistant Staphylococcus epidermidis MSSE methicillin-sensitive Staphylococcus epidermidis NCTC national collection of type cultures NICU neonatal intensive care unit PBP penicillin-binding protein PCR polymerase chain reaction PFGE pulsed-field gel electrophoresis RE restriction enzyme or restriction endonuclease rRNA ribosomal RNA SSCmec staphylococcal cassette chromosome mec SmaI restriction enzyme (from Serratia marcescens) SRGA the Swedish Reference Group for Antibiotics ST sequence type VNTR variable number of tandem DNA repeat UTI urinary tract infection

10

11

Definitions (Adapted from van Belkum 2007 (1).)

Allele: a variant of the DNA sequence/gene at a specific locus. Clone: Bacterial isolates that have so many identical phenotypic and genotypic traits that the most likely explanation for this identity is a common origin within a relevant time span. Endemic strain: in a given setting, a constant presence of a strain at a significant frequency over a longer period of time. Epidemic strain: a strain suddenly present in a given setting at an unusually high frequency, well above endemic levels. Genotype: the genetic configuration/constitution of an organism as evaluated by a molecular method.

Isolate: A population of bacterial cells in pure culture derived from a single colony.

Locus: the specific location of a DNA sequence/gene on a chromosome.

Strain: an isolate or group of isolates that can be distinguished from other isolates of the same genus and species by phenotypic and genotypic characteristics.

INTRODUCTION

Introduction

Background Staphylococci are Gram-positive, non-motile, non-spore forming, facultative anaerobic bacteria, usually catalase-positive, that occur in asymmetrical grape-like clusters and, less frequently, singly and in pairs, tetrads or short chains. Staphyle means a bunch of grapes in Greek.

The word staphylococcus was first used in 1881 by Sir Alexander Ogston describing the cluster forming organism in pus in contrast to the chain forming streptococci (2). Rosenbach introduced the generic name Staphylococcus for cluster-forming cocci in 1884 and classified species of staphylococci on the basis of colony colour. He named the orange-yellow staphylococci Staphylococcus aureus (or S. pyogenus aureus) and the white staphylococci S. albus (or S. pyogenes albus). It was important for the medical bacteriologist to differentiate the pathogenic species of S. aureus from the alleged non-pathogenic S. albus, later called Staphylococcus epidermidis (3). The coagulase test was later introduced as an important tool to distinguish S. aureus from S. epidermidis. Furthermore, S. aureus ability to ferment mannitol was found to be another useful tool separating them from S. epidermidis (2).

In the 1960s, Baird-Parker divided the staphylococcus into six subgroups (biotypes) on the basis of the coagulase test and carbohydrate fermentation (4). In 1975, Kloos and Schleifer presented a simplified scheme for identifying human staphylococci based on 13 key characters (5). These were coagulase activity, haemolysis, nitrated reduction, and aerobic production from 10 different carbohydrates. Colony morphology, anaerobic time growth pattern in thioglycolate semisolid medium, phosphatase activity and susceptibility to novobiocin and lysostaphin were used as secondary characters to improve identification accuracy. In routine laboratory medicine today, gram-positive staphylococci are mainly differentiated using test for heat-stable nuclease (DNase), coagulase or latex agglutination detection of clumping factor and protein A (Staphaurex®) (6).

Currently, the genus of Staphylococcus consists of at least 40 species of gram-positive catalase-positive cocciform bacteria including the recently proposed Staphylococcus microti (7). Ten of these species also contain subdivision with subspecies designations. Among the CoNS subset, 18 species have been isolated from humans, but until now only five species of CoNS have repeatedly been implicated in nosocomial spread of genotypes (Table I).

12

INTRODUCTION

Table I. Species of coagulase-negative staphylococci isolated in clinical specimens from humans.

Species Nosocomial spread (ref.)

Only case reports (ref)

Initial description

S. auricularis - + (8, 9) Kloos & Schleifer 1983

S. capitis * + (10-12) Kloos & Schleifer 1975

S. caprae + (13) Devrise 1983

S. carnosus * - + (14) Schleifer and Fischer 1982

S. cohnii * - + (15, 16) Schleifer & Kloos 1975

S. epidermidis + (12, 17-34) Windslow & Windslow 1908

S. haemolyticus + (21, 34-36) Schleifer & Kloos 1975

S. hominis * + (37-39) Kloos & Schleifer 1975

S. lugdunensis + (40) Freney 1988

S. pasteuri - + (15, 41) Chesneau 1993

S. pettenkoferi - + (42, 43) Trülzsch 2002

S. saccharolyticus - + (44-46) Foubert & Douglas 1948

S. saprophyticus * - + (47-50) Fairbrother 1940

S. schleiferi * - + (51, 52) Kloos 1976

S. sciuri * - + (53) Freney 1988

S. simulans - + (54, 55) Kloos & Schleifer 1975

S. warneri + (21, 30, 56, 57) Kloos & Schleifer 1975

S. xylosus - + (58) Schleifer & Kloos 1975

* includes subspecies + nosocomial spread of genotypes documented - no data available Species and initial description adapted from www.bacterio.cict.fr/ Distribution Staphylococci can be found in air, dust, food and water and are primarily isolated from skin and mucous membranes of humans and other mammals. The population density of staphylococci may reach 103 to 106/cm2 in moist areas such as the anterior nares, axillae and perineal areas in contrast to 101 to 103 colony forming units/ cm2 in relative dry parts of the skin such as extremities (59). Some species have a predilection to colonize specific areas of

13

INTRODUCTION

14

the body, i.e. S. hominis and S. haemolyticus are most common on skin areas with apocrine glands (axillae, inguinal and perineal areas); S. auricularis in the external auditory meatus and S. capitis subspecies capitis on the adult human head. Other species such as S. epidermidis and S. lugdunensis are widely distributed over the body (59). Staphylococci are classified as coagulase-positive or coagulase-negative staphylococci (CoNS) depending on their ability to produce the enzyme coagulase, and therefore clot blood plasma. S. aureus is the most clinically significant among the coagulase-positive staphylococci. S. epidermidis comprises 65% to 90% of all CoNS isolated from humans and is the most clinically important species of CoNS in humans (60-62). Approximately 10-24 different phenotypes of S. epidermidis can be isolated from healthy individuals (59, 63). Neonates are colonized with multiple phenotypes of CoNS within the first week of life (64, 65).

S. saprophyticus is a urease-positive, desferrioxamine and novobiocin-resistant CoNS. S. saprophyticus are frequently isolated from the perineal region which may serve as a reservoir for inoculation of the urinary tract, leading to S. saprophyticus urinary tract infection (UTI), particularly in young women (66). Whole genome sequence has revealed a large number of genes encoding several transport proteins that enables adjustments due to changes in pH and osmotic pressure and production of urease, facilitating survival and proliferation in the urinary tract (67). S. saprophyticus has been isolated from the gastrointestinal tract of both humans and animal (cattle, pigs, chickens and rodents), and from meat and cheese products as well as vegetables (68-73).

MOLECULAR TYPING METHODS FOR CONS

Molecular typing methods for CoNS Several different epidemiological typing methods have been used in studies of CoNS. The typing systems can be based either on phenotypic or genotypic criteria.

Table II. Comparison of the main currently available S. epidermidis molecular typing methods.

Method Target Strengths Weaknesses Pulsed-field gel electrophoresis (PFGE)

Restriction polymorphism of the whole chromosome

-High discriminatory power

-Technically demanding -Slow -Limited interlaboratory portability -No standard nomenclatures

Multilocus sequence typing (MLST)

Sequencing allelic variants of housekeeping genes

-Phylogenetic structure of core genome -Interlaboratory portability -Standard nomenclature

-Limited discriminatory power -Low throughput -Cost

Multilocus VNTR* analysis (MLVA)

Polymorphism in number of chromosomal VNTR elements

-Rapid -High throughput

-No validated interpretation criteria -No standard nomenclature - Discriminatory power undecided

Raman spectroscopy

Laser light induced polymorphism in energy exchange

-Rapid -High throughput -High discrimina-tory power

-No validated interpretation criteria -No standard nomenclature -Cost

*VNTR, variable number of tandem DNA repeat.

15


Phenotyping methods, such as biochemical reactivity, antibiotic susceptibility pattern analysis, serological typing, phage typing, plasmid profile, slime production detection, and protein profile analysis, lack discriminatory power on closely related strains (74-76). Genotyping methods, which use various techniques to compare the genetic material of bacterial strains, have higher resolution and have become the method of choice for bacterial strain typing. The most discriminatory method for genotyping of S. epidermidis is so far PFGE (77) which presently, in contrast to multi locus sequence typing (MLST), also can be utilized to characterize other species of CoNS (18, 30, 78-83).

Genotyping methods used in studies on molecular epidemiology of CoNS are mainly based on two different techniques: DNA banding pattern analysis and DNA sequencing. Recently, Raman spectroscopy was introduced as a novel method for characterization of CoNS (84) (Table II).

1. Methods based on DNA banding pattern analysis a) Restriction fragment length polymorphism (RFLP) RFLP was the first widely used fragment-based method. The genome is digested by a frequent cutter restriction enzyme (RE) into several hundred fragments, which are separated by agarose gel electrophoresis (85). Utilizing labelled DNA probes enhances the RFLP analysis by limiting the number of bands. The RFLP pattern is thus mainly a result of the combination of RE and DNA probes. IS6110-fingerprinting, using probes complementary to the insertion sequence (IS) IS6110, is currently used as typing system for Mycobacterium tuberculosis (86). Due to the frequent occurrence and mobility of the IS in the chromosome, IS6110-typing has high resolution. Ribotyping, a variation of RFLP performed on the highly conserved bacterial ribosomal ribonucleic acid (rRNA) genes, have been used for genotyping of staphylococci but is less discriminatory than PFGE (87-89). Automatic ribotyping can be used for species identification in CoNS (90). b) Randomly amplified polymorphic DNA analysis (RAPD) RAPD, also called arbitrary primed polymerase chain reaction (AP-PCR), utilizes arbitrary primers to amplify genomic DNA at multiple loci followed by gel electrophoresis. Several studies have shown that RAPD is less discriminatory than PFGE (18, 29, 30, 74, 91, 92) but it is more rapid and easier to perform.

16


c) Repetitive sequence-based PCR (Rep-PCR) Naturally, repetitive DNA sequences are interspersed throughout the bacterial genome. These repetitive consensus gene fragments can be used to amplify flanking regions. Rep-PCR is quick, but less discriminatory than PFGE and is difficult to standardize in order to get reproducible and portable data (93). d) PCR- Restriction fragment length polymorphism (PCR-RFLP) PCR-RFLP utilizes RFLP-analysis of a specific locus amplified by PCR and has been used in CoNS for the detection of the G2576T mutation in the 23S rRNA, indicating linezolid-resistance (94).

e) Amplified fragment length polymorphism analysis (AFLP) AFLP uses one or two REs to digest genomic DNA and end-specific adapters to ligate restriction fragments followed by selective amplification. Resolution can be further increased using fluorescently labelled primers and an automated DNA sequence for fragment detection. AFLP has high discriminatory power, approximately the same as PFGE, for genotypic identification of Staphylococcus (95-98). Although AFLP have been used for the study of S. aureus, the experience of typing in CoNS is limited (98, 99). Shortcomings with AFLP are, due to the high number of fragments produced, the need for automatic analysis equipment and, with the inherent limitations of band pattern technologies; ambiguity in interpretation of results (97).

f) Multilocus variable number tandem repeats (VNTR) analysis (MLVA) MLVA has been used to study Staphylococcus since 2003 (100). It detects the variability in the numbers of short tandem repeats in a limited amount of both coding and noncoding genes. Individual strains are categorized according to the number of repeats at each investigated loci (Table II). MLVA, a rapid and reproducible genotyping method, is considered the gold standard for epidemiological typing of Bacillus anthracis, Francisella tularensis, Yersinia pestis, and Mycobacterium tuberculosis (101). Although MLVA is highly discriminatory, the variation in different loci may evolve too quickly to permit reliable data on long term epidemiological relationship and population structure. Depending on which and how many loci that are chosen, the discriminatory power can be adjusted (102-104). MLVA has been able to discriminate indistinguishable PFGE types among S. aureus (105). Several investigations have evaluated different primers for MLVA analysis on S. aureus (102-105). In contrast, only two investigations on a limited number of S. epidermidis strains have hitherto been published (106, 107).

17


g) Pulsed-field gel electrophoresis (PFGE) Genotyping by macrorestriction profiling by PFGE was originally devised for electrophoretic separation of the chromosomes of lower eukaryotes such as yeast (108, 109) and has been considered as the gold standard typing method with high resolution and reproducibility for many bacteria(110, 111). Digestion of whole chromosomal genome by rare-cutting restriction endonucleases is followed by separation of the fragments using agarose gel electrophoresis. Restriction endonucleases are enzymes expressed by bacterial organisms that recognize unique sequences of nucleotides and cut the DNA molecule at a specific short sequence unique for each enzyme. Hundreds of such REs have been characterized. PFGE analysis of a group of strains results in banding patterns that reflects DNA variation at recognition sites and the sequence length between recognition sites. Despite the fact that the method only targets the specific enzyme’s recognition motifs in the genome, PFGE has a high capability of detecting larger genomic alterations such as insertions, deletions and rearrangements (Table II). Procedure PFGE allows separation of large DNA fragments by increasing the upper limits from 20-25 kb by normal electrophoresis to over 10 Mb. During continuous field electrophoresis, DNA larger than 20-25 kb migrate with the same speed regardless of size. However, in PFGE the orientation of the electrical field across the gel is changed at programmed intervals allowing the negatively charged large DNA fragments to squeeze through the agarose gel. The smaller the fragment, the higher mobility it will have in the gel. In contrast to RLFP, which uses restrictions enzymes with numerous restriction sites, PFGE is based on digestion by “rare-cutter” enzymes. The number of fragments is thus reduced from hundreds to usually 5-20, which facilitates the interpretation and reproducibility. With knowledge of the complete bacterial genome, the number and size of recognition fragments generated by various REs can theoretically be calculated by computer simulations (http://insilico.ehu.es/). To get fewer and more easily resolved fragments, REs that recognize GC rich DNA sequences are used in species with low GC content, such as SmaI in Staphylococci (Figure 1).

Figure 1. The recognition site of SmaI.

18


To protect the DNA fragments from mechanical shearing and auto digestion, whole cells and their chromosomal DNA are embedded in agarose before cell lysis and subsequent restriction enzyme digestion. The gel with its DNA fragments is stained with ethidium bromide and subsequently photographed under UV-light (108).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Figure 2. SmaI PFGE patterns of MRSE genotypes identified in Study I-II. Line 2-7, 9, 11-12, 14: S. epidermidis (PFGE type A1 in line 7, 11, 12 and 14; PFGE type A2 in line 3, 4, 9; PFGE type A3 in line 2; PFGE type C in line 6). Line 10 and 13: S. haemolyticus. S. aureus NCTC 8325 included as a reference in line 1, 8 and 15. Standardization To standardize the different PFGE patterns within a gel and in-between different gels, a strain with known and useful banding pattern is used as a reference (112). When genotyping staphylococci, S. aureus national collection of type cultures (NCTC) 8325 is used. The reference strain is run as every 6th to 7th isolate, to facilitate the comparison over the gels (Figure 2). The GelCompar program uses this “normalization” process in the analysis model to compensate for minor differences in running length and distortions in and between different gels. Furthermore, the GelCompar software utilizes different parameters when comparing band patterns from several isolates: optimization and tolerance. The optimization is the shift that is allowed between any two patterns and within which the program will look for the best possible matching. The position tolerance is the maximal discrepancy allowed, in percentage of the pattern length, between two bands to consider them as matching. There is no default value, but a tolerance of 1.0% to 1.3% is often used (77, 112, 113). As a consequence, two bands may differ or seem to differ

19


in a position in the dendrogram, but if the difference is within the tolerance level the similarity coefficient will be 100%, and the strains are thus considered indistinguishable. Comparison of PFGE patterns and assignment of PFGE types The banding patterns of the DNA fragments can be analyzed visually or by using computerized systems. Independent of what system is used, the principle is to compare different patterns and to identify similarities and differences. The banding patterns are determined as “indistinguishable” if no differences are identified, as “closely related” if less than three differences are discovered and “unrelated” if the isolates shows more than three variations in their banding pattern (111). Alternatively, as one mutational event could give up to four bands variation in the PFGE profile, a difference of up to four bands have been suggested to classify “closely related” isolates (Figure 3) (1).

The computerized comparison of the banding pattern is based on the computation of similarity coefficients. The Dice coefficient is most commonly used (1). Using pair wise comparison between profiles, where the number of bands having matching positions (NC) is compared to the total number of bands (NT). Thus Dice = 2 x NC/(NC + NT). Consequently, the similarity coefficient expresses the similarity of the DNA pattern profiles between isolates as an indication for genetic relationship or similarity between isolates. The scale bar represents percent similarity of PFGE patterns where 100% corresponds to identical patterns. Thus, PFGE types consist of indistinguishable isolates and isolates with one to three differences in their banding pattern (subtypes).

