Comparison of the diagnostic performance of bacterial culture of nasopharyngeal swab and bronchoalveolar lavage fluid samples obtained from calves with bovine respiratory disease

Authors: CAPIK, SARAH - Kansas State University; WHITE, BRAD - Kansas State University; LUBBERS, BRIAN - Kansas State University; APLEY, MICHAEL - Kansas State University; DEDONDER, KEITH - Kansas State University; LARSON, ROBERT - Kansas State University; Harhay, Gregory; Chitko-Mckown, Carol; Harhay, Dayna; KALBFLEISCH, THEODORE - University Of Louisville; Schuller, Genevieve - Gennie; Clawson, Michael - Mike

Submitted to: American Journal of Veterinary Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/23/2016
Publication Date: 3/4/2017
Citation: Capik, S.F., White, B.J., Lubbers, B.V., Apley, M.D., DeDonder, K.D., Larson, R.L., Harhay, G.P., Chitko-McKown, C.G., Harhay, D.M., Kalbfleisch, T.S., Schuller, G., Clawson, M.L. 2017. Comparison of the diagnostic performance of bacterial culture of nasopharyngeal swab and bronchoalveolar lavage fluid samples obtained from calves with bovine respiratory disease. American Journal of Veterinary Research. 78(3):350-358.

Interpretive Summary: INTERPRETIVE SUMMARY: Bovine respiratory disease (BRD) is a major animal welfare and economic issue for the U.S. cattle industry, and is caused by numerous microbial and viral pathogens. Three microbial pathogens that are associated with BRD are Mannheimia haemolytica, Pasteurella multocida, and Histophilus somni; all three can be found in the upper respiratory tract of cattle with or without signs of BRD, and in the lungs of cattle with BRD. Two sampling methods used for the detection of these three pathogens are deep nasopharyngeal swabs, which target the upper respiratory tract, and bronchoalveolar lavages, which target the lower respiratory tract (lungs). In this study, the two sampling methods were compared for the detection of M. haemolytica, P. multocida, and H. somni, in a small number of cattle with BRD following treatment with gamithromycin. Gamithromycin is a macrolide antibiotic that was first approved for the treatment of BRD in the United States in 2011. Substantial but imperfect agreement was found between the two methods for the detection of all three pathogens five days post-treatment; nasopharyngeal swabs did not always predict what was in the lungs but if an organism was not present in the upper respiratory tract then it was generally not cultured from the lower respiratory tract at the same time in this group of cattle. Strain typing of M. haemolytica isolates from the upper and lower respiratory tracts of animals with BRD indicated that the same strain was present in the upper and lower respiratory tract of a given animal when both samples were positive. However, different strains of M. haemolytica were detected between animals and it was possible to obtain a mixture of strains as well as a mixture of antimicrobial resistance levels to gamithromycin in a single sample by testing multiple isolates from the sample. Antimicrobial resistance testing is routinely only performed on a single isolate from a sample, and may or may not represent all the antimicrobial resistance levels present within that sample or animal. This was a small study that should be followed up with additional ones to determine the repeatability and statistical significance of the results but it highlights the complexity of BRD and indicates the need for caution when interpreting tests for BRD pathogens and antibiotic resistance.

Technical Abstract: Objective: Examine the culture results, gamithromycin susceptibility, predictive values, and agreement of pooled bilateral nasopharyngeal swabs (NPS) and bronchoalveolar lavages (BAL) for identification of Mannheimia haemolytica genotypes, Pasteurella multocida, and Histophilus somni in calves treated for Bovine Respiratory Disease (BRD). Animals: Twenty-eight mixed-breed calves diagnosed with clinical BRD. Procedures: Bilateral pooled NPSs were obtained from all calves pre- and 5 days post- gamithromycin treatment, and blind BALs were performed on all calves 5 days post-treatment. Additional samples were obtained 0.5 and 1 day post-treatment from subsets of calves. Up to 12 M. haemolytica, 6 P. multocida, and 6 H. somni isolates were selected from each sample for gamithromycin susceptibility testing; whole-genome sequencing was performed on all M. haemolytica isolates. Kappa values and predictive values for organism presence were calculated from the 28 paired samples obtained on day 5 post-treatment. Results: Prevalence of M. haemolytica, P. multocida, and H. somni was low 5 days post-treatment and kappa values for BAL/NPS were 0.71, 0.81, and 0.78 respectively. When BAL results were considered the gold standard, the positive and negative predictive values of NPS were: 66.7% and 100% for M. haemolytica; 75% and 100% for P. multocida; and 100% and 96% for H. somni, respectively. Gamithromycin susceptibility of isolates varied within the same sample and between paired NPS and BAL samples. Conclusions and Clinical Relevance: In this population, NPS and BAL showed high agreement and a high negative predictive value of NPS along with the potential for different susceptibility profiles from paired NPS and BAL samples.