Prevalence and Factors Associated With Fecal Shedding of Giardia spp. in Domestic Cats

Introduction

Giardia (G.) duodenalis (synonyms G. intestinalis and G. lamblia) is a flagellated intestinal protozoan parasite that affects many mammalian species, including cats, dogs, and humans.^1–10 In cats, the predominant clinical sign reported in Giardia spp. infections is acute, small-bowel diarrhea.^{1,2,4,5,11–13} Less frequently, chronic small- or large-bowel diarrhea can also be seen.^{1,2,4,5,11–13} Most cats infected with Giardia spp. continue to have a normal appetite; but in severe cases, dehydration and anorexia can occur, especially in young or debilitated animals.⁴ Although vomiting has been reported as a possible clinical sign, documented instances of vomiting in cats with giardiasis are extremely uncommon.^{2,9,11,13–17}

Evaluation of the records of the Parasitology Laboratory at Mississippi State University College of Veterinary Medicine (MSU-CVM) revealed that a major presenting complaint in over half of the cats with confirmed giardiasis was vomiting, with or without diarrhea.^a Evaluation of the records also revealed that approximately 20% of all fecal samples submitted contained Giardia spp. cysts, which was higher than the percentage previously reported for cats (1% to 11%).^1,14,18–28 These differences in clinical signs may potentially be explained by the presence of a cat-adapted genotype of G. duodenalis, as has been demonstrated in humans, and may also explain the apparent higher regional prevalence of Giardia infections.¹⁶ Alternatively, the actual prevalence of feline giardiasis in previous studies may have been higher than what was reported because of difficulties in obtaining a definitive diagnosis using standard diagnostic tests, such as zinc sulfate flotation techniques, or because of rapid and complete responses to common empirical gastrointestinal drugs, such as fenbendazole and metronidazole.^{4,5,7,11,13–15,29,30} The purposes of this study were to document the regional prevalence of Giardia spp. cysts in the feces of cats and to identify factors associated with infection, including the nature and frequency of clinical signs.

Materials and Methods

Fresh fecal samples from cats were voluntarily submitted by owners and veterinarians in response to requests attached to referral letters, announcements at regional veterinary continuing education seminars, and email announcements to the faculty and students at MSU-CVM. Samples were sought from cats of all ages, sexes, breeds, and housing circumstances, and from cats with or without signs of gastrointestinal disease (e.g., vomiting and/or diarrhea). Written owner or agent authorization, a completed risk factor questionnaire, and an adequate fresh fecal sample (>3 g) were required for cats to be eligible for participation in the study. The study was approved by the Institutional Animal Care and Use Committee.

Fecal samples were stored at 4°C and were analyzed within 5 days of collection. Samples were analyzed using a centrifugation-flotation technique and a commercially available direct immunofluorescent antibody (IFA) kit,^b as previously described.^31,32 Giardia spp. cysts and Cryptosporidium spp. oocysts were identified using a fluorescent microscope at the 100× power, and then they were confirmed at 400× magnification.^c Giardia spp. cysts identified by the IFA stain bright apple-green in color, are oval shaped, and range in length from 8 to 15 μm. In contrast, Cryptosporidium spp. oocysts are generally smaller (2 to 6 μm in length), round, and have a thicker capsule that also stains bright apple-green on the IFA [see Figure]. The IFA has been shown to be reliable on fecal samples refrigerated or frozen for up to 3 years.^d

When sufficient sample was available, fecal flotation was also performed as previously described.³⁰ Parasite eggs and oocysts were identified using 400× magnification and were primarily categorized according to type (e.g., coccidia, ascarids). Fecal samples were scored as positive if one or more eggs or oocysts were identified.

Results

A single fecal sample was obtained from each of 250 domestic cats from Mississippi and northwestern Alabama; cats were of various breeds, genders, and ages and had various clinical signs. The overall prevalence of Giardia spp. cysts was 13.6% (n=34; 95% confidence intervals [CIs]=9.6, 18.5); that is, there was a 95% certainty that the interval of 9.6 to 18.5 contained the prevalence. The cats came from 159 households, of which multiple cats were from 29 households and single cats were from 130 households. Forty-eight percent (n=14) of households with multiple cats had at least one cat shedding Giardia cysts, compared to 5% (n=7) of the cats from a single cat household. The odds of a cat shedding cysts were increased if another cat shedding Giardia spp. cysts was also present in the household (χ²=19.93; odds ratio=5.51; 95% CIs=2.57, 11.79).