Figure 3. Schematic diagram showing the theoretical changes in the PFGE pattern of an isolate as a result of possible “single genetic events”. Adapted from R Goering, Workshop 47-08, 47th ICAAC 2007.

20


Different methods can be used to construct a similarity tree diagram, a so-called dendrogram. The most common is the “unweighted pair group method using arithmetic averages” (UPGMA) (1). The two isolates with the most similar DNA profiles are first identified and the similarity coefficient is calculated. Secondly, the next most similar banding pattern is compared with the first, and a new similarity coefficient is calculated. Subsequently, these similarity coefficients build up a similarity matrix depending on the distances of similarity. However, in many cases when the DNA patterns are similar, there are several equivalent solutions (114). Limitations PFGE has a number of inherent limitations. Firstly, it is a time consuming, labour intensive and technically demanding method. Secondly, minor genetic alterations that could occur during a relative short evolutionary time could have significant impact on the macrorestriction pattern and a more detailed understanding of the relationship between different genotypes cannot be made (115). Consequently, caution has to be used when investigating isolates from different periods of time and place, especially when there are minor differences in banding patterns (1, 114). Therefore, PFGE is currently not the typing method of choice for performing long term bacterial population genetics studies (114). However, by combining PFGE genotyping and SCCmec typing, reliable data equivalent to MLST generated data was recently produced on the long-term evolutionary history of S. epidermidis (77). Thirdly, there are difficulties in evaluating and comparing PFGE generated DNA profiles from different laboratories as the method is sensitive to minor change in the technical procedure, for example composition of agarose and buffer, temperature and enzymes. Several attempts have been made to standardize the technical part of the PFGE method in order to facilitate interlaboratory exchange of data (112, 116, 117).

In summary, PFGE is a highly reproducible and discriminating method for genotyping of staphylococcal strains and is ideal for short-term epidemiological studies. Moreover, combined with staphylococcal cassette chromosome mec (SSCmec) typing, PFGE typing can give information on the long-term evolution of S. epidermidis (77). To enable interlaboratory comparable and reproducible data, PFGE analysis requires standardization (1, 112).

21


2. Methods based on DNA sequencing

a) Multilocus sequence typing (MLST) MLST was first introduced in 1998 (118). DNA sequencing is used to examine the allelic variety in a set of slowly evolving genes. Roughly 500 base pairs internal fragments are sequenced to create unambiguous and reliable categorical data. These “housekeeping genes” encode essential proteins and are probably not related to the expression of virulence or antimicrobial resistance. This set of genes, chosen for MLST analysis, is partly conserved and have an intermediary mutation frequency compared to the low mutation frequency in ribosomal genes. Consequently MLST has become the method of choice for the study of long term epidemiologic genetic relationship within bacterial populations as it measures genetic changes in these slowly evolving house-keeping genes (119). Isolates with identical MLST profiles are regarded as having descended from a recent common ancestor (120). For studies on S. epidermidis, seven genes are partially sequenced and compared to known sequences and translated into allelic numbers over the internet by the MLST database (www.mlst.net) (121). For example, sequence type (ST) 1 isolates have the allelic profile 1-2-2-2-1-1-10 in the S. epidermidis scheme. The generated profile facilitates international standardisation and comparison of genotyping data (122) (Table II).

MLST has been an expensive method for genotyping of isolates but has the advantage that it also gives information on bacterial phylogeny and evolution of population lineages. MLST generates no data on mobile genetic elements, and is less discriminatory, and thus less applicable than PFGE for investigating hospital outbreaks. Furthermore, the allelic numbering system is arbitrary and does not represent the actual gene sequence (101). Recently, to increase the resolution of a S. epidermidis population, MLST was combined with analysis of repeat regions of two genes, sdrG (fibronogen-binding protein Fbe) and aap (accumulation-associated protein) (123).

b) 16 S rRNA sequencing Analysis of the highly conserved 16S rRNA gene is useful for identifying most bacteria at the genus and species level (124). It has been evaluated in studies on staphylococci but lacks the discriminatory power even at the species level (125, 126).

22

http://www.mlst.net/


23

c) rpoB rpoB, a conserved gene contained in many bacteria, encodes the β-subunit of the bacterial RNA polymerase. rpoB gene sequence-based analysis is more discriminatory than 16S rRNA sequencing for identification of staphylococcal isolates to species and subspecies level. At the same time it is able to detect rpoB mutations conferring rifampicin resistance (127, 128). d) Staphylococcal cassette chromosome mec (SCCmec) typing The SSCmec is a mobile genetic element carrying the mecA gene, which encodes the alternative penicillin binding protein (PBP2’) conferring methicillin-resistance (129). SCCmec also includes ccr genes for exact integration into and excision from the genome (130). The SSCmec are defined according to the combination of specific ccr and mecA complexes. There are currently eight recognized types (I-VI) and several subtypes of SCCmec in S. aureus (131). Additionally, several non-typeable variants are present in CoNS (77, 132). The SSCmec may be transferred between staphylococcal species (133). Recently, the combination of SCCmec typing and PFGE analysis provided information both on the short-term as well as the long-term epidemiology of S. epidermidis consistent with MLST data analysis (77). c) Whole genome sequencing Currently, only four CoNS strains have been sequenced: S. epidermidis ATCC12228 (134), S. epidermidis RP62a (135), S. haemolyticus JCSC1435 (136), and S. saprophyticus (67). Comparative analysis shows that a substantially portion of the core genome was conserved between the species (137).

3. Raman spectroscopy Raman spectroscopy uses laser light to illuminate bacteria and generates a fingerprint representing their overall molecular composition (138, 139). Raman spectroscopy can identify bacteria to species and subspecies levels (140-142). Recently, through technical improvements, this new and promising method has been documented useful for real-time bacterial typing in clinical microbiology laboratories. The method provides data with high reproducibility and good discriminatory power in concordance with established genotyping techniques and epidemiological data (84, 143). However, specific technical equipment is required and further studies are needed to demonstrate if generated data is portable and if it provides information on bacterial phylogeny (Table II).

PATHOGENESIS

Pathogenesis

Table III. Virulence and adaptive factors in S. epidermidis (adapted from Otto, M 2009 (144)).

Factor Gene Function Biofilm formation-attachment to abiotic surfaces or matrix proteins Autolysin, adhesion (Aae) aae Binds to fibrinogen,

fibronectin, vitronectin Autolysin (AtlE) atlE Binds to vitronectin Serine aspartat binding protein sdrF Binds to collagen Fibrinogen-binding protein (Fbe) sdrG Binds to fibrinogen Elastin binding protein (Ebp) ebp Binds to elastin Intercellular aggregation Polysaccharide intracellular adhesion (PIA)

icaA/B/C/D Cell to cell adhesion

Biofilm-associated protein (Bap) bap Cell to cell adhesion Accumulation associated protein (Aap)

aap Accumulation of biofilm

Teichoic acid Binds to fibronectin Protective biopolymers PIA icaA/B/C/D Protects from IgG,

complement and phagocytosis Poly-γ-glutamic acid (PGA) capA/B/C/D Protects from phagocytosis

and antimicrobial peptides Toxins Phenol-soluble modulins (PSMs) Psmα/β/δ Immune modulation,

Biofilm dispersion δ- Toxin Immune modulation Exoenymes Cysteine protease sspB Unknown Metalloprotease sepA Immune modulation Glutamylendopeptidase and serine protease

sspA Degradation of fibronogen

Lipase gehC, gehD Allows persistence on skin Other Staphyloferrins Sfna Iron acquisition Fatty-acid-modifying-enzyme (FAME)

Allows persistence on skin

Lantibiotics Skin colonization

24

PATHOGENESIS

25

The ability to produce biofilm is a fundamental factor in the pathogenesis of infections due to S. epidermidis. The initial process, adhesion and attachment to biomaterial, is mediated by non-specific factors, such as van der Waals forces, followed by irreversible linkage due to a number of alleged adhesions molecules (Table III). The biofilm is primarily composed of polysaccharide intracellular adhesion (PIA) usually encoded by the ica (intracellular adhesion) operon. Additionally factors are Phenol-soluble modulins (PMSs), which modulates the immune system and the cytolytic δ- Toxin (144). Mobile genetic elements (MGE) have also been investigated as putative virulence factors of S. epidermidis. One such is the arginine catabolic mobile element (ACME) that frequently occurs in S. epidermidis. Specific allotypes of ACME are thought to increase the ability to grow and survive within the host. Horizontally transferred to S. aureus, such as the highly transmissible community-associated methicillin-resistant S. aureus (MRSA) genotype USA300, ACME potentially confers advantage for bacterial colonization and transmission (145).

Clinical presentation of CoNS infections

The first reports on the pathogenicity of CoNS came in late 1950s in studies of endocarditis and septicaemia (146, 147). The clinical significance of CoNS became subsequently recognized during 1960s-70s due to surveys of CoNS in bacteraemia, native and prosthetic valve endocarditis, osteomyelitis, arthritis, and infections related to prosthetic joints, continuous ambulatory peritoneal dialysis, pacemakers, vascular grafts, intravascular catheters, and cerebrospinal fluid shunts (148, 149). Moreover, S. saprophyticus was established as a pathogen causing UTI in young women (61, 150, 151). CoNS causing nosocomial infections are frequently resistant to several of available antibiotics. Moreover, CoNS have the ability to adhere to and colonize indwelling medical devises and to produce biofilm (152). These characteristics make CoNS infections difficult to treat and eradicate (153). CoNS infection has been implicated in practically all synthetic implantable devices (62).

Bacteraemia and intravascular catheter infections CoNS are the most common cause of nosocomial bloodstream infection, responsible for approximately 30% of these infections (154). Most of the CoNS bloodstream infections are related to infections of intravascular catheters, in particular long-term catheters (155). The rate of catheter-related infections increases in parallel with the time of use. These CoNS isolates most often originate from patients own skin alternatively from health care workers

CLINICAL PRESENTATION OF CONS INFECTIONS

(HCW) or the hospital environment (29). CoNS are also the most frequent blood culture contaminants, and it is often difficult to distinguish between clinical infection and contamination, although several approaches have been investigated. Increasing the number of positive blood cultures to two or more will increase the percentage of patients with true bacteraemia (156). Diagnosis of CoNS intravascular device-related blood-stream infection is facilitated if at least two separate blood cultures are obtained, one from peripheral blood and one collected through the intravascular device (157). True CoNS bacteraemia has been found to correlate to time to positivity of blood cultures (158-160). Alternatively, as the blood culture drawn from the infected intravascular devise would have higher inoculum than blood sampled from peripheral blood, it will become positive more quickly. Using a cut-off of 2 hour or more in differential time has been demonstrated to correlate to catheter-related bacteraemia (161). Complications of catheter-related infections due to CoNS, such as exit site infections, abscesses, tunnel infections, infected trombophlebitis, sepsis, and endocarditis have been documented (162).

Vascular graft infections The incidence of prosthetic graft infections have been reported to be 0.6% to 3% with a 5 year mortality of 46% among those infected (163, 164). CoNS are the most common cause of vascular graft infections. Conservative management, antibiotic treatment without surgical removal of the infected graph, is related to increased mortality (163).

Central nervous system shunt infections CoNS cause the majority of cerebrospinal shunt infections. An infection rate from 3% to 20% (165) has been documented. Systemic prophylactic antibiotics reduce shunt infection by approximately 50% (166). Clinical symptoms can be subtle with low grade fever and signs of shunt dysfunction, i.e. drowsiness, nausea, vomiting and headache. Replacement of the shunt is necessary in most cases to eradicate the infection (167). Endocarditis Previously thought to be uncommon, CoNS has been increasingly recognized as an important pathogen in native valve endocarditis (NVE) (168). In NVE cases without history of injection drug use, CoNS are isolated in approximately 7% to 8% of the infections (169). The clinical course of NVE due to CoNS is often severe with high rates of valvular destruction, heart failure, and death. Despite higher rates of surgical therapy (60%) than NVE

26


caused by S. aureus, NVE due to CoNS is associated with 25% in-hospital mortality (169).

CoNS is the second leading cause of prosthetic valve endocarditis (PVE), which is a potentially devastating complication associated with myocardial abscesses, heart failure, additional surgery interventions and with a high mortality ranging from 20% to 70% (170). The majority of all CoNS associated PVE cases occur within 60 days to one year after surgery. Cardiac devices Cardiac pacemaker infections occur with an incidence of 0.1% to 20% with a high mortality rate (171). CoNS, causing at least 25% of these infections, are probably inoculated at the time of surgery (171). Removal of the device is required in a majority of cases to achieve successful treatment of the infection (172, 173). Prosthetic joint infections Total joint replacement, one of the most common operations performed in orthopaedic surgery today, is at present safe and cost effective (174). However, within 2 years of surgery up to 1% and 2% of hip and knee implants, respectively, are infected. Due to the high frequency of prosthetic joint surgery, this complication is associated with substantial morbidity and economic cost. CoNS are together with S. aureus the most common organism isolated from infected prosthetic joints, which is a clinical challenge due to both diagnostic and therapeutic difficulties (175-178).

Surgical site infections CoNS are, second only to S. aureus, a common cause of surgical site infections (179-181). Particularly, CoNS infections in thoracic surgery can be challenging, often requiring repeated surgical interventions and long antibiotic treatment and hospitalization (182, 183).

Endophtalmitis CoNS are the most common cause of endophtalmitis after penetrating eye injury as well as postoperative infection after vitrectomy (184). Treatment has to be timely and potent, using both systemic and intravitreal antibiotics, in order to maintain vision (185).

CoNS infections in neonates Modern healthcare in neonatal intensive units frequently requires the use of umbilical and central venous catheters, which in part explains why preterm

27


28

infants have high risk of developing CoNS bacteraemia (21). S. epidermidis, often multidrug-resistant strains persisting in hospital environment for long periods of time, is the most common species involved in these infections. Other CoNS species such as S. warneri, S. haemolyticus and S. capitis have also been implicated (10, 11, 78).

Urinary tract infections CoNS are accountable for two types of UTI in humans. Firstly, complicated nosocomial infections related to indwelling catheters or surgery primarily caused by methicillin-resistant S. epidermidis. Secondly, and more common, lower UTI in young women caused by S. saprophyticus, which is a true urinary pathogen.

The first indication of pathogenicity in humans was published in 1962 when S. saprophyticus was isolated in urine from 40 women with acute UTI (150). Henceforth, S. saprophyticus has been recognized as a major pathogen, second only to E. coli as a causative agent of lower UTI in young and middle-aged women (151). However, in men with UTI it is only rarely isolated and predominantly in men with indwelling urinary catheters or obstruction (48). Renal involvement, such as acute pyelonephritis and nephrolithiasis, has been documented but systemic infection, for example septicaemia or endocarditis is an uncommon complication of S. saprophyticus infection (49, 66, 186).

S. saprophyticus UTI have been regarded as an organism of low virulence (187). However, in a large prospective, multicentre, randomized, double-blind and placebo controlled study on the natural course of uncomplicated lower urinary tract in women, placebo-treated women infected with E. coli and staphylococci had similar eradication rates of symptoms and spontaneous clearance of bacteria (188).

ANTIMICROBIAL RESISTANCE IN CONS

Antimicrobial resistance in CoNS Staphylococci, and in particular CoNS, have a documented history of developing resistance to antimicrobials. Penicillin, introduced in the 1940s, was the first valuable antibiotic used for treatment of staphylococcal infections(189). Shortly hereafter, penicillin-resistant strains emerged. A beta-lactamase stable penicillin (isoxazolyl-penicillin) called methicillin or “Celbenin” was introduced in September 1960. However, methicillin-resistance in staphylococci may have occurred prior to that (190). Henceforth, methicillin-resistance in CoNS has increased steadily, that is from 20% to 60% during 1980s (191). Multidrug-resistant CoNS frequently colonize hospitalized patients and personnel (192), who serve as a possible reservoir for these bacteria, which antibiotic resistance genes may be transferred among CoNS and potentially to S. aureus (133).

Currently, 70% to 95% of CoNS isolated from nosocomial infections produce beta-lactamase (175, 193-195). Methicillin-resistance is caused by the mecA gene encoding a low-affinity penicillin-binding protein, PBP2a, which prohibit the effect of all beta-lactam antibiotic therapy (196). This phenotype is called heterotypic since only a minority of the isolates in the bacterial population express resistance. Promotion of the growth of resistant subpopulations, i.e. neutral pH, adjustment of temperature and the presence of 2% to 4% NaCl, facilitates the detection of methicillin-resistance (6).

Detection of the mecA gene, alternatively the resulting PBP2a, is considered the method of choice to detect methicillin-resistance (197). However, mecA PCR is still to labour intensive and costly to enable testing all staphylococcal isolates. Phenotypic screening methods have been developed and improved (198). The current recommendation to test for methicillin-resistance in CoNS is to use semi-confluent growth on Iso-Sensitest agar (ISA) using a 10mg cefoxitin disc and overnight incubation at 35°C (the Swedish Reference Group for Antibiotics (SRGA) 2009-05-06) which correlates highly with mecA PCR testing (198).

