|
Recent data of the Danish Integrated Antimicrobial
Resistance Monitoring and Research Programme (DANMAP) show that, in
Denmark, resistance levels among Salmonella enterica are modest and
that resistance in Escherichia coli isolates causing disease in
animals should be seen as a resistance reservoir rather than a threat to
the public health. Furthermore, trends of resistance in Danish E. coli
clinical isolates were consistent with Danish patterns of antimicrobial
use. Resistance in Staphylococcus aureus and Streptococcus
pneumoniae remains rare. However, the recent emergence of penicillin
and erythromycin resistance in S. pneumoniae is of concern.
Finally, monitoring of resistance in commensal bacteria from food animals
and food, shows the positive impact on resistance of interventions such as
the ban of antimicrobial growth promoters in farm animals.
In 1995, in the interest of consumer safety and the health of animals
and humans, the Danish ministry of
agriculture and fisheries, and the ministry of health, asked the Danish
Veterinary Laboratory, the Danish Veterinary and Food Administration, and
the Statens Serum Institut to collaborate in surveillance and research on
antimicrobial resistance. This request resulted in the development of the
Danish Integrated Antimicrobial Resistance Monitoring and Research
Programme (DANMAP). The Danish Medicines Agency joined the programme in
1998. The programme focuses mainly on the control of resistance in food
animals and the risks of transmission of resistant enteric pathogens and
commensal bacteria to humans through food. Data are also reported on
resistance in other bacteria, such as Staphylococcus aureus and Escherichia
coli, which are responsible for infections in food animals and humans.
This article summarises the most recent results from DANMAP, which have
been published in full in English by the Danish Zoonosis Centre as an
annual report (1). The report describes DANMAP’s objectives, principles,
and methods, which have also been described in Eurosurveillance
(1-3). Similarly, detailed data on antimicrobial use in humans and food
animals have been presented elsewhere (1,4).
Surveillance of resistance in enteric pathogens
A main objective of DANMAP is to follow variations of resistance levels
over time. This objective has been difficult to fulfil for enteric
pathogens because the isolates included in the programme do not represent
unbiased, random samples. For
example, increasing recognition of outbreaks of Salmonella enterica
serotype Typhimurium definitive phage type (DT)104 – two domestic
outbreaks in 1999 (5) – means that this pentaresistant phage type
accounts for a disproportionally high percentage of S. Typhimurium
isolates from cattle. Nevertheless, an overall assessment of the
occurrence of resistance in Salmonella enterica shows that, apart
from the bias due to S. Typhimurium DT104, resistance levels are
modest. For example, only 11% and 13% of S. Typhimurium other than
DT104 isolated from human clinical specimens in 1999 were resistant to
ampicillin and tetracycline, respectively. For S. Enteritidis,
about 2% were resistant to ampicillin or to tetracycline in 1999.
The same reasons make it difficult to study resistance trends in
campylobacter isolates. In 1999, 1% of Campylobacter jejuni
isolates from broilers, 13% from broiler meat samples, and 10% from human
stool specimens were resistant to tetracycline. Tetracycline resistance
was significantly commoner in C. jejuni isolates from imported than
from Danish broiler meat from retail outlets. Resistance to erythromycin
among C. coli isolates from pigs fell by almost 50% from 1998 to
1999, probably because tylosin’s use as an antimicrobial growth promoter
was withdrawn from Danish pig production.
Surveillance of resistance in other bacteria of clinical importance
DANMAP monitors resistance in bacteria identified in specimens
submitted for diagnosis from food animals and from humans. In 1999, the
programme showed that resistance was generally rare in Staphylococcus
aureus and coagulase-negative staphylococci isolates from mastitis in
cattle. Conversely, a high level of resistance was detected in Escherichia
coli isolated from diseased cattle and pigs, although resistance in
commensal E. coli isolated from animals at slaughter is still rare.
Thus, resistance in E. coli isolates causing disease in animals
should be seen as a resistance reservoir rather than an immediate threat
to the public health.
Resistance in E. coli isolates from human clinical specimens was
estimated from the results of susceptibility testing from the clinical
microbiology laboratories serving four counties throughout Denmark and
serving about a third of the Danish population. Since 1995, these
laboratories have provided summary reports once a year of the numbers of
clinical E. coli isolates from blood (hospital isolates only) and
urine (hospital and primary health care isolates) tested for antimicrobial
susceptibility, and the numbers of these isolates that were resistant to
selected antimicrobials. Although all laboratories were asked to remove
duplicate isolates from the same patient within a window of 30 days, only
the laboratories serving North Jutland County and Copenhagen Municipality
were able to comply with this rule. Susceptibility testing methods also
vary between laboratories. Neo-SensitabsÒ
tablets (A/S Rosco) are used in the laboratories from Roskilde County and
North Jutland County, and were also used in Copenhagen Municipality until
June 1999, when they were replaced by disk diffusion (Oxoid). A home-made
pre-diffusion disk method was used in Aarhus County laboratory during this
period. All laboratories take part in the United Kingdom National External
Quality Assessment Schemes (NEQAS).
