Thesis

European antimicrobial resistance surveillance
as part of a Community strategy

Stef Bronzwaer
Stef Bronzwaer
Colofon
Stef Bronzwaer, 2003
Lay-out: Studio aan de Werf
Cover: Studio RIVM
Printing: Wilco, Amersfoort
ISBN electronic version: 90-367-1830-9
EARSS receives funding from the European Commission.
The information contained in this publication does not necessarily reflect the opinion or the
position of the European Commission.
EARSS is also financially supported by the Dutch Ministry of Health, Welfare and Sports.
RIJKSUNIVERSITEIT GRONINGEN
Proefschrift
ter verkrijging van het doctoraat in de
Medische Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. F. Zwarts,
in het openbaar te verdedigen op
woensdag 1 oktober 2003
om 14.15 uur
door
Stephan Louis Adrianus Marie Bronzwaer
geboren op 28 april 1967
te Heerlen
Promotor: Prof. dr. J.E. Degener
Co-promotor: Dr. M.A.E. Conyn-van Spaendonck
Aan Pa
Contents
Chapter 1 Introduction 3
Chapter 2 Objectives and set up of the European Antimicrobial
Resistance Surveillance System (EARSS) 13
Chapter 3 Standardisation of Streptococcus pneumoniae and
Staphylococcus aureus susceptibility data within EARSS 19
Chapter 4 Comparability of antimicrobial susceptibility test results from
22 European countries and Israel: an external quality assurance
exercise of EARSS in collaboration with UK NEQAS 29
Chapter 5 Streptococcus pneumoniae susceptibility data in Europe 51
Chapter 6 A European study on the relationship of antimicrobial use
and antimicrobial resistance 77
Chapter 7 Staphylococcus aureus susceptibility data in Europe 89
Chapter 8 The Community strategy against antimicrobial resistance
concerning human medicine 105
Chapter 9 General discussion 121
Summary 131
Samenvatting 135
Acknowledgements 139
Curriculum Vitae 145
EUROPEAN ANTIMICROBIAL RESISTANCE SURVEILLANCE AS PART OF A COMMUNITY STRATEGY 1
Chapter 1
Introduction
The introduction of penicillin in clinical practice dates back to the 1940s, and almost
immediately the possibility for micro-organisms to develop resistance to antibiotics was
recognised. Some 60 years later, antimicrobial resistance has become a major public
health concern and a world-wide problem, requiring international approaches. The
worlds leading health authorities, such as the World Health Organisation (WHO) and the
Centers for Disease Control and Prevention (CDC), as well as the European Community
have recognised the importance to study the emergence and determinants of
antimicrobial resistance and launched strategies for its control (1-3).
Antimicrobial resistance makes infections more difficult to treat. It may also increase the
length and severity of illness, the period of infectiousness, adverse reactions (due to the
need to use less safe alternative drugs), length of hospital admission and costs (4, 5).
The emergence of resistance represents adaptive selection by micro-organisms which is
to some extent an inevitable result of the therapeutic use of antibiotics. Killing or
suppressing drug-sensitive organisms allows naturally drug-resistant ones to emerge
which can then not only spread but also transfer their resistance to other organisms.
There is an established but complex relation between the consumption of antibiotics and
the prevalence of drug resistance in micro-organisms. This problem can not be overcome
by continuously developing new drugs, as time needed may come too short. An important
complementary step is to avoid further increase in resistance by reducing unnecessary
and inappropriate use of antibiotics.
This makes it imperative that measures are taken to slow the emergence and spread of
resistance to existing antibiotics and to new ones as they come into use. This chapter
provides a brief discussion on the concept of resistance and will illustrate some clinical
implications of recalcitrant infections with resistant strains as opposed to infections with
susceptible strains.
Mechanisms of resistance
Resistance is considered to be present if a bacterium is not susceptible to a clinically-
relevant concentration of an antibiotic and/or when it is possible to demonstrate that the
bacterium possesses a mechanism or property which will render the antibiotic ineffective.
Resistance of a bacterium to an antibacterial substance may be:
Inherent: the species is not normally susceptible to a particular drug. This may be due to
an inability of the antibacterial to enter the bacterial cell and reach its target site(s), lack
of affinity between the antibacterial drug and its target (site of action), or absence of the
target in the cell. This is also called intrinsic resistance.
Acquired: the species is normally susceptible to a particular drug but certain strains
express drug resistance that may be mediated via a number of mechanisms:
i. destroying enzymatically the antimicrobial agent inside or outside the cell;
4 EUROPEAN ANTIMICROBIAL RESISTANCE SURVEILLANCE AS PART OF A COMMUNITY STRATEGY
ii. lowering the intracellular concentration of an antimicrobial as a result of reduced
uptake and/or increased excretion;
iii. altering the target site so that the antimicrobial no longer binds to it;
iv. creating an alternative metabolic pathway that bypasses the target action.
In those strains having an inherent or an acquired mechanism of resistance, minimum
inhibitory concentrations (MICs) of the antibiotic may be higher than those which may be
achieved for an adequate period at the site of infection and, hence, there is the risk of
therapeutic failure. Sometimes two or more mechanisms exist simultaneously in the
same organism and may produce an even greater degree of resistance. One single
mechanism of resistance may bring about the ability to resist actions of some or all of the
drugs of a particular class (cross-resistance). Therefore, exposure of a bacterial
population to one single antibiotic may select for organisms that display resistance to a
large number of similar agents.
Transfer of resistance
There is a genetic basis for all bacterial resistance to antimicrobial agents. Inherent
resistance is determined by the genetic composition of a particular bacterial species.
Acquired resistance is brought about either by random mutation of the DNA of the
bacterial genome, which is then passed on to offspring, or by the acquisition of DNA
containing a gene or genes which code for a mechanism(s) of resistance. DNA may be
transmitted to other bacterial cells by three processes: conjugation, transformation and
transduction (6).
In conjugative transfer, DNA passes along a tube that links two bacteria, which may occur
between bacteria of the same or similar species. Plasmids carrying genes as transposable
elements (transposons) may transfer between cells. Those carrying more than one
transposon can encode resistance to many, chemically unrelated, antibacterials.
Transformation involves the uptake of DNA from the environment. DNA acquired by this
process may come from an unrelated species, and antibacterial resistance may be
acquired even from species not usually responsible for causing disease.
Transduction involves the transfer of DNA by a bacteriophage.
Extent of the problem
There is no clear answer to the question of the extent of the resistance problem. In the
Netherlands, for example, the Rijksinstituut voor Volksgezondheid en Milieu (RIVM,
National Institute of Public Health and the Environment) and a number of associated
regional laboratories run a resistance monitoring programme. The data can be used in
order to formulate antibiotics policy both inside and outside hospitals. However, this
concentrates on selected material so that no picture is established of morbidity and
mortality in the population as a whole or of financial consequences.
Resistant strains are generally no more virulent than non-resistant ones. However, an
infection with a resistant strain can be much more serious because the chance of effective
treatment is much lower. Little is known about the frequency of problems of this kind. In
addition, it is also possible that patients will remain contagious for longer as a result of
inadequate treatment so that an infectious disease can spread more extensively.
Mild infections often improve after treatment using antibiotics to which the pathogen is
resistant (7). The reason for this may be that, despite the reduction in susceptibility,
enough effective concentrations are still attained at the location of the infection (8, 9).
Another possible cause is that the natural course of many of these infections - such as
bronchitis, otitis and sinusitis - is also generally positive without antibiotics (7).
Staphylococcus aureus: resistant (MRSA) or susceptible to methicillin
(MSSA)
In a case-control study, patients infected with MRSA and MSSA in a hospital in the United
States were compared with one another (10). Of the S. aureus infections, 31% were caused
by MRSA. Infection was associated with several prior courses of antibiotics and extension
of the period of admission by seven days. There was no increase in mortality. Another
study produced a comparable result (11). In some hospital departments, the period of
admission was increased by 30 days.
In a retrospective study Crowcroft and Catchpole used death certificates to examine the
evidence that mortality due to MRSA and staphylococcal infections in England and Wales
is increasing (12). MRSA was mentioned on 20.6% (1387/6723) of death certificates that
included an ICD-9 code for staphylococcal infection, gradually increasing from 7.5% in
1993 to 25.0% in 1998. Although recognising limitations of using routine mortality data
for monitoring the impact of MRSA, they conclude that infections due to MRSA seem to
be an increasing cause of mortality in England and Wales.
It is assumed that there are a number of risk factors for the contraction and selection of
MRSA, like frequent and extensive use of wide-spectrum antibiotics, lengthy hospital
admission, presence of decubitus ulcers and other pre-existent skin disorders,
intravascular endoprotheses, administration systems and indwelling catheters.
Recently a number of reports are published on MRSA strains possessing the Panton
Valentine Leukocidin (PVL) gene. The PVL gene encodes a highly potent toxin, which is
involved in severe skin infections and necrotising pneumonia. PVL positive MRSA strains
have been detected in the Netherlands and have also been reported in France (in healthy
individuals), in the United States (in the Los Angeles gay community, and in a large
prison), and in Scotland (small outbreaks of skin abscesses in healthcare staff ) (13). It has
been suggested that the PVL MRSA is acquired in the community (14-16).
Penicillin-resistant and penicillin-sensitive Streptococcus pneumoniae
The mechanism for penicillin resistance in pneumococci is comparable to that of MRSA
and is based upon the change in the affinity of beta-lactam antibiotics for the penicillin-
binding proteins in the bacterial cell wall (6). Penicillin resistance in pneumococci is still
only a sporadic phenomenon in the Netherlands (17), but studies in other European
countries report resistance rates of 50% or more (18). The fact that the highest MIC value
found for pneumococci in European research is 8 mg/l whereas this is normally
< 0.1 mg/l shows that resistance is not absolute. That is why, in the treatment of less
serious infections with high doses of penicillin or amoxicillin, a beneficial effect is still
usually seen. In the case of more complex infections and infections in compartments
where the antibiotic penetrates with greater difficulty, as in the case of the central nervous
system and pulmonary abscesses in emphysema, therapeutic failure should be kept in
mind (8). Penicillin-resistant pneumococci are less susceptible to cephalosporins.
Furthermore, a considerable proportion is resistant to other drugs such as the
macrolides, quinolones and doxycycline. The diffusion of teicoplanin and clindamycin is
poor in cerebrospinal fluid. Susceptibility is universal only in the case of vancomycin.
Complications seen in penicillin resistance have been described in systemic
pneumococcal infections with bacteraemia. In a retrospective study in Spain (19),
mortality was significantly higher (54%, n=24) in patients with infections involving
resistant pneumococci than in patients with susceptible pneumococci (25%, n=48).
Patients with resistant pneumococci had often been treated with antibiotics before. They
had also suffered from pneumonia more often and more of them were seriously ill. In a
later prospective study carried out by the same researchers, no increase in mortality was
found after the results had been corrected for other causes of death (20).
Implications
Resistance is a problem with logistical and economical implications. This is true in particular
of the severe infections that require hospital admission or which arise in hospitals. In the
case of multi-resistance, quarantine measures are required which are not only difficult for
the patient in psychosocial terms but which are also accompanied by a higher workload for
staff. Often, relatively expensive antimicrobial therapies are required. A number of controlled
studies have shown that, in patients with both an infection and resistance, length of
admission and costs in general are at least doubled (8).
For general practitioners, it is of major importance to follow a restrictive antibiotic policy
given the fact that many infections seen in general practice (upper airway infections) are
caused by viruses. Antibiotics should only be prescribed upon strict indication. The
Standards of the Netherlands Society of General Practitioners, partly available in English,
provide guidelines in this respect (21). If an antibiotic is indicated, the classic drugs
should be selected first. Where possible, preference will be for drugs with a narrow
spectrum. In addition, reserve drugs should never be prescribed blindly, since this can
contribute to the increase of resistance to these drugs. If an antibiotic therapy that has
been started fails, resistance should be tested by means of a culture. Subsequent
treatment should be based on the result of this test.
For a number of bacterial antigens and clinical situations, it has been demonstrated that
resistance to antibiotics is a complicating factor. Resistance constitutes a threat to
patients in risk categories such as those with reduced immunity or those who are infected
with tuberculosis or salmonella bacteria. Fast treatment that covers the susceptibility
spectrum can often save lives here. If the right antibiotic is not prescribed for this patient
group given the susceptibility of the bacterium, therapeutic failure with serious
consequences is seen more often than if the right choice had been made. Alongside
common strains of bacteria such as Staphylococcus aureus and enterococci, it is in the
nature of things that more resistant strains are found in this situation, examples being
Pseudomonas spp. and Serratia spp. For a few of the species, it has been shown that
infections with resistant strains are associated with higher rates of morbidity, mortality
and recurrent infections. This applies to the entire range of Gram-positive and -negative
species of bacteria that can cause bacteraemia. Not a single one of these species of
bacteria is an obligate pathogen; they constitute a part of the indigenous flora or of flora
in the environment that colonises the patient. It is only under exceptional circumstances
that their pathogenic properties become evident.
In the Netherlands, in October 1996, the Dutch Working Party on Antibiotic Policy (Dutch
acronym is SWAB) was established as an initiative of the Society of Infectious Diseases and
the professional societies of medical microbiologists and hospital pharmacists (22). The
mission of the SWAB is to contribute to the containment of the development of
antimicrobial resistance and of the expanding costs of the use of antibiotics. This is
achieved by optimising the use of antibiotics by means of guideline development,
education and antibiotic resistance surveillance. In December 2000, the Council on Health
Research advised the government on antibiotic resistance. The Minister of Health
responded in November 2001, stating that she would follow this advice to a large extent.
This advice by the Council on Health Research as well as the decision made by the Minister
of Health are of great importance to the SWAB, because the SWAB has since then been
designated to co-ordinate the surveillance of antibiotic resistance in the Netherlands.
Background and outline of thesis
Pathogens have never recognised the ever more fading European frontiers as barriers.
There is a clear need for European collaboration to control infectious diseases. The Treaty
of Amsterdam makes provision for action directed towards improving public health,
preventing human illness and diseases. At the EU conference The Microbial Threat, held
in Copenhagen in 1998, all EU Member States unanimously agreed that antimicrobial
resistance was no longer a national problem, but a major international issue requiring a
common strategy at European level (23). One of the recommendations made at this
conference was that a European surveillance system of antimicrobial resistance should be
set up. In the same year the RIVM (National Institute of Public Health and the
Environment) in the Netherlands had taken the initiative and received funding from the
European Commission to start with the European Antimicrobial Resistance Surveillance
System (EARSS). In 2001, at a follow-up EU-conference in Visby, Sweden, it was
concluded that all Member States should join EARSS as a minimum requirement of
national surveillance programmes.
Surveillance of antimicrobial resistance is a first step towards containment of the problem
and is generally considered to be necessary to provide local data for selection of empirical
therapy, to assess the scale of the resistance problem at local, national or international
level, to monitor changes in resistance rates, to detect the emergence and spread of new
resistances, and to provide a measure of the effectiveness of interventions aimed at
reducing resistance. Surveillance can also provide an opportunity to improve the quality
of susceptibility testing among participants in the surveillance (24).
This thesis aims to explore ways how to set up and improve European surveillance of
antimicrobial resistance as a necessary step in the containment of antimicrobial
resistance. Microbiological laboratories are using different diagnostic protocols between
and even within countries. Indications for taking clinical samples may vary as well as the
choice of antibiotics. Criteria for discriminating resistant isolates from susceptible
bacteria are often based on national, and not on international consensus. How to address
these problems, aiming to provide reproducible and comparable data from the
participating laboratories, is studied and discussed in chapters 2 and 3.
In chapter 4 the question is asked whether laboratories in different countries are able to
provide reliable results when it comes to susceptibility testing of the bacterial species
under surveillance. For this reason we initiated an external quality exercise to study the
comparability of susceptibility test results among participants.
In chapter 5 a survey is described to investigate the European geographical distribution
and a trend-analysis of the susceptibility of the community-acquired pathogen S.
pneumoniae against a number of indicator antibiotics.
In chapter 6 we investigate whether there is a relationship between the level of resistance
in a certain country and the level of antimicrobial use. We therefore study the correlation
between S. pneumoniae resistance rates and the amount of penicillin and macrolides used
at country level.
In chapter 7 a survey is described to investigate the European geographical distribution
and a trend-analysis of the susceptibility of a common hospital-acquired pathogen S.
aureus against key indicator antibiotics.
In chapter 8 we aim to provide the larger framework of which EARSS is part. We present
the comprehensive Community strategy against antimicrobial resistance with its actions
to contain antimicrobial resistance and discuss how these actions are to be co-ordinated.
Finally, in chapter 9 we discuss general findings of the studies and provide
recommendations specifically for community- and hospital-acquired pathogens.
References
1. European Community. Official Journal of the European Community. Council Resolution of 8 June 1999 on
antibiotic resistance A strategy against the microbial threat. Official Journal C 195, 13/07/1999 p.13.
Available at: http://europa.eu.int/eur-lex/en/lif/dat/1999/en_ 399Y0713_01.html. Accessed April 29, 2000.
2. World Health Organization. Report on Infectious Diseases 2000: Overcoming antimicrobial resistance.
www.who.int/infectious-disease-report/index.html. Accessed September 23, 2000.
3. Centers for Disease Control and Prevention. Preventing Emerging Infectious Diseases.
www.cdc.gov/ncidod/emergplan/plan98.pdf. Accessed May 20, 2000.
4. Metly J, Hoffmann J, Cetron M, Fine M, Farley M, Whitney C, Breiman R. Impact of Penicillin Susceptibility
on Medical Outcomes for Adult Patients with Bacteremic Pneumococcal Pneumonia. Clin Inf Dis
2000;30:520-8.
5. Kim T, Oh PI, Simor AE. The economic impact of methicillin-resistant Staphylococcus aureus in Canadian
hospitals. Infect Control Hosp Epidemiol. 2001 Feb;22(2):99-104.
6. Neu HC. The crisis in antibiotic resistance. Science 1992; 257: 1064-1072.
7. Melker RA de. Effectiviteit van antibiotica bij veel voorkomende luchtweginfecties in de huisartspraktijk. Ned
Tijdschr Geneeskd 1998; 142: 452-456.
8. Klugman KP. The clinical relevance of in-vitro resistance to penicillin, ampicillin, amoxycillin, and alternative
agents for the treatment of community-acquired pneumonia caused by Streptococcus pneumoniae,
Haemophilus influenzae and Moraxella catarrhalis. J Antimicrob Chemother 1996; 38 (suppl A): 133-140.
9. Friedland IR, McCracken GH jr. Management of infections caused by antibiotic-resistant Streptococcus
pneumoniae. N Engl J Med 1994; 331: 377-382.
10. Saravolatz LD, Markowitz N, Arking L, Pohlod D, Fisher E. Methicillin-resistant Staphylococcus aureus.
Epidemiologic observations during a community-acquired outbreak. Ann Intern Med 1982; 96: 11-16.
11. Holmberg SD, Solomon SL, Blake PA. Health and economic impacts of antimicrobial resistance. Rev Infect
Dis 1987; 9: 1065-1078.
12. Crowcroft NS, Catchpole M. Mortality from methicillin resistant Staphylococcus aureus in England and Wales:
analysis of death certificates. BMJ 2002;325:1390-1.
13. Wannet W. Virulent MRSA strains containing the Panton Valentine Leukocidin gene in the Netherlands.
Eurosurveillance weekly 7;10. www.eurosurveillance.org/ew/2003/030306.asp Accessed 7 April 2003.
14. Dufour P, Gillet Y, Bes M, , et al. Community-acquired methicillin resistant Staphylococcus aureus in France:
emergence of a single clone that produces Panton Valentine Leukocidin. Clin Infect Dis 2002; 35: 819-24.
15. Gonzalez-Zorn B, Courvalin P. VanA-mediated high level glycopeptide resistance in MRSA. Lancet Infect Dis
2003; 3: 67-8.
16. SCIEH. Community MRSA and Panton-Valentin leukocidin. SCIEH Weekly Report 2002; 36: 298.
http://www.show.scot.nhs.uk/scieh/PDF/pdf2002/0246.pdf. Accessed 7 April 2003.
17. Neeling AJ de, Pelt W van, Hendrix MGR, Buiting AGM, Hol C, Ligtvoet EEJ et al. Antibiotica resistentie in
Nederland. Deel III: Gram-positieve bacterin. Infectieziektenbulletin 1997; 8: 211-215.
18. Goldstein FW, Acar JF. The Alexander Project Collaborative Group. Antimicrobial resistance among lower
respiratory tract isolates of Streptococcus pneumoniae: results of a 1992-93 western Europe and USA
collaborative surveillance study. J Antimicrob Chemother 1996; 38 (suppl A): 71-84.
19. Pallares R, Gudiol F, Linares J, Ariza J, Rufi G, Murgui L et al. Risk factors and response to antibiotic therapy
in adults with bacteremic pneumonia caused by penicillin-resistant pneumococci. N Engl J Med 1987; 317:
18-22.
20. Pallares R, Linares J, Vadillo M, Cabellos C, Manresa F, Viladrich PF. Resistance to penicillin and
cephalosporin and mortality from severe pneumococcal pneumonia in Barcelona in Spain. N Engl J Med
1995; 333: 474-480.
21. Official web site of the Netherlands Society of General Practitioners: http://nhg.artsennet.nl/
index.asp?s=4512. Accessed 6 May 2003.
22. Official SWAB web site: www.swab.nl. Accessed 3 January 2003.
23. State Serum Institut and Danish Veterinary Laboratory, eds. The Copenhagen recommendations the
microbial threat. Ministry of Health, Ministry of Food, Agriculture and Fisheries, 1998.
24. Kahlmeter G, Brown D. Resistance surveillance studies comparability of results and quality assurance of
methods. J. Antimicrob. Chemother. 2002 50: 775-7.
Chapter 2
Objectives and set up of the European
Antimicrobial Resistance Surveillance
System (EARSS)
Adapted from: Bronzwaer SLAM, Goettsch W, Olsson-Liljequist B, Wale
MCJ, Vatopoulos AC, Sprenger MJW. European Antimicrobial Resistance
Surveillance System (EARSS): objectives and organisation.
Eurosurveillance 1999;4;4:41-4, and from Bronzwaer SLAM, Sprenger
MJW. A surveillance system for Europe - textbox. BMJ 1998; 317; 615.
Introduction
In 1997 a prioritisation exercise was carried out among heads of national surveillance
centres in the Member States of the European Union. Antimicrobial resistance ranked in
the top five areas in communicable disease surveillance for which the development of a
network was deemed a high priority (1).
Effective European surveillance must have the agreement and active involvement of all
participants, concluded the Microbial Threat conference on the need for surveillance of
resistant micro-organisms, held in September 1998 in Denmark (2). Patterns of antibiotic
resistance differ widely between member states of the EU (3, 4), and different studies
suggest that policies and guidelines on antibiotic usage may affect the prevalence of
resistance (5, 6). From an epidemiological and methodological standpoint it is difficult to
compare antimicrobial resistance rates because of differences in antimicrobial agents
tested, sampling policies, susceptibility test systems used, and breakpoints adopted.
To obtain more comparable and validated data, the European Commission, Directorate
General Health and Consumer Protection, made funds available to implement a
European Antimicrobial Resistance Surveillance System (EARSS). This system is
coordinated by the Rijksinstituut voor de Volksgezondheid en Milieu (RIVM), the National
Institute of Public Health and the Environment of the Netherlands. In 1998, more than
400 laboratories expressed willingness to take part in this European surveillance network.
This chapter describes objectives and set-up of EARSS.
Objectives
EARSS is an international network of national surveillance systems, aiming to collect
comparable and validated antimicrobial resistance data for public health purposes. Taking
into account laboratory methods as well as epidemiological principles, EARSS will explore
the feasibility of analysing regional differences, assessing risk factors, and providing
electronic feedback. EARSS started on 1 April 1998 with an 18-month feasibility study.
During the first plenary meeting, with a microbiologist and an epidemiologist
representing every country, it was decided that EARSS will concentrate on Streptococcus
pneumoniae and Staphylococcus aureus during the pilot phase; with more pathogens
added later. The system will use routine data from laboratories so that no changes to the
primary diagnostic process will be needed. The participants will gather unbiased samples
of isolates by either total or representative coverage. The objective for S. pneumoniae is to
collect susceptibility data on penicillin and cephalosporins, and possibly other drugs,
from blood and cerebrospinal fluid isolates. For S. aureus, in particular data on methicillin
resistance will be collected from isolates from blood.
EARSS aims to assist in the control of antimicrobial resistance, by performing
antimicrobial resistance surveillance at national and European level, and has set the
following objectives:
collecting susceptibility data in standardised manner, thereby improving comparability
providing information to target interventions (at local, national, and EU level)
providing official national AMR data that constitute a basis for policy decisions
analysing temporal / geographical trends: monitoring AMR data over place (among
different European countries) and time (from year to year).
providing feedback to those who need to know, for evaluating interventions and
follow the effect of policy decisions.
Furthermore, EARSS aims to stimulate:
national antimicrobial resistance surveillance and provision of information for
national policies
linkage of antimicrobial resistance data to antibiotic use data
European research in the field of antimicrobial resistance
Organisation
Each participating country has appointed a national representative microbiologist and a
representative epidemiologist. One of the representatives from each country acts as the
national coordinator. His/her main task is to coordinate activities of the participating
laboratories; arrange distribution and collection of questionnaires on susceptibility
testing; and to collect and forward resistance data each quarter for international collation.
EARSS Network
DG-SANCO (EC)
EARSS Management Team WHO
ESCMID
Public
Plenary Meeting
QA Committee
Advisory board
Lab 2
(28) National Co-ordinating Centres (PH institutes)
National representatives Data managers
Lab ... Lab 1
Figure 2.1: EARSS Network organogram
Standardisation and microbiological quality control methods are being developed in
consultation with the European Society of Clinical Microbiology and Infectious Diseases
(ESCMID). EARSS is a component of the network-of-networks being established by the
World Health Organization (WHO) for global surveillance.
Selection of participating laboratories
EARSS recommended that the national coordinators should select enough laboratories in
their countries to cover at least 20% of the total population. For the community acquired
pathogens the catchment population of the laboratories (the number of people living in the
area they serve) will be considered as the denominator. The 400 or so laboratories
participating in EARSS will cover well over 20% of the population in many countries.
Epidemiological data
EARSS collects the following data by means of isolate record forms and questionnaires:
information about an isolate and its susceptibility test results
information about patients
information about the laboratory methods used and denominator data
data about the hospital(s) served by the laboratory used to generate the denominator.
Isolate record form. This form collects information about patients and isolates. EARSS
requires the following information: sex, month and year of birth, date of specimen
collection, name or code of hospital, hospital department, origin of patient, isolate
specimen number, laboratory code, and antibiotic susceptibility testing results as
specified in the protocol. Furthermore, the isolate record form allows other optional data
to be collected: patient identifier, clinical diagnosis, and susceptibility data for other
antibiotics.
Questionnaire on susceptibility testing. This questionnaire asks about test methods used,
and collects denominator data from a laboratory and from the hospital(s) it serves. The
facilities the hospital offers (intensive care unit, renal, transplant, cardiac surgery) and the
number of bed days are requested. For nosocomial pathogens the number of bed days will
be considered as the denominator. Data on patients and isolates can be related to
information about the laboratory and hospital by means of a unique laboratory code that
will be filled out on all isolate record forms and questionnaires. We are aware that the
catchment population estimated by a laboratory may overestimate the true catchment
population. True catchment populations can be calculated through postal codes of the
patients from whom isolates were obtained. To preserve confidentiality this must be done
at a national level.
Duplicates
To prevent duplicate isolates from being reported, laboratories are asked to send
information only about the first isolate of each strain from each patient. These are referred
to as patient-isolates. To be able to correct for duplicate isolates, the isolate record form
asks for patient ID/code. This is marked as optional information, since in many countries
there are legal limitations on the inclusion of patient identifiers. For the same reason we
do not ask for date of birth, but month and year of birth. A code is needed, however, to
exclude duplicates at the national level. If a patient identifier cannot be used in a particular
country, we ask laboratories to use another (encrypted) code for a specific patient. In
other countries the patient identifier may be used to exclude repeat isolates, removing the
identifier before sending data to the central database.
Data processing
Participating laboratories are offered two methods of data entry: electronically and on
paper. Details vary from country to country, but if a laboratory opts for electronic data
transfer they can use an existing laboratory information system or make use of Whonet
(and/or Whonet-Baclink). WHO revised the existing microbiology laboratory database
software Whonet for EARSS. Laboratories that do not process data electronically will
forward the isolate record forms to their national coordinator, who will perform the data
entry and will send data each quarter to the RIVM in ASCII fixed or tab separated format.
On receipt, the data will be checked for syntax errors (for example, dates and test results).
After this validation, tables, figures, and geographical maps can be generated and
published on the internet site. The aggregated data sets will also be used for more
complex epidemiological studies, for example investigating relationships between
antimicrobial use and resistance.
Feedback
Sufficient and timely feedback is essential for all surveillance systems. Information on
resistance is needed at local, national, and international levels to guide decision making
and interventions. As well as information letters and a newsletter, data will be shared
using the electronic infrastructure Health Surveillance System of Communicable Diseases
(HSSCD) network of the EU. Feedback, in the form of standard reports, is already
provided by means of the EARSS web site, newsletters and publications.
Results
About 400 laboratories will take part by sending data via national coordinators to the
central EARSS database. Data collection began in some countries on 1 October 1998. The
EARSS protocol and issues like ownership of data and data management were agreed by
all national co-ordinators and laid down in the EARSS manual (7). The manual has been
distributed to participating laboratories. By the end of 1999, questionnaires on test
methods and denominators from 283 laboratories had been received. These laboratories
serve 450 hospitals, mainly general hospitals (76%) but also academic/tertiary hospitals
(20%) and nursing homes (4%). Ninety-five per cent of the 150 laboratories that specified
which method they used undertook susceptibility testing of S. aureus against oxacillin
and/or methicillin routinely. About half of these laboratories use Mueller-Hinton agar
(sometimes with salt) and follow the National Committee for Clinical Laboratory
Standards (NCCLS) recommended breakpoints. By the end of the pilot phase,
laboratories from 12 countries (Belgium, Denmark, Germany, Greece, Iceland, Ireland,
Italy, Luxembourg, Netherlands, Portugal, Sweden, United Kingdom) were sending data.
Conclusion
In developing the protocol and questionnaire, the challenge was to balance scientific
validity and feasibility. A first result is that consensus has been reached by leading
microbiologists and epidemiologists on the protocol and logistical framework.
The feasibility phase yielded a conclusion that EARSS is needed and feasible, and that it
must run continuously with guaranteed funding. The number of pathogens under
surveillance will be expanded as soon as the data processing has been optimised. EARSS
is already acting as a catalyst for national surveillance systems, such as in Ireland (8).
