Ironing out the wrinkles and folds in the epidemiology of skin fold dermatitis in dog breeds in the UK

Skin fold dermatitis (intertrigo) is an inflammatory process of closely apposing skin surfaces. Extreme conformations towards folded skin in many dog breeds are linked with higher risk. Using anonymised primary-care veterinary data from the VetCompass Programme, this study aimed to report the frequency, demographic risk factors and clinical management for skin fold dermatitis in the UK. Risk factor analysis used random effects multivariable logistic regression modelling. From a study population of 905,553 dogs, the one-year period prevalence in dogs overall was 0.37% (95% CI 0.35–0.39). Diagnosis was supported by laboratory testing in 4.21% cases. Systemic antibiosis was used in 42.30% cases. Compared with crossbreed dogs, the most highly predisposed breeds were English Bulldog (odds ratio [OR] 49.07, 95% CI 37.79–63.70), French Bulldog (OR 25.92, 95% CI 19.62–34.26,) and Pug (OR 16.27, 95% CI 12.20–21.69). The most protected breeds were Yorkshire Terrier (OR 0.14, 95% CI 0.03–0.56), Border Collie (OR 0.31, 95% CI 0.11–0.84), Jack Russell Terrier (OR 0.53, 95% CI 0.30–0.92) and Labrador Retriever (OR 0.57, 95% CI 0.35–0.93). This study adds further evidence to the welfare concerns around high popularity of dog breeds with extreme conformations. The three breeds with by far the highest odds of skin fold dermatitis represent an extreme brachycephalic conformation.

The frequency of recorded locations for skin fold dermatitis differed widely between the breeds with high prevalence ( Table 2). Three breeds with extreme brachycephaly (English Bulldog, French Bulldog and Pug) showed high levels of facial, nasal, and periocular locations. In contrast, the lips were the dominant location in the spaniel breeds (English Cocker Spaniel and Cavalier King Charles Spaniel) and the West Highland White Terrier.
Laboratory testing was not recorded as supporting the diagnosis of skin fold dermatitis in 933/974 (95.79%) cases. From the 41 dogs with at least one laboratory test reported, the most common laboratory tests were bacterial culture and sensitivity (n = 18, 1.85% of total), skin scraping (12, 1.23%), tape strip (6, 0.62%) and swab with cytology (6, 0.62%).
The most used medical treatments to treat skin fold dermatitis in the 974 cases were systemic antibiosis (n = 412, 42.30%), antibacterial ± antifungal shampoo/cleanser (382, 39.22%), antibacterial ± antifungal ± Risk factors. All study variables were liberally associated with skin fold dermatitis in univariable logistic regression modelling and were evaluated using multivariable logistic regression modelling (Tables 1, 3 and  4). The final breed-focused multivariable model retained five risk factors: breed, age, sex-neuter and insurance (Fig. 2). Bodyweight relative to breed-sex mean was not associated with the odds of skin fold dermatitis and was not retained in the final model. No biologically significant interactions were identified. The final model was improved by inclusion of the clinic attended as a random effect (rho: 0.05 indicating that 5% of the variability was accounted for by the clinic attended, P < 0.001). The final model showed acceptable model-fit ( Five breeds had no recorded skin fold dermatitis cases. All older age groups had higher odds compared with dogs aged under one year, with odds rising as dogs aged. Neutered animals had higher odds than entire animals within both sexes. Insured dogs had 1.97 (95% CI 1.69-2.29) times the odds of skin fold dermatitis compared with uninsured dogs (Fig. 2).
As described in the methods, breed-derived variables were introduced individually to replace breed in the final breed-focused model. Compared with crossbred dogs, purebred dogs had increased odds (OR 2.54, 95% CI 2.06-3.13) of skin fold dermatitis. Four Kennel Club breed groups showed higher odds compared to breeds that were not recognised by the Kennel Club: utility, toy, gundog and working. Compared with breeds with long

Discussion
This study estimates a one-year prevalence of 0.37% for skin fold dermatitis in the general population of dogs under primary veterinary care in the UK. The study also highlights several breeds with high prevalence values. The wider canine welfare relevance from skin fold dermatitis arises from the high (and rising) popularity of some of these commonly affected breeds, with four of the five most frequently affected breeds also being amongst the 10 most commonly registered Kennel Club breeds in the UK 20 : French Bulldog (2.7% prevalence, Kennel Club Chinese Shar-Pei 9 2 (22.22) 0 (0) 0 (0) 1 (11.11) 1 (11.11) 0 (0) 6 (66.67)  However, despite these high breed-specific prevalence values reported here, it is likely that studies based on primary-care veterinary clinical records underestimate the true frequency of skin fold dermatitis because of systematic under-recording of observed problems in electronic records, with only 64% of observed problems being recorded into the practice EPR in one study 23 . Additionally, 37% of owners of English Bulldogs with skin disease of any type in Finland failed to identify these animals as having a health problem, and so therefore many affected dogs may not even be presented for veterinary care 24 . Under-recognition by owners of skin fold dermatitis as a clinical problem may be even higher than for other types of skin disease because the abnormal skin is generally concealed deep within the skin folds. Furthermore, pernicious effects from the normalisation phenomenon of breed-related issues (i.e., 'normal for breed') may result in not just reduced owner recognition of skin fold dermatitis as a clinical problem but could also reduce veterinary recognition or recording [25][26][27] .
