The Veterinary Journal 198 (2013) 504–507

Contents lists available at ScienceDirect

The Veterinary Journal

journal homepage: www.elsevier.com/locate/tvjl

Evaluation of an in-clinic assay for the diagnosis of canine parvovirus

N. Decaro a,⇑, C. Desario a, M. Billi b, E. Lorusso a, M.L. Colaianni a,c, V. Colao a, G. Elia a, G. Ventrella a,
I. Kusi d, S. Bo e, C. Buonavoglia a
a
Department of Veterinary Medicine, University of Bari, Valenzano (Bari), Italy
b
Zoetis Italia S.r.l., Milan, Italy
c
Istituto Zooprofilattico Sperimentale della Puglia e della Basilicata, Foggia, Italy
d
Faculty of Veterinary Medicine, Agricultural University, Tirana, Albania
e
Veterinari Associati, Torino, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The results of a study designed to evaluate the performance of an in-clinic test for the detection of canine
Accepted 28 August 2013 parvovirus (CPV) are reported. A total of 150 faecal samples collected from dogs with acute diarrhoea
were tested using the in-clinic test, a haemagglutination assay (HA) and a real-time PCR assay for CPV
detection, quantification and characterisation.
Keywords: CPV was detected in 66, 73, and 101 faecal samples by in-clinic, HA and PCR testing, respectively. The
Canine parvovirus relative sensitivity and specificity of the in-clinic test were 86.3% and 96.1%, respectively, when the test
Diagnosis
was compared to HA, and 65.3% and 100%, respectively, when compared to real-time PCR. The sample
In-clinic test
CPV-2c
distribution according to the virus type was CPV-2a, n = 44; CPV-2b, n = 11; CPV-2c, n = 44, CPV-2,
n = 2, as determined by minor groove binder probe assays and/or sequence analysis. The percentage of
positive in-clinic tests was 70.5% for CPV-2a, 72.7% for CPV-2b and 75.0% for CPV-2c (P > 0.05). Using
real-time PCR as the reference standard for CPV detection, the in-clinic test was more specific than HA
and had comparable sensitivity to HA, demonstrating detection rates similar to those previously observed
for other rapid in-clinic tests. The in-clinic test was also able to detect all CPV types at equivalent rates.
Ó 2013 Elsevier Ltd. All rights reserved.

Introduction
Canine parvovirus (CPV) is one of the most common causes of
acute gastroenteritis in young dogs. The virus is included in a
unique species, Feline panleukopenia virus, of the genus Parvovirus
(family Parvoviridae), along with feline parvovirus and other parv-
oviruses of domestic and wild carnivores (Decaro and Buonavoglia,
2012). The original strain CPV-2 emerged in late 1970s and was re-
placed by three antigenic variants, CPV-2a, CPV-2b and CPV-2c,
that arose consecutively. These variants are variously distributed
in the world and CPV-2c strains are becoming the predominant
variants in many countries (Decaro et al., 2006b, 2007, 2011; Hong
et al., 2007; Calderón et al., 2011).
Several methods are currently available for the diagnosis of CPV
infection, including in-clinic assays (Schmitz et al., 2009; Decaro
et al., 2010), haemagglutination (HA), virus isolation (Desario
et al., 2005), and molecular methods (Decaro et al., 2005b). DNA-
based assays are the most sensitive, representing the reference
standard. These assays are also able to characterise the viral strains
(Decaro et al., 2005a, 2006b) and discriminate between vaccine
⇑ Corresponding author. Tel.: +39 080 467 9832.
E-mail address: nicola.decaro@uniba.it (N. Decaro).
and field viruses (Decaro et al., 2006a, 2006c). In-clinic and other
laboratory tests are less sensitive than molecular assays, due to
the sequestration of CPV particles by gut antibodies (Desario
et al., 2005). Concerns have been expressed regarding the ability
of in-clinic tests to detect the new variant CPV-2c with the same
accuracy as CPV-2a/2b (Kapil et al., 2007). However, a recent study
showed that an in-clinic assay detected CPV-2c with the same
accuracy as other variants (Decaro et al., 2010).
The aim of this study was to compare the performance of an in-
clinic test for the detection of CPV (Witness Parvo test, Synbiotics
Corporation) with HA and a real-time PCR assay.
Materials and methods
Clinical specimens
A total of 150 faecal samples were collected from dogs with acute diarrhoea and
other clinical signs suggestive of CPV (leukopenia, vomiting). Most of the dogs were
recruited from animal shelters, breeding kennels, and academic or private veteri-
nary clinics in Italy (n = 137), but a few specimens also came from Spain (n = 3)
and Albania (n = 10), so that the test validation was not restricted to CPV strains
of Italian origin. Dogs were aged from 1 month to 14 years (mean ± standard devi-
ation [SD], 7.5 ± 19.2 months). Sixty-eight dogs (45.3%) were mixed-breed and the
purebred animals (82/150, 54.7%) included Labradors (n = 11), Beagles (n = 9),
Bernese mountain dogs (n = 8), German shepherds (n = 8), Yorkshire terriers

1090-0233/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tvjl.2013.08.032
 N. Decaro et al. / The Veterinary Journal 198 (2013) 504–507
(n = 8), Chihuahuas (n = 7), American cocker spaniels (n = 6), Boxers (n = 6),
Dachshunds (n = 5) and dogs of other breeds (n = 14). Twelve specimens (8%) were
from dogs that had been recently vaccinated against CPV (<10 days before sample
collection).
Each sample was submitted masked for in-clinic, haemagglutination and real-
time PCR testing.
In-clinic test
Faecal material was collected on a test kit swab following the manufacturer’s
instructions. The extraction buffer/conjugate was dispensed into the sample tube
via the kit swab. The sample swab was then inserted into the tube containing the
liquid and placed in a vortex. The extracted faecal material/conjugate liquid was
transferred to the in-clinic test device using the swab pipette for the test kit as
per manufacturer’s instructions.
Haemagglutination (HA) assay
Twofold dilutions of the supernatant of each faecal homogenate were made in
phosphate buffered saline (pH 7.2) starting from a 1:2 dilution (Desario et al., 2005).
Tests were carried out in 96-well V-plates (50 lL of sample dilution per well); equal
amounts of a suspension containing 0.8% pig erythrocytes and 1% fetal calf serum
(FCS) were added to each dilution. Results were read after 4 h at 4 °C and expressed
as the reciprocal of the highest sample dilutions able to produce HA.
Real-time PCR assays for CPV detection, quantification and characterisation
Specimens were homogenized (10% w/v) in Dulbecco’s modified Eagle’s
medium and subsequently clarified by centrifuging at 2500 g for 10 min. Viral
DNA was extracted from the supernatants of faecal homogenates by boiling for
10 min and chilling on ice. To reduce residual inhibitors of DNA polymerase activity
to ineffective concentrations, the DNA extract was diluted 1:10 in distilled water
(Decaro et al., 2005b, 2006b). CPV DNA was detected by real-time PCR using a
conventional TaqMan probe (Decaro et al., 2005b), whereas virus characterisation
was obtained by a panel of minor groove binder (MGB) probe assays able to predict
the viral type (Decaro et al., 2005a, 2006b) and to discriminate between vaccine and
field strains of CPV (Decaro et al., 2006a, 2006c).
Sequence analysis
The CPV strains that were not characterised by the MGB probe assays were sub-
mitted for sequence analysis of the VP2 protein gene, as previously described
(Decaro et al., 2008, 2009). Briefly, sequence reactions were carried out using Big-
Dye 3.1 Ready reaction mix (Applied Biosystems) according to the manufacturer’s
instructions. The sequenced products were separated with a 3130 Genetic Analyzer
(Applied Biosystems). Sequences were imported and assembled with Bionumerics
5.0 software (Applied Maths). The VP2 sequences obtained were compared to those
of reference strains retrieved from the GenBank database.
Data analysis
Data were analysed using a web-based software program (R version 2.8.1).1 An
experiment critical value alpha of 0.05 was used to set statistical significance. Sensi-
tivity and specificity were calculated as follows (Altman and Bland, 1994):
Sensitivity ¼ true positives  ðtrue positives þ false negativesÞ
Specificity ¼ true negatives  ðtrue negatives þ false positivesÞ
The proportion of positive samples by viral load (DNA copies/mg of faeces) and
CPV type was evaluated using a Pearson’s Chi-squared test. Student’s t test was
used to compare the average viral loads associated with the CPV types.
Results
Performance of the in-clinic test
The results of the three tests used for CPV detection in the fae-
ces of dogs with clinical signs of parvovirus are summarised in
Fig. 1. The in-clinic test was able to detect CPV antigen in 66/150
(44%) of analysed samples. Using HA, 73 (48.7%) samples tested po-
sitive and 77 (51.3%) were negative. The highest detection rates
(101/150, 67.3%) were obtained using real-time PCR. The relative
sensitivity and specificity of the in-clinic test were 86.3% (95%
1
See: http://www.r-project.org/ (accessed 25 August 2013).
505
(a) (b)
Fig. 1. Comparison of the in-clinic test (Witness Parvo test, Synbiotics Corporation)
with haemagglutination (a) and real-time PCR (b). Numbers indicate the samples
positive (+) or negative () for canine parvovirus (CPV). Results obtained by
comparing different techniques are bolded. The overall agreement, sensitivity and
specificity of the in-clinic testing compared with the other assays were calculated
and are indicated.
confidence interval [CI] 76.6–92.4%) and 96.1% (CI 89.2–98.7%),
respectively, when the test was compared to HA (Fig. 1a), and
65.3% (CI 55.7–73.4%) and 100% (CI 92.7–100%), respectively, when
compared to real-time PCR (Fig. 1b).
CPV type distribution, detection rates using the in-clinic test and viral
loads
Using MGB probe assays, the CPV strains detected by real-time
PCR were characterised as CPV-2a (n = 44), CPV-2b (n = 11), CPV-2c
(n = 36) or CPV-2 (n = 2), and eight strains were not typed. When
sequence analysis of the VP2 protein gene was performed, the
uncharacterised strains were confirmed as CPV-2c, since they dis-
played the non-coding mutation A4061G in the CPV-2c-probe
binding region (Decaro et al., 2009, 2013). Thus, combining the re-
sults of MGB probe assays and sequence analysis, the sample dis-
tribution according to virus type was CPV-2a, n = 44 (43.6%);
CPV-2b, n = 11 (10.9%); CPV-2c, n = 44 (43.6%); CPV-2, n = 2 (1.9%).
Both CPV-2 strains were detected in dogs that had been recently
given CPV-2 based vaccines. None of the detected CPV-2b strains
were characterised as vaccine virus. Additionally, 10 dogs that
had been vaccinated against CPV less than 10 days before sample
collection tested negative using in-clinic and HA testing. However,
5/10 dogs tested CPV-positive using real-time PCR, being infected
with CPV-2c (n = 2), CPV-2a (n = 2), or CPV-2b (n = 1).
The in-clinic assay was able to detect 31/44 CPV-2a, 8/11
CPV-2b and 27/44 CPV-2c. Consequently, the percentage of posi-
tive in-clinic test results was 70.5%, 72.7% and 75.0% for CPV types
2a, 2b and 2c, respectively. Detection rates for the three CPV types
were not different when compared statistically (P > 0.05). Neither
of the samples containing the original CPV-2 strain tested positive
using the in-clinic assay, which was expected on the basis of the
low viral titres detected in these specimens.
CPV DNA titres calculated by the TaqMan assays ranged from
1.09  103 to 1.03  1010 DNA copies/mg faeces. The average viral
loads in the clinical specimens were 1.11  109 (SD 1.86  109),
1.01  109 (SD 9.94  108), 1.39  109 (SD 2.77  109), and
2.51  103 (SD 2.00  103) for CPV types 2a, 2b, 2c, and 2, respec-
tively. Viral load was not significantly different for CPV-2c type
when compared to CPV-2a (P > 0.05) and CPV-2b (P > 0.05). The
in-clinic test gave positive results for specimens containing more
than 106 DNA copies/mg faeces (Fig. 2).
Comparison of in-clinic test results and viral loads showed that
the minimal CPV titre required for a positive result was about
106 DNA copies/mg faeces. Specimens with viral loads higher than
506 N. Decaro et al. / The Veterinary Journal 198 (2013) 504–507
Fig. 2. Distribution of the in-clinic test (Witness Parvo test, Synbiotics Corporation)
results by viral load (DNA copies/mg of faeces) and canine parvovirus (CPV) strain.
109 DNA copies/mg faeces were generally detected by the in-clinic
assay. One sample gave a negative in-clinic test result although it
contained 1.49  109 DNA copies/mg faeces (Fig. 2).
Discussion
In-clinic tests are rapid assays suitable for the diagnosis of CPV
in veterinary practices, and are the only tests available for a rapid
diagnosis under field conditions. In this study, we evaluated the
performance of an in-clinic test by comparing it with HA and
real-time PCR assays. The sensitivity of the in-clinic test was
65.3% compared to real-time PCR, which is considered to be the
reference standard (Desario et al., 2005).
There have been some doubts raised regarding the ability of in-
clinic rapid tests to recognise the new variant, CPV-2c. Those con-
cerns were based on circumstantial evidence that the increase in
rapid test failure paralleled the emergence and spread of CPV-2c
(Kapil et al., 2007). At that time, it was recommended that the
monoclonal antibodies contained in the in-clinic assays should
be tested against the additional mutations detected in strains cur-
rently circulating (Hong et al., 2007). In the present study, speci-
mens representative of the three antigenic variants, including
44 samples containing CPV-2c, were tested. The detection rate of
CPV-2c was not different to the other two CPV types, thus discred-
iting any concerns regarding the ability of the in-clinic test to de-
tect the new variant. This finding confirmed previously published
studies of a different in-clinic test used for CPV diagnosis (Decaro
et al., 2010; Markovich et al., 2012).
As previously observed, the in-clinic test was able to detect CPV
in samples containing viral loads higher than 106 DNA copies/mg
faeces (Decaro et al., 2010), but several high-titre specimens gave
negative results (Fig. 2). However, the viral loads reported here
are estimations only, since the number of CPV DNA copies does
not correlate directly with the number of viral particles. It is also
possible that infected cells could contain relatively more viral
nucleic acid than virions, which are the targets of in-clinic tests.
Previous studies have shown that in-clinic assays and more tra-
ditional diagnostic methods commonly used for CPV detection,
including HA and virus isolation, are less sensitive than PCR-based
methods, but are highly specific and unlikely to detect modified
live vaccine virus (Desario et al., 2005; Schultz et al., 2008; Schmitz
et al., 2009; Decaro et al., 2010). The lower sensitivity of traditional
methods is associated with the very small amounts of virus shed in
the faeces during the late stage of infection and/or the presence of
high CPV antibody titres in the gut lumen, which could sequestrate
most viral particles (Desario et al., 2005).
Although PCR-based methods are generally not affected by
the host immune response, they can be time-consuming,
labour-intensive, and they need the expertise of specialist labora-
tory technicians. In addition, the in-clinic test did not detect CPV
in either of the samples that contained CPV-2, probably because
of the low vaccine viral loads shed by vaccinated dogs (Desario
et al., 2005).
Conclusions
This study demonstrated that an in-clinic test was able to detect
the new CPV variant CPV-2c at the same rate as other CPV variants.
However, faecal samples that are CPV-negative on in-clinic assays
should be tested using PCR-based methods, especially when there
is a strong clinical suspicion of CPV.
Conflict of interest statement
This study was funded by grants from Zoetis. M. Billi is em-
ployed by Zoetis. None of the other authors of this paper has a
financial or personal relationship with other people of organisa-
tions that could inappropriately influence or bias the content of
the paper.
References
Altman, D.G., Bland, J.M., 1994. Diagnostic tests. 1: Sensitivity and specificity.
British Medical Journal 308, 1552.
Calderón, M.G., Romanutti, C., D’Antuono, A., Keller, L., Mattion, N., La Torre, J., 2011.
Evolution of canine parvovirus in Argentina between years 2003 and 2010:
CPV2c has become the predominant variant affecting the domestic dog
population. Virus Research 157, 106–110.
Decaro, N., Elia, G., Campolo, M., Desario, C., Lucente, M.S., Bellacicco, A.L.,
Buonavoglia, C., 2005a. New approaches for the molecular characterization of
canine parvovirus type 2 strains. Journal of Veterinary Medicine B, Infectious
Diseases and Veterinary Public Health 52, 316–319.
Decaro, N., Elia, G., Martella, V., Desario, C., Campolo, M., Di Trani, L., Tarsitano, E.,
Tempesta, M., Buonavoglia, C., 2005b. A real-time PCR assay for rapid detection
and quantitation of canine parvovirus type 2 DNA in the feces of dogs.
Veterinary Microbiology 105, 19–28.
Decaro, N., Elia, G., Desario, C., Roperto, S., Martella, V., Campolo, M., Lorusso, A.,
Cavalli, A., Buonavoglia, C., 2006a. A minor groove binder probe real-time PCR
assay for discrimination between type 2-based vaccines and field strains of
canine parvovirus. Journal of Virological Methods 136, 65–70.
Decaro, N., Elia, G., Martella, V., Campolo, M., Desario, C., Camero, M., Cirone, F.,
Lorusso, E., Lucente, M.S., Narcisi, D., et al., 2006b. Characterisation of the canine
parvovirus type 2 variants using minor groove binder probe technology. Journal
of Virological Methods 133, 92–99.
Decaro, N., Martella, V., Elia, G., Desario, C., Campolo, M., Buonavoglia, D., Bellacicco,
A.L., Tempesta, M., Buonavoglia, C., 2006c. Diagnostic tools based on minor
groove binder probe technology for rapid identification of vaccinal and field
strains of canine parvovirus type 2b. Journal of Virological Methods 138, 10–16.
Decaro, N., Desario, C., Addie, D.D., Martella, V., Vieira, M.J., Elia, G., Zicola, A., Davis,
C., Thompson, G., Thiry, E., et al., 2007. Molecular epidemiology of canine
parvovirus, Europe. Emerging Infectious Diseases 13, 1222–1224.
Decaro, N., Desario, C., Miccolupo, A., Campolo, M., Parisi, A., Martella, V., Amorisco,
F., Lucente, M.S., Lavazza, A., Buonavoglia, C., 2008. Genetic analysis of feline
panleukopenia viruses from cats with gastroenteritis. Journal of General
Virology 89, 2290–2298.
Decaro, N., Desario, C., Parisi, A., Martella, V., Lorusso, A., Miccolupo, A., Mari, V.,
Colaianni, M.L., Cavalli, A., Di Trani, L., et al., 2009. Genetic analysis of canine
parvovirus type 2c. Virology 385, 5–10.
Decaro, N., Desario, C., Beall, M.J., Cavalli, A., Campolo, M., Dimarco, A.A., Amorisco,
F., Colaianni, M.L., Buonavoglia, C., 2010. Detection of canine parvovirus type 2c
by a commercially available in-house rapid test. The Veterinary Journal 184,
373–375.
Decaro, N., Desario, C., Billi, M., Mari, V., Elia, G., Cavalli, A., Martella, V., Buonavoglia,
C., 2011. Western European epidemiological survey for parvovirus and
coronavirus infections in dogs. The Veterinary Journal 187, 195–199.
Decaro, N., Buonavoglia, C., 2012. Canine parvovirus – A review of epidemiological
and diagnostic aspects, with emphasis on type 2c. Veterinary Microbiology 155,
1–12.
Decaro, N., Desario, C., Amorisco, F., Losurdo, M., Elia, G., Parisi, A., Ventrella, G.,
Martella, V., Buonavoglia, C., 2013. Detection of a canine parvovirus type 2c
with a non-coding mutation and its implications for molecular characterisation.
The Veterinary Journal 196, 555–557.
Desario, C., Decaro, N., Campolo, M., Cavalli, A., Cirone, F., Elia, G., Martella, V.,
Lorusso, E., Camero, M., Buonavoglia, C., 2005. Canine parvovirus infection:
Which diagnostic test for virus? Journal of Virological Methods 121, 179–185.
 N. Decaro et al. / The Veterinary Journal 198 (2013) 504–507
Hong, C., Decaro, N., Desario, C., Tanner, P., Pardo, M.C., Sanchez, S., Buonavoglia, C.,
Saliki, J.T., 2007. Occurrence of canine parvovirus type 2c in the United States.
Journal of Veterinary Diagnostic Investigation 19, 535–539.
Kapil, S., Cooper, E., Lamm, C., Murray, B., Rezabek, G., Johnston 3rd, L., Campbell, G.,
Johnson, B., 2007. Canine parvovirus types 2c and 2b circulating in North
American dogs in 2006 and 2007. Journal of Clinical Microbiology 45, 4044–
4047.
Markovich, J.E., Stucker, K.M., Carr, A.H., Harbison, C.E., Scarlett, J.M., Parrish, C.R.,
2012. Effects of canine parvovirus strain variations on diagnostic test results
507
and clinical management of enteritis in dogs. Journal of American Veterinary
Medical Association 241, 66–72.
Schmitz, S., Coenen, C., König, M., Thiel, H.J., Neiger, R., 2009. Comparison of three
rapid commercial canine parvovirus antigen detection tests with electron
microscopy and polymerase chain reaction. Journal of Veterinary Diagnostic
Investigation 21, 344–345.
Schultz, R.D., Larson, L.J., Lorentzen, L.P., 2008. Effects of modified live canine
parvovirus vaccine on the SNAP ELISA antigen assay. Journal of Veterinary
Emergency and Critical Care 18, 415.