Skin and ear health in a group of English bulldogs in Finland - a descriptive study with special reference to owner perceptions.

BACKGROUND: Dermatological conditions are common in English bulldogs (EBs). HYPOTHESIS/
OBJECTIVES: This cross-sectional study describes the dermatological health status of a group of EBs and compares the results with owner perceptions and its possible impact on quality of life (QoL). Computed tomographic (CT) findings of the ear canals were compared between EBs and mesaticephalic dogs. ANIMALS: Twenty-seven EBs participating in a health study in Finland. METHODS AND MATERIALS: A QoL questionnaire was completed for EBs with owner-reported clinical signs referable to the skin or ear. Clinical evaluation included recording the Canine Atopic Dermatitis Extent and Severity Index, the Otitis Index Score, false paw pad grading and the presence of interdigital furunculosis. These were summed to form a total clinical score (TCS). The cross-sectional surface areas of the horizontal ear canals were measured from CT images and compared with respective images of 14 mesaticephalic dogs collected from a patient database.
RESULTS: All 27 EBs had abnormal findings on dermatological examination, but 37% of the owners had not recognized skin or ear signs. The median QoL score was 5.0 (range 0-12) and correlated with TCS (correlation coefficient = 0.507, P < 0.05). English bulldogs had narrower horizontal ear canals than mesaticephalic dogs (P < 0.001). CONCLUSIONS AND CLINICAL IMPORTANCE: All EBs had abnormal dermatological findings that were unnoticed or considered to be of minor significance to the QoL by most owners. Narrow ear canals were common, possibly related to the brachycephalic conformation of the breed.

Introduction

A variety of dermatological conditions are common in English bulldogs (EBs).1, 2, 3, 4, 5 A cross‐sectional study of show EBs reported that all dogs had abnormal findings on dermatological examination and pedal lesions were especially frequent.3 Pododermatitis is a multifaceted disease complex with a diversity of underlying causes.2, 6, 7, 8 A wide variety of skin disorders2, 7 as well as conformational abnormalities of the paws due to obesity8 or orthopaedic diseases such as osteoarthritis and joint disease6 may lead to inflammation of the paws. Chronic inflammation may further lead to altered distribution of weight‐bearing to the haired interdigital skin adjacent to digital pads. The skin can then become thicker and fibrotic, forming a callus‐like surface known as false paw pads (FPP).9 Pads with FPP may further predispose to the formation of interdigital furunculosis (i.e. interdigital inflammatory nodules).6, 9 Dogs with chronic skin conditions are more likely to be exposed to antimicrobial treatments and frequent veterinary visits, both of which, in addition to skin lesions, increase the likelihood of acquiring antimicrobial‐resistant bacteria.10, 11

Brachycephalic dog breeds, such as the EB, have a congenitally flattened facial anatomy. This conformation is associated with several malformations of the head and neck and also has been shown to affect the structure of the ear.12, 13, 14 Brachycephalic dogs have been reported to possess a significantly thicker tympanic bulla wall12 and a lower tympanic bulla volume12, 13 than nonbrachycephalic dogs.12, 13 Fluid accumulation in the middle ear as an incidental finding appears to be more common in brachycephalic breeds.12, 13, 14 Narrow horizontal ear canals, possibly due to conformational abnormality, are a common anecdotal finding in clinical practice and may be a predisposing factor for otitis.15

Skin and ear conditions may affect not only the wellbeing of the dog but also the quality of life (QoL) of the dog's owner. QoL surveys have been developed to assess the owner‐perceived impact of dermatological conditions on the wellbeing of dogs and their owners.16, 17 When these surveys are used in parallel with clinical parameters, a more extensive view of the effect of skin disease on the wellbeing of the dog and the owner may be obtained.

