Guide to Antimicrobial Use in Animals

88 Guide to antimicrobial use in animals

For most antimicrobial drugs, biotransformation is can either enhance or retard absorption from the
of importance primarily because it normally reduces gastrointestinal tract (and hence affect bioavailabil-
or abolishes activity. Biotransformation, together with ity), depending on whether the drug is actively taken
renal excretion, is the mechanism whereby antimicro- up by enterocytes or extruded from enterocytes back
bial activity is terminated. Examples of the pharma- into the gastrointestinal liquor. Drug transporters also
cokinetic consequences of varying degrees of drug exist in other tissues and are responsible, for example,
lipid solubility on biotransformation are given in for active secretion of penicillins and cephalosporins
Table 6.4. The liver is the main organ responsible for by renal proximal tubular cells into tubular fluid and
drug biotransformation, although it is now recognised thence into urine, and active extrusion of penicillins
that many other organs, tissues and cells, for example from CSF into plasma. There is increasing recognition
kidney and enterocytes, can metabolise some drugs. that for some antimicrobial drugs, transporters can
It should be noted that species variation in the rate influence both intestinal absorption and penetration
of metabolism is the rule rather than the exception, to sites of action.
so that there are profound inter-species differences in
clearance and terminal half-life for individual drugs As some antimicrobial drugs (e.g. most aminogly-
(Table 6.5). This phenomenon determines major cosides, penicillins and cephalosporins) are rapidly
species differences in both dose requirements and cleared and have short elimination half-lives (Tables 6.4
half-life, determining dose intervals. Moreover, mech- and 6.5), therapeutic concentrations are maintained
anisms and pathways of hepatic metabolism and renal for only a few hours after a single intravenous or intra-
excretion are not fully developed in neonates, leading muscular administration in aqueous solution (e.g.
to slower clearance and a longer elimination half-life. sodium salt of benzylpenicillin). Their use requires
Reduced hepatic and renal functions may also occur repeated dosing at short intervals and is therefore
in aged animals. Finally, in fish, clearance and elimi- impractical under clinical conditions. Hence, a com-
nation rates vary markedly with body temperature, mon practice is to formulate antimicrobial prepa-
and drugs such as sulfadimidine, trimethoprim and rations as aqueous or oily suspensions of poorly
oxytetracycline have up to three-fold longer half-life water-soluble salts (e.g. procaine benzylpenicillin) or
values at low (10–12°C) compared to high (20–25°C) as solutions in organic solvents (e.g. oxytetracycline),
environmental (and body) temperatures. especially for use in farm animals. These depot prod-
ucts, when injected by a non-vascular route (usually
A further consideration in relation to antimicrobial intramuscularly but sometimes subcutaneously), are
pharmacokinetics is the existence of active transport taken up slowly into solution at the injection site and
mechanisms for some antimicrobial drugs. These provide more persistent concentrations in plasma

Table 6.5 Examples of species differences in elimination half-life
Half-life (h)

Drug Cow Horse Dog Man

Benzylpenicillina 0.7 0.9 0.5 1.0
Ampicillina 1.0 1.2 0.8 1.3
Gentamicina 1.8 2.2 1.3 2.8
Trimethoprimb 1.3 3.2 4.6 10.6
Norfloxacinb 2.4 6.4 3.6 5.0
Sulfadiazineb 2.5 3.6 5.6 9.9
Metronidazoleb 2.8 3.9 4.5 8.5
Sulfadimethoxineb 12.5 11.3 13.2 40.0

aFor lipid-insoluble drugs eliminated mainly by renal excretion, half-life is relatively short, with little variation between species.
bFor drugs of moderate or high lipid solubility, elimination is usually dictated mainly by biotransformation (most commonly but not

exclusively) in the liver, but with some renal excretion of parent drug. Species variation in biotransformation rates are commonly

considerable, being rapid in ruminants and horses and slower in humans.

Antimicrobial treatment on selection of resistant bacteria 89

and other biological fluids. The terminal half-life of interval). AUC is the area under the plasma or blood
antimicrobial drugs in depot formulations usually concentration–time curve over 24 h, Cmax is the maxi-
represents the absorption rather than elimination mum concentration, and T>MIC is the time for which
half-life (flip-flop pharmacokinetics). Another basis concentration exceeds MIC (Figure 6.5). PK–PD indi-
for prolonged duration of action is the recent intro- ces are predictive parameters for treatment outcome
duction of cefovacin, a cephalosporin with a very based on empirical observations. This PK–PD inte-
high degree of binding to plasma protein, which is gration approach utilises both of the pharmacological
both slowly excreted in urine and not significantly properties, pharmacokinetics and pharmacodynam-
metabolised in the liver. Therapeutic efficacy of long- ics, which determine the outcome of therapy. Based
acting formulations may therefore be obtained with on these surrogates, killing actions have been classi-
single, or at most two, doses. This avoids the stress fied as time-dependent, concentration-dependent or
caused by repeated injections, minimises the risks of co-dependent (Table 6.2). PK–PD indices should be
non-compliance and limits marked variations in drug derived from non-protein-bound drug concentra-
plasma concentration during therapy. For drugs with tions in plasma, as only the free drug is microbiologi-
time-dependent killing mechanisms, the presence of cally active.
peaks and troughs may predispose to resistance devel-
opment and this provides an additional advantage for For β-lactam agents, in general (1) increasing
depot formulation products. plasma concentrations above 4XMIC does not pro-
vide greater or more rapid bacterial killing and (2)
6.4 Efficacy breakpoints based maximum bactericidal activity may be achieved when
on concentration–time–effect T>MIC exceeds 40–50% of the dosage interval. For
relationships optimal antimicrobial effect, however, a dosage which
provides T>MIC of 80–100% (72), particularly for
6.4.1 Integrated pharmacokinetic Gram-negative pathogens, may be required. On the
and pharmacodynamic other hand, for concentration-dependent killing
(PK–PD) indices drugs of the fluoroquinolone group, it has been widely
recommended that: (1) AUC0–24 h/MIC should exceed
As a pharmacological basis for dosage to optimise 125 h, that is, the average daily plasma concentration
the kill of pathogens and minimise the emergence of should be approximately five times greater than MIC:
resistance, three PK–PD indices linking concentra- and (2) Cmax/MIC should be at least 10 (73–75). An
tion (usually in serum or plasma) to MIC have been AUC/MIC of 125 h correlates well with bacteriologi-
proposed (65–72): AUC/MIC, Cmax/MIC ratios and cal cure for fluoroquinolones in human clinical tri-
T>MIC (expressed as a percentage of the inter-dose als and in infection models in experimental animals.
When comparing rapid intravenous injection of
danofloxacin with slow intravenous infusion, similar

In vivo PK–PD integration in serum of goat 5 after single
IM injection at a dose rate of 1.25 mg/kg danofloxacin

Danofloxacin concentration (μg/ml) 1.000 Cmax/MIC = 12 AUC24 h = 2.32 μg.h/ml
AUIC24 h = 2.32/0.030 = 77 SIT−1.h

0.100

Figure 6.5 An example of ex vivo T>MIC =13.5 h Serum Conc.
PK–PD integration for danofloxacin
administered intramuscularly in a goat, 0.010
showing serum concentration–time
relationship and derivation of Cmax/MIC, 0.001 6 12 18 24
AUC/MIC (AUIC24 h) and T>MIC (From 0 Time (h)
Aliabadi and Lees (2001), Am. J. Vet.
Res. 62: 1979–89).

90 Guide to antimicrobial use in animals

AUC/MIC ratios but differing T>MIC and Cmax/MIC or clinical conditions of the study. For β-lactams,
ratios were achieved in a model of calf pneumonia the longer PAE against Gram-positive than against
(76). The bolus injection provided better bacteriolog- Gram-negative bacteria is a potential cause of vari-
ical and clinical responses, confirming the concentra- ability, and strain differences in the values of PK–PD
tion-dependent killing action of fluoroquinolones. indices have been reported. Veterinary data are espe-
cially sparse. However, in one study Guyonnet et al.
Different routes of administration may result in dif- (88) obtained AUC0–5 h/MIC values of 29.2, 7.9,
ferent pharmacokinetic profiles, which in turn may 6.8 and 6.2 h against four strains of porcine E. coli
affect the intensity of selection for resistance. The for an in vitro bactericidal action of colistin. Thus,
impact of the administration route on the emergence of one strain was a clear outlier. For many antimicro-
resistance has often been debated but few studies have bial breakpoint surrogate values, based on rate and
addressed the question. A Cmax/MIC of 10 has been pro- extent of killing and resistance emergence, are not
posed as a breakpoint value for minimising emergence available. When breakpoints have been determined in
of resistance to aminoglycosides and fluoroquinolones experimental animals (usually mice or rats), infection
(74, 77, 78). Assuming that achieving a high plasma models and/or in human clinical trials, the numerical
Cmax is essential to prevent resistance development, values must be extrapolated to veterinary clinical cir-
pharmacokinetic studies of fluoroquinolone treatment cumstances with caution. There is an urgent need for
in healthy animals indicate that the intramuscular route further research to either validate breakpoint values
is superior to oral (and intragastric) administration determined in experimental animals and in humans
routes (79–82). In pigs, experimentally infected with a for application in veterinary medicine, or yield alter-
99:1 mixture of Salmonella Typhimurium susceptible native values.
and resistant to nalidixic acid, intramuscular adminis-
tration of enrofloxacin resulted in reduced selection of Most authors have proposed MIC 90 as the relevant
the resistant variant compared to oral administration, pharmacodynamic index for optimising dosage, but
and increased intramuscular doses were more efficient others have suggested MIC 50 (75) as less stringent and
in preventing resistance development (83). When using more readily determined than MIC 90. For fluoroqui-
intramuscular injection and designing dosage schedules, nolones there is evidence from many sources, based
tissue tolerability must be assessed (84). Subcutaneous on in vitro data, animal disease models and clini-
and intramuscular administration of danofloxacin in cal trials in humans, that the AUC/MIC ratio is the
cattle has been shown to produce virtually identical PK–PD surrogate that best correlates with efficacy. It
plasma concentration profiles and AUCs following one, is widely reported that daily dosage should provide a
three or five consecutive daily doses (85). Oral admin- ratio of at least 125 h. However, doses yielding lower
istration given by continuous administration via drink- values are likely to be acceptable when the bacterio-
ing water (rather than via a nasogastric probe) further logical burden is low, especially in immunocompetent
increases the risk of development of resistance since high animals.
peak concentrations cannot be achieved. Accordingly,
reduced thirst in sick animals and hierarchy in a flock 6.4.2 Veterinary examples of PK–PD
can potentially cause underdosing in some animals. integration and modelling
For β-lactams and other time-dependent antimicrobial
agents, a long half-life may provide the best treatment An approach to designing optimal dosage schedules,
outcome. Subcutaneous administration of imipenem based on PK–PD integration and PK–PD modelling,
and meropenem to dogs results in extended half-lives has been proposed (54–58, 89). This involves conduct-
in comparison with the intramuscular and intravenous ing bacterial time–kill studies sequentially in vitro, ex
routes (86, 87). vivo and in vivo, followed by population PK–PD mod-
elling in clinical trials (70, 71, 90) (see Sections 6.6
Numerical values of Cmax/MIC, AUC/MIC and and 6.7). It is proposed that such studies are superior
T>MIC cited in the scientific literature can provide use- in many instances to the traditional approach based
ful guidelines to dosage determination. However, such on dose titration studies. The latter will generally
values can be both drug- and bacterial species-specific, yield a clinically effective dose, whilst dosage based on
and may additionally depend on the experimental

Antimicrobial treatment on selection of resistant bacteria 91

PK–PD modelling is designed to yield an optimal reduction in bacterial count of 99.9% (bactericidal
dose for bacteriological cure. Optimal dosage varies action) and eradication of organisms (Table 6.6).
with the organism type and location. When applied Subsequent calculations, based on MIC 90 values for
to a given organism, PK–PD modelling allows for the M. haemolytica, indicated, in the calf, for a bacte-
two principal sources of inter- and intra-subject vari- ricidal response or eradication of bacteria, respec-
ability in treatment outcome, pharmacokinetics and tively, doses of 4.1 and 7.5 mg/kg. If it is assumed
pharmacodynamics. Veterinary examples of PK–PD that a bacteriological effect level intermediate
integration and PK–PD modelling applied to dose between 99.9% kill and eradication of bacteria is
schedule design studies are provided in Figures appropriate, these values support the manufactur-
6.5 and 6.6, respectively. Studies conducted ex vivo er’s currently recommended dose of 6 mg/kg. This
with danofloxacin in four ruminant species (calf, intermediate effect level is similar to that proposed
goat, sheep, camel) may be cited (54–58). Against by Mouton et al. (68) in humans. They suggested
pathogenic strains of M.haemolytica (calf, goat, for fluoroquinolones a dosage based on an AUC/
sheep) and E.coli (camel), data describing the whole MIC ratio of 90% of Emax, whilst indicating that in
sweep of the AUC/MIC to bacterial count (CFU) immunocompetent subjects and when infections
relationship was obtained. Modelling the data to are not severe, the AUC/MIC ratio providing 50%
the sigmoidal Emax equation provided numeri- of Emax may be acceptable. These considerations
cal values required for four parameters of activity, relate to efficacy but may not be applicable to resist-
bacteriostasis, reduction in bacterial count of 50%, ance development (vide infra).

PK–PD integration using the sigmoidal Emax relationship for bacterial
count versus ex vivo AUIC24 h in goat serum

Figure 6.6 An example of PK–PD Log cfu/ml difference 1 Bacteriostatic Observed
modeling for danofloxacin admin- 0 AUIC24 h = 18 Predicted
istered intramuscularly in a goat, −1
illustrating the relationship between −2 Bactericidal 250 300
change in bacterial count from −3 AUIC24 h = 39
baseline and AUC/MIC (AUIC24 h) −4
and derivation of values producing −5 Elimination
bacteriostasis, bactericidal action and −6 AUIC24 h = 90
eradication of bacteria (From Aliabadi −7
and Lees (2001), Am. J. Vet. Res. 62: 50 100 150 200
1979–89). 0 AUIC24 h

Table 6.6 Critical ex vivo values of serum AUC24/MIC(h) for danofloxacina in four ruminant species

Level of growth inhibition Calf Goat Sheep Camel

Bacteriostatic 15.9 ± 2.0 22.6 ± 1.7 17.8 ± 1.7 17.2 ± 3.6
Bactericidal 18.1 ± 1.9 29.6 ± 2.5 20.2 ± 1.7 21.2 ± 3.7
Eradication 33.5 ± 3.5 52.4 ± 8.1 28.7 ± 1.8 68.7 ± 15.6

aIntramuscular administration at a dose rate of 1.25 mg/kg. Antibacterial activity was evaluated by bacterial count ex vivo after 24 h
incubation. The pathogens tested were strains of Mannheimia haemolytica (calf, goat, sheep) and E. coli (camel). The AUC24 h/MIC
relationship to log10 change in bacterial count (CFU/ml) was modellised by a Hill model.
Values are mean ± SEM (n=6).
Data from Aliabadi and Lees (2001, 2003) and Aliabadi et al. (2003a,b).

92 Guide to antimicrobial use in animals

6.4.3 Limitations and pitfalls in using recommended doses of the triamilide tulathromycin
PK–PD indices are extremely low: AUC/MIC = 7.6 h, Cmax/MIC =
0.04 and T>MIC = 0 h (94, 95). The explanation for
One variable significantly affecting breakpoints is these extraordinarily low values is unclear. Lung tis-
inoculum size (see Section 6.6). Moreover, numerical sue concentrations greatly exceed those in plasma,
values of one or more of the surrogate PK–PD indi- and it has been postulated that concentrations in
ces, Cmax/MIC, AUC/MIC and T>MIC, required to epithelial lining fluid (the biophase for bovine res-
provide a given level of efficacy for example a bacteri- piratory disease) may also exceed those in plasma.
cidal response, will not necessarily be the ame as those Certain macrolides are known to possess immu-
which avoid or reduce selective pressure for resist- nomodulatory and anti-inflammatory properties,
ance. Although research in this crucial area is limited, so that non-antimicrobial actions in the host may
and in the veterinary field is virtually absent, some contribute significantly to therapeutic response.
recent studies have investigated the potential optimal In calves, tilmicosin induces neutrophil apoptosis,
requirements for resistance avoidance. reduces pulmonary inflammation and controls
P. haemolytica infection. A similar action has been
Drusano (91) has defined the conditions for reported for erythromycin. Reported anti-inflam-
designing counterselective dosing schedules to avoid matory and immunomodulatory actions of mac-
resistance emergence by mutations as follows: (a) the rolides include increased airway epithelial cell ciliary
total organism burden should substantially exceed the motility, reduced leucocyte accumulation, decreased
inverse of the mutational frequency to resistance; (b) secretory functions of airway cells and reduced epi-
there should be a high probability of a resistant clone thelial cell synthesis of pro-inflammatory cytokines,
being present at baseline; and (c) the step size change for example IL6 (96–102). Similar considerations
in MIC of the mutated population should be relatively apply to other macrolides used to treat respiratory
small, not more than 10-fold. Under these conditions, infections in farm animals, such as tilmicosin, which
it is possible to select a dosage regimen suppressing is rapidly cleared from plasma but accumulates in
both the wild-type susceptible population and more lung tissue. Irrespective of the mechanism(s), mac-
resistant sub-populations. rolides and triamilides are effective in vivo at plasma
concentrations markedly below those predicted
Mouton (92) has noted that AUC/MIC values of from correlating plasma concentration obtained in
some fluoroquinolones are similar for a given magni- vivo to MIC measured in vitro.
tude of bacteriological effect. However, other studies
have suggested differences between marbofloxacin and Another circumstance in which the conventional
danofloxacin against a single strain of the calf patho- PK–PD paradigm may not be applicable in a sim-
gen Mannheimia haemolytica (55, 56). First, AUC/ plistic manner comprises bacteria in biofilms. In
MIC values measured ex vivo in calf serum for the two biofilms bacteria exist as consortia rather than as
drugs differed as the serum concentration multiples planktonic or non-aggregated cells. The biofilm
of MIC required to eradicate the selected strain of comprises biopolymers that provide a permeabil-
M. haemolytica were 1.40 for danofloxacin and 4.96 ity barrier to drug penetration. Such organisms are
for marbofloxacin. This difference possibly reflects a less susceptible to antimicrobial agents than free-
difference in MPC for the two drugs. Second, slopes living cells. Infections associated with biofilms are
of the AUC/MIC–bacterial count relationship also increasingly recognised. In addition to protection
differed, suggesting possible differences in potency against antimicrobial drugs, biofilms containing
and sensitivity for these two fluoroquinolones. human clinical isolates contain more resistance
phenotypes (103, 104). Organisms in biofilms have
Macrolides and triamilides comprise drug classes slow growth rates and most antimicrobial agents
that have proved less easy to categorise in terms of act only on dividing organisms. Another microbial
the PK–PD paradigm than agents of other classes. protective mechanism of organisms in biofilms is
Generally, they are classified as time-dependent reduced apoptosis (105), and a hyper-mutability state
killing drugs, but some newer agents in human causing antimicrobial resistance has been reported in
use, such as azithromycin, exert a significant PAE P. aeruginosa associated infection in human patients
and the marker correlating best with outcome is (106). Hence, several protective mechanisms result
AUC/MIC (93). For P. multocida of bovine origin,
indices based on plasma concentration obtained using

Antimicrobial treatment on selection of resistant bacteria 93

in the general inapplicability of the PK–PD approach resistance (108) and, in a murine peritonitis
to dosage determination for biofilm organisms. model, resistance to ciprofloxacin was lower for
P. aeruginosa when Cmax/MIC was 20 than for a
6.5 Resistance breakpoints based value of 10 (109).
on concentration–time–effect 2. A classical study in human pneumococcal patients
relationships receiving fluoroquinolone therapy showed that
resistance had emerged after 5 days in 50% of
6.5.1 General considerations subjects when AUC/MIC values were <100 h.
After 3 weeks of therapy this had increased to
Bacterial populations are heterogeneous, comprising 93% of patients (110). However, when AUC/MIC
sub-populations each with its own susceptibility to a was greater than 100 h the probability of organ-
given antimicrobial drug. Exposure to an antimicro- isms remaining susceptible exceeded 90%. In an
bial drug exerts selective pressure, so that the most earlier study, the same group (111) reported that
susceptible sub-populations are eradicated, leading to resistance selection for fluoroquinolones was
overgrowth of those sub-populations of least suscep- greater when Cmax/MIC was less than 8. Both
tibility. It is exposure, and especially repeated expo- investigations were conducted in seriously ill and
sure, to sub-optimal drug concentrations that is the possibly immunocompromised human patients
most important single factor in resistance emergence and the numerical values of AUC/MIC and Cmax/
and its subsequent spread (12). As exposure is dose MIC required to avoid resistance may be lower in
related, there is a direct link between administered animals and humans that are immunocompetent.
dose and resistance development. These fundamental Nevertheless, to avoid resistance with fluoroqui-
principles apply to commensal as well as pathogenic nolones the value of AUC/MIC may need to be
organisms, so that even with adequate exposure of higher than that required to achieve bacteriologi-
pathogens, commensals may be underexposed. This cal cure, as the next example illustrates.
situation may lead to development of resistance in the 3. In an in vitro study with Staphylococcus aureus,
commensal flora and subsequent transfer of resistance Firsov et al. (112) investigated the ability of the
genes both within and between pathogenic organisms fluoroquinolones, gatifloxacin, ciprofloxacin,
and commensals at a later stage (13). moxifloxacin and levofloxacin, to selectively
enrich resistant mutants in a dynamic model,
Numerical values of one or more of the surrogate which reproduced in vivo pharmacokinetic pat-
PK–PD indices, Cmax/MIC, AUC/MIC and T>MIC, terns in man.A relationship for fluoroquinolones
required to provide a given level of efficacy for exam- between AUC/MIC and resistance emergence at
ple a bactericidal response, will not necessarily be the 72 h was established. When resistance frequency
same as those which avoid or reduce selective pressure was plotted against the 24 h AUC/MIC ratio a
for resistance. Although research in this crucial area is bell-shaped curve was obtained, and for AUC/
limited, and in the veterinary field is virtually absent, MIC values of less than 10 h or greater than 200 h
some recent studies have investigated examples that there was no resistance. The maximum degree
can be cited to illustrate the potential optimal require- of resistance occurred with AUC/MIC values
ments for resistance avoidance. of 24–62 h. When AUC/MIC was in the range
201–244 h, concentrations exceeded MPC for
6.5.2 Concentration-dependent 80% of the dosage interval. The AUC/MIC for
killing drugs fluoroquinolones generally accepted for opti-
mal efficacy is 125 h, which is less than the 200
1. Preston et al. (107) reported that, for the fluo- h value reported by Firsov et al. (112) for resis-
roquinolone levofloxacin, in human patients, a tance avoidance.
Cmax/MIC > 12.2 provided 100% microbiologi- 4. In a hollow fibre infection model, and using a
cal kill. Other in vitro studies with ciprofloxacin dense inoculum, Tam et al. (113) investigated
and sparfloxacin have confirmed that high Cmax/ resistance emergence of P. aeruginosa to garenox-
MIC ratios are required to avoid emergence of acin; Cmax was held constant and AUC/MIC ratios
ranged from 0 to 200 h. With AUC/MIC values of

94 Guide to antimicrobial use in animals

10, 48 and 89 h most susceptible organisms were was no resistance, while peak mutation frequency
replaced by resistant mutants; and all susceptible occurred when T>MIC=70%. In fact, mutation
organisms were replaced by resistant mutants to resistance was very low when T>MIC was 87%
when AUC/MIC = 108 and 137 h, but resistant or greater. This contrasts with experimental ani-
mutants did not emerge when AUC/MIC = mal and human clinical data, which have often
201 h. reported that the optimal T>MIC for bacterial
5. Jumbe et al. (114) used P. aeruginosa in a mouse kill is of the order of 40–50% of the interdose
thigh infection model to study the effect of esca- interval for β-lactams. These findings suggest
lating doses of levofloxacin on the amplification that a higher value of T>MIC, perhaps ideally
of resistance. When AUC/MIC was 52 h resistant 90 or even 100%, should be the objective for
mutant amplification was maximal; when AUC/ minimising resistance development for time-
MIC = 157 h there was no amplification. dependent killing cephalosporins.
6. Based on the results of an E. coli model of sep- 2. Odenholt et al. (118) investigated whether certain
ticaemia in chickens (115), Toutain et al. (72) concentrations of benzylpenicillin were criti-
determined for a fluoroquinolone, an ED 50 of cal for the selection of resistant subpopulations.
8 mg/kg for reduction in mortality and an ED 50 They exposed a mixed culture of Streptococcus
of 13 mg/kg for bacteriological cure. Although pneumoniae (containing susceptible, intermedi-
the higher dose compared to the lower dose does ate and resistant bacteria) in vitro to the antibiotic
not ensure avoidance of emergence of resistance, for different times above their respective MICs;
it must be much less likely, as outcome is based they showed that selection of resistant bacteria
on bacteriological cure. occurred when concentrations were targeted only
7. For the aminoglycoside netilmicin acting on against fully susceptible strains.
E. coli and S. aureus, Blaser et al. (77) reported 3. Tam et al. (119) using an in vitro hollow fibre
a correlation between Cmax/MIC and emergence infection model, suggested an alternative PK–PD
of resistance. Organism regrowth was prevented index for β-lactams of Cmin/MIC, Cmin being the
when Cmax/MIC was greater than 8. It has been minimum plasma concentration over the dosing
found that for aminoglycosides in general, a interval required to prevent resistance emergence.
Cmax/MIC of 8–10 or higher is required to pre- Against a dense population of P. aeruginosa with
vent resistance emergence (116). drug concentrations up to 4 × MIC, resistant
8. In an in vitro bacterial time–kill study, using four populations formed after exposure to piperacil-
strains of E. coli isolated from pigs, regrowth lin, ceftazidime and meropenem. Interestingly,
occurred when the concentration of colistin (a meropenem was the most effective drug in reduc-
polypeptide belonging to the polymixin group) ing bacterial numbers at a concentration 4 × MIC
was less than 8 or 16 times MIC; growth inhibi- but it also provided the greatest regrowth of resis-
tion with lower multiples of MICs was almost tant sub-populations. These workers found that,
complete at 5 h but re-growth had occurred by even when T>MIC = 100% and Cmin/MIC ratios
24 h. When the concentration was equal to or were less than 1.7, resistance emerged. They con-
greater than 8 or 16 × MIC, re-growth at 24 h did cluded that a Cmin of 6 × MIC was required to
not occur (90). suppress resistance emergence.