References
1. Weinberg J, Grimaud O, Newton L. Establishing priorities for European collaboration in communicable
disease surveillance. Eur J Public Health 1999; 9: 236-40.
2. Thamdrup Rosdahl V, Borge Pederson K. Report from the invitational EU conference on he microbial
threat. September 1998.
3. Rahal K, Wang F, Schindler J, Rowe B, Cookson B, Houvinen P, et al. Reports on surveillance of antimicrobial
resistance in individual countries. Clin Infect Dis 1997; 24 (Suppl 1): S69-75.
4. Kresken M, Wiedemann B. Development of resistance in the past decade in central Europe. J Antimicrob
Chemother 1986; 18(suppl C): 235-42.
5. Seppala H, Klaukka T, Vuopio-Varkila J, Muotiala A, Helenius H, Lager K, et al. The effect of changes in the
consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland. N Engl
J Med 1997; 337: 441-6.
6. Pradier, C, Dunais H, Carsenti-Etesse, Dellamonica P. Pneumococcal resistance in Europe. Eur J Clin
Microbiol Infect Dis 1997; 16: 644-7.
7. EARSS Management Team and national co-ordinators. EARSS Manual 1998.
8. OFlanagan D. Development of a strategy to combat antimicrobial resistance in Ireland. Eurosurveillance
Weekly 1999; 3: 991104. (http://www.eurosurv.org/1999/991104.html)
Chapter 3
Standardisation of Streptococcus
pneumoniae and Staphylococcus aureus
susceptibility data within EARSS
Adapted from: Goettsch W, Bronzwaer SLAM, Neeling de AJ, Wale MCJ,
Aubry-Damon H, Olsson-Liljequist B, Sprenger MJW, Degener JE.
Standardisation and quality assurance for antimicrobial resistance of
Streptococcus pneumoniae and Staphylococcus aureus within the European
Antimicrobial Resistance Surveillance System (EARSS). Clin Microbiol
Infect 2000; 6; 59-63.
Introduction
In several countries of the European Union increased resistance of micro-organisms to
antimicrobial agents is reported; the rise of methillicin-resistant Staphylococcus aureus
infections, the occurrence of vancomycin resistant enterococci and the presence of
penicillin-resistant Streptococcus pneumoniae
1
cause severe problems. In a geographical
context it seems that resistance problems become urgent especially in southern European
countries.
1,2,3,4
From an epidemiological and methodological standpoint the comparison
of antimicrobial resistance from different countries is very difficult. Reasons for these
difficulties are that:
1. Different antimicrobial agents are tested.
2. Different systems for antimicrobial susceptibility testing are used.
3. Different breakpoints for antimicrobial susceptibility are used.
4. Data from point prevalence studies are used for longitudinal comparisons; e.g.
studies on antibiotic resistance performed in 1970 and 1990 are compared, in spite of
differences in study conditions and methodology.
5. Only the resistant strains are tested.
6. Differences between the prevalence of resistant strains from local practices and
university hospitals are not taken into account.
In order to obtain more comparable and validated data, the European Commission has
funded a European Antimicrobial Resistance Surveillance System (EARSS). In this chapter
we present how susceptibility data of Staphylococcus aureus and Streptococcus pneumoniae
within EARSS are standardised in order to address difficulties as mentioned above.
Methods
During the feasibility phase of EARSS it is important to use a limited number of
pathogenic bacterial species, in order to manage the set-up of the surveillance system.
More than 400 laboratories have agreed to participate in this European surveillance
network. EARSS depends on national surveillance data, so input of the participants of the
different member states is essential. The methodology of the surveillance system was
decided during the first plenary EARSS meeting with all national representatives (May 18-
20, 1998). The first decision to be taken was which species are to be to included under
surveillance. Before the meeting a working group prepared a discussion paper with
objective criteria for selection. The rationale of selection of a community acquired
pathogen and a hospital-acquired (nosocomial) pathogen for the pilot phase of EARSS is
summarised in the discussion paper:
1. Relevance for Public Health. Most participants believe that S. pneumoniae and S.
aureus are the most relevant species. They are both proven pathogens, clinically
relevant on population level for the community or the hospitalised population, have a
high potential for spread in community and/or hospital setting and are known to
acquire resistance against currently used and recommended antibiotics. Other
relevant species are Campylobacter jejuni, Haemophilus influenzae, Streptococcus
pyogenes, Pseudomonas aeruginosa and Escherichia coli.
2. Political Interest. Most participants believe that especially S. aureus, S. pneumoniae
and Enterococcus faecium/faecalis are of interest to policy makers. These species are
the ones most often attracting media and political attention. In addition, in relation to
the present discussion on the use of fluoroquinolones in animals and humans,
resistance in micro-organisms such as Salmonella typhimurium, C. jejuni and E. coli are
of interest.
3. Availability of quantitative data. The participants believe that quantitative resistance
data in sufficient numbers are present for several species, including S. aureus, S.
pneumoniae, H. influenzae and E. coli.
4. Reliability of data. For some species, such as E. coli in urinary tract infections, sample
bias can occur in resistance surveillance. Physicians only send isolated bacteria to the
laboratories when they have patients with treatment failures. For S. aureus and S.
pneumoniae blood isolates that always cause patients to be severely ill, irrespective of
the susceptibility of the isolate, it is to be expected that clinicians in the hospital send
all isolates for susceptibility testing.
5. Quality assurance. S. aureus, S. pneumoniae, E. faecium/faecalis, E. coli, P. aeruginosa
and H. influenzae are common micro-organisms and are often included in quality
assurance systems. Therefore susceptibility data for these species are normally
quality-assured.
6. Interaction with other resistance surveillance systems. For some species such as
Salmonella spp., Mycobacterium tuberculosis and Neisseria gonorrhoea other well-
organised European surveillance systems are already in place, like Enternet
(www.phls.co.uk/inter/enter-net/menu.htm) and EuroTB (www.eurotb.org).
Results consensus meeting
Having circulated the discussion paper on the rationale for selection of pathogens for
comments before, consensus was reached relatively fast during the meeting. Realising
that in the phase of establishing a surveillance programme not the most complex species
should be selected, it was decided that S. pneumoniae and S. aureus are the most relevant
species for the pilot phase in EARSS. In the future it is expected that EARSS will be a
permanent surveillance system for the most public health relevant species and after the
feasibility phase more species will be added.
In order to minimise sample bias, it was decided to test only S. pneumoniae isolates from
blood and cerebral spinal fluid (CSF), and S. aureus isolates from blood for antimicrobial
resistance. During the same meeting the protocols for antimicrobial testing of S. aureus
en S. pneumoniae were developed.
Protocol Staphylococcus aureus testing
Objective
To study the (methicillin)-resistance of S. aureus, in blood isolates in hospitals in Europe.
Case definition
Resistance data on the first isolate only of each strain from the blood of each patient with a S. aureus
infection (confirmed by a coagulase test). We exclude duplicate isolates of the same species from the
same patient, and collect information only on the first isolate from each patient (patient-isolate).
Test procedure
(1) Oxacillin screen plates (6 g/ml according to NCCLS) or oxacillin disks (1 g or 5 g) will be used.
When S. aureus is tested for oxacillin resistance a disk with a load of 1 g oxacillin (NCCLS) is used;
non-susceptibles are strains with a zone size of 10 mm or less ( 10 mm). When a disk with a load of
5 g oxacillin (according to the French guidelines, SFM) for oxacillin susceptibility testing is used, non-
susceptibles are isolates with a zone size of 19 mm or less ( 19 mm).
(2) In the case of oxacillin non-susceptible S. aureus, the participating laboratories are asked additionally
to determine the MIC for oxacillin (range of dilutions: 0.016-256) and MIC for vancomycin specifying
the method used: agardilution, microdilution or E-test (range of dilutions: 0,016 256). A participating
country can decide whether the local laboratory will perform the second step of the protocol or that a
reference laboratory will collect the non-susceptible strains and perform the MIC for oxacillin and
vancomycin.
S. aureus (blood)
oxacillin agar screen plate or oxacillin disk (1)
susceptible non-susceptible
MIC oxacillin or PCR mecA-gene (2) MIC vancomycin (2)
Protocol Streptococcus pneumoniae testing
Objective:
To study the penicillin-resistance of S. pneumoniae blood- and CSF-isolates in Europe.
Case definition
Resistance data on the first isolate only from the blood or CSF of each patient with a S. pneumoniae
infection (confirmed by an optochin test). We exclude duplicate isolates of the same species from the
same patient, and collect information only on the first isolate from each patient (patient-isolate).
Test procedure
(1) For testing of S. pneumoniae an oxacillin disk (1 g or 5 g) will be used. When S. pneumoniae is
tested, non-susceptible penicillin resistant S. pneumoniae are strains with a zone size of 20 mm or less
( 20 mm). An alternative in oxacillin susceptibility testing is a disk with a load of 5 g oxacillin
(according to the French guidelines, SFM). Non-susceptible penicillin resistant S. pneumoniae are
isolates with a zone size of 26 mm or less ( 26 mm).
(2) In the case of oxacillin non-susceptible S. pneumoniae, the participating laboratories are asked
additionally to determine the MIC of penicillin, cefotaxime or ceftriaxone and ciprofloxacin, specifying
the method used: agardilution, microdilution or E-test (range of dilutions: 0,016256 (penicillin) or
0,00232 (cefotaxime/ ceftriaxone and ciprofloxacin)). A participating country can decide whether the
local laboratory will perform the second step of the protocol or that a reference laboratory will collect
the non-susceptible strains and perform the MIC for penicillin, cefotaxime/ceftriaxone and
ciprofloxacin.
Discussion
EARSS is designed in order to minimise epidemiological or microbiological difficulties as
were summarised in the introduction:
S. pneumoniae (blood + CSF)
oxacillin disk (1)
susceptible non-susceptible
MIC penicillin (2) MIC cefotaxime/ceftriaxone (2) MIC ciprofloxacin (2)
1. Different antimicrobial agents are tested.
In EARSS, resistance for two species (S. aureus and S. pneumoniae) is tested against a
restricted set of specified antimicrobials. The choice for oxacillin, instead of
methicillin, for determination of MRSA (ORSA) is a practical one. Because methicillin
is becoming less available in the near future, we think that oxacillin is a reliable
alternative. For S. pneumoniae, testing of oxacillin, as a first step, in combination with
a penicillin minimum inhibitory concentration (MIC) for non-susceptibles is now
generally accepted.
5
We believe that the introduction of a new generation of
fluoroquinolones for the therapy of respiratory tract infections necessitates us to
follow ciprofloxacin resistance in S. pneumoniae.
2. Different systems for antimicrobial susceptibility testing are used.
The protocols for S. pneumoniae and S. aureus are clearly defined. Next to a simple first
line screening method, a second step, in which the MIC is determined, is included.
Such a protocol combines easy accessibility with careful quantitative examination of
antimicrobial resistance.
For a reliable comparison of resistance against oxacillin in S. aureus oxacillin agar
screen plates can be used.
6,7
However, results from a survey among national co-
ordinators illustrate that agar screen plates are only used in a few countries.
Because one of the key features of EARSS is easy accessibility, the protocol will also
accept data from the oxacillin disk diffusion test.
7,8
The golden standard for confirmation of an MRSA is testing for the presence of the
mecA-gene. However, when a participating laboratory is not able to perform a PCR,
determination of a MIC for oxacillin (range of dilutions: 0.016-256) will be done to
confirm that an MRSA is not false positive.
Testing of MRSA for resistance against vancomycin is very relevant but under
debate. Vancomycin intermediate resistant S. aureus (VISA) strains, which were
first reported in Japan
9
, are often heterogeneously resistant. Only a very limited
percentage of the total population of isolated bacteria is intermediately
resistant.
9,10
The presence of these VISAs can be missed measuring a MIC under
standard conditions. At this moment there is not an established protocol to test
for VISA. We propose to test the MRSA for vancomycin using the E-test, with a
standardised protocol which is also used testing the oxacillin MIC, realising that
some intermediate VISA strains might be missed. The determination of the
vancomycin MIC will preferably be done at a central reference lab in each country.
In case of finding a VISA strain, arrangements will be made for further analysis
(e.g. sequence analysis).
3. Different breakpoints for antimicrobial susceptibility are used.
Breakpoints are defined in the two protocols, according to US - National Committee
for Clinical Laboratory Standards (NCCLS) or in some cases the Socit Francaise de
Microbiologie (SFM) guidelines.
For all S. pneumoniae and S. aureus isolates, we ask the participating laboratories
to register the inhibition zone (in case of the disk method). The collection of zone
diameters has an additional value in case medium and disk load are standardised.
Firstly, zone diameters will give more insight in the distribution of S. pneumoniae
or S. aureus strains with different susceptibilities to oxacillin, e.g. high resistance
versus intermediate resistance.
11
Secondly, the distribution of zone diameters may
be used to study the quality of resistance data from different participating
laboratories.
12
It is acknowledged that laboratories in some participating countries
are not able to collect zone diameters.
In the second step (MIC testing), the validity of categorising the strains as
susceptible or non-susceptible, according to the SFM and the NCCLS guidelines,
is evaluated. Correction of false positive (resistant) strains is possible by MIC
testing.
Also a monthly testing of quality control strains assesses correct use of
breakpoints for the categorisation of strains into susceptible and resistant.
4. Data from point prevalence studies are used for longitudinal comparisons; e.g.
studies on antibiotic resistance performed in 1970 and 1990 are compared, in spite
of differences in study conditions and methodology.
For a longitudinal analysis on developments in resistance continuous data are
essential. EARSS wants to provide continuous data that is generated according to a
standardised protocol. Sudden increases or decreases in resistance percentages could
be caused by changes in the surveillance method and should be closely monitored.
5. Only the resistant strains are tested.
Selection of strains can easily occur in case of less invasive infections. For instance,
antimicrobial susceptibility testing of Enterobacteriaceae from urinary tract infections
depends on the response of the patient on the initial therapy. Resistance testing will
be more likely performed when the patient returns after therapy failure. Resistance
surveillance on basis of routine samples may overestimate the problem.
13,14
In the
EARSS pilot, sampling of the species is restricted to invasive isolates, which are
routinely tested for antimicrobial susceptibility.
6. Resistance in local practices or general hospitals is compared to resistance in
university hospitals.
In order to tackle this problem we asked the national co-ordinators to ensure
reasonable coverage in their country. In case of the community-acquired pathogen, a
coverage of more than 20% of the total national population is necessary. Participating
laboratories are providing the national co-ordinator with information on the
catchment population (the number of people living in the area they serve). In case of
S. aureus we believe that 20% of the total number of patient-days in every country is a
minimum. Therefore, the participating laboratories provide the national co-ordinators
with information on the number of patient-days of every hospital they serve. The
national co-ordinator selects laboratories; not only laboratories that serve university
hospitals but also laboratories, which serve small regional hospitals and general
practitioners, are part of EARSS.
We believe that EARSS can evolve as a good framework to monitor antimicrobial
resistance in the EU for the coming years. The first result of EARSS is that this system has
activated several countries to establish or to update their national resistance surveillance
system in order to follow national resistance patterns and to compare these to
developments in Europe.
Addendum
At the third plenary EARSS meeting in November 2000 it was decided with all the national
representatives to extend surveillance to three other species: E. coli, E. faecium and E.
faecalis. The protocol for testing for these species was agreed and most countries started
data collection in January 2001.
The EARSS manual was updated accordingly and sent to the participating laboratories.
Next to testing protocols the EARSS Manual 2001 provides an overview of the
organisation and infrastructure of EARSS, and of data management.
15
In annex it provides
the data exchange format, updated isolate record forms, an updated laboratory/hospital
questionnaire and a template Memorandum of Understanding between national EARSS
representatives and participating laboratories.
In further chapters susceptibility results will be presented only for S. pneumoniae and S.
aureus, for which data collection began in 1999.
References
1. Appelbaum PC. Antimicrobial resistance in Streptococcus pneumoniae: an Overview. Clin Infect Dis 1992; 15:
77-83.
2. Goldstein FW, Acar JF. Antimicrobial resistance among lower respiratory tract isolates of Streptococcus
pneumoniae: results of a 1992-93 western Europe and USA collaborative surveillance study. The Alexander
Project Collaborative Group. J Antimicrob Chemother 1996: 38 (Suppl A): 71-84.
3. Felmingham D, Gruneberg RN. A multicentre collaborative study of the antimicrobial susceptibility of
community-acquired, lower respiratory tract pathogens: The Alexander Project. J Antimicrob Chemother
1996; 38 (Suppl A): 1-57.
4. Rahal K, Wang F, Schindler J, Rowe B, Cookson B, Houvinen P, Marton A, Lalitha MK, Semina N, Kronvall
G, Guzman M. Reports on surveillance of antimicrobial resistance in individual countries. CID 1997;
24(Suppl 1): S69-S75.
5. Sinave C, Jette LP. Use of oxacillin disk screening test for detection of penicillin- and cefalosporin-resistant
pneumococci. Abstract ICAAC, September 1998, San Diego, D-67.
6. Frebourg NB, Nouet D, Lemee L, Martin E, Lemeland JF. Comparison of ATB Staph, Vitek, and E-test
methods for detection of oxacillin heteroresistance in staphylococci possessing mecA. J Clin Microbiol 1998;
36: 52-57.
7. Cormican MG, Wilke WW, Barrett MS, Pfaller MA, Jones RN. Phenotypic detection of mec A-positive
staphylococcal blood stream isolates: high accuracy of simple disk diffusion tests. Diagn-Microbiol-Infect-
Dis. 1996 Jul; 25(3): 107-112
8. Ramotar K, Bobrowska M, Jessamine P, Toye B. Detection of methicillin resistance in coagulase-negative
staphylococci initially reported as methicillin susceptible using automated methods. Diagn-Microbiol-
Infect-Dis. 1998 Apr; 30(4): 267-273.
9. Hiramitsu K, Hanaki H, Ino T, Yabuta K, Oguri T, Tenover FC. Methicillin-resistant Staphylococcus aureus
clinical strain with reduced vancomycin susceptibility. J Antimicrob Chem 1997; 40: 135-136.
10. Hiramitsu K, Aritaka N, Hanaki H, Kawasaki S, Hosoda Y, Hori S, Fukuchi Y, Kobayashi I. Dissemination in
Japanese hospitals of strains of Staphylococcus aureus heterogeneously resistant to vancomycin. Lancet
1997; 350: 1670-1673.
11. Ringertz S, OlssonLiljequist B, Kahlmeter G, Kronvall G. Antimicrobial susceptibility testing in Sweden II.
Species-related zone diameter breakpoints to avoid interpretive errors and guard against unrecognized
evolution of resistance. Scand J Infect Dis 1997; Suppl. 105 : 8-12.
12. Blanc DS, Petignat C, Moreillon P, Wenger A, Bille J, Francioli P. Quantitative antibiogram as a typing
method for the prospective epidemiological surveillance and control of MRSA: Comparison with molecular
typing. Infect Cont Hosp-Epidemiol 1996; 17: 654-659.
13. Neeling de AJ, Pelt van W, Hendrix MGR, e.a. Antibiotic resistance in the Netherlands. Part II: Gram-
negative bacteria [NL]. Antibioticumresistentie in Nederland. Deel II : Gram-negatieve bacterin.
Infectieziekten Bulletin 1997; 8: 192-195.
14. Neeling AJ de, Jong de J, Overbeek BP, Bruin RW de, Dessens-Kroon M, Klingeren B van. Quantitative
susceptibility research with intra- and extramural Escherichia coli isolates [NL]. Kwantitatief
gevoeligheidsonderzoek met intra- en extramurale isolaten van Escherichia coli. RIVM report nr 359001002,
November 1990.
15. EARSS Management Team and Advisory Board EARSS, in collaboration with national representatives.
EARSS Manual 2001. January 2001. Pages: 56. Available at www.earss.rivm.nl. Accessed 1 May 2003.
Chapter 4
Comparability of antimicrobial
susceptibility test results from 22
European countries and Israel:
an external quality assurance exercise of
EARSS in collaboration with UK NEQAS
S. Bronzwaer, U. Buchholz, P. Courvalin, J. Snell, G. Cornaglia,
A. de Neeling, H. Aubry-Damon, J. Degener, and EARSS participants.
Journal of Antimicrobial Chemotherapy (2002) 50, 953964
Abstract
The goal of this exercise was to organize external quality assurance (QA) of antibiotic
susceptibility testing for laboratories participating in EARSS and to assess the
comparability of susceptibility test results across countries, and guidelines. In September
2000, UK NEQAS distributed a set of three Streptococcus pneumoniae strains, two
Staphylococcus aureus strains and one Staphylococcus haemolyticus strain. Laboratories
reported the guideline followed, the interpretation of the susceptibility test result and the
MIC, if tested. In this study we considered results concordant if the reported
interpretation of the participating laboratory agreed with the designated interpretation of
reference laboratories. Overall, 433 (92%) of 471 laboratories from 23 countries reported
back. Of the 8685 tests that were assessed, 8322 (96%) were interpreted correctly by the
participants. Concordance for detection of penicillin non-susceptibility in the three S.
pneumoniae strains was 96%, 90% and 87%, respectively. Laboratories performed
extremely well in detecting oxacillin resistance in the homogeneously methicillin-resistant
S. aureus (MRSA) strain, but the concordance rate dropped from 100% to 77% in the
heterogeneously resistant MRSA strain. Concordance for detection of teicoplanin
resistance in the S. haemolyticus strain was 82%. We stratified concordance rates first
for country and then for guideline used, but observed only minor differences
among countries and guidelines. Quantitative methods yielding an MIC were more
concordant than non-MIC methods for penicillin resistance in the
S. pneumoniae strains (94% versus 79%). The NCCLS guideline was the most frequently
followed, by 61% of laboratories from 19 countries. This exercise shows that, overall,
countries participating in EARSS are capable of delivering susceptibility data of good
quality.
The comparability of susceptibility data for penicillin resistance in S. pneumoniae and for
homogeneous methicillin resistance in S. aureus is satisfactory among European
countries and across guidelines. However, we emphasize the importance of determining
an MIC for suspected penicillin non-susceptible S. pneumoniae and for suspected
glycopeptide non-susceptible S. aureus. Laboratories, particularly in some countries, may
need to improve their capability to detect oxacillin resistance in heterogeneously resistant
MRSA. For continuous external quality assessment we recommend that laboratories
participate in national and international schemes with frequent distribution of control
strains.
Introduction
Since 1999, the European Antimicrobial Resistance Surveillance System (EARSS) has
been monitoring antimicrobial resistance in an increasing number of European countries.
Funded by the European Commission, EARSS is an international network of national
surveillance systems aiming at collecting comparable and valid resistance data. The
purpose of EARSS is to document variations in antimicrobial resistance over time and
place, to provide the basis for policy decisions and assess the effectiveness of
interventions. EARSS is an ongoing system monitoring resistance of invasive infections of
Streptococcus pneumoniae and Staphylococcus aureus. Since 2001, invasive isolates of
Escherichia coli and enterococci have also been under surveillance, and a similar external
quality assurance exercise for these pathogens was organized in September 2001.
Summary results from the 2001 quality assurance (QA) exercise as well as the EARSS
database are accessible through the EARSS web site (www.earss.rivm.nl).
Antibiotic susceptibility of clinical isolates of bacteria is usually tested as part of routine
laboratory investigations to establish the most adequate therapy for an infection.
Detection of resistance relies on specimen collection from the patient, isolation,
identification and susceptibility testing of the bacterial pathogen. Only recently a
reference method for the determination of minimum inhibitory concentrations (MIC) has
been proposed by the European Committee for Antimicrobial Susceptibility Testing
(EUCAST),
1
but there is still no European agreement on breakpoint criteria for
interpreting the results into clinical categories [susceptible (S), intermediate (I), or
resistant (R)]. As a result, methods for most agents still differ between countries, and
interpretation of test results may differ.
The goal of this exercise was to organize external quality assurance of antibiotic
susceptibility testing for laboratories participating in EARSS and to assess the
comparability of susceptibility test results, as collected according to the EARSS protocol
2
across countries and guidelines. Furthermore, this exercise assessed the comparability of
MIC-yielding methods versus non-MIC-yielding methods (e.g. agar diffusion tests), and
provided an overview of the frequency of use of various guidelines among EARSS
laboratories. Quality assessment is essential in order for EARSS to assess the validity of
comparing S. pneumoniae and S. aureus susceptibility data from a large number of
laboratories from numerous countries and pooling it into a European database.
Materials and methods
A set of six strains (three S. pneumoniae, two S. aureus and one Staphylococcus
haemolyticus) was provided by the French Reference Center for AntibioticsInstitut
Pasteur. The strains were characterized and tested by three reference laboratories: one in
France, one in Italy, and one in The Netherlands. MICs were determined by an agar
dilution method in two laboratories and by Etest in the third. Each reference laboratory
interpreted the results according to its own breakpoint criteria, respectively: Comit de
lAntibiogramme de la Socit Franaise de Microbiologie (CA-SFM), National Committee
for Clinical Laboratory Standards (NCCLS) and the Commissie Richtlijnen
Gevoeligheidsbepalingen (CRG). A designated interpretation and a reference MIC was
determined for every organismantimicrobial combination. In cases where there were
differences in MIC between reference laboratories of more than one dilution step, strains
were tested repeatedly until agreeing on a reference MIC or accepting a narrow MIC
range of reference laboratories.
The S. pneumoniae strains UA1283 and UA347 were intermediately resistant to penicillin
G, and S. pneumoniae strain UA1449 was fully penicillin resistant. S. aureus strain UA1432
was homogeneously resistant to methicillin, and strain UA1450 was a heterogeneously
resistant MRSA strain. The phenotypic expression of methicillin resistance of the S. aureus
strains was analysed by performing two independent population analyses on agar plates
containing different concentrations of the antibiotic, as described by Tomasz et al.
3
The S.
haemolyticus strain UA1434 was resistant to teicoplanin. The United Kingdom National
External Quality Assurance Scheme (UK NEQAS) reference laboratory at the Central
Public Health Laboratory, Colindale, London, organized the logistics of this study and
arranged the shipment of the strains. The strains were prepared as freeze-dried cultures
and sent by air-freight to EARSS national co-ordinating centres in 23 countries, who
distributed the strains to the 471 laboratories participating in EARSS. Laboratories were
asked to identify the control strains and to test them for susceptibility to specified
antimicrobials using their routine procedures (for invasive specimens). They were asked
to report the clinical categorization (S, I or R) and the MIC, if performed, as well as the
breakpoints and guideline(s) followed. Five weeks were allowed for return of the results
to UK NEQAS. Immediately after the closing date for return of results, brief details of the
intended results were posted to participant laboratories, sent by e-mail to participants
with e-mail addresses, and made available on the UK NEQAS web site. Laboratories
received their individual results and a summary of the aggregate results.
4
Where 10 or
more laboratories within a country participated, tables of coded results specific to the
country were produced.
Analysis comprised three parts: bacterial identification, antimicrobial susceptibility test
results and the use of guidelines. We assessed participants results as being concordant
or discrepant with the designated interpretation where all three reference laboratories
agreed on the interpretation (S or I/R), and where the range of MICs of the reference
laboratories allowed unambiguous interpretation by different guidelines.
For assessing concordance we used only two categories: susceptible (S) versus non-
susceptible [i.e. intermediate plus resistant (I/R)]. Results were assessed for correct
interpretation of susceptibility/non-susceptibility for oxacillin, penicillin G and
erythromycin against S. pneumoniae; for oxacillin, methicillin, gentamicin, vancomycin,
teicoplanin and erythromycin against the S. aureus strains; and for gentamicin,
vancomycin, teicoplanin and erythromycin against the S. haemolyticus strain.
In this study we considered results concordant if the reported interpretation of the
participating laboratory agreed with the designated interpretation of the reference
laboratories. The term concordance rate denotes the proportion of susceptibility tests
with a correct result. For each countryexcept for France, Hungary and Malta, with only
one laboratory participatingwe calculated the average concordance of participating
laboratories. We also calculated for every guideline the average of the concordance of
laboratories following that guideline, using Microsoft Excel (Microsoft Corporation,
Release 97 SR-2; Redmond, WA, USA). We used SAS software (SAS Institute Inc., Release
8.01; Cary, NC, USA) for the calculation of the confidence intervals (CI), weighting the
results for the number of tests performed in each country and considering that
observations within one country are not independent.
Results
The designated interpretations and MIC reference values for the strains investigated are
listed in Table 4.1. Overall, 433 (92%) of 471 laboratories from 23 countries reported
results (Table 4.2). Analysis of results at a national level from countries where only one
laboratory participated (France, Hungary and Malta) are not presented, for confidentiality
reasons and also because the results from one laboratory may not be a true
representation of national performance.
Bacterial identification
Strains were identified at the genus and species level. The three S. pneumoniae strains
were correctly identified by: 425/428 (99%), 421/425 (99%) and 413/419 (99%) of the
participating laboratories. Twelve laboratories from different countries did not identify
one of the three S. pneumoniae strains correctly at species level and one laboratory failed
to identify the genus correctly. Four laboratories identified one of the strains as
Streptococcus mitis, four as Streptococcus viridans, two as Streptococcus sanguis, one as
Streptococcus oralis, one as Streptococcus sp., and one as Aerococcus sp.
The two S. aureus strains were correctly identified by: 422/427 (99%) and 422/423 (100%)
of the laboratories. Three laboratories identified one of the strains as coagulase-negative
staphylococci, two as S. haemolyticus and one as Staphylococcus intermedius.
The S. haemolyticus strain was identified by 364/424 (86%) of the laboratories as S.
haemolyticus or coagulase-negative staphylococcus and by 46/424 (11%) of the
laboratories as staphylococcus species other than S. aureus. Thirteen laboratories (3%)
misidentified the strain as S. aureus and one laboratory misidentified it as Enterococcus
faecalis.
Table 4.1. The designated interpretation and reference MIC or MIC range of reference laboratories for every
organismantimicrobial combination that was assessed
Designated interpretation Reference MIC or MIC range (mg/L)
Strain UA1449, S. pneumoniae
oxacillin R 16
penicillin G R 34
erythromycin R >256
oxacillin R 2
penicillin G I 0.250.5
erythromycin R >256
oxacillin R 4
penicillin G I 0.5
erythromycin R >256
Strain UA1432, S. aureus
oxacillin R >256
methicillin R >256
gentamicin R 64128
vancomycin S 2
erythromycin R >256
Strain UA1434, S. haemolyticus
gentamicin R 64
vancomycin S 24
teicoplanin R 3264
erythromycin R 64256
Strain UA1450, S. aureus
oxacillin R 864
methicillin R 3264
gentamicin S 0.120.25
vancomycin S 1
teicoplanin S 0.51
erythromycin S 0.25
Antimicrobial susceptibility testing
Of the 8685 tests that were reported and assessed in this exercise, 8322 (96%) were
interpreted correctly by the participants. The average of the concordance of all
antimicrobial test results across countries surpassed 90% in all control strains (Figure
4.1). This figure shows for every control strain the average and the range across countries
of the concordance of all antimicrobial test results that were assessed. The lower end of
the ranges in the S. pneumoniae strains varied between 90% and 72%. In the S. aureus
strains the lower ends ranged between 96% and 82%. We found similar results after
stratification for guidelines (Figure 4.2), with the lowest concordance rate of 67% for one
guideline for strain UA347.
Oxacillin, penicillin G and erythromycin susceptibility in S. pneumoniae. The overall
concordance rate to detect penicillin non-susceptibility with an oxacillin screen disc was
97%, and ranged from 96% to 99% for the three strains tested (Table 4.3). The
performance of most countries was excellent, although the group of Greek laboratories
showed a lower concordance rate. Ninetyseven per cent of the participants used an
oxacillin disc loaded with 1 g.
Table 4.2. Proportion of participants returning reports specified per country
Number of Number of Number of Number of
QA samples returning QA samples returning
Country sent reports (%) Country sent reports (%)
Austria 11 10 (91) Israel 3 3 (100)
Belgium 59 57 (97) Italy 63 53 (84)
Bulgaria 23 20 (87) Luxembourg 5 5 (100)
Czech Republic 34 33 (97) Malta 1 1 (100)
Denmark 5 5 (100) Netherlands 27 25 (93)
Germany 35 31 (89) Poland 20 19 (95)
Finland 29 25 (86) Portugal 20 16 (80)
France 1 1 (100) Slovenia 10 10 (100)
Greece 18 17 (94) Spain 32 31 (97)
Hungary 1 1 (100) Sweden 26 25 (96)
Iceland 3 3 (100) UK 25 24 (96)
Ireland 20 18 (90) Total 471 433 (92)
QA, quality assurance
0
UA347 UA1283 UA1449 UA1432 UA1450 UA1434
90
80
70
60
40
50
30
20
10
100
S. pneumoniae S. aureus S. haemolyticus
c
o
n
c
o
r
d
a
n
c
e

