• Open Access

Prevalence of Selected Bacterial and Parasitic Agents in Feces from Diarrheic and Healthy Control Cats from Northern California

Authors


  • This study was undertaken at the University of California, Davis, Veterinary Medical Teaching Hospital, and all fecal testing with the exception of cultures for Clostridium perfringens and C. difficile was performed at IDEXX Reference Laboratories, West Sacramento, CA.

Corresponding author: Dr Stanley Marks, University of California, Davis, School of Veterinary Medicine, Department of Medicine and Epidemiology, One Shields Ave, Davis, CA 95616; e-mail: slmarks@ucdavis.edu.

Abstract

Background

Bacterial and parasitic agents are commonly implicated as causes of diarrhea in cats, but there is a paucity of information evaluating epidemiological and prevalence factors associated with most of these organisms in cats.

Objectives

Determine the prevalence of selected enteropathogens in diarrheic and nondiarrheic cats.

Animals

A total of 219 diarrheic and 54 nondiarrheic cats.

Methods

Prospective study. Fresh fecal specimens were submitted for centrifugation flotation, culture, ELISA (Giardia, Cryptosporidium, Clostridium perfringens enterotoxin [CPE], and C. difficile toxin A [TcdA]) and polymerase chain reaction (PCR) testing (Tritrichomonas foetus and Campylobacter spp.). An epidemiologic questionnaire was completed for each cat.

Results

Campylobacter was isolated from significantly fewer diarrheic (21/219 or 9.6%) versus nondiarrheic cats (15/54 or 27.8%, = .001), and was detected in 74 of 131 cats (56.5%) via PCR. Campylobacter jejuni, C. helveticus, and C. upsaliensis were detected in 6.8, 100, and 44.6% of the 74 cats. Multiple Campylobacter spp. were identified in 47.3% of these cats. All cats were negative on fecal culture for Salmonella and for C. difficile TcdA via ELISA. CPE was detected in 9/219 diarrheic (4.1%) and in 1/54 nondiarrheic cats (1.9%, = .69). Cats < 2 years were significantly more likely to be infected with intestinal parasites (< .001).

Conclusions and Clinical Importance

Routine fecal cultures and toxin immunoassays for detection of bacteria are of limited diagnostic value in diarrheic cats. Molecular-based testing is superior to fecal cultures for detection and identification of Campylobacter spp., but positive test results do not correlate to the presence of disease.

Abbreviations
BA

Brucella agar

CCFA

cycloserine-cefoxitin-fructose-agar

CDI

Clostridium difficile infection

CP

crossing points

CPE

Clostridium perfringens enterotoxin

EYA

egg yolk agar

PCR

polymerase chain reaction

reverse-CAMP test

reverse-Christie, Atkins, and Munch-Peterson test TcdA and TcdB Clostridium difficile toxin A and B

VMTH

Veterinary Medical Teaching Hospital

XLT4

xylose-lysine-tergitol 4

Diarrhea is common in domestic cats[1] and can be associated with severe morbidity. A number of bacteria and parasites are detected in feces of cats with diarrhea[2-5]; however, large-scale epidemiological studies assessing Clostridium perfringens, Clostridium difficile, Campylobacter spp., and Salmonella and their prevalence factors in diarrheic and nondiarrheic domestic cats are mostly lacking. Veterinarians are faced with a dilemma when attempting to diagnose cats with suspected bacterial-associated diarrhea, because the isolation rates for putative bacterial enteropathogens are often similar in diarrheic and nondiarrheic cats, and the incidence of bacterial-associated diarrhea in cats is extremely variable.[5] In addition, the potential for zoonotic transmission of some gastrointestinal parasites and bacteria and the human health risks associated with domestic cat ownership are now being realized.[4, 5] Accurate information regarding the potential reservoir of infection for humans is warranted.

