J Antimicrob Chemother 2010; 65, Suppl. 1, i3 – i17
doi:10.1093/jac/dkp433
Antimicrobial-resistant pathogens in animals and man: prescribing,
practices and policies
Pamela A. Hunter1, Susan Dawson2, Gary L. French3, Herman Goossens4, Peter M. Hawkey5, Ed J. Kuijper6,
Dilip Nathwani7, David J. Taylor8, Chris J. Teale9, Rod E. Warren10, Mark H. Wilcox11, Neil Woodford12,
Mireille W. Wulf13 and Laura J. V. Piddock14*
1
Burnthouse, Burnthouse Lane, Cowfold, West Sussex RH13 8DH, UK; 2Faculty of Veterinary Science, University of Liverpool, Liverpool, UK;
St Thomas’ Hospital and Kings’ College, London, UK; 4Vaccine and Infectious Disease Institute, University of Antwerp, Antwerp, Belgium;
5
Health Protection Agency, West Midlands Public Health Laboratory, Heart of England NHS Foundation Trust, Birmingham, UK;
6
Department of Medical Microbiology, Leiden University Medical Center, Leiden, The Netherlands; 7Ninewells Hospital and Medical
School, Dundee, Scotland, UK; 8Faculty of Veterinary Medicine, University of Glasgow, Glasgow, Scotland, UK; 9Veterinary Laboratories
Agency (Weybridge), New Haw, Addlestone, Surrey, UK; 10Department of Microbiology, Royal Shrewsbury Hospital, Shrewsbury, UK;
11
Department of Microbiology, The General Infirmary, Leeds, UK; 12Antibiotic Resistance Monitoring and Reference Laboratory, Health
Protection Agency Centre for Infections, London, UK; 13PAMM Laboratory for Medical Microbiology, Veldhoven, The Netherlands;
14
Antimicrobial Agents Research Group, School of Immunity and Infection, University of Birmingham, Birmingham, UK
3
*Corresponding author. Tel: þ44-121-414-6966; Fax: þ44-121-414-6819; E-mail: [email protected]
This meeting focused on infections in humans and animals due to methicillin-resistant Staphylococcus aureus
(MRSA), extended-spectrum b-lactamase (ESBL)-producing bacteria and Clostridium difficile, and their corresponding treatments. MRSA is predominantly a human pathogen, and molecular typing has revealed that
certain clones have spread widely both between humans and from humans to animals. ESBL-producing bacteria, particularly those that express the CTX-M b-lactamases, have been disseminated worldwide. Whilst
such strains are usually isolated from humans, some animal isolates also produce CTX-M enzymes. In
humans, one clone of CTX-M-producing Escherichia coli, sequence type (ST)131, has been particularly successful. C. difficile, often ribotype 027, commonly colonizes the hospital environment and causes serious infections
in humans. In animals, ribotype 078 is more often found, and is an important cause of diarrhoea in piglets.
There is a concern that the numbers of MRSA or other antimicrobial-resistant bacteria might increase further
when human isolates become established in animals, as this can amplify the numbers of such bacteria by dissemination within animal groups with subsequent spread back to humans. Certain antimicrobials have been
implicated in the selection of MRSA, ESBL-producing bacteria and predisposition to infection by C. difficile.
Guidelines for treatment and prevention of infections by MRSA, ESBL-producing bacteria and C. difficile were
discussed and evidence-based policies were recommended for both humans and animals.
Methicillin-resistant Staphylococcus aureus
Methicillin-resistant Staphylococcus aureus in man
Peter M. Hawkey
Methicillin-resistant Staphylococcus aureus (MRSA) is carried by
approximately one-third of the human population and another
third are occasional carriers. Data from the European Antibiotic
Resistance Surveillance System (EARSS) have shown that the incidence of bacteraemia caused by MRSA in hospitals in Europe has
increased in many countries between 2001 and 2006 although
the incidence in Scandinavia remains low when compared with
the UK, France, Spain and Italy.1
Strains of MRSA arose from methicillin-susceptible S. aureus
(MSSA) by the acquisition of the mecA gene which encodes an
additional penicillin-binding protein (PBP20 ). This PBP has a low
affinity for isoxazolyl penicillins, such as methicillin, cloxacillin
and oxacillin, and the cell is thus able to maintain peptidoglycan
synthesis in the presence of what would otherwise be lethal
b-lactam concentrations. The mecA gene is located on a genetically mobile chromosomal determinant termed staphylococcal
cassette chromosome mec (SCCmec).
The earliest recorded MRSA was identified in England shortly
after the introduction of methicillin.2 Hospital-acquired strains
of MRSA (HA-MRSA) increased in prevalence over the next
45 years throughout the world, particularly in the last two
decades. It was originally suggested that all the clones of
MRSA had derived from a single introduction of SCCmec into
S. aureus with subsequent evolution of the host strains of
S. aureus.3 However, there is now irrefutable evidence that
SCCmec has been introduced into a number of different lineages
of S. aureus (multiclone theory).4 The development of multilocus
sequence typing (MLST) is the technique that has given this
insight. Seven genes responsible for essential cellular processes
# The Author 2010. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: [email protected]
i3
Hunter et al.
ST30-MSSA
537kb
Chromosomal
insertion from
ST30
OtherMRSA single
locus variants of
ST250-MRSA-I
ST8-MRSA-II
ST8-MRSA-I
ST8-MSSA
1 SNP arcC
+ insertion
ST239-MRSA-III
Brazilian/Hungarian clone
1SNP
yqiL
ST250-MSSA
PVL
ST250-MRSA-I
1 SNP gmk
Archaic clone
(Jevon strain)
ST8-MRSA-IV
ST247-MRSA-I
USA 300 clone
Iberian clone
Figure 1. Molecular evolution pathway of S. aureus ST8 derivatives determined by MLST (adapted from Deurenberg et al. 20086). SNP, single nucleotide
polymorphism in a specified gene.
i4
150
Number of hospitals
are sequenced, and both synonymous and non-synonymous
changes in DNA bases are scored. Because there is no direct
selection of these genes, the accumulation of single nucleotide
polymorphisms (SNPs) is directly correlated with the evolutionary
pathway of the clone [defined as isolates with identical sequence
type (ST) and SCCmec type]. The original MRSA (‘Jevon strain’)
was derived from a successful ST8 clone of MSSA, which
caused disease worldwide, by acquisition of SCCmec from
another bacterial species together with an SNP in the yqiL
gene, which gave rise to ST250. Further acquisition of SNPs
gave rise to ST247 MRSA (the Iberian clone). The Brazilian–Hungarian clone (ST239), which arose by the recombination of
537 kb of DNA from the successful ST30 MSSA into ST8 MRSA,
is widely distributed across the world, particularly the Far East,
and was recently shown to be the dominant clone in the
Indian subcontinent.5 The evolutionary relationships of the
major MRSA clones have been reviewed6 and the evolution of
ST8 derivatives is summarized in Figure 1.
In UK hospitals, data from the former UK Public Health Laboratory Service [now the Health Protection Agency (HPA)]
showed that the incidence of the epidemic MRSA (EMRSA)
strain EMRSA-3 (ST5) consistently remained fairly low between
1993 and 1996. In contrast, the incidence of EMRSA-15 (ST22)
and EMRSA-16 (ST36) increased dramatically over this period
(Figure 2). We still have little understanding of the factors, intrinsic to both the MRSA strain and the environment (antibiotic
selection pressures, routes of transmission, etc.), that allow
some clones to be so successful.
Resistance to a number of antibacterial agents among EMRSA
is widespread, with isolates of EMRSA-15 and EMRSA-16 in the
UK being frequently resistant to ciprofloxacin and erythromycin.7
Similarly, in a survey of MRSA isolates from 112 Belgian hospitals,
among 511 isolates comprising nine clones [including some that
produced Panton– Valentine leucocidin (PVL)], 97% were resistant to ciprofloxacin, 60% to erythromycin, 45% to tobramycin
and 5% to gentamicin.8 All were susceptible to tigecycline, daptomycin, ceftobiprole and linezolid, suggesting that, as yet,
resistance to these novel agents is not common.
It was thought that when colonized patients are discharged,
MRSA might transmit from the hospital setting to the community. However, data from several countries, particularly the
USA, have shown the independent emergence of strains of
EMRSA-15
EMRSA-16
125
100
75
50
25
EMRSA-3
0
1993
1994
1995
Year
1996
Figure 2. The occurrence of EMRSA-3, -15 and -16 among isolates
submitted to the national Reference Laboratory by hospitals in England
and Wales, 1993 –1997.
MRSA in the community. There are differences between these
community-associated (CA-MRSA) and HA-MRSA strains and
the infections they tend to cause. In particular, the age group
affected is different, with CA-MRSA commonly (but not exclusively) infecting children with no underlying classical risk
factors. CA-MRSA predominantly cause skin and soft tissue infections, which are often clinically severe, although CA-MRSA are
still susceptible to most classes of antibacterials other than
b-lactams. CA-MRSA strains are genetically unrelated to
HA-MRSA and have diverse lineages as they are frequently
enriched with genes encoding SCCmec IV, PVL and other exotoxins. In Europe, the predominant strains are ST80, ST8 and ST30,
while in the USA they are ST8, ST5 and ST59.
Although the vanA gene, which confers high-level transferable
vancomycin resistance, has been acquired by S. aureus from
Enterococcus spp., this is still a rare occurrence (,10 documented occasions) but is a worrying prospect, as infection with
vancomycin-resistant strains has a poor outcome when treated
with vancomycin.
