5

REVIEW

Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1
Accepted: 02.02.2022  |  Available on line: 29.06.2022

1National and WOAH Reference Centre for Brucellosis,
Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise ‘G. Caporale’, Campo Boario, 64100 Teramo, Italy.

2Direzione Generale della Sanità Animale e dei Farmaci Veterinari, Ministero della Salute,
Viale G. Ribotta 5, 00153 Rome, Italy.

* Corresponding author at: National and WOAH Reference Centre for Brucellosis,
Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise ‘G. Caporale’, Campo Boario, 64100 Teramo, Italy.

E‑mail: f.demassis@izs.it. 

Fabrizio De Massis1*, Flavio Sacchini1, Antonio Petrini1, Fabio Bellucci1,2, Margherita Perilli1,
Giuliano Garofolo1, Giovanni Savini1 and Manuela Tittarelli1 

Keywords
Brucellosis,
Brucella canis,
Dog,
Zoonosis. 

Canine brucellosis due to Brucella canis:
description of the disease and control measures

Summary
Brucellosis is a contagious disease caused by bacteria of the genus Brucella, which 
can affect different animal species. Dogs may occasionally be infected with B. abortus, 
B. melitensis or B. suis, or by the endemic form of the disease, caused by B. canis. Among the 
brucellosis‑affecting domestic animals, that of the dog is certainly the least frequent, but also 
the least studied. Canine brucellosis due to B. canis represents the dog‑specific brucellosis, 
both because it is the main susceptible animal species, and because it constitutes its 
fundamental reservoir of infection. The disease can also affect humans, although its course 
does not assume the characteristics of severity typical of the infection determined by the 
‘classical’ species of the genus Brucella. In Italy, there are frequent imports of dogs from 
countries where the disease is present, often with non‑controlled movements and without 
sanitary controls. Considering that the zoonotic potential of the disease can be favored by the 
close cohabitation between man and dog, which occurs especially in urban environments, 
canine brucellosis has to be regarded as a public health problem susceptible to introduction 
and spread in the Italian territory.

dog‑specific brucellosis. Dogs in fact are the main 
susceptible animal species, and the fundamental 
reservoir of infection.

The disease can also affect humans, although its 
course does not assume the characteristics of 
severity typical of the infection due to the ‘classical’ 
species of the genus Brucella; however, in countries 
where the disease is present (particularly in the 
American continent) it is considered an authentic 
zoonosis.

In Italy, there are frequent imports of dogs from 
countries where the disease is present, often with 
non‑controlled movements and without sanitary 
control. Considering that the zoonotic potential of 
the disease can be favored by the close cohabitation 
between man and dog, which occurs especially in 
urban environments, canine brucellosis has to be 
regarded as a public health problem susceptible to 
introduction and spread across the Italian territory.

Introduction
Brucellosis is a contagious disease caused by 
bacteria of the genus Brucella, which can affect 
different animal species. Dogs may occasionally 
be infected with B.  abortus, B.  melitensis or B.  suis, 
or be affected by the endemic form of the disease, 
caused by B.  canis. Among the brucellosis‑affecting 
domestic animals, that of the dog is certainly the 
least frequent, but also the least studied.

Canine brucellosis due to B.  abortus or B.  melitensis 
is uncommon in dogs living in close contact with 
infected ruminants. In these cases, it only represents 
an epiphenomenon of the infection circulating in 
the affected farm, and the dog does not play the 
role of disease reservoir. The same applies for dog 
brucellosis due to B. suis, which is rarely detected in 
dogs and, in any case, always in connection with a 
coexisting infection in pig farms.

Canine brucellosis due to B.  canis represents the 



6 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

investigation on several outbreaks of abortion and 
infertility in dogs occurring in different areas of 
the United States of America (Carmichael 1966). 
Subsequently, B.  canis was isolated in different 
countries of the world, while in others its presence 
has been suspected based on serological positivity. 
Recently, Hensel and colleagues (Hensel et  al. 2018) 
have collected the information available from 
the international literature regarding serological 
investigations carried out for canine brucellosis due to 
B. canis worldwide (Figure 1). These studies show that 
there is a wide level of variability in serum positivity 
(from 1% to 28%), depending on the country studied 
and the sample taken (Hensel et al. 2018).

Seroprevalence is strongly influenced by the type 
of serological test used and the interpretation of its 
results, due to the fact that B.  canis shares antigen 
determinants with all R‑phase (rough) Brucellae and 
with microorganisms of other genera not related to 
the genus Brucella (Carmichael 1990). Furthermore, 
apart from the true disease prevalence in the 
area under study, the wide spread of high level of 
seroprevalence recorded could also be attributed to 
the sampling design and the algorithm used for the 
interpretation of test results (Hensel et al. 2018).

In Italy, several surveys reported the presence of 
seropositivity at various levels and in different 
geographical areas (Tolari and Pizzirani 1978, Colella 
et  al. 1980, Ciuchini et  al. 1982, Vesco et  al. 1987, 
Valente et al. 1991, Buonavoglia et al. 1992, Prosperi 
et al. 1994, Casalinuovo et al. 1996). The only clinical 
case described in scientific literature dates to 2008, 
when Corrente and colleagues (Corrente et al. 2010) 
detected B. canis by PCR in a half‑breed dog suffering 
from chronic prostatitis and discospondylitis. In 
2020, B.  canis has been detected and isolated for 
the first time in a commercial breeding kennel 
(De Massis et al. 2021).

Decription of the disease 

Etiology
Brucellae are Gram‑negative coccobacilli or short 
rods measuring from 0.6 to 1.5 µm long and from 
0.5 to 0.7 µm wide. They are usually arranged singly, 
and less frequently in pairs or in small groups. The 
morphology of Brucella is fairly constant, except 
in old cultures where pleomorphic forms may be 
evident. Bacteria of the genus Brucella are normally 
non‑motile. They do not form spores, or pili; true 
capsules are not produced. They are aerobes and 
asporigenic.

With regards to B.  canis, it has no biovar, does not 
require carbon dioxide to grow in first isolation, and 
grows in presence of thionin but not with basic fuchsin. 
In first isolation, colonies of B.  canis always occur in 
phase R (rough) or M (mucoid), while their existence 
in phase S (smooth) has never been reported. B. canis 
does not agglutinate in the presence of Brucella A‑ and 
M‑ monospecific antisera, while it agglutinates with 
specific antisera towards the R antigen of B. ovis (REO 
198 strain). Furthermore, cross‑reactions with surface 
antigens of other species of the genus Brucella in a 
non‑smooth phase are also possible. B. canis does not 
produce H

2
S, does not oxidize the substrate based on 

L‑Asparagine or D‑Xylose, does not reduce nitrates to 
nitrites, and does not show lysis in the reaction with 
different phages, except with the phage R/C (Corbel 
and Brinkley‑Morgan 1989).

Epidemiology

Geographical distribution

B.  canis was first reported in 1966 during an 

Figure 1. Location and outcome of serological investigations for the detection of antibodies against B. canis recorded in the international scientific 
literature. Each point represents a published study; the color of the point represents the identified source (from Hensel et al. 2018)

Seroprevalence (%)

0-6
6.1-20
20.1-≈30



7Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

of infected males is high; the elimination of the 
microorganism in low concentrations for a period 
exceeding 60 weeks was also observed (George et al. 
1979). Under experimental conditions, the isolation 
of B.  canis from prostate and epididymis was 
possible up to more than two months after the end 
of bacteremia, and the venereal transmission of the 
disease continued for at least two years in clinically 
normal appearing dogs. Even after castration, 
males may remain a source of infection because the 
bacterium can persist in the prostate and lymphatic 
tissues (Carmichael 2012).

Urinary elimination begins a few weeks after the 
onset of bacteremia and continues for at least 
three months. In the urine of infected males, 
concentrations of microorganisms from 103 to 106 
Colony Forming Units (CFU)/ml have been found, 
while for females this number is lower (Serikawa and 
Muraguchi 1979).

The bacteria have also been isolated from saliva, nasal 
and ocular secretions of dogs (Carmichael and Joubert 
1988, Moore 1969). Unlike what occurs in the male, it 
seems that in females urine is not important for the 
transmission of the disease, due to the low number 
of microorganisms present (Carmichael 1990). The 
fact that the male eliminates the microorganism 
with urine in greater quantities is probably related 
to the specific localization of B.  canis in the prostate 
and epididymitis or contamination with seminal fluid 
(Carmichael and Joubert 1988). Despite this evidence, 
according to Henesel and colleagues (Hensel et  al. 
2018), the role of urine as a means of transmission of 
canine brucellosis from B. canis has not yet been fully 
clarified (Hensel et al. 2018).

Pathogenesis and pathological lesions
Although the pathogenesis of B. canis infection has 
not yet been fully elucidated, it is likely it follows 
the general pattern of Brucella infection common 
to other animal species. The route of entry of the 
pathogen are the conjunctival, oral, and genital 
mucous membranes (Carmichael 1990). In the dog, 
the minimum oral infectious dose is around 2 × 106 
CFU, while the minimum infectious dose via venereal 
route has not been determined yet, although it 
is assumed to be slightly lower (Carmichael and 
Joubert 1988). The infectious dose via conjunctival 
route is between 104 and 105 CFU (Serikawa and 
Muraguchi 1979).

After oral infection, the incubation period is variable, 
however a bacteremia associated with leukocytes 
can normally be detected one or two weeks after 
infection (Carmichael and Joubert 1988).

A characteristic feature of the infection, also 
significant from a diagnostic point of view, is the 
prolonged bacteremia. Commonly, it is possible to 

Susceptible species

Unlike S‑phase (smooth) Brucellae, which are able 
to infect different animal species, B.  canis has a 
limited range of possible hosts. Dogs and wild 
canids are thought to be the only significant hosts 
for B. canis. Cattle, sheep and swine were found to be 
highly resistant to the infection. Cats appear to be 
moderately sensitive and, following oral exposure to 
B. canis, can develop bacteremia with low antibody 
titers. Guinea pigs, mice, rats, and non‑human 
primates are also susce ptible to experimental 
infection. (Carmichael 1990). The rabbit is more 
sensitive than other laboratory species, developing 
orchitis and peritoneal abscesses following high 
doses inoculated through intraperitoneal route 
(Carmichael and Bruner 1968).

Humans can get B.  canis infection through direct 
contact with infected dogs, with their reproductive 
secretions, or with their blood (Lawaczeck et  al. 
2011, Lucero et  al. 2010). Clinical disease has been 
described.

Transmission

Canine brucellosis due to B.  canis is especially 
common among stray dogs, in shelter kennels, 
in commercial breeding kennels, or in places 
where they live in large groups (Carmichael and 
Joubert 1988). When B.  canis is introduced into a 
kennel, the spread is rapid (Carmichael and Bruner 
1968). B.  canis can be transmitted from infected to 
healthy adult males after a few weeks or months 
of cohabitation. Conversely, it is not transmitted 
between non‑sexually mature males or females 
(Carmichael and Joubert 1988).

A number of natural pathways of transmission of 
canine brucellosis have been observed, but the most 
common is the contact with placenta, fetal tissues 
and vaginal discharges resulting from abortion 
due to B. canis infection. Moreover, transmission is 
also possible during estrus or mating (Carmichael 
1990). Vaginal discharge can contain more than 
1010 microorganisms per ml, and elimination by 
this route can continue for several weeks after 
abortion (Carmichael and Joubert 1988). Other body 
secretions contain lower concentrations of B.  canis 
and are less important for the spread of infection.

