Breeder-reported patterns of antimicrobial use and point prevalence of methicillin-resistant Staphylococcus spp. among breeding bitches in the southwestern United States

Abstract

Methicillin-resistant Staphylococcus pseudintermedius skin infections are an increasing concern in veterinary medicine, especially when found in juvenile dogs with no prior antimicrobial exposure. The objectives of this study were to assess the prevalence of methicillin-resistant staphylococci in breeding bitches and survey antimicrobial administration by their breeders. A total of 17 breeders and 54 bitches were included. Bitches were housed in diverse environments throughout the Phoenix, Arizona, USA metropolitan region. Nasal and oral gingival swabs were submitted for selective culture. Methicillin-resistant Staphylococcus aureus was not present, while methicillin-resistant Staphylococcus pseudintermedius was found in only 1/54 (1.9%) dogs. Survey results revealed that 16 bitches, which did not include the 1 MRSP-positive dog, had received antimicrobials within 6 months prior to sampling. Regardless of the low point prevalence identified, veterinarians and breeders should be cognizant of risks for bacterial resistance with the overuse of antimicrobials.

Introduction

The use of antimicrobials during canine pregnancy is a reported controversial practice among dog breeders that has been debated both within the veterinary medical community and among dog breeders. Prophylactic antimicrobial use occurs with the intention of decreasing the vaginal bacterial load within a breeding bitch in the hopes that this will increase chances of conception as well as the viability of the neonates (1,2). Antimicrobial use recommendations are disseminated widely in online breeder forums and are perpetuated commonly through word-of-mouth. However, data supporting this practice are currently lacking.

Staphylococci are common mucosal commensals and opportunistic pathogens, which are transmitted from the bitch to the puppies immediately at birth and persist as the same clone in the offspring for up to 48 mo after separation from the dam (3,4). The persistence of staphylococci among canine breeding populations, and the possible persistence of resistant clones highlight the importance of antimicrobial stewardship in the breeding community (3,5).

Methicillin-resistant (MR) Staphylococcus spp. have been isolated with increasing frequency worldwide in recent years and use of antimicrobials is an important risk factor for isolation of methicillin-resistant Staphylococcus spp. (6–11). Resistance to methicillin in Staphylococcus spp. is typically due to the presence of the mecA gene, which encodes for an alteration of the penicillin-binding protein 2a (PBP2a) and decreased affinity for beta-lactam antimicrobials (5). With coagulase-positive staphylococci (e.g., S. pseudintermedius, S. aureus) being both commensal and opportunistic pathogens in most species, concern has grown that inappropriate use of antimicrobials by dog breeders could serve as a potential stepping-point for multi-drug resistant infections among juvenile and adult animals.

A previous study of milk and colostrum swabs taken from healthy breeding female dogs reported significantly higher numbers of isolated Staphylococcus spp. from both the bitches and puppies that were multi-drug resistant, methicillin-resistant, and mecA positive in dog kennels with perceived excessive antimicrobial use (12).

The objectives of this study were to investigate point prevalence of coagulase positive Staphylococcus spp. in breeding bitches among a subset of breeders within the Phoenix, Arizona metropolitan area and to evaluate their antimicrobial use practices.

Materials and methods

Animals

Breeders’ contact information was obtained through their registration located on the American Kennel Club website (www.AKC.org). A total of 86 breeders listed within 100 miles of the Phoenix, Arizona metropolitan area were contacted by both phone and e-mail when available and invited to participate in the study. Detailed information regarding the study was posted on the Canine Health Foundation website (www.akcchf.org; Raleigh, North Carolina, USA) in an effort to recruit more breeders to participate. Original power calculations prior to starting this study indicated that within the population of 86 breeders listed, assuming at least 1 dog per breeder listed and a 95% confidence interval (CI), for a 50% response distribution, ideally 71 dogs would participate in the study if possible. Requirements for inclusion of bitches were as follows: a history of previous pregnancy or current pregnancy, no current superficial pyoderma at time of sampling based on examination by the primary author, and the breeding female being housed within 100 miles of the Phoenix, Arizona region. All dogs that fit the inclusion criteria from each breeder were included. Fifty-four dogs from 17 breeders housed in separate locations across the Phoenix, Arizona metropolitan area were included in the study. All bitches were in good health at the time of sampling with no clinical signs consistent with superficial skin infection, wounds, or erythema on examination.

