The association between the signalment, common causes of canine otitis externa and pathogens



Objective: To determine whether associations exist between pathogens, allergies, conformational abnormalities, endocrinopathies and signalment in canine otitis externa (OE).

Methods: Medical records of 149 dogs which met predetermined inclusion criteria were evaluated retrospectively. Correlations between pathogens and the presence of allergy, endocrinopathy, conformational abnormalities and signalment were evaluated statistically.

Results: The shar-pei, German shepherd and cocker spaniel breeds were over-represented compared with the hospital's breed distribution (P<0·001). German shepherd dogs and cocker spaniels were statistically more prone to infection with rod-shaped organisms and Labrador retrievers less than other breeds (P=0·034). Almost all dogs that were older than five years when diagnosed with OE had cocci (P=0·01) and also had higher levels of rods (P=0·028). The incidence of rods was higher in endocrinopathies (P=0·004), while that of Malassezia spp. tended to be higher in allergies (P=0·098). There were no statistically significant differences among the groups for all the other parameters examined.

Clinical Significance: OE infection is usually not influenced by primary causes or predisposing factors. Endocrinopathies may be followed by a more severe otitis, however. OE may be more severe when it affects older dogs.


Otitis externa (OE) is a common problem in dogs, estimated to occur in 10 to 20% of canine patients seen by veterinarians (Scott 2001a). The most common primary causes of OE are allergies such as atopic dermatitis and adverse food reactions. Fifty-five percent of dogs diagnosed with atopic dermatitis may show signs of OE, and in 3 to 5% of dogs, it may be the only clinical sign (Griffin and DeBoer 2001). Up to 80% of dogs diagnosed with food allergy have OE, and this may be the only sign in 25% of them (Rosser 1993). Keratinisation disorders, either primary, as in idiopathic seborrhea or secondary, as in hypothyroidism and sex hormone imbalance, are also common primary causes affecting the secretions of the ceruminous and sebaceous glands lining the ear canal (Rosser 2004). Anatomical changes in the ear canal such as conformational abnormalities are commonly represented among the predisposing factors. Of the breeds that are genetically affected by conformational abnormalities, German shepherd dogs and cocker spaniels are prone to seborrhea, German shepherd dogs have high moisture levels in their ear canals, cocker spaniels have pendulous pinnae, shar-peis have hypoplastic and stenotic ear canals and poodles have a high density of hair in their ear canals (Muller 1990, Carloti 1991, Huang and Huang 1999, Scott 2001a, Angus and others 2002, Yoshida and others 2002, Rosser 2004).

Primary causes and predisposing factors (PCPFs) create suitable conditions for the proliferation of secondary organisms such as bacteria and yeasts in the ear canal (August 1988, Rosser 2004). The common secondary pathogens are bacteria such as Staphylococcus intermedius (now reclassified as Staphylococcus pseudointermedius), Pseudomonas aeruginosa and Proteus species and yeasts such as Malassezia pachydermatis (Blue and Wooley 1977, Cole and others 1998, Scott 2001a, Graham-Mize and Rosser 2004). Most of the microorganisms isolated from inflamed ears are found in small numbers in normal ears (Scott 2001a, Rosser 2004). In many cases of OE, more than one kind of microorganism is either isolated by bacterial culture or found in cytology smears (Graham-Mize and Rosser 2004).

The aim of this study was to determine whether associations exist between certain pathogens, three common PCPFs, allergies, endocrinopathies and conformational abnormalities, and signalment of dogs with OE.

Materials and methods

The medical records of 430 dogs diagnosed with OE on presentation during the years 1999 to 2004 at the Dermatology Department of the Veterinary Teaching Hospital of the Koret School of Veterinary Medicine, The Hebrew University of Jerusalem, were examined. In a previous study, allergies, conformational abnormalities and endocrinopathies were found to be the commonest PCPFs of OE in the dogs of the studied population (Zur and Lifshitz 2008).

Inclusion criteria were the presence of signalment of the dogs, diagnosis of OE as due to either allergy, endocrinopathy or conformational abnormalities, results of cytology smears and their validation by bacteriological isolation. A cohort of 149 files met these criteria and was included in the study.