Fecal enteric panels comprising fecal culture for C. difficile, Campylobacter spp., and Salmonella and toxin immunoassays for C. perfringens enterotoxin (CPE) and C. difficile toxin A or B (TcdA or TcdB) in diarrheic cats are relatively expensive, and these panels have been shown to be of limited diagnostic value in diarrheic dogs.[6] A similar study evaluating the diagnostic value of fecal enteric panels combined with molecular testing has not been published in cats to date. Veterinary microbiology and reference laboratories are increasingly embracing polymerase chain reaction (PCR)-based fecal panels for the diagnosis of bacterial and parasitic enteropathogens; however, there has been a paucity of studies comparing molecular-based testing to conventional culture and toxin immunoassays in domestic cats with spontaneous infections.[2, 7] Knowledge of the prevalence factors associated with these agents should optimize the selection of fecal tests for the diagnosis of bacterial and parasitic agents in diarrheic cats.

The purpose of this study was to determine the prevalence of selected bacterial and parasitic agents in diarrheic cats from 3 sources (general practitioner hospitals, specialty practices, and shelters), as well as a control group of nondiarrheic cats, and to determine epidemiologic and prevalence factors associated with these agents.

Materials and Methods

This was a prospective study carried out between August 2007 and September 2009.

Fecal Specimen Collection and Preparation

Fresh fecal samples were obtained from diarrheic cats presenting to board certified internists in private practice, general practitioners, and from 3 regional shelters in northern California. Fresh fecal samples from apparently healthy, nondiarrheic cats were obtained from veterinary students and staff at the University of California, Davis, Veterinary Medical Teaching Hospital (VMTH). Fecal consistency was determined based on a fecal scoring system where a score of 1 was considered very firm, a score of 2 was considered well-formed, a score of 3 was considered soft-formed, and a score of 4 was considered watery. Only specimens with a score of 3 or 4 were included in the diarrheic group.

All fresh fecal specimens were processed within 2–3 hours after submission. The fecal specimens were processed into 3 aliquots: 3–5 g was immediately refrigerated at 3°C; 3–5 g was placed in anaerobic culture transport media1 at room temperature; and < 0.1 g was inoculated directly into Tritrichomonas foetus Feline InPouch TF culture media2 at room temperature. Fecal specimens were transported to IDEXX Reference Laboratories, West Sacramento, CA, for fecal immunoassays, InPouch fecal culture for T. foetus, fecal centrifugation flotation, bacterial culture for Campylobacter spp., Salmonella, Shigella, Aeromonas, Edwardsiella, and Plesiomonas, and PCR testing for Campylobacter spp. and T. foetus within 24 hours of collection. Fecal cultures for C. perfringens and C. difficile were performed at the UC Davis VMTH Microbiology Laboratory.

Fecal Flotation

Fresh feces were examined for parasite ova, cysts, and oocysts by use of a zinc sulfate single centrifugation flotation technique, as previously described.[8]

Immunoassays

Four enzyme-linked immunoassays (ELISAs) were used in this study: a separate microplate ELISA for Giardia3 and microplate ELISA for Cryptosporidium4; C. difficile ColorPac toxin A5; and C. perfringens enterotoxin (CPE)6 All ELISAs were performed according to manufacturers’ instructions. Color development was determined visually for C. difficile and spectrophotometrically at a single wavelength of 450 nm for C. perfringens and for Giardia/Cryptosporidium spp.

Fecal Culture

Tritrichomonas foetus Isolation

A T. foetus Feline-InPouch culture2 was performed on each fecal specimen according to manufacturer's instructions.2 All Feline InPouch cultures were shipped to IDEXX Reference Laboratories after inoculation at the point of fecal collection. Once delivered to IDEXX Reference Laboratories, the cultures were incubated at 35°C for 10 days and observed on a daily basis for trophozoites.

Campylobacter spp. Isolation

Fresh feces were plated and streaked for isolation via a sterile swab onto a selective Campylobacter agar with cefoperazone, vancomycin, and amphotericin B.7 Plates were incubated microaerophilically at 42°C in an AnaeroPack.8 Plates were examined for growth at 48–72 hours, and suspect colonies were Gram stained and subcultured to 5% sheep blood agar.9 Biochemical testing and sensitivity to 30-μg disks of cephalothin and naladixic acid were performed.

Clostridium perfringens Isolation

A quantity of 0.5 g of feces was inoculated onto McClungs egg yolk agar (EYA)9 to test for lecithinase production. All plates were incubated anaerobically at 37°C for 24–48 hours. Identification of C. perfringens was based on a positive reverse-CAMP test, lecithinase production, and a Gram stain demonstrating large, nonspore forming, “boxcar”-shaped Gram-positive bacilli.