Sampling

Before sampling, signed consent was obtained from each owner regarding use of their animals. Samples were collected by the primary author from the nostrils and oral gingiva of the mouth using sterile aerobic culture swabs (2,3). For the nasal and oral gingival testing, a single sterile swab was inserted into the proximal portion of both nostrils and again along the oral gingiva respectively, and rotated for approximately 5 s. Swabs were refrigerated and shipped to the microbiology laboratory within 24 h of collection.

Bacteriologic culture

Each swab was inserted into an enrichment broth that consisted of tryptone (10 g/L), sodium chloride (75 g/L), D-mannitol (10 g/L) and yeast extract (2.5 g/L), and incubated at 35°C for 24 h. Approximately 10 μL of broth from each test tube was then spread onto mannitol salt agar that contained oxacillin (2 μg/mL) and incubated at 35°C for 48 h. Bacterial colonies that were phenotypically consistent with Staphylococcus spp. were then subcultured onto Columbia blood agar that contained 5% sheep blood and incubated at 35°C for 24 h. Isolates were identified as staphylococci by colony morphology, Gram stain appearance, and positive catalase reaction. Tube catalase positive isolates were subjected to Staphylococcus aureus latex agglutination test (Pastorex Staph-plus; Bio-Rad, Mississauga, Ontario). Isolates that were negative on this test were subjected to a S. pseudintermedius-specific polymerase chain reaction (PCR) assay (13), which included both positive and negative controls. Methicillin resistance was confirmed using penicillin-binding protein 2a latex agglutination testing (MRSA latex agglutination test; Denka Seiken, USE, Campbell, California, USA). Methicillin-resistant S. pseudintermedius (MRSP) was typed by mec-associated dru typing (14,15). The dru repeats and types were compared to those assigned by use of the established dru database (www.dru-typing.org/search.php).

Survey

An online survey regarding antimicrobial administration, duration, date given, and source of antimicrobial was prepared using the online survey engine SurveyMonkey (Surveymonkey.com; San Mateo, California, USA) (Appendix 1). Only the primary author was aware of any identifying information which could link the results to a specific breeding kennel. Furthermore, breeders were assured that any identifying information would not be shared with other individuals or used in a publication. Participating breeders were notified by e-mail regarding the survey and were sent a reminder about completion every 2 d until the survey was completed. The questionnaire included a total of 14 closed, semi-closed, and open-ended questions. The questions were critically reviewed by all co-authors before sending to breeders participating in the study. Answers were recorded for each dog including previous antimicrobial use, duration of antimicrobials used, source of antimicrobials, type of antimicrobials, dose of antimicrobials used, how many courses of antimicrobials were used, whether antimicrobials were routinely used before insemination of bitches, antimicrobial use during pregnancy and postpartum, pregnancy status of the bitches, and how many active breeding dogs were in the facility. Breeder responses were recorded and linked to the aerobic culture results for each dog.

Results

Breeders

Seventeen breeders across the Phoenix, Arizona metropolitan area agreed to participate in the study. Housing for the dogs ranged from outdoor individual kennels (n = 2) to within the home with the owners (n = 10), or a mixture of both (n = 5). Ages of the dogs ranged from 10 mo to 10.5 y. At time of sampling, 5/54 (9.3%) dogs were spayed.

Fifty-four female intact dogs of varying breeds were included in the study. Twelve breeds were represented with the most common being cane corso (n = 12), German shepherd dog (n = 10), French bulldog (n = 6), Cavalier King Charles spaniel (n = 5), and golden retriever (n = 5). The total number of dogs per breeder ranged from 1 to 10 (median: 3). Four tested dogs were pregnant at the time of sampling.

MRSP colonization

Methicillin-resistant Staphylococcus pseudintermedius was isolated from 1 dog (1.9%), a 4-year old female intact golden retriever. The MRSP isolate was classified as dt11cg. This dru type is consistent with the dt cluster 11 which is a sequence type 68-associated group that is common in North America (16). This dog did not have any antibiotic exposure in the 6 mo prior to sampling.