The incidence of methicillin-resistance differs depending on the origin of the CoNS population studied and on the type of method used to isolate MR-CoNS. Methicillin-resistance often reach 75% to 80% in nosocomial CoNS isolates and in particular intensive care unit (ICU) associated CoNS (24, 195) compared with 27% in CoNS isolates obtained from Swedish primary care centres (199). In CoNS, the prevalence of methicillin-resistance often correlates with the rate of resistance to other classes of antibiotics (26).

29


Aminoglycosides Aminoglycosides, protein synthesis inhibitors, are highly bactericidal to susceptible CoNS, but unfortunately resistance rates are increasing. Genes mediating resistance to several aminoglycosides have been found both on plasmids and in the chromosome of CoNS (200). Chromosomally located resistance is encoded by a transposon, frequently surrounded by copies of an insertion element (IS256) (201), which often coexists with intercellular adhesion genes (ica) and is detected in clinical invasive multidrug-resistant isolates (202, 203). IS256 could therefore be an informative marker of multidrug-resistance in clinically relevant strains (22, 204).

Fluoroquinolones Initially, quinolones had excellent anti-staphylococcal activity, but resistance has emerged rapidly (205). Quinolones acts on the DNA gyrase which are required for the supercoiling of DNA. Single point mutations in the gene encoding the A subunit of DNA gyrase (gyrA) have been identified in S. epidermidis resistant to ciprofloxacin. Mutations in norA, the gene mediating the efflux pump of quinolones, additionally increases the resistance level to quinolones (206).

Resistance to levofloxacin was reported in 17%, 28% and 47% of CoNS isolates obtained from patients attending primary care centres, general hospital wards and intensive care units in northern Europe, respectively (195). Fusidic acid Fusidic acid, utilizing a unique mechanism of action that lacks cross-resistance to other antibiotics, has been used in Europe since 1962. However, is not currently available in the USA (207). Resistance to fusidic acid is due to mutations in the fusA gene, which encodes elongation factor G, or more commonly plasmid-mediated acquisition of resistance gene fusB (208). Resistance rates among CoNS are currently low in the US (1.8%), which probably reflects the low usage of this agent. However, in European countries were fusidic acid has been available, resistance rate are higher (199, 207).

Glycopeptides Acquired bacterial resistance to glycopeptide antibiotics in CoNS was first reported in 1986, in methicillin-resistant strains of the species S. haemolyticus (209). Ever since, teicoplanin MICs against CoNS have shown an inclination to be in a wide range, mostly for some methicillin-resistant isolates of S. haemolyticus and S. epidermidis, while vancomycin MICs tended to be more stable within the limits of susceptibility (210). Other

30


glycopeptide-resistant CoNS species have infrequently been reported, such as S. capitis (11). Recently, due to an impaired clinical response of glycopeptides against intermediate resistant strains of CoNS, the I/R breakpoint for glycopeptides was reduced to 2 mg/L (211).

The mechanism for glycopeptide resistance in staphylococci appears to be endogenous: that includes multiple mutational and metabolic changes. The vancomycin-intermediate resistance phenotype has been linked to several abnormal physiological properties, including a thickened cell wall, leading to decreased affinity and diffusion rate of glycopeptides to its target (210). Linezolid Linezolid is the first drug in a new class of synthetic protein synthesis inhibitors, oxazolidinones. Resistance to linezolid among CoNS is currently very rare (<1%) (199, 212). However, several reports have recently documented clonal outbreaks of linezolid-resistant S. epidermidis, carrying the G2576T mutation of 23S rRNA, often related to increased use of linezolid in the outbreak department or hospital (94, 213, 214). Macrolides and lincosamides Main mechanisms of macrolide and lincosamide resistance in staphylococci are methylation of the 23S rRNA (erm genes) and efflux (mostly msrA gene). Resistance is mediated by a specific inducible methylase, encoded by emrC (plasmid) or emrA (transposon). Mutation from inducible to constitutive can occur during treatment with clindamycin (200). Alternatively, msrA, encoding an ATP-dependent efflux pump, not specific for clindamycin, is a common cause of resistance in European CoNS isolates (215).

Rifampicin Rifampicin inhibits the bacterial RNA polymerase and is highly active against the majority of CoNS (200). Introduced in 1971, Rifampicin has been of increasing interest due to its association with favourable outcome in orthopaedic infections, which probably reflects its capability to penetrate biofilm and effect bacteria in stationary growth phase (178). Spontaneous rifampicin resistance rapidly occurs at a frequency of 1 in 106 to 107 cells due to mutations in the rpoB gene (p. 21)(216). Thus, resistant isolates are rapidly selected both in vitro (217) and in vivo (218, 219). Therefore, rifampicin should be used in combination with another effective antimicrobial agent (178). In a study from northern Europe, resistance was found among 0%, 4% and 18% of CoNS isolates obtained from patients in primary care centres, general hospital wards and intensive care units, respectively (195).

31


32

Tetracycline Tetracycline resistance in staphylococci can develop by two mechanisms, efflux or ribosomal protection. Efflux is usually encoded by tet(K) found on plasmids or transposons often integrated within the SCCmec elements. tet(K) mediates resistance to tetracycline and doxycycline, but not to minocycline (220). The gene encoding ribosomal protection tet(M) is chromosomally located and mediates resistance to tetracycline, doxycycline as well as to minocycline (221). Both mechanisms are overcome by tigecycline, a new class of tetracyclines (“glycylcycline antibiotic”) with broad-spectrum activity against Gram-positive and Gram-negative bacteria. Tigecycline-resistance is still rarely encountered in CoNS (39, 99, 127, 199). Trimethoprim Interfering with the production of tetrahydrofolic acid, the combination of trimethoprim and sulfamethoxasole (TMP-SMX) has excellent activity against susceptible staphylococci and has been used with success in serious S. aureus infections (222, 223). First reports of increasing resistance in CoNS were published in early 1980s - a plasmid-encoded production of the trimethoprim target enzyme, dihydrofolate reductase, dfrA (224), which was not inhibited by TMP-SMX. Resistance was found in a study from northern Europe among 7%, 25% and 43% of isolates obtained from patients attending primary care centres, general hospital wards and intensive care units, respectively (195).

MOLECULAR EPIDEMIOLOGY OF CONS

Molecular epidemiology of CoNS

The origin of infections due to CoNS was previously regarded to be endogenous (225). During the last two decades there has been a considerably increase in the amount of information on the molecular epidemiology of nosocomial S. epidermidis. There is considerably less information regarding species other than S. epidermidis. Several studies investigating hospital-associated S. epidermidis have demonstrated an extensive genetic diversity among investigated strains (79, 226, 227). However, the majority of these studies have regularly detected clusters of CoNS, often methicillin-resistant S. epidermidis (MRSE), which may dominate and persist in the hospital environment (12, 18, 23, 26, 29-33, 228-231). The majority of these genotypes have been documented in one ward during a limited period of time (18, 30, 31) and often with an antibiotic resistance pattern reflecting the antibiotic consumption in that particular ward (18). A few studies have also detected indistinguishable genotypes among isolates obtained from different wards (32, 231) or hospitals (25, 228). Prior to 2003, only one report had investigated S. epidermidis strains obtained from several countries (26). One survey documented the persistence of clones of S. epidermidis in the environment of one hospital during an 11-year period (19). On the other hand, a number of papers have reported the absence of identical genotypes when analyzing isolates obtained from different wards (18, 29, 232) or hospitals (227, 230).

In Scandinavia, limited genotyping information, restricted to hospital-associated strains, was available on S. epidermidis prior to 2003 (28, 34, 55, 182, 233) and no data existed regarding the molecular epidemiology of community-associated strains of S. epidermidis or S. saprophyticus.

33

AIMS

Aims

1. To assess the distribution of different species and the molecular

epidemiology of methicillin-resistant CoNS among clinical isolates from patients treated at the Östersund Hospital (Study I).

2. To compare the molecular epidemiology of clinical isolates of multidrug-

resistant S. epidermidis among patients treated at the Umeå University Hospital and the Östersund Hospital (Study I).

3. To determine the possible presence of multidrug-resistant genotypes of

S. epidermidis among clinical isolates of MDRSE from eleven hospitals in northern Europe (Study II).

4. To document the molecular epidemiology and the frequency of antibiotic

resistance among community-associated S. epidermidis (Study III).

5. To elucidate the molecular epidemiology of S. saprophyticus and to investigate if specific clones are associated with UTI in women (Study IV).

34

STUDY DESIGN AND BACTERIAL ISOLATES

Study design and bacterial isolates

Table IV. Basic characteristics of Study I-IV.

Characteristics Study I Study II Study III Study IV

Population sampled

428 inpatients at Östersund Hospital

219 inpatients in eleven hospitals in northern Europe

223 persons attending a travel health clinic in Umeå

126 women with lower UTI from five locations in northern Europe

Species MR-CoNS MDRSE S. epidermidis S. saprophyticus

No. analyzed 140 173 124 126

Study type and time period

Prospective 2003

Retrospective/ prospective 2003-2008

Prospective 2008

Retrospective 1994-5 Prospective 2005-6

Location Hospital Hospital Community Community

Source of culture Clinical, various Clinical, various Survey, hand Clinical, urine

Simpson index of diversity *

0.949 (0.930-0.969)

0.837 (0.797-0.877)

0.99 (0.983-0.996)

0.968 (0.957-0.980)

Genotyping method

PFGE PFGE, MLST PFGE, MLST PFGE

* Calculated according to Hunter et al. (J Clin Microbiol 1988 Nov;26(11):2465-6) and

Grundmann et al. (J. Clin Microbiol 2001 Nov;39(11):4190-2). Study I Study I was a prospective clinical observation study where all isolates of CoNS identified in consecutive routine cultures from in-patients treated at the Östersund Hospital between January and October 2003 were screened for methicillin-resistance. No specific selective media were used in the screening procedure. Duplicate isolates with identical PFGE patterns from the same patient were only included if more than 14 days had elapsed between the samples.

35

STUDY DESIGN AND BACTERIAL ISOLATES

Study II Study II was a clinical observational follow-up study to determine the possible presence of multidrug-resistant S. epidermidis (MDRSE) genotypes among clinical isolates from 11 hospitals in northern Europe and included both prospectively and retrospectively collected isolates of multidrug-resistant MDRSE.

Study III Study III was a prospective clinical study to determine the molecular epidemiology and the frequency of antibiotic resistance among community-associated S. epidermidis in 223 healthy persons.

Study IV Of the 124 S. saprophyticus strains examined, 76 originated from a subset of women included in a prospective, multicenter, randomized, double-blind, placebo-controlled trial comparing different dosing regimens of pivmecillinam for community-acquired lower UTI. In addition, 45 prospectively collected clinical isolates of S. saprophyticus were obtained in 2005 and 2006 from urine samples from five different locations in northern Europe. Finally, five strains collected in Copenhagen 1995 were also included in the study.

Methods Species identification CoNS were identified by standard methods and further to species level according to a previously described species identification system (234). In short, the system combines 15 key tests (including carbohydrate fermentation) performed in micro-well strips and antimicrobial disc diffusion susceptibility tests to furazolidine, desferrioxamine, polymyxin B, novobiocin and bacitracin on PDM or Iso-Sensitest Agar.

Analyses on chromogenic agar (CM0949, Chromogenic UTI Medium, Oxoid Ltd, Basingstoke, UK) were included to identify the additional isolates of S. saprophyticus collected during 2005 to 2006. All isolates of S. saprophyticus demonstrated a zone diameter breakpoint of <13mm using a 5μg disc of novobiocin (Oxoid Ltd, UK, ISA-medium).

Antimicrobial susceptibility testing Antibiotic susceptibility for oxacillin and other clinically relevant antibiotics was in Study I determined by disc diffusion on Paper Disc Media (PDM-ASM

36

METHODS

agar, AB Biodisc, Solna, Sweden) in accordance with the recommendations at that time of SRGA, (www.srga.org, SRGA Version 3: 2002-11-07). In Study II-IV antibiotic susceptibility to clindamycin, co-trimoxazole, gentamicin and fusidic acid was determined by the disc diffusion test on Iso-Sensitest Agar (Oxoid Ltd, Basingstoke, UK) according to the revised recommendations of SRGA (Version 2008-10-22). Testing for methicillin-resistance was performed according to Table V. Table V. Method for detection of methicillin-resistance in Study I-IV.

Study Disc Medium Zone diameter breakpoints (mm) S/R

I Oxacillin 1ug PDM-ASM agar, 12/9

II Cefoxitin 10ug Mueller-Hinton agar 22/21

III Cefoxitin 10ug Mueller-Hinton agar 27/21*

IV Oxacillin 1ug Mueller-Hinton agar 12/9 * MecA PCR was used to identify methicillin-resistance in isolates showing an indeterminant result in the disc diffusion test to cefoxitin, i.e. a zone diameter of 21-27 mm.

Definition of multidrug-resistance Isolates resistant to methicillin and to four (Study I) or at least three (Study II, IV and in the thesis) of the following antibiotics: clindamycin, co-trimoxazole, gentamicin, fusidic acid or vancomycin were defined as multidrug-resistant. PFGE DNA was prepared from a 3-mL bacterial culture in Todd-Hewitt broth (Difco Laboratories, Detroit, MI, USA). DNA was digested using SmaI (MBI Fermentas, Vilnius, Latvia or Takara Bio Inc. Japan), and the DNA fragments were separated using a contour-clamped homogenous electric field apparatus, i.e. GenePath machine or CHEF-DR III Variable Angle System (Bio-Rad Laboratories, Hercules, CA, USA) using the program for S. aureus (Program 14, 50-500 kb) for 19,7 hours or equivalent settings according to the manufacturer’s instructions (Bio-Rad). Gels were stained in 1.0 mg/L ethidium bromide, rinsed and photographed under ultraviolet illumination. Digitalized images were stored as TIFF formatted files. Genetic similarity between isolates was calculated using GelCompar II software (Applied Maths, Kortrijk, Belgium) using the Dice coefficient and unweighted pair group method with arithmetic mean (UPGMA) as a method for clustering with 1.0% to 1.30% tolerance and 0.5% to 0.8% optimization settings. S. aureus NCTC 8325 was included as a reference strain in every sixth to seventh lane to allow

37

http://www.srga.org/

METHODS

38

subsequent normalization of the electrophoretic pattern across the gel. Bands less than 36 kb were not included for analyses.

Identification of PFGE types PFGE types were visually identified according to established criteria, i.e. isolates with more than three bands of variation in the PFGE pattern were defined as genetically unrelated. Using the cluster analysis, this corresponded to a similarity coefficient of 88% in Study I, 90% in Study II, III and 85% in Study IV. The shifting similarity threshold to define PFGE types in the cluster analysis paralleled with changes in the tolerance and optimizations settings used in the different studies. The reason for the different settings in the cluster analysis varied. In Study I settings from previous studies were used (27). In Study IV, settings from the Harmony report was utilized (112). Finally, in Study II and III settings were changed according to international standards recognised at that time (77). Table VI. Cluster analysis settings used in Gel Compare software in Study I-IV.

Study Tolerance (%) Optimization (%) Similarity coefficient (%)

I 1.2 0.8 88*

II 1.3 0.8 90

III 1.3 0.8 90

IV 1.0 0.5 85

* estimated MLST In Study II and IV, selected strains representing the most prevalent PFGE types and one reference strain (S. epidermidis ATCC 12228) were analysed using the MLST scheme described by Thomas et al (121). The seven housekeeping genes were amplified by PCR and both strands of all amplicons were sequenced with an Applied Biosystems 3730xl DNA Analyzer by using Big Dye Terminator v3.1 cycle sequencing kit. The numbers for allelic profiles and sequence types (STs) were obtained from the S. epidermidis MLST database (http://sepidermidis. mlst.net/).

Statistical methods Statistical analyses in Study IV were performed using the SPSS (version 14, SPSS Inc., Chicago, USA) software package. Fisher's exact test was used to test for association in all two-way tables. A p-value of <0.05 was considered significant.

RESULT AND DISCUSSION

Results and discussion Aim I: Distribution of different species of methicillin-resistant CoNS among clinical samples obtained from patients treated at the Östersund Hospital and identification of genotypes (Study I). Although MR-CoNS are increasingly recognized as major nosocomial pathogens, there are presently only a limited number of studies that have addressed the question whether specific pathogenic genotypes of MR-CoNS occur and persist in the hospital environment and if such genotypes subsequently colonize patients and cause difficult-to-treat infections.

During the study period, 428 isolates of CoNS were detected, of which 188 (44%) were methicillin-resistant. Some isolates were lost, duplicate isolates disregarded and finally 140 isolates were included in Study I (Table IV). According to the in-house species identification system (234), the majority of these isolates were identified as S. epidermidis (86%), 10 as S. haemolyticus (7%), one as S. hominis (0.7%), one as S. intermedius (0.7%), and eight isolates were not identifiable (6%) (Table IX).

Table VII. Distribution of 140 methicillin-resistant isolates of CoNS, of PFGE types A-Cö and non cluster isolates according to source of culture.