Most puppies become infected in the uterus. 
Infected mothers or their milk also represent a 
potential source of infection for those which survive 
infection.

Seminal fluid and urine of dogs harboring 
B.  canis in the prostate and epididymis have been 
demonstrated as an important source for disease 
spreading. During the first 6‑8 weeks of infection, 
the concentration of B.  canis isolated from semen 



8 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

is a unilateral or bilateral lymph node enlargement 
at sites of bacterial entry (usually retropharyngeal 
lymph nodes for the oral route and superficial 
inguinal lymph nodes for the vaginal route). The 
lymphadenopathy is the only detectable symptom 
in the non‑pregnant adult female, while prostatitis 
and epididymitis can be observed in the adult male.

In pregnant females, the main symptom is late 
abortion, which generally occurs between the 45th 
and 55th day of gestation (Carmichael 2012). The 
abortion is followed by a yellow to brown odorless 
vaginal discharge, that persists from one to six 
weeks (Carmichael 1968).

Cases of embryonic resorption or abortion from 
the 10th to the 20th day of mating have also been 
described. They often go unnoticed in practice, as 
they are normally considered as an infertile mating 
(Carmichael 1990).

In exceptional cases, pregnancy can be brought to the 
end, with simultaneous birth of both live and dead 
puppies (Feldman and Nelson 1996). Most live‑born 
puppies die within a few hours or days; those that 
survive normally show a generalized increase in 
lymph node volume, which is the main symptom 
until they reach sexual maturity (Carmichael 1990). 
These puppies may be bacteriemic but apparently 
healthy (Carmichael 2012). In addition, as a systemic 
manifestation of B. canis infection, surviving puppies 
usually show persistent hyperglobulinemia, and 
some of them may have transient fever, leukocytosis, 
or convulsions (Carmichael 2012). It is possible that 
apparently healthy puppies born from infected 
mothers spread B.  canis to other dogs and humans 
(Dentinger et al. 2012).

As with brucellosis affecting other animal species, 
B. canis infection in dogs does not interfere with the 
normal oestrus cycle. Actually, it has been found 
that more than 85% of females which have had 
abortions because of B.  canis infection can have 
normal gestations with regular births, while the 
remaining females may still experience reproductive 
problems, which can also occur intermittently. More 
rarely, infected females can have abortions more 
than four consecutive times, or have more than 
three unsuccessful mating (Carmichael 1990). These 
animals represent a reservoir of infection for the still 
healthy dogs inside the kennel (Carmichael 1968, 
Carmichael 2012).

In infected adult males, epididymitis, unilateral or 
bilateral testicular atrophy and scrotal dermatitis are 
normally found. Palpation of the scrotum or testicle 
generally does not cause acute pain. Generally, there 
is a decrease in the volume of ejaculate without 
loss of libido by the subject; however, it is possible 
to notice some suffering of the subject at the time 
of ejaculation. The seminal liquid of infected males 

find in the blood the presence of a number of bacteria 
even more than 103 per ml, starting 2‑3 weeks after 
infection. Bacteremia commonly persists for at least 
one or two years. In some cases, it has also been 
observed for a period of five years (Carmichael et al. 
1984). The bacteremia, constant in the acute phase 
of the disease, becomes intermittent in the chronic 
phase (Alton et al. 1988). B. canis can be isolated for 
several months from spleen, lymph nodes, bone 
marrow, prostate, and epididymis even when the 
bacteria cannot be found in the blood any more 
(Carmichael 1990).

No specific lesions due to B.  canis infection 
have been described; lymph node hypertrophy 
and splenomegaly can be observed in infected 
adults and  surviving puppies. Acute or chronic 
inflammatory lesions (Carmichael 1990) can also 
be found in the genital system. Microscopically, a 
generalised lymphoreticular hyperplasia affecting 
all lymphoid organs is costantly observed despite 
the bacteremia (Serikawa and Muraguchi 1979). 
In chronic bacteremia, lymph node and splenic 
sinuses can show infiltration of plasma cells and 
macrophages containing phagocytised bacteria. In 
all organs of the urogenital system, a widespread 
lymphocytic infiltration can be observed at the 
submucosa level, which mainly affects the prostate, 
epididymis, renal pelvis and uterus (Carmichael 2012). 
Subacute or chronic endometritis, granulomatous 
prostatitis, and testicular atrophy  and fibrosis are 
common in chronically infected dogs. Other lesions 
may involve the kidney, with ialine thickening 
of the glomerulus basal membrane, with poor 
cell infiltration. Other described alterations 
include focal hepatic necrosis, myocarditis, and 
meningoencephalitis (Carmichael 1990). Alterations 
at the eye level are represented by granulomatous 
iridociclitis and exudative retinitis, characterized 
by widespread infiltration of lymphocytes, plasma 
cells and neutrophils (Saegusa et al. 1978). Aborted 
fetuses show subcutaneous edema, congestion, 
and hemorrhages in the subcutaneous areas of 
the abdominal region (Carmichael 2012). Placental 
annexes might contain foci of coagulative necrosis 
in the chorionic villi, and a large number of bacteria 
within epithelial trophoblastic cells (Feldman and 
Nelson 1996).

Symptomatology
Although B. canis infection is systemic, infected adult 
dogs rarely show obvious symptoms (Carmichael 
1990). Apart from rare cases, the disease evolves in 
the absence of fever (Carmichael and Bruner 1968). 
There are no pathognomonic symptoms (Hensel 
et al. 2018).

A common symptom for both males and females 



9Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

affecting the fractions β and  γ can be indicative of 
B. canis chronic bacteriemia. Urine biochemical tests 
are generally normal, despite the possible presence 
of bacteriuria.

In cases of meningoencephalitis, analysis of 
cerebrospinal fluid reveals the presence of 
neutrophils and an increase in protein concentration, 
changes that are not evident if only discospondylitis 
is present. Radiological findings of intervertebral disc 
infection should be followed by serological testing 
for B.  canis and, where possible, bacteriological 
confirmation (Carmichael 2012).

Indeed, the examination of the seminal fluid may 
be of greater help in the presumptive diagnosis of 
the disease. Sperm abnormalities which may affect 
more than 90% of spermatozoa, start being present 
5 weeks after infection and become evident from 
the 8th week onwards even after 20 weeks (George 
et al. 1979). In any case, the sperm disorders induced 
by B. canis infection are not specific, so the detection 
of an abnormal spermogram is not indicative of 
brucellosis (Berthelot and Garin‑Bastuji 1993).

Laboratory diagnosis

Direct methods

Isolation

The isolation of the bacterium represents the only 
proof of B. canis infection. It is not difficult to isolate 
B.  canis since the microorganism grows in the 
culture media normally used for all other species of 
the genus Brucella (Nicoletti and Chase 1987b).

Isolation of B.  canis is possible from blood, fresh 
samples (vaginal discharged material, placental and 
fetal tissues, urine, semen, milk) and samples taken 
from necropsy (lymph nodes, spleen, prostate, 
epididymis, uterus, bone marrow, eye, intervertebral 
discs) (Feldman and Nelson 1996).

Blood culture is the method of choice for confirming 
B.  canis infection, provided that they have not 
been treated with antibiotics (Carmichael 2012). In 
dogs experimentally infected per os, bacteremia 
begins 7 days after exposure, while blood culture 
begins to provide positive results from 2‑3 weeks 
after exposure. If in the meantime the subject has 
not been treated with antibiotics, the positivity to 
blood culture usually persists for at least 6 months, 
and may last up to 2 years (Carmichael 1990). In this 
regard, the international scientific literature also 
reports cases of experimentally infected dogs that 
remained positive to the blood culture for a period 
of 5.5 years (Carmichael et al. 1984).

has a considerable number of sperm abnormalities 
and inflammatory cells, especially during the first 
three months of infection. Chronic infection can 
lead to bilateral testicular atrophy with consequent 
aspermia (Carmichael 2012).

In addition to symptoms affecting the reproductive 
sphere, although less frequently, B.  canis infection 
can cause disorders in other organs as well. In some 
subjects, for example, generalized lymphadenopathy 
may be accompanied by splenomegaly (Carmichael 
2012). A well‑known and described clinical 
manifestation of B. canis infection is discospondylitis 
(Henderson et al. 1974). This can arise in apparently 
healthy dogs or those which have had history of 
reproductive disorders and have been treated with 
antibiotics (Kervin et  al. 1992, Hurov et  al. 1978). 
Dogs initially experience pain in the spinal cord, 
then, if compression of the spinal cord increases, 
they present paresis and ataxia (Kornegay 1983). 
The incidence of discospondylitis is higher in males 
than in females, probably due to the localization of 
B. canis in the prostate, which can cause intermittent 
bacteremia even in castrated males (Carmichael and 
Joubert 1988, Kerwin et al. 1992, Hurov et al. 1978).

Cases of osteomyelitis affecting the appendicular 
skeleton and resulting in lameness on the affected 
limb have also been described (Smeak et  al. 1987). 
Although the onset of meningoencephalitis 
has only been observed as a consequence of an 
experimental infection, some authors have detected 
the presence of behavioral changes, anisocoria, 
ataxia, hyperesthesia and circle movements in dogs 
with B.  canis infection, in which such neurological 
signs started three weeks after mating (Carmichael 
2012). In some dogs with chronic B.  canis infection, 
the onset of recurrent anterior uveitis with corneal 
edema has also been described (Saegusa et al. 1978, 
Reike and Rhodes 1975). Other manifestations of 
B. canis infection may be represented by polyarthritis 
(Carmichael 2012).

Diagnosis

Clinical diagnosis
The clinical signs may only allow a suspicion of 
B.  canis infection. Since B.  canis infection is one 
of the most common causes of reproductive 
disorders in dogs, it should be ruled out before 
investigating other causes of infertility or abortion 
(Carmichael 2012). However, if reproductive failure 
is not observed, canine brucellosis can be difficult to 
diagnose.

The hematological and biochemical values remain 
generally unchanged or in any case they don't 
show characteristic changes. A hyperglobulinemia 



10 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Given the potential impact that a positive or 
negative result can have on individual dogs, on the 
dog population of origin, on a single customer or 
on a kennel operator, quality control and assurance 
of the B.  canis diagnostic tests are fundamental, in 
particular for PCR (Cosford 2018).

Assuming that the accuracy of the results can be 
improved, the advantages of PCR include: the 
possibility of detecting the species and (sometimes) 
the Brucella biovar involved in the infection, the 
high sensitivity and specificity, the minimum 
requirements needed in terms of biological 
containment, the relatively short time to obtain the 
results, and the possibility of performing a genetic 
fingerprinting which is useful for epidemiological 
investigations aimed at controlling the disease (Yu 
and Nielsen 2010).

Most of the PCR assays reported in the literature are 
based on the identification of the Brucella genus. 
In the past, multiplex PCRs have also been set up 
to distinguish between some Brucella species, 

To avoid the excessive development of contaminants, 
blood samples should be sowed on selective 
media (Carmichael 2012). Blood samples should 
be collected in a sodium citrate or sterile lithium 
heparin blood collection tube, stored at refrigeration 
temperature (not frozen) and submitted within 24 
hours at the laboratory, where they can be cultured 
on a Farrell's medium or on a modified Thayer‑Martin 
medium (Carmichael and Greene 2006, CFSPH 2010, 
GDA 2020).