There are a number of important strategies for controlling the
spread of MRSA in hospitals.9 The early identification of carriers is
essential, and has recently been shown to reduce transmission
when combined with decolonization10 followed by the isolation
or cohorting/labelling of patients. However, a major problem in
many hospitals is a lack of available side-rooms to enable the
isolation of patients. Transmission can also be interrupted by
JAC
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
MRSA in animals
2000
Cases of MRSA bacteraemia
hand washing and environmental cleaning. Appropriate treatment needs to be initiated promptly and vancomycin is often
the drug of choice. The use of certain antibacterials to which
MRSA are commonly resistant, such as third-generation cephalosporins and quinolones, should be restricted. Colonized patients
can be decolonized by the use of triclosan baths and mupirocin
nasal cream. Surveillance is essential, optimally with molecular
typing, and recent data from the HPA have shown encouraging
results, with a gradual decline in cases of bacteraemia due to
MRSA in the UK (Figure 3).
Future challenges include whether the reduction of HA-MRSA
can be maintained and whether the occurrence of CA-MRSA in
the UK increases. Unknowns include how common and how
important is low-level vancomycin resistance, will high-level vancomycin resistance occur and, if so, will it have the ability to
spread?
1000
0
Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1
2006
2007
2008
2009
Figure 3. Number of cases of MRSA bacteraemia in England reported
under the quarterly mandatory reporting scheme. (Data from
Health
Protection
Agency,
http://www.hpa.org.uk/web/
HPAweb&HPAwebStandard/HPAweb_C/1233906819629
date
last
accessed 31 July 2009.)
Mireille W. Wulf
There are differences in the occurrence of MRSA between companion animals (pets and horses) and livestock (mostly pigs,
poultry, cattle and sheep). Up to 2006, MRSA was rare in pets
but is now increasing. Carriage in healthy dogs is still rare but it
can be found in 2% of those with skin conditions. There are
occasional outbreaks seen in some animal hospitals, including
those treating horses, with the majority of these infections
appearing to be caused by the same strains that are seen in
humans.
MRSA infections are more common in livestock but the strains
are often, but not always, different from those found in humans.
The occurrence in dairy cattle is still low in various countries and
is usually associated with mastitis. There is a limited overlap with
humans, and transmission to humans is rare. Dairy cows have
been treated with penicillins for decades, predominantly for
mastitis, but the incidence of resistance nonetheless remains
low.
The situation with regard to MRSA carriage in veal calves is
different from that in adult milking cows, with a far higher incidence of MRSA. A study carried out on 102 farms in the Netherlands found that 28% of calves carried MRSA and 88% of the
farms sampled had calves with MRSA. The farmers and their
family members were also sampled, and 33% of the farmers
carried MRSA but only 8% of family members. The isolates
from both animals and humans belonged to the clonal
complex ST398.11,12
A high incidence of MRSA is found in pigs in the Netherlands,
Denmark, Belgium and Canada (up to 40%) and in their handlers.
In a slaughterhouse study in the Netherlands 80% of herds and
39% of the pigs carried strain ST398 MRSA.13 The isolates were
resistant to doxycycline but were not PVL producers. Dust
samples from 50% of the farms contained the same strain
and 30% of the farm workers were carriers. There was an association with use of antibiotics. In contrast, the incidence of MRSA in
chickens appears to be lower, with a study in Belgium finding
only 5/39 chickens from broiler farms carrying MRSA and 10/98
S. aureus being MRSA ST398.14 ST398 has also been isolated
from manure and broiler farms in the Netherlands.15
The most common MRSA isolates from animals are ST398, the
main reservoirs being pigs and veal calves. This type, which is
also isolated from chickens and horses, can be transferred to
humans. Most isolates are multidrug resistant, and some
PVL-positive isolates are found. MRSA is rarely found in meat
and then only in low quantities; the source is thought to be
the butcher/meat handler rather than animals.
The routine use of antibiotics, which disrupts the normal
bowel flora in piglets, could be a selective pressure and thus a
contributory factor in the occurrence of ST398 in this species.
It is possible that a biovar acquired the mecA gene and this
thrives in animals. There is some evidence that organisms can
be disseminated from contaminated dust, as Gibbs et al. 16
detected resistant S. aureus inside and downwind of partially
open swine confinement facilities, at levels of 104 cfu/m3. This
would be regarded as a potential human health hazard and
the recommendation is that swine facilities should be placed
at least 200 m from residential areas to avoid detrimental
effects on human health.17
Guidance for the treatment of MRSA in man
Dilip Nathwani
The optimal antibacterial treatment of a range of MRSA infections
continues to be the subject of numerous reviews and guidelines,
some of which address specific sites of infection.18 – 20 The key
challenges when tackling this particular disease area and pathogen are the complex array of diseases caused by MRSA, the variation in epidemiology and disease burden, the emergence of
different strains (e.g. PVL-producing CA-MRSA) and the difficulty
in establishing a rapid microbiological diagnosis combined with
relatively blunt clinical instruments for predicting the likely causative pathogen to guide empirical therapy. Therefore, the
decision as to whether to treat, how and with what agent, is
complex, and there are a number of factors that must be considered (Figure 4).
i5
Hunter et al.
• Is an antibacterial needed? Is it infection or colonization?
• Can simple surgical debridement be used? Is topical decolonization required?
• What is the site and severity of infection and is there a positive culture?
• If there is no positive culture, can the likelihood of MRSA be predicted?
When are rapid diagnostic tests likely to be available and will they be
helpful?
• Should a differentiation be made between community- or hospital-acquired
MRSA and how likely is the presence of a PVL strain?
• What is most effective treatment for MRSA? Is oral or parenteral therapy
required?
• Is the effectiveness of vancomycin declining and if so should the alternative
new therapies be used and in which clinical settings?
• How should we measure the impact of treatment for a variety of clinical,
microbiological, economic, patient-related outcomes and ecological impact
(for example resistance and C. difficile-associated diarrhoea)?
Figure 4. Factors to consider before treating MRSA infections.
A strong evidence base for how best to treat serious MRSA
infections in humans is generally lacking since different methodological approaches and grading systems have been used,
leading to conflicting conclusions. The optimal therapy for
uncomplicated or non-serious infections is usually considered
to be oral therapy with a range of ‘old’ well-established antibiotics. Some are used alone or in combination with rifampicin.
On the other hand, serious infections have been managed traditionally by glycopeptides such as parenteral vancomycin,
which has been a well accepted standard of care for some
time. However, whether we can still confidently rely on this
agent for treating severe MRSA infections such as
healthcare-associated pneumonia (particularly ventilatorassociated pneumonia) or complicated bacteraemia has been
subject to intense study and debate over the last 10 years. For
example, a consensus expert panel in 2008 tackled the question
of whether vancomycin was obsolete in the treatment of MRSA
infection and systematically reviewed the evidence to support
this.21 Among a panel of experts who reviewed the evidence
(n ¼ 11), 72% tended to agree (albeit with reservations) with
the statement whereas among a group of 744 members of the
Infectious Disease Society of America, 83% disagreed with this
statement either totally or with some reservations. This clearly
reflects a significant gap between how evidence is perceived by
a group of (primarily US) experts and what ‘non-expert’ clinicians
perceive and do in their (predominantly North American)
practice.
UK Guidelines published in 200622 suggested that treatment
with flucloxacillin should remain the first-line empiric treatment
if the incidence of MRSA locally is low. However, depending on
the severity of the infection and if MRSA is suspected based on
local epidemiology and risk factors, then the use of an agent
active against MRSA (vancomycin or linezolid) at the onset of
infection was recommended. No distinction between the two
agents was made because of a lack of unequivocal evidence of
superiority of linezolid over vancomycin. In 2008 these guidelines
were updated for a range of infections due to MRSA, including
bone and joint, endovascular, respiratory, urinary tract, and
skin and soft tissue.23 In spite of this, 3 years on it is still
i6
• Balance between desirable and undesirable effects
• Quality of evidence
• Values and preferences
• Cost (resource allocation)
Figure 5. Evidence-based recommendations for therapy or intervention.
acknowledged that there is a poor evidence base for treating
many infections, especially those involving bones and joints,
and that the evidence for clear and unequivocal superior clinical
effectiveness of newer agents such as linezolid, daptomycin and
tigecycline was absent. Nevertheless, recommendations were
made, primarily based on opinion and experience or evidence
of equivalence, as to when certain of the new agents may be
appropriate. Some older agents are also recommended but,
compared with the 2006 guidance, there is a move away from
the use of combination therapy.
In the absence of new clinical trial data showing clear evidence of clinical superiority, perhaps a more global assessment
of the true value and consequences of new and traditional therapies is required. One such approach, increasingly recognized by
health technology appraisal groups, guideline groups and
quality improvement organizations, is termed GRADE.24 This
method of providing evidence-based recommendations for clinical practice uses four domains related to the therapy or intervention (Figure 5).
For the treatment of MRSA infections, future guidelines could
perhaps consider using this broad risk –benefit approach to
decision making and guidance. This process may also inform
the development of high-level evidence-based standards of
care. Finally, on a positive note, whilst MRSA infections in
humans continue to represent a significant burden, a combination of good infection control, antimicrobial stewardship and
appropriate use of a range of established and new agents,
should give us good reason to be optimistic about reducing its
public health threat as we head towards the end of the first
decade of the new millennium.
