Serological Reactivity to Borrelia burgdorferi Sensu Lato
in Dogs and Horses from Distinct Areas in Romania
Timea Kiss,1Daniel Cadar,1Alexandra Florina Krupaci,1Armela Bordeanu,1
Gheorghe Florinel Brudas xca ˘,2Andrei Daniel Mihalca,3Viorica Mircean,3
Lucia Gliga,1Mirabela Oana Dumitrache,3and Marina Spı ˆnu1
Lyme disease is a perfect model of the complex relationship between host, vector, and the vector-borne bacteria.
Both dogs and horses in Romania are exposed to infection. The aim of the present study was to assess the
seroreactivity against Borrelia burgdorferi sensu lato in dogs and horses from different regions of Romania. 276
samples from dogs and 260 samples from horses located in different regions of Romania were analyzed by
ELISA and IFA, respectively. The effect of several factors potentially affecting seroreactivity (location, age,
gender, occupation, and vector exposition risk) was evaluated using Fisher’s exact test (R 2.12.0). The overall
prevalence of anti-Borrelia antibodies was 6.52% (18/276) in dogs, with a significantly higher positivity (46.15%,
6/13, p=0.0005) recorded in a midcountry region. Seroreactivity was correlated with occupation, with working
dogs being more exposed. The results may indicate that Lyme borreliosis foci are restricted to small areas, but
further studies on Borrelia prevalence in tick populations are needed to confirm this hypothesis. In horses, a
global seroprevalence of 11.92% (31/260) was observed. No correlations were found between positive results
and age, sex, county, or occupation. This is the first serological survey on antibodies to B. burgdorferi sensu lato in
Romanian dogs and horses.
Key Words: Borrelia burgdorferi Sensu Lato—Dogs—ELISA—IFA—Horses—Seroprevalence.
and transmitted by ticks belonging to the Ixodes genus. Be-
sides causing a multisystemic disease in humans, it affects a
wide rangeof wild anddomestic animals; among these, Lyme
disease has been well studied in dogs and horses. Even
though only 5%–10% of the exposed individuals show clinical
symptoms (Manion and Bushmich 1998, Bhide et al. 2008,
Leschnik et al. 2010), data regarding exposure to B. burgdorferi
s.l. in these species can provide valuable information on the
infectious potential for humans in a specific location (Salinas-
Mele ´ndez et al. 2001, Duncan et al. 2005). The importance of
dogs and horses as sentinel species had been previously un-
2010, Maurizi et al. 2010) because of the fact that they are host
species for the vector ticks, sharing habitats and being in close
yme borreliosis is a vector-borne disease caused by
spirochetes of the B. burgdorferi sensu lato (s.l.) complex
has been acknowledged in Romania more than 20 years ago
(Cra ˘cea et al. 1988). Studies conducted a decade ago have
measured Borrelia seroprevalence in healthy blood donors and
forestry workers (Hristea et al. 2001). Data regarding the
pathogen’s circulation in enzootic areas is scarce, with limited
studies conducted on lizards and their ticks (Majla ´thova ´ et al.
2008) as well as on unfed ticks collected from the vegetation
(Coipan and Vladimirescu 2010). The aim of this study was to
assess the presence of antibodies against B. burgdorferi s.l. in
several risk factors (age, gender, habitat, exposure to ticks, and
preventive acaricide treatments) was also investigated.
Materials and Methods
Animal sera originated from submontane and montane
dogs from eight counties and 260 horses from four counties
were sampled between 2008 and 2010. Animals were selected
based on availability and owner’s consent, where applicable,
Agricultural Sciences and Veterinary Medicine Cluj-Napoca, Cluj, Romania.
1Infectious Diseases,2Game and Wildlife Diseases, and3Parasitology, Faculty of Veterinary Medicine, University of
VECTOR-BORNE AND ZOONOTIC DISEASES
Volume 11, Number 9, 2011
ª Mary Ann Liebert, Inc.
with samples being obtained from veterinary clinics, shelters,
studs, and private homes.