At necropsy lymph nodes and spleen are the 
recommended organs for culturing the bacterium. 
It is also possible to collect samples from the eye in 
case of uveitis or the intervertebral disc in case of 
discospondylitis (Feldman and Nelson 1996). In all 
these cases, the negativity of the culture does not 
exclude the presence of infection (Carmichael 1990).

Unfortunately, the chances of growing this 
microorganism in culture depend on several factors, 
including B.  canis concentration in the collected 
sample, the intermittent germ elimination from 
infected subjects, due to the quality of samples, the 
sample handling, the forms of the microorganism, 
or how culture media have beeen used (Carmichael 
and Greene 2006). Therefore, a negative culture 
cannot exclude the presence of infection, since 
the low sensitivity of the test can lead to relatively 
high number of false negatives. Although culture 
is to be considered inappropriate as a screening 
test, it still represents the ideal test for confirming 
infection. The test can give positive results as early 
as 2‑4 weeks after infection and the dog can remain 
positive for several years (Hollett 2006).

Isolating the bacterium from samples is the best 
method for the diagnosis of early infection in dogs 
that have not received antibiotic treatment.

Expected results of the culture according to the 
time of sampling and the material examined are 
summarized in Table I. 

Polymerase chain reaction (PCR)

Several PCR primers have been used to detect 
B. canis DNA in whole blood, vaginal secretions and 
semen. Due to its high sensitivity and specificity, 
PCR can be used as a rapid screening test or as 
confirmation test for serum positive dogs (Keid et al. 
2007, Kauffman et  al. 2014, Kang et  al. 2014). There 
is a good correlation between PCR test results and 
those obtained with the 2ME‑RSAT test (Mol et  al. 
2020). The matrix of choice is the whole blood taken 
with anticoagulant sodium citrate (light blue cap 
tubes). PCR testing on serum has little diagnostic 
value as it provides unsatisfactory results in terms 
of sensitivity, much lower than those of blood 
culture, whole blood PCR, RSAT and 2ME‑RSAT 
(Keid et al. 2010). 

Table I. Possible results of the culture of B. canis in relation to the time 
at which it is performed and the material under examination (adapted 
from Feldman & Nelson 1996).

Culture material
Optimal moment of 

cultivation
Expected results

Post-abortion 
discharges

When present + + +

Placenta When present + +

Fetus When present Possible negativity

Seminal fluid

3-11 weeks PI + + +

12-60 weeks PI Reduction in number

> 60 weeks PI -

Blood

5-30 weeks PI 100% +

After 30 weeks PI Intermittently

6-12 months PI >80% +

28-48 months PI 50-80% +

48-58 months PI 25-50% +

> 58 months PI < 25% +

Epididymis
35-60 weeks PI 50-100% +

> 100 weeks PI -

Urine 8-30 + weeks PI
Emission of more 

microorganisms in males 
than in females

Prostate Up to 64 weeks PI Usually +

Lymph nodes, bone 
marrow and spleen

During bacteremia Usually +
During the 

non-bacteriemic 
phase phase

+/-

Eye In case of uveitis + + +

Intervertebral disks
In case of 

discospondylitis
+/-

PI = Post-infection;    + = Positivity;    - = Negativity.



11Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

serological tests aimed at detecting antibodies 
against B. canis should be based on a wall antigen of 
B.  canis or B.  ovis (complete cross‑reaction), or on a 
cytoplasmic antigen of B. abortus or B. canis. Whichever 
test or antigen is used, B. canis antibodies in the blood 
are usually detectable not earlier than 5‑8  weeks 
after infection (Carmichael 1990). The results then 
may be negative in the first 3‑4  weeks of infection, 
although the dog may be bacteremic as early as two 
weeks after infection (Carmichael and Greene 2006). 
When the bacteremia ceases, the antibody titer starts 
descreasing. However, level of detectable antibodies 
can be found for further 4‑6 months. After that period, 
serological tests have negative or doubtful results 
(Carmichael 1990). The antibody titer may fluctuate 
even in the presence of persistent bacteremia and 
its value does not reflect the disease progress. 
Furthermore, a possible decrease of this value does 
not necessarily indicate the efficacy of the therapy 
(Feldman and Nelson 1996).

The characteristics of the main serological tests 
for B.  canis antibodies have been summarized by 
Carmichael and Shin (Carmichael and Shin 1996, 
Table III), and by Cosford (Cosford 2018, Table IV).

Rapid slide agglutination test (RSAT and 
2ME‑RSAT )

RSAT in the original technique elaborated by George 
and Carmichael (George and Carmichael 1978) 
uses a heat‑inactivated B.  ovis or B.  canis, colored 
with Rose Bengal. Ebani and colleagues (Ebani et al. 
2003) performed the test with an antigen from 
heat‑inactivated and Rose Bengal‑stained B.  ovis 
strain 63/290 prepared according to Alton and 
colleagues (Alton et  al. 1988). Currently, the strain 
recommended by the OIE for the production of 

including B. canis (López‑Goñi et al. 2011). In recent 
years, specific PCRs for B.  canis (Kaden et  al. 2014, 
Kauffman et  al. 2014, Kang et  al. 2014, Boeri et  al. 
2018) have also been developed. The performance 
of these assays however needs to be completed 
in order to assess sensitivity and specificity. Until 
then, data from PCR trials should only be used in 
conjunction with clinical and serological information 
(Cosford 2018).

Cosford (Cosford 2018) has recently reviewed the 
state‑of‑the art of the different researches available 
to date on PCR for B. canis (Table II).

Indirect methods

Several serological methods are available 
for detecting antibodies against B.  canis. The 
Rapid Slide Agglutination Test (RSAT), the 
2‑Mercaptoethanol‑Rapid Slide Agglutination 
Test (2ME‑RSAT), and the Tube Agglutination 
Tests (TAT, and 2ME‑TAT) represent, in countries 
where the disease is present, the tests most widely 
used in the field. The Agar Gel Immuno‑Diffusion 
test (AGID), that uses either cell‑wall antigens 
(AGIDcwa), or cytoplasmic antigens (AGIDcpa), 
is used as confirmatory test. Other available 
tests include indirect fluorescent antibody 
assay (IFA), Complement Fixation Test (CFT) 
and immunoenzymatic tests (Enzyme‑Linked 
Immunosorbent Assay, ELISA). They, however, have 
only been used in experimental studies (Carmichael 
and Shin 1996, Ebani et  al. 2003, Carmichael and 
Greene 2006, Hollett 2006, Keid et al. 2009).

It is important to note that B.  canis does not have 
smooth‑phase cell wall antigens such as B.  abortus, 
B.  melitensis or B.  suis, but only rough antigens 
(Carmichael and Bruner 1968), like B.  ovis. Therefore, 

Table II. PCR assays for the detection of Brucella canis in dogs from Cosford, edited (Cosford 2018).

16S-23S rDNA interpace region (44) Brucella genus Whole blood 100% 100%
16S-23S rDNA interpace region (45)a Brucella genus Vaginal swabs Not available Not available

16S-23S rDNA interpace region (46)a Brucella genus Semen Not available Not available

16S-23S rRNA sequence (47) Brucella genus Whole blood 100% 100%

16S-23S rRNA interpace region (48) Brucella genus Inguinal lynph nodes 100% 100%

16S-23S rRNA interpace region (49) Brucella genus
Whole blood

Serum
Whole blood: 97.14%

Serum: 25.71%

Intergenic Spacer IS711 823)b Brucella genus
Males: proputial swabs, semen or urine

Female: vaginals swabs
Not available Not available

Gene fragment on chromosome 1 (23)b Brucella canis specie-specific As intermediately above Not available Not available
B. canis outer membrane protein 25 DNA 

quantitative PCR (50)
Brucella canis specie-specific

Vaginal swabs
Whole blood

Vaginal swabs: 92.31%
Whole blood: 16.67%

BCAN_B0548-0549 region in chromosome 2 
of Brucella canis (51)c

Brucella canis specie-specific
Whole blood
Buffy coats

Not available Not available

aPCR on vaginal swabs and semen in these studies correlates with blood PCR and blood coltures assessed by a Kappa-coefficient and the Mc Nemar test.
bBoth Brucella canis genus-based and Brucella canis specific PCRs used in Swedish outbreak investigation.
cBrucella canis inoculated samples; PCR on Buffy coat separated from whole blood was approximately 100 times more sensitive that from whole blood.



12 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Table IV. Comparison of traditional serological tests for the diagnosis of canine brucellosis. From Cosford, edited (Cosford 2018).

Test Antigen Sensitivity Specificity How to use the test

RSAT
B. ovis

(M-) strain B. canis

Moderate to high Low to moderate

Screening testolder studies suggest high older studies suggest 40%-50%

newer studies suggest 70.58% newer studies suggest 83.34%

2-ME-RSAT (M-) Strain of B. canis Lower than RSAT (31.76% vs. 70.58%) Higher than RSAT (100% vs. 83.34%) Confirmatory test

TAT B. canis High Low Screening test

IFA
Anti-canine immunoglobulin (Ig)G 

directed against antibodies
to B. canis

Unknown Unknown Screening test

AGID
cwa

Lipopolysaccharide antigen from
the cell wall of B. canis

High Lower than AGID
cpa

Screening test

AGID
cpa

LPS-free soluble, internal 
cytoplasmic proteins extracted

from B. canis or B. abortus

Low High

Confirmatory test52.94 sensitivity 100%

47.06 false negatives

cross‑reactions between the antigen used and specific 
antibodies, possibly present in the tested serum. 
Possible cross‑reactions have been observed against 
bacteria, such as Pseudomonas  spp., Bordetella  spp., 
Streptococcus  spp. and, more generally, some 
Enterobacteriaceae (Carmichael 2012).

In order to reduce the incidence of false positivity, 
the RSAT test was then modified to include the 
addition of 2‑mercaptoethanol (2ME‑RSAT) to 
the test serum before mixing with the antigen. 
2‑mercaptoethanol reduces the incidence of false 
positivity substantially because it inactivates the less 
specific IgM antibodies (Badakhsh et  al. 1982). The 

B.  ovis antigen is B.  ovis REO 198 (OIE 2018). RSAT is 
a quick, easy to perform and read test, commercially 
available (e.g. D‑Tec CB, Synbiotics, San Diego CA, 
USA). Positive reactions are detectable as early as 
3‑4 weeks after the onset of infection (Hollett 2006). 
The test has a sensitivity (defined as the probability 
of the test not to give rise to false negative reactions) 
of 99% (Carmichael 1990). On the contrary, specificity 
(defined as the probability of the test not to give rise 
to false positive reactions) is rather limited, counting 
false positive rates that commonly range from 20% 
up to even 50% (Carmichael and Shin 1996). False 
positive results would apparently be produced by 

Table III. PCR assays for the detection of Brucella canis in dogs (Cosford 2018).

Test
Nature of the 

antigen
Positivity limitsa Comments

2ME-RSAT Cell wall From 5-8 Sept. PI up to 3 months after the cessationof the bacteremia (followed by variable results)

High sensitivity (99%);
low specificity (50%-80%);
quick and easy performing.

