Mannheimia haemolytica isolated from feedlot cattle were tested for tulathromycin resistance. Cattle were sampled over a 3-year period, starting 12 months after approval of tulathromycin for prevention and treatment of bovine respiratory disease. Nasopharyngeal samples from approximately 5,814 cattle were collected when cattle entered feedlots (N = 4) and again from the same cattle after ≥60 days on feed. The antimicrobial use history for each animal was recorded. Mannheimia haemolytica was isolated from 796 (13.7%) entry samples and 1,038 (20.6%) ≥ 60 days samples. Of the cattle positive for M. haemolytica, 18.5, 2.9, and 2.4% were administered therapeutic concentrations of tulathromycin, tilmicosin, or tylosin tartrate, respectively. In addition, 13.2% were administered subtherapeutic concentrations of tylosin phosphate in feed. In years one and two, no tulathromycin-resistant M. haemolytica were detected, whereas five isolates (0.4%) were resistant in year three. These resistant isolates were collected from three cattle originating from a single pen, were all serotype 1, and were genetically related (≥89% similarity) according to pulsed-field gel electrophoreses patterns. The five tulathromycin-resistant isolates were multi-drug resistant also exhibiting resistance to oxytetracycline, tilmicosin, ampicillin, or penicillin. The macrolide resistance genes erm(42), erm(A), erm(B), erm(F), erm(X) and msr(E)-mph(E), were not detected in the tulathromycin-resistant M. haemolytica. This study showed that tulathromycin resistance in M. haemolytica from a general population of feedlot cattle in western Canada was low and did not change over a 3-year period after tulathromycin was approved for use in cattle.