Information regarding age, gender, service, and in dogs,
exposure to ticks during the previous year, living conditions
(outdoor versus indoor), and application of prophylactic
acaricide treatments was obtained for each animal. Subse-
quently, animals were divided into groups, according to
gender, age, and occupation. Dividing of the animals ac-
cording to their age was done as follows: young: 0–2 years in
dogs, 0–5 years in horses; adult: 3–7 years in dogs, 6–15 years
in horses; old: 8+years in dogs, 16+years in horses. Three
levels of occupation were considered: companion/sport,
stray/wild, and working animals, including shepherd dogs,
hunting dogs, guard dogs, and draft horses. In addition, for
dogs, two risk levels (high versus low) were established. High
risk was defined by the presence of at least one of the fol-
lowing characteristics: detection of ticks during the past year
or at the moment of consultation, living outdoors, and lack of
preventive treatment against ticks. No formal randomized
selection of the animals was applied, but a good approxima-
tion of the general population was aimed at by sampling
animals of different sex, age, and use at each site. Animals
younger than 2 months were excluded from this study. None
of the animals were vaccinated against Lyme disease.
Blood samples were drawn aseptically by cephalic vein
puncture in dogs and jugular vein puncture in horses, using
tubes without anticoagulant, and in some horses, tubes con-
taining EDTA. Sera and plasma, respectively, were separated
and stored at -20?C until processing.
Dog sera was analyzed using a commercial enzyme-linked
immunosorbent assay (Lyme Borrelia Canine IgG - ELISA;
NovaTec Immunodiagnostica, GmbH) for the detection of
specific antibodies of the immunoglobulin G (IgG) class
against B. burgdorferi, according to the manufacturer’s in-
structions, with sera diluted 1:101. Absorbance values were
Interpretation was done by comparing absorbance values of
the samples to that of the provided cutoff.
Indirect immunofluorescence assay (MegaScreen?Fluo-
Borrelia horse;MegaCorDiagnosik) was usedfor thedetection
of IgG antibodies in horse serum or plasma, according to
the manufacturer’s specifications, with samples diluted in
phosphate-buffered saline 1:64. Slides were read using a fluo-
rescence microscope (Axioskop 40 FL Microscope; Carl Zeiss
GmbH) at 400·magnification, comparing each sample to the
visual intensity and appearance of the B. burgdorferi s.l. fluo-
rescence pattern seen in the positive and negative controls.
Statistical analysis of the results was performed using
R.2.12.0 for Windows. Fisher’s exact test was used to analyze
the effect of each explanatory factor (location, age, gender,
occupation, and vector exposure risk) on Borrelia ser-
oprevalence. Values of p<0.05 were considered significant.
A total mean prevalence of anti-B. burgdorferi IgG anti-
bodies of 6.52% (18/276) ranging between 0% and 46.15% in
different counties was observed for dog sera. Detailed results
are presented in Table 1. Significant differences in ser-
oprevalence among locations were observed (p=0.0005),
The presence of antibodies against B. burgdorferi s.l. was sig-
nificantly higher (p=0.0441) in working dogs than in pets
(Table 2). Interestingly, no positive serological reactions were
detected in stray animals. No statistical associations were
found between positive results and age, sex, and risk cate-
The mean seroprevalence of anti- B. burgdorferi IgG in
horses was 11.92% (31/260). The highest seroprevalence was
recorded in Tulcea county (15.51%, 9/58) and the lowest was
found in Bistrit xa-Na ˘sa ˘ud and Constant xa counties (10%, 9/90,
and 2/20, respectively). These differences among locations
were not statistically significant. No significant correlations
were found between age, gender, or service and seror-
Lyme borreliosis has been reported in many countries of
Europe. However, data on the epidemiology of Lyme disease
spirochetes in Romania are deficient, even if the vector tick,
Ixodes ricinus, is widespread (Coipan and Vladimirescu 2010).
To our knowledge, there has been no survey regarding the
presence ofanti-B.burgdorferi antibodiesindogs andhorsesin
Romania up to this date.
Literature-based data document the exposure of various
species of domestic animals to borreliae worldwide. Ser-
oprevalence recorded in dogs in our study was higher than
figures reported in Sweden (3.9%) (Egenvall et al. 2000) and
close to values obtained in Spain (6.9%) (Amusategui et al.
