Content uploaded by Dorothee Bienzle
Author content
All content in this area was uploaded by Dorothee Bienzle
Content may be subject to copyright.
644 CVJ / VOL 50 / JUNE 2009
Article
Seroprevalence of feline leukemia virus and feline immunodeficiency
virus infection among cats in Canada
Susan Little, William Sears, Jessica Lachtara, Dorothee Bienzle
Abstract — The purposes of this study were to determine the seroprevalence of feline leukemia virus (FeLV) and
feline immunodeficiency virus (FIV) infection among cats in Canada and to identify risk factors for seropositivity.
Signalment, lifestyle factors, and test results for FeLV antigen and FIV antibody were analyzed for 11 144 cats
from the 10 Canadian provinces. Seroprevalence for FIV antibody was 4.3% and seroprevalence for FeLV antigen
was 3.4%. Fifty-eight cats (0.5%) were seropositive for both viruses. Seroprevalence varied geographically. Factors
such as age, gender, health status, and lifestyle were significantly associated with risk of FeLV and FIV seropositiv-
ity. The results suggest that cats in Canada are at risk of retrovirus infection and support current recommendations
that the retrovirus status of all cats should be known.
Résumé — Séroprévalence de l’infection par le virus de la leucémie féline et le virus de l’immunodéficience
féline chez les chats au Canada. Les buts de cette étude étaient de déterminer la séroprévalence du virus de la
leucémie féline (FeLV) et du virus de l’immunodéficience féline (FIV) chez les chats au Canada et d’identifier les
facteurs de séropositivité. Le signalement, les facteurs de style de vie et les résultats de tests pour l’antigène du FeLV
et les anticorps du FIV ont été analysés pour 11 144 chats provenant de 10 provinces canadiennes. La séroprévalence
pour l’anticorps du FIV était de 4,3 % et la séroprévalence pour l’antigène du FeLV était de 3,4 %. Cinquante-huit
chats (0,5 %) étaient séropositifs pour les deux virus. La séroprévalence variait selon la géographie. Les facteurs
comme l’âge, le sexe, l’état de santé et le style de vie étaient significativement associés au risque de séropositivité
pour le FeLV et le FIV. Les résultats suggèrent que les chats au Canada sont à risque d’infection par les rétrovirus
et appuient les recommandations actuelles que le statut rétroviral de tous les chats devrait être connu.
(Traduit par Isabelle Vallières)
Can Vet J 2009;50:644–648
Introduction
Feline leukemia virus (FeLV) and feline immunodeficiency
virus (FIV) are retroviruses that represent 2 of the most
common and important infectious diseases of cats worldwide.
The American Association of Feline Practitioners (AAFP)
recommends that the retrovirus status of all cats should be
known and has published guidelines for retrovirus testing and
management (1).
Several studies have evaluated the seroprevalence of FeLV
and FIV infection in North American cats, but most of the
available data is for the United States. An earlier study evalu-
ated FIV seroprevalence in 2254 high-risk cats and 511 cats of
low or unknown risk from the United States and Canada (2).
Of the high-risk cats, 42 were reported to be from Canada, but
no specific location was given. Feline immunodeficiency virus
seroprevalence was 19% (8/42) in this group of cats. Of the low
or unknown risk cats, 352 were reported to be from Canada
with no specific location indicated. The FIV seroprevalence
was reported for the group as a whole (1.2%, 6/511) with no
breakdown between cats from Canada versus those from the
United States.
A recent study of 18 038 cats tested at North American veteri-
nary clinics and animal shelters found 2.3% of cats seropositive
for FeLV antigen and 2.5% seropositive for FIV antibody (3).
Data on 325 cats from 7 unidentified Canadian provinces
were included, with 2.5% of cats being seropositive for FeLV
antigen and 3.1% being seropositive for FIV antibody. Three
other studies have evaluated seroprevalence of FeLV and FIV in
specific populations of Canadian cats. In 1 study, seroprevalence
was evaluated in 246 cats from 3 demographic populations in
Ottawa, Ontario (4). Seroprevalence for FIV antibody was
highest in urban stray cats (23%; 17/74) and lower in client-
owned cats (5.9%; 9/152) and in a feral cat colony (5%; 1/20).
