Abstract
Objectives
An outbreak of diarrhoea involving 16 cats at a cattery in Norway was investigated. Treatment and control of the outbreak were the primary objectives, but the effects of treatment on the antimicrobial resistance profiles of Escherichia coli isolated from faeces were also investigated.
Methods
Faecal samples were investigated for Giardia cysts by immunofluorescence microscopy, and multi-locus genotyping was performed to determine the Giardia genotype. Faecal E coli were assessed, before and after treatment for giardiasis, for antimicrobial resistance.
Results
The outbreak was probably caused by Giardia duodenalis, Assemblage F. Although infection was eliminated in most cats following treatment with fenbendazole, over 30% of the infected cats required a second treatment round (combined fenbendazole and metronidazole). Investigation of sensitivity to antibacterial drugs of E coli that had been isolated both prior to and following treatment demonstrated that fenbendazole treatment may select for resistant bacteria.
Conclusions and relevance
Controlling Giardia infections in dense cat populations can be challenging, and requires strict hygiene measures. In cases where fenbendazole alone does not result in treatment success, a combination treatment with fenbendazole and metronidazole may be effective. Although this study did not include untreated controls, we suggest that the potential for changes in gut microbiota and antimicrobial resistance development should be considered when choosing antiprotozoal drugs, particularly in cases of treatment failure and where repeat treatment is required.
Introduction
Giardia duodenalis is a unicellular intestinal parasite that can infect a range of different animals and humans. The genus consists of eight genotypes (A–F), of which two have been found in humans and other animals. The remaining six are considered to be more species specific, and felids are mainly infected with genotype F. 1 The prevalence of Giardia infection in cats in Europe and North America ranges from <1% to around 15%, and is more common in younger animals and shelter populations. 2 Nevertheless, outbreaks of giardiasis in catteries are seldom reported. The only cattery outbreak report that we could identify occurred more than 25 years ago and involved a period of chronic and intermittent diarrhoea in a group of 14 Persian cats; Giardia infection was eradicated by furazolidone treatment and good hygiene. 3 A more recent report of several Giardia infections at a cattery also involved infection with Tritrichomonas foetus (syn Tritichomonas blagburni), and it is unclear which of these parasites was most responsible for the signs. However, it would appear that the Giardia infection may have been of lesser clinical significance in that case. 4
Giardiasis may present treatment challenges among human patients, and post-infection changes in gut architecture and flora may result in the continuation of clinical signs despite successful treatment. 5 However, whether a similar situation occurs in animal infections has not been investigated. It has been speculated that intestinal microbiota may be of importance in evelopment of clinical signs in Giardia infections in both humans and animals. 6
Here, we describe an outbreak of diarrhoea most likely due to giardiasis at a cattery in Norway. Its control and effects on resistance patterns in intestinal Escherichia coli possibly associated with treatment are also discussed.
Material and methods
Cattery description
The outbreak occurred in a Norwegian cattery used for temporary housing of 26 cats. A non-profit organisation that arranges rehoming of cats runs the cattery with help from volunteers. Cats were fed a standard pellet diet and tap water was supplied ad libitum. Litter boxes with standard clumping cat litter were provided, with one litter box per three cats. The cattery does not use a quarantine system, but before cats are accepted, they have a routine health check, including spaying/castration, vaccination (feline panleukopenia virus, feline viral rhinotracheitis and calicivirus) and parasite control. Cats with diarrhoea are not accepted. None of the people in contact with the cats had gastrointestinal problems prior to or during the outbreak.
Outbreak detection and sampling
During summer 2012, a female cat that had entered the cattery 1 year previously developed diarrhoea and was diagnosed with giardiasis by faecal antigen detection at the local veterinary clinic. Within days, 15 other cats at the cattery developed similar clinical signs. The cats were subject to a general clinical evaluation, and faecal samples for parasitology and bacterial analysis were collected from the 16 cats presenting with diarrhoea on days 1 (first day of study), 5 (end of first treatment), 21 and 27 for parasitological analysis and bacterial analysis (for cats in which giardiasis was diagnosed) at the Norwegian University of Life Sciences (NMBU). Individual samples were collected either rectally with cotton swabs (sample size <1 g) or when cats were observed to defecate (sample size 3 g).
Parasitology analyses
Faecal samples (n = 16) were analysed at the Parasitology Laboratory, Norwegian University of Life Sciences, using standard procedures. Faecal samples were concentrated and examined by routine McMaster egg counting technique for helminth eggs and other protozoa. Subsamples were examined by immunofluorescent antibody test (IFAT; Aqua-Glo, Waterborne) for detection of G duodenalis cysts and Cryptosporidium species oocysts, and quantified using a semi-quantitative scoring method based on number of parasites per field of view at × 200 magnification (1+ = <10 cysts per field of view, 2+ = 10–50 cysts per field of view and 3+ = >50 cysts per field of view). Samples were also analysed for T foetus using a standard PCR protocol. 7
For samples in which Giardia cysts were detected, isolation of intact cysts was attempted by zinc sulfate flotation. DNA was isolated from cysts that had been successfully isolated (nine samples), using a commercial kit (QIAamp DNA Stool Mini Kit; Qiagen) and following the manufacturer’s instructions and PCR run using previously published primers for the following genes: triose phosphate isomerase (TPI), glutamate dehydrogenase (GDH), and β-giardin (BG) genes.8–10 Amplicons were sequenced in both directions by a commercial company, and the sequences obtained were assembled and compared with sequences in GenBank.
Bacterial analyses
Routine bacteriological identification protocols were used to describe intestinal E coli both prior to and following treatment in the Giardia-positive cats. 11 Sensitivity to penicillin, streptomycin, sulfamethoxazole with trimethoprim, cephalosporin, tetracycline, amoxicillin and enrofloxacin was tested by agar disc diffusion tests on two randomly selected E coli colonies per sample.