Table 1. Seroprevalence in Dogs and Horses
from Romania: Distribution by Counties
Bistrit xa-Na ˘sa ˘ud
Table 2. Distribution of Seroreactive Dogs
and Horses Regarding Sex, Age, Occupation,
and Risk Group (Dogs)
1260KISS ET AL.
2008) and the Czech Republic (6.5%–10.3%) (Pejchalova ´ et al.
2006,Kybicova ´ etal.2009),butlowerthanvaluesfromTurkey
(27.75%) (Bhide et al. 2008), Bulgaria (22.6%) (Zarkov et al.
2003), Croatia (40%) (Bhide et al. 2004), and Slovakia (33.5%)
(Sˇtefan? cı ´kova ´ et al. 2008b).
In our study, the seroprevalence in horses was higher
compared with data from Turkey (6.3%) (Bhide et al. 2008),
but below the values observed in Sweden (16.8%) (Egenvall
et al. 2000), Slovakia (26.5%) (Sˇtefan? cı ´kova ´ et al. 2008b), Po-
land (25.6%) (Sˇtefan? cı ´kova ´ et al. 2008a), Denmark (29%)
(Hansen et al. 2010), or France (33%) (Maurizi et al. 2010).
Low positivity in horses may be due to a lower infestation
with ticks of one of the studied groups of racing and leisure
animals when compared with the general population (un-
published data). Thus, the obtained results could lead to an
underestimation of the real seroprevalence.
In Europe, at least seven species of the B. burgdorferi s.l.
genospecies complex have been found and the geographical
distribution of these species is variable (Poupon et al. 2006).
Borrelia afzelii, Borrelia garinii, and Borrelia lusitaniae were iden-
tified in ticks from Romania (Majla ´thova ´ et al. 2008, Coipan
and Vladimirescu 2010). The IFA test performed in this study
used whole-cell antigen from B. afzelii, B. garinii, and B. burg-
dorferi Sensu stricto, whereas the ELISA test used purified and
recombinant antigens from the same genospecies. All strains
are either reference strains or strains isolated from Central
Europe. Although no information is available on the extent to
which these antigens allow the detection of antibodies against
various Borrelia species, previous studies indicate that use of
local antigens would insure a higher sensitivity of serological
tests (Sˇtefan? cı ´kova ´ et al. 2008a, 2008b).
The evidence of IgG antibodies can only serve as an indi-
cator of a mere contact with the agent of Lyme borreliosis, but
do not stand for either acute infection or reinfection (Sˇte-
fan? cı ´kova ´ et al. 2008b).
It has been shown that IgG antibodies generally start
emerging about 1–2 months past infection or reinfection and
can persist up to 1–1.5 years in dogs and between 9 months
and 2 years in horses (Bhide et al. 2004, Hansen et al. 2010). In
horses, it is thought that detectable titers of antibodies are
maintained in endemic areas as a result of frequent re-
infestation (Sˇtefan? cı ´kova ´ et al. 2008b). The short persistence of
anti-B. burgdorferi antibodies in dogs and horses, as opposed
equine seroprevalence estimates as a sensitive environmental
risk indicator of the changes in the epidemiology of Lyme
disease, considering that recent reinfection would be needed
to attain detectable antibody titers (Goossens et al. 2001).
The presence of IgM-type antibodies would indicate an
early infection. In this study, we did not quantify IgM-type
antibodies, possibly omitting recently infected animals.
In accordance with the literature, no relationship has been
established between seroreactivity to B. burgdorferi s.l. and
gender (Sˇtefan? cı ´kova ´ et al. 2008a) or age (Couto et al. 2010),
although some authors reported a higher prevalence in older
animals (Amusategui et al. 2008). The lack of correlation may
be a consequence of the limited persistence of anti-B. burg-
dorferi antibodies, which would explain that older individuals
would not show higher seroprevalence as a result of higher
opportunities to be infected throughout their lives.
In dogs, occupation was correlated with seropositivity in
this study, with working dogs being more likely to have a
detectable level of antibodies against B. burgdorferi s.l. As
working dogs were sampled in all locations and not only in
the one with the highest seroprevalence, this correlation is
probably the result of a more frequent exposure to infected
ticks, considering that this group included shepherd dogs,
guard dogs, and hunting dogs. A higher seropositivity in
working and hunting dogs was also documented in Slovakia
(Sˇtefan? cı ´kova ´ et al. 2008b).
