Occurrence and Predictors of Bacterial Respiratory Tract Infections and Antimicrobial Resistance Among Isolates From Dogs Presented With Lower Respiratory Tract Infections at a Referral Veterinary Hospital in South Africa

Daniel Nenene Qekwana; Vinny Naidoo; James Wabwire Oguttu; Agricola Odoi

doi:10.3389/fvets.2020.00304

Occurrence and Predictors of Bacterial Respiratory Tract Infections and Antimicrobial Resistance Among Isolates From Dogs Presented With Lower Respiratory Tract Infections at a Referral Veterinary Hospital in South Africa

Front Vet Sci. 2020 Jun 2:7:304. doi: 10.3389/fvets.2020.00304. eCollection 2020.

Authors

Daniel Nenene Qekwana¹, Vinny Naidoo², James Wabwire Oguttu³, Agricola Odoi^{1

3

4}

Affiliations

¹ Section Veterinary Public Health, Department of Paraclinical Sciences, Faculty of Veterinary Science, University of Pretoria, Pretoria, South Africa.
² Faculty of Veterinary Science, University of Pretoria Biomedical Research Centre, Pretoria, South Africa.
³ Department of Agriculture and Animal Health, College of Agriculture and Environmental Sciences, University of South Africa, Johannesburg, South Africa.
⁴ Department of Biomedical and Diagnostic Sciences, College of Veterinary Medicine, University of Tennessee, Knoxville, Knoxville, TN, United States.

Abstract

Background: Respiratory tract infections (RTIs) associated with Pasteurella multocida, Bordetella bronchiseptica, Streptococci, Staphylococci, and Pseudomonas species have been reported in dogs. The objective of this study was to investigate the occurrence and predictors of bacterial RTIs and antimicrobial resistance among samples from dogs with lower RTIs at a referral veterinary teaching hospital in South Africa. Methods: Records of 157 dogs with lower RTIs presented to the veterinary teaching hospital between 2007 and 2013 were included in the study. Crude and factor-specific proportions of RTIs and antimicrobial resistance by breed, season, year, sex, age category, and specimen type were computed. Chi-square or Fisher's exact tests were used to compare proportions of RTIs and antimicrobial resistant isolates across categorical variables. Associations between breed, season, year, sex, age, specimen, and odds of RTIs or multidrug resistance were assessed using Generalized Estimating Equations. Results: There was only one sample per clinical case and bacterial RTIs were observed in 53.5% of the samples tested. Pasteurella species (23.5%) were more common than other species. Almost all (99.5%) isolates were resistant to at least one antimicrobial, while 64.7% were multidrug resistant (MDR). Additionally, 17.0% and 3.3% showed evidence of extensive drug resistance (XDR) and pan-drug resistance (PDR), respectively. The majority of MDR isolates were resistant to penicillin-G (90.9%), lincomycin (100%), tylosine (75.8%), lincospectin (73.7%), ampicillin (72.5%), and kanamycin (68.4%). None of the investigated predictors had significant association with RTIs or antimicrobial resistance. Conclusion: Pasturella species were the most common causes of RTIs. The high levels of MDR and the presence of both XDR and PDR isolates raise the question of the effectiveness of the current antimicrobial therapy used in patients with RTIs in referral hospitals. Given the high level of resistance observed in this study, it is advisable that the choice of antimicrobials for treatment of RTIs be based on antibiograms. This will ensure use of the most efficacious antimicrobials and will minimize treatment failures among cases presented with RTIs.

Keywords: Generalized Estimating Equations; South Africa; antimicrobial resistance; canine; dog; logistic regression; multidrug resistance; respiratory tract infections.