Data from a survey of Clostridium perfringens and Clostridium difficile shedding by dogs and cats in the Madrid region (Spain), including phenotypic and genetic characteristics of recovered isolates

This article contains information related to a recent survey of the prevalence of fecal shedding of Clostridium perfringens and C. difficile by dogs and cats attended in veterinary clinics located in the Madrid region (Spain). Specifically, we provide detailed information about the clinics that participated in the survey, the demographic and clinic characteristics of recruited animals and the genetic and phenotypic characteristics (including antimicrobial susceptibility data), of recovered bacterial isolates.


Specifications
Dogs and cats attended in veterinary clinics, and the Clostridium perfringens and C. difficile isolates obtained from their feces Experimental features Analysis of general data about participating clinics, and the demographic and clinical features of recruited animals; genetic and phenotypic profiling of isolates Data source location

Universidad Complutense de Madrid, Madrid, Spain
Data accessibility Data is provided with this article

Value of the data
First detailed analysis of the prevalence of Clostridium perfringens and Clostridium difficile shedding by small animals (dogs and cats) in the Madrid region (Spain).
Detailed phenotypic and genetic data of recovered isolates is provided, which may be useful for comparison in future epidemiological surveys.
Given the recent emergence of antibiotic-resistant strains of C. difficle, information on the antimicrobial susceptibility profiles of the isolates obtained in this survey may be particularly valuable.

Data
The data shown in Section 1.1 of this article provide detailed information on the veterinary clinics that participated in a recent survey of the prevalence of fecal shedding of Clostridium perfringens and C. difficile by dogs and cats which was carried out in the Madrid region (Spain) [1]. Furthermore, the demographic and clinical features of recruited animals are detailed in Section 1.2, and Section 1.3 provides extensive data on the genetic and phenotypic characteristics of recovered bacterial isolates.

General data about participating clinics
An overview of the 17 veterinary clinics that participated in the study (hereafter referred to as clinics A to Q) is provided in Table 1. Two clinics (L and P) did not return a questionnaire of general data about their centre (see Section 2) and in two other cases (clinics H and K) the returned questionnaire was incomplete. Participating clinics were scattered within the Madrid region (14 were located in the capital city, two in other municipalities within the metropolitan area and one in a rural location) and varied widely in their year of opening (from 1981 to 2014), number of cases attended per week (x 7S.D. ¼37.6 7 18.7 and 16.5 7 11 for dogs and cats, respectively), number of fecal cultures requested per week (1.6 72.4 and 0.9 71.8), and other parameters (Table 1). These clinics also differed in the antibiotics used for the treatment of diarrhea, but 12 of them (80% for which pharmacological data were available) reported the use of metronidazole for the treatment of these conditions. Only three clinics (20%; F, J and O) acknowledged frequent request of microbiological culturing for anaerobes, and five clinics (33.3%; D, F, H, N and O) reported occasional suspicion of C. difficile and/or C. perfringens involvement in severe cases of diarrhea.

Demographic and clinical features of recruited animals
The demographic characteristics of recruited animals are summarized in Table 2 and Figs. 1 and 2. A total of 142 animals, including 105 dogs and 37 cats (73.9% and 26.1% of total, respectively; Fig. 1A) of diverse breeds (Fig. 2), were recruited for the study. The male/female ratio of animals varied widely , the overall number of animals is indicated between parentheses. B) Bar plot of the overall age distribution of dogs (n¼ 105; green bars) and cats (n¼ 37; orange bars) recruited for the study. C) Pie charts of the overall sex distribution of dogs (n¼ 105) and cats (n ¼37) included in the study. Blue and pink sectors represent male (M) and female (F) subpopulations, respectively. D) Bar plot of the antimicrobial treatments administered to recruited dogs and cats (green and orange bars, respectively) r30 days before sampling. E) Bar plot of recent pharmacological treatments administered to recruited dogs and cats (green and orange bars, respectively) r 30 days before sampling. Abbreviations: ACE, angiotensin-converting-enzyme; NSAIDs, nonsteroidal anti-inflammatory drugs; PPIs, proton pump inhibitors. In panels D and E, pharmacological treatment data were not available for a total of 48 animals (32 dogs and 16 cats). among clinics, with the overall values for dogs and cats being similar (56.2%, 43.8% and 56.8%, 43.2%, respectively; Fig. 1C). The age distribution of sampled animals also showed ample variation among clinics, but the overall values were similar for the dog and cat subpopulations: 20%, 45.7%, 32.4% of dogs and 10.8%, 54.1%, 32.4% of cats had o 1 year, 1-6 years and Z 7 years, respectively ( Table 2).
The overall proportion of dogs and cats with diarrhea on the sampling date were very similar (13.3% and 13.5%, respectively), and in both cases most animals had not suffered any episode of diarrhea within the preceding 30 days (61% and 75.7%, respectively) ( Table 2). Only 24.7% (18/73) of dogs and 14.3% (3/21) of cats for which medication data was available were under antibiotic treatment on the sampling date or within the previous 30 days, with metronidazole and amoxicillin ranking first and second, respectively (Fig. 1D). Other pharmacological treatments of sampled animals are shown in Fig. 1E.
The demographic data and clinical features of animals yielding positive fecal cultures for C. perfringens and/or C. difficile are detailed in Table 3 (see also Álvarez-Pérez et al. [1]). Table 3 includes an overview of the genetic and phenotypic characteristics of the bacterial isolates obtained from recruited animals. Additionally, the toxinotypes, PCR ribotypes (only for C. difficile isolates), amplified fragment length polymorphism (AFLP) genotypes and antimicrobial susceptibility profiles of C. perfringens and C. difficile isolates are detailed in Tables 4 and 5, respectively.