JAC
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
Points raised in discussion
† One difficulty in controlling MRSA spread in UK hospitals is that
hospital bed occupancy is frequently high, thereby necessitating the strategy of isolating carriers and infected cases
† Transfer of MRSA from animals to man is currently not a major
problem in the UK, although the presence of ST398 strains in
pigs and calves is a potential threat, especially since such
strains are frequently multiresistant These strains do not
seem to transfer between humans as readily as many epidemic strains, but they may nevertheless prove to be a reservoir of CA-MRSA
† Although infections and carriage in pets is low, the strains are
often the same as those commonly found in humans, namely
EMRSA-15 and EMRSA-16
† The knowledge base for the efficacy of treating MRSA infection
is weak and stresses the need for larger clinical trials and
better definition of key risk factors. (The point was also
made that better support systems were needed and that
most surveillance studies were inadequate.)
Extended-spectrum b-lactamases (ESBLs)
Extended-spectrum b-lactamases in man
Neil Woodford
Extended-spectrum b-lactamases (ESBLs) are enzymes produced
by Gram-negative bacteria that inactivate oxyimino cephalosporins, conferring resistance to them. Bacteria producing ESBLs are
resistant to cephalosporins such as cefuroxime, cefotaxime, ceftazidime and ceftriaxone, and, through genetically linked resistance
mechanisms, they are often resistant to other antibacterials
including quinolones and aminoglycosides. Most ESBL-producing
organisms are thus multidrug resistant and many are only susceptible to carbapenems, along with little-used agents such as
fosfomycin.
Since the start of the 21st century there has been an explosive
worldwide increase in the prevalence of organisms producing
ESBLs. Currently, in England, Wales and Northern Ireland,
20000 cases of bacteraemia caused by Escherichia coli are
reported per annum, of which 12% were resistant to cefotaxime
and/or ceftazidime in 2007 (Figure 6). Of these ESBL-producing
isolates, most are from urinary tract infections (UTIs). To a large
extent, this reflects a massive clonal spread of E. coli producing
CTX-M-15 ESBL in the UK; of 1500 isolates of E. coli producing
CTX-M ESBLs examined in the HPA’s Antibiotic Resistance Monitoring and Reference Laboratory, 91% contained alleles encoding
group 1 enzymes (mainly CTX-M-15), with many isolates belonging to the international O25:H4-ST131 clone.25 The remaining isolates were mostly diverse, 8.5% producing Group 9 CTX-M
enzymes, with fewer producing Group 2 enzymes and only one
isolate producing a Group 8 enzyme. The O25:H4-ST131 clone producing CTX-M-15 ESBL has spread across many countries, including France, Canada, Korea, Lebanon, Portugal, Spain, Switzerland
and Turkey as well as the UK. Isolates of this clone have diverse
PFGE profiles, but share .60% banding pattern similarity; they
also have variable virulence factors.
The genes that encode CTX-M ESBLs escaped from the
chromosomes of Kluyvera spp.26 However, even highly related
CTX-M ESBLs may have distinct origins. For example, the genes
encoding CTX-M-3 and CTX-M-15 differ by a single nucleotide,
but are often found on distinct plasmid types with different
flanking sequences. These clearly represent separate acquisitions
from Kluyvera spp., rather than diversification by mutation after
acquisition by E. coli.
Prolonged faecal carriage of ST131 clonal variants has been
noted and provides the potential for persistence in community
settings. Household members often share E. coli clones, which
can persist. In a 3 year study of one household, 14 clones of
E. coli were isolated, six of which were shared by more than
one person. Four of the clones persisted in the household for
1 –3 years.27 It is not clear whether intestinal clonal complexity
is a relatively stable trait or if it varies over time, perhaps influenced by age, sex or diet. It is also not clear whether poor intestinal clonal diversity (and therefore faecal abundance of
particular E. coli clones) might predict the risk of UTI.
16 000
No. of isolates tested
% resistant
12 000
10 000
10
8000
6000
5
4000
Percentage resistant
Number of isolates tested
14 000
15
2000
0
0
94 95 96 97 98 99 00 01 02 03 04 05 06 07
19 19 19 19 19 19 20 20 20 20 20 20 20 20
Figure 6. Ceftazidime resistance in E. coli blood isolates in England, Wales and Northern Ireland, 1994– 2007. (Data from Health Protection Agency.
Antimicrobial Resistance and Prescribing in England, Wales and Northern Ireland, 2008; http://www.hpa.org.uk/web/HPAwebFile/HPAweb_C/
1216798080469 31 July 2009, date last accessed.)
i7
Hunter et al.
E. coli producing CTX-M ESBLs have been isolated from raw
chicken meat in the UK, following the suspicion that spread
could be via the food chain. However, none of the isolates
from chicken produced CTX-M-15 enzyme, which is the dominant
ESBL type in human infections in the UK.28 It is possible that
more foods should be studied, since faecal carriage of
antibacterial-resistant E. coli has been found in the USA in both
meat eaters and vegetarians;29 thus avoidance of meat does
not seem to be protective. Foreign travel and household exposure
were claimed to be important risk factors, but there are insufficient data to indicate the importance of foreign travel as a risk
factor in the UK.
The dominance of clone ST131 is not understood, but its
spread is global, implying that some variants of this clone may
possess biological attributes that provide an advantage; these
could be an intrinsic property of the host clone or reflect acquired
attributes perhaps relating to the resistance plasmid(s). As carbapenems become the standard intravenous therapy for infections caused by E. coli and Klebsiella producing ESBLs and/or
AmpC, they will drive the development of carbapenem resistance, whether by porin loss among ESBL producers or via the
de novo spread of true carbapenemases. To date though, carbapenemases remain rare in Enterobacteriaceae in the UK but are
increasing.30,31
ESBLs in animals
Chris J. Teale
Previous studies have shown that the emergence of antibacterial
resistance in animals can follow a general pattern, illustrated by
the use of the aminoglycosidic growth promoter nourseothricin,
which was used in farm animals in the former East Germany in
the 1980s. No equivalent antimicrobials were used in humans
over the same period. Resistance to nourseothricin emerged and
was mediated by plasmids containing a transposon which
encoded the enzyme streptothricin acetyltransferase.32,33 Initially,
resistance was detected in E. coli from pigs in the second year after
introduction of nourseothricin, but subsequently was also
detected in E. coli isolated from pig farmers. During subsequent
years nourseothricin resistance was detected in E. coli isolated
from humans in the wider community and then finally in Salmonella and Shigella isolates from humans. These observations
strongly support the premise that resistance genes present in
the commensal flora of animals can spread to bacteria which
can colonize or infect humans.
E. coli carrying an ESBL were first detected in livestock in the
UK in diarrhoeic calves on a Welsh dairy farm in 2004. The
ESBL present was CTX-M-14 and this was located on a highly promiscuous IncK plasmid which could be detected in a large
number of different strains of E. coli. Some of these strains persisted between repeat visits to the farm and were also found
in adult cows.34
The second case of ESBL-positive E. coli in animals in the UK
also involved calves, though in this case the ESBL involved was
CTX-M-15.35 The gene encoding CTX-M-15 was co-located on
the same plasmid as the resistance genes for OXA-1 and
AAC60 -Ib-cr. A plasmid bearing the same three resistance genes
with the same AAC60 -Ib-cr mutation had been previously
described in the literature from human sources.36 The most
i8
# Crown copyright 2010
plausible explanation for these findings is that the plasmid has
been transmitted to cattle either directly or indirectly from a
human source, because it would seem highly unlikely for such a
similar plasmid to evolve independently in the gut flora of cattle.
These findings reinforced the hypothesis that ESBL-producing
E. coli can pass into the environment from humans and that
animals both on farms and in the wild may subsequently be
exposed to them. Recent work from Portugal has shown that
such passage into the environment does occur.37 Epidemiological investigations on farms on which ESBL-producing E. coli
have been detected have shown that cattle on some farms
have a history of possible exposure to potential human sources
of ESBL-positive E. coli including water courses, leaking drains
or pipes, recently hospitalized farm staff and sewage sludge.
Flooding events may also provide a route for the wider environmental dissemination of bacteria of human origin.
Secondary spread can occur within the animal population
after initial introduction or emergence and is likely to be influenced by factors such as faecal –oral re-cycling and animal
movements. The amplification which might occur as a result of
secondary spread could surpass and swamp the trickle of an
initial primary introduction. If elements carrying genes encoding
ESBLs become established in bacteria in animals and amplification occurs on a wide scale, the consequence will be an
increased occurrence in the food chain.
A wide range of ESBLs has been detected in bacteria isolated
from veterinary samples across Europe, but reports of clinical
infections due to ESBL-producing strains in animals have so far
been very limited, with only low numbers of animals affected.
Examination of bovine veterinary diagnostic samples tested at
the Veterinary Laboratories Agency in England and Wales
between 2006 and 2007 showed that the predominant
enzymes produced by veterinary isolates were CTX-M-14 and
CTX-M-15 (Table 1; C. Teale, unpublished results). Knowledge of
the epidemiological aspects relating to the spread of ESBLproducing E. coli in animals may facilitate the development of
possible interventions to control the further spread of resistance
in the animal population and prevent entry into the food chain.
If the hypothesis regarding the importance of initial environmental transfer to the animal population is correct, then the prediction may be made that the next antibacterial resistance
appearing in human strains of E. coli is likely to appear in
animal strains of E. coli a short interval after its first widespread
appearance in humans.