6.5.3 Time-dependent killing drugs In summary, it seems likely that the PK–PD break-
point commonly recommended for clinical and
1. To investigate the time-dependent killing drug bacteriological efficacy (usually T>MIC of 40–80% of
ceftizoxime, Stearne et al. (117) used a mixed dosing interval) may be too low for avoidance of resist-
infection murine model. Resistant clones were ance emergence. Finally, it should be noted that opti-
monitored and mutation frequency was shown mal dosing strategies for minimising resistance may
to be related to the surrogate T>MIC, expressed achieve this objective in two ways, first by eradicating
as a percentage of dosing interval. When T>MIC all disease causing pathogenic bacteria and second by
values were either <40 or equal to 100% there exerting minimal selection pressure on commensals.
Of course, these two aims may often differ.

Antimicrobial treatment on selection of resistant bacteria 95

6.6 Pharmacodynamic and the population approach lies in the differences that
pharmacokinetic variability and are likely to occur between PK–PD indices obtained
the requirement for population in the early in vitro investigations and in studies using
studies: use of Monte Carlo healthy animals on the one hand and those that are
simulations appropriate to ‘field’ circumstances on the other.
The former may adequately or even precisely predict
At the first stage of dosage schedule design, it is usual outcome when a drug is used in immunocompetent
to use mean values of integrated PK–PD surrogate healthy animals, for example prophylactically, and
markers to select dosage regimens for subsequent when accurate per animal dosing is possible in indi-
evaluation in the target species, initially in disease vidually treated companion animal subjects. However,
model studies and finally in confirmatory clinical such data are unlikely to apply in all instances when
trails. However, with a normal or near-normal distri- treating infectious disease. Therefore, the final and
bution in such studies, approximately half the animal crucial step in dosage determination should, when
population will have an index below the mean value possible, be taken using the population PK–PD
and the remainder will have a higher value. The final approach within a clinical trial, using a sparse sam-
dosage, therefore, should not be based on mean or pling strategy on a large number of animals. This
median values of PK–PD indices. To minimise resist- provides the opportunity for minimising, at the
ance development, focus is required on those animals population level, the selection and spread of resistant
in the population that do not achieve the desired pathogens. Inadequate exposure of pathogens to the
breakpoint. Toutain et al. (72) have therefore used required drug concentration, in even a small minor-
an approach based on Monte Carlo simulations, ity of animals within a group, may result in establish-
integrating population pharmacokinetics and MIC ment of a resistant sub-population that may transfer
values obtained from field cases. The objective is to resistance genes both vertically and horizontally. This
allow for variability in both pharmacokinetics and factor accounts for individual clinical failures as well
pharmacodynamics in clinical subjects of the target as a stepwise loss of efficacy and development of
species and thereby generate PK–PD indices appro- resistance to a high level.
priate for most animals and not only for the popula-
tion mean (Figure 6.7). Inter-individual variability, resulting in under expo-
sure of a significant proportion of treated animals, is
Population pharmacokinetics and population PK– an inevitable consequence when dose selection targets
PD provide new and potentially improved means for the population mean. There are also differences arising
establishing optimal dosage regimens, that is, those between healthy and diseased animals. A third source
which provide bacteriological cure and minimise, pre- of variability arises in veterinary medicine for those
vent or delay resistance emergence. The necessity for animals, for example fish, pigs, poultry and calves,

Figure 6.7 Hypothetical plasma Concentration (μg/ml) Simulated plasma concentration–time curves
concentration–time relationship for and their mean in a small population
an antimicrobial drug, administered
by a non-vascular route, illustrating 3
the curve for mean concentration and Possible toxicity
each symbol representing an individual
animal. Horizontal lines indicate con- 2.5
centrations above which host toxicity
may occur, below which resistance is 2
likely and the concentration window for Optimal efficacy
optimal efficacy.
1.5
Possible resistance

1

0.5

0
0 6 12 18 24
Time (h)

96 Guide to antimicrobial use in animals

which normally or commonly receive drugs orally in antibiograms, that is, to determine the MIC above
feed or in drinking water with dosing on a group which an organism can be classed as clinically resist-
basis. The nominal mg/kg dose is not the dose actu- ant. This will be the MIC for which a defined dos-
ally received by any animal in the group. The compe- age schedule fails to guarantee that 90% of the target
tition between animals for access to medicated water animal population will be exposed to a mean plasma
or feed leads to variability in the ingested dose. The drug concentration equal to one of the a priori MICs
magnitude of the variability may also be increased of the MIC distribution.
when the drug is given metaphylactically to a group
of animals, only some of which show clinical signs 6.7 Validation and extension of the
of infection. The latter animals may receive lower doses population PK–PD approach
of drug than their more healthy companions. Hence, a to determination of optimal
reduced exposure of those animals carrying the largest dosage regimens
bacterial load and most likely to have the highest path-
ogen mutational frequency is a likely additional factor At the present time, there is a lack of established
predisposing to selection for resistance. The population PK–PD breakpoints derived from population stud-
PK–PD paradigm offers the opportunity of optimising ies in veterinary medicine. Ideally, these should be
dosage to minimise resistance based on the response of set separately for therapeutic, prophylactic and meta-
a given quantile of the target population, say 90% or phylactic use of drugs, with the objective of prevent-
even 95%, rather than the population mean. ing the emergence of resistance. Generally, the initial
bacterial burden under prophylactic and metaphy-
Based on measured pharmacokinetic and pharma- lactic conditions will be lower than that pertaining
codynamic variabilities in the target animal popula- when therapy is required, and inter-animal vari-
tion, Monte Carlo simulations are used to establish ability is also less likely under the former conditions.
the statistical distribution of the selected PK–PD Different breakpoint values and hence differing
index (82, 120). In these simulations, a hypothetical dosage requirements are therefore likely to apply.
population of outcomes is generated, and this permits Jumbe et al. (114) showed that for levofloxacin the
determination of the probability of attaining a pre- breakpoint AUC/MIC values against P. aeruginosa in
selected PK–PD breakpoint in a chosen proportion of mice inoculated in the thigh with 107 or 108 bacteria,
the population. In veterinary medicine, Regnier et al. were 31 and 161 h respectively. Thus a 10-fold increase
(121) and co-workers have used Monte Carlo simu- in the pathogen burden increased five-fold the drug
lations to establish, in the dog, a dosage regimen of exposure required for the same antibacterial effect.
marbofloxacin appropriate for treating infections in The increase in the pathogen burden also increased
the anterior segment of the eye. The Toutain group the size of the resistant population. On the other
also investigated the pharmacokinetic variability of hand, for time-dependent drugs, the value of T>MIC
doxycycline in the pig in a population field study providing an optimal antibacterial effect was not
(122). Pharmacokinetic and pharmacodynamic dis- affected by inoculum size or mechanism of resistance
tributions were modelled to define the percentage but was influenced by host immune status (124).
of pigs attaining a given AUC/MIC value for several
dosage rates of doxycycline. They concluded that a A further consideration for future research is the
dose of 20 mg/kg or greater was required to attain a impact of population PK–PD approaches on zoonotic
PK–PD breakpoint value of 24 h (based on total drug and commensal flora. Antimicrobial drugs are com-
concentration) in 90% of pigs. This is equivalent to monly administered orally in food-producing animals.
obtaining a mean total plasma concentration equal Systemic bioavailability is sometimes low and this
to the actual (but unknown) MIC over a 24 h dosage increases exposure of the gastrointestinal (GIT) flora.
interval. The extent of protein binding of doxycycline This might account for the shedding of zoonotic bacte-
in the pig is approximately 90% (123), indicating that ria such as Salmonellae with resistance to drugs used in
the currently recommended dosage of 10 mg/kg daily humans. In addition, the GIT flora might increase the
does not attain an appropriate breakpoint for AUC/ gene pool of resistance, with possible transmission to
MIC for the free plasma concentration. humans in the food chain. Exposure of the GIT flora
is not confined to orally administered drugs, but may
Monte Carlo simulations can be applied to
the establishment of MIC resistance breakpoints for

Antimicrobial treatment on selection of resistant bacteria 97

also occur with parenterally administered drugs as a 9. Baquero, F., Negri, M.C., Morosini, M.I. and Blazquez, J.
consequence of active efflux by enterocytes, as dem- (1998). Antibiotic-selective environments. Clin. Infect.
onstrated for fluoroquinolones, or through active Dis. 27 Suppl 1: S5–S11.
secretion into bile. The GIT ecosystem is complex
and it is generally the local rather than plasma drug 10. Negri, M.C., Lipsitch, M., Blazquez, J., Levin, B.R. and
concentrations which must be used when applying Baquero, F. (2000). Concentration dependent selection
the PK–PD paradigm to the GIT flora. Moreover, of small phenotypic differences in TEM beta-lactamase-
the pros and cons of long-acting versus short-acting mediated antibiotic resistance. Antimicrob. Agents
products on commensal flora (as well as pathogenic Chemother. 44: 2485–91.
organisms) is not well understood and data are lack-
ing. On the one hand, maintained concentrations 11. Zhao, X. and Drlica, K. (2001). Restricting the selection
may assist bacterial kill of pathogens but, on the of antibiotic-resistant mutants: a general strategy derived
other, prolonged exposure may encourage resistance from fluoroquinolones studies. Clin. Infect. Dis. 33:
development by commensals. Also of significance in S147–S156.
relation to resistance is the use of local therapy as an
alternative to systemic administration. Although lit- 12. Burgess, D.S. (1999). Pharmacodynamic principles of
tle data are available, it would seem likely (in general) antimicrobial therapy in the prevention of resistance.
that the former (a veterinary example is intramam- Chest 115: 195–233.
mary infusions) will expose pathogenic organisms
to higher concentrations than can be achieved with 13. Baquero, F., Negri, M.C., Morosini, M.I. and Blazquez, J.
systemic dosing, whilst sparing the exposure of com- (1997). The antibiotic selective process: concentration-
mensals for example within the GIT. It will rarely, if specific amplification of low-level resistant populations.
ever, be the case that an optimal dosage regimen for CIBA Foundation Symposia, 207: 93–111.
target pathogens will also be optimal to spare the GIT
flora from the emergence and spread of resistance. 14. Lenski, R.E. (1997). The cost of antibiotic resistance–from
the perspective of a bacterium. Ciba Found Symp. 207:
References 131–140; discussion 141–51.

1. Wright, G.D. (2005). Bacterial resistance to antibiotics: 15. Besier, S., Ludwig, A., Brade, V. and Wichelhaus, T.A.
enzymatic degradation and modification. Adv. Drug Deliv. (2005). Compensatory adaptation to the loss of biological
Rev. 57: 1451–70. fitness associated with acquisition of fusidic acid resistance
in Staphylococcus aureus. Antimicrob. Agents Chemother.
2. Kumar, A. and Schweizer, H.P. (2005). Bacterial resistance 49: 1426–31.
to antibiotics: active efflux and reduced uptake. Adv. Drug
Deliv. Rev. 57: 1486–1513. 16. Bjorkman, J., Nagaev, I., Berg, O.G., Hughes, D. and
Andersson, D.I. (2000). Effects of environment on com-
3. Piddock, L.J. (2006). Clinically relevant chromosomally pensatory mutations to ameliorate costs of antibiotic
encoded multidrug resistance efflux pumps in bacteria. resistance. Science 287: 1479–82.
Clin. Microbiol. Rev. 19: 382–402.
17. Levin, B.R., Perrot,V. and Walker, N. (2000). Compensatory
4. Martinez, J.L. and Baquero, F. (2000). Mutation frequencies mutations, antibiotic resistance and the population
and antibiotic resistance. Antimicrob. Agents Chemother. 44: genetics of adaptive evolution in bacteria. Genetics 154:
1771–7. 985–97.

5. Baquero, F. (2001). Low-level antibiotic resistance. In 18. Maisnier-Patin, S., Berg, O.G., Liljas, L. and Andersson,
Andersson, D.H.a.D.I. (ed.), Antibiotic Development and D.I. (2002). Compensatory adaptation to the deleterious
Resistance. Taylor and Francis Group, London, pp. 117–36. effect of antibiotic resistance in Salmonella typhimurium.
Mol. Microbiol. 46: 355–66.
6. Lipsitch, M. and Levin, B.R. (1997). The population dynam-
ics of antimicrobial chemotherapy. Antimicrob. Agents 19. Nagaev, I., Bjorkman, J., Andersson, D.I. and Hughes, D.
Chemother. 41: 363–73. (2001). Biological cost and compensatory evolution in
fusidic acid resistant Staphylococcus aureus. Mol. Microbiol.
7. Tenover, F.C. and McGowan, J.E., Jr. (1996). Reasons for 40: 433–9.
the emergence of antibiotic resistance. Am. J. Med. Sci. 311:
9–16. 20. Nilsson, A.I., Kugelberg, E., Berg, O.G. and Andersson,
D.I. (2004). Experimental adaptation of Salmonella
8. Baquero, F. and Negri, M.C. (1997). Strategies to minimize typhimurium to mice. Genetics 168: 1119–30.
the development of antibiotic resistance. J. Chemother. 9
Suppl 3: 29–37. 21. Reynolds, M.G. (2000). Compensatory evolution
in rifampin-resistant Escherichia coli. Genetics 156:
1471–1481.

22. Schrag, S.J., Perrot, V. and Levin, B.R. (1997). Adaptation
to the fitness costs of antibiotic resistance in Escherichia
coli. Proc. Biol. Sci. 264: 1287–91.

23. Nilsson, A.I., Zorzet, A., Kanth, A., Dahlstrom, S.,
Berg, O.G. and Andersson, D.I. (2006). Reducing the
fitness cost of antibiotic resistance by amplification of
initiator tRNA genes. Proc. Natl. Acad. Sci. USA 103:
6976–81.

98 Guide to antimicrobial use in animals

24. Andersson, D.I. (2006). The biological cost of mutational 37. Robicsek, A., Strahilevitz, J., Sahm, D.F., Jacoby, G.A.
antibiotic resistance: any practical conclusions? Curr. and Hooper, D.C. (2006). qnr prevalence in ceftazidime-
Opin. Microbiol. 9: 461–5. resistant Enterobacteriaceae isolates from the United States.
Antimicrob. Agents Chemother. 50: 2872–4.
25. Criswell, D., Tobiason, V.L., Lodmell, J.S. and Samuels,
D.S. (2006). Mutations conferring aminoglycoside and 38. Wang, M., Sahm, D.F., Jacoby, G.A. and Hooper, D.C.
spectinomycin resistance in Borrelia burgdorferi. (2004). Emerging plasmid-mediated quinolone resistance
Antimicrob. Agents Chemother. 50: 445–52. associated with the qnr gene in Klebsiella pneumoniae
clinical isolates in the United States. Antimicrob. Agents
26. Enne, V.I., Delsol, A.A., Davis, G.R., Hayward, S.L., Chemother. 48: 1295–9.
Roe, J.M. and Bennett, P.M. (2005). Assessment of the fit-
ness impacts on Escherichia coli of acquisition of antibi- 39. Bradford, P.A. (2001) Extended-spectrum beta-lactamases
otic resistance genes encoded by different types of genetic in the 21st century: characterization, epidemiology, and
element. J. Antimicrob. Chemother. 56: 544–51. detection of this important resistance threat. Clin. Microbiol.
Rev. 14: 933–51, table of contents.
27. Gillespie, S.H., Voelker, L.L. and Dickens, A. (2002).
Evolutionary barriers to quinolone resistance in 40. Hasman, H. and Aarestrup, F.M. (2005). tcrB, a gene
Streptococcus pneumoniae. Microb. Drug Resist. 8: 79–84. conferring transferable copper resistance in Enterococcus
faecium: occurrence, transferability, and linkage to mac-
28. Ramadhan, A.A. and Hegedus, E. (2005). Survivability rolide and glycopeptides resistance. Antimicrob. Agents
of vancomycin resistant enterococci and fitness cost Chemother. 46: 1410–6.
of vancomycin resistance acquisition. J. Clin. Pathol.
58: 744–6. 41. Liebana, E., Batchelor, M., Hopkins, K.L., et al. (2006).
Longitudinal farm study of extended-spectrum beta-
29. Sander, P., Springer, B., Prammananan, T., et al. (2002). lactamase-mediated resistance. J. Clin. Microbiol. 44:
Fitness cost of chromosomal drug resistance-conferring 1630–4.
mutations. Antimicrob. Agents Chemother. 46: 1204–11.
42. Riano, I., Moreno, M.A., Teshager, T., Saenz,Y., Dominguez,
30. Bjorkman, J., Hughes, D. and Andersson, D.I. (1998). L. and Torres, C. (2006). Detection and characterization
Virulence of antibiotic-resistant Salmonella typhimurium. of extended-spectrum {beta}-lactamases in Salmonella
Proc. Natl. Acad. Sci. USA 95: 3949–53. enterica strains of healthy food animals in Spain.
J. Antimicrob. Chemother. 58: 844–7.
31. Cohen, S.P., McMurry, L.M., Hooper, D.C., Wolfson, J.S.
and Levy, S.B. (1989). Cross resistance to fluoroquino- 43. Weill, F.X., Lailler, R., Praud, K., et al. (2004). Emergence
lones in multiple antibiotic-resistant (Mar) Escherichia of extended-spectrum-beta-lactamase (CTX-M-9)-
coli selected by tetracycline or chloramphenicol: decreased producing multiresistant strains of Salmonella enterica
drug accumulation associated with membrane changes serotype Virchow in poultry and humans in France.
in addition to OmpF reduction. Antimicrob. Agents J. Clin. Microbiol. 42: 5767–73.
Chemother. 33: 1318–25.
44. Carattoli, A. Lovari, S., Franco, A., Cordaro, G., Di Matteo,
32. Giraud, E., Cloeckaert, A., Kerboeuf, D. and Chaslus- P. and Battisti, A. (2005). Extended-spectrum beta-
Dancla, E. (2000). Evidence for active efflux as the primary lactamases in Escherichia coli isolated from dogs and cats
mechanism of resistance to ciprofloxacin in Salmonella in Rome, Italy, from 2001 to 2003. Antimicrob. Agents
enterica serovar typhimurium. Antimicrob. Agents Chemother. 49(2): 833–5.
Chemother. 44: 1223–8.
45. Feria, C., Ferreira, E., Correia, J.D., Goncalves, J. and
33. Kern, W.V., Oethinger, M., Jellen-Ritter, A.S. and Levy, S.B. Canica, M. (2002). Patterns and mechanisms of resistance
(2000). Non-target gene mutations in the development of to beta-lactams and beta-lactamase inhibitors in uro-
fluoroquinolone resistance in Escherichia coli. Antimicrob. pathogenic Escherichia coli isolated from dogs in Portugal.
Agents Chemother. 44: 814–20. J. Antimicrob. Chemother. 49: 77–85.

34. Oethinger, M., Kern, W.V., Jellen-Ritter, A.S., McMurry, 46. Sidjabat, H.E., Hanson, N.D., Smith-Moland, E.,
L.M. and Levy, S.B. (2000). Ineffectiveness of topoisom- et al. (2007). Identification of plasmamid-mediated
erase mutations in mediating clinically significant fluo- extended-spectrum and AmpC beta-lactamases in
roquinolone resistance in Escherichia coli in the absence Enterobacter spp. isolated from dogs. J. Med. Medicrobiol.
of the AcrAB efflux pump. Antimicrob. Agents Chemother. 56: 426–34.
44: 10–13.
47. Blondeau, J.M., Zhao, X., Hanson, G. and Drlica, K.
35. Corkill, J.E., Anson, J.J. and Hart, C.A. (2005). High preva- (2001). Mutant prevention concentrations of fluoroqui-
lence of the plasmid-mediated quinolone resistance deter- nolones for clinical isolates of Streptococcus pneumoniae.
minant qnrA in multidrug-resistant Enterobacteriaceae Antimicrob. Agents Chemother. 45: 433–8.
from blood cultures in Liverpool, UK. J. Antimicrob.
Chemother. 56: 1115–7. 48. Catry, B., Laevens, H., Devriese, L.A. Opsomer, G. and
De Kruif, A. (2003). Antimicrobial resistance in livestock.
36. Mammeri, H., Van De Loo, M., Poirel, L., Martinez- J. Vet. Pharmacol. Ther. 26: 81–93.
Martinez, L. and Nordmann, P. (2005). Emergence
of plasmid-mediated quinolone resistance in 49. Drlica, K., Zhao, X., Blondeau, J.M. and Hesje, C. (2006).
Escherichia coli in Europe. Antimicrob. Agents Chemother. Low correlation between MIC and mutant prevention con-
49: 71–6. centration. Antimicrob. Agents Chemother. 50: 403–404.

Antimicrobial treatment on selection of resistant bacteria 99

50. Balaban, N.Q., Merrin, J., Chait, R., Kowalik, L. and 66. Lees, P. and AliAbadi, F.S. (2002). Rational dosing of anti-
Leibler, S. (2004). Bacterial persistence as a phenotypic microbial drugs; animals versus humans. Int. J. Antimicrob.
switch. Science 305: 1622–5. Agents 19: 269–84.