TAT Cell wall Similar to 2ME-RSAT.
False positivity as in 2ME-RSAT;

semi-quantitative;
titles above 1:200 are indicative of infection in progress.

2ME-TAT Cell wall Similar to 2ME-RSAT. Slightly higher specificity than TAT;longer laboratory process.

AGID
cpa

Cell wall Similar to 2ME-RSAT,possibility of detection 1-2 weeks before.

higher Sensitivity than RSAT;
permanence of frequent non-specific reactions;

complexity of execution;
difficulties of interpretation

AGID
cwa

Cytoplasm From 8-12 weeks to at least 12 monthsafter the end of the bacteremia, up to 36 months

Greater specificity (97%);
but less sensitivity;

reveals chronic cases negative to other tests;
reveals infections from other Brucellae.

ELISA Cell wall Cytoplasm Unknown, believed to be similar to TAT
Technique in experimental phase;

high specificity when using wall antigens of B. canis
in phase M or cytoplasmic antigens.

IFA Cell wall Unknown Unpublished data;appears to have less sensitivity than 2ME-TAT.
aTimes are approximate;    RSAT = Rapid slide agglutination test;    ME = Test with 2-mercaptoethanol;    TAT = Tube agglutination test;    AGID = Agar gel immunodiffusion test; 
ELISA = Immunoenzymatic test;    IFA = Immunofluorescence assay;    PI = Post-infection;    + = Positive;    - = Negative.



13Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

other species of the genus Brucella (AGIDcpa), 
B.  abortus in particular (Carmichael and Shin 1996). 
The technique has been described by Zoha and 
Carmichael (Zoha and Carmichael 1982).

The use of cytoplasmic‑derived antigen further 
increases the specificity of the test, lowering 
the percentage of false positivity to 3% (Zoha 
and Carmichael 1982). Conversely, the same test 
performed using the cell wall antigen, suffers 
from the same false positivity problems as the 
agglutination tests previously described (Carmichael 
2012). To produce the antigen, Ebani and colleagues 
(Ebani et al. 2003) used a hot saline extract (HSE) of 
B.  canis strain RM6/66. The lipopolysaccharide wall 
antigen is less specific than the cytoplasmic antigen. 
Therefore, AGIDcwa has a high sensitivity but it still 
has the probability to give false positive results. 
Positive results appear 8‑12 weeks after infection 
and can remain for 3‑4 years (Hollett 2006). 

The assay which uses cpa antigen is more specific 
but less sensitive since it can react with antibodies 
against other Brucella species (e.g. B. canis, B. abortus, 
B. suis) (Hollett 2006). The disadvantage when using 
cytoplasmic‑derived antigen is the long period 
of time necessary for the test to become positive, 
which is around 8‑12 weeks after the exposure. This 
makes the test not indicated for revealing the early 
stages of the infection. However, the assay remains 
positive for a longer period, about 12 months after 
the cessation of bacteremia. So this test can be more 
useful in detecting chronic infections (Carmichael 
2012). In the literature, positive results to this test 
have been reported to persist more than 5 years 
after infection (Hollett 2016).

AGIDcpa test is the most effective technique in 
kennels infected with B.  canis. Provided that the 
definitive diagnosis of B. canis infection always 
requires confirmation by blood culture, AGIDcpa 
test it can be used as a confirmatory test for those 
sera resulting positive to the agglutination tests 
(Carmichael and Shin 1996, Cosford et al. 2018).

A limited number of veterinary diagnostic 
laboratories are capable of carrying out AGID (Hollett 
2006) due to the difficulties of antigen preparation 
and purification, as well as the requirement of 
specialized personnel (Feldman and Nelson 1996, 
Hollett 2006).

Indirect Immunofluorescence Assay (IFA)

This test can be considered as alternative when 
RSAT and TAT are not avalaible (Weber and Hussein 
1976). However, results from the Cornell University's 
Diagnostic Laboratory indicate a high rate of false 
positive reactions with the IFA test (Wanke 2004). 
and, since the sensitivity of the IFA is not yet fully 
known, there is also a probability that infected dogs 

RSAT test has been further modified by replacing the 
B. ovis antigen with an antigen derived from B. canis 
in phase M (mucoid) (Carmichael and Joubert 1988). 
This resulted in a reduction of the rate of false 
positivity to about 10%.

Therefore, since false negativity to this test is rare, 
it can be used as screening test to identify and 
separate negative subjects (Wanke 2004).

Tube agglutination test (TAT )

The technique has been described by Carmichael 
and Kenney (Carmichael and Kenney 1968) and 
Alton and colleagues (Alton et  al. 1975). The TAT is 
able to detect antibodies to B.  canis in dogs tested 
positive for RSAT or 2ME‑RSAT. The test begins to 
provide positive results already 2‑4 weeks after 
exposure (Hollett 2006). TAT consists in the addition 
of a fixed dose of heat inactivated B. canis antigen to 
different test serum dilutions. It is able to determine 
the antibody titer (Feldman and Nelson 1996). 
The test is sensitive but not very specific, allowing 
false positive results (Hollett 2006). As for RSAT, the 
addition of 2‑mercaptoethanol (2ME‑TAT) reduces 
false positive reactions (Carmichael 1990). Although 
more data on the reliability of this test are needed, 
the serum samples are considered negative when 
the agglutinating titer is less than 1:50 and doubtful 
when agglutinating titers are between 1:50 and 
1:200. Titers above 1:200 are considered as positive. 
However, blood culture is always required to confirm 
the infection (Fredrickson and Barton 1974, Rhoades 
and Mesfin 1980, Flores‑Castro and Carmichael 
1977, Henderson et  al. 1974, Carmichael and Shin 
1996). There is a good correlation between TAT titer 
≥  1:200 and the isolation of the microorganism by 
blood culture (Hollett 2006).

In the United States and in countries where the 
disease is present, 2ME‑TAT is no longer used in 
laboratories. They prefer to use 2ME‑RSAT which has 
the same diagnostic accuracy, is easier to perform, 
standardizable and capable of giving comparable 
results between laboratories (Carmichael and Shin 
1996). Being however a semi‑quantitative test, 
2ME‑TAT is still used in kennels where brucellosis, due 
to B.  canis infection, has been diagnosed. It allows 
to indirectly evaluate the response to antibiotic 
therapy, through the decrease of the agglutination 
antibody titer (Carmichael and Shin 1996), although 
this correlation has not yet been sufficiently 
demonstrated (Nicoletti and Chase 1987a).

Agar gel immunodiffusion test (AGID)

AGID can employ two different types of antigens: 
a B.  canis cell wall antigen (AGIDcwa) or antigenic 
proteins extracted from cytoplasm of B.  canis or 



14 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Table V. Comparison of ELISA tests for the diagnosis of canine 
brucellosis. From Cosford, edited (Cosford 2018).

Antigens Sensitivity Specificity
Lipopolysaccharides-free cytoplasmic 

proteins of B. abortus
92% 96.7%

Hot-saline extract of B. canis containing 
outer membrane antigens

92% 94.3%

Luminaze synthase of Brucella spp. 81% 96.7%

18kDa cytoplasmic protein of B. canis 87%a 98%b

Bacterial whole cell extract from wild 
isolate of B. canis used as solid phase 

antigen
95% 91%

Heat soluble bacterial extract from wild 
isolate B. canis antigenc

91.18% 100%

M-strain B. canis antigen 100% 98.8%

B. ovis strain #11 antigen 100% 98.8%

B. abortus RB51 strain antigen 100% 98.8%
aReal sensitivity not reported as percentage was calculated considering 26/30 known 
cases that tested positivie with this ELISA.
bSpecificity not reported as percentage was calculated from the data set as 2/103 
animal tested falsely positive with this ELISA in the healthy population.
cHeat soluble extract were more useful than ultrasonic homogenates of bacteria isolates 
to generate candidate capture antigens, a sonicated antigens were associated with more 
cross reactivity, and, therefore, false positives in both ELISA and Western Blotting. 

1975, Weber and Krauss 1977, Ebani et al. 2003). The 
antigen used by Ebani and colleagues (Ebani et  al. 
2003) was a HSE of B. ovis strain 63/290.

Immunoblotting (IB)

Ebani and colleagues (Ebani et  al. 2003) 
explored the performances of Sodium dodecyl 
sulfate‑polyacrilamide gel electrophoresis 
(SDS‑PAGE) using B. ovis and B. canis HSE. The results 
of the study suggest that immunoblotting may 
be the most specific serodiagnostic method for 
detecting antibodies to B. canis. The authors had no 
false positive reactions with any of the evaluated 
sera. Serum samples positive to AGID or CFT were 
all negative to immunoblotting test. Immunogenic 
bands were evidenced with both B. canis and B. ovis 
HSE antigens only when positive control sera were 
tested. On the basis of the results obtained, the 
authors recommended to use the immunoblotting 
test as a confirmatory test. However, because of the 
intensive labor and time required for running the 
test, IB is not performed in laboratories on a routine 
basis (Ebani et al. 2003).

Interpretation of diagnostic tests

The different serological tests available have 
different levels of sensitivity and specificity 
depending on the stage of the disease and the type 
of method and antigen used. Medical history and 
clinical data, where available, should always be used 
in conjunction with the laboratory results to achieve 
a definitive diagnosis (Wanke 2004).

False negative results may occur following sampling 
carried out prior to seroconversion or due to low 
titers of circulating antibodies in some chronically 
infected subjects (Carmichael and Greene 2006).

False positive results, on the other hand, are the 
major problem when using these serological tests. 
These drawbacks can depend on specific and 
non‑specific cross reactions with surface antigens 
of other microorganisms, such as Pseudomonas 
aeruginosa, Bordetella bronchiseptica, Actinobacillus 
equuli, Streptococcus spp., Staphylococcus spp., 
Moraxella‑type microorganisms and Gram‑negative 
bacteria (Carmichael and Greene 2006, Hollett 2006, 
CFSPH 2020, Yu and Nielsen 2010). These tests can be 
used as screening tests but positive results should be 
confirmed by a high‑specific confirmatory test such 
as 2ME‑RSAT or AGIDcpa (Carmichael and Greene 
2006, Hollett 2006, Keid et al. 2009, Cosford 2018).

It has been reported that agglutinating antibodies 
may not protect the dog from infection (Pollock 
1979) or from bacteriemia (Serikawa and Muraguchi 
1979, CDC 1977).

may not be detected when tested (Carmichael and 
Greene 2006).

Immunoenzymatic test (ELISA)

Data on the performance of the ELISA for the 
detection of B.  canis antibodies have recently been 
summarized by Cosford (Cosford 2018) (Table V).

These tests have been developed by using either 
cell wall of B. canis (M− and RM 6/66), or cytoplasm 
of B.  abortus, as antigens (Baldi et  al. 1997). The 
cytoplasmic antigens, common to all strains of the 
genus Brucella, have the advantage of not showing 
cross‑reactivity with bacteria belonging to genera 
other than Brucella spp. However, it cross reacts with 
all bacteria of the genus Brucella. ELISAs which use 
cell wall antigens of Brucella strains in phase M− are 
highly specific but not highly sensitive (Serikawa 
et  al. 1989, Mateau de Antonio et  al. 1993). High 
false positive rates were instead observed when 
the ELISAs which use the RM 6/66 strain as antigen, 
were employed (Mateau de Antonio et  al. 1993, 
Ebani et al. 2003).