Only 5 dogs had previous antimicrobial exposure in the 6 mo before sampling. Of these, only 3 had antimicrobial administration in the month prior to sampling; 2 dogs received amoxicillin/clavulanate potassium and 1 dog received cefpodoxime. Two of the dogs were housed in the same household, while the other 3 were all in separate home breeding locations. The low prevalence of MRSP precluded statistical evaluation.

Survey responses

Seventeen owners completed the questionnaire. Thirteen owners replied that they had never used antimicrobials during a bitch’s pregnancy (apart from treatment of identified infectious diseases), while 4 owners replied that they sometimes used antimicrobials. Drugs that were used by the breeders during their bitches’ pregnancy at any point consisted of amoxicillin (n = 2), amoxicillin and clavulanate potassium (n = 2), cephalexin (n = 2), and 1 each for enrofloxacin and cefpodoxime. All antimicrobials that had been used by breeders during parturition had been obtained from their veterinarian with most antimicrobials being used for a duration between 7 and 14 d.

Discussion

Concern for multidrug-resistant Staphylococcus spp. perpetuating in some breeding facilities has led to studies assessing Staphylococcus spp. carriage as well as the overall prevalence of methicillin-resistance in staphylococci from breeding dogs. Our study showed low point prevalence of MRSP (1.9%) within this population of dogs with survey responses indicating a low number of individual breeders reporting the use of antimicrobials perinatally.

In this study, regardless of a multitude of factors including prior antimicrobial exposure and housing conditions, only 1 multidrug-resistant isolate was found and no isolates were recovered for methicillin-resistant Staphylococcus aureus (MRSA). As these results gave us point prevalence only, this finding is difficult to compare directly to previous studies that discussed overall prevalence in set groups of individuals. A study by Rota et al (12) identified a high (7.9%) overall prevalence of mecA positive methicillin-resistant strains in the bitches tested, but it is difficult to compare this study with ours as the sample sources obtained from the breeding bitches varied as well as the selected housing environments. The samples in this study were acquired from a more diverse population of breeding bitches in their various environments, while the samples acquired by Rota et al (12) were obtained from breeding females within controlled breeding kennels with a history of MRSP positive cultures. Because of the potential for long-term shedding of MRSP (17), it is possible for MRSP to persist and be amplified within kennels. The point prevalence of MRSP in the present study (1.9%) was similar to the values that have historically been reported for a population of healthy dogs (18,19).

Due to the low point prevalence of MRS species and the lower numbers of participating dogs than required based on our power calculations, no statistically significant conclusions could be made in the current study. Previous studies by De Graef et al (20) found higher percentages of resistance in dogs housed in breeding kennels rather than individual houses. Indeed, the study by Rota et al in 2011 (21) demonstrated a high prevalence of MRSP present in separate kennels, although the kennels themselves varied in their antimicrobial exposure. While these studies were performed in different locations across the world, the concern for prevalence of MRS species transcends most borders.

Overall, most of the breeders responded in the survey portion of the study that they had been breeding for 20 or more years (n = 10). When questioned on antimicrobial use at different stages of a bitch’s pregnancy, 4 breeders replied that they had sometimes used antimicrobials while 13 replied that they had never used antibiotics during this time. This reply was obtained from both breeders who had been breeding for < 5 y and those who had been breeding for 20 y. The high number of breeders not using antimicrobials could be due to education on appropriate antimicrobial usage among the breeders themselves, or from personal experience on the effectiveness of antimicrobials during pregnancy in their breeding dogs. In the instances in which antimicrobials were used during pregnancy, the most commonly used antimicrobials were amoxicillin and amoxicillin combined with trihydrate/clavulanate potassium. These antimicrobials had all been obtained from a veterinarian; however, the reason for prescribing these medications was not within the scope of the online survey. Overall, the survey results showed a low level of antimicrobial usage by the breeders participating in this study. However, due to the low number of breeders participating in the study, no statistically significant conclusions could be made.