Experimental design, materials and methods
Our survey was performed during one week (from November 24 to December 1, 2015) in a total of 17 primary care veterinary clinics from the Madrid region (Spain). The staff of participating clinics received training for data and sample collection, and email and telephonic support was available throughout the duration of the study. Veterinarians of participating centers were asked to select two swab samples of all feces shed by dogs and cats at their clinic, regardless of the age, origin and clinical condition of the animals, and to send those samples to a central reference laboratory at the Faculty of Veterinary Medicine of Complutense University of Madrid. Additionally, the staff of each participating clinic had to complete a questionnaire of general data about the centre and a second questionnaire for each pair of fecal swabs obtained requesting data on the sample (collection date, consistency of feces and presence of blood) and the animal of origin (species, breed, sex, age, clinical status and episodes of diarrhea and medication(s) within the previous 30 days). An informed consent and agreement to participate in the study was obtained from the owners of each animal before enrolment. Animals    Fig. 1E. e AFLP genotypes were arbitrarily designated by a lower case 'p' followed by a number (see Table 4). In vitro resistance to benzylpenicillin (PEN), erythromycin (ERY), imipenem (IPM), levofloxacin (LVX), linezolid (LZD), metronidazole (MTZ) and/or tetracycline (TET) is indicated between parentheses. The number of isolates belonging to each strain type (when different to one) is shown between square brackets. Dashes mean that C. perfringens was not isolated from the corresponding animal. NA: not analyzed (a single fecal swab was available for some animals, and this was used for C. difficile testing). f In this case, ribotype (RT) and AFLP fingerprinting information is included. AFLP genotypes were arbitrarily designated by a lower case 'd' followed by a number (see Table 5). In vitro resistance to benzylpenicillin (PEN), clindamycin (CLI) and/or erythromycin (ERY) is indicated between parentheses. Additionally, all C. difficile isolates displayed resistance to levofloxacin and imipenem. The number of isolates belonging to each strain type (when different to one) is shown between square brackets. RT?: unknown ribotype. Dashes mean that C. difficile was not isolated from the corresponding animal. g These animals were sampled twice during the study period.   Table 3). Daggers (and the number 2 after the clinic's code) indicate isolates that were obtained in the follow-up analysis. b cpe: possession of the gene encoding for enterotoxin. cpb2: possession of the gene encoding for β2 toxin, with asterisks indicating atypical forms of the gene (as determined by PCR amplification). c According to the UPGMA dendrogram shown in Fig. 1 Table 3). b Toxin profiles: RT009-like, A-B-CDT-(but with a positive PCR result for tcdB); RT106, A þB þ CDT-; RT154, A þB þ CDT-; RT? (unknown ribotype), A-B-CDT-. c According to the UPGMA dendrogram shown in Fig. 2  were always handled by experienced veterinary practitioners in strict accordance with good animal practice and the Spanish legislation. The owners of animals yielding a positive culture for C. difficile and/or C. perfringens were invited to participate in a follow-up survey performed four months after the first study (in March 2016). In this case, fecal swab samples and clinical information of animals was obtained as explained above.
The microbiology procedures used for C. perfringens and C. difficile isolation from fecal samples, and the methods used for toxin profiling, PCR ribotyping, AFLP subtyping and in vitro antimicrobial susceptibility testing of recovered isolates are detailed in our previous publication [1].

Funding sources
This work was supported by grant AGL2013-46116-R from the Spanish Ministry of Economy and Competitiveness. The funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.