The dogs were grouped according to three separate PCPFs. The allergy group included atopic dermatitis, adverse food reactions and flea bite hypersensitivity. The diagnoses of these allergies had been made according to history and clinical signs that had been found to be related to these allergies (Scott 2001b), response to tight flea control in the cases of flea bite hypersensitivity, response to 8 to 12 weeks of elimination dietary trials and worsening of clinical signs after re-challenging in adverse food reactions. Atopic dermatitis was also diagnosed according to Willemse criteria and by ruling out other causes for pruritus (Willemse 1986). The endocrinopathy group included hypothyroidism, hyperadrenocorticism and sex hormone imbalance. These diagnoses had been substantiated by histopathology or special blood tests in cases of hypothyroidism and hyperadrenocorticism (Zerbe 2000, Feldman and Richard 2004). The conformational abnormality group included breeds with known anatomical problems such as spaniels, shar-peis, poodles, German shepherd dogs (August 1988, Carlotti 1991, Scott 2001a, Yoshida and others 2002), when other PCPFs could not be identified (i.e. no signs of dermatological or other clinical abnormalities and no response to a 10-week elimination dietary trial) and also other breeds where otoscopic examination validated a primary conformational abnormality. Dogs suspected as having more than one underlying condition were excluded from the study.

The pathogens were divided into three main classes, cocci, rods and yeasts, according to cytology results. Ear swabs for cytology examination were obtained from each ear in every dog using sterile cotton swabs. The samples were obtained from the distal horizontal ear canal, heat-fixed and stained with Diff Quick® (Jorgensen Laboratories Inc.). The relative number of organisms in each class was assessed by semi-quantitative analysis for every case and graded on a scale of 0 to 4 according to the following criteria: grade 0, no organisms in high power field (HPF; X1000); grade 1, 1 to 5 organisms/HPF; grade 2, 6 to 10 organisms/HPF; grade 3, 11 to 30 organisms/HPF; grade 4, above 30 organisms/HPF. At least 10 HPFs were examined in every case. All cytological samples were examined by one author (GZ) at the dog’s first visit. This grading system has been regularly applied in the first author’s practice. No attempt was made to separate the results of each ear in cases of bilateral OE because any of the three PCPFs that were evaluated in this study could have been responsible for bilateral OE. If different results were obtained from each ear, they were analysed together, i.e. the dog had a combination of secondary causes.

Microbiological culture

These results were used only to identify the bacteria found on cytology smears. Samples from ears were collected using sterile cotton-tipped applicators from the distal horizontal canals in cases where bacteria were detected in cytology smears. These samples were submitted to a specialist microbiology laboratory (Kimron Veterinary Institute) for culture and susceptibility tests.


Breeds were assessed and compared with the general hospital population as seen by the emergency and internal medicine departments during the same period. These departments, in which cases from all over the country were referred to, reflected the breed distribution within the country. Breeds represented by three or less dogs were grouped together in an “other breeds” category.

The age of onset of otitis was classified as three categories: less than one year, one to five years and more than five years.

Statistical analysis

The associations between allergies, endocrinopathies, conformational abnormalities, signalment and pathogens were examined at four stages: stage 1 – each PCPF was examined with each pathogen separately. Stage 2 – each PCPF was examined for a combination of pathogens. Stage 3 – each PCPF was examined for the different amounts of the pathogens. Stage 4 – a combination of stages 2 and 3: associations between the three PCPFs were examined in relation to the different amounts of the combination of pathogens.

The chi-squared and Fisher’s exact tests were applied for assessing the associations between two categorical variables. This included testing the associations between the pathogens (cocci, rods and Malassezia spp.) and allergies, endocrinopathies, conformational abnormalities, age of first signs (categorised) and breed. This was applied when pathogens were defined as present or absent or combinations between them. The levels of the pathogens, being an ordinal variable, were compared among the three PCPFs for OE categories and the three age levels, using the Kruskal–Wallis non-parametric test.

All tests applied were two-tailed and a P-value of < 0.05 was considered statistically significant.

Analyses were carried out using the SPSS statistical software (PASW statistics, 18, 2010).


Allergies were the most common PCPF (117 dogs, 78·5%) and conformational abnormalities were found as the second most common PCPF (23 dogs, 15·4%). Endocrinopathies were diagnosed in nine dogs (6·1%). Of the pathogens, cocci and Malassezia spp. were more common than rods (111 dogs, 74·5%; 107 dogs, 71·8% and 78 dogs, 52·3%, respectively). The combination of all three was most frequently present (39 dogs, 26·3%), then cocci and Malassezia spp. and cocci and rods (33 dogs, 21·1% and 29 dogs, 19·5%, respectively). In most of the dogs, the pathogens were in grades 2 to 3 and very few graded 4 (Table 1). The higher grades were considered more severe infections.