51. Keren, I., Kaldalu, N., Spoering, A., Wang, Y. and Lewis, 67. McKellar, Q.A., Sanchez Bruni, S.F. and Jones, D.G. (2004).
K. (2004). Persister cells and tolerance to antimicrobials. Pharmacokinetic/pharmacodynamic relationships of
FEMS Microbiol. Lett. 230: 13–18. antimicrobial drugs used in veterinary medicine. J. Vet.
Pharmacol. Ther. 27: 503–14.
52. Levin, B.R. and Rozen, D.E. (2006). Non-inherited
antibiotic resistance. Nat. Rev. Microbiol. 4: 556–62. 68. Mouton, J.W., Dudley, M.N., Cars, O., Derendorf, H. and
Drusano, G.L. (2002). Standardization of pharmacoki-
53. Wiuff, C., Zappala, R.M., Regoes, R.R., Garner, K.N., netic/pharmacodynamic (PK/PD) terminology for anti-
Baquero, F. and Levin, B.R. (2005). Phenotypic toler- infective drugs. Int. J. Antimicrob. Agents 19: 355–8.
ance: antibiotic enrichment of noninherited resistance
in bacterial populations. Antimicrob. Agents Chemother. 69. Toutain, P.L. (2002). Pharmacokinetics/pharmacodynam-
49: 1483–94. ics integration in drug development and dosage regimen
optimisation for veterinary medicine. AAPS Pharmaceuti.
54. AliAbadi, F.S. and Lees, P. (2001). Pharmacokinetics and Sci. 4: 1–25, article 38.
pharmacodynamics of danofloxacin in serum and tissue
fluids of goats following intravenous and intramuscular 70. Toutain, P.L. (2003a). Pharmacokinetics/pharmacodynam-
administration. Am. J. Vet. Res. 62: 1979–89. ics integration in dosage regimen optimisation for veteri-
nary medicine. J. Vet. Pharmacol. Ther. 26 (Suppl 1): 1–8.
55. AliAbadi, F.S. and Lees, P. (2002). Pharmacokinetics
and pharmacokinetic/pharmacodynamic integration of 71. Toutain, P.L. (2003b). Antibiotic treatment of animals–a
marbofloxacin in calf serum, exudate and transudate. different approach to rational dosing. Vet. J. 165: 98–100.
J. Vet. Pharmacol. Ther. 25: 161–174.
72. Toutain, P.L., del Castillo, J.R. and Bousquet-Melou, A.
56. AliAbadi, F.S. and Lees, P. (2003). Pharmacokinetic– (2002). The pharmacokinetic–pharmacodynamic approach
pharmacodynamic integration of danofloxacin in the calf. to a rational dosage regimen for antibiotics. Res. Vet. Sci.
Res. Vet. Sci. 74: 247–59. 73: 105–14.

57. Aliabadi, F.S., Badrelin, H. Ali, Landoni, M.F. and Lees, 73. Craig, W.A. (1998). Pharmacokinetic/pharmacodynamic
P. (2003a). Pharmacokinetics and PK-PD modelling of parameters: rationale for antibacterial dosing of mice and
danofloxacin in camel serum and tissue cage fluids. men. Clin. Infect. Dis. 26: 1–12.
Vet. J. 165: 104–18.
74. Drusano, G.L., Johnson, D.E., Rosen, M. and Standiford,
58. AliAbadi, F.S., Landoni, M.F. and Lees, P. (2003b). H.C. (1993). Pharmacodynamics of a fluoroquinolone
Pharmacokinetics (PK) pharmacodynamics (PD) and antimicrobial agent in a neutropenic rat model of
PK–PD integration of danofloxacin in sheep biological Pseudomonas sepsis. Antimicrob. Agents Chemother. 37:
fluids. Antimicrob. Agents Chemother. 47: 626–35. 483–90.

59. Koritz, G.D., Kilroy, C.R. and Bevill, R.F. (1994). 75. Schentag, J.J. (2000). Clinical pharmacology of the fluo-
Pharmacokinetics–pharmacodynamic modelling of roquinolones: studies in human dynamic/kinetic models.
antibacterial therapy in vitro. In Proceedings of the 6th Clin. Infect. Dis. Suppl. 2: 540–544.
International Congress of the European Association for
Veterinary Pharmacology and Therapeutics. Edinburgh, 76. Sarasola, P., Lees, P., AliAbadi, F.S., et al. (2002).
UK, Blackwell Scientific Publications. Pharmacokinetic and pharmacodynamic profiles of dan-
ofloxacin administered by two dosing regimens in calves
60. Yancey, Jr., R.J., Evans, R.A., Kratzer, D.D. Paulissen, J.B. infected with Mannheimia (Pasteurella) haemolytica.
and Carmer, S.G. (1990). Efficacy of ceftiofur hydrochlo- Antimicrob. Agents Chemother. 46: 3013–9.
ride for treatment of experimentally induced colibacillosis
in neonatal swine. Am. J. Vet. Res. 51: 831–47. 77. Blaser, J., Stone, B.B., Groner, M.C. and Zinner, S.H.
(1987). Comparative study with enoxacin and netilmicin
61. Dalhoff, A. and Shalit, I. (2003). Immunomodulatory in a pharmacodynamic model to determine importance
effects of quinolones. Lancet Infect. Dis. 3: 359–71. of ratio of antibiotic peak concentrations to MIC for bac-
terial activity and emergence of resistance. Antimicrob.
62. Hoyt, J.C. and Robbins, R.A. (2001).Macrolide antibiot- Agents Chemother. 31: 1054–60.
ics and pulmonary inflammation. FEMS Microbiol. Lett.
205: 1–7. 78. Marchbanks, C.R., McKiel, J.R., Gilbert, D.H., et al. (1993).
Dose ranging and fractionation of intravenous ciproflox-
63. Ianaro, A., Ialenti, A., Maffia, P., et al. (2000). Anti-inflam- acin against Pseudomonas aeruginosa and Staphylococcus
matory activity of macrolide antibiotics. J. Pharmacol. aureus in an in vitro model of infection. Antimicrob. Agents
Exp. Ther. 292: 156–63. Chemother. 37: 1756–63.

64. Lindecrona, R.H., Friis, C. and Nielsen, J.P. (2000). 79. Bugyei, K., Black, W.D. and McEwen, S. (1999).
Pharmacokinetics and penetration of danofloxacin into Pharmacokinetics of enrofloxacin given by the oral, intra-
the gastrointestinal tract in healthy and inSalmonella venous and intramuscular routes in broiler chickens.
typhimurium infected pigs. Res. Vet. Sci. 68: 211–6. Can. J. Vet. Res. 63: 193–200.

65. Lees, P. and AliAbadi, F.S. (2000). Rationalising dosage 80. Ding, H.Z., Zeng, Z.L., Fung, K.F., Chen, Z.L. and Qiao,
regimens of antimicrobial drugs; a pharmacological per- G.L. (2001). Pharmacokinetics of sarafloxacin in pigs
spective. J. Med. Microbiol. 49: 943–5.

100 Guide to antimicrobial use in animals

and broilers following intravenous, intramuscular, and 94. Benchaoui, H.A., Nowakowski, M., Sheripngton, J.,
oral single-dose applications. J. Vet. Pharmacol. Ther. 24: Rowan,T.G.andSunderland,S.J.(2004).Pharmacokinetics
303–308. and lung tissue concentrations of tulathromycin in swine.
81. Fernandez-Varon, E., Bovaira, M.J., Espuny, A., J. Vet. Pharmacol. Ther. 27: 203–10.
Escudero, E., Vancraeynest, D. and Carceles, C.M. (2005).
Pharmacokinetic–pharmacodynamic integration of mox- 95. Nowakowski, M.A., Inskeep, P.B., Risk, J.E., et al. (2004).
ifloxacin in rabbits after intravenous, intramuscular and Pharmacokinetics and lung tissue concentrations of
oral administration. J. Vet. Pharmacol. Ther. 28: 343–8. tulathromycin, a new triamilide antibiotic, in cattle. Vet.
82. Intorre, L., Mengozzi, G., Bertini, S., Bagliacca, M., Ther. 5: 60–74.
Luchetti, E. and Soldani, G. (1997). The plasma kinetics
and tissue distribution of enrofloxacin and its metabolite 96. Goswami, S.K., Kivity, S. and Marom, Z. (1990).
ciprofloxacin in the Muscovy duck. Vet. Res. Commun. Erythromycin inhibits respiratory glycoconjugate secre-
21: 127–36. tion from human airways in vitro. Am. Rev. Resp. Dis. 141:
83. Wiuff, C., Lykkesfeldt, J., Svendsen, O. and Aarestrup, F.M. 72–8.
(2003). The effects of oral and intramuscular administra-
tion and dose escalation of enrofloxacin on the selection 97. Morikawa, K., Oseko, F., Morikawa, S. and Iwamoto, K.
of quinolone resistance among Salmonella and coliforms (1994). Immunomodulatory effects of three macrolides,
in pigs. Res. Vet. Sci. 75: 185–93. midecamycin acetate, josamycin and clarithromycin,
84. Fernandez-Varon, E., Ayala, I., Marin, P., et al. (2006). on human T-lymphocyte function in vitro. Antimicrob.
Pharmacokinetics of danofloxacin in horses after intra- Agents Chemother. 38: 2643–7.
venous, intramuscular and intragastric administration.
Equine Vet. J. 38: 342–6. 98. Roche, Y., Gougerot-Pocidalo, M.A., Fay, M., Forest, N.
85. Giles, C.J., Magonigle, R.A., Grimshaw, W.T., et al. (1991). and Pocidalo, J.J. (1986). Macrolides and immunity:
Clinical pharmacokinetics of parenterally administered effects of erythromycin and spiramycin on human
danofloxacin in cattle. J. Vet. Pharmacol. Ther. 14: 400–10. mononuclear cell proliferation. J. Antimicrob. Chemother.
86. Barker, C.W., Zhang, W., Sanchez, S., Budsberg, S.C., 17: 195–203.
Boudinot, F.D. and McCrackin Stevenson, M.A. (2003).
Pharmacokinetics of imipenem in dogs. Am. J. Vet. Res. 99. Takeyama, K., Tamaoki, J., Chiyotani, A., Tagaya, E. and
64: 694–9. Konno, K. (1993). Effect of macrolide antibiotics on cili-
87. Bidgood, T. and Papich, M.G. (2002). Plasma pharmacoki- ary motility in rabbit airway epithelium in vitro. J. Pharm.
netics and tissue fluid concentrations of meropenem after Pharmacol. 45: 756–8.
intravenous and subcutaneous administration in dogs.
Am. J. Vet. Res. 63: 1622–8. 100. Takizawa, H., Desaki, M., Ohtoshi, T., et al. (1995).
88. Guyonnet, J., Monnoyer, S., Manco, B., Aliabadi, F.S. and Erythromycin suppresses interleukin 6 expression by
Lees, P. (2003). In vivo pharmacokinetics and in vitro phar- human bronchial epithelial cells: a potential mechanism
macodynamics as a basis for predicting dosage of colistin of its anti-inflammatory action. Biochem. Biophys. Res.
in piglet g.i.t. disease. J. Vet. Pharmacol. Ther. 26 Suppl. 1: Comm. 210: 781–6.
148–9.
89. Lees, P., Aliabadi F.S. and Toutain, P-L. (2004). PK–PD 101. Tamaoki, J., Noritaka, S., Tagaya, E. and Konno, K. (1994).
modelling: an alternative to dose titration studies for Macrolide antibiotics protect against endotoxin-induced
antimicrobial drug dosage selection. J. Reg. Affairs 15: vascular leakage and neutrophil accumulation in rat
175–80. trachea. Antimicrob. Agents Chemother. 38: 1641–3.
90. Toutain, P.L. and Lees, P. (2004). Integration and model-
ling of pharmacokinetic and pharmacodynamic data to 102. Umeki, S. (1993). Anti-inflammatory action of erythro-
optimise dosage regimens in veterinary medicine. J. Vet. mycin: its inhibitory effect on neutrophil NADPH oxi-
Pharmacol. Ther. 27: 467–77. dase activity. Chest 104: 1191–1193.
91. Drusano, G.L. (2004). Antimicrobial pharmacodynamics:
critical interactions of ‘bug and drug’. Nat. Rev. Microbiol. 103. Delissalde, F. and Amabile-Cuevas, C.F. (2004).
2: 289–300. Comparison of antibiotic susceptibility and plasmid
92. Mouton, J.W. (2005). Impact of pharmacodynamics on content, between biofilm producing and non-producing
dosing schedules: optimising efficacy, reducing resistance, clinical isolates of Pseudomonas aeruginosa. Int. J.
and detection of emergence of resistance. In Antibiotic Antimicrob. Agents. 24: 405–8.
Policies Theory and Practice (Eds. Gould, I.M. and van
der Meer, J.W.M.). Kluwer Academic/Plenum Publishers, 104. Drenkard, E. and Ausubel, F.M. (2002). Pseudomonas
New York, pp. 387–407. biofilm formation and antibiotic resistance are linked to
93. Mazzei, T. and Novelli, A. (1999). How macrolide pharma- phenotypic variation. Nature 416: 740–3.
codynamics affect bacterial killing. Infect. Med. 16: 22–8.
105. Gilbert, P., McBain, A. and Rickard, A.H. (2003). Biofilms
and bacterial multi-resistance. In Multiple Drug Resistant
Bacteria (ed. Amabile-Cuevas, C.F.). Horizon Scientific
Press, Wymondham.

106. Oliver, A., Canton, R., Campo, P., Baquero, F. and
Blazquez, J. (2000). High frequency of hypermutable
Pseudomonas aeruginosa in cystic fibrosis lung infection.
Science 288: 1251–3.

107. Preston, S.L., Drusano, G.L., Berman, A.L., et al. (1998).
Pharmacodynamics of levofloxacin: a new paradigm for
early clinical trials. J. Am. Med. Assoc. 279: 125–9.

Antimicrobial treatment on selection of resistant bacteria 101

108. Thorburn, C.E. and Edwards, D.I. (2001). The effect 116. Moore, R.D., Smith, C.R. and Lietman, P.S. (1984).
of pharmacokinetics on the bactericidal activity of Association of aminoglycoside plasma levels with thera-
ciprofloxacin and sparfloxacin against Streptococcus peutic outcome in Gram-negative pneumonia. Am. J.
pneumoniae and the emergence of resistance. J. Med. 77: 657–62.
Antimicrob. Chemother. 48: 15–22.
117. Stearne, L.E., Lemmens, N., Goessens, W.H.F., Mouton,
109. Michae-Hamzehpour, M., Auckenthaler, R., Regamey, J.W. and Gyssens, I.C. (2002). In European Conference
P. and Pechere, J.C. (1987). Resistance occurring after Clinical Microbiology and Infectious Diseases, Milan.
fluoroquinolone therapy of experimental Pseudomonas
aeruginosa peritonitis. Antimicrob. Agents Chemother. 118. Odenholt, I., Gustafsson, I., Lowdin, E. and Cars, O.
31: 1803–1808. (2003). Suboptimal antibiotic dosage as a risk factor for
selection of penicillin-resistant Streptococcus pneumoniae:
110. Thomas, J.K., Forrest, A., Bhaveni, S.M., et al. (1998). in vitro kinetic model. Antimicrob. Agents Chemother.
Pharmacodynamic evaluation of factors associated with 47: 518–23.
the development of bacterial resistance in acutely ill
patients during therapy. Antimicrob. Agents Chemother. 119. Tam, V.H., Schilling, A.N., Neshat, S., Poole, K.,
42: 521–7. Melnick, D.A. and Coyle, E.A. (2005b). Optimization
of meropenem minimum concentration/MIC ratio to
111. Forrest,A.,Nix,D.E.,Ballow,C.H.,Goss,T.F.,Birmingham, suppress in vitro resistance of Pseudomonas aeruginosa.
M.C. and Schentag, J.J. (1993). Pharmacodynamics Antimicrob. Agents Chemother. 49: 4920–7.
of intravenous ciprofloxacin in seriously ill patients.
Antimicrob. Agents Chemother. 37: 1073–81. 120. Lees, P., Concordet, D., Aliabadi, F.S. and Toutain, P.-L.
(2006). Drug selection and optimization of dosage
112. Firsov, A.A., Vostrov, S.N. Lubenko, I.Y., Drlica, K., schedules to minimize antimicrobial resistance. In
Portnoy, Y.A. and Zinner, S.H. (2003). In vitro phar- Antimicrobial Resistance in Bacteria of Animal Origin
macodynamic evaluation of the mutant selection win- (ed. Frank, M.). Aerestrup, ASM Press, Washington, D.C.
dow hypothesis using four fluoroquinolones against pp. 49–71.
Staphylococcus aureus. Antimicrob. Agents Chemother. 47:
1604–13. 121. Regnier, A., Concordet, D., Schneider, M., Boisrame, B.
and Toutain, P.L. (2003). Population pharmacokinetics
113. Tam, V.H., Louie, A., Deziel, M.R., Liu, W., Leary, R. and of marbofloxacin in the aqueous humour after
Drusano, G.L. (2005a). Bacterial-population responses intravenous administration in dogs. Am. J. Vet. Res.
to drug-selective pressure: examination of Garenoxacin’s 64: 889–93.
effect on Pseudomonas aeruginosa. J. Infect. Dis. 192:
420–8. 122. del Castillo, J.R., Laroute, V., Pommier, P., et al. (2006).
Interindividual variability in plasma concentrations
114. Jumbe, N., Louie, A., Leary, R., et al. (2003). Application after systemic exposure of swine to dietary
of a mathematical model to prevent in vivo amplifica- doxycycline supplied with and without paracetamol:
tion of antibiotic-resistant bacterial populations during a population pharmacokinetic approach. J. Anim. Sci.
therapy. J. Clin. Invest. 112: 275–85. 84: 3155–66.

115. Charleston, B., Gate, J.J., Aitken, I.A., Stephan, B. and 123. Riond, J.L. and Riviere, J.E. (1989). Effects of tetracyclines
Froyman, R. (1998). Comparison of the efficacies of three on the kidney in cattle and dogs. J. Am. Vet. Med. Assoc.
fluoroquinolone antimicrobial agents, given as continu- 195: 995–7.
ous or polsed-water medication against Escherichia coli
infection in chickens. Antimicrob. Agents Chemother. 124. MacGowan, A.P. (2004). Elements of design: the knowl-
42: 83–7. edge on which we build. Clin. Microbiol. Infect. 10 Suppl
2: 6–11.

7 Swine Chapter 7

GUIDELINES FOR ANTIMICROBIAL
USE IN SWINE

David G. S. Burch, C. Oliver Duran and Frank M. Aarestrup

The demand for meat increased substantially during exception is the USA where 50% of the $1.3 billion
the post-war years and increased pig production antimicrobial market was in cattle, primarily due to
was the main driver of the industry. The increased the feedlot system, and only 20% in pigs ($0.23 bil-
demand, coupled with a decrease in people engaged lion). Precise figures regarding tonnages of active
in agriculture, has led to intensified and more effi- ingredient and actual use in pigs are rarely published,
cient production. Because of these socio-economic but national bodies are starting to collate total antimi-
changes, over the last 30 years there has been a steady crobial usage in animals and some, like Denmark, can
decline of small farms and an increase in the larger break them down by family class of antimicrobial (4, 5)
ones, particularly in the USA. In 2005, 53% of all hog and the species of animal in which they are used (6).
inventories were in farms with more than 5000 pigs The Danish antimicrobial market is not completely
(1). More recently, the development of large corpo- representative of the worldwide swine industry, as
rate farms and production systems has led to further total antimicrobial usage is relatively low compared
concentration in the ownership of pigs. Currently, to other countries and usage in pigs accounts for over
three large companies produce 20% of pigs in the UK, 80% of all animal consumption in terms of kg active
while in the USA the 20 biggest companies own one- compounds. However, tetracyclines dominate in most
third of all the breeding sows (2). These changes have markets, followed by the macrolide/lincosamide/
resulted in increased disease challenges due to larger pleuromutilin group of compounds. The use of peni-
units, increased population density and throughput cillins, trimethoprim/sulfonamide combinations and
and to a certain extent a reduction in the quality of aminoglycosides is also important (see Figure 7.1).
stockmanship. As a result, they have frequently con-
tributed to increased use of antimicrobials in pork Total use of antimicrobials in Europe has prob-
production to compensate. ably been reduced following the ban on use of
Antimicrobial Growth Promoters (AGPs) in 2006
Antimicrobials have been widely used in swine pro- (see Chapter 1). Recent Scandinavian data showed
duction over several decades and are reported to be an overall reduction in total veterinary antimicro-
worth an estimated $1.7 billion dollars or 34% of the bial usage (7). In Denmark, an increase in the use of
global animal health antimicrobial market, closely fol- antimicrobials used for therapy in recently weaned
lowed by poultry (33%) and cattle (26%) (3). A major and grower pigs due to Escherichia coli and Lawsonia

Guide to Antimicrobial Use in Animals. Edited by Luca Guardabassi, Lars B. Jensen and Hilde Kruse
© 2008 Blackwell Publishing Ltd. ISBN: 978-1-4051-5079-8

Guidelines for antimicrobial use in swine 103

35 PMWS diagnosed 30
GP ban 25
Antimicrobial (tonnes) 20 Pigs killed (millions)
30 15
10 7 Swine
25 5 Figure 7.1 Therapeutic antimicrobial
usage (tetracyclines, trimethoprim/sul-
20 fonamides, macrolides/lincosamides/
pleuromutilins (MLT) and aminoglyco-
15 sides) in Denmark after the withdrawal
of growth promoters in 1999 (9) and
10 number of pigs killed/year (millions).

5

0 1999 2000 2001 2002 2003 2004 0
1998 TMP/S 2005

Tetracyclines M,L,T Aminoglycosides Pigs killed

100

Clinical incidence/risk (%) 80

Wean Move PHE

60 Move

40

20

0 20 24
L. intracellularis
0 4 8 12 16
Weeks

E. coli B. hyodysenteriae B. pilosicoli Figure 7.2 Enteric disease patterns
in swine.

intracellularis respectively, was observed shortly after to, and causing infections in, humans, such as
the ban of AGPs. However, the situation has stabilised Salmonella and Campylobacter. Thus, the use of anti-
following changes in management practices by farm- microbials in pigs also leads to selection of resistance
ers and veterinarians (8, 9). The appearance of severe in these zoonotic bacteria and thereby potentially
clinical disease related to infections with Porcine complicates treatment of human infections. This
Circovirus Associated Diseases (PCVAD) from 2000 aspect has to be taken into account when choosing
has also resulted in increased antimicrobial use to antimicrobials for treatment of bacterial infections in
combat secondary bacterial infections, and it is diffi- pigs and in other food animals. The purpose of this
cult to separate this effect from that of the AGPs ban. chapter is to describe the current use of antimicrobial
agents in swine production and to suggest possible
The most common bacterial pathogens and diseases strategies to reduce their overall use and to use them
requiring antimicrobial use in swine are summarised more effectively, prudently and responsibly.
in Table 7.1 and their disease patterns are highlighted
in Figures 7.2 (enteric diseases), 7.3 (respiratory dis- 7.1 Antimicrobial usage in swine
eases) and 7.4 (septicaemic diseases). Any use of anti- production
microbial agents leads to development of bacterial
resistance. Resistance development in swine patho- The antimicrobial compounds most commonly used
genic bacteria complicates treatment of infections and in pig production are described in Table 7.2 together
therefore has to be regarded as both an animal health with their modes of administration, dosage rates and
problem and an economic burden. In addition, pigs
are often colonised by bacteria capable of transferring

104 Guide to antimicrobial use in animals

Table 7.1 Common bacterial pathogens and diseases in swine

7 Swine Bacterial species Disease Age

Enteric Neonatal scours. 1–3 days.
Escherichia coli Piglet scours. 7–14 days.
Post-weaning diarrhoea. 5–14 days after weaning.
Clostridium perfringens Type C – necrotic enteritis. 1–7 days.
Clostridium difficile Type A – diarrhoea. 10–21 days, weaned pigs.
Salmonella spp. Diarrhoea, ill thrift. 3–7 days.
Typhimurium – occasional Grower pigs 6–16 weeks.
Lawsonia intracellularis diarrhoea, septicaemia, death.
Derby – occasional diarrhoea. Grower pigs 6–16 weeks.
Choleraesuis – septicaemia Finishing pigs 12–16 weeks.
diarrhoea, death.
Porcine proliferative enteropathy Grower pigs.
(ileitis).
Regional/necrotic ileitis. Grower pigs.
Porcine haemorrhagic enteropathy. Finishing pigs and young adults
16–40 weeks.
Brachyspira hyodysenteriae Swine dysentery. Growers and finishers, 6–26 weeks.
All ages in primary breakdown.
Brachyspira pilosicoli Intestinal spirochaetosis ‘colitis’. Grower pigs.
Respiratory
Pasteurella multocida (D) Atrophic rhinitis. 1–8 weeks.
Bordetella bronchiseptica Nasal distortion, lasts for life.
Mycoplasma hyopneumoniae Enzootic pneumonia. Grower and finisher pig.
Pasteurella multocida Mycoplasma-induced respiratory Grower and finisher – secondary invader.
disease (MIRD).
Actinobacillus pleuropneumoniae Pleuro-pneumonia. Grower and finisher – MDA last for
10 weeks.
Actinobacillus suis Septicaemia. 5–28 days.
Pleuro-pneumonia. Weaning to slaughter.
Septicaemic/bacteraemic
Streptococcus suis Meningitis, endocarditis, arthritis, 2–10 weeks.
peritonitis.
Haemophilus parasuis Glässer’s disease (arthritis, 2–10 weeks.
pericarditis, peritonitis).
Mycoplasma hyosynoviae Mycoplasmal arthritis. 16 weeks plus.
Erysipelothrix rhusiopathiae Erysipelas (dermatitis, arthritis, Growers, finishers and sows.
endocarditis).

indications. Their relative activity against various six weeks of age) to prevent post-weaning diarrhoea,
porcine pathogens is highlighted in Tables 7.3–7.12. or in the mid to late nursery phase (in the USA, six to
Antimicrobials are used in swine production for ten weeks of age) or after moving and mixing of pigs.
therapeutic, metaphylactic and prophylactic purposes Only occasionally is prophylaxis used in the finishing
as well as for growth promotion, although the latter stage after initial placement in the barn (10 or 12 weeks
use is banned in EU countries and in Australia (see to slaughter), where medication becomes very costly,
Chapter 1 for definitions on therapy, metaphylaxis unless pigs from a number of sources are co-mingled.
and prophylaxis). Prophylactic treatments coincide Feed traditionally lends itself well to prophylactic/meta-
with defined times in the production cycle, occasion- phylactic medication of pigs, as it can be built easily into
ally at birth to reduce Streptococcus and Haemophilus a medication programme for disease control, without
transmission, frequently after weaning (say three to having to physically handle the animal.