(
%
)
0
UA347 UA1283 UA1449 UA1432 UA1450 UA1434
90
80
70
60
40
50
30
20
10
100
S. pneumoniae S. aureus S. haemolyticus
c
o
n
c
o
r
d
a
n
c
e

b
y

g
u
i
d
e
l
i
n
e

(
%
)
Figure 4.1. The average and the range across countries of the con cordance of antimicrobial test results,
specified for every control strain.
Figure 4.2. The average and the range across guidelines of the concordance of antimicrobial test results,
specified for every control strain.
The proportion of laboratories that reported penicillin non-susceptibility correctly after
testing penicillin G varied from 96% for strain UA1449 to 90% for strain UA1283 and 87%
for strain UA347 (Table 4.4). The peak of the frequency distribution of the penicillin G
MICs for strain UA1449 (2 mg/L), yielding the highest concordance is well in the non-
susceptible region compared with the peak of the frequency distribution of strain UA347
(0.25 mg/L), yielding a lower concordance. Again, almost all countries showed high
concordance rates, with the exception of Bulgaria and Greece.
The overall concordance rate for the detection of penicillin G non-susceptibility among
guidelines followed in Europe was 91%, as specified for all three S. pneumoniae strains
and for every guideline in Table 4.5. Again, performance was best for strain UA1449 and
decreased for the other two test strains.
Guidelines yielding somewhat lower concordance rates, such as those set by the Deutsches
Institut fr Normung (DIN), as well as the guidelines specified under Other [Czech 98 and
Mesa Espaola de Normalizacion de la Suseptibilitad y Resistencia a los Antimicrobianos
(MENSURA)] were used by only a few participants. The guideline used most frequently for
penicillin testing of S. pneumoniae was NCCLS, with an average concordance of 91%.
Erythromycin resistance in all three S. pneumoniae control strains was detected correctly
by 99% of the participants.
Oxacillin, gentamicin, erythromycin, teicoplanin and vancomycin susceptibility in S. aureus.
The overall concordance for detection of oxacillin (i.e. methicillin) resistance in the
homogeneously resistant S. aureus strain UA1432 was 100%. The overall concordance for
the heterogeneously resistant MRSA strain UA1450 was much lower, at 77%. Three
countries (Czech Republic, Greece and Iceland) yielded notably lower concordance rates
(Table 4.6), but no difference was found among different guidelines followed in Europe
(data not shown).
We found a very high concordance for detection of vancomycin susceptibility in the two
MRSA strains, of 98% and 100%, respectively.
Respectively 100% and 99% of the participants detected erythromycin and gentamicin
resistance in the homogeneously resistant MRSA strain. Erythromycin and gentamicin
susceptibility in the heterogeneously resistant MRSA strain was interpreted correctly by
98% and 99% of the participants, respectively. For teicoplanin susceptibility in the
heterogeneously resistant MRSA strain, we found a concordance rate of 100%.
Teicoplanin, vancomycin, gentamicin and erythromycin susceptibility in S. haemolyticus. For
detection of teicoplanin resistance in the S. haemolyticus strain, the overall concordance
rate was 82% (Table 4.7), with five countries (Belgium, Denmark, Luxembourg, The
Netherlands and UK) scoring low concordance rates. Vancomycin susceptibility of this
strain was interpreted correctly by 94% of the participants.
Table 4.3. Detection of penicillin non-susceptibility in S. pneumoniae by country as tested with an oxacillin screen
disc
Strain UA1449 Strain UA1283 Strain UA347
(oxacillin MIC of (oxacillin MIC of (oxacillin MIC of
ref. labs: 16 mg/ L; ref. labs: 2 mg/ L; ref. labs: 4 mg/ L;
intended intended intended
interpretation: interpretation: interpretation:
resistant) resistant) resistant) Total
number number number
of labs of labs of labs total
doing % doing % doing % of %
Country test correct test correct test correct tests correct
Austria 10 100 10 90 10 100 30 97
Belgium 51 100 50 98 48 100 149 99
Bulgaria 16 100 16 94 15 100 47 98
Czech R. 28 100 28 100 28 100 84 100
Denmark 5 100 4 100 5 100 14 100
Finland 21 100 21 95 20 95 62 97
Germany 30 100 30 87 30 97 90 94
Greece 12 83 12 75 12 67 36 75
Iceland 3 100 3 100 3 100 9 100
Ireland 17 100 17 100 17 100 51 100
Israel 3 100 3 100 3 100 9 100
Italy 43 98 42 95 41 95 126 96
Luxembourg 4 100 4 100 4 100 12 100
Netherlands 22 100 22 95 22 91 66 95
Poland 17 100 17 100 17 100 51 100
Portugal 15 100 15 100 15 100 45 100
Slovenia 7 100 7 100 7 100 21 100
Spain 23 100 23 96 22 100 68 99
Sweden 20 100 20 100 20 100 60 100
UK 21 100 21 100 20 100 62 100
Overall concordance 99 96 97 97
95% Confidence interval 94 99
Table 4.4. Detection of penicillin non-susceptibility in S. pneumoniae by country after testing for penicillin G
(penicillin MIC of (penicillin MIC of (penicillin MIC of
ref. labs: 4 mg/ L; ref. labs: 0.5 mg/ L; ref. labs: 0.5 mg/ L;
resistant) intermediate) intermediate) Total
Austria 8 100 8 100 8 88 24 96
Belgium 49 98 48 90 47 87 144 92
Bulgaria 11 82 11 64 11 55 33 67
Czech R. 28 100 27 93 27 93 82 95
Denmark 4 100 4 100 4 100 12 100
Finland 22 95 22 91 21 86 65 91
Germany 29 93 28 82 28 79 85 85
Greece 13 85 13 69 13 54 39 69
Iceland 2 100 2 100 2 100 6 100
Ireland 17 94 17 94 17 94 51 94
Israel 3 100 3 100 3 100 9 100
Italy 47 87 47 79 45 78 139 81
Luxembourg 5 100 5 100 5 100 15 100
Netherlands 24 100 23 96 24 83 71 93
Poland 17 100 17 100 17 100 51 100
Portugal 14 100 14 100 14 100 42 100
Slovenia 10 100 10 100 10 100 30 100
Spain 30 97 30 97 30 93 90 96
Sweden 20 100 21 100 21 100 62 100
UK 22 100 22 95 21 95 65 97
Table 4.5. Detection of penicillin non-susceptibility in S. pneumoniae by guidelines used
(penicillin MIC of (penicillin MIC of (penicillin MIC of
ref. labs: 4 mg/ L; ref. labs: 0.5 mg/ L; ref. labs: 0.5 mg/ L;
resistant) intermediate) intermediate) Total
BSAC 10 100 9 100 7 100 26 100
CRG 7 100 7 100 7 100 21 100
DIN 10 100 10 70 10 70 30 80
NCCLS 180 94 170 92 163 87 513 91
SRGA 13 100 14 100 13 100 40 100
Other 23 100 25 76 23 70 71 82
Not indicated 135 96 140 91 148 89 423 92
Other = participants using NeoSensitabs or the Stokes method or following a different guideline or following
more than one guideline. For abbreviations see footnotes to Table 4.9.
Gentamicin and erythromycin resistance were detected by 97% and 99% of participants,
respectively. The overall concordance for detection of oxacillin susceptibility in the S.
haemolyticus strain was 83%, with three countries (Bulgaria, Israel and Slovenia) clearly
scoring lower.
Concordance of MIC yielding methods versus non-MIC methods. Of the 433 laboratories
participating, 375 used methods yielding an MIC: 11 (3%) used agar dilution, 21 (6%)
(micro-) broth, 202 (54%) Etest and 52 (14%) used exclusively an automated method.
Eightynine laboratories used more than one method. One-quarter of all laboratories
(110/433) made use of an automated method. The most frequently used system was one
of the different generations of bioMrieux Vitek (Table 4.8). The MICs determined by
automated systems should be considered as semi-quantitative data because only a very
limited range of dilutions is used. However, in this study we group automated methods
under the MIC yielding methods as opposed to non-MIC methods.
Table 4.6. Detection of oxacillin non-susceptibility in S. aureus by country
Strain UA1432 Strain UA1450
(oxacillin MIC of ref. labs: (oxacillin MIC of ref. labs:
>256 mg/ L; intended 8-64 mg/ L; intended
interpretation: resistant) interpretation: resistant) Total
number of number of
labs doing % labs doing % total %
Country test correct test correct of tests correct
Austria 10 100 10 70 20 85
Belgium 55 100 56 73 111 86
Bulgaria 19 100 19 95 38 97
Czech R. 33 97 33 42 66 70
Denmark 4 100 4 75 8 88
Finland 26 100 26 81 52 87
Germany 28 100 28 86 56 93
Greece 16 100 16 63 32 81
Iceland 3 100 3 33 6 67
Ireland 10 100 10 80 20 90
Israel 3 100 3 100 6 100
Italy 48 100 47 79 95 89
Luxembourg 5 100 5 80 10 90
Netherlands 22 100 23 87 45 93
Poland 18 100 17 88 35 94
Portugal 15 100 14 71 29 86
Slovenia 10 100 10 90 20 95
Spain 29 100 29 90 58 95
Sweden 24 100 23 91 47 96
UK 10 100 10 70 20 85
Overall concordance 100 77 89
95% Confidence interval 8592
Quantitative methods yielding a penicillin G MIC [number of tests done (n) = 882] were
more frequently concordant than non-MIC methods (n = 242) for S. pneumoniae strains
(94% versus 79%). The same was true for detection of teicoplanin resistance in the S.
haemolyticus strain, with a concordance of 91% for teicoplanin MIC methods (n = 182)
versus 71% for non-MIC methods (n = 160).
Table 4.7. Detection of oxacillin susceptibility and teicoplanin resistance in S. haemolyticus by country
Strain UA1434 (oxacillin MIC of Strain UA1434 (teicoplanin MIC of
ref. labs: 0.5 mg/ L; intended ref. labs: 32-64 mg/ L; intended
interpretation: susceptible) interpretation: resistant)
number of labs number of labs
Country doing test % correct doing test % correct
Austria 10 100 9 100
Belgium 55 84 43 65
Bulgaria 19 53 14 93
Czech R. 33 76 31 90
Denmark 4 75 2 50
Finland 26 81 13 77
Germany 28 100 28 96
Greece 16 81 14 79
Iceland 3 100 0
Ireland 10 90 17 71
Israel 2 50 2 100
Italy 46 83 49 90
Luxembourg 5 100 5 60
Netherlands 23 91 18 61
Poland 17 82 9 100
Portugal 15 93 12 83
Slovenia 10 50 8 100
Spain 30 93 28 89
Sweden 25 80 17 76
UK 10 90 20 65
Overall concordance 83 82
For the two S. aureus strains, oxacillin MIC methods (n = 363) reached 99% concordance
in the homogeneously resistant MRSA, and 76% concordance in the heterogeneously
resistant MRSA. Other methods (n = 405) yielded a concordance of 100% and 78%,
respectively.
Use of guidelines
Of the 395 laboratories specifying which guideline they used, 242 (61%) in 19 countries
followed the NCCLS guideline (Table 4.9). Any other guideline was not followed by more
than 6% of the laboratories, in at most two countries. With only one reference laboratory
participating in France, and Norway not participating, the CA-SFM and Norwegian
Working Group on Antibiotics (NWGA) guidelines were not represented.
5
Thirty-eight of the 433 laboratories (9%) did not specify the guideline they followed.
Discussion
This Europewide QA exercise was characterized by an excellent response rate. It
confirmed that an exercise of these dimensions is feasible and demonstrated the
commitment of EARSS participants to quality. Strains were identified correctly at the
genus and species level, and the average concordances over all control strains were high.
We distributed strains that tested the laboratories capability to identify the most clinically
relevant resistances (penicillin G in S. pneumoniae, methicillin in S. aureus and
glycopeptide in staphylococci) and feel reassured to continue using surveillance data
generated by the participating national surveillance systems.
In this exercise, 8685 tests were reported and assessed but some 850 more results were
expected. Laboratories were asked to test all antimicrobial agents listed on the report
form, but in case they normally test for another agent from the same class they were asked
to specify the name of this agent in the same box. This may have given rise to
misunderstanding. Laboratories were asked furthermore to test the susceptibility using
routine procedures. Apparently this was interpreted by a number of laboratories to test
only those organismantimicrobial combinations that they test routinely. Laboratories
should be solicited in future QA exercises to test and report all the requested
organismantimicrobial combinations.
To screen for penicillin resistance in S. pneumoniae, almost all participants used the
oxacillin 1 g disc, achieving a very high concordance rate. This indicates that the oxacillin
screen disc reliably discriminates susceptible from non-susceptible strains.
The concordance for penicillin resistance in S. pneumoniae is somewhat lower when
laboratories test for penicillin G. This lower concordance rate is partly due to the fact that
a substantial number of laboratories use non-MIC-based penicillin confirmation
techniques. Indeed, we observed a far higher concordance (94%) for quantitative
methods yielding a penicillin MIC than for non-MIC-yielding methods (79%), confirming
the rationale of the EARSS protocol to perform MIC determination on S. pneumoniae
strains found to be non-susceptible by a screen test. The difficulty of laboratories using
disc diffusion tests to recognize reduced penicillin susceptibility in S. pneumoniae has
recently been described in another international QA survey.
6
The differences in concordance among the three S. pneumoniae control strains are a
reflection of how many dilution steps the penicillin MIC for the strain is distant from the
susceptible breakpoint. Indeed, more laboratories misinterpreted strain UA347 as being
penicillin susceptible than strain UA1449.
Some guidelines yielded lower concordance rates for the determination of penicillin
resistance, like the DIN guideline as well as the guidelines specified under Other in Table
4.5. For the DIN guideline, this may be related to the susceptible breakpoint, which is one
dilution step higher. Almost all national guidelines in Europe, as well as the NCCLS
guideline, consider isolates of S. pneumoniae to be non-susceptible to penicillin if the MIC
is >0.06 mg/L.
711
The DIN guideline considers isolates to be non-susceptible to
penicillin if the MIC is >0.12 mg/L.
12
However, it should be noted that the DIN, as well as
the guidelines specified under Other, were used only by relatively small numbers of
laboratories, allowing for larger variation. All three S. pneumoniae control strains were
non-susceptible, and the high concordance rates represent a high sensitivity of EARSS
laboratories to detect penicillin non-susceptibility in S. pneumoniae. It is not possible from
this exercise to infer the specificity of EARSS laboratories to detect penicillin susceptibility
in S. pneumoniae. Because all three S. pneumoniae strains were highly resistant to
erythromycin, they were not really a challenge to participating laboratories. Virtually all
laboratories correctly determined erythromycin resistance.
For the detection of oxacillin resistance in S. aureus, we included one strain that was
homogeneously resistant and another strain that was heterogeneously resistant to
oxacillin. Laboratories performed extremely well in detecting oxacillin resistance in the
homogeneously MRSA strain, but the concordance rate dropped from 100% to 77% in the
heterogeneously resistant MRSA strain. A notable proportion of laboratories in three
countries failed to detect the heterogeneously resistant MRSA strain. However, although
we observed differences in concordance among countries, we found no significant
differences among guidelines. Detecting heterogeneously resistant MRSA possibly
depends more on test methods used by individual laboratories than on differences in
guidelines. Laboratories in most countries, and in some countries in particular, should
scrutinize carefully their capability of detecting low-frequency resistant subpopulations,
and ensure that proper laboratory methods are used to detect heterogeneously resistant
MRSA strains.
Detection of glycopeptide resistance in staphylococci is of paramount importance. For
safety reasons we chose not to distribute a vancomycin intermediate (or resistant) strain
among laboratories all over Europe, but instead distributed a S. haemolyticus strain that
was resistant to teicoplanin. Vancomycin susceptibility of the two S. aureus control strains
was interpreted correctly by participating laboratories, but teicoplanin resistance of the S.
haemolyticus strain was often missed. Quantitative methods yielding an MIC were more
frequently concordant than non-MIC methods for the detection of teicoplanin resistance
against the S. haemolyticus strain (91% versus 71%).
Only a few participants misinterpreted gentamicin and erythromycin susceptibility in
staphylococci, indicating that most participating laboratories are capable of determining
gentamicin and erythromycin resistance.
This exercise provides a good overview of the guidelines being followed in Europe, with
exception of the French and Norwegian guidelines. The NCCLS guideline is widely
followed in Europe. In 10 countries NCCLS seems to be the only guideline in use; but in
the countries that have issued national guidelines (Germany, The Netherlands, Sweden
and Spain) some laboratories also follow the NCCLS guideline. The BSAC and Swedish
Reference Group for Antibiotics (SRGA) guidelines are the only European guidelines used
in more than one country. Because France and Norway are not represented in this study,
we cannot infer on the use of guidelines there. It is probable, however, that the CA-SFM
and NWGA guidelines are not widely followed in other European countries.
We found that 9% of participating laboratories did not specify which guideline they follow.
Apart from the obvious reason that some laboratories simply may not have reported
which guideline is followed, it may also be that some laboratories use in-house guidelines,
Table 4.8. Manufacturer and type of automated system used by participating laboratories
Automatic method Total Automatic method Total
BBL/BD Sceptor 14 Dade Behring Microscan 5
Becton Dickinson Pasco 5 Dade Behring Microscan Walkaway 11
bioMrieux ATP 2 Dade Behring Autoscan 2
bioMrieux Vitek1 6 Sensititre Aris 2
bioMrieux Vitek2 8 Soria Helgguipo Wider 8
bioMrieux Vitek32 6 Pasteur-Sanofi Pneumo PAC 1
bioMrieux Mini API 6 >1 method 6
bioMrieux API-ATB 5
bioMrieux Vitek unspecified 23 Total 110
Table 4.9. The usage of guidelines by number of laboratories per country
Guideline CZECH MEN- CA- Not
used BSAC CRG 98 DIN SURA NCCLS SFM
a
SRGA Stokes >1 specified
Austria 9 1
Belgium 31 10 16
Bulgaria 20
Czech R . 15 4 11 3
Denmark 1 1 3
Finland 25
France 1
Germany 15 6 9 1
Greece 14 3
Hungary 1
Iceland 3
Ireland 5 1 9 2 1
Israel 3
Italy 50 2 1
Luxembourg 5
Malta 1
Netherlands 6 8 9 2
Poland 18 1
Portugal 11 1 4
Slovenia 7 3
Spain 1 25 5
Sweden 25
UK 12 6 3 3
Total 17 6 15 15 1 242 1 26 15 57 38
Grand total 433
a
French laboratories did not participate in this QA exercise, with the exception of one national reference centre.
BSAC, British Society for Antimicrobial Chemotherapy; CRG, Commissie Richtlijnen Gevoeligheidsbepalingen;
DIN, Deutsches Institut fr Normung; MENSURA, Mesa Espaola de Normalizacion de la Suseptibilitad y
Resistencia a los Antimicrobianos; NCCLS, National Committee for Clinical Laboratory Standards; CA- SFM,
Comit de lAntibiogramme de la Socit Franaise de Microbiologie; SRGA, Swedish Reference Group for
Antibiotics; >1, more than one guideline followed.
or other non-documented guidelines. An overview of the antimicrobial susceptibility test
breakpoints of national societies has been published recently.
13
The authors recommend
that other national guidelines (e.g. Czech 98) are also documented in international
literature and that every laboratory works according to well-documented guidelines so
that susceptibility test results are reproducible and comparable. Guidelines should also
be freely accessible through the Internet.
It should be noted that overall the breakpoints defining susceptibility or resistance of
bacteria to antimicrobial agents do not differ greatly between guidelines used in Europe.
Baquero
14
argued that it is possible to establish a theoretical consensus standard list of
breakpoints, such that more than 95% of the breakpoints proposed by the different
systems differ from the consensus standard by no more than one dilution.
We hope that these findings may add to the process of standardizing breakpoints in
Europe as brought forward by the EUCAST.
It is shown that, overall, countries participating in EARSS are capable of delivering
susceptibility data of good quality. The comparability of susceptibility data for penicillin
resistance in S. pneumoniae and for homogeneous methicillin resistance in S. aureus is
satisfactory among European countries and across guidelines. However, we emphasize
the importance of determining an MIC for suspected penicillin non-susceptible S.
pneumoniae and for suspected glycopeptide non-susceptible S. aureus.
Laboratories, particularly in some countries, may need to improve their capability of
detecting oxacillin resistance in heterogeneously resistant MRSA and teicoplanin
resistance in S. haemolyticus.
A number of laboratories did not fill out the form completely, for example by not reporting
the species identification or not performing all the susceptibility tests requested. Not
doing (or reporting) a test is considered as non-performance and hinders the assessment
of the performance of laboratories. This should be avoided in future QA studies by
organizers and participants.
Not every laboratory produced good results in this exercise, and the performance of some
individual laboratories could probably be improved. For continuous external quality
assessment we recommend that laboratories participate in national and international
schemes with frequent distributions of control strains.
However, we feel reassured by this exercise that overall the antimicrobial susceptibility
testing data as monitored through the national surveillance systems that participate in
EARSS are of good quality. It is hoped that laboratories participating in this EARSSUK
NEQAS quality assurance are encouraged to maintain and improve their performance, as
has been observed in other surveillance schemes.
15,16
Acknowledgements
We express our thanks and appreciation for the organization by UK NEQAS, for the
countries coordinating centres who distributed the strains swiftly and for the
overwhelmingly good response rate of the 471 laboratories participating in EARSS. We
thank N. Nagelkerke for help in the statistical analysis and thank the national
representatives of EARSS in the participating countries. We also welcome the comments
on this article by D. Brown and G. Kahlmeter from EUCAST.
EARSS is funded by the European Commission, DG SANCO [Agreement SI2.123794
(99CVF4-018) European Antimicrobial Resistance Surveillance System (EARSS)].
Participating countries and national representatives in EARSS during 2000: Austria, H.
Mittermayer, W. Koller; Belgium, H. Goossens, F. van Loock; Bulgaria, B. Markova; Czech
Republic, P. Urbaskova; Denmark, T. L. Srensen, D. Monnet; Finland, P. Huovinen, O.
Lyytikinen; France, P. Courvalin, H. Aubry-Damon; Germany, W. Witte, T. Breuer; Greece,
N. Legakis, G. Vatopoulos; Hungary, M. Konkoly-Thege; Iceland, K. Kristinsson, H. Briem;
Ireland, O. Murphy, D. OFlanagan; Israel, R. Raz; Italy, G. Cornaglia, M. L. Moro;
Luxembourg, R. Hemmer; Malta, M. Borg; Netherlands, A. de Neeling, W. Goettsch;
Norway, E. Hoiby, P. Aavitsland; Poland, V. Hryniewicz; Portugal, M. Cania, M. Paixao;
Slovenia, M. Gubina; Spain, F. Baquero, J. Campos; Sweden, B. Olsson-Liljequist, O. Cars;
United Kingdom, A. Johnson, M. Wale.
References
1. EUCAST. (2000). The setting of antimicrobial breakpoints Clinical Microbiology and Infection 5, 12.
2. EARSS website. (2001). EARSS Manual. EARSS management Team. [Online.] http://www.earss.rivm.nl (21
June 2002, date last accessed).
3. Tomasz, A., Nachman, S. & Leaf, H. (1991). Stable classes of phenotypic expression in methicillin-resistant
clinical isolates of staphylococci. Antimicrobial Agents and Chemotherapy 35, 1249.
4. UK-NEQAS website. (2000). Report of External Quality Assessment Exercise EARSS-NEQAS 2000. UK-
NEQAS. [Online.] http://www.pcug.co.uk/~ukneqasm/ and at http://www.earss.rivm.nl (21 June 2002, date
last accessed).
5. Bergan, T., Bruun, J., Digranes, A., Lingaas, E., Melby, K. & Sander, J. (1997). Susceptibility testing of bacteria
and fungi. Report from the Norwegian working group on antibiotics. Scandinavian Journal of Infectious
Diseases, Suppl. 1, 103.
6. Tenover, F. C., Mohammed, M. J., Stelling, J., OBrien, T. & Williams, R. (2001). Ability of laboratories to
detect emerging anti-microbial resistance: proficiency testing and quality control results from the World
Health Organizations external quality assurance system for antimicrobial susceptibility testing. Journal of
Clinical Microbiology 39, 24150.
7. The British Society for Antimicrobial Chemotherapy (BSAC) website. (1998). Standardized disc sensitivity
testing method, The Newsletter of the British Society for Antimicrobial Chemotherapy, 1998. BSAC,
Birmingham, UK. Update available online at: http://www.bsac.org.uk/discdiff/sensitivity2.pdf (21 June
2002, date last accessed).
8. Commissie Richtlijnen Gevoeligheidsbepalingen. (1996). Nederlands Tijdschrift voor Medische Microbiologie
4, 5.
9. Cars, O. (Ed.) (1997). Antimicrobial susceptibility testing in Sweden. Scandinavian Journal of Infectious
Diseases, Suppl. 1, 105.
10. Socit Franaise de Microbiologie. (1996). Report of the Comit de lAntibiogramme de la Socit
Franaise de Microbiologie. Clinical Microbiology and Infection 2, Suppl. 1, S149.
11. MENSURA. (2000). Recomendaciones del grupo MENSURA para la seleccin de antimicrobianos en el
estudio de la sensibilidad y criterios para la interpretacin del antibiograma. Revista Espaola Quimoterapia
13, 7386.
12. Deutsches Institut fr Normung. (1998). Methods for the Determination of Susceptibility of Pathogens
(Except Mycobacteria) to Antimicrobial Agents. MIC Breakpoints of Antibacterial Agents, Suppl. 1, pp.
589404. DIN, Berlin.
13. Degener, J. E. & Phillips, I. (2001). Comparison of antimicrobial susceptibility test breakpoints of national
societies. Clinical Microbiology and Infection 7, 514.
14. Baquero, F. (1990). European standards for antibiotic susceptibility testing: towards a theoretical
consensus. European Journal of Clinical Microbiology and Infectious Diseases 7, 4925.
15. Forster, D. H., Krause, G., Gastmeier, P., Ebner, W., Rath, A., Wischnewski, N. et al. (2000). Can quality
circles improve hospital-acquired infection control? Journal of Hospital Infection 45, 30210.
16. Snell, J. S. & Brown, D. F. J. (2001). External quality assessment of antimicrobial susceptibility testing in
Europe. Journal of Antimicrobial Chemotherapy 47, 80110.
Chapter 5
Streptococcus pneumoniae susceptibility
data in Europe
Introduction
Pneumonia has been a major cause of morbidity and mortality among humans
throughout history. Despite advances made by medical science it is still the major cause
of infection related mortality world-wide. In 1998, the World Health Organisation reported
over 3.7 million deaths due to lower respiratory tract infections [1, 2].
Streptococcus pneumoniae is an important pathogen in many community-acquired
respiratory infections, including acute bacterial sinusitis, acute otitis media, community
acquired pneumonia, and acute exacerbations of chronic bronchitis, as well as more
invasive infections such as meningitis and bacteremia [3].
Until the early 1990s, clinical isolates of S. pneumoniae were nearly uniformly susceptible
to penicillin [4], but it was only a matter of time until the first reports of penicillin-resistant
pneumococci emerged and became more widespread [2, 5]. Together with the emergence
of penicillin resistance in pneumococci, strains with multiple resistance, not only to
betalactams, but also to macrolides, chloramphenicol, tetracyclines, and cotrimoxazole
appeared in various parts of the world [2, 6].
The European Antimicrobial Resistance Surveillance System (EARSS) has been collecting
antimicrobial susceptibility data (AST) data for S. pneumoniae since 1999, currently in 26
European countries, and has been monitoring variations in antimicrobial resistance
geographically and in time. In this chapter the susceptibility data for S. pneumoniae from
1999, 2000, and 2001 are presented and discussed.
Methods
EARSS protocol for Streptococcus pneumoniae testing
AST data were collected from European countries participating in EARSS of the first
invasive S. pneumoniae isolate (from blood or cerebrospinal fluid (CSF)) per patient per
quarter over the period 1999-2001. By the EARSS protocol laboratories were asked to
report penicillin susceptibility determined with an oxacillin disk test (1 g or 5 g). Using
a 1-g oxacillin disc load, a S. pneumoniae strain with a zone size of 20 mm or less is
considered presumably non-susceptible to penicillin. With a 5-g oxacillin disk load, the
zone size must be 26 mm or less in order to be presumably non-susceptible. If the isolate
was oxacillin non-susceptible by the oxacillin disk test, the minimum inhibitory
concentration (MIC) of penicillin, cefotaxime or ceftriaxone and ciprofloxacin had to be
determined. In addition, participants could also opt to report AST data of: clindamycin,
erythromycin, rifampin, tetracycline and vancomycin. EARSS collects routinely generated
data and as such accepts the interpretations, sensitive (S), intermediate (I) and resistant
(R) of the laboratories.
Antimicrobial susceptibility breakpoints
All laboratories perform antimicrobial susceptibility tests and interpret their results
according to the guidelines used in their laboratories. Almost all national guidelines in
Europe, as well as the United States guideline, consider isolates of S. pneumoniae to be
non-susceptible to penicillin if the MIC is >0.06 mg/L [7-11]. The German guideline
considers isolates penicillin non-susceptible if the MIC is >0.12 mg/L [12]. An overview of
the breakpoints according to the various AST guidelines used by laboratories participating
in EARSS is shown in Table 5.1.
Table 5.1. Antimicrobial susceptibility testing breakpoints for Streptococcus pneumoniae used in EARSS*.
Antibiotic S R National breakpoint committees Range of breakpoints
(mg/L) (mg/L) S R
Penicillin G 0.06 - NCCLS
0.06 2 MENSURA, SFM 0.06 - 0.12 2
SRGA , BSAC, CRG
0.12 2 DIN
Ceftriaxone / 0.12 2 SRGA
Cefotaxime 0.5 2 NCCLS
0.5 4 SFM 0.12 - 4 2 - 32
1 2 BSAC
4 32 CRG, DIN
Ciprofloxacin 0.12 4 SRGA
0.25 4 NWGA 0.12 - 1 4
1 4 CRG, DIN, MENSURA
Erythromycin 0.25 1 NCCLS
0.5 1 BSAC, SRGA
0.5 2 MENSURA 0.25 - 1 1 - 8
1 4 CRG, NWGA
1 8 DIN, SFM
BSAC, British Society for Antimicrobial Chemotherapy; CRG, Commissie Richtlijnen Gevoeligheidsbepalingen;
DIN, Deutsches Institut fr Normung; MENSURA, Mesa Espaola de Normalizacion de la Suseptibilitad y
Resistencia a los Antimicrobianos; NCCLS, National Committee for Clinical Laboratory Standards; NWGA,
Norwegian Working Group on antibiotics; SFM, Comit de lAntibiogramme de la Socit Franaise de
Microbiologie; SRGA, Swedish Reference Group for Antibiotics.
* For comparability, the resistance breakpoints of the SFM and CRG guidelines were adapted, changing the sign
> (greater than) into the sign (equal to or greater than) and adding one dilution step.
Data processing and validation
Laboratories send data to the national EARSS data manager, who checks and forwards the
data to the Dutch National Institute of Public Health and the Environment (RIVM).
National data managers receive standard feedback reports for approval. The national data
is then collated into the EARSS database and are accessible at the interactive EARSS web
site, and are made available for statistical analysis using the SAS software.
Data lacking mandatory information i.e., laboratory code, date of sample collection,
patient identifier or month and year of birth, pathogen code, antibiotic code, or test result
(S, I or R) were rejected. Data were de-duplicated to only the first invasive isolate per
patient per year. The proportion of PNSP and erythromycin non-susceptible S.
pneumoniae was determined per country per year (1999, 2000 and 2001). Non-
susceptible S. pneumoniae isolates include both intermediately and fully resistant isolates.
Only the countries that reported AST data for all three years (1999-2001) were included
for the trend analysis and seasonal distribution over the years.
The AST data from all countries reporting over the years were included for the age
distribution analysis. The proportion of PNSP isolates non-susceptible to 3
rd
generation
cephalosporins and fluoroquinolones was determined.
The S. pneumoniae isolates from France could not be included in the age distribution and
seasonal variation analysis, because France delivered aggregated data of invasive S.
pneumoniae isolates only for the first 2 quarters of 2001.
A quality assurance exercise was performed in September 2000 by 471 laboratories from
23 countries participating in EARSS to assess the comparability of susceptibility test
results (see chapter 4). The overall concordance rate for detecting penicillin non-
susceptibility with an oxacillin screen disk was 97% and ranged from 96% to 99% for the
three strains tested. Ninety-seven percent of the participants used an oxacillin disk loaded
with 1 g. Erythromycin resistance was detected correctly by 99% of the participants.
Results
Streptococcus pneumoniae penicillin susceptibility
In total 26 European countries reported AST data of 15 288 invasive S. pneumoniae isolates
to EARSS over the period 1999-2001 (3899, 5449, and 5940 respectively), of which 1712
were reported non-susceptible to penicillin (11.4% overall). Confirmation of the penicillin
non-susceptible S. pneumoniae (PNSP) isolates by determining the penicillin MIC, as
specified in the EARSS protocol, was performed for 91% of the PNSP isolates. In total 93%
of the S. pneumoniae isolates were derived from blood samples versus 7% from
cerebrospinal fluid (CSF) (French data not included). The highest average proportions of
PNSP isolates (> 30%) were found for the Mediterranean countries. The lowest average
proportions of PNSP isolates were found in northern countries (<3%) (Figures 5.1 and 5.2).
The exact number of invasive penicillin non-susceptible S. pneumoniae (PNSP) isolates
reported to EARSS in the period 1999-2001, per country, per year can be found in
Appendix 5.1.
Time trends in country-specific PNSP proportions
Figure 5.3 displays S. pneumoniae penicillin non-susceptibility in invasive isolates from
1999 to 2001 and shows a stable pattern in most countries. The total number of S.
pneumoniae isolates reported (by the countries with data of all three years) in 1999, 2000
and 2001 were 3871, 3839, and 3611, respectively. The average proportion of PNSP was 7%,
8% and 6% for the respective years.
LU MT
< 3
% I, R
Missing
3 9
10 30
> 30
Figure 5.1. Mean proportion (1999-2001) of Streptococcus pneumoniae penicillin non-susceptibility (PNSP) in
invasive isolates reported per country. France delivered aggregated data for the first 2 quarters of 2001.
0
n
o
n
-
s
u
s
c
e
p
t
i
b
l
e