Clostridium difficile Isolation

Feces preserved in anaerobic culture transport media were plated and streaked for isolation with a sterile swab onto prereduced cycloserine-cefoxitin-fructose agar (CCFA).7 After anaerobic incubation of plates for 24–48 hours at 37°C, they were examined for nonswarming yellow colonies exhibiting a “ground glass” appearance. Yellow colonies were subcultured to prereduced Brucella agar (BA)9 and incubated anaerobically for 24 hours at 37°C. Colonies were Gram stained and identification of C. difficile was made on the basis of lack of aerotolerance, colony morphology, fluorescence, odor, and detection of L-proline-aminopeptidase activity on both CCFA and BA.

Salmonella, Shigella spp., Aeromonas spp., Edwardsiella spp., and Pleisiomonas shigelloides Assays

Fresh feces were plated and streaked for isolation via a sterile swab onto MacConkey II agar,9 Hektoen Enteric agar,9 Triple Sugar Iron agar,9 EYA,9 and Lysine Iron agar slants.9 Preserved feces were also inoculated into a 4% selenite F enrichment broth.9 Incubated selenite enrichment broth was subcultured to xylose-lysine-tergitol 4 (XLT4) agar.9 All cultures, with the exception of the selenite broth, were incubated without CO2 at 37°C for 24–48 hours. The selenite broth was incubated for 24 hours. Lactose-negative colonies from MacConkey II and H2S-positive colonies from XLT4 were subcultured to biochemical media according to identification schema used by IDEXX Reference Laboratories.

Real-Time TaqMan Polymerase Chain Reaction Assays

Five real-time PCR assays were used to test DNA extracts of samples for the presence of T. foetus and 4 strains of Campylobacter: C. jejuni, C. lari, C. upsaliensis, and C. helveticus. Target genes for each application were T. foetus: ribosomal RNA gene AF339736; Campylobacter: UDP-N-acetylglucosamine acyltransferase gene (lpxA): C. jejuni: AL111168; C. coli: AY531496, C. lari: AY531481, C. upsaliensis: AY531470, C. helveticus: AY628395.

Sample Preparation and Real-Time Polymerase Chain Reaction Analysis

Native fecal samples were collected, stored, and shipped at 4°C and processed separately with protocols adapted as previously published.[9, 10] Briefly, 1 g of fecal material was reconstituted in the lysis solution and incubated for 10 minutes. Lysates were centrigued at low velocity and supernatants were extracted using Whatman filters in a Corbett X-Tractor platform.10 Nucleic acids were eluted into 150 μL of PCR-grade nuclease-free water and 5 μL amplified in subsequent real-time PCR reactions. Analysis was performed on a Roche LightCycler 48011 and raw data analyzed using the 2nd derivative maximum method to generate crossing points (CP). Real-time PCR was run with 7 quality controls including (1) PCR positive controls, (2) PCR negative controls, (3) negative extraction controls, (4) DNA pre-analytical quality control targeting the host ssr rRNA (18S rRNA) gene complex, (5) RNA pre-analytical quality control targeting the host ssr rRNA gene complex, (6) an internal positive control spiked into the lysis solution, and (7) an environmental contamination monitoring control.

Questionnaire

For each cat, a comprehensive epidemiologic questionnaire was completed by the owner and veterinarian at the time of fecal submission. The questionnaire provided information pertaining to the cat's signalment, presenting problem, environment and lifestyle (indoor versus outdoor), any history of predation, duration (acute versus chronic) and nature of diarrhea (large bowel, small bowel, or diffuse disease), presence or absence of blood (melena or hematochezia), appetite, presence or absence of weight loss, diet and supplements, concurrent medical issues, previous infectious causes of diarrhea, current medical treatment, deworming history, and results of any diagnostic testing within the past 7 days.