Guidance for the treatment of ESBL producers in humans
Gary L. French
Previously, ESBL producers were usually opportunistic Gramnegative bacteria such as Klebsiella pneumoniae that produced
infections and outbreaks in hospitalized patients. Recently,
however, ESBL-producing E. coli have appeared worldwide,
often producing a CTX-M-type ESBL and causing colonizations
and infections in non-hospitalized people, including those in
care homes for the elderly. In some geographical areas one or
two ESBL-producing E. coli clones have predominated, while in
others (such as London) organisms are multiclonal and the
resistance determinants have appeared in other species in
addition to E. coli. The risk factors and epidemiological features
JAC
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
Table 1. CTX-M ESBLs recovered from bovine veterinary clinical diagnostic submissions in England and Wales
Quarter
Confirmed ESBL
CTX-M-1
CTX-M-3
CTX-M-14
CTX-M-15
CTX-M-20
CTX-M-32
Total examined (% ESBLs)b
3/2006
4/2006
1/2007
2/2007
3/2007
4/2007
1/2008
8
1
–
3
4
–
–
272 (3)
6
–
–
2
3
–
–
621 (1)
7
–
–
3a
4
–
–
802 (0.9)
5
–
1
–
4
–
–
514 (1)
4
1
–
1a
1
1
–
355 (1)
9
2
–
2 þ1a
3
–
1
463 (2)
13
1
–
4 þ3a
5
–
–
NA
a
CTX-M-14 variant.
Total E. coli examined from bovine diagnostic samples.
NA, not yet available.
b
Table 2. Characteristics of infections caused by ESBL-producing bacteria38
Older (Klebsiella spp., etc.)
Virulence/place
Type of ESBL
Infection
Resistances
Molecular
epidemiology
Risk factors
Newer (Escherichia coli)
less virulent/HCAI
SHV and TEM types
UTI, respiratory, intra-abdominal, blood
all b-lactams: usually quinolones and co-trimoxazole: usually
aminoglycosides
most often clonally related
longer hospital stay; severity of illness; time in ICU; ventilation;
urinary or vascular catheterization; previous exposure to
antibiotics (especially cephalosporins)
more virulent/community and HCAI
CTX-M (especially CTX-M-15)
usually UTIs, but also blood, gastroenteritis
all b-lactams: usually quinolones and co-trimoxazole: usually
aminoglycosides
usually not clonally related, but clonal outbreaks described
worldwide, including UK
repeat UTIs/underlying renal pathology; previous antibiotics
(including cephalosporins and quinolones; previous
hospitalization; nursing home residents; the elderly;
diabetes; liver pathology
HCAI, healthcare-associated infection.
of the older and newer infections caused by ESBL-producing coliforms are shown in Table 2.38
Affected patients may present from the community with
unexpectedly multiresistant urinary or abdominal infection
and may develop bacteraemias and/or be the source for
hospital cross-infection after admission. This type of
community-associated infection is more difficult to identify and
control than the more familiar healthcare- or hospital-associated
infection. The size of the problem is illustrated by the dramatic
rise in blood isolates of E. coli resistant to ceftazidime between
1994 and 2007 (Figure 6).
The key factors in controlling ESBL-producing coliforms are to
have an antimicrobial stewardship system and an infection prevention and control system in place, both in hospital and in the
community. A surveillance system is also needed to identify
cases rapidly as they occur, and for this to be effective it is essential that the local microbiology laboratory can identify ESBLs.
CTX-M enzymes may not be detected by standard laboratory
tests using ceftazidime; the BSAC and HPA recommend that laboratories should routinely test isolates against both cefotaxime
and ceftazidime, or against cefpodoxime, and then perform an
ESBL confirmatory test on any resistant isolates.39
Antibacterial treatment policies need to take into account the
fact that plasmids carrying ESBL genes often encode resistances
to other antimicrobial agents including aminoglycosides,
trimethoprim, sulphonamides, tetracyclines and chloramphenicol.
Frequently, the host bacterium has mutated to fluoroquinolone
resistance. A reduction in use of cephalosporins and quinolones
can lead to a reduction in the risk of ESBL-producing coliforms. Conveniently, policies to limit the use of cephalosporins and quinolones
are also commonly used to help control Clostridium difficile infection.
Risk factors for sepsis need to be controlled. This includes
removing urinary/vascular catheters if possible and stopping
unnecessary antibiotics. Asymptomatic patients, with colonization only, do not need antibiotics and those with asymptomatic
UTI can await the results from susceptibility testing of cultures.
Septic patients, however, need to be treated urgently and
aggressively, frequently before susceptibility results are known.
In these cases initial empirical treatment should be based on
recent local antimicrobial susceptibilities.
In areas where ESBL-producing coliforms are diverse and multiclonal, susceptibilities of different strains may be highly variable
(Table 3)40 and only the carbapenems are (at present) reliably
effective. Therefore, for serious sepsis, immediate treatment
with carbapenems is advisable, which can be ‘stepped down’
or ‘de-escalated’ to simpler therapy if subsequent test results
permit. Some ESBL producers are susceptible to some aminoglycosides and these are possible initial agents if local results
i9
Hunter et al.
Table 3. Prevalence and mechanisms of cephalosporin resistance in Enterobacteriaceae in London and South-East England40
Percentage resistant
Organism and number of isolates examined
FOX
CIP
GEN
NIT (UTI)
TMP (UTI)
MEC (UTI)
IPM
MEM
ETP
CTX-M producers
Enterobacteriaceae (502)
E. coli (292)
K. pneumoniae (190)
40
35
42
89
91
91
60
47
80
52
25
96
89
87
91
29
7.5
62
0
0
0
0
0
0
0.4
0.3
0.5
ESBL producers (non-CTX-M)
Enterobacteriaceae (149)
E. coli (89)
K. pneumoniae (25)
49
30
60
68
75
70
60
57
44
38
14
85
85
85
75
49
37
75
0
0
0
0
0
0
0.7
0
0
FOX, cefoxitin; CIP, ciprofloxacin; GEN, gentamicin; NIT, nitrofurantoin; TMP, trimethoprim; MEC, mecillinam; IPM, imipenem; MEM, meropenem; ETP,
ertapenem.
indicate likely susceptibility. Tigecycline, a new broad-spectrum
intravenous agent, is another option, but it is not excreted in
urine and is therefore not suitable for the treatment of UTI.
Mecillinam and colistin have also been used to treat ESBLproducing coliforms but susceptibilities vary and there is
limited information on their clinical effectiveness.
Points raised in discussion
† It was noted that selective pressure by antimicrobials to maintain resistance plasmids and/or resistant strains was an important factor, although it was noted that a number of
ESBL-producing strains can be maintained in both humans
and animals without any identifiable selective pressure
† The survival of some strains of E. coli in the environment has
been investigated; there are indications that they are quite
robust and can survive various composting procedures
† The risk of resistance determinants transferring to other
related species with possible amplification was noted
† Studies from China and India, where communities lived under
poor hygiene conditions, have shown the widespread presence
of resistant strains in the environment, including rivers
† The effect of heat treatment on effluents and whether plasmids survive when the host bacterium is killed are areas for
further research
† Infections caused by ESBL-producing organisms are often
multidrug resistant, often requiring prompt treatment of
septic patients before susceptibility results are available;
general guidelines will be of limited value since initial treatment should be based on recent local susceptibility data.
Clostridium difficile
Clostridium difficile in humans
Mark H. Wilcox
C. difficile is an environmental organism, and surveys have shown
that it is widespread in rivers (88%), lakes (47%) and soil (21%).
It is also found in the hospital environment (20%).41 A survey of
a hospital elderly care ward has shown that spores can be found
in toilets, sluice floors and commodes,42 and even when cleaning
procedures are thorough they can be difficult to eradicate,
i10
persisting in the hospital environment and maintaining a reservoir of potential infection.
The number of death certificates mentioning C. difficile either as
the underlying cause or as involved in the final disease has increased
markedly from 1000 in 2001 to 8234 in 2007 in England and
Wales (Figure 7). The rate of increase has, however, slowed, from
71% between 2005 and 2006 to 28% during 2006–2007. C. difficile
infection (CDI) is still predominantly a disease of the elderly, with
3400 deaths/million population aged .85 years and only 7
deaths/million population in those aged ,45 years.
Ribotype 027 is now common and appears more virulent than
other strains of C. difficile. It is now believed that the probable
reasons for this virulence are not, as had been hypothesized, the
production of two toxins or a peak in toxin production or resistance
to fluoroquinolones. The most likely reasons are prolonged production of toxin, the duration of germination, and increased sporulation which increases the risk of transmission.43 – 45
A recent survey of cases of CDI in the community revealed that
both prior treatment with antibacterials and hospitalization were
important predisposing factors.46,47 For example, exposure to antibacterials in the previous 4 weeks, particularly with multiple
agents (P ,0.001), aminopenicillins (P ,0.05) and oral cephalosporins (P ,0.05), was significantly more frequent among cases
than controls. Hospitalization in the preceding 6 months was
also a significant risk factor (45% compared with 23%;
P¼0.022). However, almost half the cases had not received antibacterial therapy in the month before infection was detected and
approximately one-third had neither exposure to antibacterials
nor recent hospitalization. This highlights current gaps in our
knowledge about the aetiology of community-associated CDI.