Complement fixation test (CFT )

The Complement Fixation test was described by 
Alton and colleagues (Alton et  al. 1975), and Weber 
and Krauss (Weber and Krauss 1977). Altough 
showing a good correlation with TAT (Weber and 
Krauss 1977), CFT is not used on a routine basis since 
dog serum often is anticomplementary (Alton et al. 



15Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

seem to be desirable, since it would interfere with 
the serological tests carried out to identify infected 
subjects (Carmichael 1990).

Direct prophylaxis

Management of kennels free from B. canis 
and prevention of spreading between kennels

Since a valid immunological product is not available, 
the prevention of canine brucellosis due to B. canis, 
in territories where the disease is present, should 
be based on the classic sanitary measures of direct 
prophylaxis, based on regular serological testing of 
the animals hosted in the kennel.

Strict biosecurity measures should be implemented, 
together with adequate kennel management and 
strict environmental controls (cleaning, disinfection, 
ensuring adequate temperature and humidity). It is 
necessary to take special care to properly clean and 
disinfect on a daily basis the delivering sites and 
the spaces where the newborn puppies are housed 
(USDA 2015).

Quarantine

Kennels free from B.  canis should keep any newly 
introduced dog in quarantine, subjecting them 
to at least two specific serological tests carried out 
at four weeks of distance, admitting them to be in 
contact with the others only if both tests are negative 
(Carmichael 1996). The best way to keep brucellosis 
out of the kennel is then to isolate and test all 
incoming dogs, and testing them negative before 
placing into the kennel. This is best achieved by 
isolating newly purchased dogs in a separate building 
or facility, away from the rest of the population, for a 
minimum of eight weeks. All incoming dogs should 
be tested for B. canis on arrival and again after eight 
weeks. Only after having obtained two negative 
screening tests on all dogs of the isolation facility, 
they can be safely transferred in contact with the rest 
of the kennel population. If during this eight‑week 
isolation period a newly introduced and isolated 
dog results positive to B.  canis serology, it should 
be immediately removed from the facility. The 
eight‑week isolation period will then restart for the 
remaining dogs in isolation. The quarantine isolation 
approach combined with laboratory tests has been 
shown to be the safest way to introduce new dogs 
into an established reproductive population without 
the fear of introducing brucellosis or other infectious 
agents (USDA 2015).

Control of breeders

B.  canis infection is of major relevance for breeders 

Therapy
Treatment of canine brucellosis due to B.  canis 
is possible, although the results are often 
disappointing, because of the intracellular 
localization of the bacterium for long periods, and its 
ability to generate episodic bacteremia (Carmichael 
1990). For this reason, although in vitro B.  canis is 
sensitive to different antibiotics, often the therapy is 
not effective and relapses of infection are common.

Among the various proposed therapeutic protocols, 
the scheme which combines tetracycline (25 mg/Kg 
t.i.d., PO) for four weeks, and dihydrostreptomycin 
(10 mg/Kg b.i.d., IM) in the first and last week of 
therapy gave the best results (Carmichael 1996). This 
experimental therapeutic scheme led to serological 
negativization of 94% of treated animals within two 
months after treatment (Nicoletti 1991).

The therapy is less successful in males than in 
females, probably due to the greater difficulty 
encountered in eliminating foci of infection from 
the male genital tract, especially from the prostate 
(Carmichael 1990). Despite the treatment, male 
subjects can still develop irreversible sterility and 
this, along with the difficulty of eliminating the 
infection from the prostate, suggests that these 
subjects should be excluded from reproduction 
anyway (Carmichael 2012).

The variability of the results obtained by the 
various authors in the application of experimental 
therapeutic protocols (Flores‑Castro and Carmichael 
1981, Zoha and Walsh 1982, Nicoletti and Chase 
1987a, Nicoletti, 1991) also depends on the criterion 
adopted to define recovery (i.e. negativization 
vs bacteriological or serological examinations, 
respectively) (Carmichael and Shin 1996). In any 
case, in light of the fact that the infection can 
reoccur even in weeks or months after the end of 
therapy, it is recommended to carry out serological 
monitoring of the dogs for at least three months 
after completing the therapeutic cycle, repeating 
it in case serological positivity is found (Carmichael 
1996). In addition, sterilization of the subject is 
recommended (Carmichael 1990).

The clinician should inform the owner about the 
problems associated to the therapy, its costs, length, 
and possible failures (Feldman and Nelson 1996).

Prophylaxis

Indirect prophylaxis
As far as indirect prophylaxis is concerned, all 
attempts to produce an efficacious vacine have not 
provided encouraging results. On the other hand, 
the existence of a vaccine for some aspects does not 



16 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Isolation and disinfection at delivery 

It would also be appropriate to isolate females at the 
time of delivery, as well as the regular disinfection of 
kennel premises, particularly the delivering rooms 
(Berthelot and Garin‑Bastuji 1993).

Management of infected kennels

In case the infection is identified in a kennel, the 
entire existing population should be confined 
(prohibition of entry/exit of animals) and subjected 
to serological tests. All the dogs present should be 
considered suspected cases. 

Protection of operators

In order to reduce the likelihood of exposure to 
B.  canis, personnel who work in infected kennels 
should wear disposable protective gloves when 
assisting delivery, including handling newborn 
puppies, placenta, fetal membranes or possible 
contact with urine or vaginal secretions. Extreme 
care should be taken when handling miscarriage, 
including dead or partially developed puppies, 
their fetal membranes and placentas. Protective 
gloves should also be used during assistance for 
insemination, both natural and artificial. To prevent 
B.  canis infection during cleaning and disinfection 
of premises or handling of animals in quarantine 
and isolation situations, appropriate use of personal 
protective equipment (i.e. respiratory and ocular 
protection) would be advisable in addition to 
the simple use of gloves. It is also recommended 
that veterinarians, staff, owners, and laboratory 
personnel be careful when collecting and handling 
blood, serum, fluids or tissues for laboratory 
testing (USDA 2015).

Census of animals present

Before starting the checks, a thorough survey of 
the correspondence of the kennel register and 
the animals really hosted should be carried out. 
Subsequently, a register of controls should be set 
up where animals are distinguished by sex, race, and 
age, taking care to specify for each one of them the 
date of birth, so that it would be possible to easily 
identify the subjects to be monitored.

Animal tracing

It is necessary to identify the source of entry of the 
infection and its possible origin. Similarly, all animals 
having left the infected premise should be tracked. 
The animals and the sheltering kennel traced in this 
way should be subject to the same provisions as 
those adopted in the premise of origin.

as it is normally a problem that forces to end their 
reproductive career (Carmichael 1990). In breeding 
kennels located in areas where the disease is 
endemic, annual serological testing of breeding 
dogs should be carried out, combined with a further 
check to be performed at least three weeks before 
each mating (Feldman and Nelson 1996).

Ideally, breeding dogs should never leave the 
breeding facilities except to be visited by a 
veterinarian for the necessary care (e.g. caesarean 
section, serious injuries or illnesses). It is advisable 
to keep the breeding dog population sheltered 
in the kennel, and to avoid sending females out 
for mating. These subjects could pose a risk of 
introduction of the disease; therefore, any dog that 
leaves the structure for mating or for any reason 
other than a caesarean section should be tested 90 
days after its return to confirm its negative status. 
It is best to isolate these subjects from the rest of 
the kennel population upon their return, although 
doing so may not be practical. It may in fact require 
isolation during gestation and delivery, which may 
be problematic in some facilities (USDA 2015).

An alternative approach to genetic improvement of 
the breeding kennel population could be the use 
of artificial insemination (AI) on breeding females, 
using semen obtained from external breeding dogs 
proven negative for B.  canis (at least eight weeks 
prior to semen collection). If male breeding dogs are 
subjected to outdoor mating, the safest approach 
would be to offer this service only through the 
use of AI, using semen collected in the kennel and 
then shipped to the requesting kennel without the 
female being introduced into the premises or the 
male leaving them (USDA 2015).

Control of other introductions

Stray dogs should be prevented from entering 
breeding kennels. Likewise, contact with animals or 
groups of animals of unknown or doubtful health 
status should be avoided, particularly in the case of 
competitions or mating (Carmichael 1996).

Visitors (including the customers themselves) should 
not have visited any other breeding kennel on the 
same day; they should wear clean clothing, disinfect 
their shoes, wear disposable protective shoe covers 
and wash their hands properly. Ideally, visitors 
should not touch or handle dogs or equipment. 
One solution could be to provide direct contactless 
access using a video display of the mother and father, 
as well as the litter during the period of parental 
care. This would eliminate most of the direct risk of 
transmission of the disease until the puppy is moved 
to its new home (USDA 2015).



17Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

of all dogs present, for whatever reason, until the 
kennel reaches the B. canis‑free status.

Depending on the epidemiological situation of 
the kennel and on the problems that may be 
present in relation to the welfare of the hosted 
dogs, official veterinarians should verify whether a 
suspicious or infected kennel should be subject to 
a strict quarantine (which means a ban on the sale 
or transfer of any subject until the achievement of 
the B. canis‑free status) or to a partial quarantine, in 
which it is forbidden to sell or transfer only positive 
animals and puppies born to positive mothers.

The private veterinarians responsible for the kennel 
should closely link up with the respective official 
veterinarians to agree on how to manage quarantine 
in relation to the suitability of the facilities, periodic 
serological checks, the removal and isolation of 
positive animals and compliance with the rules on 
animal welfare.

Destruction of fetal membranes and aborted 
fetuses

Females should deliver in separate rooms that are 
properly washable and disinfectable. The placenta 
and fetal membranes should be removed and 
destroyed. Similarly, aborted fetuses or puppies 
that die before weaning should be removed and 
destroyed.

Cleaning and disinfection

B.  canis does not survive for long periods in the 
environment, and it is normally sensitive to common 
disinfectants, such as ammonium quaternary salts 
and iodophores (Carmichael 1990) or to direct 
sunlight (USDA 2015). Brucella is also sensitive to 
1%, sodium hypochlorite, to 70% ethanol, alcohol/
iodine solutions, glutaraldehyde and formaldehyde 
(Hollett 2006). In the presence of organic debris, 
B. canis remains stable in the environment for up to 
two months (proper cleaning and disinfection are 
therefore essential). B.  canis is resistant to drying 
in the presence of organic debris. It can withstand 
freezing and can survive in water, dust and soil. The 
combination of organic debris, high humidity, low 
temperatures and little or no sunlight promotes 
the organism’s survival (this corresponds in most 
kennels to winter conditions).

An important and often neglected part of kennels 
management is the correct cleaning and disinfection 
of environments. When dealing with brucellosis or 
other diseases, cleaning and disinfection serve to 
limit the spread of infections and are fundamental 
components for their prevention. Proper cleaning 
and disinfection needs time and should be carried 

Controls on animals

Once brucellosis is diagnosed, the only way to 
regain the freedom from the disease is to remove all 
positive dogs from the kennel.

All dogs over six weeks of age should be serologically 
tested with a first screening test (e.g. RSAT, 2ME‑RSAT 
or TAT). Dogs to be tested must not have received 
any antibiotic treatment in the three months prior to 
sampling (Hollett et al. 2006). Dogs tested negative 
to the first screening test should undergo a second 
screening test to be carried out at least 4 weeks later. 
Dogs that tested negative to this second test can be 
considered as not infected with B. canis.