The significantly higher number of reactive dogs in Bras xov
county could be explained by the following hypotheses:
(1) On one hand, the presence within the sampling sites of
Lyme borreliosis foci, restricted to small areas, where
the seroprevalence attains a much higher value than in
other regions of the same county, and
(2) On the other hand, the presence of the spirochaete at a
higher rate than in 2001, when Hristea recorded in
Bras xov county a mean seroprevalence of 4.8% in hu-
mans, with higher values in other counties.
To confirm one of these two hypotheses, studies regarding
tick density and prevalence of infection with B. burgdorferi s.l.
in questing ticks from multiple areas should be conducted in
this county. Similarly, to design balanced samplings with re-
spect to the studied explanatory factors, future serological
studies should take into account the notable differences in
seroprevalence between various locations.
Dogs act as reservoirs for B. burgdorferi s.l., and even if
experimental studies on horses are limited, one can assume,
based on the lack of borreliacidal efect of horse serum against
B. burgdorferi Sensu stricto, that this species could also have
1994, Kurtenbach et al. 1998, Bhide et al. 2008). In humans, no
increased risk of seropositivity against Lyme disease was as-
sociated with ownership of dogs, but nevertheless, similar
seropositivity in hunting dogs and humans frequenting the
same areas has been established (Goossens et al. 2001). Such
data highlight the practical value of serological surveys on
these animals as useful epidemiological tools to establish the
emergence of this disease in further regions, to locate endemic
risk of Lyme borreliosis.
This article was supported by the PNII/IDEI/PCCE 7/
2010 grant of CNCSIS (National Council of Scientific Uni-
versity Research of Romania).
No competing financial interests exist.
Amusategui, I, Tesouro, MA, Kakoma, I, Sainz, A´. Serological
reactivity to Ehrlichia canis, Anaplasma phagocytophilum, Neorick-
ettsia risticii, Borrelia burgdorferi and Rickettsia conorii in dogs from
Northwestern Spain. Vector Borne Zoonot Dis 2008; 8:797–803.
Bhide, M, Travnicek, M, Curlik, J, Sˇtefan? cı ´kova ´ A. The impor-
tance of dogs in eco-epidemiology of Lyme borreliosis: a re-
view. Vet Med Czech 2004; 49:135–142.
Bhide, M, Yilmaz, Z, Golcu, E, Torun, S, et al. Seroprevalence of
anti-Borrelia burgdorferi antibodies in dogs and horses in Tur-
key. Ann Agric Environ Med 2008; 15:85–90.
Borrelia burgdorferi S.L. SEROREACTIVITY IN DOGS AND HORSES 1261
Coipan, EC, Vladimirescu, AF. First report of lyme disease spi- Download full-text
rochetes in ticks from Romania (Sibiu County). Exp Appl
Acarol 2010; 52:193–197.
Couto, CG, Leif Lorentzen, L, Beall, MJ, Shields, J, et al. Ser-
ological study of selected vector-borne diseases in shelter dogs
in Central Spain using point-of-care assays. Vector Borne
Zoonot Dis 2010; 10:1–4.
Cra ˘cea, E, Constantinescu, S, Balaci, L, Vizitiu, O, et al. Lyme
borreliosis in Romania. Arch Roum Path Exp Microbiol 1988;
Duncan, AW, Correa, MT, Levine, JF, Breitschwerdt, EB. The
dog as a sentinel for human infection: prevalence of Borrelia
burgdorferi C6 antibodies in dogs from southeastern and mid-
Atlantic States. Vector Borne Zoonot Dis 2005; 5:101–109.
Egenvall, A, Bonnett, BN, Gunnarsson, A, Hedhammar, A, et al.
Sero-prevalence of granulocytic Ehrlichia spp. and Borrelia
burgdorferi sensu lato in Swedish dogs 1991–94. Scand J Infect
Dis 2000; 32:19–25.