Some surveillance results for E. coli from 1995 to 1999 are
presented in the figure. In 1999, each laboratory reported data on 162 to
585 E. coli isolates from hospital blood specimens, on 2009 to 5228
E. coli isolates from hospital urine samples, and on 2501 to 4107 E.
coli isolates from urine specimens from primary health care. Because
of differences in susceptibility methods, breakpoints, data handling, and
possibly frequencies of sampling, comparisons among counties are
difficult. One should also be aware that a substantial proportion of urine
specimens from primary health care is submitted to the laboratory because
of treatment failure and therefore represent a selected population.
Despite these methodological limitations, resistance levels showed little
variations between counties. Additionally, the four laboratories generally
reported similar resistance trends - for example, to ampicillin - which
indicates that the data can be used to estimate the levels and trends of
resistance in Danish E. coli clinical isolates. In 1999, resistance
to ampicillin was detected in 35% to 45% of E. coli isolated from
blood and urine, to sulphonamides in 30% to 40% of E. coli (urine
isolates only), to cefuroxime in less than 3% (blood isolates only), and
to ciprofloxacin in less than 2%. These resistance levels were consistent
with Danish patterns of antimicrobial use. For example, fluoroquinolone
use in Denmark amounted only 0.26 DDD (Defined daily dose) per 1,000
inhabitant-days in 1999 (or 1.8% and 4.7% of total use in primary health
care and in hospitals, respectively), which is among the lowest usages of
fluoroquinolones in the European Union. Sulphonamides are often the drug
of choice for uncomplicated urinary tract infections in Denmark.
With the exception of penicillin resistance, less than 5% of S.
aureus blood isolates are resistant to antimicrobials; methicillin
resistant isolates only represent about 0.5% of S. aureus isolated
from blood specimens collected nationwide and registered by the national
reference centre for staphylococci. Between 1994 and 1999, the percentage
of penicillin resistant or intermediate (MIC ³
0.125 mg/ml) Streptococcus pneumoniae
isolates from blood and cerebrospinal fluid, as reported by the
national reference centre for pneumococci, increased from 0.3% to 3.8%.
This increase has been paralleled by an increase in erythromycin
resistance. There has been no major change in the annual amount of
penicillins and macrolides used in Denmark in recent years, but the
emergence of penicillin and erythromycin resistance in S. pneumoniae
coincided with the introduction of azithromycin and has paralleled its use
in primary care. While resistance in S. pneumoniae is still very
rare in Denmark, this recent and steady increase is of concern.
Surveillance of resistance in commensal bacteria
Finally, DANMAP monitors resistance in commensal bacteria from food
animals and food. A practical approach to the regular monitoring of
resistance in commensal bacteria from healthy humans is being
investigated. Since the programme was set up there has been a marked
decrease in the percentage of Enterococcus faecium isolated from
food animals at slaughter with resistance to avoparcin, virginiamycin,
avilamycin, and macrolides. With a delay of about one year, these changes
in resistance have paralleled similar changes in the consumption of the
same antimicrobials used as growth promoters. These results show that
interventions on antimicrobial use in farm animals, such as the ban of
antimicrobial growth promoters initiated in Denmark in 1995, have an
effect on resistance. A decrease in resistance has also been observed in E.
faecium from food samples collected at retail outlets. For example,
the proportion of E. faecium isolates from beef, broiler and pork
meat with erythromycin resistance has fallen by 30% to 40% between 1998
and 1999. Although the pressure due to vancomycin resistant E. faecium
(VRE) in the food chain has fallen dramatically, studies by the Danish
Veterinary Laboratory using selective methods show that small numbers of
VRE are still commonly found in broiler flocks.
Conclusion
Until now the programme has focused mainly on the control of resistance
in food animals and the risks of transmission of resistant bacteria to
humans through food. A wide range of pathogens from human clinical
specimens are tested for resistance in Danish clinical microbiology and
reference laboratories, but only data on enteric pathogens, E. coli,
staphylococci, and S. pneumoniae are presently made available to
DANMAP. Recently, the Statens Serum Institut received additional funding
from the Danish Ministry of Health to implement a national system for the
surveillance of antimicrobial resistance through a network of clinical
microbiology laboratories. This new initiative should result in a better,
standardised, system for the collection of susceptibility data on bacteria
isolated from human clinical specimens. In the five years since DANMAP was
launched, it has provided a basis for controlling antimicrobial
resistance, especially in the food chain, and has demonstrated the value
of an integrated, practical, and active approach to monitoring and
controlling bacterial antimicrobial resistance.
Acknowledgements
We thank CS Elsberg, J Engberg, P Gerner-Smidt, HB Konradsen, K
Mølbak, JK Møller, HC Schønheyder, H Westh, and N Frimodt-Møller for
providing resistance data in bacteria isolated from clinical
microbiological samples, and K Hovgaard and HL Johansen for providing data
on antimicrobial use in humans.
|