The aim of this study was, first, to describe the clinical dermatological findings from a group of EBs and compare the results with owner perceptions of their dog's dermatological health status and its possible impact on QoL. Secondly, we aimed to evaluate the ear canal structure of EBs by comparing computed tomographic (CT) findings between EBs and dogs with a medium‐length skull (mesaticephalic dogs). In addition, dogs were screened for carriage of antimicrobial resistant bacteria. We hypothesized that dermatological conditions are common in EBs and that their ear canals are narrow compared with mesaticephalic dogs.

Methods and materials

Study design and population

We utilized the dogs participating in a cross‐sectional EB health study, conducted at the Veterinary Teaching Hospital (VTH), University of Helsinki, Finland, from December 2014 to June 2015. The study included three co‐projects evaluating the respiratory (including exercise assessment),18 musculoskeletal and dermatological health of the dogs.

The inclusion criteria consisted of a purebred EB status (registered by the Finnish Kennel Club), an age of two to five years, and no history of airway or orthopaedic surgery nor conditions prohibiting anaesthesia. The owners of the dogs were recruited via the national breed club and social media. In the first stage, the owners completed an online pre‐study questionnaire (Table S1) concerning their dog's physical activity and overall health. Altogether, 54 responses were received and dogs were classified according to their activity habits into three groups: (1) “active” which included dogs with daily exercise of ≈5 km in 60 min, (2) “average” which included dogs with daily exercise of 1–5 km in a maximum of 45 min and (3) “nonactive” which included dogs with daily exercise of a maximum of 1.5 km in 40 min. Ten dogs from each group were chosen to participate in the order of enrolment. Only one dog per owner was chosen; however, if more dogs were suitable from one owner, the owner was free to decide which dog participated in the study. As two of the 30 chosen dogs could not participate due to acute illness, a total of 28 EB were included in the study. One dog had the first visit before the start of the dermatological study, leaving 27 dogs to participate in this co‐project.

All study dogs were privately owned pet animals, whose owners signed an informed consent form. The Committee for Experimental Animals of Southern Finland and the Regional State Administrative Agency of Southern Finland (ESAVI) approved the study and this subproject (ESAVI/11519/04.10.07/2014, ESAVI/9184/04.10.07/2014 and ESAVI/827/04.10.07/2015).

Study protocol

At the first visit owners were interviewed to determine the dermatological disease history and a QoL questionnaire16 was completed. The health of the skin and ears was evaluated by one of two clinicians (see details below).

At the second visit the dogs were anaesthetized. The FPP were graded, dermatological diagnostic samples were taken and computed tomography (CT) of the head was performed. For anaesthesia, the dogs were premedicated with intramuscular butorphanol 0.2 mg/kg (Torbutor, Richter Pharma; Wels, Austria), induced after preoxygenation with intravenous lidocaine 1 mg/kg (Lidocaine, Orion; Espoo, Finland) and alfaxalone (Alfaxan, Jurox; West Sussex, UK) to effect (approximately 2 mg/kg); anaesthesia was maintained after intubation with sevoflurane (Sevoflurane, AbbVie; Espoo, Finland) and continuous infusion of lidocaine 2 mg/kg/h. CT studies from mesaticephalic dogs in the VTH database were used as a historical control to compare the ear canal surface areas.

Dermatological disease history

The dermatological history was obtained in an interview with the owners by one of the two clinicians who performed the ear and skin evaluations, and the interviews included questions about pruritic behaviour (head shaking or scratching of the ears, paw licking, rubbing of the muzzle, scratching or rubbing of the flanks) and the history of treatment for dermatological conditions (otitis, pododermatitis or interdigital furunculosis, demodicosis, skin folds) (Table S2). Information on the previous use of systemic antimicrobial treatments was collected.

Quality of life questionnaire

The pre‐question was “Has your dog had skin or ear signs?” If the response was positive, the QoL questionnaire16 was completed. The first QoL question asked the owners for their perceptions of the general severity of the disease, after which the questions concerned the QoL of the dog (Q2–8) and the owner (Q9–15). Each question had response options scored from 0 to 3 [(0) “not at all”, (1) “a little”, (2) “quite a bit” and (3) “very much”). The maximum score was 3 for the first question and 42 for the QoL part of the questionnaire.