Guidelines for antimicrobial use in swine 105

60

Clinical incidence/risk (%) 50 Move 7 Swine

Atrophic rhinitis
40

30

Move
20

10

0 16 20 24
0 4 8 12 A. pleuropneumoniae
Figure 7.3 Respiratory disease
Age (weeks) patterns in swine.

M. hyopneumoniae P. multocida

Clinical incidence/risk (%) 100 Stable PRRS
90
80 8 12 16 20 24
70 Age (weeks)
60 Unstable PRRS
50 Strep suis H. parasuis S. Typhimurium
40
30
20
10
0
04

PRRS

Figure 7.4 Septicaemic disease patterns in swine.

In growth promotion practice, antimicrobials can deviate from the label, not even veterinarians.
are usually included in feed at low concentrations In Europe, since the banning of AGPs, antimicro-
that preclude systemic disease-controlling effects. bial premixes for medicated feeding stuffs (MFS) are
Particularly in the USA, antimicrobial feed premixes supplied under a veterinary controlled MFS prescrip-
may be approved for treatment,prevention and growth tion for prevention or treatment purposes, although
promotion depending on the dose administered. For the mills generally purchase and include the pre-
example, tiamulin was licensed for inclusion in feed at mixes in the feed. In some countries, like Denmark,
220 ppm for treatment, 38.5 ppm for prevention and the medicines are supplied via the pharmacist,
11ppm as a growth promoter respectively (10). Some although this is changing as it was considered too
products, like tylosin, are approved at 44–110 ppm restrictive and uncompetitive. Interestingly, in the
for prevention and treatment, but also at 11–110 ppm USA, only one antimicrobial in feed, tilmicosin,
for growth promotion depending on the diet. None currently requires the equivalent of a veterinary pre-
of these dosages require a veterinary prescription, scription or what is called a Veterinary Feed Directive
but the inclusion levels strictly follow Food and Drug (11). All others can be included in rations at the dis-
Administration (FDA) label instructions. Nobody cretion of the farmer, nutritionist or mill manager.

106 Guide to antimicrobial use in animals

Table 7.2 Routes of administration, dosages (mg/kg bodyweight) and target pathogens of antimicrobials used in swine

7 Swine Administration and dosage

Antimicrobial class/ Injection In water In feed Target pathogens
compound

Tetracyclines 10 (LA forms 20–30) 10–30 20 M. hyopneumoniae
Oxytetracycline — 20 10–20 P. multocida
Chlortetracycline — 20–40 – A. pleuropneumoniae
Tetracycline 4–6 12.5 12.5 H. parasuis
Doxycycline L. intracellularis
15 30 15 E. coli (resistance)
Trimethoprim/ Salmonella spp. (resistance)
sulfonamide
P. multocida
Penicillins 10 (LA form 15) — — B. bronchiseptica
Penicillin G 10 10 A. pleuropneumoniae
Penicillin V S. suis
15–20 S. hyicus
Synthetic penicillins 7 (LA forms 15) 20 — H. parasuis
Amoxicillin 7.5 — — E. coli
Ampicillin +1.75 — Salmonella spp.
plus clavulanic acid
S. suis
P. multocida
H. parasuis
A. pleuropneumoniae
A. pyogenes
C. perfringens
E. rhusiopathiae

S. suis
P. multocida
H. parasuis
A. pleuropneumoniae
A. pyogenes
C. perfringens
E. rhusiopathiae
E. coli
Salmonella spp.

Cephalosporins 7 — — S. suis
Cefalexin 3 (LA forms 5) — — P. multocida
Ceftiofur 1–2 — — H. parasuis
Cefquinome A. pleuropneumoniae
A. pyogenes
Fluoroquinolones 2.5 —— C. perfringens
Enrofloxacin 1.25 —— E. rhusiopathiae
Danofloxacin 2 —— E. coli
Marbofloxacin Salmonella spp.
10–30 — 10
Thiamphenicols 15 15 15 M. hyopneumoniae
Thiamphenicol P. multocida
Florfenicol A. pleuropneumoniae
H. parasuis
E. coli
Salmonella spp.

P. multocida
A. pleuropneumoniae
H. parasuis
S. suis
B. bronchiseptica

Continued

Guidelines for antimicrobial use in swine 107

Table 7.2 (Continued)

Administration and dosage 7 Swine

Antimicrobial class/ Injection In water In feed Target pathogens
compound
25 — — E. coli
Aminoglycosides – (NA) 11 11 Salmonella spp.
Streptomycin – (NA) 7.5–12.5 4–8
Neomycin – (NA) — — E. coli
Apramycin — — Salmonella spp.
Gentamicin – (NA) M. hyopneumoniae
Amikacin 10–50 1.1–2.2 L. intracellularis
Aminocyclitol — B. hyodysenteriae (resistance
Spectinomycin 50 000 IU 50 000 IU
2–10 25 tylosin)
(+ lincomycin) 3–6 (T) B. pilosicoli (resistance tylosin)
1.2–2.4 (P)
Polymixin Plus A. pleuropneumoniae
Colistin H. parasuis
P. multocida
Macrolides S. suis (resistance)
Tylosin M. hyopneumoniae
M. hyosynoviae
Acetyliso-valeryltylosin — — 2.5–5 L. intracellularis
— — 8–16 B. hyodysenteriae
Tilmicosin 2.5 (LA form) — — B. pilosicoli
Triamilide
Tulathromycin 10 4.5 5.5–11 (T) M. hyopneumoniae
2.2 (P) M. hyosynoviae
Lincosamides — L. intracellularis
Lincomycin 3.75–10 (T) B. hyodysenteriae
8.8 1–1.5 (P) B. pilosicoli
Pleuromutilins — Plus A. pleuropneumoniae
Valnemulin — 5–11 (T)
— 1.5–2 (P) C. perfringens
Tiamulin 10–15 — B. hyodysenteriae
— 10–40 ppm
Miscellaneous — — 5–100 ppm B. hyodysenteriae
Growth promoters (not EU) — — 10–250 ppm S. Choleraesuis
Avoparcin — 2–4 ppm
Virginiamycin — 20 (OD) 10–40 ppm Isospora suis
Bacitracin MD — — 11–55 ppm
Flavophospholipol — —
Avilamycin —
Carbadox 15–60 ppm
100 ppm (NA)
Anticoccidials —
Toltrazuril —
Salinomycin —
Monensin

NA – not approved; LA form – long-acting formulation; OD – oral doser; T – treatment; P – prevention; ppm – parts per million.

108 Guide to antimicrobial use in animals

Table 7.3 Occurrence of antimicrobial resistance among Salmonella enterica serovar Typhimurium from swine in
different European countries in year 2004

7 Swine Country, number and origin of isolates

The England/
Netherlands Belgium Denmark Germany Poland Wales Italy

Origin and antimicrobial agent n=77 n=175 n=814 n=299 n=10 n=147 n=216

Amoxicillin 51.9 78.6 40 82 66.2
Ampicillin a 84 22.2 1.3 0 1
0 4
Amoxicillin + clavulanic acid 0 0 0
46.2 71 32.6
Apramycin 1.4 30 0 0.5
0 10 0.9
Ceftiofur 20
44.2 30 3 1.4
Chloramphenicol 32.5 63 9.3 6.0 0 0.5

Ciprofloxacin 0 0.8 14.1 20 4 10.2
0 10 2.5
Enrofloxacin 0 1.7 80 70.8
14.1 40 90 81.9
Florfenicol 31.2 53 5.3 82.6 40
90.0 93 80.6
Gentamicin 0 0 1.4 40 20.3
77.9 0
Kanamycin 26.8 72
26.8
Flumequin

Nalidixic acid 1.3 4 0.9

Neomycin 0 2 7.8

Streptomycin 73 37.3

Sulfonamides 67.5 91 37.7

Doxycycline

Tetracycline 72.7 89 39.9

Trimethoprim 29.9 6.3

Trimethoprim + sulfonamides 36

a Indicates that this antimicrobial was not tested for in that country.

No extra-label usage of in feed antimicrobials is Injections of all pigs may also occur at weaning as
allowed by the FDA (12). part of a Medicated Early Weaning health protocol to
eliminate certain bacterial infections from the growing
Therapeutic antimicrobials may also be given in population (13). Individual older pigs are injected when
feed, but because sick pigs have a reduced appetite and they are ill, but for therapy usually only after removal
variable feed intake, and medicated feed may also take to a hospital pen. A sick pig is easier to inject in a pen
time to prepare and deliver, they are commonly given but, as they recover, it becomes more difficult. Injecting
as soluble formulations in the drinking water to groups unrestrained large sows or boars can be dangerous. Due
or houses of animals. This is becoming increasingly to these risks and the reduced availability of labour,
popular with the development of more reliable dos- long-acting injectable formulations have been developed
ing/water-proportioner machines rather than using to facilitate administration and to reduce the need for
header tanks. Injectables and piglet dosers are widely repeated injections. While these preparations improve
used, but require individual handling of the animal. compliance by removing the need for repeated injec-
When they are small piglets are easy to handle, so tions, if products are used that do not reach therapeutic
metaphylactic injection programmes of antimicro- concentrations for the target pathogens this may lead to
bials are quite common to fight off piglet infections therapy failure or encourage resistance development.
such as neonatal scours, navel infections or arthritis.

Table 7.4 Occurrence of antimicrobial resistance among Escherichia coli isolated from swine in different European countries in year 2004 (source www.arbao.org)

Country, number and origin of isolates

Origin and Spain Belgium Denmark Latvia England/ Portugal Switzerland
antimicrobial 169 137 177 19 Latvia Norway Finland Belgium Sweden Wales France 44 47
agent
11 45 61 105 386 313 758–1412

SWINE Passive Diseased O149 Infected Faeces Enteritis/ Infections Diarrhoea Intestinal, Disease Diagnostic Origin Intestine

monitoring animals animals oedema (Young diagno- outbre- laboratories/ of the K88 Guidelines for antimicrobial use in swine

programme disease animals) stic mat aks pathogenic isolates

bacteria infected

animal

Amoxicillin 33.7 a 4.2
Ampicillin
72.3 45.8 75 100 7 16 61.9 22 47 53.2
0.7 0.6
Amoxicillin + Cl 1.8 15.8 0 1 2.0 36 0.0
13.1 13.6 8 3.3 45 38.3
Apramycin 13 0.7 0 1.9
0.6
Ceftiofur 38.7 41.8 0 0 1.0 <1

Chloramphe- 40.8 2.9 0 0 4.0 7 29.5
4.4
nicol 3.6 0
12
Ciprofloxacin 14.2 00 0.0
30
Enrofloxacin 0 0 13.3 6 2 5.5
45 12.7
Florfenicol 7.1 0 0 3.8 0.9 27
41
Gentamicin 19.5 9 0 0 0 6.7 0 5.5
64
Kanamycin 33.3 63.6
91
Flumequin 18.6 98 57.4

Nalidixic acid 33.7 34.3 32 23 0 2 13 18.1
1.5 35
Neomycin 20.1 77.4 39 16.7 2 7 1.0 4 11 10.9
82
Streptomycin 74 50 100 47 54 28
77.4 91
Sulfonamides 76.3 48.6 7 51

Doxycycline 78.2 70.8 100

Tetracycline 87 82 100 24 51 73.3 27 82 82.6

Trimethoprim 66.9 7 44 27

Trimethoprim + 55.2 55 66.4

sulfonamides

aIndicates that this antimicrobial was not tested for in that country.

109

7 Swine

110 Guide to antimicrobial use in animals

Table 7.5 Minimum inhibitory concentrations (MICs) of antimicrobial agents against B. hyodysenteriae, B. pilosicoli,
C. perfringens, C. difficile and intracellular inhibitory concentrations (90% inhibition) for L. intracellularis

7 Swine Organism/antimicrobial MIC 50 (μg/ml) MIC 90 (μg/ml) Range (μg/ml)

B. hyodysenteriae – Australia (76 isolates) (63) 0.031 0.5 ≤0.016–2.0
Valnemulin 0.125 1.0 ≤0.016–2.0
Tiamulin 16 64 ≤1.0–64
Lincomycin >256 >256 ≤2.0>256
Tylosin
B. hyodysenteriae – Czech Republic (100 isolates) (64) 0.125 4.0 —
Valnemulin 0.25 2.0 —
Tiamulin 32 64 —
Lincomycin >128 >128 —
Tylosin 25 50 —
Acetylisovaleryltylosin 4 8 —
Chlortetracycline
B. pilosicoli – N. America (25 isolates) (65) 0.06 0.5 0.03–2.0
Valnemulin 0.125 1.0 0.06–8.0
Tiamulin 32 64 4.0–>128
Lincomycin >512 >512 <16–>512
Tylosin 0.06 0.06 0.03–0.125
Carbadox
L. intracellularis – UK (1–3 isolates) (55)a — — <1.0–<4.0
Valnemulin — — <2.0–<8.0
Tiamulin — — <0.25
Lincomycin — — <2.0–<100
Tylosin — — <0.125–<0.5
Tilmicosin — — <1.0–<16
Chlortetracycline — — <64
Spectinomycin
L. intracellularis – USA (4 isolates) EU (4 isolates) (56)a — — 0.125–0.25
Carbadox — — 0.25–64
Chlortetracycline — — 8–>128
Lincomycin — — 0.125–0.5
Tiamulin — — 0.25–32
Tylosin — — 0.125
Valnemulin
C. perfringens – Belgium (58 isolates) (66) — – 0.25–4.0
Tiamulin 2.0 256 0.12–>512
Lincomycin 0.12 0.5 0.06–1.0
Penicillin G 16 32 0.06–>64
Tetracycline
C. perfringens - Belgium (95 isolates) (ref 67) ≤0.12 ≤0.25 ≤0.12–>64
Tylosin
C. difficile – USA (80 isolates) (57) >256 >256 NR
Bacitracin 256 >256 NR
Ceftiofur 0.5 >256 NR
Erythromycin 8 32 NR
Tetracycline 4 8 NR
Tiamulin 0.5 >256 NR
Tilmicosin 0.25 64 NR
Tylosin 0.25 2 NR
Virginiamycin

a Based on determination of intracellular MICs.

Guidelines for antimicrobial use in swine 111

Table 7.6 Minimum inhibitory concentrations (MICs) of antimicrobial agents against M. hyopneumoniae and
M. hyosynoviae

MIC 50 (μg/ml) MIC 90 (μg/ml) Range (μg/ml) 7 Swine

M. hyopneumoniae – Hannan et al., 1997 (68) – 0.025 0.05 0.01–0.1
Global (20 isolates) 0.05 0.05 0.01–0.1
Enrofloxacin 0.1 0.25 0.025–0.25
Tiamulin 0.25 1.0 0.025–1.0
Tylosin
Oxytetracycline 0.03 0.5 0.015–>1.0
M. hyopneumoniae – Vicca et al., 2004 (69) – ≤0.015 0.12 ≤0.015–0.12
Belgium (21 isolates) 0.03 0.06 ≤0.015–>1.0
Enrofloxacin 0.25 0.5 ≤0..25–>16
Tiamulin ≤0.06 ≤0.06 ≤0.06–>8.0
Tylosin 0.25 0.5 ≤0.12–0.5
Tilmicosin 0.12 1.0 0.03–2.0
Lincomycin 0.12 0.5 0.03–1.0
Spectinomycin ≤0.12 0.25 ≤0.12–0.5
Oxytetracycline
Doxycycline 0.1 0.25 0.05–0.25
Florfenicol 0.005 0.025 0.0025–0.01
M. hyosynoviae – Hannan et al., 1997 (68) – 0.25 1.0 0.025–>10
Global (18 isolates) 0.5 5.0 0.25–10
Enrofloxacin
Tiamulin
Tylosin
Oxytetracycline

Table 7.7 Antimicrobial resistance in porcine bacterial pathogens in the UK (59)

Antimicrobial P. multocida A. pleuropneumoniae S. suis E. rhusiopathiae A. pyogenes

No. of isolates 573 201 92 44 69
Ampicillin 3 6 00 0
Penicillin – – 00 0
Tetracycline 12 31 75 33 1
Trimethoprim/sulfonamide 8 14 9 59 6
Enrofloxacin 0 1 0– 0
Ceftiofur – 1 00 0

The pharmacokinetics of antimicrobials is discussed in male castrates especially, and there can be a halving
in Chapter 6. However, there are a number of key of relative feed intake to 2.5% from 80 kg and above.
aspects to swine medicine that are both important Lactating sows are usually fed to about 2.5% and
and specific. The bulk of antimicrobials used in swine dry sows can be fed at a rate of 1% of bodyweight.
are via the feed, approximately 80% of the total con- To achieve a target dose of chlortetracycline to treat
sumption. Dose intake is linked directly to feed intake a uterine infection, five times the normal weaner
and inclusion level. One of the common problems inclusion rate is required. This is also important in
veterinarians encounter regarding efficacy is under- eradication programmes to ensure the correct dose is
dosing. If the animal is sick with a high temperature, administered to the various age groups. Many in-feed
it will stop eating. It is essential to treat pigs with no administration products give relatively lower plasma
appetite by injection to get them to start eating the levels and bioavailability (14), especially products that
medicated feed. The age of the pig is also important. are mainly metabolised in the liver (e.g. macrolides,
Most dose rates are based upon a 20 kg pig eating 1 kg pleuromutilins and lincosamides), due to the slower
of feed/day or 5% of bodyweight. Finishing pigs are passage down the gut. Products excreted via the
often given restricted feed to control fat deposition, kidney (e.g. tetracyclines, trimethoprim/sulfonamides

112 Guide to antimicrobial use in animals

Table 7.8 Minimum inhibitory concentrations (MICs) of antimicrobial agents used against A. pleuropneumoniae,
P. multocida and S. suis (60)

7 Swine Antimicrobial (NCCLS resistance breakpoints – μg/ml) MIC 50 (μg/ml) MIC 90 (μg/ml) Range (μg/ml)

A. pleuropneumoniae – N. America (89 isolates) 0.5 32 ≤0.12–64
Penicillin (≥0.25) 16 32 ≤0.12–64
Tetracycline (≥8)
≤0.06 ≤0.06 ≤0.06–0.12
Trimethoprim/sulfonamide (≥4/76) 2.0 4.0 ≤0.12–4.0
Tilmicosin (≥32) 0.25 0.5 ≤0.06–0.5
Florfenicol (≥8) ≤0.03 ≤0.03 ≤0.03–0.06
Enrofloxacin (≥2) ≤0.03 ≤0.03 ≤0.03–0.06
Ceftiofur (≥8)
≤0.12 ≤0.12 ≤0.12–>64
P. multocida – N. America (186 isolates) 2.0 24 0.25–>64
Penicillin (≥0.25)
Tetracycline (≥8) ≤0.06 0.25 ≤0.06–>8.0
4.0 8.0 ≤0.12–>64
Trimethoprim/sulfonamide (≥4/76) 0.25 0.5 ≤0.06–4.0
Tilmicosin (≥32) ≤0.03 ≤0.03 ≤0.03–0.5
Florfenicol (≥8) ≤0.03 ≤0.03 ≤0.03–4.0
Enrofloxacin (≥2)
Ceftiofur (≥8) ≤0.12 0.25 ≤0.12–32
64 64 0.25–>64
S. suis – N. America (167 isolates)
Penicillin (≥0.25) ≤0.06 0.12 ≤0.06–1.0
Tetracycline (≥8) >64 >64 ≤0.12–>64
1.0 2.0 0.12–4.0
Trimethoprim/sulfonamide (≥4/76) 0.25 0.5 ≤0.03–1.0
Tilmicosin (≥32) ≤0.03 0.06 ≤0.03–4.0
Florfenicol (≥8)
Enrofloxacin (≥2)
Ceftiofur (≥8)

Table 7.9 MICs 50, MICs 90, ranges (μg/ml) and antimicrobial resistance (%) among 229 Spanish isolates of
A. pleuropneumoniae (61)

Antimicrobial MIC 50 MIC 90 Range Breakpoint Resistance

Penicillin 1.0 64 0.12–128 ≥4.0 15
99
Amoxicillin ≤0.25 15
Cefalothin 0.4
Tetracycline 0.5 64 0.25–64 ≥8.0 74
Gentamicin 9.2
Trimethoprim 0.5 2.0 0.5–32 ≥16 NR
Sulfisoxazole 17
Florfenicol 32 64 0.25–128 ≥16 0
Nalidixic acid NR
8.0 8.0 4.0–8.0 ≥16

≤1.0 ≤1.0 ≤1.0–32 NR

64 >512 32–>512 ≤512

0.25 0.5 0.12–1.0 ≤8.0

2.0 4.0 1.0–32 NR

NR – not recorded.

and penicillins) are not normally affected. It must also Soluble products given via the drinking water or in
be remembered that some products given orally are liquid feed pass through the stomach more quickly
hardly absorbed from the gut, such as the aminogly- and are therefore more rapidly absorbed. Therapeutic
cosides (e.g. neomycin, apramycin) and aminocycli- levels of soluble products (e.g. tiamulin) can be
tols (e.g. spectinomycin). Thus, it is of little use to achieved in the lung to treat respiratory pathogens
give them orally for systemic or respiratory infections. such as Actinobacillus pleuropneumoniae, but these

Table 7.10 Recent data on the occurrence of antimicrobial resistance among S. suis isolated from swine in different countries (62)
Country, year of isolation and number of isolates

Belgium Canada Croatia Denmark England/ Portugal Spain USA Guidelines for antimicrobial use in swine
Wales France Finland Japan Netherlands Norway Poland

1999– 1986– Prior to Prior to 1995– 1984– 1987– Prior
2000 1987 1996 2003 1999– to
1988 1988 1991 1995 1996 2003 2003 2003 1986 2002 2003 2003 1992 2001 1992
87
Antimicrobial 59a 135 80 33a 180 557 34 Differs 35 689 762 21 150 151 14 65 151 48
agent

Fluoroquino- 1 — —— 9.1 2.2 0 0 2.9 — 0.3 — — 8.6 9.9 64 — 2.0 —
lones

Macrolides 71 33.9 60.7 66.3 0 41.7 29.1 36 52.9 0 — 35 28.6 28.0 — 93/71 63 90.7 58
0.0 — ——
Gentamicin — 66.1 0.0 2.5 — — —— —0 0.9 0 — — 0.0 100 — 4.6 9
—— ——
Penicillin — 0 3.0 2.5 51.1 1.1 0.9 0 — 10.6 7.9 21 7 4.0 —

Sulfona- — — — 47.5 — — 0.9 — — — —— — 96.0 44
mides 85
83.0 78.5 95.0 72.2 32.2 52.2 68 69.9 42.8 86.9 48 66.7 73.3 55.0 93/79 61 95.4 63
Tetracycline 39.0 1.5 — — 1.7 51.5 3 15.5 31.4 0.0 8

Co-trimoxazole — — 30.0 16.6 100 61 0.0 —

a Only serotype 2.