There is often a poor response to the treatment of severe
infections. Metronidazole has commonly been the favoured
treatment but reduced susceptibility is now seen48 and there is
an increasing dependence on vancomycin, which is more effective and gives a superior microbiological response. Nonetheless,
recurrence occurs in 20% –25% of cases. There is a poor evidence base for the effectiveness of probiotics.
The techniques used to diagnose CDI in the laboratory need
to be reassessed given concerns about the accuracy of current
methods.49 Recent advice from the Department of Health and
the HPA in England recommends against use of single tests for
the detection of C. difficile toxin.50
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
JAC
10 000
Number of deaths
8000
Mention
Underlying cause
6000
4000
2000
0
2001
2002
2003
2004
2005
2006
2007
Figure 7. Number of death certificates mentioning C. difficile and recording C. difficile as the underlying cause of death, England and Wales, 2001 –
2007. (Data from Office for National Statistics; http://www.statistics.gov.uk/cci/nugget.asp?id=1735 31 July 2009, date last accessed.)
C. difficile infections in animals
Ed J. Kuijper
C. difficile is a ubiquitous bacterium in the environment and has
been found in calves, ostriches, chickens, elephants, dogs, horses
and pigs,51 but its role in infection and its pathogenesis in
animals are largely unrecognized and possibly underestimated.
Similarly, any association between antibiotic usage and C. difficile
colonization or diarrhoea in animals is less well documented
than in humans.
In the past decade, two strains of C. difficile (PCR ribotypes
027 and 078) have emerged in humans. These two strains
have been reported from several European countries and the
USA,52,53 although the emergence of type 078 occurred relatively
recently. In November 2008, a pan-European hospital-based
study encompassing 106 laboratories revealed PCR ribotypes
014, 001 and 078 as the predominant types, whereas type
078 was found only sporadically in 2005.54 From February
2005 to February 2008, the incidence of type 078 among isolates
obtained from 1687 patients increased from 3% to 13% in the
Netherlands (data from the National Reference Laboratory).
Compared with patients infected with type 027, patients infected
with type 078 were younger and more frequently had
community-associated disease.55
Interestingly, type 078 has also been reported as a frequently
found type in piglets. Recently, two herds of piglets suffering
from diarrhoea in the Netherlands proved to be infected with
C. difficile toxinotype V, PCR ribotype 078.56 Thirty-one isolates
of C. difficile were obtained from 48 animals. C. difficile was not
detected in the mother sows or in 272 healthy, weaned piglets
on seven farms. None of the farm-workers and none of the
family members of the farm-owners had CDI. C. difficile type
078 isolates from humans and pigs were highly genetically
related by multilocus variable number tandem repeat analysis,
suggesting a common origin.
It is likely that type 078 is an important pathogen of piglet
diarrhoea worldwide. A study of the prevalence of C. difficile in
pigs in Spain sampled 780 animals from 13 pig farms.
Newborn piglets (541 samples) and 1 –2 month piglets were
sampled, and 25.9% of the newborns were positive irrespective
of the presence of diarrhoea. In contrast none of the 239
samples from the older piglets was positive.57 Preliminary data
suggest that all strains belong to PCR ribotype 078.
Other C. difficile PCR ribotypes are also frequently found in
animals.51 In a Canadian study of calves from 102 dairy farms
in Canada, faecal samples collected from 144 calves with diarrhoea and 134 control calves were cultured for C. difficile and
tested using an ELISA for C. difficile toxins A and B. Toxins were
detected in 39.6% of calves with diarrhoea but also in 20.9%
of healthy controls. PCR ribotyping showed eight distinct patterns, seven of which have been identified in humans; two of
these (PCR types 017 and 027) have been associated with outbreaks of severe disease.58 A study of the presence of C. difficile
in various animals in the UK sampled neonatal pigs, male dairy
calves (,2 months of age), horses and dogs. The 232 isolates
obtained comprised 19 ribotypes, with the predominant one
being 078 in calves and pigs.59
Spores of C. difficile can also be detected in meat products.60,61
Following the emergence of type 078, a pilot study was carried out
in the Netherlands by the Food and Consumer Product Safety Authority to determine the occurrence of C. difficile in 500 consumable
meat samples from calves, pigs, sheep, turkey and chicken. The
only positive samples were four from chicken—types 001, 003
(2) and 087. These results are clearly different from the US and
Canadian experiences and need further confirmation.
Overall there is variability in the occurrence of C. difficile in
different species. There is a low prevalence of human types but
they have been identified in both calves and pigs. There is still
much to be learned about the role of toxins and the role of
animals as a possible source for human infection via direct or
indirect contact, the environment or the food chain.
Guidance for the treatment of C. difficile in humans
Rod E. Warren
Antibacterial guidelines for treating many infections have been
published, but sometimes (for example with community-acquired
i11
Hunter et al.
pneumonia guidelines) they can be criticized because the empirical regimens are based on rarities and are too broad spectrum.
They are often difficult to apply or are applied incorrectly. Although
earlier guidelines on C. difficile and antibacterials are still applicable,62 it is difficult to assess the importance of antibacterial
control as opposed to control of spread in reducing the incidence
of infection. Some of the complicating factors include the fact that
the ward case mix can affect both usage and denominators.
Wards are commonly based on specialties, but frequently with
high bed occupancy the overflow of elderly patients from
medical wards into surgical areas can also introduce C. difficile.
There are quantitative variations in exposure and predisposition
to colonization or infection by C. difficile, the most common
being the length of cumulative stay of some elderly patients. Similarly, sequential and repeated prescribing make it difficult to
assess the effectiveness of policy changes. Data are needed on
managing C. difficile in large units, over long time periods, and in
many patients.
There is a lack of knowledge regarding exactly how antibacterials select for C. difficile. With regard to resistance to gut
colonization by C. difficile, it is likely that the anaerobic flora play
a key role, as they do with aerobes, but all that is certain is that
a complex flora is responsible.63 – 65 The gut flora is affected by
the antibacterial drug concentration in the gut lumen or at the
luminal surface, whereas the systemic concentration may be
less significant for C. difficile.66,67 A ratio of MICs for C. difficile
and Bacillus fragilis of .1 suggests a propensity for an agent to
cause CDI, although allowances must be made for the poor and
unreliable penetration of current b-lactamase inhibitors into the
gut lumen and the apparent protective effect of free gut
b-lactamases for b-lactam-provoked C. difficile.68,69
Over the years substantial changes in antibacterial use have
affected the incidence of CDI, e.g. metronidazole being substituted
for clindamycin.70,71 It is generally agreed that one should minimize
the use of third-generation cephalosporins and cephamycins,62,72 – 75 and this advice has been extended to include secondgeneration cephalosporins, although it is not clear what can be
safely substituted. Carbapenems appear to have a propensity to
induce CDI.71,76
There are uncertainties as to whether any classes of antibacterial agents produce a low incidence of C. difficile colonization.
For example, the use of quinolones used to be regarded as a low
risk for provocation of CDI, and there used to be homogeneous
MICs of the early quinolones. After 20 years of use this view has
been shown to be incorrect and a biphasic population of more
resistant strains is evident particularly with some of the newer quinolones. Very resistant strains are more evident in some C. difficile
ribotypes.77 – 79 The chances of linezolid and tigecycline provoking
C. difficile colonization appear to be low, but remain uncertain.
Aminoglycosides are clearly agents which when used alone
seldom provoke colonization with C. difficile.
Points raised in discussion
† The diversity of techniques used for both routine isolation and
typing of C. difficile was noted, yet few are suitable for use in
routine laboratories
† The serious deficiencies in the reliability and performance of
C. difficile toxin detection kits are of particular concern for surveillance, diagnosis and management of infected persons
i12
† Difficulties remain in linking prior antibiotic therapy with CDI
where factors such as the type of antibacterial, and the
timing and duration of treatment might all have importance;
some compounds, such as lincosamide, have a prolonged
effect in experimental hamster models
† The links between animals and man, particularly the potential
for transmission of C. difficile from animals to food to the community, remain to be determined
† The lack of data may mean that C. difficile infection could be
unrecognized and therefore underestimated in many
animals, notably pigs and cattle
† The prevalence of C. difficile in many animal species appears to
be low yet the identification in pigs of ribotype 078 with a high
homology with human types is of concern and begs the question of whether these animals could be a source for human
infection either directly or indirectly
† Investigations in the Netherlands have shown a low prevalence of C. difficile in locally produced food, but little is
known about food from other parts of the world
† The true mortality in humans is uncertain since it is difficult to
distinguish between whether a patient has died ‘of’ or ‘with’
CDI
† The quantity of toxin produced by strains and its association
with the severity of infection in both humans and animals
remains unclear
† Larger studies are required both to determine the optimum
antibacterial regimen for treatment and to better define
which agents provoke CDI
† Such studies need to be of adequate duration and follow-up,
particularly in populations where CDI is common
† Better computerized support systems that link pharmacy and
laboratory data are needed; currently most surveillance
studies are inadequate and lack denominators and controls.
Additional funding is needed to support this.
Multidrug resistance and changing
antimicrobial use in man
Herman Goossens
The seemingly simple matter of measuring the use of antimicrobials in hospitals is fraught with problems as there are so many
different methods used (Figure 8), and trying to compare these
can give misleading results. Attempts to determine cause –
effect relationships following changes in antimicrobial use are
thus difficult and are affected by numerous confounding
factors. Factors affecting the emergence and spread of
multidrug-resistant organisms are complex. Their emergence
can be caused by genetic recombination or mutations and their
spread can be a consequence of selection. The latter may
involve the amplification of a resistant strain within a colonized
or infected patient receiving antibiotics. Clonal spread may then
take place from a colonized to a non-colonized patient, possibly
followed by the dissemination of resistance genes among bacterial strains via conjugation, transduction or transformation.