Ultra-predispositions have been defined as predisposition with odds over 4 times higher compared with a broad comparator group 28 . In the current study, ultra-predispositions were shown for 8 breeds: English Bulldog, French Bulldog, Pug, Basset Hound, English Cocker Spaniel, Chinese Shar-Pei, Boxer and Cavalier King Charles Spaniel. It would be logical to prioritise review of the breed standards for each of these ultra-predisposed breeds to remove any wording that encourages harmful skin folding 29 . Welfare, veterinary and breed associations could also make greater efforts to inform the wider general public that, although humans may perceive folded skin as 'cute' , this conformation is instead a pathology that carries high welfare implications for affected breeds and individuals 27 .
The current study aimed to add information on breed protection, as well as on breed predispositions, to enable a more differentiated consideration of aetiopathogenesis. For instance, the current study highlighted contrasting frequencies of skin fold dermatitis for four breeds predisposed to atopic dermatitis but without significant skin folding 30,31 , with two breeds (West Highland White Terrier, Golden Retriever) predisposed to skin fold dermatitis while two were protected (Jack Russell Terrier, Labrador Retriever). Furthermore, breeds with both predisposed and protected statuses co-occurred within genetically close terrier and retriever clades 32 . www.nature.com/scientificreports/ Clearly, skin fold dermatitis is an aetiopathogenetically complex disorder, with much remaining to be learned about the aetiological pathways. At an individual patient level, most affected dogs (96.41%) had just one location of skin fold dermatitis recorded during the study period. Whilst this may be an accurate reflection of patient status, one of the author's (AH) referral practice experience suggests that, where skin fold dermatitis is detected in more than one location, it is often only the site of most concern that is recorded in the primary-care patient record. However, despite this limitation, the current study still identified majorly differing patterns of location between commonly affected breeds 29 . Among the three breeds with extreme brachycephalic conformations, English Bulldogs, French Bulldogs and Pugs, facial (facial, nasal and peri-ocular) locations of fold dermatitis were common; this could reflect the failure of muzzle skin to be reduced in proportion to the extreme brachycephalic facial skeleton. A high frequency of tail fold dermatitis was also recorded in English Bulldogs and French Bulldogs, two breeds that commonly show screw tails 33,34 . Conversely, high frequency of fold dermatitis in locations other than the face was shown www.nature.com/scientificreports/ in the Bassett Hound and Shar Pei, two breeds with breed standards that promote wide-spread wrinkling of body and extremities 29 . However, the high predilection of the lips as the location for fold dermatitis in affected Cocker Spaniels (93%) and West Highland White Terriers (77%) suggests that factors other than the extent of skin folding can contribute to this disorder because these two breeds show muzzle conformations that are broadly similar to protected breeds such as Labrador Retrievers, Jack Russell Terriers or Border Collies 29 . This opens speculation that factors promoting maceration, skin barrier damage and susceptibility to skin infection 35 , or concurrent inflammatory skin disease are relatively more important in these breeds than physical factors related to skin folding. The role of moisture as one such alternative risk factor may be represented by the Cavalier King Charles Spaniel, with 9% of cases in that breed recorded with periocular fold dermatitis. This breed has a mild brachycephalic conformation without significant facial folding 36 , but a high frequency of ocular disorders 37 . Excess tearing (epiphora) or use of eye treatments related to these ocular disorders may result in prolonged wetting of the shallow skin fold originating near the medial canthus and could promote skin maceration and consequent skin fold dermatitis. Compared with dogs aged under one year, the odds of diagnosis with skin fold dermatitis rose as dogs aged. The biological basis of this observation is not obvious. The limited information available on the effect of physiological ageing on skin structure and function in dogs would suggest that skin thickness, measured with calipers, reduces with age but not epidermal thickness or elasticity 38,39 , and that the bacterial community structure in healthy dogs did not differ by age 40,41 . A fuller understanding of geriatric skin changes is compounded by regional, individual, breed, sex, time, and disease effects on biophysical parameters 38,[40][41][42][43] . It is conceivable that chronic, if initially low grade, inflammation of skin folds results in chronic hyperplastic dermatitis characterized by dermal Figure 3. Results for risk factors that directly replaced the breed variable in the final breed-focused mixed effects multivariable logistic regression model (along