Clinical evaluation

The health of the skin was evaluated using the Canine Atopic Dermatitis Extent and Severity Index (CADESI)‐04.19 Dogs with an index score of < 10 points were considered normal, whereas scores of 10–34 indicated mild, 35–59 indicated moderate and 60–180 points indicated severe dermatitis.

Ear health assessment was based on the Otitis Index Score, OTIS‐3.20 Clinical scores of ≤3 per ear were considered normal. An otoscopic examination was performed in an awake animal and the status of the tympanic membrane was recorded if visualization was possible.

The severity of FPP was graded subjectively as follows: (0) no false pads, (1) mild, (2) moderate and (3) severe FPP. Example images of the grading are presented in Figure 1. The presence of interdigital furunculosis also was recorded (yes/no).

Figure 1

Grading (0‐3) of false paw pads in English bulldogs with examples of digital images used for evaluation. a) Grade 0; b) grade 1; c) grade 2; d) grade 3.

The scores from CADESI‐04, OTIS‐3 (both ears), FPP scores (four paws) and presence of interdigital furunculosis were summed to form a total clinical score (TCS) to obtain an overview of the dermatological findings.

Interdigital deep skin scrapings were taken with a scalpel blade until capillary bleeding, together with trichograms. One scraping and one trichogram were taken from each paw, from the dorsal interdigital space of the left paws and the ventral aspect of the right paws. The samples were placed on glass slides with a drop of paraffin oil and examined at ×100 magnification under a microscope to detect Demodex mites.

Bacteriological swabs to screen for the presence of meticillin‐resistant Staphylococcus pseudintermedius (MRSP), S. aureus (MRSA) and extended spectrum beta‐lactamase (ESBL)‐producing Enterobacteriaceae were taken using sterile transport swabs (M40 Transystem, Copan Diagnostics; Murrieta, CA, USA). For MRSP/MRSA, two swabs were taken: one from the nares and the mouth/lip commissure, and one from the perianal region.11 For ESBL, the swab was taken from the rectum.21 The swabs were processed and bacterial identification, including susceptibility testing and Pulsed Field Gel Electrophoresis (PFG) analysis, was performed as described previously.21, 22

Computed tomography and image processing

The CT examination of the whole head of EBs was performed in sternal recumbency with a helical dual slice scanner (Somatom Emotion Duo, Siemens AG; Forchheim, Germany) using a bone algorithm. The slice thickness was 2 mm, feed/rotation 4 mm and reconstruction increment 0.5 mm.

Selected multiplanar reconstruction images were used for measurements of the horizontal ear canal using osirix software (v9.0, OsiriX Foundation; Geneva, Switzerland). The window level and width were 300 and 1500, respectively. First, the dorsal image plane was determined from the midsagittal image as the parallel plane to the skull base. Second, the transverse image was straightened at the level of the middle ears using the cochleae as landmarks. The sagittal image for measuring the cross‐sectional surface area (CSA; mm2) of the air‐filled area in the osseous part of the ear canal (CSA1) was the one in which the bony wall was most clearly seen. The image for measuring the CSA of the widest part of the air‐filled ear canal (CSA2) was selected as the one in which the cross‐section of the canal was considered largest. The images were saved in DICOM format. In addition, the presence or absence of soft tissue attenuating material in the tympanic bullae and presence of significant external ear canal mineralization was recorded from the image series.

In order to compare the results, CT scans of mesaticephalic dogs were collected from the VTH patient database. The inclusion criteria were (i) a mesaticephalic breed, (ii) CT of the head performed for other than ear‐related reasons, (ii) weight >13 kg and (iv) slice thickness of the scan ≤2 mm. Dogs with a history of ear and/or skin disease or with abnormal CT findings in the ear canals/bullae were excluded. Nine of 14 dogs had been imaged with the same scanner as the EBs of this study and five with a helical 64‐slice scanner (Lightspeed VCT, GE Healthcare; Madison, WI, USA). All CT scans were examined by a single observer by using osirix v.9.0 (OsiriX Foundation); the observer did not participate in the clinical evaluations of the EBs.