Bytown Cat Hospital, 422 McArthur Ave., Ottawa, Ontario
K1K 1G6 (Little); Department of Pathobiology (Bienzle) and
Department of Population Medicine (Sears), University of
Guelph, Guelph, Ontario N1G 2W1; IDEXX Laboratories,
1 IDEXX Drive, Westbrook, Maine 04092, USA (Lachtara).
Address all correspondence to Dr. Susan Little; e-mail:
catvet@vin.com
Use of this article is limited to a single copy for personal study.
Anyone interested in obtaining reprints should contact the
CVMA office (hbroughton@cvma-acmv.org) for additional
copies or permission to use this material elsewhere.
CVJ / VOL 50 / JUNE 2009 645
ARTICLE
Seroprevalence for FeLV antigen was highest in urban stray cats
(6.7%; 5/74) and lower in client-owned cats (2.6%; 4/152).
No cats in the feral colony were seropositive for FeLV antigen.
Two male cats (0.8%; 2/246) were seropositive for both FeLV
and FIV infection.
In a study of a trap, neuter, and release program for feral cats
on Prince Edward Island, 7.6% (14/185) of cats were seroposi-
tive for FIV antibody and 6.5% (12/185) were seropositive for
FeLV antigen (5). Three male cats (1.6%) were seropositive for
both FeLV and FIV. The seroprevalence of FIV was also evalu-
ated in submissions of feline sera to a diagnostic laboratory in
Atlantic Canada (6). Although 671 samples were tested, 90.5%
were from Prince Edward Island. Seroprevalence of FIV antibody
was 7.6%. The study identified an increasing risk of FIV sero-
positivity with increasing age, and found the prevalence of
FIV antibodies was significantly higher in intact male cats than
in other gender categories.
More information is needed on the seroprevalence of FeLV
and FIV infection in cats in Canada to better define prophy-
lactic, management, and therapeutic measures for owned and
shelter cats. The purposes of this study were to determine
seroprevalence of FeLV antigen and FIV antibody among cats
from all 10 Canadian provinces and to identify risk factors for
seropositivity.
Materials and methods
Study participants and population
Veterinary clinics, animal shelters, cat rescue programs, and
feral cat programs in Canada were invited to participate in the
study. Potential study participants were identified as all those
who had purchased combination FeLV antigen and FIV anti-
body test kits or submitted samples for retrovirus testing to a
diagnostic laboratory (Vita-Tech, Markham, Ontario; Montreal,
Quebec; Central Laboratory for Veterinarians, Calgary, Alberta;
Edmonton, Alberta; Langley, British Columbia) in the previ-
ous 12 mo. Potential participants were sent a letter explaining
the study and inviting them to participate. An incentive was
offered to participants who submitted a minimum of 15 com-
plete test results before the end of the study period. Enrolled
participants were asked to submit results of tests for FeLV and
FIV performed between August 1, 2007 and November 15,
2007. Participants were requested to offer retrovirus testing to
the owners of all cats and kittens in accordance with testing
guidelines developed by the AAFP. The AAFP guidelines were
provided to study participants.