i
s
o
l
a
t
e
s

(
%
)
country code (number of isolates)
40
35
30
25
20
15
10
5
45
50
F
R

(
6
9
3
)
B
G

(
2
9
)
E
S

(
1
2
3
3
)
H
U

(
3
6
)
L
U

(
7
2
)
I
L

(
1
6
9
)
B
E

(
2
3
2
8
)
U
K

(
1
3
1
6
)
I
E

(
6
0
5
)
I
T

(
4
2
3
)
D
E

(
6
8
4
)
F
I

(
7
7
0
)
N
O

(
4
1
6
)
D
K

(
7
3
7
)
C
Z

(
2
6
5
)
N
L

(
2
2
1
5
)
S
E

(
2
3
9
6
)
P
T

(
3
7
9
)
S
I

(
1
9
6
)
H
R

(
2
0
)
M
T

(
2
3
)
I
S

(
1
3
3
)
A
T

(
1
1
6
)
E
E

(
2
0
)
% intermediate
% resistant
Figure 5.2. S. pneumoniae penicillin susceptibility in invasive isolates reported per country over 1999-2001, with a
minimum of 10 S. pneumoniae isolates (number of isolates indicated between brackets). France delivered
aggregated data for the first 2 quarters of 2001. For the country codes, see appendix 5.0 at the end of this chapter.
Figure 5.3. Streptococcus pneumoniae penicillin non-susceptibility in invasive isolates from 1999 to 2001. Only
countries that reported to EARSS for all three years of surveillance are presented (number of isolates indicated
between brackets).
0
20
25
15
10
5
30
n
o
n
-
s
u
s
c
e
p
t
i
b
l
e

i
s
o
l
a
t
e
s

(
%
)
year of sample collection
country code (mean
sample-size per year)
1999 2000 2001
BE (776)
DE (228)
FI (257)
IE (202)
IS (44)
IT (141)
LU (24)
NL (738)
PT (126)
SE (799)
UK (439)
Seasonal variation
Figure 5.4 shows the total number of invasive S. pneumoniae isolates and the proportion
of PNSP isolates reported per month over a period of three years from 9 countries
(Germany, Finland, Iceland, Ireland, Luxembourg, the Netherlands, Portugal, Sweden,
and the United Kingdom). The distribution of the S. pneumoniae isolates that are reported
displays a clear seasonal variation with a peak around the turn of the year, a constant
decline of isolates until August, and a successive regular increase in the fall and winter
months.
Figure 5.4. The total number of invasive Streptococcus pneumoniae (SPN) isolates and the proportion of penicillin
non-susceptible S. pneumoniae isolates (PNSP) per month (only the 9 countries with data of all quarters over
the period 1999-2001 were included).
A two-fold increase of the average number of S. pneumoniae isolates reported during the
winter months (January to March) was observed in comparison with the summer months
(June to August) (Table 5.2). No significant differences (P > 0.05) were found for the
average proportion of PNSP isolates during the winter and summer months (Table 5.2).
Every August there appears to be a repeated high prevalence of PNSP.
0
n
u
m
b
e
r

o
f

S
P
N

i
s
o
l
a
t
e
s
%

P
N
S
P
month of sample collection
350
300
250
200
10
400
0
10,5
7,5
6
4,5
1,5
12
1
9
9
9

j
a
n
f
e
b
m
a
r
a
p
r
m
a
y
j
u
n
j
u
l
a
u
g
s
e
p
o
c
t
n
o
v
d
e
c

2
0
0
0

j
a
n
f
e
b
m
a
r
a
p
r
m
a
y
j
u
n
j
u
l
a
u
g
s
e
p
o
c
t
n
o
v
d
e
c
2
0
0
1

j
a
n
f
e
b
m
a
r
a
p
r
m
a
y
j
u
n
j
u
l
a
u
g
s
e
p
o
c
t
n
o
v
d
e
c
350 9
200 3
number of SPN isolates
% PNSP
Table 5.2. The average number of invasive Streptococcus pneumoniae isolates and the proportion of penicillin
non-susceptible S. pneumoniae isolates (PNSP) during the winter (Jan-March) and summer months (Jun-Aug)
over the years.
Winter Summer
Year Average number Average number
S. pneumoniae %PNSP S. pneumoniae %PNSP
1999 301 5% 139 3%
2000 320 5% 143 5%
2001 341 6% 164 7%
Age distribution
The invasive S. pneumoniae isolates (n= 14 595, French isolates not included) reported to
EARSS over the years 1999-2001 show that S. pneumoniae isolates were most common in
the age group 4 years and younger (Figure 5.5). Also the likelihood of having PNSP was
greatest for the age group 4 years and younger (20%, p<0.05). A steadily increasing
number of S. pneumoniae isolates were reported for those between the ages of 49 and 79.
Streptococcus pneumoniae cephalosporin susceptibility
In total, 1627 of the 1712 (95%) PNSP isolates reported to EARSS were tested for
ceftriaxone or cefatoxime in the period 1999-2001. Figure 5.6 shows the proportion of
PNSP isolates non-susceptible to 3
rd
generation cephalosporins per country. Three
percent of the PNSP isolates were reported resistant and 28% intermediately resistant to
these agents (Appendix 5.2).
Streptococcus pneumoniae fluoroquinolone susceptibility
In total 915 of the 1324 PNSP isolates reported to EARSS in 1999-2001 were tested for
ciprofloxacin or ofloxacin. Figure 5.7 shows the proportion of PNSP isolates non-
susceptible to fluoroquinolones per country. Two percent of these isolates were reported
as resistant and 10% as intermediately resistant to these agents (Appendix 5.3).
Streptococcus pneumoniae erythromycin susceptibility
For 76% of all S. pneumoniae isolates also AST data of erythromycin was reported.
Therefore, it was decided at the 2001 EARSS plenary meeting to display as well
erythromycin susceptibility data. In total 11 781 of 15 288 (77%) S. pneumoniae isolates
reported to EARSS were tested for erythromycin in the years 1999, 2000, and 2001 (2419,
0
n
u
m
b
e
r

o
f

S
P
N

i
s
o
l
a
t
e
s
%

P
N
S
P
age categories
1800
1600
1200
1000
800
200
2000
0
15
10
25
0
-
4
5
-
9
1
0
-
1
4
1
5
-
1
9
2
0
-
2
4
2
5
-
2
9
3
0
-
3
4
3
5
-
3
9
4
0
-
4
4
4
5
-
4
9
5
0
-
5
4
5
5
-
5
9
6
0
-
6
4
6
5
-
6
9
7
0
-
7
4
7
5
-
7
9
8
0
-
8
4
8
5
+
20
1400
600
400 5
number of SPN isolates
% PNSP
0
60
50
40
30
20
10
70
% intermediate
% resistant
n
o
n
-
s
u
s
c
e
p
t
i
b
l
e

i
s
o
l
a
t
e
s

(
%
)
U
K

(
6
5
)
P
T

(
8
8
)
D
E

(
1
2
)
C
Z

(
1
5
)
E
S

(
4
1
6
)
B
E

(
3
3
6
)
I
L

(
6
5
)
F
R

(
3
1
5
)
S
E

(
4
1
)
N
L

(
1
5
)
S
I

(
4
0
)
I
E

(
7
4
)
I
T

(
4
7
)
F
I

(
3
6
)
D
K

(
2
3
)
Figure 5.6. 3
rd
generation cephalosporin susceptibility in invasive penicillin non-susceptible Streptococcus
pneumoniae (PNSP) isolates reported per country in 1999-2001, with a minimum of 10 PNSP isolates (number
of isolates indicated between brackets). France delivered aggregated data for the first 2 quarters of 2001.
Figure 5.5. The total number of invasive Streptococcus pneumoniae isolates and the proportion of penicillin non-
susceptible S. pneumoniae isolates (PNSP) by age reported to EARSS from 1999 to 2001.
0
90
80
70
60
40
50
30
20
10
100
% intermediate
% resistant
n
o
n
-
s
u
s
c
e
p
t
i
b
l
e

i
s
o
l
a
t
e
s

(
%
)
country code (nr of isolates)
C
Z

(
1
3
)
I
L

(
1
3
)
D
K

(
2
3
)
P
T

(
8
8
)
S
I

(
3
8
)
I
T

(
4
3
)
E
S

(
2
0
4
)
B
E

(
3
3
7
)
S
E

(
3
4
)
F
I

(
1
3
)
I
E

(
7
3
)
LU MT
< 3
% I, R
Missing
3 9
10 30
> 30
Figure 5.7. Fluoroquinolone susceptibility in invasive penicillin non-susceptible Streptococcus pneumoniae (PNSP)
isolates reported per country in 1999-2001, with a minimum of 10 PNSP isolates. Note that the Swedish guideline
recommends that pneumococci should never be reported as clinically susceptible to ciprofloxacin.
Figure 5.8. Mean proportion (1999-2001) of Streptococcus pneumoniae erythromycin non-susceptibility in invasive
isolates per country reported to EARSS. France delivered aggregated data for the first 2 quarters of 2001.
4431 and 4931 respectively). In the period 1999-2001, 2071 S. pneumoniae isolates were
reported non-susceptible to erythromycin (17.6% overall).
The highest average proportions of erythromycin non-susceptible S. pneumoniae isolates
were found in Italy, France and Belgium (>30%), whereas the lowest proportion of
erythromycin non-susceptible isolates was found in the Czech Republic (<3%), followed
by the Scandinavian countries, the Netherlands, Germany, Austria and Iceland (3-10%)
(Figure 5.8).
More than half of the PNSP isolates with reported erythromycin data in the period 1999-
2001 (1558/1712, 91%) were reported to be erythromycin resistant (54%) and 1% was
reported to be intermediately resistant. Figure 5.9 reports the proportions of erythromycin
resistance among PNSP isolates by country. For some countries the ratio of reported
erythromycin AST data was unequal for penicillin susceptible and non-susceptible
isolates. Therefore the erythromycin AST data in Appendix 5.4 is categorised as penicillin
susceptible and penicillin non-susceptible.
0
70
60
50
40
30
20
10
80
% intermediate
% resistant
n
o
n
-
s
u
s
c
e
p
t
i
b
l
e

i
s
o
l
a
t
e
s

(
%
)
country code (nr of isolates)
F
R

(
3
1
5
)
B
E

(
3
3
7
)
I
T

(
4
6
)
F
I

(
4
7
)
E
S

(
4
1
0
)
S
I

(
3
0
)
N
L

(
1
4
)
U
K

(
6
4
)
I
L

(
6
1
)
P
T

(
4
8
)
C
Z

(
1
4
)
I
E

(
6
4
)
D
K

(
2
5
)
S
E

(
4
5
)
Figure 5.9. Erythromycin susceptibility in invasive penicillin non-susceptible Streptococcus pneumoniae (PNSP)
isolates reported per country in 1999-2001, with a minimum of 10 PNSP isolates (number of isolates indicated
between brackets). France delivered aggregated data for the first 2 quarters of 2001.
Discussion
There was a clear north-south gradient for the proportion of invasive penicillin non-
susceptible S. pneumoniae. The highest proportion of penicillin non-susceptibility was
found in the southern European countries. Most countries (with data for all three years)
did not show any increase in resistance during the relatively short observation time. Some
countries reported low numbers of invasive S. pneumoniae isolates, possibly leading to an
overestimation of the proportion of non-susceptible isolates. However, in the majority of
countries only a relatively small proportion of the isolates was reported as fully resistant
to penicillin.
The prevalence of invasive S. pneumoniae was seasonal with clear peaks during winter.
The prevalence of penicillin non-susceptible S. pneumoniae displayed no seasonality. The
seasonal variation analysis of the total number of invasive S. pneumoniae isolates and the
proportion of PNSP per month includes mainly data from northern European countries
with the only exception being Portugal. Epidemiology of S. pneumoniae infections in
northern countries may possibly be different from southern and/or central European
countries also with regard to PNSP-seasonality. The same analysis should be repeated
once trend-data is available from more countries. To what degree the number of reported
S. pneumoniae strains over the years/months follow a comparable seasonality as do upper
respiratory viral infections, respiratory syncytial virus, and influenza remains to be
elucidated.
The prevalence of invasive infections with S. pneumoniae was highest in children 4 years
and younger. Elderly are also at higher risk for invasive S. pneumoniae infections. It is
known from literature that the highest rates of invasive pneumococcal disease (i.e.
bacteremia and meningitis) occur among young children, especially those less than 2
years old [13]. EARSS data show that also the likelihood of having PNSP was greatest for
the age group 4 years and younger. This emphasises the importance of physician and
parent education about the prudent use of antimicrobial agents, as well as the importance
of new conjugate vaccines from which children could benefit. Since February 2000, a new
vaccine to protect children of less than 2 years against invasive pneumococcal disease has
been widely used in the United States [14]. This vaccine has also been granted final
marketing authorisation by the European Commission in 15 European countries, and is
currently on the market in several of them. The standard treatment for pneumococcal
illnesses in infants and young children has relied on the use of antibiotics, such as
penicillin and erythromycin. According to WHO, increasing pneumococcal resistance to
antimicrobial drugs, and the rapid spread of resistant strains throughout the world
underlines the importance of the prudent use of antimicrobial agents and vaccination [15].
There were large differences in the resistance proportions among penicillin non-
susceptible S. pneumoniae for 3
rd
generation cephalosporins and fluoroquinolones. When
interpreting proportion rates from the different countries it is essential to be aware of
recommendations of the different national breakpoints.
Regarding cephalosporin susceptibility, national breakpoint committees recommend
laboratories in case of finding a PNSP either to:
a) report 3
rd
generation cephalosporins automatically as non-susceptible (unless MIC
determination has shown otherwise), such as is common practise in the UK; or
b) determine the cefotaxim or ceftriaxone MIC, leading most PNSPs in Sweden to be
classified as intermediate resistant because the SRGA susceptible breakpoint for
cephalosporins is low (S <0.12 mg/L) compared to other national susceptible
breakpoints.
These recommendations explain the high proportion of intermediate resistance to 3
rd
generation cephalosporins in Sweden and the UK.
Regarding fluoroquinolone susceptibility it is essential to know that the national guideline
SRGA, used by all Swedish laboratories participating in EARSS, recommends that
pneumococci should never be reported as clinically susceptible to ciprofloxacin. For this
reason all PNSP isolates from Sweden were reported as intermediately resistant.
Taking these recommendations into account, considerable differences remain among
countries in resistance proportions among penicillin non-susceptible S. pneumoniae for
3
rd
generation cephalosporins and fluoroquinolones. Fluoroquinolone resistance is a
problem in many areas warranting the prudent use of these agents.
In many countries the proportion of macrolide resistant S. pneumoniae is high. It is
apparent also from the EARSS data that penicillin and macrolide resistance is often
associated. This high macrolide resistance prevalence has cast doubt on the efficacy of
macrolide antibiotics for serious pneumococcal infections. Based on surveillance data
from EARSS and on outpatient antibiotic sales, a strong correlation between antimicrobial
resistance in S. pneumoniae and the use of beta-lactam antibiotics and macrolides was
demonstrated in Europe, see chapter 6. Thus, in situations where penicillin and
erythromycin resistance is common, the empirical use of macrolides should be
discouraged.
Appendix 5.0. EARSS country codes
Austria AT Italy IT
Belgium BE Luxembourg LU
Bulgaria BG Malta MT
Croatia HR Netherlands NL
Czech Republic CZ Norway NO
Denmark DK Poland PL
Estonia EE Portugal PT
Finland FI Rumania RO
France FR Russia RU
Germany DE Slovakia SK
Greece GR Slovenia SI
Hungary HU Spain ES
Iceland IS Sweden SE
Ireland IE Switzerland CH
Israel IL United Kingdom UK
Appendix 5.1. Invasive penicillin non-susceptible S. pneumoniae (PNSP) isolates reported to EARSS in the period
1999-2001, per country, per year (SP = S. pneumoniae).
Total SP
country year Nr of tested Nr %
Labs for PNSP Nr I Nr R % I % R PNSP PNSP
AT 2000 9 53 1 0 2% 0% 1 2%
2001 9 63 2 0 3% 0% 2 3%
BE 1999 96 938 87 44 9% 5% 131 14%
2000 92 973 100 51 10% 5% 151 16%
2001 83 417 55 0 13% 0% 55 13%
BG 2000 8 13 0 3 0% 23% 3 23%
2001 8 16 0 1 0% 6% 1 6%
CZ 2000 26 111 4 0 4% 0% 4 4%
2001 32 154 10 1 6% 1% 11 7%
DE 1999 11 363 5 3 1% 1% 8 2%
2000 9 168 3 1 2% 1% 4 2%
2001 9 153 4 1 3% 1% 5 3%
DK 2000 5 410 14 1 3% 0% 15 4%
2001 5 327 8 2 2% 1% 10 3%
EE 2001 5 20 0 0 0% 0% 0 0%
ES 2000 33 584 126 64 22% 11% 190 33%
2001 38 649 167 74 26% 11% 241 37%
FI 1999 14 245 8 2 3% 1% 10 4%
2000 9 176 9 0 5% 0% 9 5%
2001 9 349 27 2 8% 1% 29 8%
FR 2001 329 693 211 104 31% 15% 315 46%
HR 2001 10 20 3 0 15% 0% 3 15%
HU 2001 14 36 5 3 14% 8% 8 22%
IE 1999 10 157 26 4 17% 3% 30 19%
2000 18 202 16 10 8% 5% 26 13%
2001 21 246 24 6 10% 2% 30 12%
IL 2001 5 169 59 8 35% 5% 67 40%
IS 1999 2 49 1 0 2% 0% 1 2%
2000 2 36 3 0 8% 0% 3 8%
2001 3 48 3 0 6% 0% 3 6%
Total SP
IT 1999 42 183 21 3 11% 2% 24 13%
2000 38 119 12 1 10% 1% 13 11%
2001 39 121 6 5 5% 4% 11 9%
LU 1999 1 9 1 1 11% 11% 2 22%
2000 5 22 3 0 14% 0% 3 14%
2001 8 41 2 3 5% 7% 5 12%
MT 2000 1 11 1 0 9% 0% 1 9%
2001 1 12 1 0 8% 0% 1 8%
NL 1999 21 762 6 2 1% 0% 8 1%
2000 23 739 7 3 1% 0% 10 1%
2001 20 714 5 2 1% 0% 7 1%
NO 1999 1 28 0 0 0% 0% 0 0%
2000 1 388 7 2 2% 1% 9 2%
PL 2001 5 8 0 1 0% 13% 1 13%
PT 1999 12 119 20 0 17% 0% 20 17%
2000 11 98 28 0 29% 0% 28 29%
2001 16 162 40 0 25% 0% 40 25%
SE 1999 24 805 11 1 1% 0% 12 1%
2000 19 803 16 0 2% 0% 16 2%
2001 20 788 18 4 2% 1% 22 3%
SI 2000 7 40 9 0 23% 0% 9 23%
2001 10 156 31 0 20% 0% 31 20%
SK 2001 4 6 0 0 0% 0% 0 0%
UK 1999 22 241 8 9 3% 4% 17 7%
2000 28 503 11 20 2% 4% 31 6%
2001 26 572 10 15 2% 3% 25 4%
26 countries Total 15288 1255 457 8% 3% 1712 11%
Appendix 5.2. Invasive penicillin non-susceptible S. pneumoniae (PNSP) isolates non-susceptible to 3
rd
generation cephalosporins (CEP) reported to EARSS in the period 1999-2001 per country per year
Total PNSP Proportion
Nr of tested PNSP tested
country year Labs for CEP for CEP Nr I Nr R % I % R
AT 2000 9 1 100% 0 0 0% 0%
2001 9 2 100% 0 0 0% 0%
BE 1999 96 130 99% 44 0 34% 0%
2000 92 151 100% 44 7 29% 5%
2001 83 55 100% 2 0 4% 0%
BG 2000 8 1 33% 0 0 0% 0%
2001 8 1 100% 0 0 0% 0%
CZ 2000 26 4 100% 0 1 0% 25%
2001 32 11 100% 4 0 36% 0%
DE 1999 11 5 63% 1 1 20% 20%
2000 9 2 50% 0 0 0% 0%
2001 9 5 100% 0 0 0% 0%
DK 2000 5 14 93% 0 0 0% 0%
2001 5 9 90% 1 0 11% 0%
EE 2001 5 0 0%
ES 2000 33 179 94% 58 6 32% 3%
2001 38 237 98% 56 11 24% 5%
FI 1999 14 6 60% 0 0 0% 0%
2000 9 7 78% 0 0 0% 0%
2001 9 23 79% 2 0 9% 0%
FR 2001 329 315 100% 151 2 48% 0.6%
HR 2001 10 3 100% 0 0 0% 0%
HU 2001 14 5 63% 0 0 0% 0%
IE 1999 10 27 90% 2 0 7% 0%
2000 18 22 85% 5 0 23% 0%
2001 21 25 83% 1 0 4% 0%
IL 2001 5 65 97% 7 1 11% 2%
IS 1999 2 1 100% 0 0 0% 0%
2000 2 2 67% 0 0 0% 0%
2001 3 3 100% 0 0 0% 0%
Total SP
IT 1999 42 24 100% 2 0 8% 0%
2000 38 13 100% 0 0 0% 0%
2001 39 10 91% 3 0 30% 0%
LU 1999 1 2 100% 0 0 0% 0%
2000 5 3 100% 0 0 0% 0%
2001 8 4 80% 1 1 25% 25%
MT 2000 1 1 100% 0 0 0% 0%
2001 1 0 0%
NL 1999 21 4 50% 0 0 0% 0%
2000 23 7 70% 2 0 29% 0%
2001 20 4 57% 2 0 50% 0%
NO 1999 1 0 0%
2000 1 9 100% 4 0 44% 0%
PL 2001 5 1 100% 0 0 0% 0%
PT 1999 12 20 100% 0 0 0% 0%
2000 11 28 100% 5 3 18% 11%
2001 16 40 100% 0 10 0% 25%
SE 1999 24 10 83% 7 0 70% 0%
2000 19 14 88% 8 0 57% 0%
2001 20 17 77% 10 0 59% 0%
SI 2000 7 9 100% 1 0 11% 0%
2001 10 31 100% 6 0 19% 0%
SK 2001 4 0 0%
UK 1999 22 15 88% 6 1 40% 7%
2000 28 29 94% 11 8 38% 28%
2001 26 21 84% 12 3 57% 14%
26 countries Total 1627 95% 458 55 28% 3%
Appendix 5.3. Invasive penicillin non-susceptible S. pneumoniae (PNSP) isolates non-susceptible to
fluoroquinolones (FLUO) reported to EARSS in the period 1999-2001 per country per year.
country year Labs for FLUO for FLUO Nr I Nr R % I % R
AT 2000 9 0 0%
2001 9 2 100% 1 0 50% 0%
BE 1999 96 131 100% 5 4 4% 3%
2000 92 151 100% 1 1 1% 1%
2001 83 55 100% 1 0 2% 0%
BG 2000 8 1 33% 0 0 0% 0%
2001 8 1 100% 0 0 0% 0%
CZ 2000 26 2 50% 0 0 0% 0%
2001 32 11 100% 0 3 0% 27%
DE 1999 11 1 13% 0 0 0% 0%
2000 9 1 25% 0 0 0% 0%
2001 9 3 60% 2 0 67% 0%
DK 2000 5 14 93% 1 1 7% 7%
2001 5 9 90% 0 0 0% 0%
EE 2001 5 0 0%
ES 2000 33 136 72% 20 4 15% 3%
2001 38 68 28% 1 0 1% 0%
FI 1999 14 1 10% 1 0 100% 0%
2000 9 3 33% 0 0 0% 0%
2001 9 9 31% 0 0 0% 0%
HR 2001 10 3 100% 0 0 0% 0%
HU 2001 14 1 13% 1 0 100% 0%
IE 1999 10 27 90% 0 0 0% 0%
2000 18 21 81% 1 0 5% 0%
2001 21 25 83% 0 0 0% 0%
IL 2001 5 13 19% 0 1 0% 8%
IS 1999 2 0 0%
2000 2 0 0%
2001 3 0 0%
country year Labs for FLUO for FLUO Nr I Nr R % I % R
IT 1999 42 24 100% 3 0 13% 0%
2000 38 13 100% 2 1 15% 8%
2001 39 6 55% 1 0 17% 0%
LU 1999 1 0 0%
2000 5 3 100% 0 0 0% 0%
2001 8 4 80% 0 0 0% 0%
MT 2000 1 1 100% 0 0 0% 0%
2001 1 1 100% 0 0 0% 0%
NL 1999 21 2 25% 0 1 0% 50%
2000 23 2 20% 1 0 50% 0%
2001 20 1 14% 0 0 0% 0%
NO 1999 1 0 0%
2000 1 9 100% 9 0 100% 0%
PL 2001 5 0 0%
PT 1999 12 20 100% 4 3 20% 15%
2000 11 28 100% 3 0 11% 0%
2001 16 40 100% 2 0 5% 0%
SE 1999 24 5 42% 5 0 100% 0%
2000 19 16 100% 16 0 100% 0%
2001 20 13 59% 13 0 100% 0%
SI 2000 7 7 78% 1 0 14% 0%
2001 10 31 100% 1 1 3% 3%
SK 2001 4 0 0%
25 countries Total 915 69% 96 20 10% 2%
The proportion of PNSP isolates (n=315) non-susceptible for France was 7.3%.
Appendix 5.4. Invasive S. pneumoniae (SP) erythromycin non-susceptible isolates reported to EARSS in the
period 1999-2001 per country per year, categorised by susceptibility to penicillin.
country year PNSP Total SP Total SP Proportion tested
isolates tested for erythro for erythro Nr I (%) Nr R (%)
AT 2000 I 1 1 100% 1(100%)
S 52 49 94% 1 (2%)
2001 I 2 1 50%
S 61 30 49% 3(10%)
BE 1999 I 87 87 100% 58(67%)
R 44 44 100% 30(68%)
S 807 807 100% 203 (25%)
2000 I 100 100 100% 59(59%)
R 51 51 100% 36( 71%)
S 822 822 100% 236(29%)
2001 I 55 55 100% 44(80%)
S 362 362 100% 95(26%)
BG 2000 R 3 3 100% 1(33%) 1 (33%)
S 10 9 90% 1 (11%)
2001 R 1 0 0%
S 15 11 73% 1 (9%)
CZ 2000 I 4 4 100% 1 (25%)
S 107 89 83%
2001 I 10 9 90% 2(22%)
R 1 1 100%
S 143 135 94% 1 (1%)
DE 1999 I 5 1 20%
R 3 0 0%
S 355 235 66% 1 (0%) 16 (7%)
2000 I 3 2 67%
R 1 1 100%
S 164 132 80% 13(10%)
2001 I 4 1 25%
R 1 0 0%
S 148 74 50% 1 (1%) 9 (12%)
DK 2000 I 14 14 100% 2 (14%)
R 1 1 100% 1(100%)
S 395 395 100% 17 (4%)
2001 I 8 8 100%
R 2 2 100% 1( 50%)
S 317 317 100% 16 (5%)
EE 2001 S 20 20 100% 1 (5%)
ES 2000 I 126 121 96% 63 (52%)
R 64 55 86% 1 (2%) 26(47%)
S 394 368 93% 5 (1%) 25 (7%)
2001 I 167 162 97% 4 (2%) 96(59%)
R 74 72 97% 2 (3%) 36(50%)
S 408 389 95% 5 (1%) 50 (13%)
FI 1999 I 8 8 100% 2 (25%)
R 2 2 100% 1(50%)
S 235 207 88% 4 (2%) 7 (3%)
2000 I 9 9 100% 5 (56%)
S 167 167 100% 3 (2%) 6 (4%)
2001 I 27 26 96% 14(54%)
R 2 2 100% 1(50%)
S 320 320 100% 1 (0%) 28 (9%)
FR 2001 not 693 693 100% 312 (45%)
applicable non-susceptible
HR 2001 I 3 3 100%
S 17 17 100% 3 (18%)
HU 2001 I 5 4 80% 3 (75%)
R 3 2 67%
S 28 25 89% 3 (12%)
IE 1999 I 26 20 77% 3 (15%)
R 4 3 75% 2(67%)
S 127 98 77% 12 (12%)
2000 I 16 13 81% 3 (23%)
R 10 7 70% 3(43%)
S 176 129 73% 12 (9%)
2001 I 24 18 75% 2 (11%)
R 6 3 50% 1 (33%)
S 216 164 76% 20 (12%)
IL 2001 I 59 53 90% 12 (23%)
R 8 8 100% 2 (25%)
S 102 90 88% 3 (3%)
IS 1999 I 1 1 100% 1(100%)
S 48 31 65%
2000 I 3 3 100% 3(100%)
S 33 33 100% 1 (3%)
2001 I 3 3 100% 1 (33%)
S 45 45 100% 3 (7%)
IT 1999 I 21 21 100% 14(67%)
R 3 3 100% 1 (33%)
S 159 159 100% 1 (1%) 37 (23%)
2000 I 12 12 100% 5(42%)
R 1 1 100% 1(100%)
S 106 106 100% 28(26%)
2001 I 6 5 83% 4(80%)
R 5 4 80% 3 (75%)
S 110 91 83% 2 (2%) 30 (33%)
LU 1999 I 1 1 100% 1(100%)
R 1 1 100% 1(100%)
S 7 7 100% 1 (14%)
2000 I 3 3 100% 3(100%)
S 19 16 84% 2 (13%)
2001 I 2 2 100% 2(100%)
R 3 2 67% 1(50%) 1(50%)
S 36 35 97% 5 (14%)
MT 2000 I 1 1 100%
S 10 10 100% 4(40%)
2001 I 1 1 100%
S 11 10 91% 2(20%)
NL 1999 I 6 0 0%
R 2 0 0%
S 754 0 0%
2000 I 7 6 86% 2 (33%)
R 3 3 100%
S 729 675 93% 2 (0%) 25 (4%)
2001 I 5 3 60% 2(67%)
R 2 2 100% 1(50%)
S 707 570 81% 1 (0%) 26 (5%)
NO 1999 S 28 0 0%
2000 I 7 0 0%
R 2 0 0%
S 379 0 0%
PL 2001 R 1 1 100% 1(100%)
S 7 6 86%
PT 1999 I 20 20 100% 2(10%) 5 (25%)
S 99 99 100% 4 (4%)
2000 I 28 28 100% 6 (21%)
S 70 70 100% 5 (7%)
2001 I 40 0 0%
S 122 0 0%
SE 1999 I 11 8 73% 2 (25%)
R 1 1 100%
S 793 526 66% 12(2%) 17 (3%)
2000 I 16 15 94% 2 (13%)
S 787 628 80% 3 (0%) 16 (3%)
2001 I 18 17 94% 1 (6%) 3 (18%)
R 4 4 100%
S 766 632 83% 26 (4%)
SI 2000 I 9 5 56% 1(20%)
S 31 20 65% 1 (5%) 1 (5%)
2001 I 31 25 81% 1 (4%) 12(48%)
S 125 82 66% 6 (7%)
SK 2001 S 6 5 83% 1(20%)
UK 1999 I 8 8 100% 2 (25%)
R 9 7 78% 2(29%)
S 224 14 6% 4(29%)
2000 I 11 11 100% 3(27%)
R 20 16 80% 5 (31%)
S 472 227 48% 37 (16%)
2001 I 10 9 90% 3 (33%)
R 15 13 87% 4 (31%)
S 547 287 52% 33 (11%)
Total 15288 11781 77% 2071 (18%)
non-susceptible
References
1. Causes of annual deaths worldwide 1998. Geneva: World Health Organization, 1998.
2. Moellering RC. The continuing challenge of lower respiratory tract infections. Clin Infect Dis 34, Suppl: 1-3,
2002.
3. Appelbaum PC. Resistance among Streptococcus pneumoniae: Implications for Drug Selection. Clin Infect
Dis 34:1613-20, 2002.
4. File TM. Appropriate use of antimicrobials for drug-resistant pneumonia: focus on the significance of beta-
lactam-resistant Streptococcus pneumoniae. Clin Infect Dis 34, Suppl:17-26, 2002.
5. Hansman D, Andrews G. Hospital infection with pneumococci resistant to tetracycline. Med J Aust 11:498-
501, 1967.
6. Whitney CG, Farley MM, Hadler J, et al. Increasing prevalence of multidrug-resistant Streptococcus
pneumoniae in the United States. N Engl J Med 343:1917, 2000.
8. Commissie Richtlijnen Gevoeligheidsbepalingen. (1996). Nederlands Tijdschrift voor Medische
Microbiologie 4, 5.
13, 7386.
10. Socit Franaise de Microbiologie (1996). Report of the Comit de lAntibiogramme de la Socit Franaise
de Microbiologie. Clinical Microbiology and Infection 2, Suppl.1, S149.
Diseases, Suppl. 1,105.
13. Nielsen SV, Henrichson J. Incidence of invasive pneumococcal disease and distribution of capsular types of
pneumococci in Denmark, 1989-4. Epidemiol Infect 1996;117:411-16.
14. New conjugate vaccine- helps prevent invasive pneumococcal disease in infants and young children
(http://www.efpia.org/1_efpia/evm/pneuconfinal.htm)
15. World Health Organisation. Vaccines and Biologicals Pneumococcus. (http://www.who.int/vaccines/
intermediate/pneumococcus.htm)
Chapter 6
A European study on the relationship
of antimicrobial use and antimicrobial
resistance
S L A M Bronzwaer, O Cars, U Buchholz, S Mlstad, W Goettsch,
I K Veldhuijzen, JL Kool, M J W Sprenger, J E Degener, and EARSS
participants.
Emerging Infectious Diseases 2002;8(3):278-82
Abstract
In Europe, antimicrobial resistance has been monitored since 1998 by the European
Antimicrobial Resistance Surveillance System (EARSS). We examined the relationship
between penicillin nonsusceptibility of invasive isolates of Streptococcus pneumoniae and
antibiotic sales. Information was collected on 1998-99 resistance data for invasive isolates
of S. pneumoniae to penicillin, based on surveillance data from EARSS and on outpatient
sales during 1997 for betalactam antibiotics and macrolides. Our results show that in
Europe antimicrobial resistance of S. pneumoniae to penicillin is correlated with use of
betalactam antibiotics and macrolides.
Introduction
Antimicrobial resistance is a growing problem worldwide, requiring international
approaches. The World Health Organization (WHO) and the European Commission have
recognized the importance of studying the emergence and determinants of resistance and
the need for strategies for its control (1-3). In European countries, antimicrobial resistance
has been monitored in selected bacteria from humans since 1998 through the European
Antimicrobial Resistance Surveillance System (EARSS). Funded by the European
Commission, EARSS is an international network of national surveillance systems
intended to collect comparable and reliable resistance data. The purpose of EARSS is to
document variations in antimicrobial resistance over time and place and to provide the
basis for and assess the effectiveness of prevention programs and policy decisions.
One of the indicator organisms in EARSS is Streptococcus pneumoniae. It was included for
three reasons: it is of major clinical importance for pneumonia, bacterial meningitis, and
otitis media; many countries have reported that its resistance to penicillin is increasing;
and S. pneumoniae is representative of organisms that are transmitted in the community.
A major risk factor for the development of resistance is thought to be inappropriate use
of antimicrobial drugs. Most studies that have investigated the relationship of
antimicrobial use and antimicrobial resistance have been undertaken in hospital,
multicenter, or country settings (4-7). For infections with penicillin-non-susceptible S.
pneumoniae (PNSP), studies have demonstrated that at the individual level, previous use
of betalactam antibiotics such as penicillin is an important risk factor (8-10). Studies on
carriage of PNSP in children have shown that sulfamethoxazole-trimethoprim (co-
trimoxazole) and macrolides such as erythromycin have also been associated with
selection of PNSP (11,12). Translated to the population level, sales of betalactam
antibiotics, co-trimoxazole, or macrolides in a given geographic region may be
proportional to microbial resistance to penicillin. If on the European level a relationship
between antimicrobial resistance and antimicrobial use could be found (as in the case of
S. pneumoniae and resistance to penicillin), efforts to control antimicrobial use and
misuse could be stimulated and monitored in Europe.
We used an ecologic study design to examine the correlation between use of relevant
antibiotics in the outpatient setting and the proportion of PNSP among invasive isolates
of S. pneumoniae in 11 European countries.
Methods
Antimicrobial Resistance Data
The estimated average coverage of the populations of countries participating in EARSS is
52% (range 10% to 90%) (13). Laboratories that participate in EARSS screen invasive S.
pneumoniae isolates for oxacillin resistance (14). When an isolate is found to be non-
susceptible, the EARSS protocol requests confirmation as intermediate- or high-level
resistance to penicillin by determination of MICs. Laboratories perform microbiologic
testing and interpret results according to their own standards. National guidelines in
Europe differ; isolates of S. pneumoniae are considered non-susceptible to penicillin if the
MIC is >0.06 (15-19) or >0.12 (20) mg/L. For this report, we use nonsusceptibility and
intermediate resistance as synonyms; PNSP isolates are either intermediate or fully
resistant to penicillin. Only the first invasive isolate per patient per quarter is reported.
To assess the comparability of susceptibility test results, a quality assurance exercise was
performed in September 2000 among 482 laboratories from 23 countries participating in
EARSS. The concordance (agreement of reported results with intended results) for the
detection of penicillin resistance in the three S. pneumoniae control strains was 91% (21).
Laboratories send standardized data to the national EARSS data manager, who checks
data contents and ensures conformity with the EARSS data format. In collaboration with
WHO, an export module from the laboratory-based software WHONET was developed for
EARSS (22). Every quarter, data are forwarded to the central database at the National
Institute of Public Health and the Environment (RIVM), Bilthoven, Netherlands, where
the project is coordinated.
Antimicrobial Use Data
National outpatient sales data for antibiotics from 1997 were purchased from IMS Health
Global Services, London, United Kingdom, for 13 of the 15 member states of the European
Union. Corresponding data were obtained from the Danish Medicines Agency for
Denmark and from the National Corporation of Swedish Pharmacies for Sweden (23). The
IMS data were examined and adjusted according to the Anatomic Therapeutic
Classification (ATC) system used by WHO (24). The amount in kilograms for an
antimicrobial agent was converted to a number of defined daily doses (DDD). The DDD,
which is based on the average daily dose used for the main indication of the drug, is
appropriate for comparisons of drug use over time and in different geographic areas. For
betalactam antibiotics, we combined ATC groups J01C (extended-and narrow-spectrum
penicillins) and J01D (cephalosporins); macrolides were classified under code J01F. No
data were available for the combination of trimethoprim and sulphonamide.
Nonadherence
We considered nonadherence of patients to the physicians prescription in individual
countries as a possible confounder of antimicrobial resistance. Branthwaite et al. reported
nonadherence levels from a population-based survey in seven countries (25). Data from
four of the seven countries (Spain, Belgium, the United Kingdom, and Italy) were also
captured in EARSS.
Statistical Analysis
We calculated the proportion of PNSP among all invasive S. pneumoniae isolates from
each country reported during 1998-99. Because probabilities allow only values between 0
and 1, we modeled the natural logarithm of the odds of PNSP resistance (logodds).
Least-square linear regression analysis was used to assess correlation between
antimicrobial use (of betalactam antibiotics and macrolides, expressed in DDD per 1,000
population per day) and the logodds of resistance. We correlated nonadherence levels
with the logodds of resistance in the same way.
We calculated the Spearman coefficient of determination (r-square) and its corresponding
p value. For the calculation of the regression lines, we weighted the data points by the
inverse of the variance of each data point. We used SAS software (SAS Institute Inc.,
Release 6.03., Cary, NC).
Results
Antimicrobial Resistance
During 1998-99, 337 laboratories from 11 European Union member states (Belgium,
Finland, Germany, Ireland, Italy, Luxembourg, Netherlands, Portugal, Spain, Sweden, and
United Kingdom) and one nonmember state (Iceland) reported 4,872 invasive S.
pneumoniae isolates to EARSS. The proportion of PNSP among isolates of invasive S.
pneumoniae ranged from 1% to 34% (Table 6.1) (Figure 6.1). Southern European countries
reported higher rates than northern European countries.
Antimicrobial Use
Data on outpatient sales of betalactam antibiotics and macrolides were available for 1997
from all 15 European Union member countries. Antimicrobial use varied widely between
countries. Sales to outpatients ranged from 3.8 to 23.6 DDD per 1,000 inhabitants per day
for betalactam antibiotics and from 0.97 to 5.98 DDD for macrolides. The three countries
with the highest reported use were France, Spain, and Portugal for betalactam antibiotics
and France, Spain, and Italy for macrolides; the three countries with the lowest use were
the Netherlands, Germany, and Austria for betalactam antibiotics and Sweden, the
Netherlands, and Finland for macrolides.
Correlation
For 11 countries, information was available for both antimicrobial resistance and
antimicrobial use. Linear regression of the correlation of use of betalactam antibiotics and
the logodds of resistance showed an r-square of 0.80 (p=0.0002) (Figure 6.2). The
equation for the regression is logodds of resistance =(-3.94)+(0.16xDDD).
For the use of macrolides, we calculated an r-square of 0.46. Figure 6.3 shows the graph
for nonadherence to antibiotics and the logodds of resistance. The r-square is 0.8 (p=0.2).
Discussion
We present for the first time Europewide, country-specific, representative data on
antimicrobial resistance collected by EARSS. Using an ecologic study design, we
demonstrate through the correlation with data on antimicrobial use one aspect of the
usefulness of surveillance for antimicrobial resistance. The results from 11 European
countries show a linear relationship between use of betalactam antibiotics and
macrolides and the proportion of PNSP among all invasive S. pneumoniae isolates.
EARSS data show that resistance for PNSP follows a north-south gradient. Southern
European countries have higher proportions of PNSP than countries in northern Europe.
A possible reason for this observation could be the difference in antimicrobial use, which
also tends to be higher in southern European countries. If use of relevant antibiotics
(betalactam antibiotics and macrolides) and the logodds of resistance are modeled
through linear regression, a strong linear and statistically significant relationship is
demonstrated.
Our findings agree with those of Austin et al., who modeled the relationship between
antimicrobial use and endemic resistance, based on population genetic methods and
epidemiologic observations (26). The correlation in Figure 6.2 is consistent with the
model developed by Austin et al. on theoretical grounds.
We correlate antimicrobial sales data for 1997 with antimicrobial resistance data for 1998
and 1999. Others have observed that after a lag time of 1 or more years, changes in
antimicrobial use may be followed by changes in antimicrobial resistance (27,28).
Therefore, we believe that it is reasonable to correlate antimicrobial sales data in 1997
with antimicrobial resistance data from 1998-99.
We address several limitations in our study. First, because it is an ecologic study, we can
make no inferences on the individual level. Second, resistance rates in some countries
(Table 6.1) are calculated from a relatively limited number of isolates.
However, based on communications with EARSS country representatives, our data are
Table 6.1. Number of submitting laboratories, number of isolates of Streptococcus pneumoniae, number (#) and
percent (%R) of penicillin nonsusceptible S. pneumoniae isolates, logodds of resistance (ln(%R/[1-%R]), and
outpatient sales of betalactam antibiotics and macrolides
Penicillin Outpatient sales of
nonsusceptible antibiotics in DDD
*
/
S. pneumoniae 1,000 inhabitants/day
No. of S.
No. of labo- pneumoniae %R ln Betalactam
Country ratories isolates No. (95% CI) (%R/[1-%R]) antibiotics Macrolides
Austria - - - - - 6 3.7
Belgium 96 940 131 14 (12-16) -1.82 14 4.1
Denmark - - - - - 7 2
Finland 11 211 8 4 (2-8) -3.18 8 1.9
France - - - - - 24 6.0
Germany 15 222 4 2 (1-5) -3.89 5 2.5
Iceland 2 54 1 2 (0-11) -3.89 N.a. N.a.
Ireland 12 157 30 19(13-26) -1.45 11 2.5
Italy 46 194 26 13 (9-19) -1.87 15 5.1
Luxembourg 1 11 2 18 (3-52) -1.52 14 4.7
Netherlands 20 760 8 1 (0-2) -4.6 4 1.2
Portugal 12 134 25 19 (13-27) -1.45 16 3.7
Spain 76 1,240 418 34 (31-36) -0.66 21 5.9
Sweden 24 706 21 3 (2-5) -3.48 8 1
United Kingdom 22 243 21 9 (6-13) -2.31 9 3.2
N.a.= not available; DDD* = defined daily doses; CI = confidence interval.
consistent with antimicrobial resistance levels derived from other sources (29). Third, an
explanation for the differences in antimicrobial resistance could be sampling bias:
clinicians in northern European countries may request blood cultures more frequently than
their southern European colleagues, who may sample only in case of empirical treatment
failure. Fourth, we have not addressed other, potentially important contributing factors for
the development of antimicrobial resistance of organisms that are transmitted in the
community, particularly nonadherence and over-the-counter sales of antimicrobial agents.
Both these factors are difficult to measure. However, in 1993 nonadherence to prescribed
antimicrobial agents was assessed in a survey in six European countries (25). Although the
Figure 6.1. Proportions of invasive isolates of Streptococcus pneumoniae resistant to penicillin (PNSP) among 12
European countries, 199899.
< 3
%PNSP (countries)
(4)
(3)
(3)
(5)
(2)
no data
3 9
10 30
> 30
LU
DDD betalactam antibiotics/1000
UK
BE
IT
LU IE
FI
SE
DE
NL
ES
PT
0 5 10 15 20 25
1
0
-1
-2
-5
-3
-4
l
n