Source

of culture

No. of

isolates

Distribution of PFGE types

and non cluster isolates

A B Cö Non cluster

Wound 71 16 3 4 48

Blood 26 1 5 4 16

Urine 23 3 0 5 15

Biopsy 8 2 0 1 5

Body fluid 7 0 1 0 6

Airway 5 0 4 0 1

Total 140 22 13 14 91

39


Several other studies have found corresponding figures for the species distribution among clinical samples of CoNS (60, 194, 235). It should be noted that we only included methicillin-resistant CoNS in our study. According to the PFGE results in Study I we identified 13 possible isolates of S. hominis, of whom 12 according to the in-house species identification system initially were categorized as S. epidermidis due to a negative reaction to trehalose. It has previously been shown that phenotypic species identification of CoNS lacks accuracy (125) and that PFGE analysis can be used as a complementary method for species identification (39, 81, 83, 236, 237).

In Table VII-IX the details of the isolates in Study I are shown. The rates of antibiotic resistance among isolates in Study I are comparable to findings in other studies investigating methicillin-resistant CoNS (194, 195, 199, 212).

Three PFGE types of S. epidermidis, each consisting of more than 10 isolates, were identified (Table IX, Figure 4). Re-evaluation and comparing the PFGE data from Study I with Study II showed that isolates denoted “cluster Group A” and “cluster Group B” in Study I, were indistinguishable to PFGE type A1 and type B1 identified in Study II, respectively. In addition, five PFGE type A2 isolates and two type A3 isolates were detected among the isolates from Study I. Thus, in retrospect 22 PFGE type A isolates obtained from 19 patients were detected (Table IX, Figure 4).

Table VIII. Distribution of 140 methicillin resistant isolates of CoNS, PFGE type A-Cö and non cluster isolates among different ward units within the county hospital.

Ward unit No. of

isolates Distribution of PFGE types and

non cluster isolates

A B Cö Non cluster

Internal medicine 34 1 2 5 26

ICU 45 15 11 2 17

Infectious diseases 21 2 3 16

Surgery 16 2 3 11

Orthopaedic surgery 15 2 1 12

Paediatric 9 9

Total 140 22 13 14 91

40


Table IX. Antimicrobial resistance among 140 methicillin resistant isolates of CoNS, different species of CoNS and among PFGE types A-Cö.

Category No. of isolates

Antimicrobial resistance No. (%)

Co-trimoxazole

Genta-micin

Clinda-mycin

Fusidic acid

All strains 140 83 (59) 80 (57) 68 (49) 47 (34)

S. epidermidis 120 77 (64) 68 (57) 60 (50) 42 (35)

S. haemolyticus 10 4 (40) 7 (70) 4 (40) 2 (20)

Other CoNS 10 2 (20) 5 (50) 4 (40) 3 (30)

PFGE type A * 22 22 (100) 21 (95) 20 (91) 22 (100)

PFGE type B1 # 13 12 (92) 12 (92) 10 (77) 0 (0)

PFGE type Cö § 14 1 (7) 0 (0) 5 (36) 1 (7) *PFGE type A includes PFGE type A1 (which in Study I was denoted Group A), PFGE type A2, and PFGE type A3 # PFGE type B1 was in Study I denoted Group B § PFGE type Cö was in Study I denoted Group C

The second cluster, PFGE type B1, consisted of 13 isolates from 10 patients

of whom eight had been treated in the ICU. In comparison, 15 of 19 PFGE type A isolates and only 2 of 14 PFGE type CÖ isolates were found among patients previously treated in the general ICU.

Five of the PFGE A1 isolates and two A2 isolates were categorized with MLST and all showed an identical allelic profile, designated ST215. One isolate each from PFGE type B1 and type CÖ was identified as ST2 and ST5, respectively (Figure 4).

In conclusion, S. epidermidis comprised the majority (86%) of MR-CoNS isolates obtained from clinical cultures in patients treated at the Östersund Hospital during a 10-month period in 2003. Three major PFGE types of S. epidermidis were identified, of whom two were multidrug-resistant and associated with ICU admission.

41


Aim 2: To compare the molecular epidemiology of multidrug-resistant S. epidermidis detected in clinical samples among patients treated in the Umeå University Hospital and the Östersund Hospital (Study I). The PFGE patterns obtained in this study were compared with 69 previously described multi-drug resistant isolates of S. epidermidis, identified during 2001-2002 from patients treated at the referral Hospital in Umeå (27).

Identical PFGE profiles were detected for isolates in PFGE type A in Study I and in a cluster of 45 isolates from the referral hospital.

Eight of 19 patients with PFGE type A isolates had undergone surgery at the University hospital during the two months preceding isolation of the organism compared with PFGE type B and type Cö, respectively.

In summary, indistinguishable isolates of MDRSE were detected among clinical cultures in both the Östersund Hospital as well as the Umeå University Hospital indicating a possible inter-hospital spread.

Figure 4. Dendrogram (generated using BioNumerics v 5.1) depicting the PFGE patterns of the major genotypes of S. epidermidis identified in Study I and II. The horizontal upper bar represents genetic similarity (percent). The dotted lines in the centre represent digitalized transformation of the PFGE-DNA pattern. The number of isolates in each PFGE type identified in Study I and II, sequence type and antimicrobial susceptibility, respectively, are described in the columns to the right.

42


Aim 3: To determine the possible presence of multidrug-resistant S. epidermidis genotypes among clinical isolates of MDRSE from eleven hospitals in northern Europe (Study II). Prior to 2003, only one report had investigated the molecular epidemiology of MRSE that originated from more the one country. We aimed to determine if PFGE type A and type B, identified in Study I, also could be detected among MRSE isolates obtained from other hospitals in northern Europe. A total of 173 isolates of MDRSE, detected in clinically samples of patients treated in eleven hospitals in northern Europe, were included in Study II (Table IV). The distribution of the isolates in Study II according to origin and major PFGE types is shown in Figure 5. DNA macrorestriction analysis of the 173 isolates revealed 35 different genotypes. Two dominating PFGE types were detected, PFGE type A and PFGE type B, which included 56 (32%) and 38 (22%) isolates, respectively. Three minor PFGE types (type C, type D and type E) were also identified containing 11, 8 and 13 isolates, respectively (Figure 4).

PFGE type A isolates were observed in all Swedish hospitals as well as in the Norwegian hospital (Figure 5). Three isolates from each subtype A1, A2 and A3, were analysed with MLST and all showed an identical and novel allelic profile (1,6,2,1,1,16,1) designated ST215 (Figure 4).

PFGE type B isolates including two subtypes, B1 and B2, were detected in seven Swedish hospitals as well as in the German and the Danish hospital (Figure 5). PFGE type C comprised 11 isolates and was identified in two Swedish hospitals. PFGE type D consisted of eight strains that were recovered in two Swedish hospitals. PFGE type E contained 13 isolates from five hospitals; four in Sweden and one in Norway (Figure 5). MLST analysis of six, one, three and three isolates belonging to PFGE type B, C, D and E, respectively were all identified as ST2 (Figure 4).

The PFGE patterns of the 173 strains included in Study II were compared with the previously categorized 69 MDRSE collected from the Umeå University Hospital during 2001 to 2002 (27) and the 120 MRSE from the Östersund Hospital during 2003 in Study I. PFGE type A1 and subtype B1 proved indistinguishable to the Group A type and Group B type previously described in Study I. However, it should be noted that the DNA macrorestriction pattern denoted PFGE type C in Study II is different from the Group C type (also denoted PFGE type Cö) described in Study I (Figure 4).

43


Figure 5. Distribution of 173 multi-drug resistant isolates of S. epidermidis in Study II according to origin and number of isolates (n). Figures below the city name refers to the number of isolates that belonged to PFGE type A-B-C-D-E from each city, respectively.

Antimicrobial susceptibility Resistance to co-trimoxazole, gentamicin, clindamycin and fusidic acid was detected in 96%, 95%, 94% and 75% of the isolates, respectively, and 60% of the isolates were resistant to all four of these antimicrobials. No isolate was resistant to vancomycin. The majority of PFGE type A and type B isolates, 78% and 76% respectively, were resistant to all four antimicrobials compared with 19% of all PFGE type C, D and E isolates, and 47% of the ‘non-cluster’ isolates (Figure 4).

In conclusion, two dominating genotypes, previously indentified in 2001-2003, were detected among MDRSE isolates obtained from 9 of 11 hospitals in northern Europe during 2003-2008 indicating possible inter-hospital spread and persistence.

44


Aim 4: To document the molecular epidemiology and the frequency of antibiotic resistance among community-associated S. epidermidis in healthy persons (Study III). Presently, there is limited research with respect to community associated strains of S. epidermidis. Only a few studies have investigated the frequency of antibiotic resistance in S. epidermidis obtained from healthy individuals in northern Europe (238). Moreover, the PFGE pattern of MRSE and MSSE in this setting has not previously been studied. The primary objective of this study was to determine the frequency of antibiotic resistance among community-associated S. epidermidis and also to evaluate if the previously identified nosocomial MRSE genotypes are spread among healthy persons in the community.

A total of 223 subjects were included in the study, 124 (56%) women and 99 (44%) men. The mean age was 35 years (range 3 to 75 years). Isolates of S. epidermidis were identified in 124 (56%) of the subjects (Table IV, Figure 6).

0102030405060708090

100

1‐9 10‐19 20‐29 30‐39 40‐49 50‐59 60‐69 70‐

Number Age

No of SE subjects Total no of subjects

Figure 6. Age distribution of all subjects compared with the 124 subjects from whom S. epidermidis (SE subjects) was isolated.

45


Among these 124 subjects, 23% were medical care employees and another

12% reported having a family member that was employed in medical care. A minority, 9% had been treated with antibiotics or had a family member, 14%, that been treated with antibiotics in the preceding six months, 25% had travelled abroad and 4% had been treated in as an inpatient during that time period.

In comparison, 248 of 390 (64%) and 38 of 45 (84%) of MR-CoNS isolates respectively, were classified as S. epidermidis in two other studies of community-associated CoNS (132, 239). However, different methods were used for species identification. Furthermore, we analysed CoNS from hands whereas methicillin-resistant CoNS from noses were examined in the other reports. All these factors may contribute to the varying distributions of staphylococcal species identified in the different studies. Antimicrobial resistance All 223 staphylococcal isolates were sensitive to vancomycin and gentamicin. On the whole, the prevalence of antibiotic resistance was very low (Table X). The only statistical difference in antibiotic resistance rate between S. epidermidis and non-epidermidis in the studied CoNS population was the rate of co-trimoxazole resistance (p=0.01). However, overall the S. epidermidis population did not exhibit more resistance than the non-epidermidis population (Table X). No definite conclusion could be made due to the low prevalence of resistant strains and the low number of subjects in each group.

Table X. Antimicrobial resistance among 124 and 90 isolates of S. epidermidis and non-epidermidis CoNS, respectively.

S. epidermidis

n= 124 (%) Non-epidermidis CoNS

n= 90 (%) Fusidic acid 17 (14) 13 (14) Co-trimoxazole 9 (7) 0 Clindamycin 3 (2,4) 4 (4) Methicillin 2 (1,6) 0 Gentamicin 0 0 Vancomycin 0 0

46


PFGE results DNA macrorestriction analysis of the 124 S. epidermidis isolates revealed a heterogeneous population with 86 different PFGE types, of which 62 were singletons and 23 other genotypes with two to three isolates each (Figure 1, Manuscript III). Only one cluster (PFGE type S) consisting of more than three isolates was detected. Six of 10 of the PFGE type S isolates were obtained from hospital employers and one from a person who had a family member working in medical care.

The PFGE profiles of isolates in Study III were compared with that of 610 hospital-associated S. epidermidis isolates, collected from the Umeå University Hospital (270 isolates) and the Östersund Hospital in northern Sweden (164 isolates) as well as isolates from nine other northern European hospitals (176 isolates).

Only three PFGE types from the present study had identical profiles to that of the isolates from that collection. Two of these PFGE types, one of which belonged to PFGE type S, were identical to two MRSE blood isolates, respectively, collected during 1994 to 1995 at the neonatal intensive care unit (NICU), Umeå University Hospital. The third PFGE type, identified in a student nurse, was indistinguishable to PFGE type A described in Study I and II (Figure 4). However, this isolate, obtained from the student nurse, showed a novel ST profile (1,1,2,1,1,1,10) and was submitted and designated ST257 in the MLST database.

A subset of the 124 isolates was further characterized using MLST. The majority of PFGE type S isolates were identified as ST73, earlier only described once (122). The two MRSE isolates were both identified in HCW. These two isolates belonged to ST2 and ST88, respectively, which previously have been identified among hospital-associated isolates (122, 240).

In summary, these data indicate that the community-associated methicillin-sensible S. epidermidis are more diverse and comprise different genotypes compared with hospital-associated MRSE. The latter are presently infrequent in our community.

47


Aim 5: To elucidate the molecular epidemiology of Staphylococcus saprophyticus and to investigate if specific clones are associated with UTI (Study IV). S. saprophyticus is the second leading cause of uncomplicated UTI in young, sexually active women (49, 66). Prior to Study IV, no information was available on the molecular epidemiology of S. saprophyticus causing UTI and it was unknown if specific genotypes were associated with UTI in women.

In the original study, using routine methods in practice at that time, only 30 of 86 isolates of CoNS identified at study inclusion were typed as S. saprophyticus (188). For the present study all staphylococcal isolates were reanalyzed according to the previously described species identification system (18). This re-evaluation identified 76 isolates of S. saprophyticus and 10 isolates of other CoNS species. The discrepancy, between the original and the current study in the results of CoNS species identification, is probably related to methodical problems regarding novobiocin susceptibility testing. In the original study, CoNS isolates resistant to 10 mg/L of novobiocin were classified as S. saprophyticus while in the current study a zone diameter breakpoint of <13mm using a 5μg disc of novobiocin (Oxoid Ltd, UK, ISA-medium), and a more diversified biotyping scheme, was utilized to identify this species. The relative high concentration of novobiocin used in the original identification probably contributed to the misclassification of a number of S. saprophyticus isolates. Patient population and antibiotic susceptibility Among the 76 women included in the original trial, the median age was 26 (18-77). In contrast to previous reports, a relative high proportion of the S. saprophyticus isolates (41%) were identified in women older than 35 years of age (47).

All of the 126 examined isolates of S. saprophyticus were sensitive to ampicillin, clindamycin, gentamicin, nitrofurantoin, and trimethoprim but resistant to nalidixic acid. Only two isolates were resistant to oxacillin. The low prevalence of methicillin-resistance among S. saprophyticus isolates in the present study (2/126, 1.6%) is comparable with findings in two other studies, including a large international investigation of uncomplicated UTI showing 0.9% resistance to cefadroxil (241, 242).

The bacteriological cure rate among pivmecillinam treated participants was 92% and 98% at the first and last follow-ups, respectively. Corresponding figures in the placebo group was 24% and 50%, respectively. One patient in the pivmecillinam group had a relapse with the primary isolate at the last follow-up (Table XI).

48


The outcome for patients treated with pivmecillinam in the original clinical trial, is in agreement with the clinical and bacteriological cure rates of approximately 90% reported in other studies (243-245). Routine susceptibility testing of S. saprophyticus to mecillinam is not recommended in Sweden (http://www.srga.org). We found that the five strains obtained from mecillinam treated patients with persistence or relapse at follow-up, all had a zone diameter to mecillinam <8mm. The advantageous clinical and bacteriological outcome of pivmecillinam treatment may be related to the high concentrations of mecillinam obtained in the urinary tract. In concordance with previous reports, we found that short-term persistence rate and spontaneous cure rate for S. saprophyticus UTI is highly similar to that of E. coli infections (188, 246, 247).

Table XI. Number of patients (n) in the original trial randomized to placebo or treatment with different dosages of pivmecillinam at inclusion, and number of patients with persistence of bacteriuria with S. saprophyticus among investigated number of patients at follow-up.

Treatment Inclusion (n) Day 8-10 (n) Day 35-49 (n) Placebo 22 16/21 3/6 200mg bid x 7 days* 19 1/18 0/13 200 mg tid x 7 days# 11 0/11 0/11 400 mg bid x 3 days 24 3/23 1/18 Total 76 20/73 4/48

* bid = twice daily # tid = three times daily PFGE results DNA macrorestriction profiles of the 126 S. saprophyticus isolates, 76 from the original trail and 50 additional strains, showed a rather heterogeneous population (Table IV; Figure 3, Manuscript IV). However, several groups consisting of five to 12 isolates were identified from various geographic locations and obtained at different time periods. For instance, indistinguishable isolates were detected in Umeå during 1994 and 2006 as well as in Bochum and Copenhagen. The 21 patients with persistence or relapse demonstrated, for each patient, separate and indistinguishable isolates. In conclusion, PFGE analyses of S. saprophyticus causing UTI in women revealed a quite heterogeneous population, comprising small multiple clusters of related isolates

49


50

0

5

10

15

20

25

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Num

ber o

f UTI

Month

Figure 7. Number of UTI/month caused by S. saprophyticus among the 76 women included in the original trial.

S. saprophyticus infection differs intriguingly, compared with other

uropathogens, in its seasonal variation and geographic distribution. In the northern hemisphere, the prevalence of colonization varies with season both in humans and animals and corresponds to a higher incidence of UTI during late summer and autumn (49, 66, 186, 243). In agreement, the majority of isolates identified in the original trail were obtained during late summer and autumn (Figure 7).

