Comparison of Microbial Isolates and Susceptibility Patterns From the External Ear Canal of Dogs With Otitis Externa

Introduction

Otitis externa is defined as inflammation of the epithelium of the external ear canal.¹ Infectious otitis externa commonly occurs as a secondary complication of hypersensitivity disorders (e.g., atopic dermatitis, cutaneous adverse reactions to food, contact hypersensitivities) and of other primary diseases or predisposing factors. Such factors include conformational defects (e.g., stenotic ear canals, excess hair in ear canals), excess moisture (e.g., “swimmer’s ear”), obstructive ear disease (e.g., neoplasia, polyps), foreign bodies, ectoparasites, metabolic diseases, keratinization abnormalities, and autoimmune diseases.¹

Many studies have described the more common microbial isolates and susceptibility patterns from ears with otitis externa.^2–6 Recent reports have proposed that different microbial isolates and susceptibility patterns may be present in the same ear of dogs with otitis externa and otitis media.^7–8 These studies pose the question as to how useful culture and susceptibility testing are in the clinical management of canine otitis externa. One of the reports recommended that culture and susceptibility testing be done from both the external ear canal and middle ear cavity in all cases of chronic otitis externa.⁸ This recommendation was based on the finding of bacterial otitis media in 82.6% of cases of bacterial otitis externa where the tympanic membrane was intact. In addition, different bacteria were sometimes isolated, or more than one strain of the same organism with different susceptibility patterns was isolated from both the external ear canal and the middle ear canal of the same ear.⁸

A more recent study suggested that in addition to differences in the microbial population between the external ear canal and the middle ear cavity in cases of chronic otitis externa, there also exists a difference in the microbial population and their susceptibilities in the horizontal ear canal from a single sampling site within the same ear canal.⁷ This suggestion would imply that the information obtained from culture and susceptibility testing may not be as valuable in guiding the selection of appropriate antibiotic therapy as previously suggested. Therefore, it may be most appropriate to select the initial treatment protocol empirically, based only on the stained microscopic appearance of the microbial population in the otic exudate, and reserve the use of bacterial culture and susceptibility testing for circumstances when the infection fails to improve after empirical therapy.

The results of the above studies leave the clinician with conflicting recommendations regarding when to perform culture and susceptibility testing and the value of the test results. If the microbial populations and susceptibility patterns differ from the same site within the external canal, then culture and susceptibility testing may not be as useful as previously believed. One of the purposes of this study was to identify the microbial isolates and their susceptibility patterns from two samples taken from the same location within the external ear canal of dogs with otitis externa. The null hypothesis of this study was that the two samples taken from the same location should be identical and have the same susceptibility pattern. The culture results of this study were also compared with the stained morphological appearance of the microbes in the otic exudate from the same site. The objective was to determine the frequency of which the morphological appearance of the stained microbial organisms from exudate correlated with the species of microbial organisms isolated on culture.

Materials and Methods

Cases were selected from dogs with otitis externa that were presented for evaluation to the Veterinary Medical Teaching Hospital’s dermatology service at Michigan State University. To be included in the study, no systemic antibiotics and no topical otic medications were allowed for at least 4 weeks prior to evaluation. The study was designed to include either a total of 25 dogs or a total of 50 ears, as some patients had otitis externa in only one ear canal at the time of presentation. Any underlying diseases confirmed at the time of presentation were also to be recorded.

An ear canal was considered to have otitis externa based on clinical evaluation, otoscopic examination, and microscopic examination of stained otic exudates. Organism counts of ≥5 cocci per oil immersion field (OIF), ≥1 rod per OIF, or ≥5 yeast per OIF upon examination of at least 20 microscopic fields at 1000×(OIF) were used as indicators of over-colonization of the external ear canal.^7–9 The number of inflammatory cells present in each case was not quantified, and their presence or absence was not used as an inclusion or exclusion criterion in this study.

For culture and susceptibility sampling, two minitip culturettes^a were simultaneously placed through a sterile otoscopic cone and guided down the vertical ear canal to the juncture of the horizontal canal (both ears were sampled separately in cases of bilateral otitis externa).

After samples were collected for culture, swab samples were taken from the external ear canal at the same juncture of the horizontal and vertical ear canals for immediate microscopic examination. The swab samples were transferred onto glass slides, gently heat-fixed, and stained with a modified Wright’s stain.^b The slides were scanned on low power (10×) to find a representative area for estimation. Organisms were counted from 20 OIFs, the number of each type of organism observed per field was recorded, and the mean number of each type of organism was calculated per field.

If the ear canal was defined as having otitis externa based on the criteria listed above, the samples collected for culture were then submitted for bacterial isolation and susceptibility testing. The culture samples were labeled group A and group B for the author’s records, but they were submitted to the Veterinary Teaching Hospital’s microbiology laboratory with only a number identification, so that the laboratory would be blinded to the fact that the samples were taken from the same ear. Both swabs were plated onto enriched blood agar, Columbia Colistin Nalidixic acid agar, cycloheximide chloramphenicol agar, and MacConkey’s agar. They were incubated at 30° to 37°C for 18 to 24 hours for isolation of microorganisms. Susceptibility patterns were determined using minimum inhibitory concentrations, with two to three colonies being selected for susceptibility testing for each bacterial isolate. The frequency of isolation of the same organisms from groups A and B was evaluated using a binomial distribution equation analysis.