113

7 Swine

114 Guide to antimicrobial use in animals

Table 7.11 Antimicrobial resistance of 124 US isolates of H. parasuis (70) and MIC 50, MIC 90 and Range (μg/ml) of
Danish isolates (71)

7 Swine Antimicrobial agent (disc conc) Resistance (%) MIC 50 MIC 90 Range

Penicillin (10 IU) 2.1 0.015 0.25 0.015–0.25
Ampicillin (10 μg) 0.0 1.0 1.0 1.0
Cefalothin (30 μg) 0.0 — — —
Ceftiofur (30 μg) 2.1 0.03 0.03 0.03
Tetracycline (30 μg) 14.9 1.0 2.0 0.06–2.0
Trimethoprim/sulfonamide (5 μg) 6.4 0.03 2.0 0.015–8.0
Gentamicin (10 μg) 4.3 — — —
Amikacin (30 μg) 6.4 — — —
Enrofloxacin (5 μg) 0.0 — — —
Ciprofloxacin — 0.015 0.125 0.015–0.5
Florfenicol — 0.25 0.5 0.125–0.5
Tilmicosin — 2.0 2.0 2.0–4.0
Tiamulin — 4.0 8.0 1.0–16

Table 7.12 Occurrence of antimicrobial resistance in S. hyicus from different countries (62)

Country, year, (number of isolates) and percentage of resistance

Belgium Denmark Germany Japan United Kingdom

Antimicrobial agent 1974–1976 (46) 2003 (68) 1989 (32) 1979–1984 (124)a 1988 (37)

Chloramphenicol — 0 90 0
Florfenicol — 0 —— —
Fluoroquinolones — 4 —— —
Gentamicin — 0 —0 0
Macrolides 74 21 3 41 11
Penicillin 60 84 25 38 32
Streptomycin 72 44 43 23 51
Sulfonamides — 2 100 — —
Tetracycline 60 35 66 54 41
Trimethoprim — 24 — —

aBoth healthy and diseased.

levels are not reached when the same drug is given consumption of contaminated pork, infection with
in the feed. less known zoonotic agents like Streptococcus suis and
methicillin-resistant Staphylococcus aureus (MRSA)
7.2 Zoonotic infections transmitted usually occurs in people working closely with pigs,
by pigs such as farmers, slaughterhouse workers, butchers or
veterinarians. These zoonotic infections are reviewed
The two main food-borne pathogens associated with in further detail below.
swine are Salmonella (especially S. Typhimurium
and S. Derby) and Campylobacter (primarily C. coli), 7.2.1 Salmonella
both causing mainly enteric disease in humans, but
occasionally more severe infections. While these The main serovar found in pigs is S. Typhimurium,
zoonotic agents are mainly transmitted to humans by which accounts for about 65% of isolations from pigs
in the UK (VLA Salmonella 2004, 2005), while S. Derby

Guidelines for antimicrobial use in swine 115

accounts for about 17%. Among human cases in the Salmonella and found that infections with resistant 7 Swine
UK in 2004, 64% were due to S. Enteritidis, which Salmonella were associated with a higher death rate,
is found predominantly in poultry, 11% was due to increased hospitalisation rate and longer illness than
S. Typhimurium, which can be found in most animal infections with susceptible isolates (18–21). In partic-
species, and <1.0% due to S. Derby (15). Some cases ular, infections with quinolone-resistant Salmonella
of S. Typhimurium are more commonly associated are associated with increased mortality and morbid-
with phage types from pigs (such as U288 and 193) ity compared to infections with quinolone-suscepti-
so a definite link to pigs as a source has been estab- ble Salmonella (22,23). Not all types of resistance are
lished (15). However, many phage types can be found of equal importance. Fluoroquinolones are in many
in a variety of animal and poultry species, and it is countries the drug of choice for treatment of salmo-
therefore difficult to apportion the number of cases nellosis in humans. Cephalosporins are the drugs of
directly due to pigs/pork. Danish and imported pork choice for the treatment of salmonellosis in children,
was estimated to account for approximately 29% of as fluoroquinolones cannot be administered due
all attributed human Salmonella infections acquired to toxic effects on children. Accordingly, the use of
in Denmark (16). However, the question is contro- fluorquinolones and cephalosporins in pigs should
versial. If S. Derby is used as an indicator, on a pro- be avoided unless the pathogen is resistant to any
portional basis, the percentage of all UK human cases other antimicrobial agent available in the veterinary
due to pork consumption could be under 5% (D.G.S. arsenal.
Burch – unpubl. obs.).
7.2.2 Campylobacter
The resistance patterns of S. Typhimurium, from a
number of EU countries, are highlighted in Table 7.3, The majority of Campylobacter infections in man are
with high susceptibility being recorded for amoxi- due to Campylobacter jejuni. Burch (24), in a litera-
cillin/clavulanic acid, ceftiofur (third-generation ture review, estimated that 92% of human infections
cephalosporin), ciprofloxacin (fluoroquinolone) and are caused by C. jejuni and only 8% by C. coli. The
gentamicin (aminoglycoside). In Poland, slightly dominant Campylobacter species isolated from pigs
higher resistance levels have been reported on the is C. coli (96%). The prevalence of C. jejuni is higher
basis of a low number of isolates (Table 7.3). It can be in chickens (90%) and in cattle (99%), which is simi-
concluded that pork does pose a potential zoonotic lar to the incidence of human Campylobacteriosis
hazard in relation to Salmonella, but that it is a rela- associated with these species. Conversely to chickens,
tively small risk in comparison with poultry meat and turkeys can have quite a high prevalence of C. coli,
products, especially if handled and cooked properly. although C. jejuni tend to be the dominant species
The likelihood of treatment failure due to antimi- (25). Several studies have indicated that poultry is the
crobial resistance, on those rare occasions when anti- most important reservoir of Campylobacter infections
microbial therapy should be required, is generally in man (26, 27).
considered to be low. However, antimicrobial resis-
tant Salmonella such as S. Typhimurium DT104 can Macrolides and fluoroquinolones are the drugs
have serious human health consequences both due of choice for treatment of severe campylobacte-
to the occurrence of infections that would otherwise riosis in humans. There is only limited information
not have occurred and due to treatment failures and on the importance of antimicrobial resistance in
increased severity of infections. The use of antimicro- Campylobacter for human health. In the only available
bial agents in humans disturbs the intestinal micro- study, Helms et al. (28) found that patients infected
flora. Individuals taking an antimicrobial agent for with quinolone or macrolide-resistant strains of
respiratory infections, for example, are therefore at C. jejuni had an increased risk for an adverse event
an increased risk of becoming infected with intestinal compared to patients infected with susceptible strains.
pathogens resistant to that agent. Barza and Travers This finding still needs to be verified in other studies.
(17) estimated that in the USA resistance to antimi- Macrolide (erythromycin) resistance in C. coli iso-
crobial agents results annually in an additional 29 lated from pigs can be quite high, up to 85% in the
379 Salmonella infections, leading to 342 hospitalisa- UK (29) and it cannot be discounted that this could
tions and 12 deaths. Several studies have examined have some human health implications. However, since
the severity of infections with antimicrobial resistant most infections in man are caused by C. jejuni, the

116 Guide to antimicrobial use in animals

7 Swine importance of macrolide resistance in C. coli should With our current knowledge, it seems quite evident
not be overestimated. that ST398 is a MRSA clone transmitted from pigs to
humans; its origin is unknown, though it – or its ante-
7.2.3 Streptococcus suis cedents – could have originated in humans. Further
studies are underway in several countries, but it seems
Human infections with S. suis serotypes 2 and 14 that MRSA ST398 is widespread in the pig popula-
have been reportedly associated with people produc- tions, at least in the Netherlands and Denmark, but
ing or processing pigs in the UK (30). Among the most likely in all European countries with intensive
41 laboratory confirmed cases between 1981 and 2000 swine production. ST398 mainly colonises animals,
(approximately 2 per year), 27% were pig farmers or but has been found to cause infections, in a few cases.
stockmen, 22% retail butchers, 12% abattoir workers, The limited number of reports is probably due to the
12% no apparent risk and 27% no epidemiological difficulties of isolating this bacterium from animals,
data. Recently, a large S. suis outbreak in China was because it is necessary to use selective enrichment.
related to slaughterhouse employees working in poor It must be expected that several new reports will be
health and safety conditions (31,32). Although the published in the near future. The reason for the colo-
risk of infection is low, it can cause fatalities in man. nisation of MRSA ST398 in pigs and the epidemiol-
Penicillin still remains highly active against S. suis ogy of this clone is currently not known. It possibly
(>90% of isolates are susceptible). first emerged in 2003, as it was not detected in 2002 in
the human monitoring being done in the Netherlands
7.2.4 Methicillin-resistant nor in monitoring from 1992 to 2003 of human iso-
Staphylococcus aureus lates in Germany. It can be speculated that the use of
cephalosporins and other antimicrobials has provided
MRSA colonisation in animals has been implicated a niche for this clone, but until further studies are car-
in infections in humans in several cases, and MRSA ried out this is merely speculation. The importance
should today be considered as a zoonosis. Conversely for human health and the possibilities for infection
to pets, which are usually infected or colonised with control are currently unclear. In the Netherlands, the
classical human variants of MRSA (33), a new clone advice is to keep pig breeders in isolation when they
(multi-locus sequence type ST398) appears to have are admitted to a hospital, until surveillance cultures
emerged in pigs. This MRSA clone was first reported are proven negative. This also applies to veterinarians
in a family of pig farmers in the Netherlands (34). and slaughterhouse personnel. For cattle breeders,
Subsequently, in a survey, MRSA was found in 23% screening without isolation on admission to a hospi-
of Dutch farmers and in 4.6% of veterinarians and tal is sufficient.
veterinary students in the Netherlands (34, 35). In the
same country, a recent study of the pig population has 7.3 Prudent antimicrobial use
revealed a colonisation rate of 39% of all slaughter in pigs
pigs and 80% of pig slaughter batches (36). All isolates
belonged to clone ST398, which seems to have estab- 7.3.1 Principles of disease control in
lished itself in the pig population in the Netherlands. swine medicine
In the autumn of 2006, the same clone was detected
in patients in Denmark, most of which had close Disease control is not only about using medicines.
contacts to production animals, mainly swine. A ret- Frequently, what has gone wrong is the production
rospective study has shown that MRSA ST398 had system. Hence, the challenge is to correct the underly-
already occurred in Danish slaughter pigs in 2005 (37). ing management problems. Post-weaning diarrhoea is
The same sequence type has also been described in the classic example. If the temperature of the weaning
S. aureus strains isolated from pigs and farmers in accommodation is kept high and constant (26–29°C),
France, though the isolates were not methicillin resis- and drafts are avoided, there is normally little trou-
tant (38). MRSA ST 398 has also been described in ble. Pigs can be weaned into deep straw in barns
Germany, including in veterinarians (nasal carriage), at four weeks of age, in the middle of a UK winter,
human patients, companion animals and a single with no clinical problems. The ‘right’ environment is
pig (39).

Guidelines for antimicrobial use in swine 117

very important to the pig and to disease prevention. Veterinarians have established protocols and eco- 7 Swine
In general, various approaches can be used to avoid nomic justification for depopulation of pig herds
or eliminate the infectious agents and to avoid the affected by certain costly infections (40). Sometimes,
clinical disease. it is more cost effective or easier to eradicate using
herd closure and mass medication. Many different
Avoid or eliminate the infectious agent protocols have been developed for eradication of
enzootic pneumonia, A. pleuropneumoniae, swine
The health status of a herd is critical in relation to dysentery and atrophic rhinitis (41). Basically, the
antimicrobial use. A herd can be started up free of herd is closed to new pig introductions, growing
specific infectious agents. Alternatively, the infectious pigs are moved off site to clean accommodation,
agent can be eliminated by carrying out depopula- and the breeding herd immunity is usually well devel-
tion and repopulation with clean stock or by using oped, either by natural infection or vaccination if
herd closure and antimicrobial medication. Having suitable, while disease transmission is eliminated
sources of high-health or Specific Pathogen Free (SPF) by antimicrobial prophylaxis and metaphylaxis.
stock is essential to prevent introduction of common Once a high-health herd is established, it is essen-
bacterial pathogens such as Mycoplasma hyopneu- tial to keep diseases out. Ideally, there should be no
moniae, toxigenic Pasteurella multocida type D, A. other pig farm within three miles (5 km). There are
pleuropneumoniae and Brachyspira hyodysenteriae. strict biosecurity protocols to keep infections out,
Ideally, this list should be extended to include several so there is no direct access for feed deliveries, pig
viruses such as those causing Porcine Reproductive loading or animal entry. Ideally, the site should be
and Respiratory Syndrome (PRRS), Aujeszky’s disease closed to new pig replacements, as they still remain the
and Transmissible Gastroenteritis. Many countries most common cause of herd breakdown, in spite of
have eradicated Classical Swine Fever and Foot and separate quarantine or isolation facilities (41). Semen
Mouth Disease, but such diseases are still endemic in can also be a carrier of viral infections, as demon-
certain areas of Asia and South America. It is diffi- strated by the Swine Fever outbreak in Holland (42)
cult for producers to obtain stock free of Parvovirus, and by well-documented PRRS breakdowns follow-
L. intracellularis, S. suis or Haemophilus parasuis, ing introduction of PRRSv contaminated semen into
unless they use caesarean-derived pigs and extreme naive herds (43). However, import of live animals
isolation and biosecurity measures. Breeding stock for breeding and finishing remains the more com-
companies and large production systems operate mon means of bringing diseases onto a farm. Persons
health pyramids to flow high health pigs in linear can carry infectious agents on their skin and clothes
fashion. The top of the pyramid is usually the genetic and pig farms should not be entered by external visi-
nucleus. This way the highest health pigs always flow tors without them going through a shower, a clothes
down the pyramid towards the lowest health status changing system, and a one to three-day down time
pigs, and crossing over of pyramids is avoided in order without pig contact.
to reduce transmission of infectious agents.
Avoid clinical disease
Sourcing of new genetic stock is important, and
stricter protocols for diagnosis should possibly be There are a variety of production systems in every
imposed to ensure absence of specified infections. pig-producing region of the world and almost
Often herd veterinarians will only certify ‘no clinical every farm is different with respect to structure
disease is observed’, but the pigs may still be carriers and management problems. Working within the
or even test positive for a disease-causing organism farm system and improving it to reduce infectious
by polymerase chain reaction (PCR). In some cases, challenge is the key to being a successful swine
unwanted antimicrobial resistance may be introduced veterinarian. However, some basic principles still
by this route into a herd or country. Matching the apply. Various key areas need to be addressed to
health status of a herd with a supplier is also important. avoid clinical disease: herd management and size,
Supplying enzootic pneumonia-free gilts or boars to an population density, parity segregated produc-
infected farm is not recommended, as the replacements tion, weaning age, pig housing environment and
and their offspring will come down with the acute form immunity.
of the disease when exposed to the causative agent.

118 Guide to antimicrobial use in animals

7 Swine Herd management and size although ileitis, S. suis meningitis and Glässer’s disease
still occur. However, this is not a system that can be
Attention to detail is critical. Small closed breeding easily adopted, especially in Europe where herd sizes
finisher herds, which are family owned, usually do are usually smaller and land is not readily available.
well when compared with farms where employees
look after pigs. Stress is reduced by avoiding movement and mix-
ing of pigs during the production cycle. Many larger
Mixing pigs from different sources should be avoided. systems are adopting wean-to-finish production
The old cooperative farm system, where grower (C.O. Duran, pers. comm.). This production sys-
pigs from many breeding farms were mixed together tem consists of weaning pigs offsite, into accom-
for finishing, usually had severe disease problems. modation where they remain until slaughter. These
The more mixing that took place, especially from systems incorporate the benefits of all-in-all-out and
different farms, the greater the stress to the pigs, and no mixing of different ages and reduces the need for
the greater the risk of introducing new diseases to transport between phases. The weaned piglet ther-
which they had little or no immunity. In the past, fin- mal requirements are met with temporary provision
ishing units would mix pigs from 20 different sources of supplemental heat in the form of heat lamps or
and, basically, the lowest common disease denomina- brooder heaters. Disease challenges are much reduced
tor applied. In smaller single-site or farrow-to-finish by this system, particularly enzootic pneumonia.
farms, all-in-all-out housing/room systems coupled
with good hygiene have very successfully reduced the Population density
spread of disease from one group of pigs to another.
Eventually, following the principles developed for dis- For respiratory diseases particularly, a reduction in
ease elimination, without depopulation, the three-site pigs per airspace has resulted in less severe infections,
production system was developed in the USA, to try although some of these benefits have been reached
to halt or limit the spread of disease from one pro- with correct ventilation and management. Increased
duction stage to another (13). The three-site system pig density in pens or barn has also been linked to
consists of having the breeder sites, the nursery sites increased stress and disease transmission resulting in
and finishing sites not just in separate houses, but higher mortality and reduced growth. Ideal space per
on separate farms. Larger sow farms (site 1) allow head recommendations will vary depending on the
for more cost effective use of growing pig buildings. housing type and production system (44).
Specialised labour can be employed on the different
sites. Often pigs are weaned at between 14 and 18 days Production based on segregated sow parities
to the nursery site to prevent the spread of a number
of infections, but the usefulness of early weaning in Recently, the benefits of raising pigs segregated by the
preventing disease is still under debate. In the EU, for parity of the sows have been applied in Canada and
welfare reasons, piglets are required to be weaned at the USA (45). This system reduces disease challenges
28 days. Initially, nursery sites (also known as site 2) by reducing variation in the immune status of the pig-
consisted of 7 or 8 rooms to accommodate 1 week of lets by grouping them with piglets of like maternally
production. As a consequence, a large number of sows derived antibody (MDA) and disease carrier status.
were required to efficiently fill a nursery in one week, Generally, this system raises the gilt offspring sepa-
hence the development during the 1990s of the large rate from offspring of sows of parity one and above.
integrated systems in the USA. Following the increase These systems reduce antimicrobial and vaccine use
of PRRS infections, it was determined that even the by allowing finely tuned interventions due to a very
age variation from weaning to 10 weeks on one site predictable time of disease challenge.
was detrimental to pig health, so multi-site production
systems were developed. In these systems, the entire Weaning age
site is filled with pigs of the same age and moved all-
in-all-out. Generally, age varies by 7–14 days maxi- One of the other key elements to disease elimina-
mum. This in turn has fuelled a further increase in tion without depopulation was early weaning below
the sow herd size to allow for efficient filling of large 21 days of age, sometimes below 10 days (13). This
nurseries. The closer the farms follow the multi-site was found to reduce the transfer of infections from
production system, the fewer disease problems occur, the sow to the piglets, utilising the transfer of MDAs
to protect the piglets and preventing colonisation.
This was effective for several viral infections, notably

Guidelines for antimicrobial use in swine 119

pseudorabies virus (PRV), porcine parvovirus (PPV), is essential for the piglets to absorb sufficient IgG 7 Swine
TGE and also A. pleuropneumoniae, atrophic rhinitis, antibodies and acquire circulating immunity.
M. hyopneumoniae, although it was less effective Continued production of IgA antibodies during lac-
against some bacterial infections, where transmis- tation provides mucosal protection in the piglet gut.
sion from sow to piglet occurs around the time of
parturition (e.g. S. suis and H. parasuis). In the EU, Vaccine use in growing pigs was geared towards
welfare legislation encourages weaning no earlier than prevention of acute respiratory or systemic infections
28 days of age. This has been helpful in reducing the like A. pleuropneumoniae or erysipelas. These vaccines
effects of post-weaning diarrhoea and has been found were traditionally whole cell, killed and adjuvanted,
to mitigate the severity of PMWS/PCVAD, but could requiring two injections in growing pigs. The vacci-
increase the risk of infection transfer from sows to nation of piglets was revolutionised with the intro-
piglets. Recent studies in the USA investigating the duction of the M. hyopneumoniae vaccines, where
profitability of weaning pigs below 21 days have fur- piglets as young as a week old were shown to develop
ther encouraged moves away from early weaning (46). immunity against this endemic infection. The piglet’s
immune system appears to be sufficiently well devel-
Housing environment oped to respond to vaccination at this age, although it
is not fully matured until four weeks of age. However,
It is beyond the scope of this chapter to detail all the it has been shown that high levels of MDA can reduce
ideal conditions for pig environments at different ages, the vaccinal response in some cases (50,51). Some
but it is essential to avoid conditions that increase mycoplasmal vaccines were coupled with H. parasuis
pathogen exposure and challenge. Most important to give early protection against Glässer’s disease. Other
are providing the correct temperature to maintain common vaccines, against viral infections like PRV,
pigs in their thermo-neutral zone, and avoiding drafts PRRSV and SIV, can help reduce secondary bacterial
while removing moisture, gases and pathogens with challenges in growing pigs (52). A recent development
adequate ventilation. Removal of manure and soiled is the availability of live oral vaccines for Salmonella
bedding, plus sanitation processes, are also critical to choleraesuis, Lawsonia intracellularis, E. rhusiopathiae
the reduction of build-up of microbes in the envi- and F18 E. coli. These improve protection against
ronment. Disease prevention by appropriate housing these costly diseases and it is expected that they will
and environmental management has been reviewed reduce antimicrobial usage for treatment and control.
elsewhere (47). Excitingly, new PCV2 vaccines have been developed
for young pigs in North America and it is hoped that
Herd/population immunity this scourge will be successfully controlled worldwide
in the future.
Disease control relies heavily on the protection of the
pigs via natural or induced immunity. Early exposure There appears to be a need for new vaccines, as well
to herd pathogens for replacement breeding stock can as improving old vaccines, particularly in the areas
result in solid protection of their offspring (acclimati- of mucosal immunity, improved practical delivery
sation) (48, 49). This is often achieved by exposure to and application of needle-less technology. Recent
manure, mature cull sows or sick pigs, supplemented work on development of sub-unit vaccines, which
by vaccines. Stimulation of immunity by vaccination permit poly-vaccination to cover a spectrum of com-
also plays a major role in pig production. Vaccination mon pig diseases in one shot, provides hope for the
of the breeding herd to protect against infections in future (53).
the sow is essential for good production and helps to
build up herd immunity. These include viral infec- 7.3.2 Choice of antimicrobials for
tions such as PPV, PRRSV, PRV and in some countries swine diseases
porcine circovirus type 2 PCV2 as well as erysipelas
(Erysipelothrix rhusiopathiae). Vaccination of the sow Besides using approaches to avoid infectious agents
in late gestation is also used for stimulating MDA pro- and clinical disease and the use of vaccines where
tection in the baby piglet against E. coli, Clostridium possible, there are definite cases where antimicrobial
perfringens type C and type A, atrophic rhinitis in therapy is required. In such cases, the authors advise
the farrowing house and erysipelas in the grower. that an accurate diagnosis is made, and that cultures
Early intake of sow colostrum in the first 24 h of life and antimicrobial susceptibility data are used to

120 Guide to antimicrobial use in animals

7 Swine support the therapy of choice. Once the most appro- used under certain extreme conditions, but prudent
priate antimicrobial is chosen, then a suitable route of use guidelines recommend their use only as a last
administration and dosage should be used to reach the resort (see Table 7.13).
organism and to ensure clinical efficacy. The follow-
ing sections provide pathogen- and disease-specific Diarrhoea/enteritis
guidelines for efficacious antimicrobial use in swine
practice. In addition to clinical efficacy, an important Data on the occurrence of antimicrobial resistance
additional factor is to try to limit the use of antimicro- among E. coli from enteric infections in pigs in dif-
bials that are critical in human medicine, such as fluo- ferent countries are given in Table 7.4. Major differ-
roquinolones and the cephalosporins (see Chapter 4). ences in the occurrence of resistance are obvious. In
In the USA, prudent use guidelines would also include general, there is a frequent occurrence of resistance to
trimethoprim/sulfonamide combinations. They may be

Table 7.13 First, second and last resort choices for antimicrobial therapy of common porcine infections

Infection/disease First choice Second choice Last resort

E. coli: Trimethoprim/S* (OD**, Inj) Neomycin, Apramycin (OD) Amoxicillin (OD, Inj)
Neonatal scours Colistin
(Sow vaccination) Amoxicillin/clavulanate (Inj)
Piglet scours Colistin (OD)
Post-weaning diarrhoea Zinc oxide (IF) Cephalosporins*** (Inj),