with age, sex/neuter and insurance status). These results report associations between these risk factors and skin fold dermatitis in dogs under primary veterinary care in the VetCompass Programme in the UK. Clinic attended was included as a random effect. *CI confidence interval. www.nature.com/scientificreports/ and epidermal thickening through oedema, inflammatory infiltrates, glandular hyperplasia, and fibrosis. The progression of these structural and functional changes over time may precipitate or exacerbate chafing, secondary infection and thus increase the odds of diagnosis with aging. Long hair was protective from diagnosis with skin fold dermatitis in the current study. Whilst there may be an effect of hair coat type and length on friction, moisture retention or skin microbiome and thus predisposition to skin fold inflammation, at present this remains speculative. It is also possible that signs of skin fold dermatitis may be more easily detected in dogs with a short hair coat and therefore this may lead to a correspondingly higher rate of diagnosis.
The current results revealed that 96% of diagnoses of skin fold dermatitis were not accompanied by laboratory testing. This finding concurs with skin fold dermatitis being firmly regarded as a clinical diagnosis based on inflammatory skin lesions limited to the area of skin apposition [5][6][7] . In line with this view, where clinical signs were specifically recorded in this study, these reflected signs such as inflammation and associated malodour and pain that are detectable organoleptically. Pain was noted in nearly one in five dogs that had a record of clinical signs, underscoring the welfare relevance of skin fold dermatitis as a chronic condition. Whilst laboratory testing was uncommon in this primary-care study, this is recommended to rule out primary causes of dermatitis such as demodicosis, and to identify the presence of deeper infection or the presence and nature of the microbial overgrowth that could alter the therapeutic approach 5,6 . Although bacteria and Malassezia yeasts are considered to play a central role in skin fold dermatitis, their involvement is frequently limited to high populations on the skin surface 7 .
In the absence of a published evidence base for treatment of skin fold dermatitis in dogs, there is agreement on the general principles of treatment, if not the specific treatment modalities. Removal of surface microbial load, debris and exudation is considered a crucial treatment principle, along with resolution of tissue infection (if present), and glucocorticoid-aided control of inflammation is also deemed helpful for resolution 5,6 . For both approaches, topical treatment modalities are generally recommended. Acceptance and adoption of these principles is reflected in the current study where 84% of cases were prescribed antimicrobial ± glucocorticoid topical treatment. Lasting resolution of skin fold dermatitis is rarely achieved where friction and moisture retention continue at sites of close skin apposition, and therefore longer-term topical preventive or maintenance treatment is usually required [5][6][7] . Although skin barrier repair products are used for treatment and prevention of skin fold dermatitis in humans 44 , these have not yet been proposed in the veterinary literature but would seem an interesting concept to explore prospectively for prevention in predisposed dogs.
In line with responsible antimicrobial stewardship principles in dogs, systemic antibiotics are recommended for use only after confirmation of deep bacterial infection by cytology, and culture and susceptibility testing to direct drug choice 5,6 . This strict approach is in contrast with the current results, where systemic antibiosis was prescribed in over 40% of cases, whereas bacterial culture was performed in less than 2%. Extrapolating from the overall period prevalence of skin fold dermatitis and the antibiotic treatment frequency in the current study, 0.16% of dogs in primary care practice received systemic antimicrobials for treatment of skin fold dermatitis over the one-year period. While this shows that skin fold dermatitis treatment contributes significantly less to systemic antimicrobial use in canine practice compared with anal sac disorders and pyoderma, with an estimated 1% of UK dogs being treated with systemic antibiosis each year for each of these disorders 45,46 , the current results still highlight the widespread use of systemic antibiosis without documented diagnostic justification. Selection bias towards formal diagnosis of more severe presentations of skin fold dermatitis may partially explain the high frequency of systemic antimicrobial prescriptions documented in the current study that seems misaligned with the heavy focus on treating using only topical modalities in published, empirical treatment recommendations [5][6][7] . Assessment of disease severity, or depth of infection, was not possible in the current study and therefore fuller speculation about the clinical justification for systemic antibiotic treatment in such a large proportion of dogs was not possible.
