Comparison of bacterial organisms from otic exudate and ear tissue from the middle ear of untreated and enrofloxacin‐treated dogs with chronic end‐stage otitis

Abstract

Cultures of otic exudate from the tympanic bulla to identify the infective organism(s) are usually performed in dogs suspected of otitis media. If the infection extends into the otic tissue and antibiotic selection is made based on exudate results only, treatment failure may result. In addition, dogs may be on systemic antibiotics when cultures are performed, potentially affecting culture results. The objective of this study was to compare the organisms in the exudate and tissue obtained from the middle ear of dogs with chronic end‐stage otitis. Additionally, the effect of concurrent antibiotic (enrofloxacin) administration on this comparison was assessed. It was hypothesized that the organisms identified from the exudate and tissue would match biochemically and morphologically, regardless of antibiotic administration. Samples for bacterial culture, one from tissue and one from exudate, were obtained from the middle ear of 23 dogs undergoing a unilateral total ear canal ablation and bulla osteotomy. These dogs were selected from 30 dogs enrolled in a study measuring concentrations of enrofloxacin in ear tissue. The dogs were randomized to one of four treatment groups or a control group. Seven dogs were not included since they did not have a sample from the middle ear exudate. Nineteen dogs (treatment group) received two intravenous doses of enrofloxacin (four dogs received 5 mg/kg, five dogs each received 10, 15 or 20 mg/kg) within a 24‐h period of being sampled, while four dogs (control group) did not receive any antibiotic. Samples for the tissue and exudate were routinely processed for aerobic bacterial culture. Bacterial organisms were identified biochemically and morphologically. The data were analysed using the Fisher's Exact test and Kruskal–Wallis test to analyse discrepancies between organisms identified. One hundred and thirty‐eight organisms were identified, 71 from tissue and 67 from exudate. One hundred and eight organisms (78%), 54 from tissue and 54 from exudate, matched biochemically and morphologically. Thirty organisms (22%) were unmatched, 17 from tissue and 13 from exudate. There were no significant differences (P = 0.23) detected between the dose of enrofloxacin administered and the number of matched and unmatched organisms identified in the tissue and exudate. When treatment groups were combined, discrepancies between organisms in the tissue and exudate were identified in 15 of 19 (79%) dogs that received enrofloxacin compared to only one of four (25%) dogs in the control group (P = 0.067). Based on these results, samples for culture should be obtained from both tissue and exudate to identify all organisms. If only one sample is obtained, tissue would be best. Although there were no significant differences between the dose of enrofloxacin administered and the number of matched and unmatched organisms in the tissue and exudate, when the treatment groups were combined and compared to the control group, the differences approached significance (P = 0.067). This suggests it would be best to obtain cultures off antibiotics. These differences would need to be assessed using more dogs, since the lack of significance was likely due to the low numbers in the control group. Funding: Bayer Animal Health, The Ohio State University Canine Research Fund, American College of Veterinary Dermatology, Ohio Animal Health Foundation.