In subgroup analysis, EBs that had either received veterinarian‐prescribed care for otitis or had an OTIS‐3 score of >3, or had discharge ≥2 in the OTIS‐3 score, were excluded.

Data analysis

Descriptive statistics were used to summarize the data, which are presented as medians and ranges. For categorical data, counts are provided. The Wilcoxon–Mann–Whitney U‐test was used for non‐normally distributed data to compare the TCS between dogs with and without owner‐recognized skin or ear signs, and the distribution of basic demographics (except for gender) as well as CSA1 and CSA2 measurements between EBs and mesaticephalic dogs. The distribution of gender between the groups was compared with the chi‐square test. Spearman's rank correlation was used to evaluate the correlation of QoL with TCS in dogs with owner‐recognized skin or ear signs. Statistical analyses were performed using IBM SPSS v.25 (IBM Corporation; Chicago, IL, USA). The significance level was set at P < 0.05.

Results

Twenty‐seven EBs (12 females and 15 males) were examined. The median age was three years (range 2–5 years) and the median weight was 25 kg (range 18.5–33.1 kg).

In the interview focused on pruritic behaviour and the treatment history, 18 of 27 (67%) owners reported at least one type of pruritic behaviour; 13 of these owners reported paw licking and 10 head shaking or scratching of the ears in their dog. Fourteen (52%) dogs had received prescribed treatment for at least one of the following diseases: otitis (11 of 27), pododermatitis or furunculosis (8 of 27) or other dermatitis (5 of 27). One dog had received treatment for demodicosis. Of 27 dogs, nine (33%) had received systemic antibiotic treatments during the previous year and 10 (37%) more than one year before. In five (19%) dogs, systemic antibiotics had never been used and in three this information was missing. None of the dogs had ongoing medical treatments for skin or ear conditions at the time of the visits.

Quality of life

The QoL questionnaire was completed for 17 of 27 dogs (63%), as their owners responded “yes” to the preliminary question. The median score for owner‐assessed disease severity was 1 (Q1, range 0–3 points) and for the QoL part of the questionnaire (Q2–15) the median was 5.0 (range 0–12). Questions concerning therapies, time loss and expenditure had a median score of 1 point (range 0–2). All the other questions had a median of 0 (range 0–2).

Regarding CADESI‐04, all dogs received scores indicating either mild (20 of 27, 74%) or moderate (7 of 27, 26%) dermatitis.

In four of 27 dogs, the OTIS‐3 score was >3 (two dogs bilateral, two dogs unilateral) whereas the rest had normal scores. Of these, tympanic membranes could be visualized in only one dog. The presence of a tympanic membrane could not be assessed in an additional 10 dogs (16 ears) due to poor visibility associated with narrow ear canals or discharge in the ear canal.

Twenty‐five of 27 dogs (93%) had FPP, of which 14 (52%) were mild, seven (26%) were moderate and four (15%) were severe. In the two most advanced cases, the dogs had severe FPP on all four paws. Nine dogs (33%) had interdigital furunculosis.

Regarding the TCS, the median was 26 (range 17–45) for dogs without and 36 (range 25–80) for dogs with owner‐recognized skin and ear signs (P = 0.005; Wilcoxon–Mann–Whitney U). A correlation between the TCS and QoL scores in dogs with owner‐recognized skin and ear signs (n = 17) was observed (Spearman rho 0.507; P < 0.05) (Figure 2).