There are a few possible causes for most findings in this study. The varying, but overall infrequent, antimicrobial usage in dogs most likely played a strong role as previous antimicrobial administration has been proven to increase the risk of developing multi-drug resistance (12,14,22–25). Additionally, the dogs had various housing accommodations, which may have affected drug resistance and carriage. Self-selection bias may have also played a role in the lower number of positive findings, as breeders who obtain antimicrobials through non-traditional routes and use them repeatedly may have elected to not participate in a study on inappropriate antimicrobial use despite the promise of anonymity. While our study assessed MRSP and MRSA carriage as well as the presence of the mecA gene, we did not find a similar level of carriage even in the Loeffler et al study (26). Study bias including social desirability bias and demand characteristic bias may have influenced the study participants in becoming influenced by the study, as the very general theory of the study was known to the participants in order to encourage participation.

In addition to the selection bias described, other limitations included the low number of participants in the study preventing statistical significance from being achieved. Secondly, S. pseudintermedius colonization in dogs is not monoclonal with dogs typically carrying multiple different clones of S. pseudintermedius at any time (25). Therefore a finding of a single Staphylococcus spp. isolate is not truly representative of overall carriage; however, the intent of this study was restricted to the identification of methicillin-resistant Staphylococcus spp. Sampling more locations on each bitch could have possibly been more representative of true carriage, or using other screening methods to determine overall Staphylococcus spp. carriage; however, the decision was made to limit samples to the nares and oral gingiva to encourage owner participation.

Future studies could include a longitudinal study to assess persistence of MRSP carriage throughout breeding populations in multi-housed conditions similar to this study with potentially enough participation to achieve statistical significance. Assessment of persistence and transmission of MRSP isolated to newborn puppies may also be of benefit, as subjectively these authors have observed a higher prevalence of MRSP infections in progressively younger dogs in their clinical private practice. Additionally, assessment of bacterial populations with multiple groups of dogs in different geographical locations and over different time periods may be of benefit.

The results herein were suggestive of a similar prevalence of MRSP among these healthy breeding dogs as in the general population of healthy dogs. Overall self-reported antimicrobial use during pregnancy was also limited among this subset of breeders. Longitudinal studies involving multiple geographical locations are needed to better determine the overall incidence of MRSP among breeding dogs. Despite the need for further studies, continued prevalence of MRSP and MRSA cultures in clinically diseased animals, as well as the concern for breeding females serving as a reservoir for MRS species, should create awareness of appropriate antimicrobial use and the concern for MRSP as a pathogen for certain dog populations, namely neonates (4). CVJ

Appendix 1

Breeder Questionnaire

What is the name of your kennel/Your name? (This is used only to link your survey results with your culture results) Open answer How many active breeding bitches do you currently have in your program? Open answer How many of your dogs are currently pregnant? Please state their name and suspected stage of pregnancy. Open answer In the last 2 years, what is the average number of litters per dog per year in your program? Open answer For how long have you been breeding dogs? Open answer Do you use antibiotics routinely prior to insemination of your bitches? Always Never Sometimes Do you use antibiotics routinely at any time during a bitch’s pregnancy? Always Never Sometimes Do you use antibiotics routinely during a bitch’s postpartum period? Always Never Sometimes Which of the following antibiotics have you used during this time? Penicillin Amoxicillin Clavamox® (amoxicillin trihydrate/clavulanate potassium) Baytril® (enrofloxacin) Keflex® (cefpodoxime) Erythromycin Bactrim® (trimethoprim/sulfamethoxazole) I have not used antibiotics in my breeding bitches Other (please specify which antibiotic used) IF you have given a bitch antibiotics during any of her periods of pregnancy, where did you obtain them? Veterinarian Other breeders From an online store From other pets prescriptions I have not used antibiotics in my breeding bitches Other (please specify where you obtained them) If or when you use antibiotics in your breeding bitches, how is the dosage of antibiotics determined? From my veterinarian From other breeders From my own knowledge of antibiotics From internet sources (breeding sources, internet searches, etc) From the label of the antibiotic I do not use antibiotics in my breeding bitches Other (please specify how you determined dose) What is the typical duration of use if/when you do use antibiotics? Less than 7 days Between 7–14 days Between 15–21 days Greater than 21 days I do not use antibiotics in my breeding bitches How many courses of the antibiotic do you typically give? Open answer For each breeding female that has been/will be sampled for this study, please list their name, when they last were given antibiotics and what antibiotic it was to the best of your knowledge (approximate times are fine). This is to link their culture results with their recent antibiotic history Open answer