Quarantine and infection prevention measures
Following diagnosis, infected cats were confined together in housing that was separate from the cats that tested negative for Giardia, and quarantine was instigated: (i) no movement of animals in/out of the facility; (ii) no visitors; (iii) disposable shoe covers, gloves and mouth mask worn; (iv) daily disinfection of surfaces with a detergent containing ammonia, including feed and water bowls; and (v) increased frequency of cleaning.
Treatment
Infected cats were treated with fenbendazole (50 mg/kg PO q24h for 5 days). Cats with persistent diarrhoea and recurrent infections and/or infections refractory to treatment were then treated with a combination therapy of fenbendazole (50 mg/kg PO q12h) and metronidazole (25 mg/kg PO q12h) for 5 days. The dose rates are approximate; data on precise mg/kg doses for each cat are not available.
Results
Clinical inspection
At first examination, 16/26 cats (62%) were clinically ill. The predominant clinical sign was foul-smelling diarrhoea, but the cats also exhibited decreased appetite, abdominal pain, lethargy, stress and obvious discomfort. Body temperature was 38.1–38.6°C and capillary refill time was 2–3 s.
Parasitology results before treatment
Of the 16 cats with clinical signs, all were excreting large quantities (score of 3+) of Giardia cysts. Those cats not demonstrating clinical signs were apparently uninfected with Giardia (no cyst excretion detected). No other parasites were detected in any of the faecal samples.
Of nine Giardia isolates examined by PCR, nine (100%), seven (78%) and six (67%) were positive for the TPI, GDH and BG genes, respectively. All sequences obtained (TPI, GDH and BG) had 100% homology with Assemblage F sequences in GenBank (TPI: AB569402.1; GDH: AB569384.1; BG: AY647264.1).
Parasitology results and clinical presentation following treatment
After the first treatment regimen (fenbendazole), 11/16 cats stopped excreting Giardia cysts. However, five cats still had diarrhoea and continued shedding low numbers of Giardia cysts (score of 1+). Following the second treatment regimen of these five cats (fenbendazole and metronidazole), clinical signs ceased in all cats and Giardia cyst excretion could not be detected. None of the cats investigated on days 21 and 27 were excreting Giardia cysts.
Bacteriology results before treatment
Culture plates from faecal samples from cats with giardiasis were dominated by E coli. Very few lactic acid-producing bacteria were observed.
Bacteriology results following treatment
After the first treatment regime, a significant reduction in enterobacteriaceae was observed. Antimicrobial resistance of E coli was increased post-treatment for streptomycin, sulfamethoxazole with trimethoprim, tetracycline and amoxicillin; antimicrobial-specific resistance alterations in infected cats on days 5, 21 and 28 are described in Table 1. Although alterations seemed to be ameliorating by day 28, for some cats the alteration persisted.
Development of antimicrobial resistance in Escherichia coli following treatment regimens with fenbendazole in cats diagnosed with giardiasis
Number (%) of cats with resistant E coli isolates
Discussion
Here, we describe an outbreak of diarrhoea associated with giardiasis (G duodenalis, Assemblage F) at a Norwegian cattery that was successfully controlled by appropriate treatment regimens. The original source of infection could not be identified, but as the Giardia was Assemblage F, the infection source was probably feline. It is possible that one or more of the cats harboured asymptomatic infections that became symptomatic due to unknown factors or that were transmitted to other cats that then developed diarrhoea.
Two rounds of treatment were necessary for 5/16 infected cats. This treatment, together with effective quarantine (isolation of infected cats) and improved hygiene, apparently eliminated symptomatic infection from the cattery.
Avoidance of ingestion of medication may explain the persistence of giardiasis in five cats following the first treatment. However, lack of treatment success has been reported from human giardiasis when treatment compliance should probably be satisfactory. 12 Refractory giardiasis may also be due to various other factors, or these cases may represent reinfection with cysts remaining in the environment. It was not possible to determine why the infection was not eliminated from five of the cats by the first treatment round.
Development of antimicrobial resistance is rarely considered when selecting treatment for protozoan infections, but natural selection of resistant bacteria is clearly a possible side effect of antimicrobial therapy. Although fenbendazole has been reported to have no significant antibacterial activity, antimicrobial treatment of otherwise healthy animals has previously been demonstrated to result in increases in gram-negative bacteria and with higher resistance.13–16 Similarly, in this study, a single fenbendazole treatment apparently affected the antimicrobial resistance profile of the regular E coli of the faecal microbiota in all or nearly all the cats treated, and for some cats this lasted for more than 3 weeks. The antimicrobial resistance profile of E coli in the untreated cats was not tested. This represents a limitation of our study, and these results should be interpreted with caution.
Conclusions
In this paper, we describe the challenges regarding treatment and control of an outbreak of diarrhoea associated with giardiasis in a cattery in Norway. In addition, changes in gut microbiota associated with treatment were observed that lasted for a period of weeks. These results suggest that antimicrobial resistance development and the potential for changes in gut microbiota should be considered when choosing antiprotozoal drugs. This may be particularly important in cases of treatment failure where repeat treatment is required.
Footnotes
Acknowledgements
We thank the cattery personnel for assistance during the sampling, and Professor Henning Sørum, Norwegian University of Life Sciences, for his insights into development of antimicrobial resistance.
Funding
This study was financed by internal funds at the Norwegian University of Life Sciences, related to the PhD projects of Kristoffer Tysnes and Leon Cantas. Katrien Luyckx’s stay at the Parasitology Lab at NMBU was funded through the Erasamus Programme.
Conflict of interest
The authors declare that there are no conflicts of interest.