Table 1. Levels of pathogens in dogs with otitis externa
Number of organismsPathogen
Cocci (number of dogs, %)Rods (number of dogs, %)Malassezia (number of dogs, %)
  1. *1 to 5 organisms/high power field (HPF; ×1000)

  2. 6 to 10 organisms/HPF

  3. 11 to 30 organisms/HPF

  4. §Above 30 organisms/HPF

None38 (25·5)71 (47·6)42 (28·2)
Level 1*14 (9·4)11 (7·4)19 (12·7)
Level 246 (31·7)24 (16·1)51 (34·2)
Level 345 (30·2)38 (25·5)29 (19·5)
Level 4§6 (4·0)5 (3·3)8 (5·4)

Culture results

The vast majority of cocci (>95%) were S. pseudointermedius; thus, all cocci were analysed as one group. Different kinds of rods were cultured but statistical analysis was done by treating all rods as one group based on cytology results.

The associations between allergy, endocrinopathy, conformational abnormalities and pathogens in dogs with OE

When each pathogen was analysed in association with each of the three conditions (stage 1 in the statistical analysis), no statistically significant differences were found between the groups. Albeit but not significantly cocci and Malassezia spp. were more frequently present in allergic dogs (76·9 and 71·8%, respectively) than rods (50·4%; Table 2).

Table 2. Distribution of pathogens in dogs with allergies, conformational problems and endocrinopathies
 CocciPathogen RodsMalassezia
Number of dogsPercentage in dogs with the same PCPF*Percentage in dogs with the same secondary causeNumber of dogsPercentage in dogs with the same PCPF*Percentage in dogs with the same secondary causeNumber of dogsPercentage in dogs with the same PCPF*Percentage in dogs with the same secondary cause
  1. *Primary cause or predisposing factor


No statistically significant differences were found between the three different conditions when combinations of pathogens were analysed (stage 2 of the statistical analysis). Also, but not significantly, rods and Malassezia spp. were much less frequently present in allergic dogs than the other combinations (~7% had rods and Malassezia spp. versus ~30% had the other combinations).

The levels of rods were significantly higher in endocrinopathies than in allergies and conformational abnormalities (P=0·004). The levels of Malassezia spp. were higher in allergies than in endocrinopathies, but this was not statistically significant (P=0·098). No other statistically significant differences were found between allergies, endocrinopathies and conformational abnormalities and the levels of the pathogens, either separately (stage 3 of the statistical analysis) or their combinations (stage 4 of the statistical analysis).

Signalment and its association with the pathogens of OE

Breeds The breeds and age distribution of the dogs in the study are provided in Tables 3 and 4.

Table 3. Distribution of pathogens among the various breeds of dogs with OE
BreedNumber of dogs in the studyCocciPathogen Rods*Malassezia
Number of dogs% within breedNumber of dogs% within breedNumber of dogs% within breed
  1. *Chi-squared analysis indicated a significant difference in the presence of rods among the different breeds, P=0·034

German shepherd dog272385·21866·71970·4
Cocker spaniel10880990·0550·0
Labrador retriever161062·5425·01381·3
Golden retriever4375250·04100
American Staffordshire bull terrier9666·7444·4777·8
Mixed breeds241666·71041·71979·2
Other breeds372976·31847·42463·2
Table 4. Distribution of pathogens at different ages of onset of otitis externa
Age of onsetNumber of dogs in the studyCocci*Pathogen RodsMalassezia
Number of dogs% in age groupNumber of dogs% in age groupNumber of dogs% in age group
  1. *Chi-square analysis indicated a significant difference in the presence of cocci among the different age groups, P=0·01

0 to 1 year463678·32043·53371·7
1 to 5 years754965·34053·35877·3
>5 years282692·91864·31657·1

The shar-pei, cocker spaniel and German shepherd breeds were represented significantly more with OE when compared with the general hospital population (P<0·001). All the Pekingese dogs had cocci and rods, almost all the cocker spaniels (9 of 10) had rods but only 4 of 16 Labrador retrievers had rods. The differences between breeds for the presence of rods was statistically significant (P=0·034; Table 3). No other statistically significant differences were found between the various breeds and the combination and/or levels of the secondary causes.

Sex There were 80 (53·7%) males and 69 (46·3%) females, 65·6% were neutered and 34·4% were entire. No statistically significant associations were found for sex and neutering and the different primary or secondary causes for OE.

Age at onset of OE The age at onset of OE was divided into three groups: 46 dogs (30·9%) had their first episode when they were less than one year old, 75 dogs (50·3%) were one to five years old and 28 dogs (18·8%) were more than five years old at first diagnosis. Almost all dogs diagnosed with otitis at the older age (26 of 28) had cocci (P=0·01). However, the level of cocci was not different among the various age groups. Although not statistically significant, rods were more frequently found at the older age of onset (>five years) and Malassezia spp. in younger ages (<five years; Table 4). The levels of rods were higher in older dogs (P=0·028). No statistically significant differences were found between the various ages at onset of OE and the combination and/or other levels of the pathogens.