Figure 2

Correlation between total clinical scores (TCS) and quality of life (QoL) scores in English bulldogs with owner‐recognized skin or ear signs. Individual TCS (scale 0–217) and QoL scores (scale 0–42) in English bulldogs (EBs) are plotted (n = 27). The correlation between TCS and QoL in dogs with owner‐recognized skin or ear signs was statistically significant (n = 17, r = 0.507, P < 0.05). The TCS for EBs without owner‐perceived skin and ear signs (QoL questionnaire not completed, n = 10) are placed at the zero point on the QoL scale.

All trichograms and skin scrapings were negative for Demodex mites.

Five dogs (19%) tested positive for MRSP and one (4%) tested positive for ESBL as part of their microbiota. None of the dogs had MRSA isolated. The MRSP strains were unrelated according to PFGE analysis (Figure S1)

Computed tomography image evaluations

Control images were obtained from 14 mesaticephalic dogs representing 13 breeds (breeds and reasons for a CT referral are presented in Table S3). There were no significant differences in gender or age between EBs and control dogs, but EBs were slightly heavier (median 25 kg, range 18.5–33.1 kg) than the controls (median 18 kg, range 13.7–35.9 kg, P = 0.04; Wilcoxon–Mann–Whitney U).

Horizontal ear canals of EBs were significantly narrower than mesaticephalic dogs in CSA1 and CSA2 (P < 0.001; Wilcoxon–Mann–Whitney U; Table 1). Subgroup analysis which excluded the 14 EBs that had received veterinarian‐prescribed care for otitis and/or had an OTIS‐3 score >3 or discharge ≥2 did not change the results (Table 1). Of the remaining 13 EBs included in the subanalysis, one or both tympanic membranes could not be visualized in six dogs (10 ears) due to narrow horizontal canals.

Table 1

Cross‐sectional area (CSA) measurements of the ear canals in English bulldogs (EBs) and mesaticephalic dogs

CSA (mm2) Median (range)	EBs (n = 27)	Subgroup of EBs* (n = 13)	Mesaticephalic dogs (n = 14)	P‐value
CSA1 ‐ osseous air‐filled part of the horizontal ear canal	9.1 (0–22.0)	10.3 (0–22.0)	24.0 (11.2–31.4)	0.000**
CSA2 ‐ the widest air‐filled part of the horizontal ear canal	30.4 (0–65.2)	31.5 (15.5–65.2)	46.8 (33.5–75.6)	0.000**

Subgroup analysis excluding 14 EBs that had either received veterinarian‐prescribed care for otitis and/or, had an OTIS‐3 score of >3, and/or had discharge ≥2 in the OTIS‐3 score.

P‐values from comparisons between EBs and mesaticephalic dogs, as well as the subgroup of EBs and mesaticephalic dogs.

Two EBs had fluid/soft tissue attenuating material unilaterally in the middle ear. Both dogs had a history of ear treatment and one had an abnormal OTIS‐3 score in the affected ear. The tympanum could not be evaluated in the ears with effusion. One of the dogs with middle ear effusion had marked unilateral mineralization of the ear canal. Another dog without effusion had marked mineralization in both ear canals. This dog had a history of ear treatments and an abnormal OTIS‐3 score in both ears and lack of visibility of the tympanum.

Discussion

All EBs in our study had abnormal findings on dermatological examination. Pedal lesions were frequent and all the dogs had abnormal CADESI‐04 scores. These results are in accordance with a previous study of show EBs, which all had dermatological signs with frequent pedal lesions.3 As all dogs had abnormal CADESI‐04 scores and many owners reported pruritic behaviour in their dogs, allergic skin diseases may have been among the underlying causes of pododermatitis in some dogs in our study. Demodicosis, another common cause of pododermatitis, was not observed. Obesity was not considered to be a predisposing factor in this study, as the body condition scores were normal in most dogs.18 Interestingly, 37% of the owners of our study stated that their pet did not have skin or ear‐related problems. It is possible that especially mild signs are not noticed by the owner, as these dogs had lower TCS. The majority of those owners who reported skin problems did not consider them to affect the QoL of the dog or the family, or gave relatively low scores (mean 4.6, median 5.0, range 1–12) compared to the mean QoL scores reported previously for similar skin conditions (pododermatitis, 17.3 points, uncomplicated atopic dermatitis, 15.2 points).23

The EBs had narrow ear canals compared to mesaticephalic dogs. To determine whether this was conformational or caused by the ear disease, we repeated our analysis after excluding dogs with a history of ear treatment and/or elevated OTIS‐3 scores and/or discharge in the ear canal, which could cause artificial narrowing of the canals in CT images. This had no effect on our results, suggesting that EBs have narrow ear canals due to conformational reasons. We are aware of the methodological restrictions of CT image comparisons between EBs and mesaticephalic dogs. In some EBs, the air‐filled area of the ear canal, especially in CSA1, was so small (even < 5 mm2) that precise measurement was challenging. It also is possible that we were unable to exclude all the EBs with a history of ear diseases or that these had not been noticed in some of the control dogs, because retrospective data can be incomplete. However, as the difference between the groups was large, it is unlikely that such matters had a major effect on our results. Clinical evaluations of the ears supported our results, as in many EBs the ear canals were so narrow that the presence of a normal tympanum could not be assessed. It is known that narrow conformation of the ear canals may predispose to otitis,15 but an abnormal OTIS‐3 score was a surprisingly rare finding in the dogs in our study. Previous studies have demonstrated a higher prevalence of subclinical middle ear effusion in brachycephalic dogs of various breeds.12, 13, 14 However, only a small number of dogs in these studies were EBs. Two dogs in our study had fluid/soft tissue attenuating material unilaterally in the middle ear. Both dogs had a history of ear treatment and one also had an abnormal OTIS‐3 score in the affected ear.

Discrepancies between the owner‐reported skin problems and clinical findings raise the question of whether the owners of EBs are unaware of dermatological signs, especially when milder. In three cases, the owners reported either pruritic behaviour or previous treatment for dermatitis, although they responded that their dog did not have skin‐related problems. We do not know whether pruritic behaviour was not considered by the owners as a sign of skin problems, or whether this discrepancy was simply a misinterpretation of the questions. In this study, owner‐reported paw licking was infrequent (13 of 27, 48%) compared to the presence of pedal lesions. In the previous study of EBs, the frequency of owner‐reported paw licking did not differ from a population of healthy dogs, although pedal lesions were very common.3 It is possible that some EBs may be unable to express certain types of pruritic behaviour, such as paw‐licking, due to their body conformation (i.e. inability to bring their paws near to the mouth) as has been suggested.3 Another explanation might be that not all pedal lesions were pruritic. Of four dogs with an abnormal OTIS‐3 score, only two had a history of prescribed treatment for otitis and one of head shaking or scratching of the ears. Thus, it may be that not all the owners recognized skin or ear lesions alone as a problem if no other clinical signs were displayed.

The prevalence of MRSP was higher in this group of EBs (19%) than estimated for the normal canine population in Finland (3%).10, 22 However, the results of this study were not surprising considering that all dogs had abnormal findings on dermatological examination and many had received systemic antibiotics. Pyoderma,24 skin lesions, antimicrobial treatment and frequent veterinary visits10, 11 are known risk factors for MRSP. In Finland, the prevalence of MRSP in high‐risk patients (e.g. dogs that suffer from skin‐related diseases and/or have a history of heavy antimicrobial use) is approximately 10%.10 Considering that all EBs of this study had clinical dermatological findings and that the majority of them had been exposed to antimicrobial products, the prevalence of 19% (95% CI 8–36%) is consistent with the prevalence observed in dogs with risk factors.

Various health problems, such as brachycephalic obstructive airway syndrome,18, 25 osteoarthritis26 and certain dermatological conditions,1, 2, 3, 4, 5 are common in EBs, although the Official United Kennel Club Breed Standard for the EB states that any conditions or exaggerations that are detrimental to the health, welfare, essence or soundness of the breed must be avoided.27 Health problems of EBs also have received public attention in the media.28, 29 EB as a breed has low genetic diversity, which may be an obstacle to recreating healthier phenotypes from the existing genetic pool.30 Moreover, in one study a frameshift mutation in the gene DISHEVELLED 2 (DVL2) was found in EBs and other breeds that share many similar morphological and disease‐predisposition traits such as a brachycephalic skull morphology and vertebral malformations.31 Analogous frameshift mutations in the human DVL1 and DVL3 genes cause Robinow syndrome, a congenital disorder characterized by similar craniofacial, limb and vertebral malformations to those exhibited in EBs. If the breed‐associated characteristics are linked to a gene causing congenital disorders, it should be questioned why a characteristic that should be regarded as “abnormal” has actually become “normal” or even desired in the breed, and whether a better health status could be achievable with cross‐breeding. It is not known whether dermatological conditions are related to this gene variant, but the dermatological health status of the breed is nevertheless concerning, because none of the EBs were clinically healthy and neither were the EBs of the previous study, even though show dogs should represent the ideal standard of the breed and are frequently used for breeding.3

There were certain limitations in this study. The sample size was small and the dogs were from a selected population. Dermatological health was not among the inclusion criteria for selection. All of these facts can cause bias in the results. It may be that the results of this small study do not necessarily reflect the situation in the EB population in Finland and other countries. However, the included dogs were young adult EBs, and according to the owners’ statements, most were without medical or other conditions affecting the dog's daily life.18 In addition, although the dermatological health was not among the inclusion criteria, one‐third of EBs were without owner‐perceived skin and ear signs, but all of these had abnormal findings in the dermatological examination. It would, therefore, be unlikely that the high frequency of abnormal dermatological findings is due to bias or chance alone. However, as the dogs with owner‐perceived skin and ear signs had higher TCS than the dogs without, the abnormal dermatological findings would have possibly been milder if the inclusion criteria had been dogs with no history or perceived history of dermatological or otic disease.

Another limitation is related to the questionnaires and clinical assessments. Even though the QoL questionnaire has been validated and is applicable to all dermatological conditions,16 the pre‐question and interview were not validated. The CADESI‐04 score has been validated, but only for dogs with atopic dermatitis.19 Nevertheless, it was considered the best available option for this study. OTIS‐3 has not been validated, although in a small pilot study scores of ≥4 differentiated affected ears from healthy ears with a sensitivity of 91.1% and a specificity of 100%,20 and was thus considered adequate for this study. A subjective grading system was created and used for FPP. Even though this was not validated, example pictures were used to standardize the assessments between the observers. Ideally, the same investigator would have performed all the evaluations. Otoscopic examinations were performed in EBs during the first visit while the dogs were awake, as we did not want anaesthesia to affect the OTIS‐3 assessments. It would have been optimal for evaluation of the tympanic membranes to repeat the otoscopic examination while the dogs were under anaesthesia during the second visit.

In conclusion, all EBs had abnormal dermatological findings that were unnoticed or considered to be of minor significance to QoL by most owners. The prevalence of MRSP was higher than estimated for the normal canine population in Finland. However, as the sample size was small, these findings may not be generalizable, although they indicate that further studies on the occurrence of dermatological conditions and the presence of MRSP are warranted. The EBs had narrow ear canals when compared to mesaticephalic dogs. Further studies with larger sample sizes are needed to verify the results and investigate whether this could be a breed‐specific finding or applicable to brachycephalic dogs in general.

Figure S1. smaI restriction enzyme PFGE patterns and antimicrobial susceptibility profiles of five MRSP strains carried by English bulldogs in the study.

Table S1. Health survey of English bulldogs; online pre‐study questionnaire.

Table S2. Dermatological disease history questionnaire.

Table S3. Breeds and reasons for referral for the mesaticephalic control dogs in computed tomography evaluations (n = 14).

Click here for additional data file.