Results

Samples were obtained from 33 different dogs. Sixteen dogs had unilateral otitis externa, with samples taken from only the affected ear canal. Seventeen dogs had bilateral otitis externa and samples were taken from both ears. A total of 50 ear canals were sampled. A diagnosis of the primary disease considered as the cause for the otitis externa was not established at the time of presentation for a majority (19; 57.6%) of the dogs. Nine (27.3%) dogs were diagnosed with atopic dermatitis, four (12.1%) had a cutaneous adverse reaction to food, and one (3.0%) was hypothyroid.

Discussion

The most important findings of this study are the apparent differences in the microbial populations from the same sampling site of the external ear canal in cases of otitis externa. These findings are in agreement with recent studies examining the infectious organisms present in cases of chronic otitis externa.⁷⁸ In the study reported here, different bacterial organisms were isolated from the same anatomical location in 20% of the cases, and the same bacterial organisms, but with different antimicrobial susceptibilities were isolated from the same anatomical location in 20% of the cases. One possible reason for previous treatment failure was the assumed development of in vivo resistance to the antibiotic used during the initial treatment. However, this study suggests that another possible reason for treatment failure when using culture and susceptibility testing as a guide for the selection of topical or systemic antibiotic(s), is that a single swab sample submitted for culture and susceptibility testing may not reveal the total population of organisms that are present in the external ear canal.

Another interesting finding in this study is that 16/33 (48.5%) of the dogs with otitis externa had only unilateral disease. Previous studies reported unilateral disease in 27.3%,⁸ 35%,³ and 59.5%¹⁰ of otitis cases. When bilateral otitis externa was present, only 4/17 (23.5%) dogs had identical culture results from both ears. One previous study also reported a low rate of similarity between affected ears.⁸

Although establishing the primary underlying disease responsible for the otitis externa was not an objective of this study, the authors were able to determine at least one of the underlying causes in 14/33 (42%) cases on initial presentation.

Malassezia spp. and Staphylococcus intermedius were the most common microbial isolates identified in this study, and these findings were similar to previous reports.⁸¹⁰ The majority of the bacterial isolates were susceptible to a number of antimicrobials; however, Pseudomonas aeruginosa and Enterococcus spp. were the most resistant bacteria. Pseudomonas aeruginosa has previously been reported to be the most resistant microbe.³⁸¹⁰

The morphology of the organisms identified on the stained otic swab samples was consistent with the culture results in only 34 (68%) of the 50 otic samples from the same anatomical location. The primary discrepancy was that the cultures revealed a greater and more diverse population of microorganisms than that suggested from examination of the stained slides. A likely reason for this discrepancy is that examination of stained slides evaluated only 20 OIFs, rather than the entire slide. Complete examination of the entire slide might have revealed additional microorganisms, but it would be a very time consuming and unrealistic procedure in clinical practice. Also, the minitip culturette samples used for culturing would only need a small number of organisms for growth to occur on the media under optimal laboratory conditions. This small number of organisms could easily have been missed on a single stained slide preparation. Assuming this last statement to be true, it is possible that examination of stained slides of the otic exudate may be a more realistic representation of the primary organisms responsible for the otitis externa.

In regard to the examination of the yeast population, the authors chose five yeast per OIF as significant, versus three or four yeast per OIF (as cited in previous studies) in order to increase the threshold for inclusion in the study. In the authors’ experience, otic samples have occasionally been negative cytopathologically for yeast but been positive for both bacteria and yeast on culture. The presence of yeast on these cultures was reported as an additional finding by the laboratory even though a fungal culture was not specifically requested. The results of this study further support this discrepancy, because four samples had cytopathological examinations that were negative for yeast, yet culture results were positive.

Conclusion

When interpreting cytopathology and culture results from dogs with otitis externa, the clinician must use caution in assuming that these samples truly represent the full spectrum of microorganisms present. Findings in this study support the concept that initial treatment of otitis externa can be empirically based on historical information of the most common isolates and their susceptibility patterns, in conjunction with examination of stained otic swab samples. When treatment failure occurs, repeated examinations and modification of the treatment may be aided by examination of multiple samples of otic exudate and by submitting cultures. Based on this study there appears to be greater limitations in the usefulness of these procedures, especially when used alone, than previously realized. Further studies are needed to address the efficacy and value of cytopathology and culturing of otic exudates from cases of otitis externa. Additionally, it is very important to consider addressing the underlying causes of otitis externa in chronic and recurrent cases.

Acknowledgments

The authors thank The American College of Veterinary Dermatology for funding this project, the veterinary microbiology laboratory at Michigan State University for performing the cultures and susceptibility testing, Joseph Hauptman for statistical assistance, Elaine Striler for technical assistance, and Annette Petersen for contribution of cases to the database.