GENERAL DISCUSSION

General discussion Molecular epidemiology of S. epidermidis S. epidermidis is one of the most important pathogens involved in hospital-associated bloodstream infections and infections related to vascular catheters and prosthetic devices. Most nosocomial strains of S. epidermidis are increasingly showing resistance to the majority of available antibiotics. Although S. epidermidis more and more cause difficult-to-treat infections, there is limited information regarding the clonal characteristics and transmission of nosocomial S. epidermidis infections.

Studies performed after 2003 have contributed to the understanding of the global epidemiology of S. epidermidis. Reports have demonstrated extensive diversity among investigated isolates of S. epidermidis obtained both from the hospital and community settings (132, 248). On the other hand, indistinguishable genotypes of S. epidermidis have been identified in samples obtained from one to several wards (20, 21, 24, 25). Moreover, persistence of genotypes of S. epidermidis has been documented in the hospital environment during an 11-year period (21, 23). In agreement, we identified two multidrug-resistant genotypes of S. epidermidis in Study I, in particularly from patients treated at the ICU for a longer period of time. These genotypes persisted in the hospital setting during the study period of 10 months and were also identified at the referral hospital.

Several factors seem important for the persistence of specific MRSE genotypes in the hospital environment. Firstly, the antimicrobial resistance profiles of hospital-associated MRSE genotypes mirror the antibiotic utilization in the particular unit (18, 20, 21, 24, 94, 194, 213, 214, 249). Secondly, endemic hospital-associated genotypes of S. epidermidis have higher rates of biofilm production than non-cluster isolates (21, 24, 31, 250). Thirdly, high level of genetic flexibility, including presence of SSCmec and IS256, has been demonstrated in nosocomial strains of S. epidermidis, possibly reflecting higher ability to adapt to the hospital environment through genetic exchange (24, 134, 135, 202-204, 248).

Less information is currently available on the molecular epidemiology of S. epidermidis obtained from more than one hospital. Endemic clones of CoNS were indentified among isolates from blood cultures obtained in two neonatal intensive care units (NICU) in Rotterdam (n=70) and Utrecht (n=31), respectively, but the genotypes in the two centres were not related (230). The paper did not comment on the exchange of patients or personnel. Likewise, no dominating genotype was detected among 71 isolates of S. epidermidis obtained from patients admitted to three different hospitals in

51

GENERAL DISCUSSION

England (227). However, the type of infection varied and no information was given regarding possible exchange of patients or HCW.

A more recent study, performed at two adjacent NICU in New York, demonstrated the presence of a prevalent genotype of S. epidermidis causing infections in neonates and colonizing HCW in both investigated wards. However, patients were only rarely transferred between the centres, which apparently did not share staff (25). A Dutch study compared 53 and 58 isolates of S. epidermidis, respectively, identified in surveillance cultures obtained from neonates admitted to two adjacent NICU (31). A prevalent and closely related multidrug-resistant genotype was detected in both hospitals. Colonization of the MDRSE genotype was related to longer admittance period of stay and with antibiotic use. There was a frequent exchange of personnel between the two hospitals. In agreement, there is a regular and quite extensive exchange of patients and, to a lesser extent physicians and students, between the Östersund Hospital and the Umeå University Hospital (Study I). Furthermore, in a subgroup of four patients who were sampled repeatedly during their stay in the ICU in Study I, isolates belonging to PFGE type A and B only occurred after at least 11 days in the ICU. The different results in these investigations have several possible explanations: 1) different selection of population and clinical settings, 2) diverse number and selection of isolates, 3) varied antibiotic consumption among the studied population and 4) the amount of transferral of patients and exchange of HCW between the investigated hospitals.

A few studies have analysed the molecular epidemiology of S. epidermidis strains from different countries (22, 26, 122, 123). Among the 230 predominately hospital-associated MRSE isolates obtained from six different countries, indistinguishable or closely related strains isolated were detected, suggesting geographic spread (26). In agreement, both PFGE type A and type B (Study II) were identified in nine of the 11 hospitals indicating possible dissemination among the investigated hospitals in northern Europe (Figure 5). Moreover, although restricted to MDRSE isolates, two PFGE types comprised more than half of the examined isolates indicating a less diverse bacterial population. A recent study, using MLST to analyse 217 nosocomial S. epidermidis isolates from 17 countries, also found disseminating genotypes and the most prevalent genotype, comprising 31% of the isolates, was identified as ST2 (122).

Reports from other countries show that ST2 currently is the most prevalent hospital-associated MRSE genotype. ST2 includes isolates with numerous PFGE patterns and different antibiotic susceptibility profiles (20, 21, 24, 122, 240). In concurrence, ST2 included four of the five most prevalent PFGE types

52

GENERAL DISCUSSION

(PFGE type B, C, D and E) in Study II and comprised at least 41% (70/173) of the examined isolates (Figure 4). However, the most prevailing genotype in Study II, PFGE type A, demonstrated a new allelic profile, designated ST215, suggesting that several genotypes have the ability to survive and disseminate in the hospital setting. Currently, there is lack of epidemiological information to substantiate the dissemination of PFGE type A among all investigated Swedish hospitals. However, the exchange of patients among Swedish hospitals, not only between the Östersund Hospital and the Umeå University Hospital, is quite extensive (personal communication, Ingela Jönsson, Controller, County of Jämtland, Sweden).

Currently, limited information is available regarding the molecular epidemiology of community-associated S. epidermidis, especially methicillin-susceptible isolates. Only two surveys, restricted to methicillin-resistant strains, have previously used PFGE to study the clonality among community-associated S. epidermidis. In one study, no dominating genotype was identified among 161 MRSE nasal isolates obtained from Japanese children attending day-care centres (132). Among 29 MRSE isolates, obtained from healthy individuals at one public school and two military quarters in Brazil, 17 PFGE types were detected (239). The results in these two studies are in agreement with the wide genetic diversity among the S. epidermidis found in Study III; 86 PFGE types were identified among 124 isolates. Only three of these PFGE-types were identified in our collection of 610 nosocomial isolates of S. epidermidis. Similarly, two studies reported that the dominating hospital-associated MLST type rarely was identified among community isolates (22) (Poster no 631, ESCMID 2009; Marta Hercun, Institute for Medical Microbiology and Hygiene, University of Luebeck, Germany).

In summary, community-acquired isolates of S. epidermidis show lower rate of antibiotic resistance and higher degree of heterogeneity compared to S. epidermidis in clinical cultures from hospitalized patients, especially those treated in the intensive care unit (17, 21, 33, 169, 195).

53

GENERAL DISCUSSION

54

Molecular epidemiology of S. saprophyticus The source of UTI still remains unclear. Several factors correlate with S. saprophyticus UTI, such as outdoor swimming, sexual intercourse and work in meat production (47). S. saprophyticus have been isolated from the gastrointestinal tract of a variety of animals and from contaminated food products as well as from the environment (66, 68, 70, 71, 251, 252). Eating contaminated food products may precede colonization and subsequently infection with S. saprophyticus (70, 71). Moreover, rectal, vaginal or urethral colonization with S. saprophyticus has been associated with UTI, suggesting a similar pathogenesis to that of UTI caused by Escherichia coli (66). However, there is scarce data regarding the molecular epidemiology of S. saprophyticus UTIs and it is presently unknown if specific strains or clones are related to urinary tract infections. A recent investigation in France, using PFGE to compare 50 isolates from food and seven clinical isolates, found no common S. saprophyticus genotypes (68). A high genetic diversity was also found among 31 S. saprophyticus strains obtained from fermented dry sausages (72).

We found that the isolates identified in patients with persistence or relapses were indistinguishable to the isolate identified at inclusion in the original trial. Although the isolates were obtained within a short time span, this (unaltered PFGE profile) indicates short-term genomic stability. PFGE analysis of the 126 strains showed a rather consistent banding pattern that could signify a relatively stable core genome with low level of genetic rearrangement (67). On the other hand, the diversity among the uropathogenic genotypes, identified in Study IV, could be consistent with a deviating genetic background and may indicate that the isolates originated from a large environmental reservoir.

To further elucidate the molecular epidemiology and origin of these S. saprophyticus strains, larger patient population and isolates from both UTI and food from various geographical locations should be investigated and preferably also using other typing methods such as MLST or MLVA.

REFLECTIONS AND FURTHER STUDIES

Reflections and future studies

To further elucidate the molecular epidemiology of S. epidermidis and S. saprophyticus there is a need for:

Developing and evaluating simpler methods for genotyping CoNS/ S. epidermidis such as MLVA or Raman spectroscopy.

Comparative genetic studies on successful genotypes such as sequence

studies of ST215 in order to identify markers essential for virulence and pathophysiology.

Prospective clinical studies to assess when, how and why a patient is

colonized by MDRSE. Prospective clinical studies to document if preoperative colonization with

genotypes, such as ST 2 and ST215, confer increased risk of prosthetic joint infection.

Comparative molecular studies on S. saprophyticus isolates obtained

from both UTI and food from various geographical locations to explore the population structure and origin of S. saprophyticus causing UTI in women.

55

CONCLUSIONS

Conclusions

1. S. epidermidis comprised the majority of MR-CoNS isolates obtained from clinical cultures from in-patients at the Östersund Hospital. Two dominating MDRSE genotypes were identified. Isolation of MDRSE was associated to ICU admission.

2. Indistinguishable isolates of multi-drug-resistant S. epidermidis were identified within a county hospital and the referral hospital, indicating possible interhospital spread.

3. Two dominating MDRSE genotypes were detected in clinical cultures

from hospitals located in northern Europe demonstrating clonal persistence and potential wide-spread dissemination.

4. Community-associated S. epidermidis showed high level of genetic

diversity with no predominating genotype. Methicillin-resistance was infrequent.

5. Several different genotypes of S. saprophyticus cause lower UTI in

women with no dominating identified genotype. S. saprophyticus isolates with identical PFGE-pattern were identified in different countries several years apart, indicating the persistence and geographical spread of some clones of S. saprophyticus. Persistent S. saprophyticus infection was common among placebo-treated patients with UTI, demonstrating a low spontaneous bacteriological cure rate.

56

ACKNOWLEDGEMENTS

Acknowledgements

A large number of friends and colleagues have contributed with help and support to the present studies which have been carried out at the Department of Clinical Bacteriology, Umeå University Hospital and the Department of Clinical Microbiology, Östersund Hospital, Östersund. I want to express my profound gratitude to all of them and especially to;

Johan Wiström, my tutor, supervisor and friend, for relentless support, meticulous guidance with serenity, joviality, and enthusiasm as well as scientific stringency during the years. You have shared your wisdom and experience with such generosity.

Tor Monsen, my co-tutor, friend and travel companion for outstanding encouragement, sanguinity, generosity and supervision with delight, amusement and laughs.

Professor Anders Sjöstedt for exceptional advice and for giving me the opportunity to conduct the laboratory work at the department.

Dr Anders Johansson for invaluable scientific contribution and rewarding collaboration.

Carina Karlsson for dazzling technical assistance and enjoyable guidance. You have taught me all I know about PFGE.

Helén Edebro for cheerful and swift co-operation and immaculate technological work.

Elin Ek and Maria Nyström for skilful technical assistance and eternal endeavour.

Elisabeth Lövgren, Satu Koskiniemi and Per Gottfridsson at the research department for pleasurable collaboration.

Jessica Nääs, Elisabeth Marklund and Mattias Backman and all staff at the Department of Microbiology, Östersund Hospital, for invaluable assistance.

Susanne Granholm for precious contribution with the MLST analysis.

57

ACKNOWLEDGEMENTS

58

Sven Ferry for providing us with isolates of S. saprophyticus and valuable collaboration.

Stig Granström, Marianne Rönnmark and all staff at the “substrate” department for excellent collaboration.

Gunborg Eriksson and Inger Molander for speedy assistance and excellent secretarial knowledge.

My colleagues and the entire staff at the Department of Infectious Diseases and the Department of Communicable Disease Control and Prevention, Östersund County Council for support and friendship through a lot of hard work.

Colleagues who provided us with Staphylococcus isolates: Anders Nyberg, Sundsvall Hospital; Magnus Thore, Västerås Hospital; Jalakas Pörnull, Karolinska University Hospital, Solna Stockholm; Christina Welinder-Olsson, Sahlgrenska University Hospital Gothenburg; Marita Skarstedt, Ryhovs Hospital, Jönköping; Professor Gunnar Kahlmeter, Växjö Hospital; Karin Ejrnaes, National Centre for Antimicrobials and Infection Control, Statens Serum Institute, Copenhagen, Denmark; and Florian Szabados, Institute for Hygiene and Microbiology, University Hospital Bochum, Germany.

Ragnar Asplund, Christina Reuterwall and Susanne Johansson at the Research and Development Unit, Jämtland County Council for support and for giving me the opportunity to do research.

Kerstin Fessé and Agneta Bergman for considerate generosity giving me the opportunity to conclude this work.

My family Tage, Ingrid, Charlotta and Mattias for loving support and encouraging knowledge and research.

My wonderful wife Birgitta and astonishing daughters Rebecka, Lovisa, and Matilda for never-ending love, assistance, tolerance and understanding, and for reminding me of real life.

This work was financially supported by grants from the Research and Development Unit, Jämtland County Council, Sweden and the Faculty of Medicine, Umeå University, Umeå, Sweden.

REFERENCES

References 1. van Belkum A, Tassios PT, Dijkshoorn L, et al. Guidelines for the validation and application

of typing methods for use in bacterial epidemiology. Clin Microbiol Infect 2007;13(s3):1-46. 2. Kloos WE. Natural populations of the genus Staphylococcus. Annu Rev Microbiol

1980;34:559-92. 3. Winslow W. The Systematic Relationships of the Coccaceae. New York: John Rily 1908. 4. Baird-Parker AC. The Classification of Staphylococci and Micrococci from World-Wide

Sources. J Gen Microbiol 1965 Mar;38:363-87. 5. Kloos WE, Schleifer KH. Simplified scheme for routine identification of human

Staphylococcus species. J Clin Microbiol 1975 Jan;1(1):82-8. 6. Bannerman TL PS. Staphylococcus, Micrococcus, and other Catalase-Positive Cocci. In:

Murray PR BE, editor. Manual of clinical microbiology Vol 1. 9. ed. Washington, D.C.: ASM Press; 2007. p. 390-411.

7. Novakova D, Pantucek R, Hubalek Z, et al. Staphylococcus microti sp. nov., isolated from the common vole (Microtus arvalis). Int J Syst Evol Microbiol 2010 Mar;60(Pt 3):566-73.

8. Hoffman DJ, Brown GD, Lombardo FA. Early-onset sepsis with Staphylococcus auricularis in an extremely low-birth weight infant - an uncommon pathogen. J Perinatol 2007 Aug;27(8):519-20.

9. Lew SQ, Saez J, Whyte R, et al. Peritoneal dialysis-associated peritonitis caused by Staphylococcus auricularis. Perit Dial Int 2004 Mar-Apr;24(2):195-6.

10. Gras-Le Guen C, Fournier S, Andre-Richet B, et al. Almond oil implicated in a Staphylococcus capitis outbreak in a neonatal intensive care unit. J Perinatol 2007 Nov;27(11):713-7.

11. Van Der Zwet WC, Debets-Ossenkopp YJ, Reinders E, et al. Nosocomial spread of a Staphylococcus capitis strain with heteroresistance to vancomycin in a neonatal intensive care unit. J Clin Microbiol 2002 Jul;40(7):2520-5.

12. de Silva GD, Justice A, Wilkinson AR, et al. Genetic population structure of coagulase-negative staphylococci associated with carriage and disease in preterm infants. Clin Infect Dis 2001 Nov 1;33(9):1520-8.

13. Ross TL, Fuss EP, Harrington SM, et al. Methicillin-resistant Staphylococcus caprae in a neonatal intensive care unit. J Clin Microbiol 2005 Jan;43(1):363-7.

14. Cid J, Lozano M, Nomdedeu B, et al. Staphylococcal sepsis associated with platelet transfusion: report of a new case and review of the literature. Sangre 1997 Oct;42(5):407-9.

15. Arciola CR, Campoccia D, An YH, et al. Prevalence and antibiotic resistance of 15 minor staphylococcal species colonizing orthopedic implants. Int J Artif Organs 2006 Apr;29(4):395-401.

16. Basaglia G, Moras L, Bearz A, et al. Staphylococcus cohnii septicaemia in a patient with colon cancer. J Med Microbiol 2003 Jan;52(Pt 1):101-2.

17. Agvald-Ohman C, Lund B, Hjelmqvist H, et al. ICU stay promotes enrichment and dissemination of multiresistant coagulase-negative staphylococcal strains. Scand J Infect Dis 2006;38(6-7):441-7.

59

REFERENCES

18. Burnie JP, Naderi-Nasab M, Loudon KW, et al. An epidemiological study of blood culture isolates of coagulase-negative staphylococci demonstrating hospital-acquired infection. J Clin Microbiol 1997 Jul;35(7):1746-50.

