Dog echinococcosis in northern Spain: Comparison of
coproantigen and serum antibody assays with coprological exam
Aitziber Benitoa, David Carmenaa,1, Lawrence Josephb,
Jorge Martı ´neza, Jorge A. Guisantesa,*
aDepartment of Immunology, Microbiology and Parasitology, Faculty of Pharmacy, University of the Basque Country,
P.O. Box 450, 01080 Vitoria, Spain
bDepartment of Epidemiology and Biostatistics, Faculty of Medicine, McGill University, Montreal, Canada
Received 30 November 2005; received in revised form 25 April 2006; accepted 16 June 2006
A large sheep-dog population from the province of A´lava (northern Spain) has been investigated in order to determine the
prevalence of the cestode parasite Echinococcus granulosus. Worms were detected in 14.0% of 721 dog faecal supernatants by
coproantigen ELISA, and in 9.1% of 754 dog serum samples by serum antibody ELISA. A weak but statistically significant
correlation (Spearman’s r = 0.103,95% CI: 0.023–0.178) between the two immunoassay results was found. In addition, eggsof the
family Taeniidae were detected in 10.3% of 726 faecal samples examined by coproparasitological (flotation and sedimentation)
tests. The overall E. granulosus infection rate, based on a Bayesian latent class model that accounts for the imperfect sensitivities
andspecificitiesofalldiagnostictests used,was estimatedtobe8.0%(95%credibleinterval:5.4–11.4%),corroboratingthatsheep-
dog is the dog class most vulnerable to acquiring the infection. Dog sex did not influence the prevalence of E. granulosus,
independently of the diagnostic test used or the dog region of origin. No significant linear correlation was found between the
coproantigen ELISAOD values and the dog age (Spearman’s r = ?0.049, 95% CI: ?0.234 to 0.135), suggesting that therewere no
differences inprevalence ofE.granulosusbetween oldandyoungdogs.Theobtainedresults highlighttheimportanceofinitiatinga
control program based on regular treatment of the sheep-dogs with praziquantel in the province of A´lava.
# 2006 Elsevier B.V. All rights reserved.
Keywords: Echinococcus granulosus; Intestinal dog echinococcosis; Epidemiology; Coproantigen; Spain
Cystic echinococcosis (CE) is an important zoonosis
caused by the taeniid tapeworm Echinococcus granu-
losus with a considerable impact in both human and
animal health in endemic areas (Schantz et al., 1995).
The disease has a wide geographical distribution, with
emerging and re-emerging regions mainly in Central
Europe and China (Eckert and Deplazes, 2004). Spain,
together other Mediterranean countries, is currently
considered as hyper endemic area (McManus et al.,
The parasite’s domestic life cycle is maintained
through dogs (which harbour the adult tapeworm) and a
range of domestic livestock intermediate host species,
generally sheep and cattle. Due to the high biotic
potential of E. granulosus, infected dogs can excrete a
Veterinary Parasitology xxx (2006) xxx–xxx
* Corresponding author. Tel.: +34 945 013804;
fax: +34 945 013014.
E-mail address: email@example.com (J.A. Guisantes).
1Present address: MRC Clinical Sciences Centre, Faculty of Med-
icine, Imperial College, Hammersmith Hospital Campus, Du Cane
Road, London W12 0NN, UK.
0304-4017/$ – see front matter # 2006 Elsevier B.V. All rights reserved.
VETPAR-3669; No of Pages 10
large number of parasite’s eggs with their faeces,
contaminating wide extensions of soil, and spreading
the disease (Gemmell, 1990). Because its pivotal role in
the transmission dynamics of CE, detection of E.
granulosus in the definitive host is a key point in
developing of epidemiological studies and implementa-
tion of hydatid control programmes in endemic areas
Necropsy of dogs and examination of the small
intestine is the reference method for the detection of
intestinal infections with E. granulosus, but this
laborious and ethically questionable procedure is not
suitable for mass screening. Thus, a number of
antemortem methods have been developed for diag-
nostic purposes. Arecoline purging, although 100%
specific, has a highly variable sensitivity (a negative
result even after two or more treatments does not
guaranteed the animal is Echinococcus-free), is labour
intensive, biohazardous and some dogs suffer undesired
side-effects (Wachira et al., 1990; Eckert et al., 1984,
2001). Coprological exams have low specificity and
sensitivity, as eggs from different taeniid cestodes
cannot be differentiated by light microscopy, and egg
production may be erratic. Finally, ELISAs for
detecting parasite-specific antibodies in serum have
showed variable sensitivities, ranging from 40 to 90%
(Benito et al., 2001; Gasser et al., 1994; Jenkins et al.,
1990), and cross-reactivity with other parasite species is
often detected (Gasser et al., 1988).