There are many confounding factors including infection
control practices, colonization pressure, any underlying illness,
the length of stay, the duration of treatment and also any use
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
JAC
DOT (Days of Treatment)
Percentage of patients exposed to antimicrobials
Weight (g or kg or units of treatment)/year)
Grams/1000 patient days
Vials/month
DDD (defined daily dosage) or PDD (prescribed daily dose)/1000 inhabitant-days
DDD or PDD/100 or 1000 patient-days
DDD or PDD/100 or 1000 administrative bed-days
DDD or PDD/100 or 1000 occupied bed-days
DDD or PDD/100 or 1000 admissions
DDD or PDD/100 or 1000 discharges
DDD or PDD/month/occupied bed
Figure 8. Methods used to measure antimicrobial use in hospitals.
of the drug in the community or in animals. These factors complicate any attempts to determine the impact of the use of an
antimicrobial. An example is afforded by the impact that the duration of treatment can have. A retrospective, cross-sectional
study was carried out to estimate the probability of carbapenem
resistance among Pseudomonas aeruginosa isolates from adult
inpatients with respiratory tract infection.80 The predictors of carbapenem resistance were found to be prior receipt of mechanical
ventilation for 11 days and prior exposure to fluoroquinolones
and to carbapenems for 3 days.
Use of antimicrobials in animals can be important as shown
by Johnson et al. 81 In this study of 931 geographically and temporally matched E. coli isolates from human volunteers and commercial poultry products, drug-resistant human isolates were
similar to poultry isolates but drug-susceptible human isolates
differed considerably from poultry isolates.
The relationship between antibacterial use and resistance in
hospitals is complex, with no overall agreement on which
methods are most appropriate for the measurement of drug use
in hospitals and no agreement on how to correlate their use with
selection of resistance in hospitals. There is a limited understanding
of the association of exposure to antibacterials with acquisition or
progression from colonization towards infection with multidrug
resistance.
Guidance for the use of antimicrobials and
husbandry practices in veterinary medicine
Susan Dawson and David J. Taylor
There are major differences between veterinary and human medicine. The most obvious is the wide range of animals that may
need treatment. These are usually divided into companion
animals and farm or food animals. Companion animals include
horses, cats, dogs, rabbits, reptiles, fish and various other less
common pets. After death, their bodies generally go to incineration
or burial. Care for companion animals is intimate, and human
contact extends to all age groups in the home. Antimicrobial use
is not dissimilar to that in human medicine for the treatment of bacterial infections in individuals; however, for animals payment for
treatment must be covered by the owner. Some pets may be
insured and these policies may cover more sophisticated treatment
to the limit of the policy. More elaborate operations are now available as in human medicine (hip replacement, more reconstructive
work). This may require the administration of more veterinary medicines and could potentially lead to the use of more antimicrobials.
Veterinarians usually operate in small groups—there is no
equivalent to the NHS for humans! Most vets specialize in treating either small companion animals, or horses, or farm animals.
There is increasingly more specialization, however, for example in
poultry or pigs, usually for the big producers.
Veterinary medicines are registered for use in animals by the
Veterinary Medicines Directorate (VMD) in the UK or the European
Medicines Evaluation Agency (EMEA). All antibacterials for veterinary use are prescription-only products which must be prescribed by a veterinary surgeon (POM-V). There is a ‘cascade’
system for the use of medicines in animals—veterinary surgeons
must prescribe and use veterinary medicines where available but
if no medicine is authorized they can then use a veterinary medicine authorized in the UK for another species or another condition. Failing this they may use a medicine authorized for
human use in the UK or one imported from another member
State.
The Responsible Use of Medicines in Agriculture alliance
(RUMA) was set up 10 years ago with input from various Veterinary Associations. RUMA provides guidelines for the responsible
use of antimicrobials in livestock production, with individual
guidelines for all the major food-producing species. Its major
aim is to highlight strategies that may reduce the need for antimicrobials, e.g. management conditions, reduction of stress.
There are strict conditions for the use of antimicrobials in foodproducing animals; treatment is restricted to animals on a
single holding and any medicine imported from another
member State must be authorized for food-producing species
in the other member State. For food-producing animals, withdrawal periods are set for each medicinal product whereby that
i13
Hunter et al.
animal, or products from that animal such as milk, cannot enter
the food chain for a specified period of time.
In the EU, antimicrobial growth promoters were withdrawn in
2006 although one class continues to be administered in feed,
namely the ionophores which have anti-coccidial activity. They
are categorized as Zootechnical Feed Additives and fed to
broiler chickens and layers to 12 weeks of age, and to rabbits.
Their antimicrobial activity is incidental to their primary
purpose. There are consequences to the removal of growth promoters, namely an increase in disease in some species.
The VMD reported on the annual antimicrobial usage in
2007.82 This revealed an overall decrease in sales of veterinary
antimicrobials. The reduction was accounted for by reduced
sales of tetracyclines and macrolides but an increase in sales
of fluoroquinolones and cephalosporins. In spite of an increase
in fluoroquinolone usage, their sales represented ,1% of the
total. Approximately half the total sales of therapeutic antimicrobials were tetracyclines. Eighty-seven percent of the antimicrobials were sold for use in food-producing animals. Total sales in
the veterinary field are greatly biased by the weight of the type
of animals treated—cows and horses require larger amounts of
drugs than other smaller animals.
Cephalosporins are now used quite widely, including thirdand fourth-generation compounds, but these are used mostly
for cattle, pigs, poultry and horses. A short withdrawal period
can give a market advantage to a particular product in cattle.
For small animals, it is mostly the first- and second-generation
cephalosporins that are used.
The British Small Animal Veterinary Association has issued
guidelines for the use of antimicrobials for the treatment of
MRSA. These state that antibacterial drugs should only be used
where significant bacterial disease has been confirmed or can
reasonably be suspected, and that the use of human products
under the cascade system should be restricted to cases only
where there is compelling evidence from culture and susceptibility
testing that there is no suitable veterinary product. When selecting a drug, the likely infecting pathogen and breadth of antibacterial spectrum should be considered. Prophylactic antibacterial
drugs are not indicated for routine clean surgery of ,90 min.
MRSA is uncommon in animals, and those at greatest risk of
infection are ill hospitalized animals, especially those with intravenous catheters, those undergoing surgery (especially with
implants) and immunocompromised animals. Bacterial strains
are usually of common human types and most are present on
the animals on arrival but rarely persist in the hospital. There is
the potential for veterinary staff to act as a reservoir of MRSA
infection. Most UK veterinary isolates of MRSA are usually susceptible to routine antibacterials, including potentiated sulphonamides, tetracyclines, fusidic acid and mupirocin, although these
are not all licensed for use in animals.
Vaccines for bacterial diseases such as enzootic pneumonia
and proliferative enteropathy of pigs have directly reduced the
need for some treatments. Similarly, vaccines for E. coli septicaemia in poultry can reduce antimicrobial use.
Points raised in discussion
† Animals may present with totally different spectra of infecting
organisms which are not or only very rarely found in humans.
Examples are Bordetella bronchiseptica in dogs, non-S. aureus
i14
strains in canine skin infections and bronchopneumonia in
foals caused by Rhodococcus equi. Most farm animals also
have a variety of non-human pathogens. Mastitis is frequently
caused by coagulase-negative staphylococci, Streptococcus
agalactiae, Streptococcus uberis, Streptococcus dysgalactiae
or even Actinomyces pyogenes. Pasteurella species can cause
pneumonia in cattle as can non-human species of Mycoplasma. Pigs are susceptible to Mycoplasma hyopneumoniae
which causes enzootic pneumonia
† Surveillance data on discontinued compounds by the Veterinary Laboratories Agency still continues for pathogens such as
salmonellae and E. coli; there has been no major problem with
resistance
† Guidelines for the use of antimicrobials in veterinary practice
are not as specific as those for humans; however, some felt
that most guidelines for human therapy were often
inadequate
† Veterinary students are given a good background in the use of
drugs, and formularies are widely available. New formularies
are sent out automatically free of charge as they become
available
Some questions that remain unresolved
Despite considerable advances having been made in our understanding of MRSA, ESBLs and their host strains and C. difficile, to
allow effective treatment and/or eradication, questions still
remain. These include:
† Why do certain types of bacteria dominate and then decline;
are certain organisms biologically fitter or better able to
infect particular patients; what is the role of the host?
† What is the evidence that C. difficile and ESBL-producing bacteria from food-producing animals survive and infect humans?
† How can antibiotic-resistant bacteria be prevented from transferring between humans and animals?
† Can any successful farming practices that reduce the prevalence of bacterial colonization or infection be adapted for
use in humans? Likewise, are there any successful strategies
used in human medicine that could be modified for use in
animals?
Transparency declarations
G. L. F., P. A. H., D. N., M. H. W., N. W. and R. E. W. have received honoraria
for conference attendance, advisory boards or consultancy from various
pharmaceutical companies.
Other authors: none to declare.
References
1 The European Antimicrobial Resistance Surveillance System (EARSS).
http://www.rivm.nl/earss/ (30 July 2009, date last accessed).