Twelve (35%) of the 34 cats shedding Giardia spp. cysts had diarrhea as a clinical sign (95% CIs=19.1, 52.2). Only two (5.9%) cats shedding Giardia spp. cysts had acute vomiting as a clinical sign (95% CIs=0.7, 19.7). For one cat, vomiting was the only clinical sign, while the second cat also had diarrhea. Twenty-five (74%; 95% CIs=55.6, 87.1) of Giardia-positive cats were concurrently shedding Cryptosporidium spp. oocysts, whereas nine cats were shedding only Giardia spp. cysts.

The prevalence of Giardia cysts in fecal samples from apparently healthy cats (n=6) was 6.1% (95% CIs=2.9, 14.0). The prevalence in cats (n=117) showing gastrointestinal signs (acute or chronic) was 20.5% (n=24; 95% CIs=2.3, 12.7). Of the 34 cats with histories of chronic, nongastrointestinal problems, 11.8% were shedding Giardia spp. cysts in their feces (95% CIs=3.3, 27.5) [Table 1].

Owners were asked to report test results for feline leukemia virus (FeLV) and feline immunodeficiency virus (FIV). Of the 250 cats in the study, 99 were untested, 149 were negative for both viruses, and two were positive for FIV. Consequently, these variables were not included in the univariate analysis. Univariate analysis showed no associations with type of housing (i.e., indoor only, indoor/outdoor, outdoor only), the source from which the animal was acquired (i.e., private home, shelter/stray, within or outside Mississippi), or the presence of a number of endoparasites (i.e., hookworms, ascarids, capillarids, Paragonimus kellicotti, Taenia spp). Age, sex, clinical signs, presence of cryptosporidial oocysts, and presence of coccidial oocysts were identified as meeting the criteria for inclusion in the multivariable model [Table 1]. Refitting the regression model resulted in the deletion of the age, sex, and acute diarrhea variables. Addition of two-way interactions did not contribute to the model, and these were not included in the final model. Because there was an increased odds ratio of cats shedding cysts in households where at least one other cat was shedding cysts, the variable household was retained in all models to control for individual household effects. Therefore, the final logistic regression model included household, presence of cryptosporidial oocysts, presence of coccidial oocysts, and a clinical history of gastrointestinal disorders beginning 2 weeks prior to sample submission [Table 2]. The Hosmer-Lemeshow goodness of fit statistic was 3.74 (degrees of frequency=8, P=0.88) and indicated that the overall fit of the model was good.

Discussion

This study demonstrated a prevalence of Giardia spp. in cats from northeastern Mississippi and northwestern Alabama that was slightly higher than what has been reported elsewhere.^1,14,18–28 The prevalence reported here may still have underestimated the actual prevalence of infection, since only a single fecal sample was analyzed from each cat. Because Giardia cysts are intermittently shed in the feces, the current recommendation for their detection is the collection of at least three separate fecal samples over 3 to 5 days that are analyzed using centrifugal-flotation techniques and zinc sulfate solution.^2,4,5 The reported sensitivity of this latter protocol is approximately 95%.^2,4,5 Identification of Giardia spp. cysts has also been historically difficult, because the cysts are small, transparent, and shed intermittently.^2,4,5 Additionally, routine fecal flotation solutions can cause crenation of the cysts, making their identification difficult.^2,4,5 Unlike trophozoites, cysts can survive for several days if stored at 4°C.¹⁰ Freezing of the samples or storage in 10% formalin is not recommended if identification of Giardia cysts is required.¹⁰ These limitations can be minimized with the use of the IFA test, which has been shown to have improved sensitivity and specificity over flotation/concentration techniques using light microscopy in cattle, sheep, and humans.^35–37 The use of the IFA technique to detect Giardia spp. cysts in the feline fecal samples in the study reported here was expected to have similar improved performance and may have helped to offset some of the limitations associated with testing only one sample from each cat. Therefore, the apparent increased prevalence of Giardia spp. in the present study may be genuine and may reflect the development of improved methods for detecting the organism in feces. Because fecal samples were voluntarily submitted by local veterinarians or owners, a sample selection bias for cats that received routine veterinary care may also have influenced the reported prevalence.

No statistically significant association was found between the presence of Giardia cysts in the feces and the presence of vomiting. This result may reflect a true lack of association or the low number of vomiting cats enrolled in the study, since very few cats (n=26) had vomiting reported as a clinical sign [Table 1]. Only 7.1% of cats with vomiting as the only clinical sign and 8.3% of cats with vomiting as a component of the clinical signs were shedding Giardia spp. cysts. These results did not support the original hypothesis of an increased association of Giardia infection with vomiting in cats.