Goossens, HAT, van den Bogaard, AE, Nohlmans, MKE. Dogs
as sentinels for human Lyme borreliosis in The Netherlands. J
Clin Microbiol 2001; 39:844–848.
Hansen, MGB, Christoffersen, M, Thuesen, LR, Petersen, MR.
Seroprevalence of Borrelia burgdorferi sensu lato and Anaplasma
phagocytophilum in Danish horses. Acta Vet Scand 2010; 52:3.
Hristea, A, Hristescu, S, Ciufecu, C, Vasile, A. Seroprevalence of
Borrelia burgdorferi in Romania. Eur J Epidemiol 2001; 17:891–
Kybicova ´, K, Scha ´nilec, P, Hulı ´nska ´, D, Uherkova ´, L, et al. De-
tection of Anaplasma phagocytophilum and Borrelia burgdor-
feri sensu lato in dogs in the Czech Republic. Vector Borne
Zoonot Dis 2009; 9:655–661.
Kurtenbach, K, Sewell, HS, Ogden, NH, Randolph, SE, et al.
Serum complement sensitivity as a key factor in Lyme disease
ecology. Infect Immunol 1998; 66:1248–1251.
Leschnik, MV, Kirtz, G, Khanakah, G, Duscher, G, et al.
Humoral immune response in dogs naturally infected with
Borrelia burgdorferi sensu lato and in dogs after immuniza-
tion with a Borrelia vaccine. Clin Vacc Immunol 2010; 17:828–
Majla ´thova ´, V, Majla ´th, I, Hromada, M, Tryjanowski, P, et al.
The role of the sand lizard (Lacerta agilis) in the transmission
cycle of Borrelia burgdorferi sensu lato. Int J Med Microbiol
Manion, TB, Bushmich, SL. Interpretation criteria of Western
blots for the serodiagnosis of Lyme disease in horses. AAEP
Proc 1998; 44:144–145.
Mather, TN, Fish, D, Coughlin, RT. Competence of dogs as
reservoir for Lyme disease spirochetes (Borrelia burgdorferi).
J Am Vet Med Assoc 1994; 205:186–188.
Maurizi, L, Marie ´, JL, Aoun, O, Courtin, C, et al. Seroprevalence
survey of equine Lyme Borreliosis in France and in Sub-
Saharan Africa. Vector Borne Zoonot Dis 2010; 10:535–537.
Pejchalova ´, K, Za ´kovska ´, A, Fucı ´k, K, Scha ´nilec, P. Serological
confirmation of Borrelia burgdorferi infection in dogs in the
Czech Republic. Vet Res Commun 2006; 30:231–238.
Poupon, MA, Lommano, E, Humair, PF, Douet, V, et al. Pre-
valence of Borrelia burgdorferi sensu lato in ticks collected from
migratory birds in Swizerland. Appl Environ Microbiol 2006;
Salinas-Mele ´ndez, JA, de la Garza, SG, Riojas-Valde ´s, VM, Wong
Gonza ´lez, A, et al. Antibody detection against Borrelia burg-
dorferi in horses located in the suburban areas of Monterrey,
Nuevo Leo ´n. Rev Lat Am Microbiol 2001; 43:161–164.
Sˇtefan? cı ´kova ´, A, Adaszek, q, Petˇko, B, Winiarczyk, S, et al.
Serological evidence of Borrelia burgdorferi sensu lato in horses
and cattle from Poland and diagnostic problems of Lyme
borreliosis. Ann Agric Environ Med 2008a; 15:37–43.
Sˇtefan? cı ´kova ´, A, Derda ´kova ´, M, Sˇkardova ´, I, Szesta ´kova ´, E, et al.
Some epidemiological and epizootiological aspects of Lyme
borreliosis in Slovakia with the emphasis on the problems of
serological diagnostics. Biologia 2008b; 63:1135–1142.
Zarkov, IS, Marinov, MM. The lyme disease: results of a sero-
logical study in sheep, cows and dogs in Bulgaria. Rev Med
Vet 2003; 154:363–366.
Address correspondence to:
Department of Infectious Diseases
Faculty of Veterinary Medicine
University of Agricultural Sciences and Veterinary
Ma ˘na ˘s xtur Street, No. 3-5
1262KISS ET AL.