Statistical Analyses

Diarrheic and nondiarrheic cats were compared in terms of their prevalence of bacterial and parasitic agents using Fisher's Exact test. A 2-step procedure was followed in identifying prevalence factors for bacterial and parasitic agents. The 4 demographic variables evaluated were age, breed, sex and housing location (indoor versus outdoor). The impact of these demographic variables on the prevalence of a given agent was evaluated univariately using the 2-sample t-test (age) and Pearson's chi-square test (breed, sex, housing location). Demographic variables demonstrating risk potential univariately (P < .12) were entered as predictor variables in a logistic regression. P ≤ .05 identified statistical significance of a given demographic variable in the logistic regression. All statistical computations were made using SPSS software.12 Descriptive results are presented as mean ± SD.

Results

Fecal specimens were collected from a total of 273 cats, comprising 219 diarrheic and 54 nondiarrheic cats. These cats represented 22 different breeds, which were grouped according to ancestry for statistical purposes. Breeds were categorized into Western European breeds (n = 235 comprising 155 Domestic Shorthair, 31 Domestic Longhair, 29 Domestic Mediumhair, 5 Persian, 5 Ragdoll, 2 each of Russian Blue, American Shorthair, and Himalayan, and 1 each of Persian mix, Scottish Fold, Exotic Shorthair and Maine Coon); Southeast Asian breeds (n = 23 comprising 14 Siamese mix, 4 Siamese, 2 each of Tonkinese and Snowshoe mix, and 1 Burmese mix); and Abyssinian breeds (n = 15 comprising 7 Abyssinians, 4 Bengals, 2 Abyssinian mix, and 1 each of Somali and Bengal mix). There were 115 male neutered cats, 101 female neutered cats, 26 female intact cats, 25 male intact cats, 3 female cats of unknown neuter status, 1 male cat of unknown neuter status, and 2 cats of unknown sex or neuter status. The mean (±SD) age of all cats was 5.8 ± 5.3 years; range 0.1–19.0 years; however, the exact age was unknown for 6 of the shelter cats. One hundred and eighty-six of the cats were housed exclusively indoors, 59 of the cats were either exclusively or partially housed outdoors, and the housing status was unknown for 28 cats. Two hundred and sixteen cats were fed a variety of commercial canned or kibble diets, and 3 cats were fed raw-meat diets. Of the 219 diarrheic cats, 93 (42%) were from internists, 72 (33%) were from general practitioners, and 54 (25%) were from shelters. The mean age for diarrheic cats evaluated by internists was 8.6 ± 5.9 years; range 0.1–19.0 years, by general practitioners was 5.1 ± 5.2 years; range 0.2–17.0 years, from shelters was 2.2 ± 3.0 years; range 0.1–11.0 years, and from the nondiarrheic control group was 5.3 ± 3.3 years; range 0.5–11.0 years. The cats evaluated by internists were significantly older than the other 2 diarrheic groups of cats, and the cats from the shelter population were significantly younger than the other 2 diarrheic groups (< .05).

The overall prevalence of each of the selected bacteria in the diarrheic and nondiarrheic cats was evaluated (Table 1). Campylobacter spp. was isolated from significantly fewer diarrheic cats (21/219 or 9.6%), compared with the nondiarrheic cats (15/54 or 27.8%, = .001). C. perfringens was also isolated from significantly fewer diarrheic cats (92/215 or 42.8%) compared with the nondiarrheic cats (34/54 or 63.0%, = .009); however, there was no significant difference in the isolation of C. difficile from diarrheic cats (3/215 or 1.4%) versus nondiarrheic cats (0/54, = 1), or in the isolation of P. shigelloides from diarrheic cats (6/219 or 2.7%), versus nondiarrheic cats (3/54 or 5.6%, = .39). There were no positive fecal culture results for Salmonella, Shigella spp., Aeromonas spp., and Edwardsiella spp. in any of the cats from both the diarrheic and nondiarrheic groups. C. perfringens CPE was detected in 9/219 diarrheic cats (4.1%) and in 1/54 nondiarrheic cats (1.9%, = .69), whereas TcdA was not detected in any of the diarrheic or nondiarrheic cats.