MMA syndrome Trimethoprim/S*(Inj) Fluoroquinolone (OD, Inj)***
Salmonella spp:
Diarrhoea Neomycin, As above

Trimethoprim/S* (OD; Inj)

Colistin (IF, IW), As above

Neomycin (IF, IW)

Trimethoprim/S* (IF, IW)

Amoxicillin (Inj) As above

Ampicillin (Inj)

Colistin (IW, IF) Neomycin (IF, IW) As above
Trimethoprim/S* (IF, IW)
Spectinomycin (IF, IW)

Septicaemia Trimethoprim/S (Inj) Amoxicillin (Inj) As above
C. perfringens: Amoxicillin (Inj)
Necrotic enteritis Penicillin (Inj) Amoxicillin/clavulanate (Inj)
(Sow vaccination) Tetracyclines (IW, IF) Tylosin (Inj)
L. intracellularis: Tylosin (Inj)
Ileitis Pleuromutilins Tylosin (Inj),
(Inj, IW, IF) Macrolides (IW, IF)
PHE (Pig vaccination) Lincomycin (IW, IF)
B. hyodysenteriae: Tiamulin (Inj)
Swine dysentery Lincomycin (Inj, IW, IF) Macrolides (Inj, IW, IF)
Pleuromutilins
B. pilosicoli: (Inj, IW, IF) Lincomycin (Inj, IW, IF) Macrolides (Inj, IW, IF)
Colitis
Pleuromutilins Tetracyclines (Inj, IW, IF) Macrolides (Inj, IW, IF)
M. hyopneumoniae: (Inj, IW, IF) Lincomycin (Inj, IW, IF)
Enzootic pneumonia Pleuromutilins
(Inj, IW, IF)
(Pig vaccination)

Continued

Guidelines for antimicrobial use in swine 121

Table 7.13 (Continued) First choice Second choice Last resort 7 Swine

Infection/disease Penicillin (Inj) Florfenicol (Inj) Amoxicillin (Inj, IW, IF)
Penethamate (Inj)
P. multocida:
Pneumonia

Tulathromycin (Inj)

Trimethoprim/S* (Inj, IW, IF)

Tetracyclines (Inj, IW, IF)

A. pleuropneumoniae: Penicillin (Inj) Florfenicol (Inj) Amoxicillin (Inj, IW, IF)
Pleuropneumonia Penethamate (Inj)
(Pig vaccination) Tulathromycin (Inj) Cephalosporins***,
H. parasuis:
Glässer’s disease Penicillin (Inj) Tetracycline (Inj, IW, IF) Fluoroquinolone (Inj)***
(Pig vaccination)
S. suis: Trimethoprim/S* (Inj, IW, IF) Tilmicosin (IF)
Meningitis Penicillin (Inj, IW, IF)
E. rhusiopathiae: Florfenicol (Inj) Amoxicillin
Erysipelas Penicillin (Inj)
M. hyosynoviae: (Sow/pig vaccination) Tetracycline (Inj, IW, IF) Amoxicillin/
Arthritis Tiamulin (Inj, IW, IF)
Trimethoprim/S* (Inj, IW, IF) clavulanate, Florfenicol

Amoxicillin, (Inj, IW, IF)
Trimethoprim/S* (Inj, IW, IF)

Amoxicillin (Inj, IW, IF)

Lincomycin (Inj, IW, IF)

* Trimethoprim/sulfonamides on US critical antimicrobial list, but widely used in the rest of the world in pig medicine.
** Inj – injectable; OD – oral dosing; IW – in water; IF – in feed.
*** Wherever possible, the use of fluoroquinolones and cephalosporins should be reserved for human use.

ampicillin, streptomycin, sulfonamides and tetracy- (54) consider that both organisms are so commonly
clines. However, it is not possible to predict for E. coli resistant to tylosin and lincomycin that these antimi-
whether an isolate is susceptible or resistant, and crobials cannot be recommended.
choice of empiric treatment has to be made on the
basis of knowledge of the individual herd and local Pleuromutilins are also the drugs of choice for
data on resistance patterns. This is why routine sub- treatment of L. intracellularis infections (Table 7.13).
mission of samples to a microbiological laboratory is This bacterium grows inside the cell and MIC testing
important to generate records on susceptibility data. requires growth in cell cultures. This is not an easy
or possibly sensitive procedure. Mackie (55) looked
The MICs for B. hyodysenteriae and B. pilosicoli are at the effects of antimicrobial concentrations in the
very low for pleuromutilins (valnemulin and tiamu- culture fluid and their effects on inhibition of growth
lin), and although resistance has now been reported of the organism inside the cell. Not all were titrated to
in a number of EU countries to a few strains (see their lowest concentration. A recent study (56), which
Table 7.5), one would expect over 90% of isolates titrated eight US and EU isolates down to 0.125 µg/
to be susceptible. Pleuromutilins are not used in ml, showed that tiamulin, valnemulin and carbadox
human medicine and apparently do not cross-select were highly active, and that lincomycin, tylosin and
for resistance to critically important antimicrobials. chlortetracycline had a more variable activity.
Accordingly, these antimicrobials should be regarded
as the first choice for treatment of swine dysen- Clostridium perfringens is very susceptible to
tery colitis enteritis associated with Brachyspira spp. penicillin (first choice) and tylosin (second choice),
Lincomycin achieves much higher levels in the colon, whereas there is a variable degree of susceptibility to
so much higher MICs are found, but most isolates tiamulin, lincomycin and tetracycline. Clostridium
(>70%) would be resistant to tylosin. Some authors difficile seems to show some susceptibility to macro-
lides and virginiamycin (57).

122 Guide to antimicrobial use in animals

7 Swine Respiratory/systemic diseases When evaluating data on antimicrobial susceptibil-
ity, it is very important to take into consideration the
The variability in susceptibility/resistance patterns of methods and breakpoints used to measure resistance.
the various respiratory and systemic bacterial patho- Guttierez-Martin et al. (61) reported on the suscepti-
gens is shown in Tables 7.6–7.12. Culture and sus- bility of 229 Spanish isolates of A. pleuropneumoniae,
ceptibility testing of Mycoplasma species is difficult which had been isolated between 1997 and 2004. They
and only carried out at a few reference laboratories used a microdilution doubling-dilution technique
(see Table 7.6). However, M. hyopneumoniae, and resistance was based on CLSI interpretation.
M. hyosynoviae and M. hyorhinis are generally suscep- The levels of penicillin resistance differed substan-
tible to tiamulin, whereas low levels of resistance have tially depending on the interpretive breakpoint used.
been reported towards tylosin, tilmicosin, lincomycin, High prevalence (99%) of resistance to penicillin
tetracycline and fluoroquinolones (58). was recorded according to the lower MIC breakpoint
(Table 7.9), suggesting that penicillin was likely to be
A. pleuropneumoniae is generally susceptible to all ineffective if given by the oral route. However, the
commonly used antimicrobials, including the penicil- higher breakpoint indicated that possibly only 15% of
lins, which should therefore be considered as the first the strains would be resistant if penicillin was given
choice antimicrobial for treating this pathogen. Teale by injection, as much higher blood levels are achieved
et al. (59) reported on porcine isolates submitted for by this administration route. There would appear to
diagnostic investigation to the Veterinary Laboratories be a very high level of resistance to the tetracyclines in
Agency in the UK between 1999 and 2002 (see Table Spain, similar to the USA.
7.7). Resistance was measured by the Kirby-Bauer disc
method using a 13 mm-diameter as the breakpoint for Penicillin is active against most strains (>90%)
ceftiofur resistance. This differs from the CLSI (for- of S. suis, (60) and resistance to trimethoprim/
merly NCCLS) definition of resistance, which is based sulfonamide, florfenicol and ceftiofur has never been
on a higher zone interpretive breakpoint (R≤17 mm). reported. In Europe, S. suis often shows MICs of fluo-
roquinolones close to the breakpoint and thus these
Penicillins also remain very active against S. suis, antimicrobials cannot be recommended (62).Similarly,
E. rhusiopathiae and A. pyogenes. Some resistance to tetracyclines and macrolides are scarcely active against
ampicillin, trimethoprim/sulfonamide and tetracy- this Gram-positive pathogen (Table 7.10).
cline is shown by A. pleuropneumoniae and, in uncon-
trollable resistance situations, the fluoroquinolones H. parasuis is generally susceptible even to the more
or cephalosporins have been effectively used by injec- commonly used antimicrobials such as penicillin,
tion. However, these antimicrobials should be used as although low-level resistance to tetracycline has been
last resorts, limited to the hypothetical situations in reported (Table 7.10). Altogether, parenteral adminis-
which the target strain is resistant to all first, second tration of penicillin is particularly useful for treating
and third choice drugs listed in Table 7.13. High MICs respiratory and systemic bacterial pathogens in pigs
of penicillins and tetracyclines have been reported as it results in markedly higher serum concentrations
among A. pleuropneumoniae isolates in North America compared with oral administration. However, unfor-
(60), but there does not appear to be any resistance tunately, in the USA no injectable formulations of
to trimethoprim/sulfonamide, tilmicosin, florfenicol, penicillins are labelled for use in pigs and this factor
enrofloxacin and ceftiofur. P. multocida can also be limits the use of these antibiotics.
resistant to penicillin, tetracyclines or tilmicosin, but
is generally susceptible to trimethoprim/sulfonamide, Others
florfenicol, enrofloxacin or ceftiofur (Table 7.8).
Trimethoprim/sulfonamide is not available as a feed Staphylococcus hyicus can be quite a difficult condi-
premix or water soluble in the USA for use in pigs, tion to treat on some farms (Table 7.12) and the
which may account for its high activity, in contrast occurrence of resistance varies considerably between
to tetracyclines, which are also commonly used in the countries (62). Resistances to macrolides, tetracy-
EU. Fluoroquinolones are also not licensed for use in clines, penicillin, streptomycin or sulfonamides are
swine in the USA, which may account for their lack frequently observed in most countries. Thus, like for
of resistance, but ceftiofur is widely available and no E. coli, resistance in S. hyicus is difficult to predict and
resistance is displayed. choice of the most appropriate antimicrobial drug

Guidelines for antimicrobial use in swine 123

for empiric treatment has to be done on the basis of 7. Grave, K., Jensen V.F., Odensvik, K. Wierup, M. and 7 Swine
knowledge of the individual herd and data on suscep- Bangen, M. (2006). Usage of veterinary therapeutic anti-
tibility at the farm level. microbials in Denmark, Norway and Sweden following
termination of antimicrobial growth promoter use. Prev.
7.4 Conclusions Veter. Med. 75 (1–2): 123–32.

Antimicrobials are, and will be, essential to maintain 8. Nielsen, J.P. and Stege, H.S. (2006). Consumption of pre-
the health, welfare and productivity of pigs. It is criti- scribed antimicrobials after the growth promoter termi-
cal to retain the use and maintain the effectiveness nation in Denmark. Proceedings of the 19th International
of the antimicrobials that are currently available to Pig Veterinary Society Congress. Copenhagen, Denmark, 1:
combat current and future disease problems as they 191.
arise. There is therefore a responsibility on behalf of
the users of these medicines to preserve their efficacy. 9. Danmap 2005 (2006). Antimicrobial consumption. In Use
If antimicrobials are withdrawn, as has been seen in of Antimicrobial Agents and Occurrence of Antimicrobial Res-
the USA with fluoroquinolones for poultry, they will istance in Bacteria from Food Animals, Foods and Humans in
never be restored and important last resort drugs Denmark (eds. Heuer, O.E. and Hammerum, A.E.). Danish
will be lost for future use in veterinary medicine. Institute for Food and Veterinary Research and Danish
The responsibility lies with the veterinarians or the Zoonosis Centre, Søborg, Denmark, pp. 15–16.
farmers, depending on the national legislation, to use
antimicrobials in a responsible way, to minimise their 10. AnimalHealthInstitute(1995).FeedAdditiveCompendium.
use and seek alternatives without compromising the The Miller Publishing Company, Minnetonka, Minnesota,
health of the animal. USA.

References 11. Anonymous (1999). Proposed rules. US Federal Register
64 (127): 35966–72.
1. Anonymous (2006). Farms, land in farms and livestock
operations, 2005 Summary, Agriculture Statistics Board, 12. Anonymous (2002). Extralabel drug use in animals. 21
National Agricultural Statistical Service, USDA, 34–5. Code of Federal Register, Chapter 1, part 530.II. 4/1/2002
Available at: http://usda.mannlib.cornell.edu/usda/nass/ Edition. US Federal Government, Washington DC.
FarmLandIn//2000s/2006/FarmLandIn-01-31-2006.pdf2
Accessed on: May 2007. 13. Harris, D.L. (2000). Exclusion and elimination of microbes.
In Multi-site Pig Production (ed. Harris, D.L.). Iowa State
2. Anonymous (2005). Pork Powerhouses, 2005. Successful University Press, Ames, Iowa, USA, pp. 57–78.
farming, http://www.agriculture.com/ag/pdf/2005-pork-
powerhouses.pdf, 16 October. Accessed on: May 2007. 14. Nielsen, P. (1997). The influence of feed on the oral bio-
availability of antibiotics/chemotherapeutics in pigs.
3. Vivash-Jones, B. (2000). COMISA report. The year in J. Vet. Pharmacol. Ther. 20 (Supplement 1): 30–1.
review. COMISA, Brussels, Belgium.
15. Veterinary Laboratories Agency (VLA) (2005).
4. Veterinary Medicines Directorate (2005). Sales of antimi- Antimicrobial sensitivity in Salmonella. In Salmonella in
crobial products authorised for use as veterinary medicines, Livestock Production in GB – 2004 (eds. Davies, R. and
antiprotozoals, antifungals, growth promoters and coccid- Kidd, S.A.). VLA, Weybridge, UK, pp. 144–55.
iostats, in the UK in 2004 (ed. Goodyear, K.). Veterinary
Medicines Directorate, New Haw, Surrey, UK. 16. Annual report on zoonoses – http://www.dfvfdk/Default.
aspx?ID=9202#74145, Accessed on: May 2007.
5. SVARM, 2005 (2006). Use of antimicrobials. In Swedish vet-
erinary resistance monitoring (eds. Bengtsson, B., Greko, C. 17. Barza, M. and Travers, K. (2002). Excess infections due to
and Grönlund-Andersson, U.). National Veterinary Institute, antimicrobial resistance: the ‘Attributable Fraction’. Clin.
Uppsala, Sweden, pp. 8–12. Infect. Dis. 34 (Suppl 3): S126–30.

6. Danmap 2004 (2005). Antimicrobial consumption. In Use 18. Holmberg, S.D., Wells, J.G. and Cohen, M.L. (1984).
of Antimicrobial Agents and Occurrence of Antimicrobial Animal-to-man transmission of antimicrobial-resistant
Resistance in Bacteria from Food Animals, Foods and Salmonella: investigations of U.S. outbreaks, 1971–1983.
Humans in Denmark (eds. Emborg, H.-D., Heuer, O.E. and Science 225: 833–5.
Larsen, P.B.). Danish Zoonosis Centre and Danish Institute
for Food and Veterinary Research, Søborg, Denmark, 19. Lee, L.A., Puhr, N.D., Maloney, E.K., Bean, N.H. and
pp. 15–22. Tauxe, R.V. (1994). Increase in antimicrobial-resistant
Salmonella infections in the United States, 1989–1990.
J. Infect. Dis. 170: 128–34.

20. Martin, L.J., Fyfe, M., Dore, K., et al. (2004). Multi-
provincial Salmonella typhimurium case–control study
steering committee. Increased burden of illness associ-
ated with antimicrobial-resistant Salmonella enterica
serotype typhimurium infections. J. Infect. Dis. 189:
377–84.

21. Varma, J.K., Mølbak, K., Barrett, T.J., et al. (2005).
Antimicrobial-resistant nontyphoidal Salmonella is asso-
ciated with excess bloodstream infections and hospitaliza-
tions. J. Infect. Dis. 191: 554–61.

124 Guide to antimicrobial use in animals

7 Swine 22. Helms, M., Vastrup, P., Gerner-Smidt, P. and Mølbak, K. 36. de Neeling, A.J., van den Broek, M.J.M., Spalburg, E.C.,
(2002). Excess mortality associated with antimicrobial et al. (2007). High prevalence of methicillin resistant
drug-resistant Salmonella typhimurium. Emerg. Infect. Dis. Staphylococcus aureus in pigs. Vet. Microbiol. 122 (3–4):
8: 490–5. 366–72.

23. Helms, M., Simonsen, J. and Mølbak, K. (2004). Quinolone 37. Guardabassi, L., Stegger, M. and Skov, R. (2007).
resistance is associated with increased risk of invasive Retrospective detection of methicillin resistant and sus-
illness or death during infection with Salmonella serotype ceptible Staphylococcus aureus ST398 in Danish slaughter
Typhimurium. J. Infect. Dis. 190: 1652–4. pigs. Vet. Microbiol. 122 (3–4): 384–6.

24. Burch, D.G.S. (2002). Risk assessment – Campylobacter 38. Armand-Lefevre, L., Ruimy, R. and Andremont, A. (2005).
infection transmission from pigs to man using erythro- Links clonal comparison of Staphylococcus aureus isolates
mycin resistance as a marker. Pig J. 50: 53–8. from healthy pig farmers, human controls and pigs. Emerg.
Infect. Dis. 11 (5): 711–4.
25. Wesley, I.V., Muraoka, W.T., Trampel, D.W. and Hurd, H.S.
(2005). Effect of preslaughter events on prevalence of 39. Witte, W., Strommenger, B., Stanek, S. and Cuny, C.
Campylobacter jejuni and Campylobacter coli in market- (2007). Methicillin-resistant Staphylococcus aureus ST398
weight turkeys. Appl. Environ. Microbiol. 71 (6): 2824–33. in humans and animals, Central Europe. Emerg. Infect.
Dis. 13: 255–8.
26. Kramer, J.M., Frost, J.A., Bolton, F.J. and Wareing, D.R.A.
(2000). Campylobacter contamination of raw meat and 40. Baker, R.B. (2005). Strategies and techniques for dis-
poultry at retail sale: identification of multiple types and ease eradication: eradication of multiple diseases from
comparison with isolates from human infection. J. Food herds. Proceedings of the 4th International Swine Disease
Prod. 63 (12): 1654–59. Eradication Symposium, Allen D. Leman Conference, Saint
Paul, Minnesota, 17–20 September 2005, 1–5.
27. Hopkins, K.L., Desai, M., Frost, J.A., Stanley, J and
Logan, J.M.L. (2004). Fluorescent amplified fragment 41. Harris, D.L. and Alexander, T.J.L. (1999). Methods of
length polymorphism genotyping of Campylobacter disease control. In Diseases of Swine, 8th edn (eds. Straw,
jejuni and Campylobacter coli strains and its relationship D’Allaire, Mengeling and Taylor). Iowa State University
with host specificity, serotyping and phage typing. J. Clin. Press, Ames Iowa, pp. 1077–110.
Microbiol. 42 (1): 229–35.
42. Hennecken, M., Stegeman, J.A., Elbers, A.R., van Nes. A.,
28. Helms, M., Simonsen, J., Olsen, K.E. and Molbak K. Smak, J.A. and Verheijden, J.H. (2000). Transmission of
(2005). A study of adverse health events associated classical swine fever virus by artificial insemination during
with antimicrobial drug resistance in Campylobacter spe- the 1997–1998 epidemic in The Netherlands: a descriptive
cies: a registry-based cohort study. J. Infect. Dis. 191 (7): epidemiological study. Vet. Quart. 22 (4): 228–33.
1050–5.
43. Guerin, B. and Pozzi, N. (2005). Viruses in boar semen:
29. Teale, C. (2002). Antimicrobial resistance in porcine detection and clinical as well as epidemiological con-
bacteria. Pig J. 49: 52–69. sequences regarding disease transmission by artificial
insemination. Theriogenology 63 (2): 556–72.
30. Barlow, A.M., Hunt, B.W., Heath, P.J. and Smith, R.M.M.
(2003). The prevalence and clinical diseases caused in pigs 44. Brumm, M. (2004). Housing decisions for the growing
by different serotypes of Streptococcus suis (June 2000 to pig. Proceedings of the London Swine Conference – Building
September 2002) and human infection (1981 to October Blocks for the Future 1–2 April, 2004, 31–44.
2002) in England and Wales. Pig J. 51: 164–76.
45. Moore,C.(2003).Parity segregations,successes and pitfalls.
31. Yu, H., Jing, H., Chen, Z., et al. (2006). Streptococcus suis Proceedings of the A.D.Leman Conference, Minneapolis,
study groups. Human Streptococcus suis outbreak, Sichuan, MN, USA, 36–42.
China. Emerg. Infect. Dis. 12 (6): 914–20.
46. Main, R.G., Dritz, S.S., Tokach, M.D., Goodband, R.D. and
32. Ye, C., Zhu, X., Jing, H., et al. (2006). Streptococcus suis Nelssen, J.L. (2004). Increasing weaning age improves pig
sequence type 7 outbreak, Sichuan, China. Emerg. Infect. performance in a multisite production system. J. Anim.
Dis. 12 (8): 1203–8. Sci. 82 (5): 1499–1507.

33. Moodley, A., Stegger, M., Bagcigil, A. F., et al. (2006). PFGE 47. Wathes, C and Whitehouse, C. (2006). Environmental
and spa typing of methicillin-resistant Staphylococcus management of pigs. In Whittemore’s Science and Practice
aureus isolated from domestic animals and veterinary of Pig Production (eds. Kyriazakis, I. and Whittemore, C.T.).
staff in the UK and Ireland. J. Antimicrob. Chemother. 58: Blackwell Publishing, Oxford, UK, pp. 533–90, Chapter 17.
1118–23.
48. Muirhead, M.R. and Alexander, T.J.L. (eds.) (1997).
34. Voss, A., Loeffen, F., Bakker, J., Klaassen, C. and Wulf, M. Reproduction non-infectious infertility. In Managing Pig
(2005). Methicillin-resistant Staphylococcus aureus in pig Health and the Treatment of Disease: A Reference for the
farming. Emerg. Infect. Dis. 11 (12): 1965–6. Farm. 5M Enterprises, Sheffield, UK, pp. 140–1.