Data collection
Information requested on each cat included test date, patient
name, whether the test was performed in a clinic or at a referral
Table 1. Study population characteristics
Samples FIV1 test results FeLV1 test results
Factor Category Number Proportion Number Prevalence Number Prevalence
Test site Shelter/Rescue 1556 13.96 100 6.43 42 2.70
Veterinary clinic 9588 86.04 380 3.96 341 3.56
Outdoor access Yes 6532 58.61 339 5.19 258 3.95
No 3607 32.37 69 1.91 84 2.33
Unknown 1005 9.02 72 7.16 41 4.08
Age Juvenile 4030 36.16 63 1.56 68 1.69
Adult 7114 63.84 417 5.86 315 4.43
Gender Spayed female 2423 21.74 58 2.39 100 4.13
Intact female 2715 24.36 59 2.17 73 2.69
Castrated male 3093 27.75 177 5.72 112 3.62
Intact male 2873 25.79 183 6.37 95 3.31
Unknown 40 0.36 3 7.50 3 7.50
FIV test result Positive 480 4.31 NA NA 58 12.08
Negative 10 664 95.69 NA NA 325 3.05
FeLV test result Positive 383 3.44 58 15.14 NA NA
Negative 10 761 96.56 422 3.92 NA NA
Health status Healthy 7200 64.61 232 3.22 143 1.99
Sick 3432 30.80 229 6.67 227 6.61
Unknown 512 4.59 19 3.71 13 2.54
Province Alberta 829 7.44 37 4.46 21 2.53
British Columbia 1388 12.46 39 2.81 31 2.23
Manitoba 656 5.89 19 2.90 39 5.95
New Brunswick 120 1.08 1 0.83 0 0.00
Newfoundland 300 2.69 15 5.00 13 4.33
Nova Scotia 603 5.41 15 2.49 40 6.63
Ontario 5613 50.37 216 3.85 144 2.57
Prince Edward Island 187 1.68 7 3.74 8 4.28
Quebec 1266 11.36 118 9.32 83 6.56
Saskatchewan 182 1.63 13 7.14 4 2.20
NA — not applicable.
646 CVJ / VOL 50 / JUNE 2009
ARTICLE
laboratory, age, gender, access to outdoors, and whether the cat
was currently ill. Test results and information were submitted
to the investigators by fax transmission using a standard report-
ing form.
Testing protocol
Cats were tested for FeLV antigen and FIV antibody using a
commercially available point-of-care ELISA (SNAP Combo
FeLV antigen/FIV antibody, IDEXX Laboratories, Westbrook,
Maine, USA), or samples were submitted for enzyme-linked
immunosorbent assay (ELISA) testing (PetChek FIV Antibody,
PetChek FeLV Antigen; IDEXX Laboratories) to a diagnostic
laboratory. The assay procedure was included in the package
insert for each point-of-care test kit. Tests were performed on
blood, plasma, or serum. Confirmatory testing was not per-
formed as part of the study.
Risk factors
To evaluate regional variations in seroprevalence, data were
grouped by the province of origin. Other risk factors that were
evaluated included age [juvenile (, 6 months old) versus adult],
gender (intact female, spayed female, intact male, castrated
male), and whether cats had access to the outdoors. General
health status of the cat at the time of testing was recorded as
presence or absence of current illness.
Statistical analysis
Seroprevalence was determined by the percentage of cats with
positive test results. Unadjusted seroprevalence estimates of
FeLV infection, FIV infection, and FeLV/FIV co-infection were
calculated for the study population as a whole and for each of
the 10 provinces. A generalized linear mixed model was fitted
(SAS 9.1.3, Proc GLIMMIX; SAS, Cary, North Carolina,
USA) to the binary outcome variables FIV or FeLV seropositive
result. Effectively, a modified “logistic regression” model was
established, with nominal explanatory variables and random
effects. Practices, nested within shelter type and province, were
treated as random effects (variance component). Fixed effects
included the kind of test site (shelter or veterinary clinic),
outdoor access, age category (juvenile or adult), health status,
gender and province. Only practices with at least 1 case were
included because practices with no cases are not informative
with respect to the relation between any of the fixed effects and
the outcome. A model was fitted to examine the main features
of relationships. Cats used in the analyses only appeared once.
Mean probabilities and odds ratios with approximate 95% confi-
dence intervals (CI) were calculated for each effect. Significance
was set at P # 0.05.
Results
A total of 343 veterinary clinics and 13 animal shelters or rescue
organizations representing the 10 Canadian provinces partici-
pated in the study. Complete FeLV and FIV test results were
received for 11 144 cats. More cats were tested by veterinary
clinics (n = 9588) than by animal shelters/rescue organizations
(n = 1556). Results are summarized in Tables 1 and 2.
Of the 11 144 cats with complete test data, 480 (4.31%) were
seropositive for FIV antibody and 383 (3.44%) were seropositive
for FeLV antigen. Fifty-eight cats (0.52%) were seropositive for
both viruses. The probability of a positive test for FIV was not
significantly higher for cats tested at shelters or rescue programs
than at veterinary clinics. The probability of a positive test result
for FeLV for cats tested at veterinary clinics was 1.73 higher
(P = 0.023) than for cats tested at shelters or rescue programs.