Dogs that test positive to the first serological test 
should be classified as suspected of B. canis infection 
(USDA 2015). They should be isolated and subjected 
to a confirmation test (e.g. AGIDcwa) to be carried 
out not earlier than 4 weeks. The dogs should remain 
isolated until the second response is obtained. If the 
second test is also positive, these dogs should be 
considered suspected of infection. They should be 
removed from the colony. To confirm or definitively 
rule out brucellosis, blood culture (Feldman and 
Nelson 1996), or a third diagnostic test (e.g. AGIDcpa) 
could be performed eight weeks after the second 
test (USDA 2015).

Puppies born to positive mothers or in any case 
younger than 6 weeks at the time of control should 
undergo three blood cultures carried out every 
24  hours. Puppies with positive blood culture 
should be removed from the plant. Puppies over six 
weeks at the time of control must follow the same 
diagnostic protocol as adults (Hollett 2006).

In order to reach the status of kennel free from 
B.  canis, the tests should be repeated with the 
remaining dogs every four weeks, until there are two 
consecutive negative tests on the entire population 
of the kennel (USDA 2015, Carmichael 1996).

Isolation of positive animals – ban on mating 
and sale

Dogs infected or suspected of infection should 
be kept physically isolated from negative dogs. 
However, this may not be feasible or sufficient to 
prevent the spread of the disease, even if strict 
hygiene measures are maintained (Carmichael and 
Joubert 1988). For this reason, it is recommended 
to remove from the plant all animals infected 
(Wanke 2004).

In kennels infected or suspected to be infected with 
B. canis, it is recommended to put in place isolation 
procedures for all positive dogs, and to stop any 
trade or exchange of positive subjects or puppies 
born to positive mothers. In addition to this, it is 
strongly recommended to stop all sales or transfer 



18 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Because of the zoonotic potential of Brucella, dogs 
that have been confirmed positive for brucellosis 
should not be given for adoption.

The decision regarding the possible adoption of 
B.  canis positive dogs can only be taken with the 
authorization of the competent health authorities 
both in the areas of origin and destination (USDA 
2015). If a decision to authorize the adoption of 
B.  canis positive dogs is taken, they should be 
subject to ovarian‑hysterectomy or castration, 
as well as to appropriate long‑term antibiotic 
therapy, with appropriate supervision by an official 
veterinarian. This should include periodic lifetime 
laboratory tests for B.  canis. Canine brucellosis is 
considered a potentially long‑life infection; even 
after undergoing surgical spaying and long‑term 
use of antibiotics, both male and female dogs can 
continue to eliminate the bacterium intermittently. 
New owners should be made aware of the potential 
risk these dogs may pose over the course of their 
lives regarding infection of humans, other dogs and 
other susceptible animal species they could come in 
contact (USDA 2015). 

Training and information for operators

Official or private veterinarians managing dog 
breeding kennels infected from B.  canis should 
discuss in depth with the staff and the owners of 
the facility the potential for legal liability (beyond 
damaging the reputation of the kennel) that would 
inevitably accompany cases of zoonosis from B. canis 
as a result from the sale of infected puppies or 
adult dogs. These puppies or adult dogs commonly 
come in contact with children, the elderly or other 
who may be immunocompromised (USDA 2015). 
A recent example was recorded in New York City 
in 2012 involving a 3‑year‑old girl. It was the first 
documented case of transmission of B.  canis from 
a puppy to a child in the United States (Dentinger 
et al. 2015).

Public health aspects
B.  canis can cause disease in humans, which can 
acquire infection through direct contact with 
infected dogs, their reproductive secretions, or their 
blood (Lawaczeck et al. 2011, Lucero et al. 2010).

Sources of infection
A potential source of spread of B. canis are breeding 
kennels, both for the nature of the disease, and for 
the fact that the animals are housed in close contact 
with each other, and for the constant movement of 
dogs for reproduction or for sale (Brower et al. 2007). 
Recent outbreaks in kennels in the USA, Hungary, 

out correctly in order for a kennel to be considered 
truly disinfected. It is important to remember that a 
clean kennel is not always a disinfected kennel. 

The kennel consultant veterinarian should be sure 
that operators thoroughly understand the entire 
cleaning and disinfection process, including the 
correct dilution and storage of detergents and 
disinfectants, as well as the fact that compliance 
with the expected contact times and proper rinsing 
are absolutely necessary. The frequency required for 
cleaning and disinfection (daily or weekly) should 
be thoroughly discussed with the operators. The 
structure should always be cleaned and disinfected 
following an order of susceptibility of animals to 
the disease, starting first of all from the areas of the 
kennel that host the most sensitive animals (puppies 
and lactating females), followed by the areas that 
house healthy adults and finally the areas that house 
animals in poor health or in isolation (USDA 2015).

Treatment and relocation of positive animals

In countries where the disease is present, although 
the infection does not expose the dog to a lethal 
risk, in the view that antibiotic therapy does not 
guarantee the bacteriological recovery of infected 
dogs and that they may represent a source of 
infection both for other dogs and for humans, 
euthanasia is recommended. Possibly an attempt of 
therapy may be reserved exclusively for high‑value 
breeding animals (Carmichael 1996). However, 
although cases of partially successful treatments 
have been reported, no treatment has been shown 
100% effective, and puppies born to mothers with 
chronic brucellosis, if they survive, are often infected. 
For this reason, it is essential that dogs positives to 
diagnostic tests are not maintained as breeders, 
even if they have a high genetic value (Wanke 2004).

If the owner does not intend to choose euthanasia, 
an attempt of therapy could be recommended 
by providing from the beginning the necessary 
information on the zoonotic potential of the disease, 
especially underlining that even a sterilized dog 
subjected to specific therapeutic treatment could 
still be a source of contagion for humans (Feldman 
and Nelson 1996).

It is important to emphasize that canine brucellosis 
due to B. canis is currently not considered a curable 
disease in dogs. Attempts at therapeutic treatment 
led to very disappointing results, with relapses 
commonly occurring. The attempt at treatment 
can also mask diagnostic test results, and has been 
shown to be an important contributing factor to the 
spread of the disease. The impact of this evidence for 
dog breeding kennel owners is that animals infected 
with B.  canis must be removed from the breeding 
population (USDA 2015).



19Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

people would have a higher risk of contracting the 
disease (Dentinger et  al. 2012, Marzetti et  al. 2013, 
Tosi and Nelson 1982, Lucero et al. 2010). Three cases 
have been reported in children under 4 years of age 
(Dentinger et  al. 2012, Marzetti et  al. 2013, Tosi and 
Nelson 1982). In one of the reports, Dentinger and 
colleagues (Dentinger et  al. 2012) described the 
transmission of B.  canis to a child by an infected 
puppy that had been purchased from a pet store and 
had been believed healthy during the preliminary 
veterinary visit. The child showed fever and B. canis 
infection was diagnosed with blood culture. The 
isolated strains from the child and the puppy 
were sent to the Centers for Disease Control and 
Prevention (CDC, USA) where the two strains showed 
a close genetic similarity, suggesting that the puppy 
had been the source of infection. However, clinical 
signs did not develop in four adults belonging to 
the same family, all exposed to the puppy. Several 
recent reports of B.  canis infection in patients with 
HIV infection also highlight this population as at risk 
(Lawaczeck et  al. 2011, Lucero et  al. 2010, Moreno 
et  al. 1998). These cases of B.  canis infection have 
been linked to the owning of non‑spayed dogs that 
had a history of reproductive failure and a following 
diagnosis of B. canis infection by serology and blood 
culture (Lawaczeck et al. 2011, Moreno et al. 1998).

Symptomatology
Symptoms of B.  canis infection in humans are 
generally similar to those of brucellosis caused by 
other Brucella species (e.g. B. abortus or B. melitensis) 
(USDA 2015). Symptoms are often non‑specific and 
may include one or more of the following: fever 
(often periodic and nocturnal), fatigue, headache, 
weakness, general malaise, nausea, chills, sweating, 
loss of weight, hepatomegaly, splenomegaly and 
lymphadenopathy (Swenson et  al. 1972, USDA 
2015). Endocarditis, meningitis, arthritis and visceral 
abscesses can represent further complications, 
however rare (Carmichael 2012).

Although there are reports in the literature that the 
course of the disease would still be less severe when 
compared to the infection caused by the ‘classical’ 
species of the genus Brucella (Swenson et  al. 1972, 
Polt et  al. 1982), however, severe manifestations 
have been also described. These include septic 
arthritis, aortic valve vegetation, osteomyelitis, 
epidural abscess, pleural effusion, oral lesions, lower 
limb aneurysms and culture‑negative endocarditis 
(NASPHV 2012).

Diagnosis
The diagnosis of infection in humans, as well as in 
dogs, is based on serological examination followed 
by blood culture. However, in humans, diagnosis 

Sweden and Colombia highlight the link between 
outbreaks and the inter‑regional or international 
movement of breeding dogs (Kaden et  al. 2014, 
Castrillón‑Salazar et  al. 2013, Gyuranecz et  al. 2011, 
Brower et  al. 2007). Uncontrolled handling of 
puppies or non‑spayed dogs is a known risk factor 
for the spread of infectious diseases, and this has 
led to human infection with B. canis (Dentinger et al. 
2015, Brower et al. 2007).

Compared to owned dogs, stray dogs are more likely 
to be not spayed, and may have a higher level of 
seropositivity to B.  canis (Flores‑Castro and Segura 
1976, Brown et al. 1976). A high incidence of canine 
brucellosis in stray dog populations may cause a 
spillover in the human population, especially in 
areas with a large number of non‑spayed stray dogs, 
as these dogs are brought into shelter kennels or 
placed in other types of premises awaiting adoption. 
In the United States, 30% of pet dogs are adopted 
from animal shelters, and testing for B.  canis is not 
a standard procedure before adoption (Brower et al. 
2007). However, there is no evidence of a direct link 
between the number of fertile stray dogs in an area 
and the potential for exposure to humans.

Another potential source of B.  canis infection may 
be represented by laboratory accidents. Brucella 
spp. is considered a high‑risk pathogen and it 
requires manipulation in a specialized laboratory 
at biosecurity level 3 (BSL 3), which if not used may 
result in laboratory acquired exposure (Yagupsky 
and Baron 2005). Dentinger and colleagues 
(Dentinger et  al. 2012) described an incident in 
which 31 laboratory technicians were exposed to 
B.  canis after handling an unknown gram‑negative 
bacterium on the bench. No one has fallen ill with 
clinical disease, even those classified as having had 
high‑risk exposures (according to CDC guidelines) 
or who have declined post‑exposure prophylaxis 
(5 out of 21 of those at high risk) (Dentinger et  al. 
2012). A case of laboratory‑acquired exposure 
has been documented in a technician who used 
oral pipetting to re‑suspend an M‑phase strain of 
B.  canis; the technician showed symptoms despite 
this particular strain being considered non‑virulent 
in dogs (Wallach et al. 2004).

Categories at risk
Laboratory staff, veterinarians and animal keepers 
are the categories at greater risk of exposure to 
B.  canis (Lucero et  al. 2010, Marzetti et  al. 2013, 
Krueger et al. 2014).