Table 1. Prevalence of bacterial enteropathogens in diarrheic and apparently healthy cats.
OrganismDiarrheicNondiarrheicP Value
Campylobacter spp. (culture)9.6% (21/219)27.8% (15/54).001
Clostridium perfringens (culture)42.8% (92/215)63.0% (34/54).009
C. difficile (culture)1.4% (3/215)0% (0/54)1
C. perfringens enterotoxin4.1% (9/219)1.9% (1/54).69
C. difficile toxin A0% (0/219)0% (0/54)1
Pleisiomonas shigelloides2.7% (6/219)5.6% (3/54).39

There was no significant difference in the prevalence of any of the intestinal parasites studied in the diarrheic versus nondiarrheic cats, and the overall prevalences were low; however, all of the cats that tested positive for 1 or more parasites were diarrheic, with the exception of 1 nondiarrheic cat that tested positive for Giardia via ELISA (Table 2). The prevalence of enteropathogens in the nondiarrheic control population was compared with the prevalence of enteropathogens in each of the diarrheic subgroups (internist, general practitioner, and shelter cats). The diarrheic cats evaluated by internists had significantly lower isolation rates of C. perfringens (diarrheic 31/91 or 34.1%; nondiarrheic 34/54 or 63.0%; = .001) and Campylobacter spp. (diarrheic 4/93 or 4.3%; nondiarrheic 15/54 or 27.8%; = .001) compared with the nondiarrheic control cats. In contrast, the diarrheic cats evaluated by general practitioners had a significantly higher incidence of Giardia (diarrheic 9/72 or 12.5%; nondiarrheic 1/54 or 1.9%; = .043) and T. foetus (diarrheic 7/68 or 10.3%; nondiarrheic 0/53 or 0%; = .018) compared with the nondiarrheic cats.

Table 2. Prevalence of parasitic enteropathogens in diarrheic and apparently healthy cats.
OrganismDiarrheicNondiarrheicP Value
Tritrichomonas foetus (culture)2.9% (5/170)0% (0/53).59
T. foetus (polymerase chain reaction [PCR])4.0% (5/125)0% (0/6)1
Cryptosporidium (ELISA)3.2% (6/190)0% (0/54).34
Giardia (ELISA)8.2% (18/219)1.9% (1/54).14
Isospora spp.3.7% (8/215)0% (0/54).36
Toxocara spp.0.9% (2/215)0% (0/54)1

Fecal cultures for Campylobacter spp. were performed in all 273 cats enrolled in the study and 36 cats (13.2%) were positive (Table 1). Of these 273 cats, 131 had 4 real-time PCR assays performed to test DNA extracts of samples for the presence of 4 strains of Campylobacter spp. (C. jejuni, C. lari, C. upsaliensis, and C. helveticus). Of the 131 samples, 74 were PCR positive for Campylobacter spp. (56.5%). Eleven of the 131 fecal specimens that had PCR performed were positive for Campylobacter spp. via culture, and all specimens were diarrheic. Ten of the 11 cats (90.9%) that were culture positive for Campylobacter spp. were also positive on PCR. In addition, 64/131 cats (48.9%) were PCR positive, but culture negative for Campylobacter spp. None of the samples were positive for C. lari, 5/74 were positive for C. jejuni (6.8%), all 74 cats were positive for C. helveticus (100%), and 33/74 were positive for C. upsalensis (44.6%). Thirty-five of 74 cats (47.3%) had more than 1 species identified on PCR. PCR testing for Campylobacter spp. was positive in 69/125 of the diarrheic cats (55.2%), and positive in 5/6 nondiarrheic cats tested (83.3%). Twenty-nine of 40 diarrheic cats (72.5%) from the general practitioner group were positive for Campylobacter spp. via PCR, 25/56 diarrheic cats (44.6%) from the internist group were positive for Campylobacter spp., and 15/29 (51.7%) diarrheic cats from the shelter group were Campylobacter spp. positive. Diarrheic cats evaluated by general practitioners were significantly more likely to be positive for Campylobacter spp. compared with cats in the internist and shelter groups (= .022). Of the 5 cats that were positive for C. jejuni, 1 was nondiarrheic and 4 were diarrheic. The latter 4 diarrheic cats were from the shelter group (n = 3) and general practitioner group (n = 1). All 5 cats that were positive for C. jejuni were also positive for other species (C. helveticus alone or C. helveticus and C. upsalensis).