Chronic gastrointestinal disorders, such as diarrhea, were significantly (P<0.0170) associated with Giardia spp. cysts in the feces [Table 2]. Diarrhea was the primary gastrointestinal clinical sign associated with Giardia infections.^1–4 It is interesting to note, however, that acute diarrhea (<2 weeks) was not associated with Giardia cysts in the feces. The prepatent period of giardiasis is reported to be 5 to 16 days; consequently, the lack of an association with acute diarrhea may have reflected the testing of animals in the prepatent period when cysts would not have been present.^1,4,5,7

Eighty-three percent of the cats in the study that were shedding cryptosporidial oocysts were concurrently shedding Giardia cysts, and a strong association between the presence of Giardia spp. and Cryptosporidium spp. was detected [Table 2]. This association, which has also been reported elsewhere, suggests that cats with the two protozoal parasites may share common risk factors.³⁸ Alternatively, infection with one parasite may predispose cats to concurrent infection with the other parasite. It has been suggested that noninvasive Giardia spp. trophozoites may damage the enterocyte epithelial surface, facilitating the invasion of cells by cryptosporidial organisms.^38,39 Further work is necessary to clarify this association.

Of the 222 cats with a sufficient fecal sample size to perform a fecal flotation, only the presence of coccidial parasites was found to be associated with shedding of Giardia cysts. Because of morphological similarities between the oocysts of a number of coccidial parasites, positive identification of a specific coccidial agent is difficult based solely upon the microscopic appearance of the oocysts.¹⁰ In the present study, potential coccidial parasites included Toxoplasma gondii, Isospora spp., Hammondia spp., and Besnoitia spp. Of these, Isospora spp. are the most common coccidial parasite reported in cats.¹⁰ The association between Giardia spp. and coccidial parasites may suggest shared risk factors (e.g., common age or shared environment) or may indicate that there is some degree of synergism between different enteric parasites, as has been suggested for Cryptosporidium spp.^38,39

The lack of association between the shedding of Giardia cysts and either housing or the source of the cats was somewhat surprising, since previous studies reported a Giardia spp. prevalence of up to 100% in infected breeding facilities.^7,40 The study reported here did not include any breeding catteries, although many of the study subjects came from multiple-cat households. While no association was found in cats living indoors, outdoors, or both, a selection bias may have been present, because samples were not randomly selected and owners were not required to submit fecal samples from all cats within the same household. Because litter pans are typically shared among cats in the same household, there is ample opportunity for contact with the immediately infective cysts. Therefore, if one cat is shedding oocysts within a household, it is expected that one or more of the other cats would also be shedding oocysts; this was an observation confirmed in the present study.⁴⁰

The inability to evaluate associations between feline retroviral status (i.e., FeLV, FIV) and the shedding of Giardia cysts was unfortunate. Some protozoan parasites, such as Cryptosporidium spp. and Toxoplasma gondii, occur with increased prevalence and morbidity in humans infected with retroviruses (i.e., human immunodeficiency virus), immunosuppressive diseases, or iatrogenic immunosuppression associated with organ transplantation or chemotherapy.^41,42 Whether this increased prevalence occurs in cats is unknown, and further studies are needed to determine the influence of retroviral infections and general immunosuppression on the shedding of Giardia cysts.

The prevalence of Giardia spp. infection, as well as the presence or absence, nature, and severity of clinical signs associated with infection, are influenced by many host and parasite factors.^16,43 One potentially important factor determining the prevalence of Giardia infections and associated clinical signs is parasite genotype. The influence of Giardia genotype on presenting symptomatology has been recently recognized in humans, and parasite genotype has the potential to similarly influence the nature of clinical signs in cats.^16,43 Determination of the effect, if any, that Giardia genotype has on the prevalence of infection, presenting clinical signs, and the agent’s zoonotic potential in cats may facilitate better characterization of giardiasis in domestic cats.

Conclusion

The present study shows a higher prevalence of cats shedding Giardia spp. cysts than what has been reported in the literature. This higher prevalence may be a reflection of the diagnostic test used or may indicate a naturally higher prevalence. The presence of Cryptosporidium spp. oocysts, of coccidial oocysts, and a history of chronic (>2 weeks) gastrointestinal signs were found to be significantly associated with the presence of Giardia cysts in the feces. There were no associations between the presence of Giardia cysts and gender, source of cat, type of housing, or evidence of acute diarrhea, vomiting, or gastrointestinal parasites.

Acknowledgments

The authors acknowledge the contributions of Dr. Carla Panuska, Mr. Erle Chenney, and Mr. Daniel Shy for their assistance in the laboratory procedures performed for this project.