In addition to these, several reports in the literature 
highlight pet dog owners as possible categories at 
risk (Swenson et al. 1972, Munford et al. 1975, Lucero 
et  al. 2010, Dentinger et  al. 2012, Tosi and Nelson 
1982). In particular, children and immunosuppressed 



20 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Therapy
Unlike canine brucellosis, the disease in humans can 
be quickly and effectively treated with tetracycline 
therapy administered for two to three weeks 
(Carmichael 1990).

Prophylaxis
In countries where the disease is present, it is 
recommended that veterinarians always inform 
the owners of infected animals about the potential 
zoonotic risk of cohabitation with their pets and 
to use the outmost caution and hygiene when 
visiting dogs suspected of infection, especially the 
female dogs who have aborted in the recent past 
(Carmichael 2012).

Similarly, caution is recommended in the laboratory 
when handling samples to be subjected to 
diagnostic tests for B. canis (Carmichael 2012).

Acknowledgements
The authors would like to thank Giulia Colacicco for 
the English text revision.

is often complicated due to non‑specific signs and 
symptoms and it is therefore associated with a low 
rate of disease suspicion by many physicians. If the 
disease is placed in a differential diagnosis, blood 
culture is the only test available to confirm B.  canis 
infection in humans. However, confirmation is not 
straightforward, due to intermittent and low‑level 
bacteremia (Rumley and Chapman 1986). Regarding 
serology, human antibodies to B. canis react with the 
same antigens used in dog serological tests, while 
they do not react with the B. abortus‑derived antigens 
(Brucella abortus strain 99, Weybridge or B.  abortus 
strain 1119‑3, USDA), which are used in routine tests 
for the diagnosis of human brucellosis caused by 
smooth strains (Carmichael 1990). Therefore, even 
if the physician may suspect B.  canis infection on 
the basis of clinical findings, the diagnosis may not 
be supported by serological tests available on the 
market, as these are aimed at detecting antibodies 
produced against Brucella in a smooth phase and do 
not detect antibodies against B.  canis (Lucero et  al. 
2005). Serological tests for the detection of B.  canis 
infection developed on dogs have been adapted for 
use in humans, but test results should be interpreted 
with caution (Hensel et al. 2018).



21Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

Alton G.G., Jones L.M. & Pietz D.E. 1975. Laboratory 
techniques in brucellosis. 2nd Ed. World Health 
Organization, Geneva, Switzerland.

Alton G.G., Jones L.M., Angus R.D. & Verger J.M. 1988. 
Techniques for the brucellosis laboratory. I.N.R.A., 
Paris, 169‑174.

Badakhsh F.F., Carmichael L.E. & Douglas J.A. 1982. 
Improved rapid slide agglutination test for presumptive 
diagnosis of canine brucellosis. J Clin Microbiol, 
15 (2), 286‑289.

Baldi P.C., Wanke M.M., Loza M.E., Monachesi N. & Fossati 
C.A. 1997. Diagnosis of canine brucellosis by detection 
of serum antibodies against an 18 kDa cytoplasmic 
protein of Brucella spp. Vet Microbiol, 57, 273‑281.

Berthelot X. & Garin‑Bastuji B. 1993. Brucelloses canines. 
Point Vétérinaire, 25 (152), 125‑129.

Brower A., Okwumabua O., Massengill C., Muenks Q., 
Vanderloo P., Duster M., Homb K. & Kurth K. 2007. 
Investigation of the spread of Brucella canis via the U.S. 
interstate dog trade. Int J Infect Dis, 11, 454‑458. doi: 
10.1016/j.ijid.2006.12.009.

Boeri E.J., Wanke M.M., Madariaga M.J., Teijeiro M.L., 
Elena S.A. & Trangoni M.D. 2018. Comparison of four 
polymerase chain reaction assays for the detection 
of Brucella spp. in clinical samples from dogs. Vet 
World, 11, 201‑208. http://dx.doi.org/10.14202/
vetworld.2018.201‑208.

Brown J., Blue J.L., Wooley R.E. & Dreesen D.W. 1976. 
Brucella canis infectivity rates in stray and pet dog 
populations. Am J Public Health, 66, 889‑891. http://
dx.doi.org/10.2105/AJPH.66.9.889.

Buonavoglia D., Malena G., Mosca R., Benchetti G., Narcisi 
D. & Corsalini T. 1992. Ricerca di anticorpi per Brucella 
canis in sieri di cane. Atti SISVet, XLIV, 1097‑1101.

Carmichael L.E. 1990. Brucella canis. In Animal brucellosis 
(Nielsen K. & Duncan J.R., eds). CRC Press, Boca Raton, 
Florida, 335‑350.

Carmichael L.E. & Bruner D.W. 1968. Characteristics of a 
newly‑recognized species of Brucella responsible for 
infectious canine abortions. Cornell Vet, 58, 579‑592.

Carmichael L.E. & Joubert J.C. 1988. Transmission of Brucella 
canis by contact exposure. Cornell Vet, 78, 63‑73.

Carmichael L.E. & Kenney R.M. 1968. Canine abortion caused 
by Brucella canis. J Am Vet Med Assoc, 152, 605‑616.

Carmichael L.E. & Shin S.J. 1996. Canine brucellosis: a 
diagnostician’s dilemma. Semin Vet Med Surg (Small 
Anim), 11, 161‑165.

Carmichael L.E., Zoha S.J. & Flores‑Castro R. 1984. Problems 
in the serodiagnosis of canine brucellosis: dog 
responses to cell wall and internal antigens of Brucella 
canis. Dev Biol Stand, 56, 371‑383.

Carmichael L.E. 2012. Canine brucellosis. In Infectious 
diseases of the dog and cat (Greene C.E., ed) 4th Ed. 
London, Elsevier Health Sciences, 398‑411.

Carmichael L.E. & Bruner D.W. 1968. Characteristics of a 

References

newly recognized species of Brucella responsible for 
infectious canine abortion. Cornell Vet, 58, 579‑592.

Carmichael L.E. & Greene C.E. 2006. Canine brucellosis. 
In Infectious diseases of the dog and cat (Greene 
C.E., ed). 3rd ed. Philadelphia, Pennsylvania, Elsevier 
Saunders, 369‑381.

Casalinuovo F., Cacia A., Lopresti A., Miceli M. E., Viterbo A. 
& Fenizia D. 1996. La brucellosi in Calabria. Ricerca delle 
fonti di contagio. Vet Ital, 32 (20), 23‑28.

Castrillón‑Salazar L., Giraldo‑Echeverri C.A., 
Sánchez‑Jiménez M.M. & Olivera‑Angel M. 2013. 
Factors associated with Brucella canis seropositivity 
in kennels of two regions of Antioquia, Colombia. 
Cad Saude Publica, 29, 1955 1973. http://dx.doi.
org/10.1590/0102‑311X00133013.

Center for Disease Control (CDC). 1977. Brucellosis 
surveillance. Annual summary 1977. CDC, Atlanta.

Center for Food Security and Public Health (CFSPH). 
2018. The canine brucellosis: Brucella canis. Ames, 
Iowa. http://www.cfsph.iastate.edu/Factsheets/pdfs/
brucellosis_canis.pdf.

Ciuchini F., Sala V., Pistoia C., Piccininno G., Pievaroli A. 
& Fantini C. 1982. Brucellosi da Brucella canis: rilievi 
anticorpali su cani della città e del suburbio di Roma. 
Clinica Veterinaria, 105 (3), 138‑144.

Colella G., Corsalini T. & Goffredo G. 1980. Indagine 
sierologica sulla brucellosi del cane in Puglia e 
Basilicata. Atti SISVet, 34, 330‑332.

Corbel M.J. & Brinkley‑Morgan W.J. 1989. Genus Brucella. 
In Bergey's Manual of Systematic Bacteriology, 
vol.  1 (Kreig N.R. & Holt S.D., eds). Williams & Wilkins, 
Baltimora, 377‑388.

Corrente M., Franchini D., Decaro N., Greco G., D’Abramo 
M., Greco M.F., Latronico F., Crovace A. & Martella V. 
2010. Detection of Brucella canis in a dog in Italy. New 
Microbiologica, 33, 337‑341.

Cosford K.L. 2018. Brucella canis: an update on research 
and clinical management. Can Vet J, 59, 74‑81.

Dentinger C.M., Jacob K., Lee L.V., Mendez H.A., 
Chotikanatis K., McDonough P.L., Chico D.M., De B.K., 
Tiller R.V., Traxler R.M., Campagnolo E.R., Schmitt D., 
Guerra M.A. & Slavinski S.A. 2015. Human Brucella 
canis infection and subsequent laboratory exposures 
associated with a puppy, New York City, 2012. 
Zoonoses Public Health, 62, 407‑414. http://dx.doi.
org/10.1111/zph.12163.

Feldman E.C. & Nelson R.W. 1996. Brucellosi. In 
Endocrinologia e riproduzione del cane e del gatto. 
UTET, Torino, 731‑739.

Flores‑Castro R. & Carmichael L.E. 1977. Canine brucellosis: 
current status of methods for diagnosis and treatment. 
27th Gaines Veterinary Symposium, 17‑24.

Flores‑Castro R. & Carmichael L.E. 1981. Brucella 
canis infection in dogs. Rev Latinoamericana 
Microbiologia, 23, 75.

Flores‑Castro R. & Segura R. 1976. A serological and 



22 Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

Canine brucellosis due to Brucella canis  De Massis et al.

Keid L.B., Soares R.M., Vasconcellos S.A., Megid J., Salgado 
V.R., Megid J. & Richtzenhain L.J. 2010. Comparison of 
a PCR assay in whole blood and serum specimens for 
canine brucellosis diagnosis. Vet Rec, 167, 96‑99. http://
dx.doi.org/10.1136/vr.c3811.

Kerwin S.C., Lewis D.D., Hribernik T.N., Partington 
B., Hosgood G. & Eilts B.E. 1992. Diskospondylitis 
associated with Brucella canis infection in dogs: 14 
cases (1980‑1991). J Am Vet Med Assoc, 201, 1253‑1257.

Kornegay J.N. 1983. Discospondylitis. In Current 
veterinary therapy (Kirk R.W., ed). WB Saunders Co., 
Philadelphia, 718‑721.

Krueger W.S., Lucero N.E., Brower A., Heil G.L. & Gray G.C. 
2014. Evidence for unapparent Brucella canis infections 
among adults with occupational exposure to dogs. 
Zoonoses Public Health, 61, 509‑518. http://dx.doi.
org/10.1111/zph.12102.

Lawaczeck E., Toporek J., Cwikla J. & Mathison B.A. 2011. 
Brucella canis in a HIV‑infected patient. Zoonoses 
Public Health, 58, 150‑152. http://dx.doi.org/10.1111/
j.1863‑2378.2010.01334.x.

López‑Goñi I., García‑Yoldi D., Marín C.M., de Miguel 
M.J., Barquero‑Calvo E., Guzmán‑Verri C., Albert D. & 
Garin‑Bastuji B. 2011. New Bruce‑ladder multiplex 
PCR assay for the biovar typing of Brucella suis and the 
discrimination of Brucella suis and Brucella canis. Vet 
Microbiol, 154, 152‑155.

Lucero N.E., Corazza R., Almuzara M.N., Reynes E., Escobar 
G.I., Boeri E. & Ayala S.M. 2010. Human Brucella canis 
outbreak linked to infection in dogs. Epidemiol 
Infect, 138, 280‑285. http://dx.doi.org/10.1017/
S0950268809990525.