Of the 273 cats in the study, 223 had T. foetus culture performed (170 diarrheic and 53 nondiarrheic cats), and 5 diarrheic cats were culture positive for T. foetus (2.9%). In addition, T. foetus was detected via PCR in 5/125 diarrheic cats (4.0%) and in 0/6 healthy, nondiarrheic cats tested. Overall, 9 cats were positive for T. foetus on fecal culture or PCR, and 1 cat was positive on both PCR and fecal culture. All of the T-foetus-positive cats were diarrheic, and none of these cats was in the shelter group.

Sixty-nine of 219 diarrheic cats (31.5%) had blood in the feces on gross inspection, and 62 of these cats (89.9%) had evidence of hematochezia only. In the remaining 7 cats, 2 had evidence of melena only, 2 had evidence of melena and hematochezia, and records from 2 cats lacked clarification on the appearance of the blood. None of the nondiarrheic cats had blood in the feces, whereas 10 of 21 Campylobacter-positive diarrheic cats (47.6%) had blood in the feces. There was no significant association between isolation of Campylobacter spp. via culture in diarrheic and nondiarrheic cats and detection of blood in the feces (= .14), and only 2 of 5 diarrheic cats (40%) that were positive for C. jejuni via PCR had evidence of hematochezia. In addition, 27 of 44 diarrheic cats (61.4%) positive for C. helveticus and 15 of 33 diarrheic cats (45.5%) positive for C. upsaliensis had blood in the feces. Five of 8 cats (62.5%) infected with T. foetus had evidence of hematochezia (= .02). C. perfringens enterotoxin (CPE) was detected in 10 cats (9 diarrheic cats and 1 nondiarrheic cat), and 2 of the 9 diarrheic cats (22.2%) had clinical evidence of colitis. There was no significant association between detection of CPE and evidence of hematochezia (= .31).

Assessment of the effects of any previous or current medical treatment on the outcome of diagnostic testing was performed. Medical treatment specifically focused on administration of antimicrobials given within 2 weeks of fecal testing or administration of anthelminthics within the prior 6-month period. Administration of antimicrobials and anthelminthics were not associated with detection of fecal enteropathogens with the exception of C. perfringens and Giardia, where significantly more of the culture negative cats received antimicrobial treatment (< .001) or anthelminthic and antimicrobial treatment (= .022).

There was no significant association between diarrhea and age (= .30). Male intact cats had a significantly higher prevalence of diarrhea compared with male neutered cats (= .0069) and female spayed cats (= .0051). Male intact cats did not have a significantly higher prevalence of diarrhea than female intact cats (= .49) because almost all had diarrhea. For breed, there was no significant difference in diarrhea prevalence between Southeast Asian and European breeds (= .91), between Abyssinian and European breeds (= .080), or between Abyssinian and Southeast Asian breeds (= .27). When evaluating prevalence factors, younger cats were found to be significantly more likely to be infected with Giardia (mean age of positive cats 1.8 ± 2.5 years; mean age of negative cats 6.1 ± 5.3 years; < .001), Cryptosporidium (mean age of positive cats 0.4 ± 0.0 years; mean age of negative cats 5.6 ± 5.2 years; < .001), T. foetus (mean age of positive cats 0.7 ± 0.3 years; mean age of negative cats 5.4 ± 5.0 years; < .001), and Isospora (mean age of positive cats 0.6 ± 0.3 years; mean age of negative cats 6.0 ± 5.3 years; < .001). Abyssinian breeds comprising Abyssinians, Somalis, and Bengals, either purebred or crossbred, were found to be significantly more likely to be positive for T. foetus (= .007). Intact male cats were found to be significantly more likely to be positive for Isospora (= .036).

Discussion

This study demonstrated that there was no significant difference in the prevalence of intestinal parasites between the diarrheic and nondiarrheic cats. This is likely a consequence of the relatively small number of cats that were ultimately diagnosed with intestinal parasites, the fact that various cofactors and host factors likely play an important role in parasite-associated diarrhea,[11] and the disproportionate number of samples from cats evaluated by internists creating a potential bias in the lack of parasites detected from this population. These differences probably had an impact on the results of the fecal testing, as younger cats are more likely to be infected with parasites whereas middle-aged to older cats may be more likely to have chronic inflammatory or neoplastic disorders such as inflammatory bowel disease, intestinal lymphoma, and hyperthyroidism causing diarrhea.[12, 13] Another important limitation of this study was that the healthy control cats that were studied came from a different population than the cats with diarrhea. Determination of epidemiological factors associated with diarrhea would have been optimized if the healthy cats had originated from the same hospitals as the diarrheic cats.