35. Wulf, M.W.H., Van Nes, A., Eiklenboom-Boskamp, A., 49. Pesente, P., Rebonato, V., Sandri, G., Giovanardi, D.,
et al. (2006). MRSA-prevalence in Dutch veterinarians Ruffoni, L.S. and Torriani, S. (2006). Phylogenetic analysis
and veterinary students. Proceedings 19th International of ORF5 and ORF7 sequences of porcine reproductive and
Pig Veterinary Society Congress. Copenhagen, Denmark, respiratory syndrome virus (PRRSV) from PRRS-positive
1: 193. Italian farms: a showcase for PRRSV epidemiology and

Guidelines for antimicrobial use in swine 125

its consequences on farm management. Vet. Microbiol. 61. Gutierrez-Martin, C.B., del Blanco, N.G., Blanco, M., 7 Swine
114 (3–4): 214–24. Navas, J. and Rodríguez-Ferri, F.E. (2006). Changes in
50. Thacker, E. and Thacker, B. (2000). Factors affecting antimicrobial susceptibility of Actinobacillus pleuropneu-
Mycoplasma hyopneumoniae vaccine efficacy. Proceedings moniae isolated from pigs in Spain during the last decade.
of the 16th International Pig Veterinary Society Congress. Vet. Microbiol. 115: 218–22.
Melbourne 17–20 September 2000, 164.
51. Jayappa, H., Davis, R., Rapp-Gabrielson, V., Wasmoen, T., 62. Aarestrup, F.M. and Schwarz, S. (2006). Staphylococci
Thacker, E.L. and Thacker, B. (2001). Evaluation of effi- and streptococci. In Antimicrobial Resistance in Bacteria
cacy of M. hyopneumoniae bacterin following vaccination of Animal Origin (ed. Aarestrup, F.M.). ASM Press,
of young pigs in the presence of varying levels of mater- Washington, DC, pp. 187–206.
nal antibodies. Proceedings of the American Association of
Swine Veterinarians, March 2001, pp. 237–41. 63. Karlsson, M., Oxberry, S.L. and Hampson, D.J. (2002).
52. Brockmeier S.L., Halbur, P.G. and Thacker, E.L. (2002). Antimicrobial susceptibility testing of Australian isolates
Porcine respiratory disease complex. In Polymicrobial of Brachyspira hyodysenteriae using a new broth dilution
Diseases (eds. Kim A. Brogden and Janet M. Guthmiller). method. Vet. Microbiol. 84: 123–33.
ASM Press, Washington, DC, USA, pp. 231–59.
53. Desrosiers, R., Clark, E., Tremblay, D. and Tremblay, R. 64. Cizek, A., Lobova, D. and Smola, J. (2002). In vitro sus-
(2007). Proceedings 38th American Association of Swine ceptibility of Brachyspira hyodysenteriae strains isolated
Veterinarians annual Meeting, March 2007. Orlando, USA, in the Czech Republic from 1996–2001. Proceedings 17th
pp. 143–5. International Pig Veterinary Society Congress. Ames, Iowa,
54. Franklin, A., Pringle, M. and Hampson, D.J. (2006). USA, 2, 191.
Antimicrobial resistance in Clostridium and Brachyspira
spp. and other anaerobes. In Antimicrobial Resistance 65. Kinyon, J.M., Murphy, D., Stryker, C., Turner, V.,
in Bacteria of Animal Origin (ed. Aarestrup, F.M.). ASM Holck, J.T. and Duhamel, G. (2002). Minimum inhibitory
Press, Washington, DC, pp. 127–44. concentration for US swine isolates of Brachyspira pilosicoli
55. Mackie, R.A. (1996). Masters by Research Thesis: to valnemulin and four other antimicrobials. Proceedings
‘An in vitro study of antimicrobial agents against the obli- 17th International Pig Veterinary Society Congress. Ames,
gately intracellular bacterium Lawsonia intracellularis.’ Iowa, USA, 2, 50.
University of Edinburgh, Scotland.
56. Wattanaphansak, S., Gebhart, C., Singer, R. and Dau, D. 66. Dutta, G.N. and Devriese, L.A. (1980). Susceptibility
(2007).In-vitro testing of antimicrobial agents for Lawsonia of Clostridium perfringens of animal origin to
intracellularis. Proceedings 38th American Association of fifteen antimicrobial agents. J. Vet. Pharmacol. Therap.
Swine Veterinarians Meeting, Orlando, USA, 255–6. 3: 227–36.
57. Post, K.W. and Songer, J.G. (2002). Antimicrobial sus-
ceptibility of Clostridium difficile isolated from neonatal 67. Devriese, L.A., Daube, G., Hommez, J. and Haesebrouck, F.
pigs. Proceedings 17th International Pig Veterinary Society (1993). In vitro susceptibility of Clostridium perfringens
Congress, Ames, Iowa, USA, 2, 62. isolated from farm animals to growth-enhancing antibi-
58. Aarestrup, F.M. and Kempf, I. (2006). Mycoplasma. In otics. J. Appl. Bacteriol. 75: 55–7.
Antimicrobial Resistance in Bacteria of Animal Origin
(ed. Aarestrup, F.M.). ASM Press, Washington, DC, 68. Hannan, P.C.T., Windsor, G.D., de Jong, A., Schmeer, N.
pp. 239–48. and Stegeman, M. (1997). Comparative susceptibilities
59. Teale, C.J., Martin, P.K. and Watkins, G.H. (2004). VLA of various animal-pathogenic mycoplasmas to fluoroqui-
Antimicrobial Sensitivity Report 2003. Crown Copyright nolones. Antimicrob. Agents Chemother. 41 (9): 2037–40.
Unit, Norwich, UK.
60. Salmon, S., Portis, E. and Lindeman, C. (2003). Minimum 69. Vicca, J., Stakenborg, T., Maes, D., et al. (2004).
inhibitory concentrations for Ceftiofur and compara- In-vitro susceptibilities of Mycoplasma hyopneumoniae
tor antimicrobial agents against bacterial pathogens field isolates. Antimicrob. Agents Chemother. 48 (11):
of swine. Proceedings of the American Association of 4470–2.
Swine Veterinarians Congress, Orlando, Florida, USA,
pp. 235–9. 70. Trigo, E., Mendez-Trigo, A.V. and Simonson, R. (1996).
Antimicrobial susceptibility profiles of Haemophilus
parasuis. A retrospective study from clinical cases submit-
ted during 1994 and 1995 to a veterinary diagnostic labo-
ratory. Proceedings of the 14th International Pig Veterinary
Society Congress. Bologna, Italy, 313.

71. Aarestrup, F.A., Seyfarth, A.M. and Angen, O. (2004).
Antimicrobial susceptibility of Haemophilus parasuis, and
Histophilus somni from pigs and cattle in Denmark. Vet.
Microbiol. 101: 143–6.

8 Poultry Chapter 8

Guidelines for antimicrobial use
in poultry

Ulrich Löhren, Antonia Ricci and Timothy S. Cummings

8.1 Antimicrobial use in poultry The Swann report (1) that was published in 1969
recommended that therapeutic antimicrobials should
Antimicrobial agents used in poultry include growth not be used as growth promoters. This resulted in most
promoters, coccidiostats and antimicrobials for countries adapting their legislation over time, so that
therapeutic or prophylactic use. All these forms of certain antimicrobial products were either banned or
antimicrobial treatment are briefly described in the had to be used either as feed additives (without veteri-
following sections. narian prescription) or as therapeutic products under
veterinarian prescription only. This attitude was more
8.1.1 Growth promoters or less strictly applied by all food animal producing
countries of Western Europe and North America.
Antimicrobials were first used for growth promotion
purposes as early as the late 1940s when it was discov- In the mid-1990s, certain poultry integrators in
ered that chickens grew faster when fed tetracycline Europe made significant efforts to improve hygiene,
fermentation by-products. Subsequently, other anti- disinfection and biosecurity to reduce bacterial loads
microbials were approved for growth promotion and as a precursor to reducing the use of antimicrobial
performance enhancement over the years. Initially, growth promoters. In 1997, avoparcin was banned
antimicrobials like tetracyclines, tylosin and bacitra- from use in the EU, followed by the ban of other anti-
cin could be used in poultry at low concentrations microbial growth promoters (virginiamycin, bacitra-
as feed additives for ‘growth promotion’, whereas cin, spiramycin and tylocin) in 1999. Following the
higher dosages were restricted to veterinary use. precautionary principle, the use of growth promot-
ers was prohibited in the EU because it was liable

Guide to Antimicrobial Use in Animals. Edited by Luca Guardabassi, Lars B. Jensen and Hilde Kruse
© 2008 Blackwell Publishing Ltd. ISBN: 978-1-4051-5079-8

Guidelines for antimicrobial use in poultry 127

to induce resistance to antimicrobial drugs used in 8.1.3 Therapeutic antimicrobials 8 Poultry
human medicine. In 2006, all remaining growth pro-
moters were banned from use in animal feeds in the Antimicrobials for therapeutic use are, in most coun-
EU, but the USA and other Third Countries have not tries, regulated by specific veterinarian or pharmaceuti-
introduced similar restrictions. These actions have cal legislation. Their use in most countries is restricted
been strenuously debated due to the lack of conclu- to veterinarian prescription. Misuse and overuse of
sive scientific evidence to support all the bans, but antimicrobials occurs more easily in countries where
the overall worldwide usage of antimicrobial use as the farmer has easy access to antimicrobials not requir-
growth promoters is a downwards trend (2). ing veterinary prescription. In these cases, antimicrobi-
als tend to be used on a trial and error basis (Donoghue,
8.1.2 Coccidiostats 2006; personal communication), in an effort to elicit a
favourable response, which can be difficult to obtain or
Intensive production of commercial poultry over the interpret depending upon the characteristics of the bac-
past 60 years has been largely due to the introduction terial pathogen being treated (see Section 8.4).
of coccidiostats in the feed. These products interfere
with various stage(s) of intestinal development of the In avian medicine, antimicrobials can be applied
coccidia. In the early days, sulfonamides were primar- to the target animal by individual injection or oral
ily used as coccidiostats, and they are still registered application (to pet birds or valuable stocks only), or
for veterinary prescription for therapy of coccidiosis. by mass application to the whole flock via drinking
The vast majority of coccidiostats are regulated by water (major way of administration) or feed (used on
feed legislation as feed additives. In the 1980s, a new a limited basis). Individual injection or application
group of coccidiostats was added: the polyether iono- (e.g. by oral gavage) will rarely be an option due to
phores. The main target and purpose of ionophores the sheer numbers of birds involved. The most practi-
is coccidiosis control, but this group also has limited cal method of application of therapeutic substances
antibacterial activity, especially against Clostridium. (including antimicrobials) is by oral administration,
For this reason, ionophores were and are used almost either via drinking water or feed. Whereas individual
exclusively as coccidiostats. Their significance has therapy in large animals is often done by the farmer
increased in the EU since antimicrobial growth pro- himself, the owners of large poultry flocks are typi-
moters were banned. Worldwide, ionophores are still cally fully integrated, commercial poultry companies
the mainstay of most programmes for the control of in industrialized nations. These companies often have
coccidiosis. Ionophores are not perceived as antimi- veterinarians on staff or have access to qualified vet-
crobials by most public health authorities, as these erinary poultry expertise and diagnostic facilities to
agents are not used in human medicine. help make prudent therapeutic decisions. In these
instances, therapeutic choices are made with eco-
Live vaccines have been developed to prevent coc- nomic, as well as efficacy and welfare, issues factoring
cidiosis. These vaccines are a valuable tool for com- into the decision, because medication of large groups
bating the loss of efficacy by coccidiostats against of animals can be very costly.
resistant Eimeria strains, and could eventually replace
coccidiostats if the ionophores are banned as feed In contrast to large animal practice, poultry vet-
additives. There is circumstantial evidence that non- erinarians or the farmer can easily justify sacrificing
specific enteritis occurs more frequently in flocks a few sick birds or taking some fresh dead carcasses
vaccinated with live vaccines, as compared to flocks from the flock to a specialized diagnostic laboratory.
reared on ionophore coccidiostats, which has been There, a necropsy is typically performed with individ-
associated with the cycling of the coccidial vaccine ual samples taken for cultivation and identification of
strains in the bird’s intestine. For this reason, specific the causative organism. Antimicrobial susceptibility
antimicrobial treatment can be needed if the enteritis patterns on any resultant isolate are routinely per-
becomes significant. Further steps into changing dis- formed as well. Hence, the modern poultry veterinar-
tribution and legislation on the use of coccidiostats ian bases his diagnosis and resultant therapy on the
should be carefully considered, as their removal could clinical picture of the flock, bird pathology, bacteriol-
have significant implications for the poultry indus- ogy and history of the problem at hand. This is stan-
tries of the world. dard procedure for poultry medicine in large poultry
producing regions of the world.

128 Guide to antimicrobial use in animals

8 Poultry It should be noted that actual treatment of poultry is a concern about the effectiveness of alternative
flocks has never been an extensive practice due to the approaches for therapy of diseased poultry flocks.
relatively good health status of poultry flocks, the rel- These approaches are often less effective, and ani-
atively few efficacious antimicrobials available for use mal welfare considerations should also be respected.
and the prohibitive expense of medicating flocks. As Although the commercial aspect of treating diseased
a result, antimicrobial therapy of chicken flocks in the flocks with antimicrobials should never compromise
USA and Europe has been decreasing in recent years public health, neither should their removal be allowed
and a shift in the use of therapeutic antimicrobials without proper scientific evidence or carefully
has been observed in Europe (3). A small increase in designed and interpreted risk assessments. On the
the use of therapeutic antimicrobials coincided with other hand, antimicrobials should never be used for
the growth promoter ban, which was suggested as an prophylactic purposes to substitute for poor hygiene
undesirable side effect of the ban (4). or management.

The issue of antimicrobial residues became more of 8.1.4 Guidelines and codes
a concern in the 1980s, so withdrawal times (different of practice
durations for the same product in different countries)
for antimicrobials were introduced. From the begin- With respect to the necessary therapeutic use of
ning of antimicrobial residue testing, the poultry antimicrobials in food animals, codes of practice
industry has always exercised responsible use of anti- have been agreed upon by various national veteri-
microbials to allow for proper withdrawal times, as nary associations (7). Although such codes of prac-
residues in edible tissues would have an impact on the tice are not as binding as legislation and are largely
use of all meat (or eggs) from the whole flock (instead voluntary, they have made a big impact on the veter-
of an individual animal) for human consumption. As inary therapeutic use of antimicrobials in veterinary
a result, antimicrobial residues in poultry flocks at medicine (8). Most of these guidelines share com-
time of slaughter or in poultry meat (or eggs) have mon principles. A good example is the ‘Guidelines
rarely been a problem. for prudent use of antimicrobials in animals’, which
was published by the German Federal Veterinarians
In Europe, only those antimicrobials for which a Association (BTK) and a Working Group of Senior
MRL value (maximum residue level) according to Veterinarians (ArgeVet) in 2000 (9). The scope of
the procedure laid down in regulation 2377/90 is set these guidelines is to minimize the impact of anti-
may be used in food delivering poultry flocks. Table microbial usage on development of resistance in
8.1 lists those antimicrobials for which a MRL value animals, and they should be regarded as the mini-
is set and therefore may be used in Europe. The with- mum requirement that must always be followed by
drawal times are based on the MRL and on the differ- veterinarians when administering antimicrobials to
ent pharmacology of the antimicrobial molecules. animals. The guidelines constitute the rules of vet-
erinary science (good veterinary practice), which
Although not a new debate, the concern about are to be complied with during any use of antimi-
the rise of antimicrobial resistance in certain human crobials in animals, and which must be observed
pathogens began to surface again in the 1990s. Despite each and every time an animal (or a poultry flock)
the general consensus that the increase in bacterial is treated properly in accordance with the national
resistance in human medicine has been strongly asso- drug legislation. Antimicrobials should only be
ciated with overuse by physicians (5), the legislative prescribed by veterinarians. They may be used by
emphasis continues to focus on what extent the use the animal (or poultry flock) owner only according
of therapeutic antimicrobials in food animals has to written instructions under veterinary supervi-
contributed to the increasing antimicrobial resistance sion. The prescribing veterinarian must check this
issue in human medicine (6). Currently in the USA, at suitable intervals by monitoring the success of
legislation is routinely proposed to ban or limit anti- treatment.
microbial use. The use of fluoroquinolones in poul-
try was banned in 2004. Many poultry companies are The use of antimicrobials is only justified
reducing antimicrobial use at the request of their cus- for therapy if it has been proven by appropriate
tomers or to meet export requirements.

There are a few other issues to be considered
concerning the use of antimicrobials in poultry. There

Guidelines for antimicrobial use in poultry 129

Table 8.1 Major antimicrobial classes used in avian medicine

Antimicrobial class Drug name Type of activity Intestine Spectrum 8 Poultry
Sulfonamides absorption
Potentiated Several compounds are Bacteriostatic of activity
sulfonamides available in this class Good
Aminoglycosides Trimethoprim and Bactericidal Gram +
sulfonamides Good Gram –
β-Lactames Apramycin Bactericidal
Gentamicin Bactericidal Poor Gram +
Neomycin Bactericidal None Gram –
Spectinomycin Bactericidal Poor
Streptomycin Bactericidal Intermediate Mainly Gram –
Dihydrostrepto-mycin Bactericidal Poor Mainly Gram –
Benzylpenicillin Bactericidal Poor Mainly Gram –
Potassium Pen. G Good Mainly Gram –
Ampicillin Bactericidal Mainly Gram –
Intermediate Mainly Gram –
Amoxicillin Bactericidal
Good Gram +
Ceftiofur Bactericidal
Cannot be given Gram +
Fluoroquinolones Enrofloxacin Bactericidal orally (Gram –)
Lincosamides Difloxacin Bactericidal Very good Gram +
Flumequin Bactericidal Good (Gram –)
Bacteriostatic Good Gram +
Lincomycin Good Gram –
Bacteriostatic
Macrolides Erythromycin Good Gram ±
Spiramycin Bacteriostatic Gram ±
Tylosin Good Gram ±
Timicosin Bacteriostatic
Good Gram +
Pleuromutilines Tiamulin Bacteriostatic Mycoplasma
Polypeptides Good
Tetracyclines Colistin sulfate Bacteriostatic Gram –
Bacteriocidal Good Mycoplasma
Tetracycline Bacteriostatic None Gram –
Chlortetracycline Bacteriostatic Intermediate Mycoplasma
Oxytetracycline Bacteriostatic Good Gram –
Doxycycline Bacteriostatic Good Mycoplasma
Good Gram –
Mycoplasma

Mycoplasma

Gram –

Gram ±
Gram ±
Gram ±
Gram ±

diagnostic measures that the animals are infected by identification of the pathogen and therapeutic
a pathogen susceptible to the antimicrobial that is to options should be guided by susceptibility testing,
be administered. In veterinary practice, prophylaxis knowledge of local resistance patterns,history of anti-
is only admissible in substantiated exceptional cases microbial efficacy in the field, expense and relative
(immunosuppressed animals, long and/or elective sur- importance of any antimicrobial options to human
gery, etc.), which are not applicable to commercial poul- medicine. Diagnosis and susceptibility testing are
try flocks. The diagnosis shall generally be based on always required when switching the therapy to

130 Guide to antimicrobial use in animals

8 Poultry another antimicrobial agent, when considering 8.2 Major antimicrobials used
compounding (mixing of drugs for use in combina- in avian medicine
tion), or when using the antimicrobial in an extra-
label manner (not used in compliance with the label The most important antimicrobials used in avian
instructions). medicine are listed in Table 8.1.

With respect to national legislation procedures, a Sulfonamides are the oldest antimicrobials with
solution needs to be found that enables the pharma- minimal application significance in human medicine,
ceutical producers to get easier registration of new and are therefore regarded as first choice products.
claims for existing antimicrobials or new antimi- Due to their toxicity, their small therapeutic scale and
crobials for food animals. This is especially true for their longer withdrawal times, they are not used to a
species like chickens, which have been restricted in great extent. Potentiated sulfonamides (combinations
usage of many feed and/or water antimicrobials in with trimethoprim) are much more suitable for the
some countries in recent years. Certain opportuni- same indications. They possess anticoccidial activity
ties exist in some regions, where some poultry spe- and should not be used within the first three weeks
cies (like turkeys and ducks) are considered a minor after live vaccination against coccidiosis.
animal species. But a dilemma can present itself
at times with regard to label claims. For example, Penicillins have been used for decades in human
enrofloxacin is one of the very few, highly effica- medicine. Some penicillins are inactivated by the
cious, registered antimicrobials for specific diseases presence of hydrochloric acid in the proventriculus.
in turkeys in certain countries. However, this is also Only benzylpenicillin and penicillin V potassium can
one of the critically important antimicrobial classes be given orally. Penicillins are first choice products
in human medicine (see Chapter 4) and according against Clostridium infections in poultry. Ampicillin
to most judicious guidelines one should first con- and amoxicillin belong to the group of aminopeni-
sider the use of antimicrobial products according to cillins. Both are regarded as first choice antimicrobi-
their label indications. als in avian medicine, although they still have some
impact in human medical usage. Their choice in avian
As alluded to earlier, previous guidelines on pru- medicine should be based on registration, withdrawal
dent use of antimicrobials are widely accepted in times and the degree of systematic efficacy, which is
avian medicine. The development of these guide- needed under the actual therapeutic situation. Both
lines are timely and beneficial to the veterinary and have limited solubility in higher concentrations (like
medical professions as a whole, in addition to help- that needed for dosatrons) and their stability in drink-
ing fulfil portions of the veterinarian’s obligation to ing water is limited. Solutions should be renewed
use scientific knowledge to promote public health every 8 h.
and relieve animal suffering. Poultry veterinarians
as a group are highly specialized, and have typically Polypeptides (like colistinsulfate or polymixin
striven for a more precise clinical and microbiologi- E) are used in human medicine mainly for topical
cal diagnosis. Whereas in the past, in some coun- application (too toxic for systemic use). They can
tries, the use of antimicrobials has been regarded as therefore be considered in avian medicine as a first or
a tool to control zoonotic bacterial infections like second choice product. Their in vitro activity against
Salmonella in poultry, it is generally accepted that Gram-positive bacteria remains excellent. Although
other measures should be used to control food- not very well absorbed, some efficacy is seen in the
borne pathogens. EU Decision 1177/06 clearly field after a prolonged oral administration (mini-
states that antimicrobials may never be used as a mum 1 week) against systematic E. coli. This may
control method within a specific Salmonella control help to avoid the usage of third choice products like
programme. The EU poultry industry takes the lead quinolones.
and unanimously supports this legislation: no use
of antimicrobials to control Salmonella. This has Lincosamides (available in combination with specti-
been the course of action in the USA as well, with nomycin or as the sole antimicrobial) are used as a
Hazard Analysis Critical Control Point (HACCP) starter medication for broiler chicks in some countries,
principles being implemented at the processing like the UK. This practice should not be encouraged.
plants to reduce pathogen loads.
Cephalosporins have a high importance in human
medicine. They are therefore considered as third choice
antimicrobial in avian medicine (product of last

Guidelines for antimicrobial use in poultry 131

reserve). Their activity against both Gram-positive and registered for these species and only newer products 8 Poultry
Gram-negative bacteria is excellent. Because cepha- with broader spectrum are available according to label
losporins have to be injected, they are very rarely used claims. For example, in Germany, two fluoroquinolo-
with poultry and then only under very limited condi- nes (enrofloxacin and difloxacin) are registered for
tions and only with very valuable poultry stocks. use in turkeys, while a first or second choice product
should only be used in the framework of the guide-
Among the quinolones, the first-generation product line cascade. This is also true for ampicillin, amoxi-
is flumequine and the second-generation products are cillin, colistin, potentiated sulfonamides, tetracyclines
enrofloxacin, difloxacin and norfloxacin. Flumequine and chlortetracycline, which all have no registration
is only registered in a few countries. Within the sec- for turkeys. The situation is even worse for ducks. To
ond-generation quinolones, enrofloxacin has by far safeguard the option of treating flocks of the so-called
the greatest significance in avian medicine. Because a minor species (turkeys, ducks) with products of first
closely related product (ciprofloxacin) is still consid- or second choice, it would be highly beneficial if some
ered as the drug of choice for many human bacterial type of international standard was adopted to provide
diseases, these antimicrobial agents should be consid- a practical, streamlined process for registration of
ered as product of last reserve (third choice product) older antimicrobials or to obtain new label claims. This
in avian medicine. Quinolones are highly water solu- type of activity should most likely arise from an inter-
ble and can reach high tissue levels after oral applica- national agency such as European Medicines Agency
tion. If used in a rational way (10 mg/kg/body weight (EMEA) or Codex Alimentarius (see Chapter 5).
for minimum 3 days), the prevalence of resistance
against these antimicrobials usually remains low. With the relatively short lifespan of the meat-
producing poultry species (broilers, turkeys and
8.3 Registration of antimicrobials ducks), the withdrawal time is of importance for
for use in poultry choosing the best antimicrobial. Many times, a health
problem will develop just prior to processing, which
It is the intention of this book, as well as of the various significantly limits available therapeutic options. As a
national guidelines and Codes of practice, to encour- safety precaution, current legislation typically overes-
age veterinarian use of ‘first choice’ products initially. timates the importance of antimicrobial residues in
These are products with an antibacterial spectrum as meat products, but does not take into account the safe-
narrow as possible and/or with limited importance guard of critically important antimicrobial agents in
in human medicine. Many of these antimicrobials human medicine. There is concern about the varying
are older products, which often require re-registra- withdrawal times in different EU Member States for
tion in the light of higher registration standards. For the same antimicrobial with the same EU MRL meat
this reason, some older products have lost their old value, whereas the final poultry meat products can
general registration for poultry in some countries. be traded globally in many countries without major
Pharmaceutical producers will have to submit separate restrictions. Certain anti-inflammatory substances,
registration data for chicken (broilers, rearing pullets like acetylsalicylic acid, may be indicated under certain
and layers) and for minor avian species like turkeys, circumstances instead of an antimicrobial, but these
ducks, geese and guineafowl. Unfortunately, many are not registered for poultry in most countries.
antimicrobials (especially those of first choice) are not
registered for use in poultry in many countries. For 8.4 Resistance trends in poultry
example, in the Netherlands, there was until recently pathogens and zoonotic bacteria
no oral penicillin (like benzylpenicillin or penicillin
G) registered for any avian species. Hence, the poultry Antimicrobial resistance has been extensively studied
veterinarians have to go for a second choice product in certain poultry pathogens and zoonotic bacteria.
with broader spectrum like ampicillin or amoxicillin. As far as Salmonella is concerned, the occurrence of
resistance seems to have increased over the years, and
With regard to turkeys and ducks, these cannot is associated with the selective pressure exerted by
be regarded as minor species. However, in most the use of antimicrobials in poultry. There are large
countries, older narrow-spectrum products are not