2 Jevons MP. ‘Celbenin’-resistant Staphylococci. BMJ 1961; 1: 124– 5.
3 Kreiswirth B, Kornblum J, Arbeit RD et al. Evidence for a clonal origin of
methicillin resistance in Staphylococcus aureus. Science 1993; 259:
227–30.
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
4 Enright MC, Robinson DA, Randle G et al. The evolutionary history of
methicillin-resistant Staphylococcus aureus (MRSA). Proc Natl Acad Sci
USA 2002; 99: 7687– 92.
5 Shabir S, Hardy KJ, Abbasi W et al. Epidemiological typing of methicillin
resistant Staphylococcus aureus isolates from India and Pakistan. In:
Abstracts of the Nineteenth European Congress of Clinical Microbiology
and Infectious Diseases, Helsinki, Finland, 2009. Abstract P532.
http://www.blackwellpublishing.com/eccmid19/abstract.asp?id=74141
(4 August 2009, date last accessed).
JAC
19 American Thoracic Society and the Infectious Diseases Society of
America. Guidelines for the management of adults with
hospital-acquired, ventilator-associated, and healthcare-associated
pneumonia. Am J Respir Crit Care Med 2005; 171: 388–416.
20 Masterton RG, Galloway A, French G et al. Guidelines for the
management of hospital-acquired pneumonia in the UK: report of the
working party on hospital-acquired pneumonia of the British Society for
Antimicrobial Chemotherapy. J Antimicrob Chemother 2008; 62: 5 –32.
6 Deurenberg RH, Stobberingh EE. The evolution of Staphylococcus
aureus. Infect Genet Evol 2008; 8: 747–63.
21 Kollef MH, Napolitano LM, Solomkin JS et al. Health care-associated
infection (HAI): a critical appraisal of the emerging threat-proceedings
of the HAI Summit. Clin Infect Dis 2008; 47 Suppl 2: S55– 99).
7 Johnson AP, Aucken HM, Cavendish S et al. Dominance of EMRSA-15
and -16 among MRSA causing nosocomial bacteraemia in the UK:
analysis of isolates from the European Antimicrobial Resistance
Surveillance System (EARSS). J Antimicrob Chemother 2001; 48: 143– 4.
22 Gemmell CG, Edwards DI, Fraise AP et al. Guidelines for the
prophylaxis and treatment of methicillin-resistant Staphylococcus
aureus (MRSA) infections in the UK. J Antimicrob Chemother 2006; 57:
589–608.
8 Denis O, Deplano A, Nonhoff C et al. In vitro activities of ceftobiprole,
tigecycline, daptomycin, and 19 other antimicrobials against
methicillin-resistant Staphylococcus aureus strains from a national
survey of Belgian hospitals. Antimicrob Agents Chemother 2006; 50:
2680– 85.
23 Gould FK, Brindle R, Chadwick PR et al. on behalf of the MRSA Working
Party of the British Society for Antimicrobial Chemotherapy. Guidelines
(2008) for the prophylaxis and treatment of methicillin-resistant
Staphylococcus aureus (MRSA) infections in the United Kingdom.
J Antimicrob Chemother 2009; 63: 849–61.
9 Calfee DP, Salgado CD, Classen D et al. Strategies to prevent
transmission of methicillin-resistant Staphylococcus aureus in acute
care hospitals. Infect Control Hosp Epidem 2008; 29 Suppl 1: S62–80.
24 Guyatt GH, Oxman AD, Kunz R et al. Rating quality of evidence
and strength of recommendations: going from evidence to
recommendations. BMJ 2008; 336: 1049 –51.
10 Hardy KJ, Price C, Szczepura A et al. Reduction in the rate of MRSA
acquisition in surgical wards by rapid screening for colonisation: a
prospective, cross-over study. Clin Microbiol Infect 2009; 15: doi:
10.1111/j.1469-0691.2009.02899.x
25 Woodford N. Successful, multiresistant bacterial clones. J Antimicrob
Chemother 2008; 61: 233– 4.
11 Graveland H, Wagenaar J, Heederik D et al. Exposure related carriage
of methicillin-resistant Staphylococcus aureus in veal calf farmers in the
Netherlands. In: Abstracts of the Nineteenth European Congress of
Clinical Microbiology and Infectious Diseases, Helsinki, Finland, 2009.
Abstract P1368. http://www.blackwellpublishing.com/eccmid19/abstract.
asp?id=74974 (4 August 2009, date last accessed).
27 Johnson JR, Clabots C, Kuskowski MA. Multiple-host sharing,
long-term persistence, and virulence of Escherichia coli clones from
human and animal household members. J Clin Microbiol 2008; 46:
4078– 82.
26 Livermore DM, Canton C, Gniadkowski M et al. CTX-M: changing the
face of ESBLs in Europe. J Antimicrob Chemother 2007; 59: 165–74.
28 Warren RE, Ensor VM, O’Neill P et al. Imported chicken meat as a
potential source of quinolone-resistant Escherichia coli producing
extended-spectrum b-lactamases in the UK. J Antimicrob Chemother
2008; 61: 504–8.
12 Graveland H, Wagenaar JA, Broekhuizen-Stins M et al.
Methicillin-resistant Staphylococcus aureus (MRSA) in veal calf farmers
and veal calves in the Netherlands. In: Abstracts of the ASM Conference
on Antimicrobial Resistance in Zoonotic Bacteria and Food-Borne
Pathogens, Copenhagen, Denmark 2008. Abstract B84. http://www.
asm.org/ASM/files/ccLibraryFiles/Filename/000000004028/Zoonotic%20
Program%20Abstract%20Book.pdf (4 August 2009, date last accessed).
29 Sannes MR, Belongia EA, Kieke B et al. Predictors of
antimicrobial-resistant Escherichia coli in the feces of vegetarians and
newly hospitalized adults in Minnesota and Wisconsin. J Infect Dis
2008; 197: 430–4.
13 van Duijkeren E, Jansen MD, Flemming SC et al. Methicillin-resistant
Staphylococcus aureus in pigs with exudative epidermitis. Emerg Infect
Dis 2007; 13: 1408 –10.
30 Woodford N, Zhang J, Warner M et al. Arrival of Klebsiella pneumoniae
producing KPC carbapenemase in the United Kingdom. J Antimicrob
Chemother 2008; 62: 1261 –4.
14 Nemati M, Hermans K, Lipinska U et al. Antimicrobial resistance of old
and recent Staphylococcus aureus isolates from poultry: first detection of
livestock-associated methicillin-resistant strain ST398. Antimicrob Agents
Chemother 2008; 52: 3817 –9.
31 Health Protection Agency. Multi-resistant hospital bacteria linked to
India and Pakistan. Health Protection Report Volume 3 No 26; 3 July
2009. http://www.hpa.org.uk/hpr/archives/2009/news2609.htm#ndm1
(29 July 2009, date last accessed).
15 Leenders A, Janssen M, Render N et al. Pig-related MRSA on a poultry
farm? Infectieziekten bulletin 2007; 18: 43– 4 (article in Dutch).
32 Witte W, Tschape H, Klare I et al. Antibiotics in animal feed. Acta Vet
Scand 2000; 93: 37–45.
16 Gibbs SG, Green CF, Tarwater PM et al. Airborne antibiotic resistant and
non-resistant bacteria and fungi recovered from two swine herd confined
animal feeding operations. J Occup Environ Hyg 2004; 1: 699– 706.
33 Hummel R, Tschäpe H, Witte W. Spread of plasmid-mediated
nourseothricin resistance due to antibiotic use in animal husbandry. J
Basic Microbiol 1986; 26: 461–6.
17 Gibbs SG, Green CF, Tarwater PM et al. Isolation of antibiotic-resistant
bacteria from the air plume downwind of a swine confined or
concentrated animal feeding operation. Environ Health Perspect 2006;
114: 1032– 7.
34 Liebana E, Batchelor M, Hopkins K et al. Longitudinal farm study of
extended-spectrum b-lactamase-mediated resistance. J Clin Microbiol
2006; 44: 1630– 4.
18 Nathwani D, Morgan M, Masterton RG et al. Guidelines for UK practice
for the diagnosis and management of methicillin-resistant
Staphylococcus aureus (MRSA) infections presenting in the community.
J Antimicrob Chemother 2008; 61: 976–94.
35 Department for Environment, Food and Rural Affairs.
Extended-Spectrum Beta-Lactamases (ESBL) in bacteria associated
with animals—Defra Position. http://www.defra.gov.uk/foodfarm/
farmanimal/diseases/atoz/zoonoses/esbl.htm (2 October 2009, date
last accessed).
i15
Hunter et al.
36 Karisik E, Ellington MJ, Pike R et al. Molecular characterization of
plasmids encoding CTX-M-15 b-lactamases from Escherichia coli strains
in the United Kingdom. J Antimicrob Chemother 2006; 58: 665–8.
56 Debast SB, van Leengoed LA, Goorhuis A et al. Clostridium difficile PCR
ribotype 078 toxinotype V found in diarrhoeal pigs identical to isolates
from affected humans. Environ Microbiol 2009; 11: 505–11.
37 Machado E, Coque TM, Cantón R et al. Leakage into Portuguese
aquatic environments of extended-spectrum-b-lactamase-producing
Enterobacteriaceae. J Antimicrob Chemother 2009; 63: 616– 8.