Isolation of enteropathogenic bacteria was relatively uncommon in both diarrheic and nondiarrheic cats; however, isolation of Campylobacter spp. was significantly more common in nondiarrheic cats, underscoring the fact that many of the Campylobacter spp. isolated from cats are nonpathogenic.[14] PCR-based testing was significantly more sensitive than fecal culture in detecting Campylobacter spp., and only 13.2% of cats were positive for Campylobacter spp. via fecal culture versus 56.5% via PCR. It has been well documented that biochemical and phenotypic characteristics of Campylobacter spp. in cat feces are insufficient to characterize the infection, and molecular-based testing allows differentiation of enteric Campylobacter spp. and Helicobacter spp., and also allows identification of multiple Campylobacter spp. in individual animals.[7, 15] This is underscored by the finding of almost 50% of infected cats having more than 1 species of Campylobacter identified on PCR. Speciation of Campylobacter via PCR is important for identification of potentially pathogenic species, particularly in situations where enteric zoonotic infection is a concern.[16] Commercial reference laboratories are increasingly utilizing real-time PCR technology on feces that allows rapid detection of DNA from Campylobacter spp. within 1–3 days, and facilitates differentiation of Campylobacter spp. including C. jejuni, C. coli, C. helveticus, and C. upsaliensis. The test utilizes a closed-tube system that minimizes the risk of laboratory contamination.

The prevalence of Campylobacter jejuni (6.8%) in this study is similar to that reported in purpose-bred cats obtained from several commercial sources (6.0%),[7] but higher than that reported in privately owned cats and cats from shelters (1.8 and 0%, respectively).[5] A likely reason for this difference is that PCR was performed in the former study and fecal culture with biochemical testing was performed in the latter study. Fecal culture for detection of Campylobacter spp. is also limited in light of the high prevalence of Campylobacter spp. in healthy, nondiarrheic cats, and the results of this study were supported by those recently published by Sucholdoski et al who documented a higher prevalence of Campylobacter spp. in healthy cats (22/33 or 66.7%) compared with diarrheic cats (42/87 or 48.3%).13 The same study documented a relatively high prevalence of C. helveticus in both healthy (66.7%) and diarrheic cats (68.0%) based on PCR. All 74 cats that were PCR positive for Campylobacter spp. in this study were positive for C. helveticus, and the pathogenicity of this species is questionable given its high prevalence in nondiarrheic cats. A limitation of this study was that there were very few nondiarrheic cats (n = 6) that were tested for Campylobacter by PCR, although fecal culture for Campylobacter spp. was performed on all 54 nondiarrheic cats. Caution should be heeded in using these data to compare Campylobacter spp. in diarrheic versus nondiarrheic cats given the small number of nondiarrheic cats tested for Campylobacter by PCR.

Results of the present study support prior presumptions that C. perfringens is of little clinical importance in cats. Isolation of C. perfringens from diarrheic cats in this study (43%) was similar to the results previously documented in privately owned diarrheic and healthy cats,[4] but substantially lower than that documented in diarrheic dogs (>80%).[17] The limitations of relying upon isolation of C. perfringens results alone are underscored by the higher prevalence of C. perfringens found in the nondiarrheic cats (63%). In contrast to dogs, in which CPE was detected in approximately 34% of diarrheic animals,[17] CPE was only detected in 4.1% of the diarrheic cats and 1.9% of the nondiarrheic cats. In addition, only 22% of the cats that were positive for CPE had clinical evidence of colitis, and these findings parallel those in dogs in which C. perfringens-associated diarrhea is more commonly associated with enteritis or enterocolitis signs.[17] The pathogenicity of C. perfringens in cats remains poorly understood, and warrants further study. No other studies documenting the prevalence of CPE in healthy and diarrheic cats has been published to date.