132 Guide to antimicrobial use in animals

8 Poultry variations among regions, sectors and sources, and 0.2% and 0.8% respectively, in Salmonella spp. MS
the ability to acquire resistance seems to vary between generally reported high proportions of fully sensitive
different serovars. Antimicrobial-resistant Salmonella S. Enteritidis isolates from chickens, ranging from
are commonly isolated from different food animal 48.8% to 95.9%.
species and food products throughout Europe (10).
Over the last decade, clones of Salmonella with mul- It should be noted that resistance to quinolones in
tiple drug resistance have been distributed widely S. Enteritidis from cases of human infection is emerg-
in many European countries; in particular multi- ing in many EU countries, and in poultry (11). In
resistant S. Typhimurium definitive phage types (DTs) 2005, detection of nalidixic acid-resistant S. Enteritidis
204b and 104. isolates from poultry was reported in Austria (3.9%),
Germany (4.9%), Italy (34.3%) and the Netherlands
Data from the EU in 2005 (10), indicate that resis- (51.2%), in Denmark, resistance to nalidixic acid in
tance to tetracycline is common in Salmonella strains S. Enteritidis increased from 0% in 2001 to 23% in
from food animals. Resistance to streptomycin, sul- 2002. Use of fluoroquinolones in food animals
fonamides and ampicillin were also often observed. in Denmark decreased markedly in 2002 after several
Although fluoroquinolone resistance in many coun- initiatives by the authorities (11). The increase in resis-
tries remains infrequent, resistance to nalidixic acid, tance was most likely a result of clonal spread caused by
which is an indicator of developing resistance to trade in day-old chicks carrying nalidixic acid-resistant
fluoroquinolones, was observed by most reporting S. Enteritidis. This illustrates how the association
countries. As regards Salmonella isolates from poultry between usage of antimicrobials and occurrence of
in 2005, the highest proportions of isolates resistant resistance may be confounded by other factors, such
to chloramphenicol, sulfonamides and tetracyclines as transmission of resistant bacterial strains between
were reported by the Netherlands and the UK. For premises. In 2002, fluoroquinolone resistance was
S. Typhimurium, the highest levels of resistance detected in 1% of poultry S. Enteritidis isolates from
among isolates from chickens were reported for ampi- Italy, 5% from Spain and 13% from Portugal.
cillin (up to 73.9%), sulfonamide (up to 69.6%) and
tetracycline (up to 73.9%) (10). In several EU countries since the 1990s, a signifi-
cant increase in the prevalence of resistance to mac-
Resistance to different types of antimicrobi- rolides and fluoroquinolones among Campylobacter
als, including quinolones, has become quite com- has been reported. This has been recognized as an
mon among S. Typhimurium and many strains are emerging public health problem due to the ability
multi-resistant (10). In several European countries of these bacteria to enter the food chain (13). In
as well as North America, a multi-resistant clone of Norway in 2001,the prevalence of quinolone resistance
S. Typhimurium DT 104 (MR-DT 104) became epi- among Campylobacter isolates from domestic poul-
demic during the 1990s. MR-DT 104 has been iso- try and from domestically acquired cases of campy-
lated from many different food animals including lobacteriosis in humans was low (2.7% versus 7%), as
poultry. It typically exhibits resistance to ampicillin, opposed to a high prevalence of quinolone resistance
chloramphenicol, streptomycin, sulfonamides and in isolates from imported human cases (60%) (12).
tetracyclines (ACSSuT). Since the mid-1990s, the In Australia, where fluoroquinolones have not been
occurrence of resistance to quinolones has increased authorized for use in animals, indigenous fluoroqui-
in MR-DT104 isolates. In the UK, the emergence of nolone resistant Campylobacter are not seen in human
quinolone-resistant MR-DT104 in poultry, cattle, (15). In the USA, fluoroquinolone resistance in human
pigs and humans followed soon after the licensing of Campylobacter isolates fluctuated between 13% and
enrofloxacin for use in food-production animals. 18% (16). After an initial rise in resistance during the
1990s (14), the fluoroquinolone approval for use in
In contrast to S. Typhimurium, S. Enteritidis iso- poultry was withdrawn in 2004. Van Boven et al. (17)
lates are, in general, susceptible to most antimicrobi- compared the selection of quinolone resistance in C.
als. In poultry isolates of S. Enteritidis, in 2005, the jejuni and E. coli in individually housed broilers, and
highest level of resistance was reported for nalidixic demonstrated that treatment with enrofloxacin at
acid (up to 51.2%). Resistance to tetracycline was doses routinely prescribed (50 ppm) rapidly reduced
generally low (from 0% to 10.5%). Italy was the only the faecal counts of E.coli below the detection limit,
country to report resistance to ciprofloxacin and enro- and did not induce resistance in this bacterial species.
floxacin. The proportions of resistant isolates were

Guidelines for antimicrobial use in poultry 133

However, the same treatment quickly selected for high classified as resistant to bacitracin. A surprisingly high 8 Poultry
frequencies of fluroquinolone-resistant strains of percentage of S. aureus (30%) were resistant to cipro-
C. jejuni. floxacin, whereas 24% were resistant to erythromycin
and 19% to sulfamethoxazole.
Resistance among clinical isolates of poultry E. coli
can be high and multiple resistance is common, as Johansson et al. (24) performed a study to determine
demonstrated in studies from Spain and the USA the in vivo susceptibility of Clostridium perfringens
(18, 19). In a collection of strains isolated from vari- isolated from poultry to antimicrobials used in poul-
ous types of poultry in the USA, 63% of the strains try production, including the ionophore coccidiostat
were found to harbour class 1 type integrons, mostly narasin. Isolates were obtained from broilers, laying
located in a transposon related to Tn21 (19). In the hens and turkeys in Sweden, Denmark and Norway.
USA, an increase in resistance to fluoroquinolones Tetracycline-resistance was the most common anti-
among avian pathogenic E. coli has been reported microbial resistance trait found in C. perfringens in
(21). In 2005, Zhao et al. (20) reported, in the USA, this study despite marked differences among the three
the presence of multiple antimicrobial-resistant phe- countries, whereas all isolates proved to be susceptible
notypes (≥3 antimicrobials) in 92% of E. coli isolated to narasin. Three per cent of the Swedish isolates and
from diagnosed cases of avian colibacillosis. The 15% of the Danish isolates were resistant to bacitracin,
majority of isolates displayed resistance to sulfame- and 13% of the Norwegian isolates were resistant to
thoxazole (93%), tetracycline (87%), streptomycin virginiamycin (a streptogramin). All isolates were sus-
(86%), gentamicin (69%) and nalidixic acid (59%). ceptible to avilamycin, erythromycin, ampicillin and
Fifty-six E. coli isolates displaying resistance to nali- vancomycin, where negligible use of these substances
dixic acid were co-resistant to difloxacin (57%), enro- occurred in poultry production in the investigated
floxacin (16%), gatifloxacin (2%) and levofloxacin countries. Similarly, in the USA, avilamycin, avo-
(2%). Similar data were previously reported by Bass parcin, penicillin and narasin were found to exhibit
et al. (19), who observed how the resistance to specific the most potent in vitro anti-clostridial activity on
antimicrobials, like streptomycin, continued to be C. perfringens strains of avian origin (25).
prevalent among avian E. coli isolate, despite the dis-
continuance of a given antimicrobial as a therapeutic Poultry products and farms have been implicated as
agent. In this study, the presence of integrons among a source of vancomycin-resistant Enterococcus (VRE)
clinical isolates was demonstrated and linked to the in humans (26). The role that non-human sources
presence of multiple-antimicrobial resistance with and reservoirs, other than hospitalized patients, may
continued resistance to antimicrobials that have been play in the spread of Enterococcus is controversial and
withdrawn from use in poultry medicine. In Ireland, poorly understood (27). In the USA, where glyco-
Cormican et al. (22) compared levels of antimicrobial peptides have not been used for production animals,
resistance in pathogenic E. coli isolated from hens and among E. faecium and E. faecalis isolated from 13
from turkeys, and observed higher levels of resistance chicken and 8 turkey farms, none was resistant to van-
in turkeys, where antimicrobial use is more com- comycin, whereas quinupristin/dalfopristin, gentami-
mon. Resistance to sulfonamides, potentiated sul- cin and ciprofloxacin resistance rates in E. faecium
fonamides and nalidixic acid were more common in were 85%, 12% and 23% in chicken, and 52%, 13%
E. coli originating from turkeys. Ciprofloxacin resis- and 24% in turkey isolates. Quinupristin/dalfopristin
tance, at a level of 2.9%, was observed only in E. coli (streptogramin) resistance in E. faecium was more
isolates from turkeys. common on chicken and turkey farms using virgin-
iamycin compared with farms not using a strepto-
As far as Staphylococcus is concerned, Aarestrup gramin. Ciprofloxacin resistance was more common
et al. (23) tested 118 isolates from infections of poul- on turkey farms using enrofloxacin compared with
try in Denmark for their susceptibility to 19 antimi- those with no enrofloxacin use (27). Conversely, it
crobial agents. All isolates were found susceptible to should be noted that one large, clinical study involv-
avoparcin, flavophospholipol, gentamicin, kanamy- ing 28 000 human clinical isolates from 200 medical
cin, monensin, nitrofurantoin, oxacillin, salinomy- centres in the USA and Canada revealed strepto-
cin, trimethoprim and vancomycin. Seven per cent gramin resistance to be 0.2% in E. faecium isolates
of S. aureus isolates and 35% of the novobiocin- despite decades of use of virginiamycin in the poultry
resistantcoagulase-negativestaphylococci(CNoS)were industry (28). In Spain, no resistance to vancomycin,

134 Guide to antimicrobial use in animals

teicoplanin, penicillin or ampicillin was detected in Table 8.2 Antimicrobial susceptibility in E. coli and
Enterococcus isolated from poultry (29), but strains Pseudomonas isolates from poultry in the UK (31)
showed high-level aminoglycoside resistance for
streptomycin (34.5%), kanamycin (27.3%) and gen- E. coli Pseudomonas
tamicin (7.3%).
8 Poultry Antimicrobial 1996 2003–2005 2000–2005
Considering the fact that resistance trends in poultry drug (%) (%) (%)
pathogens may differ greatly from country to coun-
try, it is extremely interesting to compare data con- Enrofloxacin 95 95 98
cerning resistance in different European countries Ampicillin 58 62 4
with different patterns of antimicrobial use. When Tetracyclin 21 32
comparing data, different methods used (microdilu- Spectinomycin 95 95 35
tion versus agar diffusion test), variances in suscep- Tylosin 72
tibility among different bacterial species of the same Potentiated 30
genus and various other factors must be taken into sulfonamides 55 80 0
account. For example, antimicrobial susceptibility of Apramycin 12
Ornithobacterium rhinotracheale (ORT) shows great Neomycin 96 98
differences between geographical regions (30, 31). 89 95 100
ORT is a respiratory pathogen in chicken and tur- 84
keys, and causes more problems in turkeys where it
is regarded as primary pathogen. In chicken, ORT is Table 8.3 Enrofloxacin resistance in clinical E. coli
often involved in respiratory problems as a secondary isolates from poultry in Spain and France (35)
organism. Lister (31) reported susceptibility in E. coli
(APEC) and Pseudomonas isolates from turkey in the Spain France
UK. Data from 1996 and 2003–2005 did not show dif-
ferences in susceptibility for E. coli (Table 8.2). 1991–1995 1996–2000 1991–1995 1996–2000

Bywater (32), when reviewing published litera- n = 338 n = 198 n = 154 n = 248
ture, reported susceptibility patterns in zoonotic
(Salmonella, Campylobacter) and indicator bacteria 10.3% 41.9% 7.1% 2.5%
(E. coli) from poultry flocks in Sweden, France, the
UK and the Netherlands. He concluded that the varia- diagnosis, including bacterial isolation and sensitivity
tion seen between countries could have resulted from testing (wherever possible), medical knowledge and
differences in prescribing practices, in the disease experience, economic considerations, epidemiologi-
distribution (resulting in differences in antimicro- cal background and information at the flock level.
bial demand), or in clonal distribution of particular Antimicrobials should never replace fundamental
strains (e.g. Salmonella). Easy access to antimicrobials shortcomings in poultry production such as biosecu-
of third choice (like enrofloxacin) may lead to overuse rity measures and proper hygiene. The administration
with serious consequences on resistance development. of antimicrobials in disease situations is supportive
Comparison over time between France (only original of good farm management and properly designed
enrofloxacin product registered) and Spain (cheaper immunization programmes (see Section 8.7).
generic enrofloxacin products on the market) sug-
gests that market dynamics can influence antimicro- The use of antimicrobials should meet the
bial usage patterns and thus impact on antimicrobial requirements of a valid veterinarian–client–patient
resistance development (Table 8.3). relationship.

8.5 Good veterinary practices for The veterinarian assumes the responsibility for
antimicrobial use in poultry initiation of antimicrobial therapy and the farmer
agrees to follow his instructions.
Rational choice of antimicrobial agents should The veterinarian is acquainted to the farm by reg-
be based on clinical judgement and laboratory ular visits.
The veterinarian is available for follow-up evalua-
tion and emergency visits.
Unless the clinical picture (signs, gross lesions) is
pathognomonic, a flock diagnosis should be con-
firmed by laboratory testing. In urgent situations,
a lab confirmation cannot be waited for before an

Guidelines for antimicrobial use in poultry 135

antimicrobial treatment is initiated. In this case, the reduced in an attempt to save money. It is best for the 8 Poultry
veterinarian will be guided by his professional knowl- antimicrobial dosage to be calculated on the basis of
edge and experience in similar situations.Susceptibility mg of active ingredient per kg bodyweight. In most
testing of the causative microorganism(s) in a rep- broiler flocks, actual live weight can be seen on the dis-
resentative bird sample (typically ill subjects, recent play of automatic scales, which are today standard in
deaths), prior to or concurrently with the onset of modern broiler houses. For turkeys, the actual weight
medication, is common practice in avian medicine. can be taken from the age and the growth profile of the
breed. Replacement pullets and breeders are normally
In contrast to antimicrobial use in individual human weighed every week. Modern poultry houses have
patients and in large animal practice, the avian medi- water meters, so the amount of water to be consumed
cine practitioner has to make use of antimicrobials can be taken from the day before. Consideration needs
in some instances for total flock medication, where to be made for the effect of temperature swings on
often only a small percentage of birds may be showing water consumption patterns in poultry flocks:
clinical symptoms. Individual birds are not treated in
the modern poultry industry. Although a diseased When choosing the appropriate antimicrobial, the
flock consists partially of sick and lethargic birds with veterinarian has to take into account: susceptibil-
varying degrees of symptoms, it is important to treat ity results, withdrawal times, pharmacodynamic
the flock as a whole to lower the infection pressure for and pharmacokinetic properties.
flock mates. When different products come into consideration,
the veterinarian should choose as first choice a
Certain other principles pertaining to responsible narrow-spectrum antimicrobial or an antimicro-
antimicrobial usage include: bial with limited importance in human medicine.
Treatment failures may occur if a low dosage or
The use of antimicrobials as soon as premonitory short duration of treatment is attempted.
disease signs appear. The earlier in the disease pro- An unsatisfactory clinical outcome can also be
cess therapy is initiated, the better the chance of a triggered by concurrent immune suppressive viral
favourable response. For intensively kept poultry, diseases (Chicken Infectious Anemia, Gumboro,
it is also vital to minimize the further spread of Reo, Marek), metabolic diseases, or too high infec-
the disease to adjacent flocks and neighbouring tion pressure (overwhelming infections).
farms.
By minimizing bird morbidity and mortality with 8.6 Disease-specific guidelines
properly selected and timed antimicrobial ther- for antimicrobial use
apy, the veterinarian also sustains improved ani-
mal welfare. Medication in anticipation of rising The following sections provide guidelines for specific
mortality and major disease damage is justifiable pathogens and diseases in poultry. As a rule, guide-
to minimize bird suffering as well as to improve lines tend to be generic in nature, although they
performance. emphasize important use principals that are effective.
In addition to group medication, very sick birds, Obviously, the guidelines do not take into consider-
which will not drink enough of the medicated ation national differences regarding registration and
water, may be culled (broilers, rearing pullets) or withdrawal times, which are regulated by national
individually treated (valuable breeder and turkey legislation. Antimicrobial agents are categorized into
stocks) in some cases. first, second and last choice: (i) first choice antimicro-
The careful use of antimicrobials to anticipate bials are products with no or minimal use in human
developing disease in a flock should never be con- medicine; second choice antimicrobials are products
fused with, or serve as an excuse for, the injudi- which are used in human medicine, but which are not
cious use of antimicrobials in healthy flocks to first choice products in the human medical commu-
cover shortcomings in hygiene and management. nity (see Chapter 4); and third choice products are
Antimicrobial products should be administered important antimicrobials in human medicine, which
according to the label directions established by the should therefore be regarded as reserve antimicrobials
manufacturer and approved by the regulatory author- for treatment of poultry flocks.
ity. Label directions encompass indications (claims)
and dosage (dose, duration of application). The anti-
microbial should always be used at full dose and never

136 Guide to antimicrobial use in animals

8 Poultry 8.6.1 Unspecific enteritis 8.6.2 Clostridial infections
(Dysbacteriosis) (C. perfringens and
C. colinum)
After the withdrawal of growth promoters in the EU,
the incidences of unspecific enteritis in turkeys and Clostridium are opportunistic, spore-forming bacteria
broilers have increased. The ban of growth promoters that can survive heat treatment of the feed. They
coincided in some countries with the ban of highly cause sudden mortality in a broiler flock (necrotic
digestible animal protein in the feed due to the BSE enteritis) or may lead to higher condemnation rates
crisis. As a consequence of this ban of meat and bone at slaughter (cholangiohepatitis). Chronic forms of
meal, animal proteins had to be replaced by plant necrotic enteritis have also been described. In con-
protein sources with a higher content of NSP (non trast to dysbacteriosis, these Clostridium infections
starch polysaccharides) and of potassium, namely soy typically present a clear clinical picture, which can
bean meal which is rich in potassium. NSPs are not be recognized at post-mortem examination and the
digestible to the avian gut, but may lead to a bacterial causative microorganism isolated. Again, in flocks
overgrowth in the intestine. Primarily Gram-positive vaccinated with a live coccidiosis vaccine, the likeli-
bacteria, particularly Clostridium in the upper part hood of Clostridium infections may increase com-
of the jejunum and the duodenum, are thought to be pared to flocks using coccidiostats as feed additives.
involved in this form of unspecific enteritis (dysbacte- C. perfringens is the main causative organism of
riosis). If untreated, the disease may lead to caked and necrotic enteritis. A new breeder vaccine is under reg-
wet litter with increased food pad dermatitis and hock istration in Europe, which should help protect the off-
burns (animal welfare problems). If the birds live lon- spring in the first weeks of life. If this product works
ger (turkeys) this may lead to ascending Staphylococcus under field conditions, its use should be considered to
aureus infections from the litter into the hock and avoid antimicrobial treatment against necrotic enteri-
knee joints with very serious uniformity and welfare tis. In acute outbreaks, flocks are routinely treated to
problems. Medication may be warranted under these reduce mortality and economical losses. The antimi-
circumstances. Dysbacteriosis should not be confused crobials of choice are similar to dysbacteriosis (Table
with wet litter caused by poor ventilation or leaking 8.4). Where registered for oral application, streptomy-
water supply (spillage of water). Before medication is cin or dihydrostreptomycin can be used as possible
considered for cases of suspected dysbacteriosis, any alternatives. Antimicrobial susceptibility of C. per-
potential nutritional influences should be assessed, fringens and other clostridia causing avian disease is
such as use of NSP enzymes and control of sodium predictable, thus susceptibility tests can be excluded.
in the diet. If management and nutritional inputs are
not felt to be contributing factors, then antimicrobial 8.6.3 Colibacillosis
medication for dysbacteriosis needs to be considered.
Antimicrobials of choice are those effective against Colibacillosis is the most common bacterial infec-
Clostridium spp., although these bacteria may not be tion in chickens or turkeys, and can be involved in
isolated in many instances. In most countries, ben- a number of syndromes affecting multiple ages. It is
zylpenicillin is the drug of choice. If not registered part of the yolk sac omphalitis syndrome during the
for usage in poultry, macrolides (tylosin) or amin- first week of life, when colibacillosis is transmitted by
openicillins (ampicillin or amoxicillin) represent dirty eggs or induced by poor hatchery hygiene. In
valid alternatives. Antimicrobial susceptibility testing adult layers, colibacillosis may lead to salpingitis and
is not needed, as bacterial intestine overgrowth is in egg peritonitis. Colibacillosis is typically a secondary
most cases of unspecific nature. There is circumstan- pathogen in respiratory infections resulting in pericar-
tial evidence that in broiler flocks vaccinated against iditis, perihepatitis and/or airsacculitis. Following sys-
coccidiosis, dysbacteriosis is more likely to occur. temic infections, E. coli can also result in synovitis and
Cycling of the vaccine strains has been suggested to osteomyelitis. Some E. coli are primarily pathogenic
promote selection of clostridia in the intestinal flora. to chicken (APEC, avian pathogenic E. coli). APEC
Ionophores can be used if feed additive antimicrobi- strains are E. coli O:1, O:2 and O:78 K 80. Colicin
als to prevent unspecific enteritis are banned for this and type 1 fimbriae seem to correlate with virulence,
purpose. but non-APEC strains can sometimes also cause

Table 8.4 Antimicrobial agents for treatment of common bacterial diseases in poultry

Disease/pathogen 1st choice 2nd choice Last choice

Dysbacteriosis Benzylpenicillin Aminopenicillins Tylosin Guidelines for antimicrobial use in poultry
Necrotic enteritis and other clostridial Benzylpenicillin Aminopenicillins or tylosin Tylosin
infections
Clostridium perfringens and others Potentiated sulfonamides Aminopenicillins, tetracyclines, colistin, Enrofloxacin
Colibacillosis spectinomycin, aminoglycosides Enrofloxacin
Escherichia coli Tiamulina Tetracyclines, lincomycin, (macrolides)
Mycoplasmosis Tiamulina Tetracyclines
Ornithobacterium rhinotracheale Aminopenicillins
Benzylpenicillin or potentiated sulfonamides Aminopenicillins, tetracyclines Macrolides
Stapylococcus or Streptococcus Potentiated sulfonamides Tetracyclines, spectinomycin Enrofloxacin
Fowl cholera Aminopenicillins
Pasteurella multocida Aminopenicillins Tetracyclines Enrofloxacin
Riemerella anatipestifer Sulfonamides, potentiated sulfonamides or Tetracyclines, lincomycin, spectinomycin, Enrofloxacin
Infectious Coryza macrolides
Haemophilus paragallinarum streptomycin Aminopenicillins, tetracyclines Enrofloxacin
Bordetella avium No antimicrobials Aminopenicillins Unnecessaryb
Erysipelothrix rhusiopathiae Benzylpenicillin BAST
Salmonellosis No antimicrobialsc

Aminoglycosides: streptomycin, apramycin, neomycin; aminopenicillins: amoxicillin, ampicillin; macrolides: erythromycin, spiramycin, tylosin, tilmicosin; tetracyclines: tetracycline,

oxytetracycline, doxycycline; fluoroquinolones: enrofloxacin.

BAST, based on antimicrobial susceptibility testing.
aTiamulin has neurotoxic effects when combined with ionophores and sulfonamides.
bErysipelotrix rhusiopathiae infections can normally be treated successfully with penicillin or aminopenicillins. It is therefore unnecessary to mention a last choice product.
cA Salmonella infection should only be treated on the bases of a clinical outbreak for welfare reasons. In this case a first choice antimicrobial cannot be suggested. Therapy should

always be based on antimicrobial testing. Zoonotic Salmonella infections should be eradicated by other means than antimicrobial treatment.

137

8 Poultry