(
R
/
(
1
-
R
)
)
Figure 6.2. The logodds of resistance to penicillin among invasive isolates of Streptoccus pneumoniae (PNSP;
ln(R/[1-R])) is regressed against outpatient sales of betalactam antibiotics in 11 European countries;
antimicrobial resistance data are from 1998 to 1999 and antibiotic sales data are from 1997. DDD = defined daily
dose; BE = Belgium; DE = Germany; FI = Finland; IE = Ireland; IT = Italy; LU = Luxembourg; NL = the
Netherlands; PT = Portugal; ES = Spain; SE = Sweden; UK = United Kingdom.
number of data points is limited, Figure 6.2 suggests a direct relationship between non-
adherence rates and logodds of resistance. Thus, if nonadherence is also related to sales
of antimicrobial agents, it could potentially confound the relationship between use and
resistance. Data on the degree of over-the-counter use among European countries are not
widely available; we know of one Spanish and one Greek study reporting an estimate of
over-the-counter use (30,31). The influence of these and other parameters on the level of
resistance should be quantified and understood. Finally, because children are the main
reservoir of carriage of S. pneumoniae, an age-stratified analysis would be desirable, i.e., a
correlation of resistance with antimicrobial use among children. However, this analysis
would require more detailed use data, for example, of liquid formulations of antibiotics.
At least two studies in northern Europe have demonstrated that PNSP rates can be halted
or even reversed when physicians avoid the inappropriate prescription of antimicrobial
agents (32,33). Our study is timely because it shows that even at the European level a
correlation can be observed between antimicrobial resistance (of S. pneumoniae to
penicillin) and antimicrobial use. In several European countries, national action plans for
the appropriate use of antimicrobial agents are being planned or implemented; their
effectiveness should be monitored through prospective and continuous surveillance of
antimicrobial resistance and antimicrobial sales data (34-38).
Non-adherence (% of interviewees)
UK
BE
IT
ES
0 5 10 15 20 25 30 35 40 45
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
l
n

(
R
/
(
1
-
R
)
)
Figure 6.3. The logodds of resistance of invasive isolates of Streptococcus pneumoniae to penicillin (PNSP;
ln(R/(1-R))) is regressed against nonadherence rates to antibiotic therapy in four European countries.
Nonadherence rates are from 1993; PNSP data are from 1998-99. UK = United Kingdom; BE = Belgium; IT =
Italy; ES = Spain.
Acknowledgements
We thank all the dedicated laboratories that contributed data. We specifically thank all the
national data managers and representatives of the countries participating in EARSS for
their hard work in collecting and processing the data, Karl Kristinsson for highly relevant
and constructive comments, John Stelling for help with software development, Nico
Nagelkerke for significant statistical help, Marc-Alain Widdowson for thoughtful
comments, and Jos van de Velde for helping to keep EARSS running.
EARSS is funded by the European Commission, DG SANCO (Agreement SI2.123794
[99CVF4-018] European Antimicrobial Resistance Surveillance System [EARSS]). National
data for nonhospital antibiotic sales were purchased from IMS Health Global Services
with support of the Nepi Foundation, Malm, Sweden.
Stef Bronzwaer is a public health specialist in the Department of Infectious Disease
Epidemiology of the National Institute for Public Health and the Environment, Bilthoven,
the Netherlands. He helped establish the European Antimicrobial Resistance Surveillance
System, for which he now serves as project leader.
References
1. World Health Organization. Report on infectious diseases 2000: overcoming antimicrobial resistance.
Available at URL: http://www.who.int/infectious-disease-report/index.html Accessed September 23, 2000.
2. Centers for Disease Control and Prevention. Preventing emerging infectious diseases. Available at URL:
http://www.cdc.gov/ncidod/emergplan/plan98.pdf Accessed September 20, 2000.
3. European Community. A strategy against the microbial threat. Official Journal of the European Community.
Council Resolution of 8 June 1999 on antibiotic resistance. Official Journal C 195 , 13/07/1999 p. 13.
Available at URL: http://europa.eu.int/eurlex/en/lif/dat/1999/en_399Y0713_01.html Accessed September
29, 2000.
4. Lyytikainen O, Vaara M, Jarviluoma E, Rosenqvist K, Tiittanen L, Valtonen V. Increased resistance among
Staphylococcus epidermidis isolates in a large teaching hospital over a 12-year period. Eur J Clin Microbiol
Infect Dis 1996;15:133-8.
5. Fridkin SK, Steward CD, Edwards JR, Pryor ER, McGowan JE Jr, Archibald LK, et al. Surveillance of
antimicrobial use and antimicrobial resistance in United States hospitals: Project Intensive Care
Antimicrobial Resistance Epidemiology (ICARE) hospitals, phase 2. Clin Infect Dis 1999;29:245-52.
6. Mouton RP, Hermans J, Simoons-Smit AM, Hoogkamp-Korstanje JA, Degener JE, van Klingeren B.
Correlations between consumption of antibiotics and methicillin resistance in coagulase-negative
staphylococci. J Antimicrob Chemother 1990;26:573-83.
7. Goettsch W, van Pelt W, Nagelkerke N, Hendrix MGR, Buiting AGM, Petit PL, et al. Increasing resistance to
fluoroquinolones in Escherichia coli from urinary tract infections in The Netherlands. J Antimicrob
Chemother 2000;46:223-8.
8. Tan TQ, Mason EO Jr, Kaplan SL. Penicillin-resistant systemic pneumococcal infections in children: a
retrospective case-control study. Pediatrics 1993;92:761-7.
9. Nava JM, Bella F, Garau J, Lite J, Morera MA, Marti C, et al. Predictive factors for invasive disease due to
penicillin-resistant Streptococcus pneumoniae: a population-based study. Clin Infect Dis 1994;19:884-90.
10. Deeks SL, Palacio R, Ruvinsky R, Kertesz DA, Hortal M, Rossi A, et al. Risk factors and course of illness
among children with invasive penicillin-resistant Streptococcus pneumoniae. The Streptococcus pneumoniae
Working Group. Pediatrics 1999;103:409-13.
11. Arason VA, Kristinsson KG, Sigurdsson JA, Stefansdottir G, Molstad S, Gudmundsson S. Do antimicrobials
increase the carriage rate of penicillin resistant pneumococci in children? Cross sectional prevalence study.
BMJ 1996;313:387-91.
12. Melander E, Mlstad S, Alsterlund R, Ekdahl K, Jnsson G. Macrolides and broad-spectrum antibiotics are
risk-factors for spread of pneumococci with reduced sensitivity to penicillin. Pediatr Infect Dis J
2000;19:1172-7.
13. European antimicrobial resistance surveillance system. Report on feasibility phase, p. 56. Available at URL:
http://www.earss.rivm.nl Accessed September 26, 2000.
14. Goettsch W, Bronzwaer SLAM, Neeling de AJ, Wale MCJ, Aubry-Damon H, Olsson-Liljequist B, et al.
Standardisation and quality assurance for antimicrobial resistance of Streptococcus pneumoniae and
Staphylococcus aureus within the European Antimicrobial Resistance Surveillance System (EARSS). Clin
Microbiol Infect 2000;6:59-63.
15. Commissie Richtlijnen Gevoeligheidsbepalingen. Nederlands Tijdschrift voor Medische Microbiologie
1996;4:5.
16. The National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial
susceptibility testing; eight informational supplements. Vol 18, no 1. ISBN 1-56238-337-X. Wayne (PA): The
Committee; 1998.
17. Cars O. Antimicrobial susceptibility testing in Sweden. Available at URL:
http://www.ltkronoberg.se/ext/raf/raf.htm Accessed December 21, 2001.
18. Report of the Comit de lAntibiogramme de la Socit Franaise de Microbiologie. Clin Microbiol Infect
1996;2 Suppl 1: S1-S49.
19. British Society for Antimicrobial Chemotherapy. BSAC standardized disc sensitivity testing method.
Birmingham (UK): Newsletter of the British Society for Antimicrobial Chemotherapy; 1998.
20. Deutsches Institut fr Normung. Methods for the determination of susceptibility of pathogens (except
mycobacteria) to antimicrobial agents. MIC breakpoints of antibacterial agents. Berlin: DIN; 1998. Suppl
1:58940-4.
21. Buchholz U, Bronzwaer S, Snell J, Courvalin P, Cornaglia G, de Neeling J, et al. Comparability of microbiological
susceptibility test results from 23 European countries and Israel: the European Antimicrobial Resistance
Surveillance System (EARSS)/NEQAS study. Clin Microbiol Infect 2001;7 Suppl 1:25.
22. World Health Organization, Communicable Disease Surveillance and Response. WHONET 5 software.
Available at URL: http://www.who.int/emc/WHONET/WHONET.html. Accessed September 28, 2000.
23. Cars O, Mlstad S, Melander A. Variation in antibiotic use in the European Union. Lancet 2001;357:1851-3.
24. ATC index with DDDs. Oslo: WHO Collaborating Centre for Drug Statistics Methodology; 1999.
25. Branthwaite A, Pechre J-C. Pan-European survey on patients attitudes to antibiotics and antibiotic use. J
Int Med Res 1996;24:229-38.
26. Austin DJ, Kristinsson KG, Anderson RM. The relationship between the volume of antimicrobial
consumption in human communities and the frequency of resistance. Proc Natl Acad Sci U S A
1999;96:1152-6.
27. Seppala H, Klaukka T, Vuopio-Varkila J, Muotiala A, Helenius H, Lager K, et al. The effect of changes in the
consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland.
Finnish Study Group for Antimicrobial Resistance. N Engl J Med 1997;337:441-6.
28. Stephenson J. Icelandic researchers are showing the way to bring down rates of antibiotic-resistant bacteria
[news]. JAMA 1996;275:175.
29. European antimicrobial resistance surveillance system. Annual report EARSS 2000. Available at URL:
http://www.earss.rivm.nl Accessed 17 June, 2001.
30. Gonzlez J, Orero A. Consumo de antibiticos en Espaa. Revista Espaola de Quimicoterapia 1996;9
Suppl 4:155.
31. Contopoulos-Ionnides DG, Koliofoti ID, Koutroumpa IC, Giannakakis IA, Ioannides JPA. Pathways for
inappropriate dispensing of antibiotics for rhinosinusitis: a randomized trial. Clin Infect Dis 2001;33:76-82.
32. Mlstad S, Cars O. Major change in the use of antibiotics following a national Programme: Swedish
Strategic Programme for the Rational Use of Antimicrobial Agents and Surveillance of Resistance
(STRAMA). Scand J Infect Dis 1999;31:191-5.
33. Kristinsson KG. Modification of prescribers behaviour: the Icelandic approach. Clin Microbiol Infect 1999;5
(Suppl 4):S43-7.
34. Danish Integrated Antimicrobial Resistance Monitoring and Research Programme (DANMAP).
Consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food
animals, food and humans in Denmark, July 1999. Available at URL: http://www.svs.dk/ Accessed October
27, 2000.
35. Institute de Veille Sanitaire, Paris: Propositions pour un plan national dactions pour la matrise de la
rsistance aux antibiotiques, January 1999 (English version: Proposals for a national action plan to control
antibiotic resistance in France July 1999).
36. Ministries of Health and Agriculture, Ireland. A draft strategy document for control of Antimicrobial
Resistance in Ireland - (SARI), 2000. Available at URL: http://www.ndsc.ie/ Accessed October 27, 2000.
37. National Board of Health and Welfare, Sweden. National plan against antibiotic resistance, 2000. Available
at URL: http://www.sos.se/ (Swedish). Accessed October 27, 2000.
38. United Kingdom Department of Health. UK antimicrobial resistance strategy and action plan. Available at
URL: http://www.doh.gov.uk/arbstrat.htm Accessed October 27, 2000.
Chapter 7
Staphylococcus aureus susceptibility
data in Europe
Introduction
Of the pathogens causing nosocomial infections, Staphylococcus aureus most rapidly
spreads within hospitals and is now a difficult to control problem in many European
countries. S. aureus is a Gram-positive micro-organism that is traditionally treated with
penicillinase-stable betalactams. Immuno-compromised patients are at highest risk for S.
aureus infection. Other risk factors for contracting S. aureus infections are impaired
cellular immunity (e.g., patients with diabetes mellitus or impaired function of white
blood cells), and, because S. aureus can contribute to the formation of biofilms on
artificial devices, use of catheters or other invasive devices and presence of artificial
grafts. S. aureus can cause a variety of infections, such as wound infections, septicemia,
osteomyelitis, pneumonia, and urinary tract infection. Invasive infections resulting in
bacteremia, endocarditis, metastatic infections (e.g., in respiratory system), sepsis and
the toxic shock syndrome can lead to death [1]. However, in most humans S. aureus does
not cause infections. In fact, S. aureus colonises the nasal cavities of about 30% of healthy
humans without causing any symptoms of infection.
Shortly after the introduction of methicillin - the first available betalactam stable penicillin -
reports of resistance of S. aureus to methicillin appeared [2]. The emergence and spread of
MRSA is favoured by the absence of hygienic measures to prevent the spread of the micro-
organisms in hospital settings, probably in combination with high use of antibiotics [3-5].
MRSA strains are usually no more virulent than methicillin susceptible S. aureus (MSSA),
but are often resistant to multiple drugs. Moreover, the antimicrobial agent of choice for
treating an MRSA infection is vancomycin [1], which is more toxic and less efficient than
betalactam antibiotics are. Possibly therefore, MRSA was associated with a higher case
fatality rate in some studies [6, 7]. Thus, control of the spread of MRSA is important.
Information on exactly which countries, hospitals and specific wards are most prone to
harbouring MRSA will help policy makers take measures aimed at MRSA control and
infection prevention.
S. aureus was chosen as one of the two pathogens to start EARSS data collection with,
because it meets the WHO criteria for public health relevance of bacterial pathogens,
being a proven pathogen which is clinically relevant on hospital population level, having
a high probability to spread in hospital settings and being able to develop resistance
against currently used and recommended antibiotics (i.e. betalactams).
In this chapter, we discuss S. aureus susceptibility data collected through EARSS over the
period 1999 to 2001 and in greater detail data from 2001. As presented in earlier EARSS
reports, MRSA proportions vary greatly among the European countries [8].
Methods
EARSS protocol for Staphylococcus aureus testing
Antimicrobial susceptibility test (AST) results of the first S. aureus blood isolate per
patient per quarter are reported to EARSS. The EARSS protocol for S. aureus, requests the
laboratories to report oxacillin susceptibility, preferably determined by an oxacillin screen
plate (6 mg/l) or, alternatively, by an oxacillin disk test (1 g or 5 g). For isolates that are
found to be oxacillin non-susceptible, the minimum inhibitory concentration (MIC) of
oxacillin and vancomycin should be determined. The protocol accepts determination of
the mecA gene by PCR for confirmation of MRSA. Apart from these protocol antibiotics,
other antibiotics can optionally be reported to EARSS (i.e., ciprofloxacin, erythromycin,
gentamicin, rifampin, streptomycin and tetracyclin). EARSS collects routinely generated
data and as such accepts the interpretations, sensitive (S), intermediate (I) and resistant
(R) of the laboratories.
Antimicrobial susceptibility breakpoints
All laboratories perform antimicrobial susceptibility tests and interpret their results
according to their own guidelines. Most national guidelines in Europe as well as the
NCCLS guideline consider isolates of S. aureus non-susceptible to oxacillin if the MIC is
>2 mg/L [9-12]. The German guideline consider isolates non-susceptible to oxacillin if the
MIC is >1 mg/L [13]. According to most guidelines, a MIC of 4 mg/L should be classified
as resistant to oxacillin, but the German and Swedish [14] consider a MIC of 2 mg/L as
resistant.
Data processing and validation
Laboratories send their data to the national EARSS data manager, who checks and
forwards the data to the Dutch National Institute of Public Health and the Environment
(RIVM). National data managers receive standard feedback reports for approval. The
national data is then appended to the EARSS database and are accessible at the
interactive EARSS web site, and are made available for statistical analysis using the SAS
software.
Data lacking mandatory information (i.e., laboratory code, date of sample collection,
patient identifier or month and year of birth, pathogen code, antibiotic code, and test
result (S or R) were rejected, as were observations with intermediate methicillin resistance
test results. Data were de-duplicated to only the first invasive isolate per patient per year,
and analysed by country. To test whether the MRSA proportions changed over the years,
we performed the Cochran-Armitage test for trend by taking the MRSA prevalence per
country per year; only countries reporting S. aureus AST results for all three years were
considered. We also analysed data at the hospital level, only taking into account hospitals
that sent in more than 10 isolates in each of the 3 years (51 hospitals in eight countries).
In addition, multidrug resistance was analysed, defined as resistance to three or more
antibiotics of different classes [15].
A quality assurance exercise was performed in September 2000 by 471 laboratories of 23
countries participating in EARSS to assess the comparability of susceptibility test results
(chapter 4). All laboratories detected correctly the homogeneously methicillin-resistant S.
aureus (MRSA) strain (concordance rate 100%), but the concordance rate was much
lower (77%) for the heterogeneously resistant MRSA strain. For tests of vancomycin
susceptibility, the concordance rates were 98% and 100%. For the optional antibiotics
tested (gentamicin and erythromycin), concordance rates of susceptibility results were
between 98% and 100%.
Results
In 2001, results for 14723 isolates originating from 764 hospitals in 26 countries were
reported. Over the period 1999-2001, 26 countries reported AST data for 32942 isolates.
Total adherence to the EARSS protocol was 65.5% during the period 1999 2001. Non-
susceptibility of S. aureus was confirmed either by determining the oxacillin/methicillin
Figure 7.1. Mean proportion of methicillin resistant Staphylococcus aureus (MRSA) in blood isolates over the
period 1999-2001.
< 3
% R
Missing
3 9
10 30
> 30
LU MT
MIC or by the mecA PCR for 71.8% of the MRSA isolates. Vancomycin MIC was obtained
from 66.4% of the MRSA isolates.
The mean proportion of MRSA blood isolates per country over the years 1999-2001 is
shown in Figure 7.1. Nordic countries and the Netherlands have the lowest level of
resistance, whereas the percentages in most southern European countries, the United
Kingdom, Ireland, and Israel, are much higher and even exceed 40% in some countries
(Figure 7.2).
Figure 7.2 gives the MRSA percentages in 2001 by country, ordered by relative frequency,
indicating also the number of isolates that have been reported. More detailed information
on the proportion of MRSA by country and year can be found in Appendix 7.1.
Time trend
Figure 7.3 shows the prevalence of MRSA per year per country for the 3 years of MRSA
surveillance through EARSS. Only countries with data over the complete period are
included in this Figure (the total number of isolates is 25 442). In most countries the
proportion of MRSA seems to be relatively stable, whereas in others, it has increased
Figure 7.2. Staphylococcus aureus methicillin resistance (MRSA) in blood isolates per country in 2001 (number
of isolates between brackets). For country codes, see appendix 7.0.
0
p
r
o
p
o
r
t
i
o
n