Lucero N.E., Escobar G.I., Ayala S.M. & Jacob N. 2005. 
Diagnosis of human brucellosis caused by Brucella 
canis. J Med Microbiol, 54, 457‑461. http://dx.doi.
org/10.1099/jmm.0.45927‑0.

Marzetti S., Carranza C., Roncallo M., Escobar G.I. & Lucero 
N.E. 2013. Recent trends in human Brucella canis 
infection. Comp Immunol Microbiol Infect Dis, 36, 55‑61. 
http://dx.doi.org/10.1016/j.cimid.2012.09.002.

Mateau de Antonio E.M., Martin M. & Soler M. 1993. Use 
of indirect enzyme‑linked immunosorbent assay with 
hot saline solution extracts of a variant (M−) strain of 
Brucella canis for diagnosis of brucellosis in dogs. Am J 
Vet Res, 54, 1043‑1046.

Mol J.P.S., Guedes A.C.B., Eckstein C., Quintal A.P.N., 
Souza T.D., Mathias L.A., Haddad J.P.A., Paixão 
T.A. & Santos R.L. 2020. Diagnosis of canine 
brucellosis: comparison of various serologic tests 
and PCR. J Vet Diagn Invest, 32, 77‑86. https://doi.
org/10.1177/10406387198910.

Moore J.A. 1969. Brucella canis infection in dogs. J Am Vet 
Med Assoc, 155, 2034‑2037.

Moreno S., Ariza J., Espinosa F.J., Podzamczer D., Miró 
J.M., Rivero A., Rodríguez‑Zapata M., Arrizabalaga J., 
Mateos R. & Herrero F. 1998. Brucellosis in patients 
infected with the human immunodeficiency virus. Eur 
J Clin Microbiol Infect Dis, 17, 319‑326. http://dx.doi.
org/10.1007/BF01709454.

Munford R.S., Weaver R.E., Patton C., Feeley J.C. & Feldman 

bacteriological survey of canine brucellosis in Mexico. 
Cornell Vet, 66, 347‑352.

Fredrickson L.E. & Barton C.E. 1974. A serologic survey for 
canine brucellosis in a metropolitan area. J Am Vet Med 
Assoc, 165, 987‑989.

Georgia Department of Agriculture (GDA). 2020. Canine 
brucellosis (Brucella canis). Atlanta, Georgia. http://agr.
georgia.gov/Data/Sites/1/media/ag_animalindustry/
animal_health/files/caninebrucellosis.pdf.

George L.W. & Carmichael L.E. 1974. A plate agglutination 
test for the rapid diagnosis of canine brucellosis. Am J 
Vet Res, 35 (7), 905‑909.

George L.W., Duncan J.R. & Carmichael L.E. 1979. Semen 
examination in dogs with canine brucellosis. Am J Vet 
Res, 40 (11), 1589‑1595.

Gyuranecz M., Szeredi L., Rónai Z., Dénes B., Dencso 
L., Dán Á., Pálmai N., Hauser Z., Lami E., Makrai L., 
Erdélyi K. & Jánosi S. 2011. Detection of Brucella 
canis‑induced reproductive diseases in a kennel. 
J Vet Diagn Invest, 23, 143‑147. http://dx.doi.
org/10.1177/104063871102300127.

Henderson R.A., Horlein B.F., Kramer T. & Meyer M.E. 1974. 
Discospondylitis in three dogs infected with Brucella 
canis. J Am Vet Med Ass, 165 (5), 451‑455.

Hensel M.E., Negron M. & Arenas‑Gamboa A.M. 2018. 
Brucellosis in dogs and Public Health risk. Emerg Infect 
Dis, 24, 8, 1401‑1406.

Hollett R.B. 2006. Canine brucellosis: outbreaks and 
compliance. Theriogenology, 663, 575‑587.

Hurov L., Troy G. & Turnwald G. 1978. Diskospondylitis in 
the dog: 27 cases. J Am Vet Med Assoc, 173, 275‑281.

Kaden R., Ågren J., Båverud V., Hallgren G., Ferrari S., 
Börjesson J., Lindberg M., Bäckman S. & Wahabbg T. 
2014. Brucellosis outbreak in a Swedish kennel in 2013: 
determination of genetic markers for source tracing. 
Vet Microbiol, 174, 523‑530. http://dx.doi.org/10.1016/j.
vetmic.2014.10.015.

Kang S.I., Lee S.E., Kim J.Y., Lee K., Kim J.W., Lee H.K., Sung S.R., 
Heo Y.R., Jung S.C. & Her M. 2014. A new Brucella canis 
species‑specific PCR assay for the diagnosis of canine 
brucellosis. Comp Immunol Microbiol Infect Dis, 37, 
237‑241. http://dx.doi.org/10.1016/j.cimid.2014.07.003.

Kauffman L.K., Bjork J.K., Gallup J.M., Boggiatto P.M., 
Bellaire B.H. & Petersen C.A. 2014. Early detection 
of Brucella canis via quantitative polymerase chain 
reaction analysis. Zoonoses Public Health, 61, 48‑54. 
http://dx.doi.org/10.1111/zph.12041.

Keid L.B., Soares R.M., Vasconcellos S.A., Megid J., Salgado 
V.R. & Richtzenhain L.J. 2009. Comparison of agar gel 
immunodiffusion test, rapid slide agglutination test, 
microbiological culture, and PCR for the diagnosis of 
canine brucellosis. Res Vet Sci, 86, 22‑26.

Keid L.B., Soares R.M., Vieira N.R. Megid J., Salgado V.R., 
Vasconcellos S.A.,da Costa M., Gregori F. & Richtzenhain 
L.J. 2007. Diagnosis of canine brucellosis: comparison 
between serological and microbiological tests and a 
PCR based on primers to 16S‑23S rDNA interspacer. Vet 
Res Commun, 31, 951‑965. http://dx.doi.org/10.1007/
s11259‑006‑0109‑6.



23Veterinaria Italiana 2022, 58 (1), 5‑23. doi: 10.12834/VetIt.2561.16874.1

De Massis et al.  Canine brucellosis due to Brucella canis

Tolari F. & Pizzirani S. 1978. Indagini preliminari sulla 
esistenza della brucellosi canina da Brucella canis in Italia. 
Annali Facoltà Medicina Veterinaria Pisa, 31, 107‑113.

Tosi M.F. & Nelson T.J. 1982. Brucella canis infection 
in a 17‑month‑old child successfully treated with 
moxalactam. J Pediatr, 101, 725‑727. http://dx.doi.
org/10.1016/S0022‑3476(82)80301‑6.

United States Department of Agriculture Animal and Plant 
Health Inspection Service (USDA). 2015. Best practices 
for Brucella canis prevention and control in dog breeding 
facilities. https://www.aphis.usda.gov/animal_welfare/
downloads/Brucella_canis_prevention.pdf.

Valente C., Gialletti L., Capua I. & Antognoni M.T. 
1991. Brucellosi del cane. Indagine sierologica e 
batteriologica per Brucella canis. Archivio Veterinario 
Italiano, 42, 3, 135‑139.

Vesco G., Caracappa V. & Guercio V. 1987. Indagine 
sierologica nei confronti di B. canis in cani di affezione 
nella Sicilia occidentale. Giornale di Malattie Infettive e 
Parassitarie, 39 (5), 450‑451.

Wallach J.C., Giambartolomei G.H., Baldi P.C. & Fossati 
C.A. 2004. Human infection with M‑ strain of Brucella 
canis. Emerg Infect Dis, 10, 146‑148. http://dx.doi.
org/10.3201/eid1001.020622.

Wanke M.M. 2004. Canine brucellosis. Animal Reprod Sci, 
82‑83, 195‑207.

Weber A. & Krauss H. 1977. Zum serologischen Nachweis 
von Brucella canis‑Infektionen bei Hunden mittels 
Komplementbindungsreaktion. Zentralbl Veterinarmed 
B, 24, 746‑752.

Yagupsky P. & Baron E.J. 2005. Laboratory exposures to 
Brucellae and implications for bioterrorism. Emerg 
Infect Dis, 11, 1180‑1185. http://dx.doi.org/10.3201/
eid1108.041197.

Yu W.L. & Nielsen K. 2010. Review of detection of 
Brucella  spp. by polymerase chain reaction. Croat Med 
J, 51, 306‑313.

Zoha S.J. & Carmichael L. E. 1982. Serological responses of 
dogs to cell wall and internal antigens of B. canis. Vet 
Microbiol, 28, 1‑4.

Zoha S.J. & Walsh R. 1982. Effect of a two‑stage antibiotic 
regimen in dogs naturally infected with Brucella canis. J 
Am Vet Med Ass, 180, 1474.

R.A. 1975. Human disease caused by Brucella canis. 
A clinical and epidemiologic study of two cases. 
JAMA, 231, 1267‑1269. http://dx.doi.org/10.1001/
jama.1975.03240240037023.

National Association of State Public Health Veterinarians 
(NASPHV). 2012. Public Health implications of Brucella 
canis infections in humans. http://www.nasphv.org/
Documents/BrucellaCanisInHumans.pdf.

Nicoletti P. 1991. Further studies on the use of antibiotics 
in canine brucellosis. Compend Contin Educ Pract Vet, 
13 (6), 944‑947.

Nicoletti P. & Chase A. 1987a. The use of antibiotics to 
control canine brucellosis. Compend Contin Educ Pract 
Vet, 9 (11), 1063‑1066.

Nicoletti P. & Chase A. 1987b. An evaluation of methods to 
diagnose B. canis in dogs. Compend Contin Educ Pract 
Vet, 9 (11), 1071‑1074.

Pollock R.V.H. 1979. Canine brucellosis: current status. 
Compend Contin Educ Pract Vet, 1, 255‑267.

Prosperi S., Ostanello F., Baldelli R. & Falcone A. 1994. 
Ruolo del cane nella epidemiologia della Brucellosi 
ovi‑caprina. Vet Ital, 30 (12), 2‑6.

Reike J.A. & Rhodes H.E. 1975. Brucella canis isolated from 
the eye of a dog. J Am Vet Med Ass, 165, 1734.

Rhoades H.E. & Mesfin G.M. 1980. Brucella canis infection in 
a kennel. Vet Med Small Anim Clin, 75, 595‑599.

Rumley R.L. & Chapman S.W. 1986. Brucella canis: an 
infectious cause of prolonged fever of undetermined 
origin. South Med J, 79, 626‑628. http://dx.doi.org/10.1
097/00007611‑198605000‑00027.

Saegusa J., Ueda K., Goto Y. & Fujiwara K. 1978. Ocular 
lesions in experimental canine brucellosis. Jap J Vet Sci, 
39, 181‑185.

Serikawa T. & Muraguchi T. 1979. Significance of urine 
in transmission of canine brucellosis. Jap J Vet Sci, 
41, 607‑616.

Smeak D.D., Olmstead M.L. & Horn R.B. 1987. Brucella canis 
osteomyelitis in two dogs with total hip replacement. J 
Am Vet Med Ass, 191 (8), 986‑990.

Swenson R.M., Carmichael L.E. & Cundy K.R. 1972. Human 
infection with Brucella canis. Ann Intern Med, 76, 435‑438. 
http://dx.doi.org/ 10.7326/0003‑4819‑76‑3‑435.