Table 2. Relationship of FIV and FeLV-seropositive test results with risk variables
FIV FeLV
Probability of Probability of
Factor Category positive test (%) 95% CI P-value positive test (%) 95% CI P-value
Test site Shelter/Rescue 3.35 1.56–7.08 0.619 5.39 3.07–9.30 0.023
Veterinary clinic 2.83 1.70–4.68 9.17 6.27–13.22
Outdoor access Yes 5.00 2.88–8.54 , 0.001 8.19 5.37–12.31 0.027
No 1.46 0.75–2.81 5.85 3.63–9.30
Unknown 3.98 1.10–7.76 7.30 4.38–11.93
Age Juvenile 1.51 0.78–2.90 , 0.001 5.08 6.54–14.17 , 0.001
Adult 6.20 3.71–10.19 9.70 3.13–8.14
Gender Spayed female 1.40 0.78–2.49 0.541 5.52 3.61–8.35 0.485
Intact female 2.10 1.15–3.81 7.29 4.86–10.80
Castrated male 3.01 1.77–5.05 , 0.001 4.35 2.87–6.54 0.017
Intact male 7.43 4.58–11.81 7.08 4.84–10.26
Unknown 4.13 0.95–16.26 13.62 4.03–37.22
FIV test result Positive NA NA NA 10.97 6.85–17.14 , 0.001
Negative NA NA 4.46 2.97–6.64
FeLV test result Positive 4.88 2.67–8.77 , 0.001 NA NA NA
Negative 1.93 1.11–3.33 NA NA NA
Health Healthy 2.21 1.24–3.92 , 0.001 4.26 2.75–6.54 , 0.001
Sick 4.72 2.63–8.36 12.72 8.35–18.90
Unknown 2.79 1.23–6.21 6.31 3.25–11.89
NA — not applicable.
CVJ / VOL 50 / JUNE 2009 647
ARTICLE
There were some statistically significant regional differences in
seroprevalence for both FeLV and FIV. Seroprevalence for FeLV
infection was significantly higher in Quebec (6.56%) than in
British Columbia (2.23%, P = 0.0236) and Ontario (2.57%,
P = 0.0143). Seroprevalence for FIV infection was significantly
(P = 0.0481) higher in Quebec (9.32%) than in Nova Scotia
(2.49%).
Several factors were significantly associated with risk of FeLV
and FIV seropositivity. Adult cats were significantly more likely
(P , 0.001) to be seropositive (4.43% FeLV, 5.86% FIV) than
juvenile cats (1.69% FeLV, 1.56% FIV). Cats with current ill-
ness were significantly more likely (P , 0.001) to be seroposi-
tive (6.61% FeLV, 6.67% FIV) than healthy cats (1.99% FeLV,
3.22% FIV). The probability of a positive test result for FIV
was 3.43 higher (P , 0.001) in cats with access to outdoors
when compared to cats with no outdoor access. The probability
of a positive test result for FeLV was 1.43 higher (P = 0.027)
in cats with access to outdoors when compared with cats which
had no outdoor access. The probability of a positive test result
for FIV was highest in intact males (7.43%), and for FeLV was
highest in intact females (7.29%) and intact males (7.08%).
The probability of a positive test result for FIV or FeLV was
2.53 higher in samples seropositive for the other retrovirus than
in seronegative samples (P , 0.001).
Discussion
The AAFP recommends that cats be tested for both FeLV and
FIV infection under various circumstances, such as when first
acquired, when exposed to a known retrovirus-infected cat,
and before initial vaccination against FeLV or FIV (1). Sick
cats should be tested even if previously tested negative. Cats at
ongoing risk of infection should be tested annually.
To the authors’ knowledge, this is the first broadly based
study of retrovirus seroprevalence in Canada. It is not possible
to compare the results from the current study with the previous
studies of FeLV and FIV infection in Canada due to differences
in the study populations, selection biases, test methodologies,
and study design. In addition, while prevalence rates may
change over time, trends should be monitored by repeating
similar surveys.
The seroprevalence for both FeLV and FIV in this study was
higher in cats with access to outdoors compared with cats which
had no outdoor access, in intact cats compared with neutered
cats, in adult cats compared with juvenile cats, and in sick cats
compared with healthy cats. These findings are in accordance
with similar North American retrovirus seroprevalence stud-
ies (3,7). Bite wounds due to territorial or sexual aggression
are an efficient mode of retrovirus transmission. Bite wounds
are more common in intact cats than neutered cats, as well as
in cats with access to the outdoors. Cats with no access to out-
doors are less likely to have contact with seropositive cats than
are cats kept totally indoors. Neutered cats are also more likely
to be kept indoors as companion animals and are at lower risk
for retrovirus exposure. Both FeLV and FIV cause immunosup-
pression and are associated with many disease manifestations in
cats (1), and are therefore expected to be more prevalent in sick
than healthy cats.
The seroprevalence of FIV (4.3%) and FeLV (3.4%) in this
study was higher than in a recent similar study of samples
from predominantly US cats (3). In the US study, the overall
prevalence of FIV was 2.5% (409/18 038) and FeLV was 2.3%
(446/18 038). Samples from 325 Canadian cats were included in
the study, with a prevalence of 3.1% for FIV and 2.5% for FeLV.
The prevalence of co-infected cats was similar in both studies.
The difference in seroprevalence between the studies may be
influenced by the recruitment methods used for potential par-
ticipants. In the US study, participants were recruited from a
wider pool, including members of the AAFP and lists of animal
shelters, cat rescue programs, and trap-neuter-return programs
derived from Internet directories. In the current study, recruit-
ment was focused on those centres already known to employ
retrovirus testing as a matter of convenience for developing a
list of potential participants that would cover private veterinary
practices, humane organizations, animal shelters, and cat rescue
programs.
The manufacturer reports point-of-care assay sensitivities
for FeLV antigen and FIV antibody of 97.6% and 100%
respectively, and specificities of 99.1% and 99.5% respectively
(Package insert, SNAP Combo FeLV antigen/FIV antibody,
IDEXX Laboratories). An independent study found 100%
sensitivity and specificity for FIV antibody when unvaccinated
cats were tested (8). Positive results for FeLV antigen obtained
by ELISA testing should be confirmed with a secondary test
such as an immunofluorescent antibody test or a polymerase
chain reaction (PCR) assay and positive results for FIV antibody
obtained by ELISA testing should be confirmed with a western
blot assay (1). Investigation of the sensitivity and specificity
of PCR assays for FIV offered by some laboratories has shown
widely variable results (9,10), precluding recommendation of
this test methodology. Positive test results were not confirmed
using an alternate assay in the present study, so it is possible
that false positive test results were included in the analysis.
Feline immunodeficiency virus antibodies may be detected in
uninfected cats that have been vaccinated against FIV (8) and
in kittens with passively acquired immunity from an infected
or vaccinated queen (11). An inactivated dual-subtype vaccine
against FIV became available in 2003 in Canada (Fel-O-Vax
FIV; Fort Dodge Animal Health, Overland Park, Kansas, USA),
but FIV vaccination status was not recorded or known for cats
in the present study. It is presumed that veterinarians would
be unlikely to test cats for FIV if they were known to be vac-
cinated, and the population of cats tested by animal shelters/
rescue organizations is unlikely to have a high rate of FIV vac-
cination. Therefore, it seems likely that bias of FIV prevalence
estimates caused by vaccination was minimal. After exposure
to FIV, seroconversion may take 60 d or longer (12), and
similarly, after exposure to FeLV, seroconversion with detection
by soluble antigen tests may take 30 d (13) so that some false
negative results may also have been included for both viruses.
False negative test results may also occur if the concentration
of FeLV antigen or FIV antibody is below the detection limit
of the test, or if the test is performed incorrectly.
Unadjusted seroprevalence data should be interpreted with
caution because veterinarians and organizations may have
648 CVJ / VOL 50 / JUNE 2009
ARTICLE
varying opinions about the importance of testing, and about
criteria for selection of cats for testing despite the existence
of testing guidelines. Although veterinarians may recommend
testing for all cats, it is the individual cat owner who makes the
final decision. It is reasonable to consider that sick cats are more
likely to be tested than healthy cats as part of a diagnostic plan.
Seroprevalence for both FeLV and FIV infection was higher
in sick cats than healthy cats in the present study, similar to
findings in other studies, but it is possible that these rates are
artificially increased by selection bias.
Risk factor associations for FeLV and FIV seroprevalence
were similar to those previously identified, but should also be
interpreted with caution because cats and study participants
were not selected in a random manner. For example, owned cats
that never receive veterinary care or unowned cats that do not
enter a shelter or come under the care of a rescue program were
not sampled as part of this study. This limits the extent to which
the findings can be generalized to the Canadian cat population.
However, analysis identified age, gender, health status, and life-
style as significant risk factors for FeLV and FIV seropositivity
in agreement with other studies, including a recently published
study of North American cats (3). This information can be used
to counsel owners on prevention of disease transmission, such
as by limiting access to outdoors and neutering.
Although testing for FeLV and FIV has been readily avail-
able for many years in Canada, and vaccines against FeLV have
been in widespread use for more than 15 y, this study indicates
that retroviral infections remain common in Canada. Increased
awareness of Canadian seroprevalence data among veterinarians,
animal shelters, rescue organizations, and pet owners may help
improve testing and vaccination rates. Currently available guide-
lines for feline retrovirus testing and management developed by
the AAFP should be adopted in Canada. CVJ
References
1. Levy J, Crawford C, Hartmann K, et al. 2008 American Association of
Feline Practitioners’ feline retrovirus management guidelines. J Feline
Med Surg 2008;10:300–316.
2. Yamamoto JK, Hansen H, Ho EW, et al. Epidemiologic and clinical
aspects of feline immunodeficiency virus infection in cats from the
continental United States and Canada and possible mode of transmis-
sion. J Am Vet Med Assoc 1989;194:213–20.
3. Levy JK, Scott HM, Lachtara JL, Crawford PC. Seroprevalence of feline
leukemia virus and feline immunodeficiency virus infection among cats
in North America and risk factors for seropositivity. J Am Vet Med Assoc
2006;228:371–6.
4. Little SE. Feline immunodeficiency virus testing in stray, feral, and
client-owned cats of Ottawa. Can Vet J 2005;46:898–901.
5. Gibson K, Keizer K, Golding C. A trap, neuter, and release program for
feral cats on Prince Edward Island. Can Vet J 2002;43:695–698.
6. Hitt M, Spangler L, McCarville C. Prevalence of feline immunodefi-
ciency virus in submissions of feline serum to a diagnostic laboratory
in Atlantic Canada. Can Vet J 1992;33:723–726.
7. O’Connor TP, Jr., Tonelli QJ, Scarlett JM. Report of the National FeLV/
FIV Awareness Project. J Am Vet Med Assoc 1991;199:1348–1353.
8. Levy JK, Crawford PC, Slater MR. Effect of vaccination against feline
immunodeficiency virus on results of serologic testing in cats. J Am Vet
Med Assoc 2004;225:1558–1561.
9. Bienzle D, Reggeti F, Wen X, Little S, Hobson J, Kruth S. The variabil-
ity of serological and molecular diagnosis of feline immunodeficiency
virus infection. Can Vet J 2004;45:753–757.
10. Crawford PC, Slater MR, Levy JK. Accuracy of polymerase chain reac-
tion assays for diagnosis of feline immunodeficiency virus infection in
cats. J Am Vet Med Assoc 2005;226:1503–1507.
11. MacDonald K, Levy JK, Tucker SJ, Crawford PC. Effects of passive
transfer of immunity on results of diagnostic tests for antibodies against
feline immunodeficiency virus in kittens born to vaccinated queens.
J Am Vet Med Assoc 2004;225:1554–1557.
12. Barr MC. FIV, FeLV, and FIPV: Interpretation and misinterpretation
of serological test results. Semin Vet Med Surg (Small Anim) 1996;
11:144–153.
13. Jarrett O, Golder MC, Stewart MF. Detection of transient and persistent
feline leukaemia virus infections. Vet Rec 1982;110:225–228.