19. Huebner J, Pier GB, Maslow JN, et al. Endemic nosocomial transmission of Staphylococcus epidermidis bacteremia isolates in a neonatal intensive care unit over 10 years. J Infect Dis 1994 Mar;169(3):526-31.

20. Ibrahem S, Salmenlinna S, Lyytikainen O, et al. Molecular characterization of methicillin-resistant Staphylococcus epidermidis strains from bacteraemic patients. Clin Microbiol Infect 2008 Nov;14(11):1020-7.

21. Klingenberg C, Ronnestad A, Anderson AS, et al. Persistent strains of coagulase-negative staphylococci in a neonatal intensive care unit: virulence factors and invasiveness. Clin Microbiol Infect 2007 Nov;13(11):1100-11.

22. Kozitskaya S, Olson ME, Fey PD, et al. Clonal analysis of Staphylococcus epidermidis isolates carrying or lacking biofilm-mediating genes by multilocus sequence typing. J Clin Microbiol 2005 Sep;43(9):4751-7.

23. Krediet TG, Mascini EM, van Rooij E, et al. Molecular epidemiology of coagulase-negative staphylococci causing sepsis in a neonatal intensive care unit over an 11-year period. J Clin Microbiol 2004 Mar;42(3):992-5.

24. Li M, Wang X, Gao Q, et al. Molecular characterization of Staphylococcus epidermidis strains isolated from a teaching hospital in Shanghai, China. J Med Microbiol 2009 Apr;58(Pt 4):456-61.

25. Milisavljevic V, Wu F, Cimmotti J, et al. Genetic relatedness of Staphylococcus epidermidis from infected infants and staff in the neonatal intensive care unit. Am J Infect Control 2005 Aug;33(6):341-7.

26. Miragaia M, Couto I, Pereira SF, et al. Molecular characterization of methicillin-resistant Staphylococcus epidermidis clones: evidence of geographic dissemination. J Clin Microbiol 2002 Feb;40(2):430-8.

27. Monsen T, Karlsson C, Wistrom J. Spread of clones of multidrug-resistant, coagulase-negative staphylococci within a university hospital. Infect Control Hosp Epidemiol 2005 Jan;26(1):76-80.

28. Monsen T, Olofsson C, Ronnmark M, et al. Clonal spread of staphylococci among patients with peritonitis associated with continuous ambulatory peritoneal dialysis. Kidney Int 2000 Feb;57(2):613-8.

29. Nouwen JL, van Belkum A, de Marie S, et al. Clonal expansion of Staphylococcus epidermidis strains causing Hickman catheter-related infections in a hemato-oncologic department. J Clin Microbiol 1998 Sep;36(9):2696-702.

30. Raimundo O, Heussler H, Bruhn JB, et al. Molecular epidemiology of coagulase-negative staphylococcal bacteraemia in a newborn intensive care unit. J Hosp Infect 2002 May;51(1):33-42.

31. Sloos JH, Horrevorts AM, Van Boven CP, et al. Identification of multiresistant Staphylococcus epidermidis in neonates of a secondary care hospital using pulsed field gel electrophoresis and quantitative antibiogram typing. J Clin Pathol 1998 Jan;51(1):62-7.

32. Spiliopoulou I, Santos Sanches I, Bartzavali C, et al. Application of molecular typing methods to characterize nosocomial coagulase-negative staphylococci collected in a Greek

60

REFERENCES

hospital during a three-year period (1998-2000). Microb Drug Resist 2003 Fall;9(3):273-82.

33. Villari P, Sarnataro C, Iacuzio L. Molecular epidemiology of Staphylococcus epidermidis in a neonatal intensive care unit over a three-year period. J Clin Microbiol 2000 May;38(5):1740-6.

34. Bjorkqvist M, Soderquist B, Tornqvist E, et al. Phenotypic and genotypic characterisation of blood isolates of coagulase-negative staphylococci in the newborn. APMIS 2002 Apr;110(4):332-9.

35. Rodriguez-Aranda A, Daskalaki M, Villar J, et al. Nosocomial spread of linezolid-resistant Staphylococcus haemolyticus infections in an intensive care unit. Diagn Microbiol Infect Dis 2009 Apr;63(4):398-402.

36. Tabe Y, Nakamura A, Igari J. Glycopeptide susceptibility profiles of nosocomial multiresistant Staphylococcus haemolyticus isolates. J Infect Chemother 2001 Sep;7(3):142-7.

37. Chaves F, Garcia-Alvarez M, Sanz F, et al. Nosocomial spread of a Staphylococcus hominis subsp. novobiosepticus strain causing sepsis in a neonatal intensive care unit. J Clin Microbiol 2005 Sep;43(9):4877-9.

38. Palazzo IC, d'Azevedo PA, Secchi C, et al. Staphylococcus hominis subsp. novobiosepticus strains causing nosocomial bloodstream infection in Brazil. J Antimicrob Chemother 2008 Dec;62(6):1222-6.

39. Sorlozano A, Gutierrez J, Martinez T, et al. Detection of new mutations conferring resistance to linezolid in glycopeptide-intermediate susceptibility Staphylococcus hominis subspecies hominis circulating in an intensive care unit. Eur J Clin Microbiol Infect Dis 2010 Jan;29(1):73-80.

40. Hellbacher C, Tornqvist E, Soderquist B. Staphylococcus lugdunensis: clinical spectrum, antibiotic susceptibility, and phenotypic and genotypic patterns of 39 isolates. Clin Microbiol Infect 2006 Jan;12(1):43-9.

41. Savini V, Catavitello C, Carlino D, et al. Staphylococcus pasteuri bacteraemia in a patient with leukaemia. J Clin Pathol 2009 Oct;62(10):957-8.

42. Song SH, Park JS, Kwon HR, et al. Human bloodstream infection caused by Staphylococcus pettenkoferi. J Med Microbiol 2009 Feb;58(Pt 2):270-2.

43. Loiez C, Wallet F, Pischedda P, et al. First case of osteomyelitis caused by "Staphylococcus pettenkoferi". J Clin Micro 2007 Mar;45(3):1069-71.

44. Godreuil S, Jean-Pierre H, Morel J, et al. Unusual case of spondylodiscitis due to Staphylococcus saccharolyticus. Joint Bone Spine 2005 Jan;72(1):91-3.

45. Krishnan S, Haglund L, Ashfaq A, et al. Prosthetic valve endocarditis due to Staphylococcus saccharolyticus. Clin Infect Dis 1996 Apr;22(4):722-3.

46. Steinbrueckner B, Singh S, Freney J, et al. Facing a mysterious hospital outbreak of bacteraemia due to Staphylococcus saccharolyticus. J Hosp Infect 2001 Dec;49(4):305-7.

47. Hedman P, Ringertz O. Urinary tract infections caused by Staphylococcus saprophyticus. A matched case control study. J Infect 1991 Sep;23(2):145-53.

48. Hovelius B, Colleen S, Mardh PA. Urinary tract infections in men caused by Staphylococcus saprophyticus. Scand J Infect Dis 1984;16(1):37-41.

61

REFERENCES

49. Latham RH, Running K, Stamm WE. Urinary tract infections in young adult women caused by Staphylococcus saprophyticus. Jama 1983 Dec 9;250(22):3063-6.

50. Wallmark G, Arremark I, Telander B. Staphylococcus saprophyticus: a frequent cause of acute urinary tract infection among female outpatients. J Infect Dis 1978 Dec;138(6):791-7.

51. Celard M, Vandenesch F, Darbas H, et al. Pacemaker infection caused by Staphylococcus schleiferi, a member of the human preaxillary flora: four case reports. Clin Infect Dis 1997 May;24(5):1014-5.

52. Hernandez JL, Calvo J, Sota R, et al. Clinical and microbiological characteristics of 28 patients with Staphylococcus schleiferi infection. Eur J Clin Microbiol Infect Dis 2001 Mar;20(3):153-8.

53. Hedin G, Widerstrom M. Endocarditis due to Staphylococcus sciuri. Eur J Clin Microbiol Infect Dis 1998 Sep;17(9):673-5.

54. Crichton PB, Anderson LA, Phillips G, et al. Subspecies discrimination of staphylococci from revision arthroplasties by ribotyping. J Hosp Infect 1995 Jun;30(2):139-47.

55. Tammelin A, Hambraeus A, Stahle E. Mediastinitis after cardiac surgery: improvement of bacteriological diagnosis by use of multiple tissue samples and strain typing. J Clin Microbiol 2002 Aug;40(8):2936-41.

56. Center KJ, Reboli AC, Hubler R, et al. Decreased vancomycin susceptibility of coagulase-negative staphylococci in a neonatal intensive care unit: evidence of spread of Staphylococcus warneri. J Clin Microbiol 2003 Oct;41(10):4660-5.

57. Buttery JP, Easton M, Pearson SR, et al. Pediatric bacteremia due to Staphylococcus warneri: microbiological, epidemiological, and clinical features. J Clin Microbiol 1997 Aug;35(8):2174-7.

58. Mastroianni A, Coronado O, Nanetti A, et al. Staphylococcus xylosus isolated from a pancreatic pseudocyst in a patient infected with the human immunodeficiency virus. Clin Infect Dis 1994 Dec;19(6):1173-4.

59. Kloos WE, Musselwhite MS. Distribution and persistence of Staphylococcus and Micrococcus species and other aerobic bacteria on human skin. Appl Microbiol 1975 Sep;30(3):381-5.

60. Kleeman KT, Bannerman TL, Kloos WE. Species distribution of coagulase-negative staphylococcal isolates at a community hospital and implications for selection of staphylococcal identification procedures. J Clin Microbiol 1993 May;31(5):1318-21.

61. Kloos WE, Bannerman TL. Update on clinical significance of coagulase-negative staphylococci. Clin Microbiol Rev 1994 Jan;7(1):117-40.

62. Rogers KL, Fey PD, Rupp ME. Coagulase-negative staphylococcal infections. Inf Dis Clin North Am 2009 Mar;23(1):73-98.

63. Brown E, Wenzel RP, Hendley JO. Exploration of the microbial anatomy of normal human skin by using plasmid profiles of coagulase-negative staphylococci: search for the reservoir of resident skin flora. J Infect Dis 1989 Oct;160(4):644-50.

64. Goldmann DA. Prevention and management of neonatal infections. Inf Dis Clin North Am 1989 Dec;3(4):779-813.

65. Goldmann DA, Leclair J, Macone A. Bacterial colonization of neonates admitted to an intensive care environment. J Pediatr 1978 Aug;93(2):288-93.

62

REFERENCES

66. Raz R, Colodner R, Kunin CM. Who are you--Staphylococcus saprophyticus? Clin Infect Dis 2005 Mar 15;40(6):896-8.

67. Kuroda M, Yamashita A, Hirakawa H, et al. Whole genome sequence of Staphylococcus saprophyticus reveals the pathogenesis of uncomplicated urinary tract infection. Proc Natl Acad Sci U S A 2005 Sep 13;102(37):13272-7.

68. Coton E, Desmonts MH, Leroy S, et al. Biodiversity of coagulase-negative Staphylococci in French cheeses, dry fermented sausages, processing environments and clinical samples. Int J Food Microbiol Feb 28;137(2-3):221-9.

69. Gillespie BE, Headrick SI, Boonyayatra S, et al. Prevalence and persistence of coagulase-negative Staphylococcus species in three dairy research herds. Vet Microbiol 2009 Feb 16;134(1-2):65-72.

70. Hedman P, Ringertz O, Eriksson B, et al. Staphylococcus saprophyticus found to be a common contaminant of food. J Infect 1990 Jul;21(1):11-9.

71. Hedman P, Ringertz O, Lindstrom M, et al. The origin of Staphylococcus saprophyticus from cattle and pigs. Scand J Infect Dis 1993;25(1):57-60.

72. Leroy S, Giammarinaro P, Chacornac JP, et al. Biodiversity of indigenous staphylococci of naturally fermented dry sausages and manufacturing environments of small-scale processing units. Food Microbiol Apr;27(2):294-301.

73. Mounier J, Gelsomino R, Goerges S, et al. Surface microflora of four smear-ripened cheeses. Applied and environmental microbiology 2005 Nov;71(11):6489-500.

74. Sloos JH, Dijkshoorn L, Vogel L, et al. Performance of phenotypic and genotypic methods to determine the clinical relevance of serial blood isolates of staphylococcus epidermidis in patients with septicemia. J Clin Microbiol 2000 Jul;38(7):2488-93.

75. Jarlov JO. Phenotypic characteristics of coagulase-negative staphylococci: typing and antibiotic susceptibility. APMIS Suppl 1999;91:1-42.

76. Bannerman TL, Hancock GA, Tenover FC, et al. Pulsed-field gel electrophoresis as a replacement for bacteriophage typing of Staphylococcus aureus. J Clin Microbiol 1995 Mar;33(3):551-5.

77. Miragaia M, Carrico JA, Thomas JC, et al. Comparison of molecular typing methods for characterization of Staphylococcus epidermidis: proposal for clone definition. J Clin Microbiol 2008 Jan;46(1):118-29.

78. de Silva GD, Kantzanou M, Justice A, et al. The ica operon and biofilm production in coagulase-negative Staphylococci associated with carriage and disease in a neonatal intensive care unit. J Clin Microbiol 2002 Feb;40(2):382-8.

79. Spare MK, Tebbs SE, Lang S, et al. Genotypic and phenotypic properties of coagulase-negative staphylococci causing dialysis catheter-related sepsis. J Hosp Infect 2003 Aug;54(4):272-8.

80. Widerstrom M, Wistrom J, Ferry S, et al. Molecular epidemiology of Staphylococcus saprophyticus isolated from women with uncomplicated community-acquired urinary tract infection. J Clin Microbiol 2007 May;45(5):1561-4.

81. George CG, Kloos WE. Comparison of the SmaI-digested chromosomes of Staphylococcus epidermidis and the closely related species Staphylococcus capitis and Staphylococcus caprae. Int J Syst Bacteriol 1994 Jul;44(3):404-9.

63

REFERENCES

82. Shimizu A, Berkhoff HA, Kloos WE, et al. Genomic DNA fingerprinting, using pulsed-field gel electrophoresis, of Staphylococcus intermedius isolated from dogs. Am J Vet Res 1996 Oct;57(10):1458-62.

83. Shimizu A, Kloos WE, Berkhoff HA, et al. Pulsed-field gel electrophoresis of Staphylococcus hyicus and Staphylococcus chromogenes genomic DNA and its taxonomic, epidemiologic and ecologic applications in veterinary medicine. J Vet Med Sci 1997 Jun;59(6):443-50.

84. Willemse-Erix HF, Jachtenberg J, Barutci H, et al. Proof of principle for successful characterization of methicillin-resistant coagulase-negative staphylococci isolated from skin by use of Raman spectroscopy and pulsed-field gel electrophoresis. J Clin Microbiol Mar;48(3):736-40.

85. Fey PD, Climo MW, Archer GL. Determination of the chromosomal relationship between mecA and gyrA in methicillin-resistant coagulase-negative staphylococci. Antimicrob Agents Chemother 1998 Feb;42(2):306-12.

86. van Embden JD, Cave MD, Crawford JT, et al. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J Clin Microbiol 1993 Feb;31(2):406-9.

87. Cookson BD, Stapleton P, Ludlam H. Ribotyping of coagulase-negative staphylococci. J Med Microbiol 1992 Jun;36(6):414-9.

88. Hubner J, Kropec A. Cross infections due to coagulase-negative staphylococci in high-risk patients. Zentralbl Bakteriol 1995 Dec;283(2):169-74.

89. Kluytmans J, Berg H, Steegh P, et al. Outbreak of Staphylococcus schleiferi wound infections: strain characterization by randomly amplified polymorphic DNA analysis, PCR ribotyping, conventional ribotyping, and pulsed-field gel electrophoresis. J Clin Microbiol 1998 Aug;36(8):2214-9.

90. Carretto E, Barbarini D, Couto I, et al. Identification of coagulase-negative staphylococci other than Staphylococcus epidermidis by automated ribotyping. Clin Microbiol Infect 2005 Mar;11(3):177-84.

91. Casey AL, Worthington T, Caddick JM, et al. RAPD for the typing of coagulase-negative staphylococci implicated in catheter-related bloodstream infection. J Infect 2006 Apr;52(4):282-9.

92. Marquet-Van Der Mee N, Mallet S, Loulergue J, et al. Typing of Staphylococcus epidermidis strains by random amplification of polymorphic DNA. FEMS Microbiol Lett 1995 Apr 15;128(1):39-44.

93. Deplano A, Schuermans A, Van Eldere J, et al. Multicenter evaluation of epidemiological typing of methicillin-resistant Staphylococcus aureus strains by repetitive-element PCR analysis. The European Study Group on Epidemiological Markers of the ESCMID. J Clin Microbiol 2000 Oct;38(10):3527-33.

94. Kelly S, Collins J, Maguire M, et al. An outbreak of colonization with linezolid-resistant Staphylococcus epidermidis in an intensive therapy unit. J Antimicrob Chemother 2008 Apr;61(4):901-7.