o
f

i
M
R
S
A

s
o
l
a
t
e
s

(
%
)
50
40
54.2
45.4
30
20
10
60
M
T

(
8
3
)
U
K

(
1
4
8
8
)
I
E

(
7
9
8
)
I
T

(
8
3
9
)
G
R

(
3
5
6
)
I
L

(
3
8
1
)
F
R

(
1
7
4
1
)
P
T

(
5
2
1
)
H
R

(
1
4
9
)
B
G

(
1
0
3
)
E
S

(
1
0
1
1
)
B
E

(
1
9
4
)
L
U

(
8
5
)
S
I

(
2
7
0
)
D
E

(
7
8
3
)
P
L

(
1
5
1
)
A
T

(
2
7
7
)
C
Z

(
1
0
7
4
)
S
K

(
3
7
)
E
E

(
7
9
)
H
U

(
3
0
1
)
S
E

(
1
6
3
2
)
D
K

(
5
2
0
)
N
L

(
1
2
9
2
)
F
I

(
5
2
2
)
I
S

(
6
3
)
41.7
41.0
39.3
38.6
32.9
31.9
31.5
27.2
23.1
21.6
20.0
7.6
19.6
17.5
15.2
5.9
5.4
5.1
4.7
0.9 0.8
0.5 0.4
0.0
during the past three years (see Figure 7.3). MRSA proportions are increasing by about
1.6% per year (R
2
=0.96; p< 0.0001). We found a linear increase in MRSA proportions of
almost 6% per year in the United Kingdom (from 33% in 1999 to 45% in 2001; R
2
=0.99;
p< 0.0001). Similarly, in Germany, MRSA proportions increased by almost 4% per year
from 10% in 1999 to 17% in 2001 (R
2
=0.99; p< 0.0001).
MRSA in hospitals
We restricted the analyses to hospitals with data on more than 10 isolates each year
(1999-2001), the analyses included 9 454 isolates from 66 hospitals in 7 countries
(Belgium, Denmark, Ireland, Italy, the Netherlands, Sweden and the United Kingdom).
The number of isolates per hospital per year varied and was mostly below 30, which
hinders proper statistical analysis. Generally, there was no large variation in hospitals
within countries with the highest (e.g., Ireland, Italy and the United Kingdom) or the
lowest MRSA proportions (e.g., Denmark, the Netherlands and Sweden). Although MRSA
proportions differed among hospitals within countries, these differences were generally
small (data not shown). It may be that university hospitals are over-represented in this
selection.
Figure 7.3. Staphylococcus aureus methicillin resistance (MRSA) in blood isolates from 1999 to 2001
(number of isolates between brackets). Only countries reporting to EARSS for all three years of surveillance are
presented.
0
50
40
30
20
10
60
p
r
o
p
o
r
t
i
o
n