Epidemiological studies based on primary care practice records uniquely reflect common clinical practice for the patient demographics and professional standards that are specific to the professional, economic and societal context at the time of sampling 47 . Whilst this is a strength, offering opportunities to define benchmarks and compare current practice to guidance, there are several limitations that have been previously reported to this approach of using a convenience sample of practices and to the nature of data that were not primarily recorded for research purposes and thus subject to data gaps and inaccuracies 47,48 . The diagnosis of skin fold dermatitis relies heavily on typical clinical signs evident on inspection of the skin, which should translate into a high diagnostic specificity. However as discussed above, issues around owner awareness and factors relating to thresholds for clinical diagnosis and recording may have created considerable and variable selection bias between practices and clinicians. The current study was a prevalence study that included all cases diagnosed with skin fold dermatitis in 2016 rather than an incidence study that only included cases that were first diagnosed in 2016. While this approach may give stronger inference on the disorder burden from skin fold dermatitis on dogs, it could also bias towards subsets of dogs with greater survival or earlier first diagnosis appearing to have higher odds. The statistical precision was weak for results from uncommon breeds and body sites affected, and therefore these associations should be treated with caution. Treatment information was based on medicines dispensed and may not capture recommendations of general sales items that may be purchased away from the veterinary surgery.
In conclusion, this study has explored anonymised primary-care veterinary clinical data in the UK to show that 0.37% of dogs overall were recorded with skin fold dermatitis annually. The three breeds with by far the highest odds of skin fold dermatitis represent breeds with an extreme brachycephalic conformation: English Bulldog, French Bulldog and Pug. The evident contribution of breed-related conformation for this condition adds to the discussion on the welfare and ethical challenges around current high demand for extreme conformations in dogs. The high frequency of systemic antimicrobial usage for skin fold dermatitis documented in this study www.nature.com/scientificreports/ can be regarded as a benchmark of current clinical activity and also raises questions around misalignment of UK primary-care prescribing practices with published treatment recommendations.

Methods
The study population included dogs under primary veterinary care at clinics participating in the VetCompass Programme during 2016. Dogs under veterinary care were defined as those with either (a) ≥ 1 electronic patient record (EPR) (free-text clinical note, treatment, or bodyweight) recorded during 2016 or (b) ≥ 1 EPR recorded during both 2015 and 2017. VetCompass collates de-identified EPR data from primary-care veterinary practices in the UK for epidemiological research 19 . Data fields available to VetCompass researchers include a unique animal identifier along with species, breed, date of birth, sex, neuter status, insurance and bodyweight, and also clinical information from free-form text clinical notes, summary diagnosis terms 49 and treatment with relevant dates. A cohort study design was used to estimate the one-year (2016) period prevalence of skin fold dermatitis and to explore associations with demographic risk factors in this population. In a UK study using primary-care clinical records, skin fold dermatitis did not feature among disorders with a prevalence of 0.46% and above 50 . Therefore, assuming a prevalence lower than 0.46%, power calculations estimated that a study sample with at least 38,136 dogs was needed to estimate prevalence for a disorder that occurred in 0.25% of dogs with 0.05% acceptable margin of error at a 95% confidence level from a national UK population of 8 million dogs 51,52 . Ethics approval was obtained from the RVC Ethics and Welfare Committee (reference SR2018-1652). All methods were performed in accordance with the relevant guidelines and regulations. The study is reported in accordance with ARRIVE guidelines 53 .