95. Melles DC, Gorkink RF, Boelens HA, et al. Natural population dynamics and expansion of pathogenic clones of Staphylococcus aureus. J Clin Invest 2004 Dec;114(12):1732-40.

96. Melles DC, van Leeuwen WB, Snijders SV, et al. Comparison of multilocus sequence typing (MLST), pulsed-field gel electrophoresis (PFGE), and amplified fragment length

64

REFERENCES

polymorphism (AFLP) for genetic typing of Staphylococcus aureus. J Microbiol Meth 2007 May;69(2):371-5.

97. Savelkoul PH, Melles DC, Buffing N, et al. High density whole genome fingerprinting of methicillin-resistant and -susceptible strains of Staphylococcus aureus in search of phenotype-specific molecular determinants. J Microbiol Meth 2007 Oct;71(1):44-54.

98. Sloos JH, Janssen P, van Boven CP, et al. AFLP typing of Staphylococcus epidermidis in multiple sequential blood cultures. Res Microbiol 1998 Mar;149(3):221-8.

99. Natoli S, Fontana C, Favaro M, et al. Characterization of coagulase-negative staphylococcal isolates from blood with reduced susceptibility to glycopeptides and therapeutic options. BMC Inf Dis 2009 Jun 4;9:83.

100. Malachowa N, Sabat A, Gniadkowski M, et al. Comparison of multiple-locus variable-number tandem-repeat analysis with pulsed-field gel electrophoresis, spa typing, and multilocus sequence typing for clonal characterization of Staphylococcus aureus isolates. J Clin Microbiol 2005 Jul;43(7):3095-100.

101. Li W, Raoult D, Fournier PE. Bacterial strain typing in the genomic era. FEMS microbiology reviews 2009 Sep;33(5):892-916.

102. Pourcel C, Hormigos K, Onteniente L, et al. Improved multiple-locus variable-number tandem-repeat assay for Staphylococcus aureus genotyping, providing a highly informative technique together with strong phylogenetic value. J Clin Microbiol 2009 Oct;47(10):3121-8.

103. Schouls LM, Spalburg EC, van Luit M, et al. Multiple-locus variable number tandem repeat analysis of Staphylococcus aureus: comparison with pulsed-field gel electrophoresis and spa-typing. PloS one 2009;4(4):e5082.

104. Tenover FC, Vaughn RR, McDougal LK, et al. Multiple-locus variable-number tandem-repeat assay analysis of methicillin-resistant Staphylococcus aureus strains. J Clin Microbiol 2007 Jul;45(7):2215-9.

105. Moser SA, Box MJ, Patel M, et al. Multiple-locus variable-number tandem-repeat analysis of meticillin-resistant Staphylococcus aureus discriminates within USA pulsed-field gel electrophoresis types. J Hosp Infect 2009 Apr;71(4):333-9.

106. Francois P, Hochmann A, Huyghe A, et al. Rapid and high-throughput genotyping of Staphylococcus epidermidis isolates by automated multilocus variable-number of tandem repeats: a tool for real-time epidemiology. J Microbiol Meth 2008 Mar;72(3):296-305.

107. Johansson A, Koskiniemi S, Gottfridsson P, et al. Multiple-locus variable-number tandem repeat analysis for typing of Staphylococcus epidermidis. J Clin Microbiol 2006 Jan;44(1):260-5.

108. Herschleb J, Ananiev G, Schwartz DC. Pulsed-field gel electrophoresis. Nature protocols 2007;2(3):677-84.

109. Schwartz DC, Cantor CR. Separation of yeast chromosome-sized DNAs by pulsed field gradient gel electrophoresis. Cell 1984 May;37(1):67-75.

110. Gerner-Smidt P, Hise K, Kincaid J, et al. PulseNet USA: a five-year update. Foodborne Pathog Dis 2006 Spring;3(1):9-19.

111. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995 Sep;33(9):2233-9.

65

REFERENCES

112. Murchan S, Kaufmann ME, Deplano A, et al. Harmonization of pulsed-field gel electrophoresis protocols for epidemiological typing of strains of methicillin-resistant Staphylococcus aureus: a single approach developed by consensus in 10 European laboratories and its application for tracing the spread of related strains. J Clin Microbiol 2003 Apr;41(4):1574-85.

113. McDougal LK, Steward CD, Killgore GE, et al. Pulsed-field gel electrophoresis typing of oxacillin-resistant Staphylococcus aureus isolates from the United States: establishing a national database. J Clin Microbiol 2003 Nov;41(11):5113-20.

114. Blanc DS. The use of molecular typing for epidemiological surveillance and investigation of endemic nosocomial infections. Infect Genet Evol 2004 Sep;4(3):193-7.

115. Davis MA, Hancock DD, Besser TE, et al. Evaluation of pulsed-field gel electrophoresis as a tool for determining the degree of genetic relatedness between strains of Escherichia coli O157:H7. J Clin Microbiol 2003 May;41(5):1843-9.

116. Chung M, de Lencastre H, Matthews P, et al. Molecular typing of methicillin-resistant Staphylococcus aureus by pulsed-field gel electrophoresis: comparison of results obtained in a multilaboratory effort using identical protocols and MRSA strains. Microb Drug Resist 2000 Fall;6(3):189-98.

117. Cookson BD, Robinson DA, Monk AB, et al. Evaluation of molecular typing methods in characterizing a European collection of epidemic methicillin-resistant Staphylococcus aureus strains: the HARMONY collection. J Clin Microbiol 2007 Jun;45(6):1830-7.

118. Maiden MC, Bygraves JA, Feil E, et al. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc Natl Acad Sci U S A 1998 Mar 17;95(6):3140-5.

119. Enright MC, Day NP, Davies CE, et al. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J Clin Microbiol 2000 Mar;38(3):1008-15.

120. Turner KM, Feil EJ. The secret life of the multilocus sequence type. Int J Antimicrob Agents 2007 Feb;29(2):129-35.

121. Thomas JC, Vargas MR, Miragaia M, et al. Improved multilocus sequence typing scheme for Staphylococcus epidermidis. J Clin Microbiol 2007 Feb;45(2):616-9.

122. Miragaia M, Thomas JC, Couto I, et al. Inferring a population structure for Staphylococcus epidermidis from multilocus sequence typing data. J Bacteriol 2007 Mar;189(6):2540-52.

123. Monk AB, Archer GL. Use of outer surface protein repeat regions for improved genotyping of Staphylococcus epidermidis. J Clin Microbiol 2007 Mar;45(3):730-5.

124. Clarridge JE, 3rd. Impact of 16S rRNA gene sequence analysis for identification of bacteria on clinical microbiology and infectious diseases. Clin Microbiol Rev 2004 Oct;17(4):840-62, table of contents.

125. Becker K, Harmsen D, Mellmann A, et al. Development and evaluation of a quality-controlled ribosomal sequence database for 16S ribosomal DNA-based identification of Staphylococcus species. J Clin Microbiol 2004 Nov;42(11):4988-95.

126. Ghebremedhin B, Layer F, Konig W, et al. Genetic classification and distinguishing of Staphylococcus species based on different partial gap, 16S rRNA, hsp60, rpoB, sodA, and tuf gene sequences. J Clin Microbiol 2008 Mar;46(3):1019-25.

66

REFERENCES

127. Hellmark B, Unemo M, Nilsdotter-Augustinsson A, et al. Antibiotic susceptibility among Staphylococcus epidermidis isolated from prosthetic joint infections with special focus on rifampicin and variability of the rpoB gene. Clin Microbiol Infect 2009 Mar;15(3):238-44.

128. Mellmann A, Becker K, von Eiff C, et al. Sequencing and staphylococci identification. Emerging infectious diseases 2006 Feb;12(2):333-6.

129. Katayama Y, Ito T, Hiramatsu K. A new class of genetic element, staphylococcus cassette chromosome mec, encodes methicillin resistance in Staphylococcus aureus. Antimicrob Agents Chemother 2000 Jun;44(6):1549-55.

130. Ito T, Katayama Y, Asada K, et al. Structural comparison of three types of staphylococcal cassette chromosome mec integrated in the chromosome in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 2001 May;45(5):1323-36.

131. Classification of staphylococcal cassette chromosome mec (SCCmec): guidelines for reporting novel SCCmec elements. Antimicrob Agents Chemother 2009 Dec;53(12):4961-7.

132. Jamaluddin TZ, Kuwahara-Arai K, Hisata K, et al. Extreme genetic diversity of methicillin-resistant Staphylococcus epidermidis strains disseminated among healthy Japanese children. J Clin Microbiol 2008 Nov;46(11):3778-83.

133. Hanssen AM, Kjeldsen G, Sollid JU. Local variants of Staphylococcal cassette chromosome mec in sporadic methicillin-resistant Staphylococcus aureus and methicillin-resistant coagulase-negative Staphylococci: evidence of horizontal gene transfer? Antimicrob Agents Chemother 2004 Jan;48(1):285-96.

134. Zhang YQ, Ren SX, Li HL, et al. Genome-based analysis of virulence genes in a non-biofilm-forming Staphylococcus epidermidis strain (ATCC 12228). Mol Microbiol 2003 Sep;49(6):1577-93.

135. Gill SR, Fouts DE, Archer GL, et al. Insights on evolution of virulence and resistance from the complete genome analysis of an early methicillin-resistant Staphylococcus aureus strain and a biofilm-producing methicillin-resistant Staphylococcus epidermidis strain. J Bacteriol 2005 Apr;187(7):2426-38.

136. Takeuchi F, Watanabe S, Baba T, et al. Whole-genome sequencing of staphylococcus haemolyticus uncovers the extreme plasticity of its genome and the evolution of human-colonizing staphylococcal species. J Bacteriol 2005 Nov;187(21):7292-308.

137. Ben Zakour NL, Guinane CM, Fitzgerald JR. Pathogenomics of the staphylococci: insights into niche adaptation and the emergence of new virulent strains. FEMS Microbiol Lett 2008 Dec;289(1):1-12.

138. Jarvis RM, Goodacre R. Discrimination of bacteria using surface-enhanced Raman spectroscopy. Anal Chem 2004 Jan 1;76(1):40-7.

139. Ellis DI, Goodacre R. Metabolic fingerprinting in disease diagnosis: biomedical applications of infrared and Raman spectroscopy. Analyst 2006 Aug;131(8):875-85.

140. Grow AE, Wood LL, Claycomb JL, et al. New biochip technology for label-free detection of pathogens and their toxins. Journal of microbiological methods 2003 May;53(2):221-33.

141. Maquelin K, Dijkshoorn L, van der Reijden TJ, et al. Rapid epidemiological analysis of Acinetobacter strains by Raman spectroscopy. J Microbiol Meth 2006 Jan;64(1):126-31.

142. Jarvis RM, Goodacre R. Ultra-violet resonance Raman spectroscopy for the rapid discrimination of urinary tract infection bacteria. FEMS Microbiol Lett 2004 Mar 19;232(2):127-32.

67

REFERENCES

143. Scholtes-Timmerman M, Willemse-Erix H, Schut TB, et al. A novel approach to correct variations in Raman spectra due to photo-bleachable cellular components. Analyst 2009 Feb;134(2):387-93.

144. Otto M. Staphylococcus epidermidis--the 'accidental' pathogen. Nat Rev Microbiol 2009 Aug;7(8):555-67.

145. Miragaia M, de Lencastre H, Perdreau-Remington F, et al. Genetic diversity of arginine catabolic mobile element in Staphylococcus epidermidis. PloS one 2009 Nov 6;4(11):e7722

146. Denton C, Pappas EG, Uricchio JF, et al. Bacterial endocarditis following cardiac surgery. Circulation 1957 Apr;15(4):525-31.

147. Smith IM, Beals PD, Kingsbury KR, et al. Observations on Staphylococcus albus septicemia in mice and men. AMA 1958 Sep;102(3):375-88.

148. Pfaller MA, Herwaldt LA. Laboratory, clinical, and epidemiological aspects of coagulase-negative staphylococci. Clin Microbiol Rev 1988 Jul;1(3):281-99.

149. Rupp ME, Archer GL. Coagulase-negative staphylococci: pathogens associated with medical progress. Clin Infect Dis 1994 Aug;19(2):231-45.

150. Torres Pereira A. Coagulase-negative strains of staphylococcus possessing antigen 51 as agents of urinary infection. J Clin Pathol 1962 May;15:252-3.

151. Hovelius B, Mardh PA. Staphylococcus saprophyticus as a common cause of urinary tract infections. Rev Infect Dis 1984 May-Jun;6(3):328-37.

152. von Eiff C, Peters G, Heilmann C. Pathogenesis of infections due to coagulase-negative staphylococci. Lancet Infect Dis 2002 Nov;2(11):677-85.

153. von Eiff C, Proctor RA, Peters G. Coagulase-negative staphylococci. Pathogens have major role in nosocomial infections. Postgrad Med 2001 Oct;110(4):63-4, 9-70, 3-6.

154. Wisplinghoff H, Bischoff T, Tallent SM, et al. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 2004 Aug 1;39(3):309-17.

155. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis 2009 Jul 1;49(1):1-45.

156. Tokars JI. Predictive value of blood cultures positive for coagulase-negative staphylococci: implications for patient care and health care quality assurance. Clin Infect Dis 2004 Aug 1;39(3):333-41.

157. Richter SS, Beekmann SE, Croco JL, et al. Minimizing the workup of blood culture contaminants: implementation and evaluation of a laboratory-based algorithm. J Clin Microbiol 2002 Jul;40(7):2437-44.

158. Garcia P, Benitez R, Lam M, et al. Coagulase-negative staphylococci: clinical, microbiological and molecular features to predict true bacteraemia. J Med Microbiol 2004 Jan;53(Pt 1):67-72.

159. Martinez JA, Pozo L, Almela M, et al. Microbial and clinical determinants of time-to-positivity in patients with bacteraemia. Clin Microbiol Infect 2007 Jul;13(7):709-16.

160. Kassis C, Rangaraj G, Jiang Y, et al. Differentiating culture samples representing coagulase-negative staphylococcal bacteremia from those representing contamination by use of time-to-positivity and quantitative blood culture methods. J Clin Microbiol 2009 Oct;47(10):3255-60.

68

REFERENCES

161. Raad I, Hanna HA, Alakech B, et al. Differential time to positivity: a useful method for diagnosing catheter-related bloodstream infections. Ann Intern Med 2004 Jan 6;140(1):18-25.

162. Raad I, Hanna H, Maki D. Intravascular catheter-related infections: advances in diagnosis, prevention, and management. Lancet Infect Dis 2007 Oct;7(10):645-57.

163. Saleem BR, Meerwaldt R, Tielliu IF, et al. Conservative treatment of vascular prosthetic graft infection is associated with high mortality. Am J Surg Jan 13 2010 (Epub ahead of print).

164. O'Brien T, Collin J. Prosthetic vascular graft infection. Br J Surg 1992 Dec;79(12):1262-7. 165. Prusseit J, Simon M, von der Brelie C, et al. Epidemiology, prevention and management of

ventriculoperitoneal shunt infections in children. Ped Neurosurg 2009;45(5):325-36. 166. Ratilal B, Costa J, Sampaio C. Antibiotic prophylaxis for surgical introduction of

intracranial ventricular shunts: a systematic review. J Neurosurg 2008 Jan;1(1):48-56. 167. Schreffler RT, Schreffler AJ, Wittler RR. Treatment of cerebrospinal fluid shunt infections:

a decision analysis. Pediatr Infect Dis J 2002 Jul;21(7):632-6. 168. Chu VH, Cabell CH, Abrutyn E, et al. Native valve endocarditis due to coagulase-negative

staphylococci: report of 99 episodes from the International Collaboration on Endocarditis Merged Database. Clin Infect Dis 2004 Nov 15;39(10):1527-30.

169. Chu VH, Woods CW, Miro JM, et al. Emergence of coagulase-negative staphylococci as a cause of native valve endocarditis. Clin Infect Dis 2008 Jan 15;46(2):232-42.

170. Chu VH, Miro JM, Hoen B, et al. Coagulase-negative staphylococcal prosthetic valve endocarditis--a contemporary update based on the International Collaboration on Endocarditis: prospective cohort study. Heart 2009 Apr;95(7):570-6.

171. Gandelman G, Frishman WH, Wiese C, et al. Intravascular device infections: epidemiology, diagnosis, and management. Cardiology in review 2007 Jan-Feb;15(1):13-23.

172. Rundstrom H, Kennergren C, Andersson R, et al. Pacemaker endocarditis during 18 years in Goteborg. Scand J Infect Dis 2004;36(9):674-9.

173. Arber N, Pras E, Copperman Y, et al. Pacemaker endocarditis. Report of 44 cases and review of the literature. Medicine 1994 Nov;73(6):299-305.

174. Matthews PC, Berendt AR, McNally MA, et al. Diagnosis and management of prosthetic joint infection. BMJ 2009 May 29;338:b1773.

175. Uckay I, Pittet D, Vaudaux P, et al. Foreign body infections due to Staphylococcus epidermidis. Ann Med 2009;41(2):109-19.

176. Parvizi J, Azzam K, Ghanem E, et al. Periprosthetic infection due to resistant staphylococci: serious problems on the horizon. Clin Orthop 2009 Jul;467(7):1732-9.