o
f

i
s
o
l
a
t
e
s

w
i
t
h

M
R
S
A

(
%
)
year of sample collection
IE (647)
UK (1212)
GR (301)
LU (59)
1999 2000 2001
IT (818)
BE (431)
DE (862)
FI (400)
IS (45)
NL (1303)
PT (347)
SE (1477)
DK (580)
country code (mean
sample-size per year)
Demographic data
Demographic data (age, sex and type of patient (i.e., in or outpatient)) were available for
95% of the isolates. Eighty-six percent of the isolates were those of inpatients.
Table 7.1 shows that in 2001 there were more isolates from males than from females (61%
vs. 32%, p< 0.0001), and MRSA isolates were also more frequent among males. MRSA
proportions were the highest among isolates for which the gender of the source patient
was not known. Patients with an MRSA infection were on average older than patients with
MSSA infections (Table 7.1). MRSA proportions were much higher among hospital
patients than among outpatients. With respect to hospital wards, only about 13% of all
isolates were isolated from patients being nursed at ICUs. Almost half of the isolates
came from internal medicine departments, but MRSA isolates more frequently originated
from intensive care units than from other wards. The hospital department where the
isolate originated was unknown for 13% of the isolates, which were significantly more
often identified as methicillin (oxacillin) susceptible isolates (Table 7.1). In 8 of 26
countries, the reported prevalence of MRSA was higher in isolates from surgery than from
intensive care units. In some countries, the place of isolate sampling was reported for
almost all isolates (>98%; Germany, France, Croatia, Portugal and Sweden), whereas
origin was known for only 29% of isolates from Iceland and for 37% and 39% of the
isolates reported from the Netherlands and Finland, respectively.
Non-susceptibility to non-protocol antibiotics
Table 7.2 shows all optionally reported antibiotics that were relatively frequently reported to
EARSS. Of the non-protocol antibiotics that can be reported to EARSS for MRSA isolates,
erythromycin, gentamicin and vancomycin were reported for at least half of the MRSA
isolates. Isolates were most frequently tested for susceptibility to vancomycin. The EARSS
protocol requires testing for vancomycin susceptibility of MRSA strains. In general, MSSA
isolates were less likely to be tested for susceptibility to non-protocol antibiotics than MRSA
isolates (with the exception of rifampicin). MRSA isolates were more often resistant to other
drugs than MSSA isolates. Almost all MRSA isolates were also non-susceptible to
ciprofloxacin (91%), whereas ciprofloxacin resistance occurred in only 6% of the MSSA
isolates. Vancomycin intermediate S. aureus (VISA) was confirmed in two MRSA isolates
from France. Confirmed MICs of vancomycin were respectively 6 and 12 mg/L.
Multidrug resistance
Multidrug resistance (i.e., resistance against at least three antibiotics of different classes)
occurred in 1210 (11.5%) of the 10527 isolates that were tested for at least three classes of
antibiotics. Resistance to betalactams most frequently occurred in combination with
resistance to quinolones and macrolides, sometimes in combination with aminoglycoside
resistance.
Multidrug resistance was more common among men than among women (12.5% and
8.6% respectively, p<0.0001). Patients with multidrug resistance were older (65.1 18.3
years) than patients whose isolates were resistant to at most two different antibiotics (58.5
23.5 years). The proportion of multidrug resistance was highest in the isolates from
ICUs (21.4%), followed by those from surgery (14.4%) and lowest (3.7%) in the isolates
of which the ward of origin was not reported.
Discussion
We found that methicillin resistant Staphylococcus aureus (MRSA) proportions vary
importantly within Europe. Southern European countries, the United Kingdom and
Ireland reported the highest proportions, whereas northern European countries had
proportions below 1%. These figures are consistent with those obtained through other
surveillance systems [16-19]. MRSA proportions vary more than 100-fold among the
European countries. A similar variation in proportions was previously reported [20] and
probably is mainly due to differences in hygiene policy [21] and may also in part be caused
by the large variation in the use of antimicrobial drugs [4, 5]. Increased awareness of
health professionals is essential to enable strict hygienic measures to be taken and to
optimise antimicrobial use in hospitals. Further research is needed to quantify the
Table 7.1. Staphylococcus aureus: demographic data of patients of whom isolates originated in 2001.
Characteristic MSSA MRSA Total Percentage
(n=11 589) (n=3 134) (n=14 723) of MRSA
Sex of patient Male 6 730 1 903 8 633 22.0 *
Female 4 332 1 018 5 350 19.0
Unknown 527 213 740 28.8
Age of patient, mean SD 58.2 23.5 64.8 19.4 *,
Patient admitted to
ICU 1 228 652 1 880 34.7*
Internal medicine 4 416 1 126 5 542 20.3
Surgery 1 420 527 1 947 27.1
Other wards 2 753 634 3 387 18.7
Unknown 1 772 195 1 967 9.9
* statistically significant difference, i.e. p<0.05, as determined by Chi-square test.
age of the patient of which the isolate originated was missing in 374 (MSSA), respectively 235 (MRSA)
instances.
determinants for this high level of variation, and may provide additional clues for
intervention.
MRSA proportions increased during the 1990s [18], and our data show strong indications
for a rapid increase in the prevalence of methicillin/oxacillin resistance in the United
Kingdom (almost 6% increase per year) and in Germany (almost 4% increase per year).
Especially in the United Kingdom, MRSA is a serious resistance problem, as its prevalence
has risen up to 45% by 2001. However, it should be mentioned that the follow-up period
of three years is still relatively short. Moreover, data from some countries (Portugal and
Greece) fluctuate over the years and some countries reported a low number of isolates.
Therefore, although specific trends seem to be present in some countries, no firm
conclusions about time trends overall can be drawn yet.
Generally, we found only small differences in MRSA proportions among hospitals of the
same country, indicating the importance of national policies and practices. However, to
increase our understanding of determinants of the spread among and persistence within
hospitals, it is important to collect more information about the MRSA-problem at hospital
level.
Table 7.2. Resistance of Staphylococcus aureus strains from invasive isolates to non-protocol antibiotics, by
methicillin susceptibility.
MSSA (n=11 589) MRSA (n=3 134)
Antibiotic % of isolates % of non- % of isolates % of non-
tested susceptibility (I+R) tested susceptibility (I+R)
Ciprofloxacin 43.1 6.0 61.9 91.1
Erythromycin 57.3 13.2 67.9 77.3
Gentamicin 57.2 2.0 71.9 37.6
Rifampicin 50.8 0.49 39.3 19.0
Vancomycin 82.6 0 95.0 0.07 *
Tetracycline 21.3 6.1 34.4 45.9
*
two confirmed VISA strains from France
More S. aureus isolates originated from males than from females, and the proportion
being methicillin resistant was also higher among males. These proportions may have
been higher because intensive care units, reporting more invasive infections and the
highest MRSA proportions, count more men than women [5].
There is a marked difference between the resistance profiles of methicillin susceptible
compared to methicillin resistant S. aureus isolates. Almost all MRSA isolates were non-
susceptible to ciprofloxacin, whereas ciprofloxacin resistance occurred in only a small
percentage of the MSSA isolates. The two antimicrobial drugs that are most likely to still
be active against MRSA were rifampicin and vancomycin. Methicillin resistance indeed
seems to be synonymous with, or at least indicative of multidrug resistance. This "co-
evolution" of resistance is of concern as treatment options for multidrug resistant
infections are limited.
So far, two confirmed VISA strains have been reported to EARSS. The potential emergence
of VISA strains needs to be carefully monitored. Consensus on how to define and detect
VISA is imperative at national and at European level and testing and reporting of VISA
through EARSS needs to be improved. In fact, EARSS recently issued a technical guide for
the screening of VISA/VRSA, providing a stepwise procedure for screening, testing and
reporting within EARSS.
Appendix 7.0. EARSS country codes
Austria AT Italy IT
Belgium BE Luxembourg LU
Bulgaria BG Malta MT
Croatia HR Netherlands NL
Czech Republic CZ Norway NO
Denmark DK Poland PL
Estonia EE Portugal PT
Finland FI Rumania RO
France FR Russia RU
Germany DE Slovakia SK
Greece GR Slovenia SI
Hungary HU Spain ES
Iceland IS Sweden SE
Ireland IE Switzerland CH
Israel IL United Kingdom UK
Appendix 7.1. Methicillin resistant Staphylococcus aureus (MRSA) blood isolates reported to EARSS in the period
1999-2001 per country per year.
Total nr of
country year Nr of Labs S. aureus isolates Nr MRSA % MRSA
AT 2000 9 156 8 5.1%
2001 9 277 21 7.6%
BE 1999 47 442 102 23.1%
2000 42 657 137 20.9%
2001 35 194 42 21.6%
BG 2000 16 111 41 36.9%
2001 17 103 28 27.2%
CZ 2000 31 515 22 4.3%
2001 39 1074 63 5.9%
DE 1999 12 1063 102 9.6%
2000 10 741 107 14.4%
2001 9 783 137 17.5%
DK 1999 5 718 2 0.3%
2000 4 501 1 0.2%
2001 4 520 4 0.8%
EE 2001 6 79 4 5.1%
ES 2000 31 836 235 28.1%
2001 37 1011 234 23.1%
FI 1999 13 316 3 0.9%
2000 12 362 5 1.4%
2001 9 522 2 0.4%
FR 2000 1 22 9 40.9%
2001 21 1714 572 33.4%
GR 1999 19 192 60 31.3%
2000 15 354 179 50.6%
2001 25 356 140 39.3%
HR 2001 14 149 47 31.5%
HU 2001 18 301 14 4.7%
IE 1999 11 511 200 39.1%
2000 18 632 248 39.2%
2001 19 798 333 41.7%
IL 2001 5 381 147 38.6%
Total nr of
country year Nr of Labs S. aureus isolates Nr MRSA % MRSA
IS 1999 2 32 0 0.0%
2000 2 40 1 2.5%
2001 3 63 0 0.0%
IT 1999 56 1158 473 40.8%
2000 48 456 200 43.9%
2001 53 839 344 41.0%
LU 1999 1 25 4 16.0%
2000 4 67 12 17.9%
2001 8 85 17 20.0%
MT 2000 1 76 27 35.5%
2001 1 83 45 54.2%
NL 1999 20 1224 4 0.3%
2000 24 1392 5 0.4%
2001 22 1292 7 0.5%
PL 2001 19 151 23 15.2%
PT 1999 13 369 136 36.9%
2000 8 150 38 25.3%
2001 16 521 166 31.9%
SE 1999 24 1320 13 1.0%
2000 19 1478 9 0.6%
2001 21 1632 14 0.9%
SI 2000 10 154 33 21.4%
2001 10 270 53 19.6%
2001 7 37 2 5.4%
UK 1999 23 655 219 33.4%
2000 27 1494 591 39.6%
2001 25 1488 675 45.4%
26 countries Total 32942 6360 19.3%
References
1. Lowly FD. Staphyloccoccus aureus infections. N Engl J Med 339 (8): 520-32, 1998.
2. Barrett FF, McGehee RF, and Finland M. Methicillin-resistant Staphylococcus aureus at Boston city hospital.
N Engl J Med 279: 441-448, 1968.
3. Martin MA. Nosocomial infections in intensive care units: an overview of their epidemiology, outcome, and
prevention. New Horiz 1 (2): 162-71, 1993.
4. Monnet DL. Methicillin-resistant Staphylococcus aureus and its relationship to antimicrobial use: possible
implications for control. Infect Control Hosp Epidemiol 19: 552-559, 1998.
5. Vincent JL. Microbial resistance: lessons from the EPIC study. Int Care Med 26 (Suppl. 1): S3-8, 2000.
6. Soriano A, Martinez JA, Mensa J, Marco F, Almela M, Moreno Martinez A, Sanchez F, Munoz I, de Anta
MTJ, and Soriano E. Pathogenic significance of methicillin resistance for patients with Staphylococcus aureus
bacteremia. Clin Infect Dis 30 (2): 368-373, 2000.
7. Mekontso-Dessap A, Kirsch M, Brun-Buisson C, and Loisance D. Poststernotomy mediastinitis due to
Staphylococcus aureus: Comparison of methicillin resistant and methicillin susceptible cases. Clin Infect Dis
32 (6): 877 883, 2001.
8. EARSS (European Antimicrobial Resistance Surveillance System). Annual report EARSS 2000. Bilthoven,
The Netherlands, 2001. Available at http://www.earss.rivm.nl.
13, 7386.
11. Socit Franaise de Microbiologie (1996). Report of the Comit de lAntibiogramme de la Socit Franaise
de Microbiologie. Clinical Microbiology and Infection 2, Suppl.1, S149.
12. The National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial
susceptibility testing; eight informational supplements. Vol 18, no 1. ISBN 1-56238-337-X. Wayne (PA): The
Committee; 1998.
Diseases, Suppl. 1,105.
15. Thornsberry C, Sahm DF, Kelly LJ, Critchley IA, Jones ME, Evangelista AT, and Karlowsky JA. Regional trends
in antimicrobial resistance among clinical isolates of Streptococcus pneumoniae, Haemophilus influenzae, and
Moraxella catarrhalis in the United States: Results from the TRUST Surveillance Program, 1999-2000. Clin
Infect Dis 34 (Suppl. 1): S4-16, 2002.
16. De Neeling AJ, van Leeuwen WJ, Schouls LM, Schot CS, van Veen-Rutgers A, Beunders AJ, Buiting AGM,
Hol C, Ligtvoet EEJ, Petit PL, Sabbe LJM, van Griethuysen AJA, and van Embden JDA. Resistance of
staphylococci in The Netherlands: surveillance by an electronic network during 1989-1995. J Antimicrob
Chemother 41 (1): 93-101, 1998.
17. Reacher, MH, Shah A, Livermore DM, Wale MCJ, Graham C, Johnson AP, Heine H, Monnickendam MA,
Barker KF, James D, and George RC. Bacteremia and antibiotic resistance of its pathogens reported in
England and Wales between 1990 and 1998: trend analysis. BMJ 320: 213-16, 2000.
18. Fluit AC, Jones ME, Schmitz FJ, Acar J, Gupta R, and Verhoef J. Antimicrobial susceptibility and frequency
of occurrence of clinical blood isolates in Europe from the SENTRY Antimicrobial Surveillance Program,
1997 and 1998 Clin Infect Dis 30 (3): 454-460, 2000.
19. Fluit AC, Wielders CLC, Verhoef J, and Schmitz FJ. Epidemiology and susceptibility of 3,051 Staphylococcus
aureus isolates from 25 university hospitals participating in the European SENTRY study. J Clin Microbiol 39
(10): 3727-3732, 2001.
20. Voss A, Milatovic D, Wallrauch-Schwarz C, Rosdahl VT, and Braveny I. Methicllin-resistant Staphylococcus
aureus in Europe. Eur J Clin Microbiol Infect Dis 13 (1): 50-5, 1994.
21. Witte W, Braulke C, Heuck D, and Cuny C. Analysis of nosocomial outbreaks with multiply and methicillin-
resistant Staphylococcus aureus (MRSA) in Germany: implications for hospital hygiene. Infection 22 (Suppl.
2): S128-34, 1994.
Chapter 8
The Community Strategy against
Antimicrobial Resistance concerning
human medicine
Bronzwaer S, Lnnroth A, Haigh R. Submitted to the European Journal of
Public Health.
Abstract
In 2001 the European Commission agreed a Communication to the European Parliament
and the Council outlining the Community strategy against Antimicrobial Resistance. One
important part of that Community strategy is the 'Council Recommendation on the
prudent use of antimicrobial agents in human medicine', recommending Member States
to ensure that specific strategies to contain antimicrobial resistance are implemented at
national level. The Community strategy consists of fifteen actions in four key areas:
surveillance, prevention, research and product development, and international co-
operation. Of these fifteen actions comprising the strategy, this paper discusses eleven
points of action that are directly related to human medicine, and presents relevant
Community activities.
The European Commission has initiated, through its various services, a wide range of
activities. In past years the problem of antimicrobial resistance was addressed through an
increasing number of isolated measures, but through the Community strategy the
Commission has set a comprehensive and pro-active approach. Under the new public
health programme as well as under the Commissions research programmes,
antimicrobial resistance is of key priority, and specific priorities within the area of
antimicrobial resistance are presented.
Introduction
In recent years the problem of antimicrobial resistance has received increasing attention
and many activities have been started in parallel. Member States of the European Union
(EU) are progressively taking initiatives to contain antimicrobial resistance by, for
example, implementing national surveillance systems, training health professionals,
providing information to the public, etc. With many initiatives on-going there is added
value in sharing experiences and need for co-ordinating control efforts.
At level of the European Community (hereinafter referred to as Community) the problem
of antimicrobial resistance has been recognised as a public health priority and addressed
since several years. The Treaty of Amsterdam (1997) makes provisions for action directed
towards improving public health, preventing human illness and diseases, placing the
responsibility on Community Institutions (Commission and Council) and on the Member
States. At the EU conference The Microbial Threat, held in Copenhagen in September
1998, the participants unanimously agreed that antimicrobial resistance was no longer a
national problem, but a major international issue requiring a common strategy at
European level.
1
In May 1999 a Commissions committee of independent scientists (Scientific Steering
Committee) delivered its opinion on antimicrobial resistance with recommendations for
action.
2
It emphasised that antibiotics are of utmost importance to treat and contain
communicable diseases, but that increasing prevalence of antimicrobial resistance calls
for immediate action to limit their use to diseases with strict indications. The overall use
of antibiotics should be reduced in a balanced way in all areas: human medicine,
veterinary medicine, animal production and plant protection. In its advice, the Committee
pointed out the possible need to introduce effective legislation and regulation to support
the achievement of its recommendations.
For example, Europe took action to phase out the use of antibiotics as growth promoters
in farm animals, and is now adopting a Directive to this effect. Also in the fields of plant
protection and veterinary medicine many measures are in place and more legislative acts
are being agreed to limit the use of antibiotics. Prudent use of antimicrobials in man is
an important part of a comprehensive strategy on the overall use of these agents, and this
paper limits itself to discussing the problem in the area of human medicine.
In June 1999, the Council, drawing on recommendations of the 'Microbial Threat
conference', adopted a Resolution on antibiotic resistance 'A strategy against the
microbial threat', considering that preservation of the effectiveness of antibiotics for the
treatment of infection cannot be achieved by national initiatives alone, but requires a
common strategy and co-ordinated action at Community level.
3
To follow-up on the
recommendations from the Microbial Threat conference in Copenhagen, an invitational
EU conference on antimicrobial resistance was held in June 2001, in Visby, Sweden. Also,
in June 2001 the Commission approved a Communication to the European Parliament
and the Council outlining the Community strategy against Antimicrobial Resistance
(hereinafter referred to as Community strategy).
4
A few months later during the EU conference on Antibiotic Use in Europe, held in
November 2001 in Brussels, the Council Recommendation on the prudent use of
antimicrobial agents in human medicine was launched (hereinafter referred to as Council
Recommendation).
5,6,7
This Council Recommendation constitutes an important part of
the Community strategy. It recommends that Member States ensure that specific
strategies to contain antimicrobial resistance exist and are implemented at national level.
Member States are recommended to encourage a more prudent use of antimicrobial
agents and to take measures related to surveillance, education, information, prevention
and control, and research in co-operation with the Commission. The Council
Recommendation charges the Commission with the task of supporting Member States
efforts through the Community Network on the epidemiological surveillance and control
of communicable diseases (hereinafter referred to as Community Network).
8
This paper aims to present the Community strategy against antimicrobial resistance in
the area of human medicine, including the Council Recommendation that forms an
integral part of it, and to present related Community activities funded under the public
health programmes and under the framework programmes for research.
Community strategy against antimicrobial resistance
In past years the problem of antimicrobial resistance was addressed through an
increasing number of isolated measures. In its Communication of July 2001 the
Commission proposed one comprehensive Community strategy against antimicrobial
resistance, consisting of fifteen actions in four key areas: surveillance, prevention,
research and product development, and international co-operation (Table 8.1). Of these
fifteen actions comprising the Community strategy, we describe the eleven points of
action that are directly related to practices in human medicine.
Surveillance
Action 1 of the Community strategy concerns surveillance and requires the development
of surveillance networks at the European level, encouraging the participation also of non-
EU countries.
At EU level this is undertaken through the Community Network that came into place in
1998, with antimicrobial resistance as one of its priorities. The two main pillars of this
Community Network are epidemiological surveillance of communicable diseases and an
early warning and response system. One of the surveillance networks funded within this
Community Network is dedicated to antimicrobial resistance surveillance. This large
European Antimicrobial Resistance Surveillance System (EARSS) is a network of national
surveillance systems and currently comprises about 800 laboratories from 28 countries.
9
Laboratories follow a standardised EARSS protocol and submit quarterly susceptibility
data of invasive isolates of Streptococcus pneumoniae, Staphylococcus aureus, Escherichia
coli, Enterococcus faecium and Enterococcus faecalis. The main function of EARSS is to
monitor variations in resistance of indicator pathogens of main public health relevance
for targeting interventions and assessing effectiveness of national intervention
programmes. Also a number of other surveillance networks falling under the Community
Network look into the susceptibility of the pathogens under surveillance. For instance, the
Enter-net network performs surveillance of Salmonella and verotoxin-producing E. coli
(VTEC) infections including the susceptibility to antibiotics whilst EuroTB performs
surveillance of tuberculosis including (multi) drug resistance. Monitoring of susceptibility
of meningococci, gonococci and syphilis has also begun in the context of other
surveillance networks. Table 8.2 provides an overview and specifies web site addresses of
antimicrobial resistance related projects funded under the (1996-2002) Community
action programme on public health (hereinafter referred to as PHP).
In addition, through the European Commissions Fifth Framework Programme
(hereinafter referred to as FP5) for Research and Technological Development (1998-
2002)
10
, funding is provided for the development of improved technologies for
surveillance of antimicrobial resistance in tuberculosis, Pseudomonas aeruginosa,
meningococcal disease and salmonellosis as well as for research addressing the risk of
transmission between food animals and humans.
Also, as part of the Community strategy, the Council Recommendation calls on Member
States to establish or strengthen sustainable antimicrobial resistance surveillance
systems building upon existing national and international systems using, wherever
possible, internationally recognised classification systems and comparable methods. The
same recommendation holds for the monitoring of the use of antimicrobial agents that
in fact constitutes the second point of action.
Action 2 of the Community strategy sets out to put in place and improve the collection of
data on consumption of antimicrobial agents in all sectors. Sound data on the
consumption of antimicrobial agents are needed for developing intervention strategies.
Such data already exist in many Member States but they are scattered, heterogeneous,
and in many instances not easily accessible.
The Council Recommendation asks the Member States to co-operate with the
Commission to develop indicators to monitor prescribing practices. The Commission is
funding, through the PHP, the European Surveillance of Antimicrobial Consumption in
humans (ESAC) project that started in November 2001 with first results of the
retrospective data collection presented at the 13
th
ECCMID conference in Glasgow.
Through this network about 30 participating countries deliver comprehensive national
data on cost and volume of antimicrobial consumption in ambulatory and hospital care.
Prospective and standardised data collection starts as of 2003 and indicators for the
evaluation of the appropriateness of antimicrobial use will be developed. FP5 supports a
project to implement the defined methodologies of EARSS and ESAC in the
Mediterranean region, starting in 2003. In this Antimicrobial Resistance in the
Mediterranean (ARMed) project at present seven Mediterranean countries participate
(Malta, Cyprus, Turkey, Egypt, Tunisia, Morocco and Jordan).
To guide intervention it is critical to understand the relation between antimicrobial
resistance and use. A recent study using EARSS data showed that in the EU antimicrobial
resistance of S. pneumoniae to penicillin is correlated with use of beta-lactam antibiotics
and macrolides at country level.
11
To study and monitor further the link between
antimicrobial resistance data and antimicrobial use EARSS and ESAC linked their
respective databases.
Prevention
Actions in this key area of the Community strategy aim to stimulate work on the
prevention of communicable diseases, and infection control to reduce the need for
antimicrobial agents.
Action 3 of the Community strategy aims to increase the importance of antimicrobial
resistance information for the market authorisation process. Concerns have been
expressed by regulators in various European authorities that different indications, doses,
dose regimens (duration of treatment) and different pharmacodynamic information exist
for the same and similar products already licensed in the EU. National competent
authorities in consultation with the European Agency for the Evaluation of Medicinal
products (EMEA) are currently considering the issue of divergent product information,
and Member States are asked through the Council Recommendation to initiate activities
to evaluate, and as necessary, update and harmonise the summary of product
characteristics (SPC). The EMEA has published a discussion paper on antimicrobial
resistance outlining its activities and pointing out the need to find ways to promote new
effective antibiotics.
12
Criteria for market authorisation of new antibacterial medicinal
products are outlined in three EU guideline documents.
13,14,15
Action 4 of the Community strategy sets out to support, at Community level, educational
campaigns directed at professionals and the general public to avoid overuse and misuse
of antimicrobial agents. An FP5 supported research project, the European Resistance
Intervention Study (EURIS), evaluates different approaches to reduce the prevalence of
resistant pneumococci among children in European day-care centres. These interventions
include reduced use of antibiotics through education of doctors, day-care staff, parents
and children, optimised dosing, improved hygiene, notification of resistant strains and
isolation of carriers.
Another FP5 project, the Antibiotic Resistance Prevention And Control (ARPAC), aims at
identifying hospital policies and prescription patterns which are associated with lower
resistance rates. The objective of ARPAC is to evaluate and harmonise strategies for
prevention and control of antibiotic resistance in hospitals. Results are to be expected in
2004.
In addition, the Council Recommendation encourages Member States to promote
education and training of health professionals on the problem of antimicrobial resistance
in undergraduate and postgraduate training. Member States should also promote
training on hygiene and infection control standards and on immunisation programmes in
order to reduce the spread of micro-organisms. Also the general public should be
informed on the importance of prudent antimicrobial use by raising awareness of the
problem of antimicrobial resistance, proper prescription, good patient adherence, the
value of hygiene, and the impact of vaccination.
The Commission is taking this forward in part by funding (under the PHP) a Swedish Film
and Television company to produce in 2003 a TV-documentary on the battle against
resistant bacteria to be used as an educational tool to promote appropriate antimicrobial
use. This film should be followed up at national level by TV talk shows, newspaper articles
and school discussions.
Action 5 of the Community strategy sets out to fully apply the principle that antibacterial
substances are available in human and veterinary medicine by prescription only.
Antimicrobial agents for systemic use in human medicine are by law prescription-only
medicines in all Member States, but enforcement of this regulation varies. Under the PHP
a Self-medication with Antibiotics and Resistance levels in Europe (SAR) project is funded
that aims to quantify the consumption of antibiotics sold over-the-counter (without
prescription) and of leftover (prescribed) antibiotics hoarded at home. Building on the
EARSS and ESAC networks, this project will also investigate the possible impact of non-
prescribed consumption on the development of resistance.
Action 6 of the Community strategy aims to reinforce and promote prevention
programmes of infections in human and veterinary medicine, and in particular
immunisation programmes.
In the frame of the Community network on communicable diseases, surveillance
networks have been started on vaccine preventable diseases like measles, pertussis, and
rubella. The development of vaccination registers was started to evaluate best results in
terms of vaccination coverage in Member States.
The pneumococcal disease in Europe (PNC-Euro) project, funded in FP5, studies the
epidemiology of S. pneumoniae in a variety of European countries prior to the introduction
of new conjugate vaccines. The study will produce information to design cost-effective
prevention strategies against pneumococcal infection.
Containment of antimicrobial resistance is intrinsically linked to infection control
practices. Hospitals in Europe Link for Infection Control and Surveillance (HELICS) is a
Commission funded project (PHP) to monitor hospital acquired infections developing
protocols for databases on surgical and intensive care unit infections, and to setup
evidence-based infection control standards and recommendations.
Member States are asked by the Council Recommendation to implement preventive and
control measures by developing evidence-based principles and guidelines on good
practice for the management of communicable diseases, and controlling good practice of
marketing of antimicrobial agents. Member States should ensure proper implementation
of hygiene and infection control standards in health care facilities and in the community
and encourage national immunisation programmes.
Actions 7 to 10 refer to preventive action in the fields of growth promoters, food, and
environment and although very relevant to public health do not fall within the scope of
this paper.
Research and product development
Antimicrobial resistance has been for a long time part of the Community research
priorities. Already under the Fourth Framework Programme for Research and
Technological Development (1994-1998), some actions addressing antimicrobial
resistance were initiated. Along with the Commissions increasing efforts to contain
resistance, research efforts were significantly intensified during FP5. The programme is
currently funding about 80 projects related to antimicrobial resistance at a total
Commission contribution of over 100 million. This project portfolio addresses anti-
bacterial, anti-fungal, anti-viral and anti-protozoan resistance through various
approaches, ranging from basic mechanisms of emergence and transmission of
resistance, through development of new drugs and diagnostic tests to epidemiological
and public health research. A comprehensive overview of all these projects is available at:
http://www.cordis.lu/lifescihealth/major/drugs.htm
Action 11 of the Community strategy promotes the development of new antimicrobial
agents. About one third of the antimicrobial resistance portfolio of FP5 is devoted to the
development of new classes of anti-infectives. Some of these projects focus on microbial
functional genomic approaches, in particular against tuberculosis and S. aureus. Other
projects investigate novel potential molecular targets, such as the bacterial ribosome,
protein replication initiation, or secretion and adhesion mechanisms. Yet other projects
focus on the development of new concepts for antimicrobial drugs through the
exploitation of antibiotics-producing organisms.
Action 12 of the Community strategy encourages the development of alternative
treatments and vaccines. Current FP5 research includes the development of resistance
inhibitors, such as beta-lactam inhibitors for combination treatment, bacterial
conjugation inhibitors and inhibitors of bacterial adhesion at mucosal surfaces. Lactic
acid bacteria, already widely used as probiotics for human consumption, are now subject
to a rigorous biosafety evaluation study in the scope of an FP5 project. Vaccine
development is a major priority in FP5 and several research projects are currently on-
going. Special emphasis is given to tuberculosis, malaria, HIV/AIDS, and Hepatitis C
virus through multiple approaches, but also influenza, respiratory syncytical virus,
shigellosis and Neisseria meningitidis serogroup B are being addressed. Efforts are also
devoted to the development of novel vaccine delivery systems and formula. In addition,
under Article 169 of the Treaty of Amsterdam, a unique effort has been launched in Europe
to provide an infrastructure for clinical trials of vaccines against tuberculosis, malaria and
HIV/AIDS. This European and Developing Countries Clinical Trials Partnership (EDCTP)
has been set up through a joint collaborative initiative among Member States and
developing countries with the Commission as supporting partner. The main goal is to
support phase II and phase III clinical trials of promising new clinical interventions
against HIV/AIDS, malaria and tuberculosis in, with, and for, developing countries.
Action 13 of the Community strategy sets out to support the development of rapid and
reliable diagnostic and susceptibility tests. In order to achieve a long-term sustainable
prudent use of antimicrobials, improved access to rapid read-out, reliable and
inexpensive diagnostic tests is essential. Currently funded FP5 projects include the
development of sensors for multi-drug resistant strains of tuberculosis, a DNA chip based
diagnostic test for P. aeruginosa, nucleic acid based amplification methods for the
detection of respiratory pathogens in community acquired pneumonia and a network for
automated bacterial strain fingerprinting.
International co-operation
An effective Community strategy requires close co-operation and consultation between
the Commission, the Member States and other involved parties. Action 14 of the
Community strategy encourages the development of co-operation, co-ordination and
partnership at international level, in particular via existing international organisations.
The Commission and the World Health Organisation (WHO) have signed a
Memorandum of Understanding reconfirming their common interest in health.
Antimicrobial resistance is among the agreed priorities and close co-operation with WHO
has been ensured for all antimicrobial resistance related networks. The Commission is
developing a programme with WHO on strengthening pharmaceutical policies, including
rational use of drugs and particularly supporting national programmes to contain
antimicrobial resistance, through the expansion of projects that link surveillance data to
rational prescribing programmes.
The Commission is invited to take a leading role in the implementation of the Northern
Dimension Action Plan that includes actions on antimicrobial resistance, particularly
through the Baltic Sea States Task Force on communicable diseases control.
Out of the FP5 research portfolio on antimicrobial resistance, seven projects are
specifically focused on international issues, covering a broad spectrum of issues that
range from the control of use of antibiotics and resistance in Latin America to the
problem of drug resistance in Asian aquacultural environments.
Action 15 of the Community strategy attributes special attention to applicant and
developing countries by helping putting in place the appropriate structures. The
participation of non-member countries is foreseen in the Community Network and most
applicant countries already participate in antimicrobial resistance surveillance networks
and projects funded under the PHP. Active participation of applicant countries in projects
under the framework programmes for research is also encouraged.
Discussion
The public health problem of antimicrobial resistance is receiving more and more the
attention it requires, and over the years the Commission has given due priority to the
matter. As presented here for human medicine, various Commission services have
initiated wide-ranging activities and through a number of legal acts, recommendations for
action are made to the Member States.
7,8,16
On 23 September 2002, the European Parliament and the Council adopted a Decision
establishing a new programme of Community action in the field of public health (2003-
2008).
17
This programme modernises and combines into one framework the former eight
Community action programmes on public health carried out from 1996 to 2002. The new
programme provides the framework for an annual public health workplan and a funding
mechanism for projects addressing Community priorities. In the workplan 2003,
antimicrobial resistance and guidelines for best practice on prudent antimicrobial use are
defined as key priorities, and the Commission is calling for proposals in this area.
18
The Council invites the Commission to propose, where appropriate, common methodology,
case definitions, and nature and type of data to be collected for surveillance of antimicrobial
resistance and antimicrobial use. Surveillance networks now face problems in comparability
of susceptibility data because of differences in methodology and guidelines for susceptibility
testing. The Commission has therefore defined as priority in its public health workplan 2003
to stimulate activities that propose a common or harmonised methodology and possibly
common criteria for defining resistance.
Other Community challenges in the field of antimicrobial resistance that have priority in
the workplan 2003 are to support information exchange and co-ordination of education
and intervention programmes aimed at hospitals and the open population. Recent
reports about the MRSA problem in the UK illustrate the importance of hospital infection
control to contain the spread of antimicrobial resistance.
19
To build further on results of
existing projects, the Commission also calls for developing a permanent system for
information connecting interested parties such as prescribers, pharmacists, consumers,
health insurance, etc. on consumption of antimicrobials and related trends in resistance.
Applications for activities to update product information where necessary are also called
for.
An increasing number of countries are implementing a system for laboratory-reporting in
addition to the statutory notification of communicable diseases through physicians.
These systems offer additional benefit but need to meet basic requirements such as good
quality and comparable data from participating labs. As many countries are only just
beginning, there is added value if the Commission could bring together national groups
to exchange information and harmonise issues like: minimum data set, un-duplication,
external quality control of laboratories, confidentiality and access to data, algorithms for
early warning, etc. Through the new PHP, under the frame of the Community Network,
the Commission will support actions aimed at networking and co-operation between
European laboratories, with the aim to foster communication, to enhance quality
assurance and harmonisation of laboratory methods in order to ensure comparability of
data.
Furthermore, the Commission recently launched the Sixth Framework Programme (FP6)
for Research and Technological Development (2002-2006).
20
This programme is
supported by a set of new instruments designed to ensure a more effective way of carrying
out research in Europe. These instruments are 'Networks of Excellence', which aim to
structure, integrate and co-ordinate research resources and activities around a given topic
and Integrated Projects, which bring together complementary expertise to tackle specific
ambitious research objectives in a co-ordinated fashion. Research on antimicrobial
resistance has been selected as one of the priority areas and will thus receive a further
boost in FP6. The new instruments will be used to channel microbial and human genomic
research towards applications into new molecular drug targets, alternative therapeutic
and preventive strategies, new diagnostic and susceptibility tests, epidemiological
approaches and improved knowledge of molecular mechanisms behind resistance.
Furthermore, measures to provide scientific support to antimicrobial resistance in the
context of the Community Network are placed high on the FP6 agenda under policy-
oriented research. These new initiatives aim to further complement the Commissions
contribution to the Community strategy.
Accurate information regarding antimicrobial resistance and antimicrobial use is needed
to target interventions. Hence, each Member State should have an appropriate framework
in place to monitor accurately antimicrobial use and antimicrobial resistance. Effective
implementation requires a number of key features, including a clear action plan,
delegation of authority and power to act, resources and sound mechanisms to assess the
effectiveness of interventions, allowing feedback of results to influence future strategies.
Therefore the Council Recommendation asks Member States to put in place an
intersectoral mechanism for implementing relevant measures and for effective co-
ordination with other Member States and the Commission. No specific recommendations
are made to the nature of this mechanism, but one might assume that in this body, local,
regional and national health authorities, the legislator, professionals of the different
disciplines concerned, and consumers, would be represented. This national mechanism
should co-ordinate reporting structures at local and hospital level, prioritise the action
needed, and recommend the health authorities responsible for taking action.
The Commission created a working group of representatives of the different national
intersectoral mechanisms under the auspices of the Community Network to assist in
evaluating the implementation of the Council Recommendation. Member States are to
report to the Commission on the implementation of the Council Recommendation within
two years of its adoption. The Commission monitors this implementation closely and has
recently sent out a template for reporting to facilitate and structure the reports from
Member States. The Commission intends to follow-up on these reports including other
relevant actions under the Community strategy.
In conclusion, in past years the problem of antimicrobial resistance was addressed
through an increasing number of individual measures, but through the Community
strategy the Commission has set a comprehensive and pro-active approach, giving special
attention to applicant countries. Under the new public health programme as well as under
the Commissions research programmes, antimicrobial resistance is of key priority.
Table 8.1: The four key areas and fifteen priority actions of the Community strategy against antimicrobial
resistance
Key area Action
Surveillance
1. Develop co-ordinated and coherent surveillance networks at the European level.
Encourage the participation of non-EU countries and the links between already
established surveillance networks in human and veterinary medicines
2. Put in place and improve the collection of data on consumption of antimicrobial
agents in all sectors
Prevention
3. Increase the importance of antimicrobial resistance information for the market authorisation
process in human medicine, veterinary medicine and agriculture
4. Support, at Community level, educational campaigns directed at professionals (clinicians,
veterinarians, farmers) and the general public to avoid overuse and misuse of antimicrobial
agents
5. Fully apply the principle that antibacterial substances are available in human and veterinary
medicine by prescription only and distributed in a controlled way in agriculture, and evaluate
whether the prescription-only rule should be applied to all antimicrobial agents as a precaution
6. Reinforce and promote prevention programmes of infections in human and veterinary
medicine, in particular immunisation programmes
7. Reinforce the residue monitoring system in food as regards methods of analysis, sanctions
and reporting system
8. Phase out and replace antimicrobial agents used as growth promoters in feed
9. Review the use of the two authorised antimicrobial agents in food
10. Ensure that GMOs which contain genes expressing resistance to antibiotics in use for medical
or veterinary treatment are taken into particular consideration when carrying out an
environmental risk assessment, with a view to identifying and phasing out antibiotic resistance
markers in GMOs which may have adverse effects on human health and the environment
Research and product development
11. Encourage the development of new antimicrobial agents
12. Encourage the development of alternative treatments and vaccines
13. Support the development of rapid and reliable diagnostic and susceptibility tests
International co-operation
14. Encourage strongly the development of co-operation, co-ordination and partnership at
international level in particular via the existing international organisations
15. Pay special attention to candidate and developing countries by helping them putting in
place the appropriate structures
Table 8.2. Projects funded under the public health programme related to antimicrobial resistance
Acronym Full title Focus Co-ordinated by Web site
EARSS European Antimicrobial resistance RIVM, Bilthoven, www.earss.rivm.nl/
Antimicrobial of invasive isolates Netherlands
Resistance of S. pneumoniae,
Surveillance S. aureus, E. coli,
System E. faecium / faecalis
ESAC European Scientific Evaluation on University of www.uia.ac.be/esac
Surveillance the Use of Antimicrobial Antwerp, Belgium
Antibiotic Agents in Human Therapy
Consumption
EU-IBIS European Union Invasive Haemophilus PHLS, London, UK www.phls.org.uk/
Invasive Bacterial influenzae and inter/eu_ibis/
Infections Neisseria meningitidis aims.htm
Surveillance disease
Enter-net International Salmonella, infection PHLS, London, UK www.phls.co.uk/
surveillance with E. coli O157 inter/enter-net/
network for the menu.htm
enteric infections
EuroTB Surveillance of Tuberculosis including InVS, Paris, France www.eurotb.org/
tuberculosis in multi-drug resistance
Europe
HELICS Hospitals in Nosocomial infections Universit Claude http://helics.univ-
Europe Link for Bernard, Lyon, lyon1.fr
Infection Control France
through
Surveillance
SAR Self-medication Quantification of levels University of none
with antibiotics and of self-medication Groningen,
resistance levels (OTC and use of Netherlands
in Europe leftovers of antibiotics)
in different European
countries
TV-film The battle against Production of a television Meter-film, none
antibiotic resistant film on antibiotic Stockholm,
bacteria resistance aiming to raise Sweden
awareness of this problem
in the general public
References
1
State Serum Institute and Danish Veterinary Laboratory, eds. The Copenhagen recommendations on the
microbial threat. Ministry of Health, Ministry of Food, Agriculture and Fisheries, 1998.
2 Opinion of the Scientific Steering Committee on Antimicrobial Resistance - 28 May 1999. Available at:
http://europa.eu.int/comm/food/fs/sc/ssc/out50_en.html. Accessed 18 February 2003.
3
OJ C 195, 13.07.1999, p.1. Council Resolution of 8 June 1999 on antibiotic resistance "A strategy against the
microbial threat".
4 Com (2001) 333 final 20.06.2001. Commission of the European Communities. Communication from the
commission on a community strategy against antimicrobial resistance. Available at:
http://europa.eu.int/comm/health/ph/others/antimicrob_resist/am_02_en.pdf. Accessed 11 December
2002.
5 Midday Express. European conference on antibiotic use in Europe, 14 November 2001. Available at:
http://europa.eu.int/comm/dgs/health_consumer/library/press/press206_en.pdf. Accessed 11 December
2002.
6 Byrne D. European Commissioner for Health and Consumer Protection. The EU strategy on antimicrobial
resistance in humans. European Conference on Antibiotic Use in Europe Brussels, 15 November 2001.
Available at: http://europa.eu.int/rapid/start/cgi/guesten.ksh?p_action.gettxt=gt&doc=SPEECH/
01/542|0|RAPID&lg=EN. Accessed 11 December 2002.
7 OJ L34 of 5.2.2002, p.13. Council Recommendation of 15 November 2001 on the prudent use of
antimicrobial agents in human medicine (2002/77/EC). Available at: http://europa.eu.int/eur-
lex/pri/en/oj/dat/2002/l_034/l_03420020205en00130016.pdf. Accessed 11 December 2002.
8 OJ L 268. 3.10.98, p.1. Decision no. 2119/98/EC of the European Parliament and of the Council of 24
September 1998 setting up a network for the epidemiological surveillance and control of communicable
diseases in the Community. Available at: http://europa.eu.int/eurlex/pri/en/oj/dat/1998/l_268/
l_26819981003en00010006.pdf. Accessed 11 December 2002.
9 EARSS management team, advisory board and national representatives. EARSS Annual Report 2001.
Bilthoven, the Netherlands, July 2002. Pages 95. ISBN-number: 90-6960-098-6. Downloadable from EARSS
official web-site: www.earss.rivm.nl. Accessed 24 February 2002.
10
OJ L26, 1.2.1999, p.1. Decision of the European Parliament and of the Council, of 22 December 1998,
concerning the Fifth Framework Programme of the European Community for research, technological
development and demonstration (RTD) activities (1998-2002).
11
Bronzwaer S, Cars O, Buchholz U, Mlstad S, Goettsch W, Veldhuijzen I, Kool J, Sprenger M, Degener J, and