57 Alvarez-Perez S, Blanco JL, Bouza E et al. Prevalence of Clostridium
difficile in diarrhoeic and non-diarrhoeic piglets. Vet Microbiol 2009;
137: 302–5.
38 Pitout JD, Nordmann P, Laupland KB et al. Emergence of
Enterobacteriaceae producing extended-spectrum b-lactamases
(ESBLs) in the community. J Antimicrob Chemother 2005; 56: 52– 9.
58 Rodriguez-Palacios A, Stämpfli HR, Duffield T et al. Clostridium
difficile PCR ribotypes in calves, Canada. Emerg Infect Dis 2006; 12:
1730– 6.
39 Health Protection Agency. Laboratory detection and reporting of
extended-spectrum b-lactamases. http://www.hpa-standardmethods.
org.uk/documents/qsop/pdf/qsop51.pdf (30 July 2009, date last
accessed).
59 Keel K, Brazier JS, Post KW et al. Prevalence of PCR ribotypes among
Clostridium difficile isolates from pigs, calves, and other species. J Clin
Microbiol 2007; 45: 1963–4.
40 Potz NA, Hope R, Warner M et al. Prevalence and mechanisms of
cephalosporin resistance in Enterobacteriaceae in London and
South-East England. J Antimicrob Chemother 2006; 58: 320– 26.
41 Al Saif N, Brazier JS. The distribution of Clostridium difficile in the
environment of South Wales. J Med Microbiol 1996; 45: 133–7.
42 Fawley WM, Wilcox MH. Molecular epidemiology of endemic
Clostridium difficile infection. Epidemiol Infect 2001; 126: 343– 50.
43 Freeman J, Baines SD, Saxton K et al. Effect of metronidazole on
growth and toxin production by epidemic Clostridium difficile PCR
ribotypes 001 and 027 in a human gut model. J Antimicrob Chemother
2007; 60: 83– 91.
44 Fawley WN, Underwood S, Freeman J et al. Efficacy of hospital
cleaning agents and germicides against epidemic Clostridium difficile
strains. Infect Control Hosp Epidemiol 2007; 28: 920–5.
45 Geric B, Carman RJ, Rupkik M et al. Binary toxin-producing, large
clostridial toxin-negative Clostridium difficile strains are enterotoxic but
do not cause disease in hamsters J Infect Dis 2006; 193: 1143– 51.
46 Wilcox MH, Mooney L, Bendall R et al. A case-control study of
community-associated Clostridium difficile infection. J Antimicrob
Chemother 2008; 62: 388– 96.
47 CDC. Surveillance for community-associated Clostridium difficile—
Connecticut, 2006. Morb Mortal Wkly Rep. 2008; 57: 340–3.
48 Baines SD, O’Connor R, Freeman J et al. Emergence of reduced
susceptibility to metronidazole in Clostridium difficile. J Antimicrob
Chemother 2008; 62: 1046 –52.
49 NHS Purchasing and supply agency. Clostridium difficile toxin
detection assays. http://www.pasa.nhs.uk/PASAWeb/NHSprocurement/
CEP/CEPproducts/CEP+catalogue.htm (9 July 2009, date last accessed).
50 Health Protection Agency. Questions and answers about the
laboratory diagnosis of Clostridium difficile infection (CDI) http://www.
hpa.org.uk/web/HPAwebFile/HPAweb_C/1238055363795 (9 July 2009,
date last accessed).
51 Rupnik M. Is Clostridium difficile-associated infection a potentially
zoonotic and foodborne disease? Clin Microbiol Infect 2007;13: 457– 9.
52 Bauer MP, Notermans DW, van Benthem BHB et al. First results of the
European Clostridium difficile infection survey. Clin Microbiol Infect 2009;
15 Suppl 1: S20.
53 Jhung MA, Thompson AD, Killgore GE et al. Toxinotype V Clostridium
difficile in humans and food animals. Emerg Infect Dis 2008; 14:
1039– 45.
54 Barbut F, Mastrantonio P, Delmée M et al. Prospective study of
Clostridium difficile-associated disease in Europe with phenotypic and
genotypic characterization of the isolates. Clin Microbiol Infect 2007;
13: 1048– 57.
55 Goorhuis A, Bakker D, Corver J et al. Emergence of Clostridium difficile
infection due to a new hypervirulent strain, polymerase chain reaction
ribotype 078. Clin Infect Dis 2008; 47: 1162 –7.
i16
60 Weese JS, Avery BP, Rousseau J et al. Detection and enumeration of
Clostridium difficile spores in retail beef and pork. Appl Environ Microbiol
2009; 75: 5009– 11.
61 Songer JG, Trinh HT, Killgore GE et al. Clostridium difficile in retail meat
products, USA, 2007. Emerg Infect Dis 2009; 15: 819– 21.
62 Spencer RC. The role of antimicrobial agents in the aetiology of
Clostridium difficile-associated disease. J Antimicrob Chemother 1998;
41 Suppl 3: 21– 27.
63 Borriello SP, Barclay FG. An in vitro model of colonisation resistance to
Clostridium difficile infection. J Med Microbiol 1986; 21: 299–304.
64 Larson HE, Welch A. In-vitro and in-vivo characterisation of resistance
to colonisation with Clostridium difficile. J Med Microbiol 1993; 38:
103–8.
65 Wilson KH. The microecology of Clostridium difficile. Clin Infect Dis
1993;16 Suppl 4: S214–8.
66 Lode H, Von der Hoh N, Ziege S et al. Ecological effects of linezolid
versus amoxicillin/clavulanic acid on the normal intestinal microflora.
Scand J Infect Dis 2001; 333: 899–902.
67 Wilcox MH, Brown A, Freeman J. faecal concentrations of piperacillin
and tazobactam in elderly patients. J Antimicrob Chemother 2001; 48:
155–6.
68 Edlund C, Stark C, Nord CE. The relationship between an increase in
b-lactamase activity after oral administration of three new
cephalosporins and protection against intestinal ecological disturbance.
J Antimicrob Chemother 1994; 34: 127–8.
69 Stiefel U, Nerandzie MM, Koski P et al. Orally administered
b-lactamase enzymes represent a novel strategy to prevent
colonization by Clostridium difficile. J Antimicrob Chemother 2008; 62:
1105– 8.
70 Gerding DN, Olson MM, Johnson S et al. Clostridium difficile diarrhoea
and colonization after treatment for abdominal infection regimens
containing clindamycin or metronidazole. Am J Surg 1990; 159: 212– 7.
71 Baxter R, Ray GT, Fireman BH. Case-control study of antibiotic use and
subsequent Clostridium difficile-associated diarrhoea in hospitalized
patients. Infect Cont Hosp Epidemiol 2008; 29: 44–50.
72 Hirschhhorn LR, Trnka Y, Onderdonk A. Epidemiology of
community-acquired Clostridium difficile-associated diarrhea. J Infect
Dis 1994; 169: 127– 33.
73 Ludlam H, Brown N, Sule O et al. An antibiotic policy associated with
reduced risk of Clostridium difficile-associated diarrhoea. Age Ageing
1999; 28: 578–80.
74 Settle CD, Wilcox MH, Fawley WN et al. Prospective study of the risk of
Clostridium difficile diarrhoea in elderly patients following treatment with
cefotaxime or piperacillin-tazobactam. Aliment Pharmacia Therapy 1998;
12: 1217– 23.
75 Thomas C, Stevenson M, Williamson DJ et al. Clostridium
difficile-associated diarrhea: epidemiological data from Western
Antimicrobial-resistant pathogens in animals and man: prescribing, practices and policies
Australia associated with a modified antibiotic policy. Clin Infect Dis 2002;
35: 1457– 62.
76 Warren RE, O’Neill PM, Fisher M et al. Denominator-based C difficile
incidence with exposure to specific antibiotics in a general UK hospital
before and after antibiotic and infection control interventions. In:
Abstracts of the Forty-ninth Interscience Conference on Antimicrobial
Agents and Chemotherapy, San Francisco, CA, USA, 2009. Poster K2082.
American Society for Microbiology, Washington, DC, USA.
77 Alonso R, Pelacz T, Gonzalez-Abad MJ et al. In vitro activity of new
quinolones against Clostridium difficile. J Antimicrob Chemother 2001;
47: 195–7.
78 Wilcox MH, Fawley W, Freeman J et al. In vitro activity of new
generation fluoroquinolones against genetically distinct and
indistinguishable Clostridium difficile isolates. J Antimicrob Chemother
2000; 46: 551–5.
JAC
79 Woodcock JM, Andrews JM, Boswesll FJ et al. In vitro activity of
BAY12-8039 a new fluoroquinolone. Antimicrob Agents Chemother
1997; 41: 101–6.
80 Lodise TP, Miller C, Patel N et al. Identification of patients with
Pseudomonas aeruginosa respiratory tract infections at greatest risk of
infection with carbapenem-resistant isolates. Infect Control Hosp
Epidemiol 2007; 28: 959–64.
81 Johnson JR, Sannes MR, Croy C et al. Antimicrobial drug-resistant
Escherichia coli from humans and poultry products, Minnesota and
Wisconsin, 2002– 2004. Emerg Infect Dis 2007; 13: 838–46.
82 Veterinary Medicines Directorate. Sales of antimicrobial products
authorised for use as veterinary medicines, antiprotozoals, antifungals,
growth promoters and coccidiostats, in the UK in 2007. http://www.
vmd.gov.uk/Publications/Antibiotic/salesanti07.pdf (31 July 2009, date
last accessed).
i17