177. Del Pozo JL, Patel R. Clinical practice. Infection associated with prosthetic joints. N Engl J Med 2009 Aug 20;361(8):787-94.

178. Samuel JR, Gould FK. Prosthetic joint infections: single versus combination therapy. J Antimicrob Chemother Jan;65(1):18-23.

179. Owens CD, Stoessel K. Surgical site infections: epidemiology, microbiology and prevention. J Hosp Infect 2008 Nov;70 Suppl 2:3-10.

180. Hidron AI, Edwards JR, Patel J, et al. NHSN annual update: antimicrobial-resistant pathogens associated with healthcare-associated infections: annual summary of data

69

REFERENCES

reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2006-2007. Infect Control Hosp Epidemiol 2008 Nov;29(11):996-1011.

181. Mangram AJ, Horan TC, Pearson ML, et al. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999 Apr;20(4):250-80.

182. Tegnell A, Saeedi B, Isaksson B, et al. A clone of coagulase-negative staphylococci among patients with post-cardiac surgery infections. J Hosp Infect 2002 Sep;52(1):37-42.

183. Soderquist B. Surgical site infections in cardiac surgery: microbiology. APMIS 2007 Sep;115(9):1008-11.

184. Bannerman TL, Rhoden DL, McAllister SK, et al. The source of coagulase-negative staphylococci in the Endophthalmitis Vitrectomy Study. A comparison of eyelid and intraocular isolates using pulsed-field gel electrophoresis. Arch Ophthalmol 1997 Mar;115(3):357-61.

185. Lalwani GA, Flynn HW, Jr., Scott IU, et al. Acute-onset endophthalmitis after clear corneal cataract surgery (1996-2005). Clinical features, causative organisms, and visual acuity outcomes. Ophthalmology 2008 Mar;115(3):473-6.

186. Ferry S, Burman LG, Mattsson B. Urinary tract infection in primary health care in northern Sweden. I. Epidemiology. Scand J Prim Health Care 1987 May;5(2):123-8.

187. Ishihara S, Yokoi S, Ito M, et al. Pathologic significance of Staphylococcus saprophyticus in complicated urinary tract infections. Urology 2001 Jan;57(1):17-20.

188. Ferry SA, Holm SE, Stenlund H, et al. The natural course of uncomplicated lower urinary tract infection in women illustrated by a randomized placebo controlled study. Scand J Infect Dis 2004;36(4):296-301.

189. Fletcher C. First clinical use of penicillin. BMJ1984 Dec;289(6460):1721-3. 190. Jevons MP. "Celbenin"-resistant Staphylococci. BMJ 1961 Jan 14;1(5219):125-6. 191. Schaberg DR, Culver DH, Gaynes RP. Major trends in the microbial etiology of nosocomial

infection. Am J Med 1991 Sep 16;91(3B):72S-5S. 192. Archer GL. Alteration of cutaneous staphylococcal flora as a consequence of antimicrobial

prophylaxis. Rev Infect Dis 1991 Sep-Oct;13 Suppl 10:S805-9. 193. Diekema DJ, Pfaller MA, Jones RN. Age-related trends in pathogen frequency and

antimicrobial susceptibility of bloodstream isolates in North America: SENTRY Antimicrobial Surveillance Program, 1997-2000. Int J Antimicrob Agents 2002 Dec;20(6):412-8.

194. Monsen T, Ronnmark M, Olofsson C, et al. Antibiotic susceptibility of staphylococci isolated in blood cultures in relation to antibiotic consumption in hospital wards. Scand J Infect Dis 1999;31(4):399-404.

195. Claesson C, Hallgren A, Nilsson M, et al. Susceptibility of staphylococci and enterococci to antimicrobial agents at different ward levels in four north European countries. Scand J Infect Dis 2007;39(11-12):1002-12.

196. Chambers HF. Methicillin-resistant staphylococci. Clin Microbiol Rev 1988 Apr;1(2):173-86.

197. Chambers HF. Methicillin resistance in staphylococci: molecular and biochemical basis and clinical implications. Clin Microbiol Rev 1997 Oct;10(4):781-91.

70

REFERENCES

198. Skov R, Smyth R, Larsen AR, et al. Evaluation of cefoxitin 5 and 10 microg discs for the detection of methicillin resistance in staphylococci. J Antimicrob Chemother 2005 Feb;55(2):157-61.

199. Claesson C, Nilsson LE, Kronvall G, et al. Antimicrobial activity of tigecycline and comparative agents against clinical isolates of staphylococci and enterococci from ICUs and general hospital wards at three Swedish university hospitals. Scand J Infect Dis 2009;41(3):171-81.

200. Archer GL, Climo MW. Antimicrobial susceptibility of coagulase-negative staphylococci. Antimicrob Agents Chemother 1994 Oct;38(10):2231-7.

201. Thomas WD, Jr., Archer GL. Mobility of gentamicin resistance genes from staphylococci isolated in the United States: identification of Tn4031, a gentamicin resistance transposon from Staphylococcus epidermidis. Antimicrob Agents Chemother 1989 Aug;33(8):1335-41.

202. Kozitskaya S, Cho SH, Dietrich K, et al. The bacterial insertion sequence element IS256 occurs preferentially in nosocomial Staphylococcus epidermidis isolates: association with biofilm formation and resistance to aminoglycosides. Infect Immun 2004 Feb;72(2):1210-5.

203. Montanaro L, Campoccia D, Pirini V, et al. Antibiotic multiresistance strictly associated with IS256 and ica genes in Staphylococcus epidermidis strains from implant orthopedic infections. Journal of biomedical materials research 2007 Dec 1;83(3):813-8.

204. Koskela A, Nilsdotter-Augustinsson A, Persson L, et al. Prevalence of the ica operon and insertion sequence IS256 among Staphylococcus epidermidis prosthetic joint infection isolates. Eur J Clin Microbiol Infect Dis 2009 Jun;28(6):655-60.

205. Kotilainen P, Nikoskelainen J, Huovinen P. Emergence of ciprofloxacin-resistant coagulase-negative staphylococcal skin flora in immunocompromised patients receiving ciprofloxacin. J Infect Dis 1990 Jan;161(1):41-4.

206. Piddock LJ. New quinolones and gram-positive bacteria. Antimicrob Agents Chemother 1994 Feb;38(2):163-9.

207. Pfaller MA, Castanheira M, Sader HS, et al. Evaluation of the activity of fusidic acid tested against contemporary Gram-positive clinical isolates from the USA and Canada. Int J Antimicrob Agents Mar;35(3):282-7.

208. McLaws F, Chopra I, O'Neill AJ. High prevalence of resistance to fusidic acid in clinical isolates of Staphylococcus epidermidis. J Antimicrob Chemother 2008 May;61(5):1040-3.

209. Del Bene VE, John JF, Jr., Twitty JA, et al. Anti-staphylococcal activity of teicoplanin, vancomycin, and other antimicrobial agents: the significance of methicillin resistance. J Infect Dis 1986 Aug;154(2):349-52.

210. Biavasco F, Vignaroli C, Varaldo PE. Glycopeptide resistance in coagulase-negative staphylococci. Eur J Clin Microbiol Infect Dis 2000 Jun;19(6):403-17.

211. Rybak M, Lomaestro B, Rotschafer JC, et al. Therapeutic monitoring of vancomycin in adult patients: a consensus review of the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, and the Society of Infectious Diseases Pharmacists. Am J Health Syst Pharm 2009 Jan 1;66(1):82-98.

212. Farrell DJ, Mendes RE, Ross JE, et al. Linezolid surveillance program results for 2008 (LEADER Program for 2008). Diagn Microbiol Infect Dis 2009 Dec;65(4):392-403.

213. Potoski BA, Adams J, Clarke L, et al. Epidemiological profile of linezolid-resistant coagulase-negative staphylococci. Clin Infect Dis 2006 Jul 15;43(2):165-71.

71

REFERENCES

214. Trevino M, Martinez-Lamas L, Romero-Jung PA, et al. Endemic linezolid-resistant Staphylococcus epidermidis in a critical care unit. Eur J Clin Microbiol Infect Dis 2009 May;28(5):527-33.

215. Gatermann SG, Koschinski T, Friedrich S. Distribution and expression of macrolide resistance genes in coagulase-negative staphylococci. Clin Microbiol Infect 2007 Aug;13(8):777-81.

216. Aubry-Damon H, Soussy CJ, Courvalin P. Characterization of mutations in the rpoB gene that confer rifampin resistance in Staphylococcus aureus. Antimicrob Agents Chemother 1998 Oct;42(10):2590-4.

217. Archer GL. Antimicrobial susceptibility and selection of resistance among Staphylococcus epidermidis isolates recovered from patients with infections of indwelling foreign devices. Antimicrob Agents Chemother 1978 Sep;14(3):353-9.

218. Archer GL, Armstrong BC. Alteration of staphylococcal flora in cardiac surgery patients receiving antibiotic prophylaxis. J Infect Dis 1983 Apr;147(4):642-9.

219. Liu Y, Cui J, Wang R, et al. Selection of rifampicin-resistant Staphylococcus aureus during tuberculosis therapy: concurrent bacterial eradication and acquisition of resistance. J Antimicrob Chemother 2005 Dec;56(6):1172-5.

220. Ardic N, Sareyyupoglu B, Ozyurt M, et al. Investigation of aminoglycoside modifying enzyme genes in methicillin-resistant staphylococci. Microbiol Res 2006;161(1):49-54.

221. Grundmann H, Aires-de-Sousa M, Boyce J, et al. Emergence and resurgence of meticillin-resistant Staphylococcus aureus as a public-health threat. Lancet 2006 Sep 2;368(9538):874-85.

222. Markowitz N, Quinn EL, Saravolatz LD. Trimethoprim-sulfamethoxazole compared with vancomycin for the treatment of Staphylococcus aureus infection. Ann Intern Med 1992 Sep 1;117(5):390-8.

223. Levitz RE, Quintiliani R. Trimethoprim-sulfamethoxazole for bacterial meningitis. Ann Intern Med 1984 Jun;100(6):881-90.

224. Rouch DA, Messerotti LJ, Loo LS, et al. Trimethoprim resistance transposon Tn4003 from Staphylococcus aureus encodes genes for a dihydrofolate reductase and thymidylate synthetase flanked by three copies of IS257. Mol Microbiol 1989 Feb;3(2):161-75.

225. McGowan JE, Jr. Changing etiology of nosocomial bacteremia and fungemia and other hospital-acquired infections. Rev Infect Dis 1985 Jul-Aug;7 Suppl 3:S357-70.

226. Boisson K, Thouverez M, Talon D, et al. Characterisation of coagulase-negative staphylococci isolated from blood infections: incidence, susceptibility to glycopeptides, and molecular epidemiology. Eur J Clin Microbiol Infect Dis 2002 Sep;21(9):660-5.

227. Lang S, Livesley MA, Lambert PA, et al. The genomic diversity of coagulase-negative staphylococci associated with nosocomial infections. J Hosp Infect 1999 Nov;43(3):187-93.

228. Aires De Sousa M, Santos Sanches I, Ferro ML, et al. Epidemiological study of staphylococcal colonization and cross-infection in two West African Hospitals. Microb Drug Resist 2000 Summer;6(2):133-41.

229. van Belkum A, Kluijtmans J, van Leeuwen W, et al. Monitoring persistence of coagulase-negative staphylococci in a hematology department using phenotypic and genotypic strategies. Infect Control Hosp Epidemiol 1996 Oct;17(10):660-7.

72

REFERENCES

230. Vermont CL, Hartwig NG, Fleer A, et al. Persistence of clones of coagulase-negative staphylococci among premature neonates in neonatal intensive care units: two-center study of bacterial genotyping and patient risk factors. J Clin Microbiol 1998 Sep;36(9):2485-90.

231. Worthington T, Lambert PA, Elliott TS. Is hospital-acquired intravascular catheter-related sepsis associated with outbreak strains of coagulase-negative staphylococci? J Hosp Infect 2000 Oct;46(2):130-4.

232. Schlegel L, Saliba F, Mangeney N, et al. Pulsed field gel electrophoresis typing of coagulase-negative staphylococci with decreased susceptibility to teicoplanin isolated from an intensive care unit. J Hosp Infect 2001 Sep;49(1):62-8.

233. Hedin G. A comparison of methods to determine whether clinical isolates of Staphylococcus epidermidis from the same patient are related. J Hosp Infect 1996 Sep;34(1):31-42.

234. Monsen T, Ronnmark M, Olofsson C, et al. An inexpensive and reliable method for routine identification of staphylococcal species. Eur J Clin Microbiol Infect Dis 1998 May;17(5):327-35.

235. Jarlov JO, Hojbjerg T, Busch-Sorensen C, et al. Coagulase-negative Staphylococci in Danish blood cultures: species distribution and antibiotic susceptibility. J Hosp Infect 1996 Mar;32(3):217-27.

236. Lina B, Vandenesch F, Etienne J, et al. Comparison of coagulase-negative staphylococci by pulsed-field electrophoresis. FEMS Microbiol Lett 1992 Apr 15;71(2):133-38.

237. Linhardt F, Ziebuhr W, Meyer P, et al. Pulsed-field gel electrophoresis of genomic restriction fragments as a tool for the epidemiological analysis of Staphylococcus aureus and coagulase-negative staphylococci. FEMS Microbiol Lett 1992 Aug 15;74(2-3):181-5.

238. Klingenberg C, Glad GT, Olsvik R, et al. Rapid PCR detection of the methicillin resistance gene, mecA, on the hands of medical and non-medical personnel and healthy children and on surfaces in a neonatal intensive care unit. Scand J Infect Dis 2001;33(7):494-7.

239. Silva FR, Mattos EM, Coimbra MV, et al. Isolation and molecular characterization of methicillin-resistant coagulase-negative staphylococci from nasal flora of healthy humans at three community institutions in Rio de Janeiro City. Epidemiol Infect 2001 Aug;127(1):57-62.

240. Monk AB, Boundy S, Chu VH, et al. Analysis of the genotype and virulence of Staphylococcus epidermidis isolates from patients with infective endocarditis. Infect Immun 2008 Nov;76(11):5127-32.

241. Kahlmeter G. An international survey of the antimicrobial susceptibility of pathogens from uncomplicated urinary tract infections: the ECO.SENS Project. J Antimicrob Chemother 2003 Jan;51(1):69-76.

242. Soderquist B, Berglund C. Methicillin-resistant Staphylococcus saprophyticus in Sweden carries various types of staphylococcal cassette chromosome mec (SCCmec). Clin Microbiol Infect 2009 Dec;15(12):1176-8.

243. Granlund M, Landgren E, Henning C. Pivmecillinam in treatment of Staphylococcus saprophyticus urinary tract infections. Scand J Infect Dis 1983;15(1):65-9.

244. Hovelius B, Mardh PA, Nygaard-Pedersen L, et al. Nalidixic acid and pivmecillinam for treatment of acute lower urinary tract infections. Scand J Prim Health Care 1985 Nov;3(4):227-32.

73

REFERENCES

74

245. Pitkajarvi T, Pyykonen ML, Kannisto K, et al. Pivmecillinam treatment in acute cystitis. Three versus seven days study. Arzneimittelforschung 1990 Oct;40(10):1156-8.

246. Colodner R, Ken-Dror S, Kavenshtock B, et al. Epidemiology and clinical characteristics of patients with Staphylococcus saprophyticus bacteriuria in Israel. Infection 2006 Oct;34(5):278-81.

247. Ejrnaes K, Sandvang D, Lundgren B, et al. Pulsed-field gel electrophoresis typing of Escherichia coli strains from samples collected before and after pivmecillinam or placebo treatment of uncomplicated community-acquired urinary tract infection in women. J Clin Microbiol 2006 May;44(5):1776-81.

248. Miragaia M, Couto I, de Lencastre H. Genetic diversity among methicillin-resistant Staphylococcus epidermidis (MRSE). Microb Drug Resist 2005 Summer;11(2):83-93.

249. Santos Sanches I, Mato R, de Lencastre H, et al. Patterns of multidrug resistance among methicillin-resistant hospital isolates of coagulase-positive and coagulase-negative staphylococci collected in the international multicenter study RESIST in 1997 and 1998. Microb Drug Resist 2000 Fall;6(3):199-211.

250. Milisavljevic V, Tran LP, Batmalle C, et al. Benzyl alcohol and ethanol can enhance the pathogenic potential of clinical Staphylococcus epidermidis strains. Am J Infect Control 2008 Oct;36(8):552-8.

251. Even S, Leroy S, Charlier C, et al. Low occurrence of safety hazards in coagulase negative staphylococci isolated from fermented foodstuffs. Int J Food Microbiol Apr 30;139(1-2):87-95.

252. Soge OO, Meschke JS, No DB, et al. Characterization of methicillin-resistant Staphylococcus aureus and methicillin-resistant coagulase-negative Staphylococcus spp. isolated from US West Coast public marine beaches. J Antimicrob Chemother 2009 Dec;64(6):1148-55.

Molecular epidemiology of coagulase-negative...

Documents

Transcript of Molecular epidemiology of coagulase-negative...