The case definition for a skin fold dermatitis case required evidence in the clinical records indicating a final diagnosis of skin fold dermatitis or synonym (e.g., intertrigo, skin fold pyoderma) at any date from Jan 1, 2016 to Dec 31, 2016. Each dog was defined as either a case or non-case based on the meeting the case definition above, regardless of the level of veterinary care received during 2016. Case-finding involved initial screening of all 905,554 study dogs for candidate skin fold dermatitis cases by searching the clinical free-text from July 1st 2015 to June 30th 2017 using the search terms: wrinkl*, intertrig*, pyod* + fold*, derm* + fold*, fac* + fold*, lip* + fold*, neck* + fold*, body* + fold*, vulv* + fold*, tail* + fold*, nas* + fold*, infect* + fold*, inflam* + fold*, prurit* + fold*, bact* + fold*, yeast* + fold*, chin* + fold*. The clinical notes of a random sample of candidate animals were manually reviewed to evaluate for case inclusion. Additional information was extracted for each confirmed skin fold dermatitis case: body location of the skin fold dermatitis, clinical signs, diagnostic testing and treatment/management. Breed descriptive information entered by the participating practices was cleaned and mapped to a VetCompass breed list derived and extended from the VeNom Coding breed list that included both recognised purebred breeds and also designer-crossbreed breed terms 49 . A breed purity variable categorised all dogs of recognisable breeds as 'purebred' , dogs with contrived names generated from two or more purebred breed terms as 'designer' crossbreds (purposely bred crossbreeds) and dogs recorded as mixes of breeds but without a contrived name as 'crossbred' 15 . A breed variable included individual pure breeds and designer hybrids represented by over 3000 dogs in the overall study population or with ≥ 5 skin fold dermatitis cases, along with groupings of all remaining breeds and also general crossbred dogs. This approach was taken to facilitate statistical power for the individual breed analyses 54 . Breeds were also characterised by haircoat (short, medium, long, uncategorised), skull shape (dolichocephalic, mesocephalic, brachycephalic, uncategorised) and spaniel (spaniel, non-spaniel, uncategorised) status for analysis. A Kennel Club breed group variable classified breeds recognised by the UK Kennel Club into their relevant breed groups (Gundog, Hound, Pastoral, Terrier, Toy, Utility and Working) and all remaining types were classified as non-Kennel Club recognised 15 .
Consistent with methods previously used 55 , neuter and insurance status were defined by the final available EPR value. Adult bodyweight was defined as the mean of all bodyweight (kg) values recorded for each dog after reaching 18 months old and was categorised as: < 10.0, 10.0 to < 15.0, 15.0 to < 20.0, 20.0 to < 25.0, 25.0 to < 30.0, 30.0 to < 40.0 and ≥ 40.0. Mean adult bodyweight was generated for all breed/sex combinations with adult bodyweight available for at least 100 dogs in the overall study population and used to categorise individual dogs as "at or above the breed/sex mean", "below the breed/sex mean" and "unspecified". Age (years) was defined based on the earliest date for diagnosis of skin fold dermatitis in the available clinical records for cases and on December 31, 2016 (the final date in 2016 that these dogs were not a case) for non-cases. Age was categorised as: ≤ 1.0, 1.0 to < 2.0, 2.0 to < 4.0, 4.0 to < 6.0, 6.0 to < 8.0, 8.0 to < 10.0, 10.0 to < 12.0 and ≥ 12.0.
Following internal validity checking and data cleaning in Excel (Microsoft Office Excel 2013, Microsoft Corp.), analyses were conducted using Stata Version 16 (Stata Corporation). The one-year period prevalence with 95% confidence intervals (CI) described the probability of skin fold dermatitis at any point during 2016. Because the sampling design involved verification of a subset of candidate cases, the predicted total case count was calculated using the Stata survey function as previously described 56 . The CI estimates were derived from standard errors, based on approximation to the binomial distribution 57 . Risk factor analysis used binary logistic regression modelling to evaluate univariable associations between risk factors (breed, haircoat, skull shape, spaniel, breed purity, Kennel Club recognised breed, Kennel Club breed group, adult bodyweight, bodyweight relative to breed-sex mean, age, sex, neuter, and insurance) and skin fold dermatitis during 2016. Because breed was a factor of primary interest for the study, variables that derived from the breed information and therefore were highly correlated with breed (haircoat, skull shape, spaniel, breed purity, Kennel Club recognised breed and Kennel Club breed group) were excluded from initial breed multivariable modelling. Instead, each of these variables individually replaced the breed variable in the main breed-focused model to evaluate their effects after taking account of the other variables. Adult bodyweight (a defining characteristic of individual breeds) replaced breed and bodyweight relative to breed/sex mean in the final breed-focused model. Risk factors with liberal associations in univariable www.nature.com/scientificreports/ modelling (P < 0.2) were taken forward for multivariable evaluation. Model development used manual backwards stepwise elimination. Clinic attended was evaluated as a random effect and pair-wise interaction effects were evaluated for the final model variables 58 . The area under the ROC curve and the Hosmer-Lemeshow test were used to evaluate the quality of the model fit and discrimination (non-random effect model) 58,59 . Statistical significance was set at P < 0.05.