Currently the most practical approach for the
diagnosis of the intestinal E. granulosus infection in
dogs is the detection of parasite antigens in faecal
samples (coproantigens) by ELISA using antibodies
against adult somatic antigens (Allan et al., 1992), and
(Deplazes et al., 1992) or protoscoleces (Benito and
Carmena, 2005). This method considerably improves
both diagnostic sensitivity and specificity, permits the
detection of the parasite during the prepatence period
(Ahmad and Nizami, 1998), shows the current status of
the infection (Jenkins et al., 2000), and ELISA results
correlatewell with the worm burden in the dog intestine
(Craig et al., 1995). Coproantigen ELISAs have been
successfully used in the field in Libya (Buishi et al.,
2005), Cyprus (Christofi et al., 2002), Uruguay (Malgor
et al., 1997), and Wales (Palmer et al., 1996),
demonstrating their usefulness for epidemiological
Recent field surveys have shown that sheep and stray
dogs have the highest E. granulosus infection rates,
most likely because of their greater access to offal or
casualty animals (Buishi et al., 2005; Shaikenov et al.,
2003). Therefore, those are the dog classes currently
considered to be at highest risk of infection by this
cestode. In the province of A´lava (northern Spain) we
have previously reported similar parasitological find-
ings (Benito et al., 2003), although most of the dogs
analyzed in that study came from urban environments.
In this paper we present data on the detection of E.
granulosus in a large population of sheep-dogs from the
same region, by using ELISAs for the detection of
parasite-specific coproantigens and antibodies, and
2. Materials and methods
2.1. Sampling plan design
The province of A´lava (northern Spain, 4382200N,
681200W), extends over 3037 km2, has an averaged
annual temperature of 12 8C, and an annual precipita-
tion in the range of 650–900 L/m2. A census of sheep-
dogs from livestock farms in the province was prepared
in collaboration with the Epidemiology Unit of the
Subdirection of Public Health of A´lava (Department of
Health, Basque Government, Spain) and the Depart-
ment of Agriculture (Provincial Government of A´lava,
Spain) including data on dog age and sex, and number
of head of cattle per farm. Livestock farms with less
than 10 sheep were not considered in this study. In order
to investigate whether geographical and climatologic
parameters influence the infection rate of E. granulosus
in dogs, livestock farms were distributed as follow
(Fig. 1): (i) Northern region, with an Atlantic climate
characterized by warm summers, mild winters, and
abundant precipitations through the year; (ii) Central
region, with a transition climate from Atlantic to
Mediterranean; (iii) Southern region, with a typical
mild,rainywinters.The dog sample size estimation was
performed using the Epi InfoTMV6 software (CDC,
Atlanta, USA) for epidemiologic statistics, assuming a
1% precision with a 95% confidence interval. An E.
granulosus theoretic frequency of 2.5% was estimated
on the basis of an overall prevalence of 0.5% for this
cestode found in dogs from the same area in a previous
survey (Benito et al., 2003). The total sheep-dog
population in the province of A´lava was 1858 dogs. Of
them, 978 dogs were initially registered for blood and
faecal sampling: 422 form the Northern region, 487
from the Central region and 69 from the Southern
region. However, due to inherent problems during the
specimen’s collection (unavailability of some dogs at
the time of visit, or owner’s refusal to cooperate),
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx2
a smaller number of samples were available to perform
this survey (see Section 3). Sampling took place during
a 15-month period from April 1997 to June 1998.
2.2. Blood and faecal samples
dogs, and allowed to clot at room temperature for
90 min. Serum was separated by centrifugation at
2000 ? g for 15 min, aliquotted and stored at ?20 8C
until tested. A faecal sample was taken from each dog
either rectally using a plastic spatula, or collected from
the ground after defecation. Faecal supernatants were
obtained by mixing the samples at 1:2 ratio (v/v) with
PBS buffer containing 5% formaldehyde and 5mM
EDTA. Samples were shaken vigorously until slurry
was formed, and this was centrifuged at 3500 ? g for
20 min. Supernatants were aliquotted and stored at
?20 8C. An untreated portion of each stool sample was
kept and stored at ?20 8C for coproparasitological
2.3. Coproparasitological examination
In order to identify the presence of helminths of the
family Taeniidae in the dog population, parasite eggs
were isolated from the stool samples by using both
sedimentation (Ritchie, 1948) and flotation (Sloss et al.,
2.4. Serum antibody ELISA (Ab ELISA)
ELISAs for the detection of parasite-specific
antibodies in serum samples were carried out as
described by Benito et al. (2001). Briefly, polystyrene
96-well microtitre plates (MaxiSorpTM, Nunc, Ros-
kilde, Denmark) were coated with 10 mg ml?1proto-
scolex somatic antigens in PBS buffer for 1 h at 37 8C.
Plates were blocked with PBS–1% BSA (Sigma–
Aldrich) for 1 h at 37 8C. Dog sera were assayed in
PBS–0.5% BSA in duplicate, and incubated for 1 h at
37 8C. Sera serial dilutions ranging from 1:50 to 1:400
were tested. Rabbit anti-dog IgG peroxidase conjugate
(Sigma–Aldrich) was used as secondary antibody at
1:1000 dilution in PBS–4% BSA–0.05% Tween 20 for
1 h at 37 8C. Binding was visualized with 5-aminosa-
licilic acid. The reaction was stopped by adding 25 ml/
well NaOH 1N, and the absorbancevaluewas measured
at 450 nm. Positive-negative cut-off absorbance value
of 6089 arbitrary units was based on two standard
deviations above the mean optical density (OD) value
for uninfected and heterologous helminth species
infected dogs, previously diagnosed by post-mortem
intestinal examination. Diagnostic performance of Ab
ELISAwas previously evaluated using necropsy as gold
standard method. An overall sensitivity of 80.5% and
specificity of 92.1% were obtained. Positive and
negative predictive values were 59.2 and 97.1%,
respectively, and the diagnostic efficiency was 90.7%
(Benito et al., 2001). These estimates, however, do not
account for the possibly imperfect sensitivity and
specificity of necropsy. Below we describe the
statistical methods used to account for this possibility.
2.5. Coproantigen ELISA (CpAg ELISA)
ELISAs for the detection of parasite excretory–
secretory products (ES-Ag) in faecal samples (coproan-
tigens) were performed according to Benito and
Carmena (2005). In summary, polystyrene 96-well
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx3
Fig. 1. Distribution of the sampling regions in the province of A´lava (northern Spain) according to geographical and climatological criteria.
microtitre plates (MaxiSorpTM, Nunc) were coated with
30 mg ml?1ofthe purifiedIgG anti-ES-Agfraction,and
incubated for 3 h at 37 8C. Blocking was carried out
with PBS–0.05% Tween 20 overnight at 4 8C. Faecal
supernatants were assayed at 1:2 dilution in PBS in
duplicate, and incubated for 1 h at 37 8C. Biotinylated
IgG anti-ES-Ag fraction was used as secondary
antibody at 1:250 dilution in PBS–5% pre-immune
rabbit serum for 1 h at 37 8C. Then, peroxidase-
1:1000 dilution in PBS for 1 h at 37 8C. Binding was
visualized with 5-aminosalicilic acid. The reaction was
stopped by adding 25 ml/well NaOH 1N, and the
absorbance value was measured at 450 nm. Positive–
negative cut-off absorbance value was defined as the
mean absorbance of the faecal supernatants from
helminth-free dogs (previously diagnosed by post-
mortem intestinal examination) plus two standard
deviations. This value corresponds to an ES-Ag
concentrationof0.24 mg ml?1.Diagnosticperformance
of CpAg ELISA was previously evaluated using
necropsy as goldstandard method. The assay accredited
an overall sensitivity of 78.4% and specificity of 93.3%.
Positive and negative predictive values were 72 and
95%, respectively, and the diagnostic efficiency was
90.5%. In addition, the detection limit has been
estimated in 5.12 ng ml?1of E. granulosus ES-Ag
(Benito and Carmena, 2005). As mentioned above, an
alternative method is discussed below that does not
assume necropsy to be a perfect gold standard.
2.6. Statistical analysis
A Bayesian statistical approach (Joseph et al., 1995)
was used for estimating the prevalence of E. granulosus
in the studied sheep-dog population and the properties
of the diagnostic tests utilized in this field survey. Our
estimation procedure consisted of two stages: at the first
stage, we estimated the sensitivities and specificities of
both ELISA tests from previously reported data (Benito
et al., 2001; Benito and Carmena, 2005) comparing
these tests to necropsy (Tables 1 and 2). Parasitological
examination of the small intestine at necropsy is
considered the gold standard method for the diagnosis
of E. granulosus in the definitive host, with sensitivity
and specificity each just below 100% (Eckert and
Deplazes, 2004; WHO/OIE, 2001). We carried out two
first stage analyses, one assuming 100% sensitivity and
specificity for necropsy, and the other assuming 95%
prior ranges for the sensitivity and specificity of 95–
100% and 98–100%, respectively. This allowed us to
determine robustness of our final estimates to these
Our Bayesian analysis was performed using the
BayesDiagnosticTests Version 2.1 Software Package
implements the methods of Joseph et al. (1995).
Briefly, the method operates as follows: the probability
of landing in each cell in a two by two table of data such
as those given in Tables 1, 2 and 4 depends on the
prevalence (p), and the sensitivities (S1, S2) and
specificities (C1, C2) of the two tests. For example, in
order to land in the first cell, the subject must either be
truly positive, and have both tests correctly identify this
true positive (which occurs with probability pS1S2), or
be truly negative, and have both tests incorrectly
identify the subject as positive (which occurs with
probability(1 ? p)(1 ? C1)(1 ? C2)).
these probabilities across all four cells in each table,
and raising each term to the number of subjects falling
into that cell provides the likelihood function for that
table of data. As in all Bayesian approaches, this
likelihood function must be multiplied by the joint prior
distribution over all unknown parameters, which by
Bayes Theorem gives the posterior distribution from
which all statistical inferences including 95% credible
intervals (Bayesian analogue of standard confidence
intervals) follow. At the first stage, we used uniform
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx4
Summary of the diagnostic characteristics of the serum antibody
ELISA assay (Ab ELISA) used for the development of prior distribu-
tion of this test
Necropsy results Total (n = 289)
(n = 36)
(n = 253)
Ab ELISA (+)
Ab ELISA (?)
Necropsy was considered the gold standard method (Benito et al.,
Summary of the diagnostic characteristics of the coproantigen ELISA
assay (CpAg ELISA) used for the development of prior distribution of
Necropsy resultsTotal (n = 200)
(n = 37)
(n = 163)
CpAg ELISA (+)
CpAg ELISA (?)
Necropsy was considered the gold standard method (Benito and
prior distributions over the properties of the ELISA
tests, but informative priors (as described above) for the
necropsy properties, in order to estimate the ELISA test
parameters. At the second stage, we took the results
(posterior distributions for the sensitivities and specifi-
cities of both ELISA tests) from the first stage to form
prior distributions as inputs to the analysis of the data in
Table 3. Throughout all analyses, uniform prior
distributions were used for the prevalences.
The Normal approximation to the difference
between two binomial proportions was used to estimate
the differences between the results of the diagnostic
tests and the sex and origin of the sheep-dogs. Results
derived from the dog’s origin were dichotomized into
South versus non-South regions, as sheep-dogs from the
Southern region were expected to bear higher E.
granulosus infection rates (see Section 4). Nonpara-
metric Spearman’s r was calculated to measure the
degree of association between the Ab and CpAg ELISA
results, and to evaluate whether Echinococcus coproan-
tigen OD values are related to dog age.
3.1. Coproparasitological examination
A total of 726 dog stool samples were obtained for
coproparasitological exam. The presence of intestinal
helminths were recorded in 416/726 (57.3%) of faecal
samples after laboratory examination by at least one of
the concentration methods used. Taeniid eggs were
found in 75/726 (10.3%) of the examined samples
(Table 1). Neither dog sex (difference (D) = 0.01, 95%
CI: ?0.03 to 0.06) nor dog region of origin
(D = ?0.003, 95% CI: ?0.10 to 0.09) seem to have
an effect on the copro-prevalence of members of the
family Taeniidae. Other helminth species found were
Trichuris vulpis (38.3%), Uncinaria stenocephala
(26.3%), Toxocara canis (5.2%), Toxascaris leonina
(1.4%), Capillaria spp. (0.5%), and Ancylostoma
caninum (0.3%). No important differences were found
between the detection of enteroparasite eggs and the
kind of concentration method used, except in the cases
of T. vulpis (D = 0.55, 95% CI: 0.47–0.62) and T. canis
(D = 0.75, 95% CI: 0.57–0.93), which were better
detected by sedimentation.
3.2. Immunodiagnostic assays
The Ab ELISAwas assessed against 754 dog serum
samples of which 69 (9.1%) tested positive for E.
granulosus specific antibodies. Of 721 dog faecal
supernatants assayed for E. granulosus coproantigens,
101 (14.0%) were positive (Table 1). No important
differences were found between the Echinococcus
prevalence rates obtained by Ab ELISA (D = ?0.002,
95% CI: ?0.05 to 0.04) or CpAg ELISA (D = 0.005,
95% CI: ?0.05 to 0.06) and dog sex. When tested by
both Ab or CpAg ELISA assays, sheep-dogs from cattle
farms of the Southern region showed higher E.
granulosus infection rates than those from the rest of
the province, although these results are accompanied by
wide confidence intervals (Ab ELISA – D = ?0.09,
95% CI: ?0.19 to 0.01; CpAg ELISA – D = ?0.05,
95% CI: ?0.16 to 0.07). Taking together, these results
show that CpAg ELISA detects a 4–42% more positive
samples than Ab ELISA, depending on the dog region
Matched samples (coproparasitological examination
with respective blood and stool specimens) were
obtained from a total of 666 dogs. A weak statistically
significant correlation (Spearman’s r = 0.10, 95% CI:
0.02–0.18) between the Ab and CpAg ELISA tests was
found (Fig.2). Forty-one dogs (6.1%) tested positivefor
E. granulosus by both immunoassays. Eighteen dogs
(2.7%) were Ab ELISA positive but CpAg ELISA
negative, and 56 (8.4%) animals were CpAg ELISA
positive but failed to show specific antibody levels
(Table 4). Taking together, these data indicate that 115
(18.2%) sheep-dogs tested positive for E. granulosus on
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx5
Results of the diagnostic tests for the detection of Echinococcus granulosus, according to the sheep-dog origin
(n = 726)
Serum antibody ELISA (n = 754)Coproantigen ELISA (n = 721)
Positive NegativePositive NegativePositive Negative
Total75 651 69685101 620
aFamily Taeniidae only.
at least one of the immunoassays. Using a Bayesian
statistical approach that adjusts for the imperfect
sensitivities and specificities of both ELISA tests, the
prevalence of E. granulosus infection in this sheep-dog
population was estimated to be 8.0% (95% credible
interval: 5.4–11.4%). Diagnostic performance charac-
teristics of both Ab and CpAg ELISA assays in the
studied dog population are shown in Table 5.
When the relationship betweenthe CpAg ELISAOD
values and the dog age was analyzed, no significant
linear correlation (Spearman’s r = ?0.05, 95% CI:
?0.23 to 0.13) could be demonstrated, and high E.
granulosus coproantigen levels were detected even in
dogs aged 10 years or more (Fig. 3). This seems to
of E. granulosus between old and young dogs.
Determining the rate of infection and mean
abundance in dogs is probably the best index of the
(Craig et al., 2003; Jenkins et al., 2000), a fact that is
essential for the establishment of baseline data on
prevalence, and in surveillance of hydatid control
province of A´lava (northern Spain), we previously
found a low prevalence (0.5%, 5/1040) of E. granulosus
infection in dogs at necropsy, although the majority of
the dogs studied in that survey came from urban
environments (Benito et al., 2003). Prevalences ranging
from 0.2 to 1.3% have also been reported in dogs from
diverse origins in the adjacent Autonomous Commu-
nities of Navarra (Gobierno de Navarra, 1987) and La
Rioja (Jime ´nez et al., 2002).
Bayesian methods have become a powerful statis-
tical tool for the analysis of parasitological data and
accurate estimation of levels of infection endemicity
(Basa ´n ˜ez et al., 2004). In the present study, a large
sheep-dog population from the province of A´lava has
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx6
Fig. 2. Correlation between Echinococcus granulosus antibody and
coproantigenODvaluesobtainedbyELISA(n = 666).Nonparametric
Spearman’s r = 0.103 (95% CI: 0.023–0.178).
for the detection of E. granulosus in the studied sheep-dog population
(n = 666)
Ab ELISA (+)Ab ELISA (?)
CpAg ELISA (+)
CpAg ELISA (?)
Bayesian estimation of the diagnostic characteristics of antibody (Ab)
and coproantigen (CpAg) ELISA assays in the studied sheep-dog
population, assuming necropsy is an imperfect test with 95% sensi-
tivity and 98% specificity
Mean S.D.95% Credibility
Positive predictive value
Negative predictive value
Positive likelihood ratio
Negative likelihood ratio
Positive predictive value
Negative predictive value
Positive likelihood ratio
Negative likelihood ratio
Fig. 3. Correlation between positive E. granulosus coproantigen OD
values and sheep-dog age from cattle farms of the province of A´lava
(n = 101). Nonparametric Spearman’s r = ?0.049 (95% CI: ?0.234
been investigated by using coproparasitological and
immunoenzymatic assays in order to determine the E.
granulosus infection rate. Following a Bayesian
approach which assumes necropsy is an imperfect test
with sensitivity assumed to be in the range from 95 to
to 100%, the estimated parasite prevalence in the
studied sheep-dog population was 8.0%. Avery similar
assuming necropsy is a perfect test with 100%
sensitivity and specificity, a feature that demonstrates
the robustness of our estimates to the choice of prior
distribution. This prevalence value is 6–40-fold higher
than previously reported in this province (Benito et al.,
2003), and in the bordering Autonomous Communities
of Navarra (Gobierno de Navarra, 1987) and La Rioja;
(Jime ´nez et al., 2002). As expected, these data are in
(Buishi et al., 2005; Shaikenov et al., 2003), demon-
strating that shepherd dogs are the main dog class
involved in the (domestic) transmission dynamic of E.
In our work, the coproparasitological survey
revealed the presence of taeniid eggs in 10.4% (75/
726) of the stool samples analyzed, an elevated rate
taking into account that this methodology considerably
underestimates the true prevalence of the infection
(Jenkins et al., 2000). When matched samples were
assayed using both Ab and CpAg ELISA assays, aweak
but significant correlation between the test results could
be demonstrated (Spearman’s r = 0.10, 95% CI: 0.02–
0.18). However, 56 (8.4%) sheep-dogs which were
CpAg ELISA positive have undetectable E. granulosus
specific antibody levels. Very likely many of these
results correspond to CpAg ELISA false positive
reactions. It is well-known that members of the family
Taeniidae are a common source of cross-reactivity in
the immunodiagnosis of intestinal dog echinococcosis
(see Carmena et al., 2006). This could be the case in the
present survey, where an elevated rate (10.4%) of
taeniid eggs was detected by coprological examination.
On the other hand, taking into account that a true
positive CpAg ELISA result is indicative of current
infection (Fraser and Craig, 1997), the seronegativity of
not elicited a specific antibody response. This phenom-
enon may also be due to the sequestration of antibodies
and the formation of circulating immunocomplexes
(Gasser et al., 1993; Spinelli et al., 1996), or to immune
evasion mechanisms of the parasite (Gasser et al.,
1994), or to a low immune response of the host (Gasser
et al., 1993, 1994). Host nutritional status has also been
suggested to have an impact on the antibody levels
(Jenkins et al., 1991). Eighteen (2.7%) sheep-dogs were
Ab ELISA positive but CpAg ELISA negative. This
finding may indicate the possibility of recent exposure,
with previous infection eliminated spontaneously or by
owners with anticestodal drug treatment. Because
specific serum antibody levels can persist for several
months after the worms have been removed (Gasser
et al., 1990), this fact may be an important cause of
false-positive reaction in the serodiagnosis of the
CpAg ELISA showed a higher diagnostic sensitivity
than Ab ELISA (87.0 and 82.2%, respectively). A
similar result has been previously reported by Craig
et al. (1995). Both Ab and CpAg ELISA assays showed
a very high negative predictive value (98.3 and 98.8%,
respectively), a characteristic that make them specially
suited for the mass-screening of dog populations with
low prevalence of E. granulosus. However, the
relatively low positive predictive values of these
techniques (62.7% for the Ab ELISA, and 49.2% for
the CpAg ELISA) strongly recommend their simulta-
neous use inepidemiological surveys, in order to reduce
the risk of false positive results. In an scenario of low-
medium parasite prevalence like the one described in
this study, the superior sensitivity of the CpAg ELISA,
together with its ability to estimate the current status of
the infection and the parasite burden, make this assay
the immunodiagnostic test of choice for the detection of
E. granulosus in the definitive host. However, con-
firmation of Ab and/or CpAg ELISA positive results by
detection of E. granulous DNA in faeces using PCR is
highly desirable (Deplazes et al., 2003).
Dog sex did not appear to influence the prevalence of
E. granulosus, independently of the diagnostic test used
or the dog region of origin. Concerning dog origin, the
Southern region was found to bear the highest parasite
inconclusive, this result was predictable taking into
account that the geoclimatic characteristics of this area
present favorable conditions for the dynamic transmis-
sion of the infection. In addition, this region borders the
Autonomous Communities of Navarra and La Rioja,
where the presence of the parasite infection in dogs has
been previously reported (Gobierno de Navarra, 1987;
Jime ´nez et al., 2002).
In recent years a number of field surveys based on
arecoline purgation or necropsy have been conducted in
Eastern Tibetan China (Budke et al., 2005), Kazakhstan
(Torgerson et al., 2003), and Tunisia (Lahmar et al.,
2001) in order to determine the abundance and
prevalence of E. granulosus in dogs and to investigate
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx7
the transmission dynamic of the parasite. These studies
showed a pattern of age related infection, with young
dogs bearing the highest abundance and prevalence
rates, whereas older animals had lower parasite burdens
and prevalence. Similar results have also been reported
in naturally E. multilocularis infected foxes in high
endemic areas of Germany (Tackmann et al., 2001) and
Switzerland (Hofer et al., 2000). However, other studies
have failed to demonstrate a difference in the E.
multilocularis prevalence rates in dogs or foxes of
different ages (Budke et al., 2005; Tackmann et al.,
2001). It has been suggested that the pattern of age
related infection may be due to the development of a
host protective immune response against the Echino-
coccus reinfection with dog age. The use of mathema-
tical models can provide insight into our understanding
of this phenomenon. Thus, in a high endemic region of
the Tibetan plateau (People’s Republic of China) a
model assuming the presence of host’s immunity was
the best fit for the E. granulosus natural infection in
dogs (Budke et al., 2005). However, the same model
was not valid for the E. multilocularis infection in the
same area. In Kazakhstan Torgerson et al. (2003)
reported that farm dogs with an E. granulosus
prevalence rate of 23% showed a clear age related
infection pattern, whereas a village dog population with
lower prevalence rate of 5.8% (similar to the estimated
E. granulosus prevalence of 8.0% in our study) failed to
demonstrate any decrease in the mean prevalence in
older dogs. Using the same abundance model of Budke
et al., the authors found that assuming the presence of
immunity was the best fit for farm dogs, but not for
village dogs. Taken together these data provide
evidence that in conditions of high infection pressure
the acquired host protective immune response is the
most likely explanation of the observed pattern of age
related infection (Torgerson, 2006).
In a recent studycarried out in a high endemicregion
of northwest Libya, Buishi et al. (2005) reported a
prevalence of E. granulosus of 25.8% in stray dogs by
necropsy and of 21.6% in owned dogs by CpAg ELISA.
A significant inverse correlation between the E.
granulosus CpAg ELISA OD values and the dog age
was found, with dogs aged 3 years or less seeming to
bear the heaviest adult parasite burdens, whereas
intensity of the infection seemed to decrease with the
dog age. In our study no significant linear correlation
(Spearman’s r = ?0.05, 95% CI: ?0.23 to 0.13) could
befound betweenCpAg ELISAOD values and dog age,
suggesting that there were no differences in the
prevalence of E. granulosus between old and young
dogs. As discussed before, this situation may be a
consequence of a low parasite infection pressure, where
an insufficient number of worms are not able to
2006). Other possible explanations may be the inability
of some dogs to develop a host protective immune
response against the challenging infection, or the fact
that some animals were primary infected at an advanced
In conclusion, we have shown that the overall
prevalence of E. granulosus in sheep-dogs in the
provinceof A´lava is8.0%,a rate thatrepresentsa public
health threat. This finding is in line with previous
investigations, corroborating that sheep-dog is the dog
class most vulnerable to be infected by E. granulosus.
The results indicate that the studied sheep-dog
population has access to infected viscera, being home
slaughtering the most probable cause of maintaining the
infection. Under this situation, a control program based
on regular treatment of the dogs with praziquantel
should be initiated.
The authors are very grateful to Dr. Paul Torgerson
(InstituteofParasitology,UniversityofZu ¨rich, Switzer-
land) for his critical revision of the manuscript and
helpful suggestions. We thank the personnel of the
Subdirection of Public Health of A´lava (Department of
Health, Basque Government, Spain) and the Depart-
ment of Agriculture (Provincial Government of A´lava,
Spain), for their technical assistance in the sampling
plan design and sample collection. This work was
financially supported by a grant from the Department of
Health of the Basque Government, Spain. Dr. Aitziber
Benito was a recipient of a PhD studentship from de
Ministry of Education and Science, Spain.
Ahmad, G., Nizami, W.A., 1998. Coproantigens: early detection and
suitability of an immunodiagnostic method for echinococcosis in
dogs. Vet. Parasitol. 77, 237–244.
Allan, J.C., Craig, P.S., Garcı ´a-Noval, J., Mencos, F., Liu, D., Wang,
Y., Wen, H., Zhou, P., Stringer, R., Rogan, R., 1992. Coproantigen
detection for immunodiagnosis of echinococcosis and taeniasis in
dogs and humans. Parasitology 104, 347–356.
Basa ´n ˜ez,M.G.,Marshall,C.,Carabin,H.,Gyorkos,T.,Joseph,L.,2004.
Bayesian statistics for parasitologists. Trends Parasitol. 20, 85–91.
Benito, A., Carmena, D., 2005. Double-antibody sandwich ELISA
using biotinylated antibodies for the detection of Echinococcus
granulosus coproantigens in dogs. Acta Trop. 95, 9–15.
Benito, A., Carmena, D., Postigo, I., Estı ´balez, J.J., Martı ´nez, J.,
Guisantes, J.A., 2003. Intestinal helminths in dogs in A´lava, north
of Spain. Res. Rev. Parasitol. 63, 121–126.
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx8
Benito, A., Carmena, D., Spinelli, P., Postigo, I., Martı ´nez, J., Estı ´-
balez, J.J., Martı ´n de la Cuesta, F., Guisantes, J.A., 2001. The
serological diagnosis of canine echinococcosis by an enzyme
immunoassay useful for epidemiological surveys. Res. Rev. Para-
sitol. 61, 17–23.
Budke, C.M., Jiamin, Q., Craig, P.S., Torgerson, P.R., 2005. Modeling
the transmission of Echinococcus granulosus and Echinococcus
multilocularis in dogs for a high endemic region of the Tibetan
plateau. Int. J. Parasitol. 35, 163–170.
Buishi, I.E., Njoroge, E.M., Bouamra, O., Craig, P.S., 2005. Canine
echinococcosis in northwest Libya: assessment of coproantigen
ELISA,and a survey ofinfectionwithanalysis ofrisk-factors.Vet.
Parasitol. 130, 223–232.
Carmena, D., Benito, A., Eraso, E., 2006. Antigens for the immuno-
diagnosis of Echinococcus granulosus infection: an update. Acta
Trop. 98, 74–86.
Christofi, G., Deplazes, P., Christofi, N., Tanner, I., Economides, P.,
Eckert, J., 2002. Screening of dogs for Echinococcus granulosus
coproantigensin a low endemicsituation in Cyprus. Vet. Parasitol.
Craig, P.S., Gasser, R.B., Parada, L., Cabrera, P., Parietti, S., Borgues,
C., Acuttis, A., Agulla, J., Snowden, K., Paolillo, E., 1995.
Diagnosis of canine echinococcosis: comparison of coproantigen
Parasitol. 56, 293–301.
Craig, P.S., Rogan, M.T., Campos-Ponce, M., 2003. Echinococcosis:
disease, detection and transmission. Parasitology 127 (Suppl.),
Deplazes, P., Dinkel, A., Mathis, A., 2003. Molecular tools for studies
on the tranmisssion biology of Echinococcus multilocularis. Para-
sitology 127, S53–S61.
Palacios, S., 1992. Detection of Echinococcus coproantigens by
ELISA in dogs, dingoes and foxes. Parasitol. Res. 78, 303–308.
Eckert, J., Deplazes, P., 2004. Biological, epidemiological, and clin-
ical aspects of echinococcosis, a zoonosis of increasing concern.
Clin. Microbiol. Rev. 17, 107–135.
Eckert, J., Deplazes, P., Craig, P.S., Gemmell, M.A., Gottstein, B.,
Heath, D., Jenkins, J., Kamiya, M., Lightowlers, M., 2001.
Echinococcosis in animals: clinical aspects, diagnosis and treat-
ment. In: Eckert, J., Gemmell, M.A., Meslin, F.-X., Pawlowski,
Z.S. (Eds.), WHO/OIE Manual on Echinococcosis in Humans and
Animals: A Public Health Problem of Global Concern. World
Organisation for Animal Health/OIE, Paris, pp. 72–99.
Eckert, J., Gemmell, M.A., Matyas, Z., Soulsby, E.J.L., 1984. Guide-
lines for Surveillance Prevention and Control of Echinococcosis/
Hydatidosis. World Organisation for Animal Health, Geneva.
Fraser, A., Craig, P.S., 1997. Detection of gastrointestinal helminth
approaches. J. Helminthol. 71, 103–107.
Gasser, R.B., Jenkins, D.J., Paolillo, E., Parada, L., Cabrera, P., Craig,
P.S., 1993. Serum antibodies in canine echinococcosis. Int. J.
Parasitol. 23, 579–586.
Gasser, R.B., Lightowlers, M.W., Obendorf, D.L., Jenkins, D.J.,
Rickard, M.D., 1988. Evaluation of a serological test system
for the diagnosis of natural Echinococcus granulosus infection
in dogs using E. granulosus protoscolex and oncosphere antigens.
Aust. Vet. J. 65, 369–373.
Gasser, R.B., Lightowlers, M.W., Rickard, M.D., Lyford, R.A., Daw-
kins, H.S., 1990. Serological screening of farm dogs for Echino-
coccusgranulosus infection in an endemicregion.Aust.Vet. J. 67,
Gasser, R.B., Parada, L., Acuna, A., Burges, C., Laurenson, M.K.,
Gulland, F.M.D., Reichel, M.P., Paolillo, E., 1994. Immunological
assesment of exposure to Echinococcus granulosus in a rural dog
population in Uruguay. Acta Trop. 58, 179–185.
Gemmell, M.A., 1990. Australasian contributions to an understanding
of the epidemiology and control of hydatid disease caused by
Echinococcus granulosus—past, present and future. Int. J. Para-
sitol. 20, 431–456.
Gobierno de Navarra, 1987. Informe del Programa de Hidatidosis de
1986. Departamento de Sanidad y Bienestar Social, Pamplona,
Hofer, S., Gloor, S., Muller, U., Mathis, A., Hegglin, D., Deplazes, P.,
2000. High prevalence of Echinococcus multilocularis in urban
red foxes (Vulpes vulpes) and voles (Arvicola terrestris) in the city
of Zurich, Switzerland. Parasitology 120, 135–142.
Jenkins, D.J., Fraser, A., Bradshaw, H., Craig, P.S., 2000. Detection of
Echinococcus granulosus coproantigens in Australian canids with
natural or experimental infection. J. Parasitol. 86, 140–145.
Jenkins, D.J., Gasser, R.B., Romig, T., Zeyhle, E., 1991. Antibody
responses against natural Taenia hydatigena infection in dogs in
Kenya. Int. J. Parasitol. 21, 251–253.
1990. Assessment of a serological test for the detection of
Echinococcus granulosus infection in dogs in Kenya. Acta Trop.
Jime ´nez,S.,Pe ´rez, A.,Gil,H.,Schantz,P.M.,Ramalle,E.,Juste,R.A.,
2002. Progress in control of cystic echinococcosis in La Rioja,
Spain: decline in infection prevalences in human and animal hosts
and economic costs and benefits. Acta Trop. 83, 213–221.
Joseph, L., Gyorkos, T., Coupal, L., 1995. Bayesian estimation of
disease prevalence and the parameters of diagnostic tests in the
absence of a gold standard. Am. J. Epidemiol. 141, 263–272.
Lahmar, S., Kilani, M., Torgerson, P.R., 2001. Frequency distribution
of Echinococcus granulosus and other helminths in a stray dog
population in Tunisia. Ann. Trop. Med. Parasitol. 95, 69–76.
Y., Carmona, C., Kamiya, M., 1997. Coproantigen detection in
dogs experimentally and naturally infected with Echinococcus
granulosus by a monoclonal antibody-based enzyme-linked
immunosorbent assay. Int. J. Parasitol. 27, 1605–1612.
McManus, D.P., Zhang, W., Li, J., Rishi, A.K., 2003. Echinococcosis.
Lancet 362, 1295–1304.
Palmer, S.R., Biffin, A.H., Craig, P.S., Walters, T.M., 1996. Control of
hydatid disease in Wales. Br. Med. J. 312, 674–675.
examinations. Bull. U.S. Army Med. Dept. 8, 326.
Schantz, P.M., Chai, J., Craig, P.S., Eckert, J., Jenkins, D.J., Mac-
Pherson, C.N.L., Thakur, A., 1995. Epidemiology and control of
hydatid disease. In: Thompson, R.C.A., Lymbery, A.J. (Eds.),
Echinococcus and Hydatid Disease. CAB International, Wall-
ingford, UK, pp. 233–331.
Shaikenov, B.S., Torgerson, P.R., Usenbayev, A.E., Baitursynov, K.K.,
Rysmukhambetova, A.T., Abdybekova, A.M., Karamendin, K.O.,
Sloss, M.W., Kemp, R.L., Zajac, A., 1994. Veterinary Clinical Para-
sitology. Iowa State University Press, Ames, IA.
Spinelli, P., Carol, H., Nieto, A., 1996. Niveles de anticuerpos y
antı ´genos circulantes en perros con infeccio ´n natural y experi-
mental por Echinococcus granulosus. Inmunologı ´a 15, 21–29.
Tackmann, K., Loschner, U., Mix, H., Staubach, C., Thulke, H.H.,
Ziller, M., Conraths, F.J., 2001. A field study to control Echino-
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx9
coccus multilocularis-infections of the red fox (Vulpes vulpes) in Download full-text
an endemic focus. Epidem. Inf. 127, 577–587.
Torgerson, P.R., 2006. Canid immunity to Echinococcus spp: impact
on transmission. Parasite Immunol. 28, 295–302.
Torgerson, P.R., Shaikenov, B.S., Rysmukhambertova, A.T., Abdy-
bekova, A.M., Usenbayev, A.E., Baitursinov, K.K., 2003. Model-
ing the transmission dynamics of Echinococcus granulosus in
dogs in rural Kazakhstan. Parasitology 126, 417–424.
Wachira, T., McPherson, C.N.L., Gathuma, J.M., 1990. Hydatid
disease in the Turkana district of Kenya VII analysis of the
infection pressure on definitive and intermediate hosts of E.
granulosus. Ann. Trop. Med. Parasitol. 84, 361–368.
WHO/OIE, 2001. In: Eckert, J., Gemmell, M.A., Meslin, F.-X.,
Pawlowski, Z.S. (Eds.), Manual on Echinococcosis in Humans
Organisation for Animal Health/OIE, Paris.
A. Benito et al./Veterinary Parasitology xxx (2006) xxx–xxx10