Association between allergies, endocrinopathies and conformational abnormalities and pathogens in dogs with OE

This study investigated the correlation between the most common PCPFs (Rosser 1993, 2004, Huang and Huang 1999, Griffin and DeBoer 2001, Angus and others 2002) and pathogens in cases of OE. To the best of the authors’ knowledge, this is the first report in which these kinds of associations have been studied systematically.

Malassezia spp. and cocci were found more commonly than rods. Malassezia spp. in combination with cocci was also a common finding in another study (Saridomichelakis and others 2007). In this study, in which both ears were evaluated together in cases of bilateral OE, the combination of the three types of microorganisms (i.e. cocci, rods and Malassezia spp.) was most frequently found. Three and more types of organisms were also most frequently isolated in another study in which 50 different ear canals of 33 dogs with OE were evaluated. However, in that study, a mixture of bacteria alone was more common than a mixture of bacteria and yeasts (Graham-Mize and Rosser 2004).

For most of the pathogens examined in this study, either their presence alone or in combinations or their levels in infected ear canals, there were no differences between the various conditions. Although not statistically significant, dogs with allergies were presented more frequently with Malassezia spp. and less frequently with the combination of rods and Malassezia spp. Furthermore, they also had higher levels of Malassezia spp., which is a common secondary pathogen in skin (DeBoer and Marsella 2001) and ear (Zur and others 2002) infections of allergic dogs. Allergies as the most common primary cause and Malassezia as the most common secondary cause were also found in a recent survey of 100 dogs with OE; however, a possible association between the two was not evaluated (Saridomichelakis and others 2007). In another study, Malassezia was found in a very small percentage of dogs with OE due to atopic dermatitis (Saridomichelakis and others 1999). Malassezia was frequently found in chronic cases of OE (Bernard and others 1998, Cafarchia and others 2005). However, rods are more commonly associated with more chronic and severe cases of OE (Kiss and others 1997, Scott 2001a, Rosser 2004). In this study, the highest levels of rods were found in endocrinopathies suggesting a more severe otitis due to this primary cause. Although this finding was highly significant statistically, there were too few cases of endocrinopathies and further studies are needed. Based on the findings that infections with rods are more severe than those with cocci and Malassezia, we can conclude that allergies cause less severe otitis than do endocrinopathies.

Association between signalment and pathogens in dogs with OE

This study supports other findings in which German shepherd dogs, cocker spaniels and shar-peis were predisposed to OE. These breeds are prone to conformational problems, while they are also prone to allergies (Muller 1990, Huang and Huang 1999, Yoshida and others 2002, Rosser 2004). This is probably the reason why these breeds have a more frequent incidence of OE than other breeds. Cocker spaniel and Brittany spaniel dogs were very commonly presented with OE in another recent survey, in which the cocker spaniel was highly susceptible to grass awn foreign bodies as the main cause of OE (Saridomichelakis and others 2007). This primary cause was not included in the cases presented here.

Rods were more frequently isolated in cocker spaniel, German shepherd, boxer and Pekinese breeds. Interestingly, the Labrador retriever was presented less frequently with rods than other breeds. This could be correlated with the breed’s predisposition to allergies (Zur and others 2002), and we found that there were fewer cases of allergies in which rods were detected. However, the association between the incidence of allergies and signalment was not examined in this study. We did not find a statistically significant association between the presence of Malassezia spp. among the different breeds. In another study, Malassezia was positively correlated with OE in poodles and also positively correlated in German shepherd dogs with normal ears (Girao and others 2006). In that report, Malassezia was more frequently found in young dogs (one to three years of age). In this study, there were no statistically significant differences between the various ages at onset groups and the presence of Malassezia spp., although we could demonstrate a higher prevalence in dogs less than five years old. The higher frequency of Malassezia spp. in younger dogs and in allergies is well correlated with the fact that allergies are diagnosed more frequently in dogs less than six years old (Scott 2001b). From our data, we can conclude that otitis is more severe when it is first diagnosed in older dogs. Almost all the dogs greater than five years old had cocci and had rods at significantly higher levels. This can be explained by chronic changes and the possibility that milder otitis had not been diagnosed when these dogs were younger.


From the data obtained in this study, it can be concluded that primary causes or predisposing factors do not have a major influence on the pathogens of OE. Allergies, endocrinopathies and conformational abnormalities cause chronic OE, and the stage of the disease could not be determined in this study. Moreover, the severity of the disease may be influenced by the age at onset. This could explain a possibly more severe disease in endocrinopathies, which are usually diagnosed in older dogs, and a less severe disease in allergies, which are usually diagnosed in younger dogs.


The authors thank Prof Hylton Bark and Dr Mertyn Malkinson for proof reading.

Conflict of interest

None of the authors of this article has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper.