Logo of canvetjReference to the Publisher site.Journal Web siteJournal Web siteHow to Submit
Can Vet J. 2004 Nov; 45(11): 924–930.
PMCID: PMC545982

Language: English | French

Impact of Giardia vaccination on asymptomatic Giardia infections in dogs at a research facility


Feces were collected from 107 asymptomatic dogs at a research facility in Guelph, Ontario. The prevalence of Giardia infection was 11% (12/107). To assess the effectiveness of Giardia vaccination for treatment of Giardia carriage, 9 additional asymptomatic Giardia antigen-positive dogs were brought into the facility. The Giardia antigen-positive dogs were then randomly allocated to receive either vaccine (n = 10) or a saline placebo (n = 10). Feces were then monitored monthly for 6 mo for Giardia antigen and Giardia cysts. At weeks 4, 8, 12, and 16 following vaccination, there were more Giardia-positive dogs in the vaccinated group (10/10, 9/10, 9/10, 8/10, respectively) compared with the controls (7/10, 7/10, 8/10, 4/10, respectively). At week 20, an equal number of dogs (5/10) were Giardia positive, and at week 24, fewer dogs were positive in the vaccinated group than in the control group (2/10 versus 5/10, respectively). However, there was no significant difference between the 2 groups. Vaccination was, therefore, not an effective treatment for asymptomatic canine Giardia infections in this setting.


RésuméImpact de la vaccination contre la giardiose sur les infections asymptomatiques à Giardia chez des chiens hébergés dans un établissement de recherche. Des fèces ont été recueillies chez 107 chiens asymptomatiques gardés dans des installations de recherche à Guelph en Ontario. La prévalence de l’infection à Giardia était de 11 % (12/107). Pour vérifier l’efficacité de la vaccination contre la giardiose dans le traitement du microbisme latent, 9 chiens additionnels, asymptomatiques mais positifs aux antigènes de Giardia, ont été introduits dans les locaux. Les chiens positifs aux antigènes de Giardia ont été répartis au hasard en deux groupes recevant soit le vaccin (n = 10), soit le placebo salin (n = 10). Les fèces ont été analysées mensuellement pendant 6 mois afin de détecter les antigènes et les kystes de Giardia. Aux semaines 4, 8, 12 et 16 suivant la vaccination, ily avait plus de chiens Giardia positifs dans le groupe vacciné (10/10, 9/10, 9/10, 8/10, respectivement) que dans le groupe témoin (7/10, 7/10, 8/10, 4/10, respectivement). À la semaine 20, un nombre égal de chiens (5/10) étaient Giardia positifs et à la semaine 24, moins de chiens étaient positifs dans le groupe vacciné en comparaison avec le groupe témoin (2/10 versus 5/10, respectivement). Cependant, il n’y avait pas de différence significative entre les 2 groupes. Dans le cas présent, la vaccination ne constituait donc pas un traitement efficace contre les infections asymptomatiques à Giardia.

(Traduit par Docteur André Blouin)


Giardia is an enteric protozoan parasite that is endemic throughout the world and is commonly encountered in dogs and cats (1). The parasite is known to infect a broad range of hosts, including mammals, birds, amphibians, and reptiles, and is the most frequently identified intestinal parasite of humans in North America (2).

The prevalence of Giardia infection in dogs from around the world has been reported to range from 1% to 39% (1). This variation may be due to geographical location, differences in the definition of the source population, differences in the methods used to determine prevalence, inappropriate methods of fecal analysis, difficulty in identifying Giardia cysts, and the intermittent shedding of the parasite (3). In Canada, the prevalence of Giardia in companion dogs was estimated recently to be 7.2%, based on fecal samples collected from 1216 dogs at 15 veterinary practices, with animals under 1 y of age constituting approximately 73% of the cases (3). Higher levels of infection (up to 100%) have been reported in densely housed animals, such as those in kennels, breeding operations, and research facilities (35).

Giardia infections in dogs are associated with acute or chronic diarrhea; weight loss; poor weight gain, despite a normal appetite; and, less commonly, vomiting and lethargy (1,6). However, the majority of infected animals show no clinical signs (7). In the aforementioned study of dogs seen at Canadian veterinary practices, 78% of infected dogs were asymptomatic (3). Asymptomatic carriers of Giardia, therefore, function as a reservoir and may be a potential source of infection for humans and other animals.

In humans, Giardia infections are associated with a wide spectrum of clinical signs that can range from asymptomatic to severe gastrointestinal and allergic disease (8,9). Some of the more commonly reported clinical signs include diarrhea, abdominal cramps, bloating, flatulence, fatigue, and weight loss resulting from malabsorption (10).

Traditionally, zinc sulphate fecal flotation has been used for the microscopic identification of Giardia cysts in the feces of infected animals. However, the sensitivity of a single test in dogs is only approximately 70%, due to the intermittent shedding of cysts. When performed on 3 fecal samples collected at 2- to 3-day intervals, the sensitivity of this test has been shown to increase to 96% (5,11). A number of commercial enzyme-linked immunosorbent assays (ELISA) have recently been developed that allow for the detection of Giardia-specific antigen in feces. One is an immunoassay (ProSpecT Giardia Microplate Assay; Alexon-Trend, Ramsey, Minnesota, USA) that detects a specific glycoprotein (GSA65) produced by dividing Giardia trophozoites. According to the manufacturer, this assay makes it possible to detect Giardia infection, even when trophozoites and cysts are not abundant in the feces. Previous studies in dogs have reported that this ELISA has a sensitivity and specificity similar to that obtained with zinc sulphate fecal flotation (1,5).

Current treatments for canine Giardia infections include metronidazole, fenbendazole, and febantel. Metronidazole has a treatment success rate of approximately 67% in dogs. Metronidazole-resistant strains of Giardia have been isolated from humans (12,13), and drug resistance has also been suggested as a reason for treatment failure in dogs (14,15). Fenbendazole has fewer side effects than metronidazole and 2 studies have reported success rates of 90% to 100% in small numbers of animals (7,16). However, some cases of Giardiainfected dogs have failed to respond to treatment with both metronidazole and fenbendazole (17). Albendazole has been shown to be effective in the treatment of Giardia in dogs (18) but has been associated with bone marrow toxicosis and is no longer recommended (14,19). Finally, a combination of praziquantel, pyrantel pamoate, and febantel has demonstrated efficacy against Giardia in dogs. After ingestion, febantel is absorbed and metabolized into several compounds, including fenbendazole, which is thought to result in the activity against Giardia (19).

Recently, a novel vaccine (GiardiaVax; Ayerst/Fort Dodge Laboratories, Guelph, Ontario) to reduce the shedding of Giardia cysts and to prevent clinical signs of canine giardiosis has been licensed in Canada. The vaccine consists of chemically inactivated trophozoites. Following challenge with infective Giardia cysts, 20 vaccinated puppies demonstrated fewer clinical signs and had a reduced duration of cyst shedding compared with 10 unvaccinated puppies (20). This vaccine has also been employed in the treatment of dogs with chronic symptomatic giardiosis where traditional therapy failed (17). In this latter study, signs of clinical disease were reduced and cyst shedding terminated in all of 13 client-owned dogs by a maximum of 70 d following initial vaccination. It was, therefore, concluded that immunotherapy may have application in the treatment of chronic cases of canine giardiosis. However, the study was based on 5 individual clinical cases and 1 kennel of 8 dogs, and was neither controlled nor blinded. Another study reported no differences in the reappearance of Giardia cysts in feces of dogs that had been treated concurrently with a parasiticide and vaccine versus those that had been treated with parasiticide only (15).

The impact of vaccination on asymptomatic Giardia infections has yet to be investigated. As a result, the primary objective of this study was to determine the efficacy of therapeutic vaccination with this novel vaccine in research facility dogs with asymptomatic Giardia infections. Additional objectives were to determine the prevalence of Giardia infection in a research facility dog population, and to assess the performance of a Giardia antigen-ELISA compared with that of zinc sulphate flotation.

Materials and methods


The study was conducted using the resident canine research population (107 mixed breed dogs) housed at the animal facility (AF), Animal Care Services, University of Guelph, and 13 dogs (beagles) housed initially at a conditioning unit (CU), University of Guelph. Nine of the latter dogs were moved to the AF for the vaccination component of this study. The experimental dogs were involved in other noninvasive primary research programs and were used with permission from the principal investigators, with approval of the University of Guelph Animal Care Committee, and in accordance with the Canadian Council on Animal Care Guidelines for the Care and Use of Experimental Animals. All dogs were clinically normal at the time of initial testing and had been vaccinated against Bordetella, distemper virus, adenovirus type 2, parainfluenza virus, parvovirus, and rabies virus (Progard-KC, Progard-5, Intervet, Millsboro, Delaware, USA; Imrab 3 TF, Merial, Athens, Georgia, USA). None of the dogs received immunosuppressive drugs during the course of this study. In addition, none of the dogs had been treated with albendazole, fenbendazole, or metronidazole during the 6 mo prior to this study.

The dogs used in the vaccine trial were housed individually in climate-controlled rooms (18°C to 21°C) with a daily 12-hour light/dark cycle throughout the study. Most rooms contained individual runs with 1 dog per run, and each room contained between 2 and 8 dogs. Assignment of dogs to runs or rooms changed during the course of the study, depending on the facility management needs and the primary research that the dogs were involved in. The floors of all the runs consisted of a raised plastisol-coated surface. A carpet mat was provided as a sleeping area and pens were cleaned daily by 11 am. Dogs were walked outside, or socialized indoors, for 20 min, a minimum of 5 times per week. Contact between dogs occurred randomly during these sessions. Once every 2 wk, the pens were disinfected with didecyl dimethyl ammonium chloride and dimethyl benzyl ammonium chloride (Ascend; Huntington Laboratories, Huntington, Indiana, USA). Dogs were fed a commercial dry dog chow (Eukanuba; The Iams Company, Dayton, Ohio, USA).

Prevalence determination

In January 2002, feces were collected from all dogs (n = 107) housed at the AF and 13 dogs housed at the CU. The majority of dogs (n = 100) at the AF were individually housed and fecal samples were collected from the pens of these dogs in the afternoon, after the rooms had been cleaned. Samples from dogs that had not defecated by the end of the workday were collected first thing the following morning. Feces were collected directly from the rectum for any dogs that had not defecated in their kennel or were group housed; 7 of the AF dogs and all of the dogs tested at the CU were group housed.

All fecal samples were tested within 24 h of collection for Giardia antigen by using an immunoassay that detects the antigen GSA65 (ProSpecT Giardia Microplate Assay). The assay results were read using a spectrophotometer (Microplate Autoreader EL311; Bio-Tek Instruments, Winooski, Vermont, USA) at a wavelength of 450 nm. Feces from Giardia antigen-positive dogs were also examined microscopically, within 36 h of collection, for the presence of Giardia cysts by using the zinc sulphate fecal flotation method described by Conboy (21). In every case, a sample size of approximately 5 g was used. Furthermore, the same fecal sample was used for the Giardia-ELISA analysis and zinc sulphate fecal flotation. Feces were stored at 4ºC prior to analysis. Investigators (n = 3) performing the Giardia antigen- ELISA and microscopic examinations were blinded throughout the study as to the origin of the samples. They were also blinded to the diagnostic data for each sample. All samples that were Giardia-positive by zinc sulfate fecal flotation were confirmed by at least one additional investigator.

Vaccine study

After determining the prevalence of Giardia infection in the study population, all available asymptomatic, Giardia antigen-positive dogs (11 of the 12 AF dogs positive in the prevalence study, and 9 dogs from the CU) were randomly assigned to 1 of 2 experimental groups: Group 1 (Giardia vaccine, 1 mL, SC, n = 10) or group 2 (saline control, 1 mL, SC, n = 10). The vaccine (GiardiaVax) was administered according to the manufacturer’s instructions. Day 0 of the study was the date of the primary vaccination, which took place within 1 wk of the aforementioned Giardia-prevalence study. Both groups of dogs received a booster vaccination 3 wk after the initial vaccination. Thereafter, Giardia antigen and cysts in feces were assayed every 4 wk, beginning 4 wk after the primary vaccination, for a period of 6 mo. In order to monitor the background rate of Giardia infection in the facility during the vaccine trial, 25 dogs (group 3), negative for Giardia antigen on day 0, were monitored at the same frequency as the animals in groups 1 and 2. All 25 dogs were tested with the Giardia antigen-ELISA on each sampling day, with the exception of 4 dogs that were no longer in the AF at week 24; 9 of the dogs were also tested on each occasion with the zinc sulphate flotation method. Fecal sample collection and analysis were the same as described for the initial prevalence determination.

After completion of the vaccine trial, all Giardia antigen-positive dogs were treated with fenbendazole (Safe-Guard; Intervet Canada, Whitby, Ontario) 50 mg/kg bodyweight (BW), PO, q24h for 3 d. Seven days after the last day of treatment, feces were collected from all animals and analyzed for Giardia antigen.

Statistical analyses

For the purposes of the analysis, dogs were considered to be positive for Giardia infection if they tested positive either with the Giardia antigen-ELISA or by zinc sulphate fecal flotation. All analyses comparing groups 1 and 2 were repeated after a more conservative definition was used, only considering dogs to be infected with Giardia on the basis of a positive zinc sulphate flotation. Univariable analyses and multivariable logistic regression analyses, controlling for age and exposure to group 1 and 2 dogs, were performed for each sampling period with a software system for data analysis (SAS for Windows 8.01; SAS Institute, Cary, North Carolina, USA). Values of P < 0.05 (Fisher’s Exact Test) were considered significant. A repeated measures analysis, using a generalized estimating equation (GEE) population-averaged approach, was performed using computer software (Stata, version 8.0; Stata Corporation, College Station, Texas, USA). Unlike the logistic regression analysis, the GEE models accounted for correlation between test results from the same animal, allowing inclusion of all test outcomes for all dogs in the same model (22). As for the logistic regression models, age and exposure to group 1 and group 2 dogs were controlled. Logistic regression and GEE models were constructed that compared group 1 with group 2 only, and all 3 groups together. Comparison of the ELISA results and the zinc sulphate test results was conducted by using an in-house Fortran program that precisely calculates kappa and McNemar’s test for symmetry (William Sears, Guelph, Ontario). Sensitivity and specificity of the ELISA, using zinc sulphate flotation as the nominal gold standard, were also calculated.


Prevalence determination

The prevalence of Giardia antigen, on the basis of single fecal samples, in dogs housed at the AF was 11% (12 dogs). Eight of the 12 antigen-positive dogs were also Giardia-cyst positive based on zinc sulphate fecal flotation, giving a microscopically conf irmed prevalence of at least 7.5% (note, only antigen-positive samples were tested with zinc sulphate). At the CU, the prevalence of Giardia antigen in feces was determined to be 100% among the 13 dogs that were tested (note, not all dogs at the CU were tested). Four of the 9 dogs transferred to the AF for the vaccine phase of the study were Giardia cyst-positive based on zinc sulphate fecal flotation.

All dogs were asymptomatic at the time of testing. The age of the 107 AF dogs tested ranged from 3 mo to 6.5 y (mean = 35.4 mo; median = 36 mo). However, the age range of the 12 AF Giardia antigen-positive dogs was 6 to 18 mo (mean = 13.6 mo; median = 18 mo).

Vaccine study

At weeks 4, 8, 12, and 16 following initial vaccination, the vaccinated group of dogs had a higher proportion of Giardia-positive dogs (positive on the basis of either the Giardia antigen test or zinc sulphate) at each sampling point (10/10, 9/10, 9/10, 8/10, respectively) compared with the control group of dogs (7/10, 7/10, 8/10, 4/10, respectively). At week 20, both groups had equal numbers of Giardia-positive dogs (5/10) and, at the time of the last sampling (week 24), the vaccinated group had fewer Giardia-positive dogs than did the control group (2/10 versus 5/10, respectively) (Figure 1). However, there was no statistically significant difference between the vaccinated (group 1) and control group (group 2) at any sampling time. The only point at which the difference approached significance in univariable analysis was week 16, when there were more positive dogs in the vaccinated group than in the control group (P = 0.075). There were no significant differences between the vaccinated and control groups in any of the sampling period specific multivariable logistic regression models or the overall repeated measures GEE model (P = 0.147; coefficient for group 2 versus group 1 = −0.79, CI95: −1.85 to 0.28).

Figure 1
Giardia infection status of asymptomatic Giardia-positive dogsa, administered either vaccine or saline at weeks 0 and 3, and Giardia-negative controls.

When only dogs that tested positive with zinc sulphate flotation were considered positive, the vaccinated group had a higher proportion of Giardia-positive dogs at weeks 4, 8, and 16 (5/10, 4/10, and 5/10, respectively) compared with the saline control group (4/10, 3/10, and 3/10, respectively). At weeks 12 and 24, the vaccinated group had fewer Giardia-positive dogs (2/10, 0/10, respectively) than the saline control group (4/10, 2/10, respectively). At week 20, both groups were equal (4/10) (Figure 2). However, as with results from antigen testing, there was no significant difference between the vaccinated and control groups at any time. There were also no significant differences between the vaccinated and control groups in any of the multivariable logistic regression models or the repeated measures GEE model (P = 0.117, coefficient for group 2 versus group 1 = −0.68, CI95: −1.54 to 0.17). Two of the group 1 dogs and 3 of the group 2 dogs were negative on zinc sulphate at day 0. Both of the negative group 1 dogs and 2/3 of the negative group 2 dogs had positive zinc sulphate flotation results at some point during the study.

Figure 2
Giardia infection status of asymptomatic Giardia-positive dogsa, administered either vaccine or saline at weeks 0 and 3, and Giardia-negative controls.

Among the group 3 dogs, 7/25 dogs tested positive for Giardia by antigen-ELISA or zinc sulphate at least once during the 24 wk of monitoring. Two dogs were positive at weeks 4 and 8; 1 was positive at week 4 only; 1 at weeks 8 and 12; 1 at weeks 20 and 24; 1 at weeks 4, 12, and 24; and 1 was positive at each sampling from week 4 to week 20. All but 1 of the 7 dogs were housed at least once in the same room as a group 1 or group 2 dog. None of the group 3 dogs were treated during this study for Giardia. At all sampling points except week 24, the proportion of dogs negative for Giardia was significantly higher in group 3 than in groups 1 and 2 (P < 0.005). However, there was no difference between the 3 groups at week 24. In the GEE model in which group 3 was compared with groups 1 and 2, exposure to a group 1 or 2 dog was not significantly associated with infection status in group 3.

Persistent clinical signs of giardiosis were not observed in any of the dogs throughout the 6-month study. However, 2 episodes of transient diarrhea were observed in 1 vaccinated dog in group 1 (1 d duration in week 7, and 1 d in week 9), 2 episodes of diarrhea were observed in 1 group 2 dog (3 d in week 3 and 1 d in week 10), and 1 episode of diarrhea (duration from 1 d to 5 d [median = 1 d]) was observed in each of 7 group 3 dogs. There was no apparent pattern to the occurrence of transient diarrhea and no association with group membership or housing with group 1 or 2 dogs.

All Giardia antigen-positive dogs (n = 7) treated with fenbendazole upon completion of the vaccine trial were negative for Giardia antigen 7 d after the last day of treatment.

Test comparison

Over the course of the 6-month study, 63 fecal samples tested positive for Giardia cysts (Table 1). Of these samples, 56 (88.9%) were also Giardia-antigen positive. Of 131 samples that were negative for Giardia cysts, 52 (39.7%) tested positive for Giardia antigen. McNemar’s test for symmetry was significant (P < 0.0001; OR = 7.4, CI95: 3.36 to 19.38), indicating that samples were 7.4 times more likely to be positive with the ELISA than with the zinc sulphate flotation test. Kappa was 0.42 (CI95: 0.30 to 0.53).

Table 1
Comparison of Giardia antigen enzyme-linked immunosorbent assay (ELISA) with zinc sulphate fecal flotation for detection of Giardia spp. in 194 fecal samples obtained from 29 dogs


The prevalence (11%) of Giardia antigen in the feces of dogs housed at the AF was somewhat higher than the value of 7.6% (determined with the same ELISA) that was observed in a survey of dogs presented to 15 veterinary practices located across Canada (3). However, the prevalence based on zinc sulphate alone was at least 7.5% (8 dogs), similar to the 7.2% based on formalinether concentration reported by Jacobs et al (3). For both this and the survey prevalence study, only antigenpositive samples were examined for Giardia cysts. As a result, they did not account for antigen-negative, cystpositive cases, which occurred 7 times out of 194 test pairs during the vaccine study described here. Differences with regard to housing and contact frequency may account for the higher prevalence of antigen positivity in the research facility dogs compared with the clientowned dogs. All the dogs tested at the CU in this study were positive for Giardia antigen. However, only a small number of the dogs housed at the facility were tested and all of the tested dogs were group-housed in the same room. The tested dogs had also arrived from the same facility, where it is believed that they may have acquired the Giardia infection.

The principal aim of this investigation was to determine whether vaccination of dogs with GiardiaVax would result in elimination of asymptomatic Giardia infections. However, at no point during the 6-month trial was there a significant difference between the vaccinated and control groups with respect to Giardia-infection status when defined on the basis of either the antigen-ELISA and zinc sulphate flotation test combined or the zinc sulphate test alone. This was true both for comparisons between the vaccinated and saline control groups at each sampling point, including the study end point, and for the entire study period when the sampling points were treated as repeated measures. The study sample size was small but sufficient to have detected a significant (P < 0.05) effect if the response to the vaccine was similar to that suggested by the results of Olson et al (17,20). For example, the sample size could have detected a 90% cure in the treated group versus a 20% cure in the placebo group with 95% confidence and 80% power. It is possible that the failure to detect a significant effect was due to insufficient statistical power. However, if that is the case and there is a true clinical effect, it is likely too small to have any clinical utility in comparison with anti-Giardia drug therapy. The vaccine used in this study, therefore, was not found to be an effective treatment for asymptomatic canine Giardia infections in a research facility or kennel setting.

The results described here differ from those in a report of 13 dogs with chronic symptomatic giardiosis where the use of the same vaccine was associated with resolution of clinical signs and the termination of cyst shedding within 70 d of vaccination (17). In our study, Giardia infections in some dogs appeared to persist throughout the follow-up period of 6 mo after vaccination. Furthermore, some dogs to which only saline was administered appeared to self-cure. Thus, since the report by Olson et al (17) did not include any control animals, the resolution of Giardia infections and clinical signs that was observed by these authors may not necessarily have been associated with vaccination. However, it is possible that the impact of the vaccine depends on whether dogs infected with Giardia are symptomatic or asymptomatic. Alternatively, differences in Giardia genotype may be responsible for the different results (23,24). Lastly, it should be noted that most of the dogs described by Olson et al (17) were treated with fenbendazole, albendazole, or metronidazole at approximately the same time as vaccination. As a result, conclusions about vaccine efficacy may be confounded by treatment.

The results of the current study are consistent with work in dogs by Payne et al (15) and a recent controlled trial that evaluated the use of this vaccine for treatment of experimentally induced Giardia infections in 16 cats (25). In the latter trial, the feces of cats were monitored by using an immunofluorescent antibody test for 24 wk after vaccination. No significant difference was found between the vaccinated and control groups.

With the exception of occasional transient diarrhea, all Giardia-positive dogs in the study described here were asymptomatic throughout the duration of the experiment, supporting previous reports that most Giardia infections in dogs are not associated with clinical signs (3,7). The sporadic occurrence of diarrhea in the study population was not significantly associated with Giardia infection status. However, all of the dogs that tested positive for Giardia antigen were under 2 y of age. Furthermore, age was consistently negatively associated with Giardia infection status in the logistic regression and GEE models. Since susceptibility to Giardia infection is believed to be influenced by previous exposure and maturation of the animal’s immune system, the data are consistent with the hypothesis that younger dogs are more susceptible to infection (3,6).

As indicated, unequivocal clinical signs of giardiosis were not observed in any of the Giardia-positive dogs throughout the 6-month study. This is of concern, as these animals are potential sources of environmental contamination and may increase the risk of infection for other animals, as is suggested by the results of dogs in group 3. Monis et al (24) observed the same Giardia genotype in dogs and humans and, therefore, concluded that certain genotypes may be zoonotic. However, the true risk posed by asymptomatic Giardia antigen-positive dogs to humans, and other animals, is currently unclear and requires clarification.

The sensitivity (88.9%) and negative predictive value (91.7%) of the ELISA, using the zinc sulphate flotation test as the gold standard, suggested that the ELISA, given its inherent labor savings, is a reasonable screening tool in a laboratory animal facility setting where a large number of dogs have to be screened within a short period. Data from the Giardia antigen-ELISA agreed “moderately” well (κ = 0.42) with the zinc sulphate flotation test when the traditional guidelines for assessing kappa were used; however, because the proportion of samples that tested positive to the 2 tests differed significantly, as indicated by the McNemar’s test, the kappa is of little value (20). Seven of 63 (11%) fecal samples that were positive for Giardia cysts were ELISA negative. Since there should be no false positives with the zinc sulphate flotation test provided that the test is performed correctly, false negatives may occur with the ELISA. Therefore, a negative ELISA does not necessarily rule out infection. The disagreement between the 2 tests (52 of 59 test pairs that disagreed) was largely associated with positive ELISA results when the zinc sulphate flotation test was negative. This may be a result of ELISA false positives, zinc sulphate flotation false negatives, or both. Only 1fecal sample was collected at each test point and, because the parasite can be intermittently shed, this could have resulted in a false negative result with the zinc sulphate flotation test. Lastly, although the investigators examining feces were experienced in identifying Giardia cysts, investigator error may potentially be responsible for some discrepant results.

The diagnostic results differ somewhat from a previous study that examined 164 fecal samples from 77 dogs with the same ELISA and zinc sulphate flotation test (5). That study reported a similar level of agreement (κ= 0.49) to the work reported here. However, the McNemar’s test was not significant and the results indicated a sensitivity of 56% and a specificity of 86% for the ELISA when zinc sulphate flotation was used as the gold standard. Thus, in the work of Barr et al (5), samples were equally likely to be positive with the ELISA and zinc sulphate flotation, whereas in our study, samples were more likely to be positive with the ELISA. In the study described by Barr et al (5), there was a relatively high proportion of false negatives with the ELISA compared with zinc sulphate, while in our study, there was a relatively high proportion of ELISA false positives (or zinc sulphate false negatives). In both studies, multiple samples were taken from individual dogs: At 24-hour intervals in the work described by Barr et al (5), and at 4-week intervals in the study described here. Given the time difference between tests in the same animal, it has been suggested that the lack of independence is a relatively marginal concern (5). The differences in the results may reflect differences in the study environments, the study subjects, or the way in which the tests were administered. This highlights the need for evaluation of screening and diagnostic tests in a variety of situations. At present, there is no definitive answer on how well the ELISA functions as a screening test, primarily because the zinc sulphate flotation test can only be regarded as an incomplete gold standard, since false negatives can occur. Further work is required to clarify whether the antigen test is a more appropriate screening test. Lastly, polymerase chain reaction (PCR) has been shown to be more sensitive than either microscopy or a Giardia ELISA for detection of Giardia infection in cats (26). However, work is required to determine the sensitivity of PCR for the detection of Giardia in dogs. If PCR is more sensitive and specific than other diagnostic techniques for the detection of Giardia in dogs the method may be a useful diagnostic and screening tool.


The authors thank the researchers and staff at Animal Care Services, University of Guelph, for their help and support throughout the course of this study. William Sears is thanked for his help with the statistical analysis. CVJ


Reprints will not be available from the authors.

Financial support was provided by Pet Trust, Ontario Veterinary College, University of Guelph.


1. Leib MS, Zajac AM. Giardia: diagnosis and treatment. In: Bonagura JD, ed. Current Veterinary Therapy XII: Small Animal Practice. Philadelphia: WB Saunders, 1995:716–719.
2. Marshall MM, Naumovitz D, Ortega Y, Sterling CR. Waterborne protozoan pathogens. Clin Microbiol Rev. 1997;10:67–85. [PMC free article] [PubMed]
3. Jacobs SR, Forrester CPR, Yang J. A survey of the prevalence of Giardia in dogs presented to Canadian veterinary practices. Can Vet J. 2001;42:45–46. [PMC free article] [PubMed]
4. Kirkpatrick CE. Epizootiology of endoparasitic infections in pet dogs and cats presented to a veterinary teaching hospital. Vet Parasitol. 1988;30:113–124. [PubMed]
5. Barr SC, Bowman DD, Erb HN. Evaluation of two test procedures for diagnosis of giardiasis in dogs. Am J Vet Res. 1992;53:2028–2031. [PubMed]
6. Kirkpatrick CE. Giardiasis. Vet Clin North Am Small Anim Pract. 1987;7:1377–1387. [PubMed]
7. Barr SC, Bowman DD, Heller RL. Efficacy of fenbendazole against giardiasis in dogs. Am J Vet Res. 1994;55:988–990. [PubMed]
8. Heresi G, Cleary TG. Giardia. Pediatr Rev. 1997;18:243–246. [PubMed]
9. Ortega YR, Adam RD. Giardia: Overview and update. Clin Infect Dis. 1997;25:545–550. [PubMed]
10. Wolfe MS. Managing the patient with giardiasis: clinical diagnostic and therapeutic aspects. In: Jakubowski W, Hoff JC, eds. Waterborne Transmission of Giardiasis. Cincinnati, Ohio: Environmental Protection Agency 1979:39–52.
11. Zimmer JF, Burrington DB. Comparison of four techniques of fecal examination for detecting canine giardiasis. J Am Anim Hosp Assoc. 1986;22:161–167.
12. Zimmer JF, Burrington DB. Comparison of four protocols for the treatment of canine giardiasis. J Am Anim Hosp Assoc. 1986;22:168–172.
13. Upcroft J, Upcroft P. My favourite cell: Giardia. Bioessays. 1998;20:256–263. [PubMed]
14. Lieb MS, Zajac AM. Giardiasis in dogs and cats. Vet Med. 1999;94:793–802.
15. Payne PA, Ridley RK, Dryden MW, Bathgate C, Milliken GA, Stewart PW. Efficacy of a combination febantel-praziquantel- pyrantel product, with or without vaccination with a commercial Giardia vaccine, for treatment of dogs with naturally occurring giardiasis. J Am Vet Med Assoc. 2002;220:330–333. [PubMed]
16. Zajac AM, LaBranche TP, Donoghue AR, Chu TC. Efficacy of fenbendazole in the treatment of experimental Giardia infection in dogs. Am J Vet Res. 1998;59:61–63. [PubMed]
17. Olson ME, Hannigan CJ, Gaviller PF, Fulton LA. The use of a Giardia vaccine as an immunotherapeutic agent in dogs. Can Vet J. 2001;42:865–868. [PMC free article] [PubMed]
18. Barr SC, Bowman DD, Heller RL, Erb HN. Efficacy of albendazole against giardiasis in dogs. Am J Vet Res. 1993;54:926–928. [PubMed]
19. Barr SC, Bowman DD, Frongillo MF, Joseph SL. Efficacy of a drug combination of praziquantel, pyrantel pamoate, and febantel against giardiasis in dogs. Am J Vet Res. 1998;59:1134–1136. [PubMed]
20. Olson ME, Ceri H, Morck DW. Giardia vaccination. Parasitol Today. 2000;16:213–217. [PubMed]
21. Conboy G. Giardia. Can Vet J. 1997;38:245–247. [PMC free article] [PubMed]
22. Dohoo I, Martin W, Stryhn H. Veterinary Epidemiologic Research. Charlottetown, Prince Edward Island: Atlantic Veterinary College 2003:91–92,527–531.
23. Lu SQ, Baruch AC, Adam RD. Molecular comparison of Giardia lamblia isolates. Int J Parasitol. 1998;28:1341–1345. [PubMed]
24. Monis PT, Andrews RH, Mayrhofer G, et al. Novel lineages of Giardia intestinalis identified by genetic analysis of organisms isolated from dogs in Australia. Parasitology. 1998;116:7–19. [PubMed]
25. Stein JE, Radecki SV, Lappin MR. Efficacy of Giardia vaccination in the treatment of giardiasis in cats. J Am Vet Med Assoc. 2003;222:1548–1551. [PubMed]
26. McGlade TR, Robertson ID, Elliot AD, Thompson RCA. High prevalence of Giardia detected in cats by PCR. Vet Parasitol. 2003;110:197–205. [PubMed]

Articles from The Canadian Veterinary Journal are provided here courtesy of Canadian Veterinary Medical Association
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...