EARSS participants. A European Study on the Relationship between Antimicrobial Use and Antimicrobial
Resistance. Emerg Inf Dis 2002;8(3):278-82.
12 EMEA/9880/99. EMEA discussion paper on antimicrobial resistance. Available at:
http://www.emea.eu.int/pdfs/human/regaffair/988099en.pdf. Accessed 11 December.
13 EMEA document CPMP/EWP/558/95. Note for guidance on evaluation of new antibacterial medicinal
products. Available at: http://www.emea.eu.int/pdfs/human/ewp/055895en.pdf. Accessed 11 December
2002.
14 EMEA document CPMP/EWP/520/96. Note for guidance on the pharmacodynamic section of the SPC for
antibacterial medicinal products. Available at: http://www.emea.eu.int/pdfs/human/ewp/052096en.pdf.
Accessed 11 December 2002.
15 EMEA document CPMP/EWP/2655/99. Points to consider on pharmacokinetics and pharmacodynamics in
the development of antibacterial medicinal products. Available at: http://www.emea.eu.int/pdfs/human/
ewp/265599en.pdf. Accessed 11 December 2002.
16
OJ L 244, 30.09.1993, p.35. Council Directive 92/117/EEC of 17 December 1992 concerning measures for
protection against specified zoonoses and specified zoonotic agents in animals and products of animal
origin in order to prevent outbreaks of food-borne infections and intoxications.
17 OJ L 271, 09.10.2002, p.1. Decision No 1786/2002/EC of the European Parliament and of the Council of 23
September 2002 adopting a programme of Community action in the field of public health (2003-2008).
Work Plan 2003 for the implementation of the Public Health Programme (2003-2008). Documents
regarding the call for proposals are downloadable from: http://europa.eu.int/comm/health/index_en.html.
Accessed 27 March 2003.
18
Public Health Laboratory Service. Report for the Department of Health into rates of hospital infections
caused by methicillin-resistant Staphylococcus aureus (MRSA). February 2002.
19 OJ L232, 29.8.2002, p.1. Decision No 1513/2002/EC of the European Parliament and of the Council, of 27
June 2002, concerning the Sixth Framework Programme of the European Community for research,
technological development and demonstration activities contributing to the creation of the European
Research Area and to innovation (2002 to 2006). Available at http://www.cordis.lu/fp6/find-doc.htm
Chapter 9
General Discussion
Surveillance as a tool for action
To gain control over antimicrobial resistance it is essential to undertake appropriate
surveillance. Many different initiatives are in place to monitor antimicrobial resistance
and to understand the problem. In a review by Monnet et al. more than 20 different
multinational multicentre surveillance or research projects were identified that produce
data on antimicrobial resistance in Europe (1). Each of these surveillance components
plays a valuable part in a comprehensive programme, but no single part can provide all
the answers. In fact, various approaches are needed to answer different questions at the
basis of the resistance problem.
Surveillance entails data collection to come to action. The actions taken on basis of the
data will differ depending on the level at which the data are being collected and analysed.
Thus, local surveillance data should be used to guide clinical management, to update
treatment guidelines, educate prescribers and guide infection control policies, and to
promote improvements in quality and in communications.
Nationally collected surveillance data should be used to inform policy decisions, update
national formularies or lists of essential drugs and standard treatment guidelines and
evaluate cost-effectiveness of interventions (2). With travel and trade increasing over the
years, the risk of dissemination of (resistant) pathogens grows. Certain strains have been
shown to spread between European countries (3). There is a clear need to coordinate
international surveillance, as resistance rates found in different surveys cannot easily be
compared due to differences in study design, study population, and time period.
From a public health standpoint, it would be relevant to detect the incidence of infections
caused by resistant organisms among the total number of infections in a population (4).
Because it is difficult to relate the number of infections to a denominator (the catchment
population), results of surveillance are mostly given as the proportion of resistant
organisms among all organisms tested (prevalence of resistance, often called resistance
rate).
Important parameters to study the relevance of the resistance problem are the clinical
diagnosis and patient outcome. In most surveillance systems, unfortunately, these
parameters are not collected because they are mostly not included in Laboratory
Information Systems (5). It requires active surveillance where the microbiologist contacts
treating physicians to collect clinical information. This is often not done because of time-
and cost-constraints. It is essential however that the burden of antimicrobial resistance
on morbidity and mortality is better understood and quantified, possibly through special
studies as referred to in the Introduction of this thesis.
Notwithstanding these difficulties surveillance of resistance rates is fundamental to
understand trends in resistance, to target areas for interventions, and to assess the
effectiveness of the intervention.
In chapter 2 EARSS is defined as an international network of national surveillance systems
collecting antimicrobial resistance data aiming to assist in the control of antimicrobial
resistance. EARSS opted from the beginning to build on national surveillance systems and
to collate data first at national level before entering them into the central database. In this
way EARSS helped to standardise antimicrobial resistance surveillance at national level
and in some countries it even initiated the process of collection of susceptibility data. This
is important because the responsibility for taking action to control antimicrobial
resistance lies foremost at national level. The national representatives whom are formally
recognised by their national authorities collect samples of isolates by either total or
representative coverage allowing to produce official national resistance data that
constitute a basis for policy decisions. EARSS monitors trends of antimicrobial resistance,
which enables to target problem areas and monitor the effect of interventions. The results
of the EARSS project are brought to the attention of a broad public through Newsletters,
publications and annual reports. Latest results are always freely accessible on-line at
www.earss.rivm.nl.
Routine surveillance
Surveillance of routine susceptibility tests means that susceptibility tests performed in the
course of routine clinical care are captured and analysed. The output of the analyses can
consist of aggregated summaries or detailed stratified reports. A major advantage of
routine surveillance is that it is principally a matter of data management rather than
performing extra tests with consequent costs. With the advances in the information
technology domain we witness in many European countries that next to notification of
physicians of infectious diseases more and more use is made of laboratory notification
(6). Routine surveillance is very much in line with this development. Advantages and
disadvantages are summarised below:
Advantages of routine surveillance Disadvantages of routine surveillance
geographically and demographically routinely available specimens
representative little clinical information
broad range of questions can be no specimens for common clinical
addressed, both prospectively scenarios where most treatment
and retrospectively is empirical
follow baseline trends routine quality
identification of new and unexpected often a limited number of antibiotics
problems often only qualitative data
relatively inexpensive wide variety of laboratory
permits evaluation of the quality information systems
of routine data
From the start of EARSS it was decided that the system uses routinely generated data from
laboratories so that no changes to the primary diagnostic process are needed. As
discussed above, this approach has several limitations because not all participants use the
same method and quantitative data are limited, but it allowed to build up an enormous
large network with presently 800 laboratories in 28 countries reporting. In chapter 3 we
described how these limitations are addressed. S. pneumoniae and S. aureus were chosen
as indicator pathogens because these pathogens are routinely tested and breakpoints for
the most relevant antibiotics for these pathogens do not differ significantly among the
different national breakpoint committees. The EARSS protocol for susceptibility testing is
designed in order to standardise data collection allowing for collation of data among
participants. A limitation within EARSS is the problem of sampling bias, where clinicians
in some countries may request blood cultures more frequently than their colleagues in
other countries, who may sample only in case of empirical treatment failure. Attempts have
been made within EARSS to quantify blood culture request habits in different European
countries but without satisfactory results so far.
Bacteria readily exchange information, so should we
Surveillance data are essential to target and monitor interventions and above we
discussed that there is not one golden method to perform surveillance. Instead data
from a variety of sources are available, and should be used to target and monitor
interventions. The objective of establishing surveillance systems on antimicrobial
resistance (and on the consumption of antibiotics) is a public health function, what
implies that this information has to be in the public domain. The more so because in
intervention strategies many parties are involved. For this reason anonymised
surveillance data should be freely available for public health purposes. Moreover, it can be
argued that data which are routinely generated in public laboratories is publicly funded
and therefore should be considered to be publicly owned (7). However, in any organised
control strategy the issue of data ownership and accessibility should be explicitly
addressed and agreed among participants. In this respect we could and should learn from
micro-organisms who survive thanks to their ability and practise to exchange resistance
information (genes!) rapidly. Bacteria readily exchange information, so should we ...
Quality assurance
Antimicrobial resistance surveillance also can provide an opportunity to improve the quality
of susceptibility testing among those taking part in the surveillance (8). Chapter 4 describes
an external quality exercise (EQA) of antibiotic susceptibility testing for laboratories
participating in EARSS in order to assess the comparability of susceptibility test results
across countries and guidelines. This European-wide QA-exercise was characterised by an
excellent response rate with 433 (92%) of 471 laboratories from 23 countries reporting back.
It confirmed that an exercise of this dimension is feasible and demonstrated the
commitment of EARSS participants to quality. Strains were correctly identified at the genus
and species level, and the average concordances over all control strains were high. We
distributed control strains that tested the laboratories capability to identify the most
clinically relevant resistances (penicillin G in S. pneumoniae, methicillin in S. aureus and
glycopeptide in staphylococci) and feel reassured to continue using surveillance data
generated by the participating national surveillance systems. For continuous external quality
assessment we recommend that laboratories participate in national and international
schemes with frequent distributions of control strains. EQA gives credibility to the laboratory
as a responsible approach to quality can be demonstrated, it may identify deficiencies to be
rectified, and it is a stimulus for education of staff. International EQA exercises have the
additional benefits of strengthening collaboration between national groups and possibly
highlighting limitations of particular national methods (8).
Building on the EARSS experience and network, the European Committee on
Antimicrobial Susceptibility Testing (EUCAST) is currently planning a more structural
EQA approach with a central European strain collection that could be used by different
national and international schemes.
The quality assurance exercise also provided a good overview of the guidelines being
followed in Europe, showing that the NCCLS guideline are widely used. It should be noted
that for many antibiotics the breakpoints defining susceptibility or resistance of bacteria
to antimicrobial agents do not differ greatly between guidelines used in Europe. In our
experience EARSS adds to the process of harmonising breakpoints in Europe as is
brought forward by the EUCAST. The EUCAST has taken initiatives to propose an
international reference method and agreement on epidemiological MIC breakpoints to
which all other methods can relate.
Surveillance of resistance of S. pneumoniae and S. aureus
Several striking differences in the proportions of antimicrobial resistance for the two
indicator pathogens (S. pneumoniae, S. aureus) under surveillance by EARSS exist among
European countries (chapters 5 and 7). The consistency of data over the years and the
consistency with results of other antibiotic resistance surveillance projects confirm the
reliability of the data. It is clear that there is a problem with resistance in many European
countries and it seems that this problem is on the rise in hospital settings. In order to
target and monitor interventions EARSS monitors resistance over time and place. In
doing so it is important to distinguish community-acquired infections from hospital-
acquired infections that each have specific driving mechanisms. EARSS deliberately
started with the two mentioned indicator-pathogens, with S. pneumoniae as
representative of a pathogen that is community-acquired and S. aureus as an indicator of
hospital-acquired infections.
Community-acquired infections
Considerable increases in penicillin-resistant S. pneumoniae (PNSP) have been noted over
the past decade. (9, 10) EARSS data show there is a clear north-south gradient for the
proportion of invasive PNSP (chapter 5). The prevalence of invasive and penicillin non-
susceptible (20%) S. pneumoniae is highest in children 4 years and younger. This
emphasises the importance of physician and parent education about the prudent use of
antimicrobial agents, as well as the importance of new conjugate vaccines from which
children could benefit. This vaccine is currently introduced in a number of Member States
and hopefully EARSS may document a decrease in S. pneumoniae prevalence and PNSP
proportions in the coming years.
In many countries the proportion of macrolide resistant S. pneumoniae is high. It is
apparent from chapter 5 that penicillin and macrolide resistance is often associated. Thus
in situations where penicillin and erythromycin resistance is common, the empirical use
of macrolides should be discouraged. There are large differences between countries in the
resistance proportions of PNSP for 3
rd
generation cephalosporins and fluoroquinolones.
The development of new drugs cannot be relied on to contain antimicrobial resistance.
One important driver of antimicrobial resistance is high levels of antimicrobial use
(chapter 6). Other important determinants are inappropriate use of antibiotics dependent
on drug-seeking behaviour of patients, prescription behaviour, compliance of patients,
and over the counter availability of antimicrobials (11, 12, 13). Hence, data needed to come
to intervention are resistance rates and antibiotic usage data in the community.
Furthermore, patients and doctors attitudes as well as over the counter availability of
antimicrobials need to be studied.
In chapter 6, proportions of resistance for invasive S. pneumoniae isolates from EARSS
were related to data on the usage of antibiotics. It provided evidence of a strong
correlation of penicillin resistance in S. pneumoniae, and the use of beta-lactam antibiotics
and macrolides at country level. Building on the EARSS network, ESAC (European
Surveillance of Antimicrobial Consumption) collects comprehensive national data on cost
and volume of antimicrobial consumption in ambulatory and hospital care from about 30
participating countries. The prospective and standardised data collection is starting as of
2003 and definitions of indicators for antimicrobial use will be developed. EARSS and
ESAC co-operate closely, aiming at linking antibiotic resistance and antibiotic usage data
not only at country level, but possibly at regional / local level. Self-medication with
antibiotics may be an important component of the use of antibiotics in some European
countries, but little data is available yet. Chapter 6 also suggests a direct relationship
between non-adherence rates and antimicrobial resistance. If non-adherence is also
related to sales of antimicrobial agents, it could potentially confound the relationship
between use and resistance. The influence of this parameter should be quantified and
understood. Therefore, a self-medication study will start in 2003. EARSS will also co-
operate with ESAC and the Antibiotic Resistance Prevention and Control (ARPAC) survey
to study the correlation between resistance, antibiotic usage and the policy regarding
guidelines and infection control.
In some Member States it has proved possible to bring about a change in the level of
resistance by restricting the use of antibiotics (14, 15). Such action is needed at all levels:
in hospitals, and at national and international level (16).
Hospital-acquired (nosocomial) infections
Methicillin-resistant Staphylococcus aureus is common in many centres. Among European
countries an overview is given in chapter 7, showing that southern European countries,
the United Kingdom and Ireland reported the highest proportions, whereas northern
European countries had proportions of MRSA in bacteraemia patients below 1%. The
uncompromising "search-and-destroy" policy in the Netherlands and in Nordic countries
appears to be effective in controlling the emergence of MRSA. There are strong
indications for a rapid increase in the prevalence of MRSA in the UK (with 6% increase
per year) and in Germany (almost 4% increase per year). Numerous factors responsible
for antimicrobial resistance in hospitals have been identified and can broadly be classified
in four categories (17):
1. antimicrobial use issues such as overuse, misuse and co-usage of antibiotics,
2. infection control issues such as compliance to barrier precautions, workload,
existence of outbreaks, reservoirs and patient transfers,
3. patient issues such as severity of illness and utilisation of medical devices, and
4. community issues including prevalence of resistance in the primary health care sector
and in animals.
Models for the interpretation of the results of concomitant surveillance of antimicrobial
resistance and antimicrobial use in hospitals have been proposed but should be further
validated (18, 19). There is no consensus yet on the level of stratification of results.
Antimicrobial resistance levels and antibiotic consumption are generally calculated for an
entire hospital; however, large differences can be observed among units within one
hospital (20, 21). Consequently, analyses of data from surveillance systems should
probably be stratified at the unit level or at least by type of unit.
EARSS data indicate which countries face a problem with MRSA (as an indicator of
hospital-acquired infections) and is a clear stimulus for national authorities to enforce
strict hospital infection control guidelines and programs.
Community strategy
In chapter 8 we discussed how EARSS fits into the overall Community strategy against
antimicrobial resistance. EARSS (and ESAC) are explicitly mentioned in the recitals of the
Council Recommendation on the prudent use of antimicrobial agents in humans as
European Surveillance systems. Various Commission services have initiated a wide range
of activities. Community action is not only taken in the area of surveillance but also on
prevention, education, research, and product development. In past years the problem of
antimicrobial resistance was addressed through an increasing number of isolated
measures, but through its Communication the Commission has taken a very pro-active
approach outlining one comprehensive Community Strategy against Antimicrobial
Resistance.
Accurate information regarding antimicrobial resistance and antimicrobial use is at the
basis to target interventions. Hence, each Member State should have an appropriate
framework in place to monitor accurately antimicrobial use and antimicrobial resistance.
EARSS collects routinely generated laboratory data allowing comparison of susceptibility
data among countries. However, national action plans to contain antimicrobial resistance
should incorporate aspects of different surveillance strategies, for example the addition of
intermittent national surveys to answer specific questions.
Effective implementation requires a number of key features, including a clear action plan,
delegation of authority and power to act, resources and sound mechanisms to assess the
effectiveness of interventions, allowing feedback of results to influence future strategies.
Therefore the Council Recommendation asks Member States to put in place an
intersectoral mechanism. In this body, local, regional and national health authorities, the
legislator, professionals of the different disciplines concerned, and consumers, should
possibly be represented. Differences in prevalence, health care systems and problem
diseases may all influence the national approach taken to contain antimicrobial
resistance. These intersectoral mechanisms should play an important role co-ordinate
reporting structures at local and hospital level, prioritise the action needed, and
recommend the health authorities responsible for taking action.
References
1. Monnet DL. Characteristics of multicenter surveillance and research projects on antimicrobial resistance in
Europe and the United States. Copenhagen, Denmark: Division of Microbiology, Statens Serum Institut;
1998.
2. Kim T, Oh PI, Simor AE. The economic impact of methicillin-resistant Staphylococcus aureus in Canadian
hospitals. Infect Control Hosp Epidemiol. 2001 Feb;22(2):99-104.
3. Mato R, Santos Sanches I, Venditti M, Platt DJ, Brown A, Chung M, de Lencastre H. Spread of the
multiresistant Iberian clone of methicillin-resistant Staphylococcus aureus (MRSA) to Italy and Scotland.
Microb Drug Resist 1998; 4: 107-12.
4. Williams RJ, Ryan MJ. Surveillance of antimicrobial resistance an international perspective. BMJ
1998;317:651.
5. GAO, Report to Congressional Requesters. Antimicrobial Resistance, Data to assess public health threat
from resistant bacteria are limited. Available at: http://www.cdc.gov/. Accessed May 20, 2002.
6. Hedlund J, Olsson-Liljequist B. Control programme for antibiotic-resistant pneumococci; mandatory
notification sometimes neglected by physicians. Lkartidningen 1997; 94:4914-18.
7. Bronzwaer SLAM, Buchholz U, Kool J. International surveillance of antimicrobial resistance in Europe: now
we also need to monitor antibiotic use. Eurosurveillance 2001;6;1:1-2.
8. Kahlmeter G, Brown D. Resistance surveillance studies comparability of results and quality assurance of
methods. J. Antimicrob. Chemother. 2002 50: 775-7.
9. Appelbaum PC. Antimicrobial Resistance in Streptococcus pneumoniae: An overview. Clin Infect Dis 1992;
15, 77-81.
10. Baquero F. Pneumococcal resistance to beta-lactam antibiotics: A global geographical overview. Microb
Drug Resist. 1995; 1, 115-20.
[news]. JAMA 1996;275:175.
12. Mainous AG. An evaluation of statewide strategies to reduce antibiotic overuse; 2000. Family Medicine.
13. Branthwaite A. Pan-European survey of patients attitudes to antibiotics and antibiotic use; 1996, The J of
Intern Med Research.
14. Mlstad S, Cars O. Major change in the use of antibiotics following a national programme: Swedish
Strategic Programme for the Rational Use of Antimicrobial Agents and Surveillance of Resistance
(STRAMA). Scand J Infect Dis 1999; 31(2): 191-5.
[news]. JAMA 1996; 275: 175.
16. Williams R, Ryan M. Surveillance of antimicrobial resistance an international perspective. BMJ 1998; 317;
651.
17. Monnet D. Toward multinational antimicrobial resistance surveillance systems in Europe. Int J Antimicrob
Agents. 2000 Jul;15(2):91-101.
18. Monnet DL, Archibald LK, Phillips L, et al. Antimicrobial use and resistance in eight US hospitals:
complexities of analysis and modeling. Infect Control Hosp Epidemiol 1998;19:388-94.
19. Ballow CH, Schentag JJ. Trends in antibiotic utilization and bacterial resistance. Report of the National
Nosocomial Resistance Surveillance Group. Diagn Microbiol Infect Dis1992;15:37S-42S.
20. Pierson CL, Friedman BA. Comparison of susceptibility to beta-lactam antimicrobial agents among bacteria
isolated from intensive care units. Diagn Microbiol Infect Dis 1992;15(2 Suppl):19S-30S.
21. Stratton CW 4th, Ratner H, Johnston PE, Schaffner W. Focused microbiologic surveillance by specific
hospital unit: practical application and clinical utility. Clin Ther 1993;15(Suppl A):12-20.
Summary
The emergence of resistance is to some extent an inevitable result of the therapeutic use
of antibiotics. Killing or suppressing the micro-organisms that are sensitive to
antimicrobials allows for naturally drug-resistant ones to emerge. Antimicrobial
resistance makes infections more difficult to treat, and may increase the length and
severity of illness. Some 60 years after the introduction of penicillin in clinical practice,
antimicrobial resistance has become a worldwide public health concern, requiring
international strategies for its control.
The problem can not be overcome by continuously developing new drugs. An important
complementary step is to avoid further increases in resistance by reducing unnecessary
and inappropriate use of antibiotics. Measures are needed to slow the emergence of
resistance and to limit its spread. Surveillance of antimicrobial resistance is a first step
towards containment of the problem and serves to assess the scale of the resistance
problem, to monitor changes in resistance rates, and to provide a measure of the
effectiveness of interventions aimed at reducing resistance.
This thesis describes the set up of a European antimicrobial resistance surveillance
system (EARSS) and its contribution to the Community strategy against antimicrobial
resistance.
EARSS is set up as an international network of national surveillance systems, collecting
comparable and validated antimicrobial resistance data for public health purposes. The
objectives, infrastructure and data management aspects of the surveillance system were
defined by consensus of leading microbiologists and epidemiologists in Europe. At the
kick-off meeting the community-acquired pathogen S. pneumoniae and the hospital-
acquired pathogen S. aureus were chosen as most relevant pathogens to start surveillance
for in EARSS. During the same meeting the EARSS protocol for susceptibility testing was
developed, aiming to standardise data collection to allow for comparison of susceptibility
data among participants. To minimise sample bias, it was decided to report only the first
isolate of S. pneumoniae from blood and cerebrospinal fluid and the first S. aureus isolate
from blood. This European initiative acted as a catalyst for national surveillance systems.
To assess the comparability of susceptibility test results across countries and guidelines,
an external quality assurance exercise of antibiotic susceptibility testing for laboratories
participating in EARSS was organised. Overall, 433 (92%) of 471 laboratories from 23
countries reported back. Of the 8685 tests that were assessed, 8322 (96%) were
interpreted correctly by the participants. Concordance for detection of penicillin
resistance in the three S. pneumoniae control strains was 96%, 90% and 87%,
respectively. Laboratories performed extremely well in detecting oxacillin resistance in the
homogeneously methicillin-resistant S. aureus (MRSA) strain, but the concordance rate
dropped from 100% to 77% in the heterogeneously resistant MRSA strain. The NCCLS
guideline was the most frequently followed, by 61% of laboratories from 19 countries. It
was concluded from this exercise that the comparability of susceptibility data for penicillin
resistance in S. pneumoniae and for homogeneous methicillin resistance in S. aureus is
satisfactory among European countries and across guidelines.
Over 1999, 2000, and 2001 EARSS collected susceptibility data from 15.288 S. pneumoniae
isolates from 26 European countries. Southern European countries reported higher
proportions of penicillin non-susceptible S. pneumoniae (PNSP) than countries in
northern Europe. Prevalence of invasive S. pneumoniae was seasonal with clear peaks
during winter, but the prevalence of PNSP showed no seasonality. The proportion of
invasive S. pneumoniae isolates being non-susceptible was highest in children 4 years and
younger, underlining the importance of prudent antimicrobial use and vaccination in this
age group.
In an ecological study the relationship between penicillin non-susceptibility of invasive
isolates of S. pneumoniae and antibiotic sales was examined. Information was collected
on 1998-99 resistance data for PNSP through EARSS and on outpatient sales during 1997
for beta-lactam antibiotics and macrolides. The study showed that in Europe
antimicrobial resistance of S. pneumoniae to penicillin is correlated with use of beta-
lactam antibiotics and macrolides at country level, demonstrating one aspect of the
applicability of antimicrobial resistance data.
Susceptibility data from 32.942 invasive S. aureus isolates from 26 European countries
have been collected by EARSS over the period from 1999 - 2001. Methicillin-resistant
Staphylococcus aureus (MRSA) is common in many centres and high proportions of MRSA
are found in several European countries. Southern European countries, the United
Kingdom and Ireland reported the highest proportions (over 30%), whereas northern
European countries had proportions of MRSA in bacteraemia patients below 1%. The
uncompromising search-and-destroy policy in the Netherlands and in Nordic countries
appears to be effective in controlling the emergence of MRSA. In many countries the
proportion of MRSA seems to be relatively stable over 1999 - 2001. However, EARSS data
show an overall increase in the prevalence of MRSA by about 1.6% per year, with a
particular rapid increase in the UK (6% increase per year) and in Germany (almost 4%
increase per year).
EARSS is part of the broader Community strategy against antimicrobial resistance. This
strategy defines necessary action at Community level to contain the emergence and
spread of antimicrobial resistance. Community action is not only taken in the area of
surveillance but also on prevention, education, research, and product development. In
past years the problem of antimicrobial resistance was addressed through an increasing
number of individual measures, but through its Communication the European
Commission has taken a pro-active approach outlining one comprehensive Community
strategy. Accurate information regarding antimicrobial resistance is at the basis to target
interventions. Hence, each Member State should have an appropriate framework in place
to monitor accurately antimicrobial resistance and use. National intersectoral
mechanisms have an important role to co-ordinate reporting structures at local and
hospital level, prioritise the action needed, and recommend the national health
authorities responsible for taking action.
EARSS is built on national surveillance systems through which routinely generated data
are collated first at national level, helping to standardise antimicrobial resistance
surveillance and in some countries even initiating the process of collection of
susceptibility data. The national representatives whom are formally recognised by their
national authorities collect samples of isolates by either total or representative coverage
producing official national resistance data that constitute a basis for policy decisions.
Quality assurance exercises show that susceptibility data generated by these national
surveillance systems are valid and comparable. Through EARSS, European data are
collected and analysed, and feedback is provided through Newsletters, publications and
annual reports. Latest results are always freely accessible on-line at www.earss.rivm.nl. An
enormous large network has been built up with presently 800 laboratories in 28 countries
reporting. EARSS monitors trends of antimicrobial resistance enabling to target problem
areas and monitoring the effect of interventions.
Finally, our studies in the framework of EARSS have shown that the level of antimicrobial
resistance varies markedly among countries. This is most likely the result of differences
in antimicrobial consumption and hospital infection control. As a consequence, policies
to contain resistance should be tailored to national (and local) need. In order to better
understand and quantify the impact of different determinants of antimicrobial resistance
several initiatives have been started through or in close collaboration with the EARSS
network.
Nederlandse Samenvatting
Europese antimicrobile resistentie
surveillance, onderdeel van een
Communautaire strategie
Het ontstaan en de verspreiding van resistentie is in zekere zin een onvermijdelijk
resultaat van het therapeutisch gebruik van antibiotica. Het elimineren of onderdrukken
van micro-organismen die gevoelig zijn voor antibiotica staat de natuurlijk ongevoelige
bacterin toe om te groeien. Antimicrobile resistentie maakt infecties moeilijker
behandelbaar, en kan de duur en ernst van de ziekte verergeren. Inmiddels, zo'n 60 jaar
na de introductie van penicilline in de klinische praktijk, is antimicrobile resistentie een
volksgezondheidsprobleem van wereldwijde omvang geworden, en vereist internationale
controle maatregelen.
Het probleem kan niet worden opgelost door steeds maar nieuwe antibiotica te
ontwikkelen. Maatregelen zijn nodig om het opkomen van resistentie en de verspreiding
ervan tegen te gaan, bijvoorbeeld door onnodig en oneigenlijk gebruik van antibiotica
tegen te gaan. Surveillance van antimicrobile resistentie is een eerste stap in de richting
om het probleem te beheersen en dient om de omvang van het probleem te begrijpen,
om veranderingen in resistentie percentages te monitoren, en om een maatstaf te bieden
voor de effectiviteit van resistentie-verlagende interventies.
Dit proefschrift beschrijft het opzetten van het 'European Antimicrobial Resistance
Surveillance System' (EARSS) en de bijdrage hiervan aan de 'Communautaire strategie
tegen antimicrobile resistentie'.
EARSS is opgezet als internationaal netwerk van nationale surveillance systemen, en
verzamelt vergelijkbare en gevalideerde antimicrobile resistentie data voor
volksgezondheidsdoeleinden. De doelstellingen, infrastructuur en data management
aspecten van het surveillance systeem werden op basis van consensus vastgesteld door
Europese experts op het gebied van de medische microbiologie en de epidemiologie van
infectieziekten. Tijdens de eerste bijeenkomst werden de 'open-populatie-verworven'
Streptococcus pneumoniae en de 'ziekenhuis-verworven' Staphylococcus aureus gekozen als
meest relevante ziektekiemen om te surveilleren. Tijdens dezelfde ontmoeting werd het
EARSS protocol ontwikkeld om de gegevens verzameling te standaardiseren met als doel
om gevoeligheidsgegevens tussen deelnemers te kunnen vergelijken. Om selectie bias te
minimaliseren werd besloten om enkel de eerste S. pneumoniae stam uit bloed en hersen-
liquor en de eerste S. aureus stam uit bloed te rapporteren. Dit Europese initiatief diende
als katalysator voor nationale surveillance systemen.
Om de vergelijkbaarheid van gevoeligheidsresultaten te testen tussen landen en
richtlijnen voor gevoeligheidsbepalingen werd een externe kwaliteit controle
georganiseerd voor deelnemende laboratoria in EARSS. 433 van de 471 deelnemende
laboratoria (92%) uit 23 landen rapporteerde terug. Van de 8685 testen die werden
beoordeeld zijn er 8322 (96%) correct genterpreteerd door de deelnemers. Correcte
detectie van de penicilline resistentie in de drie S. pneumoniae controle stammen was
respectievelijk 96%, 90% en 87%. Laboratoria presteerden uitzonderlijk goed in de
detectie van oxacilline resistentie in de homogeen methicilline resistente S. aureus
(MRSA) stam, maar de proportie correct detecterende laboratoria zakte van 100% naar
77% voor de heterogeen resistente MRSA stam. De Amerikaanse NCCLS richtlijn was de
meest gevolgde richtlijn door 61% van de laboratoria in 19 landen. Uit de resultaten van
de studie kan de conclusie worden getrokken dat de vergelijkbaarheid van
gevoeligheidsdata voor penicilline resistentie in S. pneumoniae en voor homogene
methicilline resistentie in S. aureus bevredigend is tussen Europese landen en tussen
richtlijnen.
Over de jaren 1999, 2000, en 2001 verzamelde EARSS gevoeligheidsdata van 15.288 S.
pneumoniae stammen uit 26 landen. Zuidelijke landen rapporteerden hogere proporties
van penicilline niet-gevoelige S. pneumoniae (PNSP) dan landen uit Noord-Europa.
Prevalentie van invasieve S. pneumoniae was seizoensafhankelijk, met duidelijke pieken in
de winter. De prevalentie van PNSP vertoonde echter geen seizoensvariatie. De proportie
van invasieve S. pneumoniae stammen die niet gevoelig waren voor penicilline was het
hoogst in kinderen jonger dan 4 jaar, wat het belang van verstandig antibiotica gebruik en
vaccinatie in deze leeftijdsgroep benadrukt.
In een andere studie werd de relatie tussen penicilline ongevoeligheid van invasieve S.
pneumoniae en antibioticumgebruik bestudeerd. EARSS resistentie data uit 1998-99 voor
PNSP werd bestudeerd in relatie tot de verkoopcijfers in de open-populatie voor
betalactam antibiotica en macroliden. De studie toonde aan dat in Europa antimicrobile
resistentie van S. pneumoniae voor penicilline is gecorreleerd met het gebruik van
betalactam antibiotica en macroliden op landsniveau. Dit demonstreert n aspect van
het nut van gegevens over antimicrobile resistentie en gebruik.
Gevoeligheidsdata van 32.942 invasieve S. aureus stammen uit 26 Europese landen
werden verzameld door EARSS over de periode 1999-2001. MRSA is endemisch in veel
ziekenhuizen en hoge MRSA proporties werden gevonden in verschillende Europese
landen. Zuidelijke Europese landen, het Verenigd Koninkrijk en Ierland rapporteerden de
hoogste proporties (>30%), terwijl noordelijke Europese landen MRSA proporties in
sepsis patinten rapporteerden lager dan 1%. Het 'search-and-destroy' beleid in
Nederland en in de Noordelijke landen lijkt effectief om het opkomen van MRSA te
controleren. In veel landen lijkt de MRSA prevalentie relatief stabiel over 1999-2001. Maar
EARSS data laat over het geheel genomen toch een stijging in de MRSA prevalentie zien
van ongeveer 1.6% per jaar, met een bijzonder snelle stijging in het Verenigd Koninkrijk
(6% stijging / jaar) en in Duitsland (bijna 4% stijging / jaar).
EARSS is deel van een brede 'Communautaire strategie tegen antimicrobile resistentie'.
Deze strategie definieert de noodzakelijke maatregelen op Gemeenschapsniveau om het
opkomen en verspreiden van antimicrobile resistentie te controleren. Niet alleen voor
surveillance wordt Communautaire actie ondernomen maar ook op de gebieden van
preventie, voorlichting, onderzoek, en productontwikkeling. Gedurende de laatste jaren
werd het probleem van antimicrobile resistentie bestreden door een toenemend aantal
losstaande maatregelen, maar in dit Communiqu stelt de Europese Commissie een
actieve aanpak en een veelomvattende Communautaire strategie voor. Exacte informatie
over antimicrobile resistentie ligt ten basis aan het bepalen van toegespitste interventies.
Iedere lidstaat moet dus in staat zijn om antimicrobile resistentie en gebruik nauwgezet
te kunnen monitoren. Nationale multidisciplinaire werkgroepen hebben een belangrijke
rol om rapportages uit lokaal en ziekenhuis niveau te cordineren, initiatieven te
prioritiseren en om aanbevelingen te doen aan nationale autoriteiten die verantwoordelijk
zijn voor actie.
EARSS stoelt op nationale surveillance systemen die allereerst op nationaal niveau routine
laboratoria data verzamelen. EARSS helpt hiermee een standaardisatie van antimicrobile
resistentie surveillance in de hand en in sommige landen heeft het zelfs het verzamelen
van gevoeligheidsdata op nationaal niveau genitieerd. De nationale vertegenwoordigers,
die formeel worden erkend door nationale autoriteiten, verzamelen gegevens over
stammen door een surveillance systeem ofwel met totale maar in ieder geval met
representatieve dekking. Dit is nodig om officile nationale resistentie data te produceren
die aan de basis liggen voor beleidsbeslissingen. Kwaliteit controle laat zien dat de
gevoeligheidsdata die door deze nationale surveillance systemen worden geproduceerd
vergelijkbaar en betrouwbaar zijn. Door EARSS worden Europese data verzameld en
geanalyseerd, en feedback wordt verzorgd in de vorm van Newsletters, publicaties en
jaarrapporten. De laatste resultaten zijn altijd vrij toegankelijk on-line via
www.earss.rivm.nl. Een enorm breed netwerk is opgebouwd waarin momenteel 800
laboratoria uit 28 landen rapporteren. EARSS volgt ontwikkelingen in antimicrobile
resistentie om probleem gebieden te detecteren en het effect van interventies te
monitoren.
Tenslotte, onze studies binnen het EARSS netwerk hebben laten zien dat het niveau van
antimicrobile resistentie aanmerkelijk verschilt tussen landen. Dit is waarschijnlijk vooral
het gevolg van verschillen in antimicrobieel gebruik en ziekenhuisinfectie controle.
Dientengevolge moet het beleid om resistentie te beperken aangepast zijn aan nationaal
(en lokaal) niveau. Om beter de invloed van verschillende determinanten van
antimicrobile resistentie te begrijpen en te kwantificeren zijn er verschillende initiatieven
gestart door of in nauwe samenwerking met EARSS.
Acknowledgements
An important part of any thesis is acknowledging the many persons that have contributed.
I would like to thank all the people I have been working with in these past years, as well
as family, friends, and focolare.
I would like to thank explicitly the many EARSS national representatives and data
managers (see table) with whom I have enjoyed working closely for almost 4 years. I
express my gratitude for your time, dedication, and confidence in building up of what has
become a largely extended and well functioning network. Indeed, EARSS is a peoples
network. We witnessed that this model of European collaboration inspired more and
more countries to participate as well as that it provided an ideal platform for related
activities. I am convinced that in bringing EARSS further successfully your endeavor will
continue to contribute to the containment over the problem of antimicrobial resistance. I
also thank the more than 800 laboratories for their essential efforts in providing so
constantly their input.
Well-appreciated occasions to meet with collaborators in EARSS were the regular
meetings of the 'EARSS Advisory Board' and 'EARSS Quality Assurance Committee'.
These meetings really set out the way to bring EARSS forward and I learned a lot from all
participants. In particular I would like to mention the close collaboration with the
European Society of Clinical Microbiology and Infectious Diseases (ESCMID), the
European Committee on Antimicrobial Susceptibility Testing (EUCAST), with the Centre
national de Rfrence des AntiBiotiques (CRAB), and with the UK National External
Quality Assurance Scheme (NEQAS), in particular with Jerry Snell. The collaboration with
WHO has been very valuable through the persons of Rosamund Williams and John
Stelling whose positive support was and is and ever shall be appreciated.
Good collaboration is built on professionalism and trust and these ingredients have been
present from the beginning in the collaboration with Herman Goossens and Monique
Elzeviers from the European Surveillance on Antimicrobial Consumption project (ESAC),
and with Floor Haaijer-Ruskamp from the Self-medication with antibiotics and resistance
levels in Europe project (SAR). I have also appreciated the collaboration with other
Commission funded surveillance projects under the umbrella of the EU Network on
Communicable disease surveillance and control. Acknowledgement for the financial
support is done elsewhere but here I would like to thank the collegial support of staff
members of the European Commission.
Thanks are due also to the Dutch Ministry of Health (in particular Marja Esveld and Trudy
van Dijk) as well as to management at RIVM (special thanks to Marc Sprenger, who is
actually at the basis of EARSS and this thesis, Jacob Kool, and Marina Conyn, as
subsequent Heads of the unit infectious disease epidemiology). I thank also Daan
Kromhout for supporting me in doing my Masters of Public Health where I appreciated
working together with Niek Klazinga. I thank the Netherlands School of Public Health and
Johan Mackenbach who triggered the idea for performing the ecological study in Chapter
6.
During these years I have received much professional satisfaction and enjoyed day-by-day
work together with the EARSS Management Team (Nienke Bruinsma, Jos Monen, Han de
Neeling, Carola Schinkel, Paul Schrijnemakers, Edine Tiemersma, Jose van de Velde, and
in earlier days also Peter Bootsma, Udo Buchholz, Sandra van Dissel, Wim Goettsch,
Mireille Greijmans, and Irene Veldhuijzen). Since I moved to Luxembourg I was happy to
see Paul and others manage EARSS so successfully and I most warmheartedly welcome
Hajo Grundmann who embodies the perfect profile of an EARSS project leader.
A very special thank you goes to my 'Paranimphen'. Udo, in a relatively short period you
have achieved more than could be expected and in particular I thank you for your work on
Chapter 6. Paul, I have great respect for your management skills, your critically positive
spirit and your sense for team-work and loyalty.
I also thank all my other colleagues from the infectious diseases epidemiology unit who
form a great team of young and enthusiastic professionals and in particular I thank the
support of Eric Elbers, Yves van de Berg, and the persons I shared office with: Liesbeth
van Eerden and Jacco Wallinga. Appreciation also for the work of the RIVM-studio.
Finally, I thank John Degener for the many times he has freed himself from his many
duties to provide essential support in keeping EARSS going. It was always a pleasure to
work with and learn from you.
Last but certainly not least I thank Pa, to whom I dedicated this thesis, and Ma for their
endless energy, love, and trust invested in my person. I thank my 4 'big' brothers for the
wonderful family they are and for keeping me alive
To conclude in supremo I think I owe the biggest grazie to my wife Karin for the infinite
support and for delivering the most important output of these years: Angelique, Laura,
and Nikita.
Karin it has been a wonderful 5 years; I look forward to many more
National representatives and data managers in EARSS over 1998-2002
Austria Belgium Bulgaria Croatia
W. Koller H. Goossens B. Markova S. Kalenic
S. Metz E. Hendrickx H. Velinov A. Tambic-Andrasevic
H. Mittermayer F. van Loock
M. Struelens
J. Verhaegen
Czech Republic Denmark Estonia France
V. Jakubu T.L. Soerensen P. Naaber H. Aubry-Damon
P. Urbaskova D. Monnet P. Courvalin
A. De Benoist
Finland Germany Greece Hungary
P. Huovinen T. Breuer N. Legakis M. Fzi
O. Lyytikinen U. Buchholz J. Papaparaskevas M. Konkoly-Thege
T. Mttnen F. Tiemann A. Vatopoulos Z. Vgh
W. Witte
Iceland Ireland Israel Italy
H. Briem S. Murchan H. Edelstein D. Boccia
K. Kristinsson O. Murphy R. Raz G. Cornaglia
S. Vilhemsson D. O'Flanagan F. D'Ancona
D. Whyte M.L. Moro
Luxembourg Malta the Netherlands Norway
O. Courteille M. Borg W. Goettsch E. Bjorlow
R. Hemmer E. Scicluna A.J. de Neeling E. Hoiby
E. Tiemersma D. Katzenelson
Poland Portugal Rumania Russia
P. Grzesiowski M. Cania C. Balotescu R. Kozlov
W. Hryniewicz P. Lavado I. Codita L. Stratchounski
M. Paixao
Slovakia Slovenia Spain Sweden
L. Langsadl M. Gubina F. Baquero O. Cars
J. Kolman J. Campos K. Ekdahl
S. Cruchaga L. Gezelius
J. Iglesias G. Kahlmeter
B. Olsson-Liljequist
UK
S. Cavendish
A. Johnson
D. Livermore
A. Noone
M.C.J. Wale
Curriculum vitae
Stef Bronzwaer was born on 28 April 1967 in Heerlen, the Netherlands. He passed his
secondary school (Lyceum) exams at the Bisschoppelijk College in Weert in 1986. He
went to the University of Amsterdam where he graduated from College of Medicine in
1992 and took his Board Exam in 1995. In 2001 he completed his Master of Public Health
degree at the Netherlands School of Public Health in Utrecht, the Netherlands.
As a medical doctor he worked shortly at the Social Medical Centre Bukas Palad in a
slum-area outside Tagaytay City, and as resident at the department of paediatrics at De
La Salle University Medical Center near Manila, the Philippines. Here he studied risk
factors for a complicated disease course in children with measles. He then moved to the
Infectious disease unit of the Istituto Superiore di Sanit in Rome, Italy, where he worked
as project manager of an EU-project making an inventory of resources and means for
controlling communicable diseases. From 1998 to 2002 he worked in the Department of
Infectious Disease Epidemiology of the National Institute for Public Health and the
Environment (RIVM), Bilthoven, the Netherlands, where he helped establish the
European Antimicrobial Resistance Surveillance System (EARSS), for which he served as
project leader. The work described in this thesis was realised within this network.
Since 2002 he works at the Communicable, rare, and emerging diseases unit of the
Directorate Public Health (DG Health and Consumer protection) at the European
Commission in Luxembourg. He holds responsibility for the proper functioning and
coherence of a number of European surveillance networks on communicable diseases
and follows the implementation of the Community strategy against antimicrobial
resistance.

Thesis

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thesis

Uploaded by

Copyright:

Available Formats

European antimicrobial resistance surveillance

as part of a Community strategy

You might also like