Clostridium difficile was isolated from only 3/215 (1.4%) of cats, and all of these cats had diarrhea. There is little information regarding C. difficile infection (CDI) in cats, with only a single report of disease in 2 household cats.[18] Whether this indicates a lower susceptibility to disease, less frequent exposure, or underreporting because of less frequent testing remains unclear. Clostridium difficile was isolated from 3/42 cats (7.1%) hospitalized in an intensive care unit, although no mention was made of whether the cats had diarrhea or not.[19] All cats in this study were negative for C. difficile TcdA via ELISA, which makes it difficult to attribute the cause of diarrhea in the 3 culture positive cats to C. difficile. Considerations for culture positive isolates that are negative for TcdA on ELISA include the isolation of a nontoxigenic strain, false-negative ELISAs because of the use of species-specific and insensitive immunoassays,[20] and detection of TcdA-negative, TcdB-positive genotypes that has been found in people, horses, and dogs.[21-23] CDI appears more common in dogs compared with cats, based on detection of TcdA in approximately 12.5% of diarrheic animals.[17] All cats were negative on culture for Salmonella, although only 1.3% of the cats were ingesting raw-meat diets, and most cats were housed exclusively indoors, eliminating predation as a source of exposure. A limiting factor in the assessment of the effects of previous or current medical treatment was that information pertaining to dose and duration of any treatment was lacking for many of the cats, and suboptimal dosing or duration of treatment could certainly have played a role in these results. In addition, it is plausible that cats that were successfully treated for Giardia 5–6 months previously could have become reinfected close to the time of fecal submission.

In conclusion, this study documented a low prevalence of Salmonella, C. perfringens, and C. difficile in diarrheic and healthy cats, and the isolation rate for Clostridial spp. was similar in both groups of cats. The fecal immunoassays used for detection of C. difficile and C. perfringens toxins are all human-based immunoassays, and it is plausible that the lack of detection of C. difficile TcdA and the low incidence of C. perfringens CPE was compounded by the lack of validated immunoassays in these cats, similar to our experience in the dog.[17, 20] There are currently no commercially available species-specific immunoassays for detection of enteropathogenic bacterial toxins in cats and dogs, and all reference laboratories rely upon human-based immunoassays for detection of these toxins. Campylobacter spp. was isolated from significantly fewer diarrheic cats compared with nondiarrheic cats, and PCR testing was superior to fecal culture for detection and identification of Campylobacter species. A limitation of this study was that PCR testing for Campylobacter spp. was only performed in 48% of the cats, and it is plausible that the prevalence of C. jejuni would have been higher than 6.8% if more cats were examined via PCR.

Acknowledgments

The authors acknowledge the technical support and expertise of Mary McIlwain at IDEXX Reference Laboratories, West Sacramento, CA, and the invaluable input provided by Drs Jane Robertson and Christian Leutenegger, IDEXX Reference Laboratories, West Sacramento, CA.

Footnotes

  1. 1

    Parapak culture & sensitivity (catalog # 900612), Meridian Biosciences, Inc, Cincinatti, OH

  2. 2

    Tritrichomonas foetus Feline InPouch™, Biomed Diagnostics, Inc; White City, OR

  3. 3

    Remel ProSpecT Giardia Microplate ELISA, Thermo Fisher Scientific, Remel Products, Lenexa, KS

  4. 4

    Remel ProSpecT Cryptosporidium Microplate ELISA, Thermo Fisher Scientific, Remel Products

  5. 5

    Clostridium difficile ColorPAC Toxin A test, Becton Dickinson Diagnostics, Sparks, MD

  6. 6

    Clostridium perfringens enterotoxin test, Techlab Inc, Blacksburg, VA

  7. 7

    Campy CVA Agar, Hardy Diagnostics, Santa Maria, CA

  8. 8

    Anaeropack Systems, Thermo Fisher Scientific, Remel Products

  9. 9

    Biological Media Service, University of California, Davis, CA

  10. 10

    Qiagen, Inc, Valencia, CA

  11. 11

    Roche Applied Science, Indianapolis, IN

  12. 12

    SPSS Statistics Professional, IBM Corporation, Armonk, NY

  13. 13

    Sucholdoski JS, Gossett NM, Aicher KM, et al. Molecular assay for detection of Campylobacter spp. in canine and feline fecal samples. ACVIM forum, 2010 (abstract)

Ancillary