Limitations of bacterial culture, viral PCR, and tulathromycin susceptibility test methods from upper respiratory tract samples in predicting the outcome of tulathromycin control or treatment of bovine respiratory disease in high risk feeder heifers

A cross-sectional prospective cohort study to correlate BRD clinical outcomes for tulathromycin metaphylaxis/treatment for bovine respiratory disease (BRD) with the results of bacterial culture and tulathromycin susceptibility from isolates of deep nasopharyngeal swabs (DNS) as well as viral polymerase chain reaction (PCR) results from nasal swabs revealed poor correlation of bacterial culture and tulathromycin susceptibility with response to tulathromycin metaphylaxis or treatment. 1031 heifers, assumed to be at high-risk (>40% expected BRD morbidity rates), were procured and transported to a research feedlot in Texas. Isolation rates from DNS collected on arrival and at first treatment respectively were: Mannheimia haemolytica (10.9% & 34.1%); Pasteurella multocida (10.4% & 7.4%); Mycoplasma bovis (1.0% & 36.6%); and Histophilus somni (0.7% & 6.3%). Prevalence of BRD viral nucleic acid on nasal swabs collected at first treatment were: PI-3V (34.1%); BVDV (26.3%); BoHV-1 (10.8%); and BRSV (54.1%). Increased relative risk of treatment failure was associated with positive viral PCR results, PI-3V (1.2644), BVDV (1.3917), BHV-1 (1.5156), and BRSV (1.3474) from nasal swabs collected at first pull and culture of M. haemolytica (1.2284) from DNS collected at arrival; however, no other statistically predictable risk of treatment outcomes were measured from DNS for bacterial isolation or tulathromycin susceptibility for M. haemolytica or P. multocida at arrival or first treatment. Predictive values of bacterial culture and tulathromycin susceptibility were substantially lower than the 85% level expected with susceptibility testing. These results indicate tulathromycin susceptibility testing of isolates of M. haemolytica or P. multocida from DNS collected on arrival or at first pull unreliably predict clinical efficacy of tulathromycin for BRD control or treatment most likely due to impacts of unpredictable risk factors and other viral and/or bacterial BRD comorbidities.


CAS Score
Appearance Descriptors 0

No BRD
No BRD clinical signs; calf is healthy and Bright, Alert, and Responsive when approached.

Mild BRD
Calf looks ill until approached by observer (calf brightens up, moves readily and appears normal); if observer loses sight of calf and upon re-exam of the pen, he/she CANNOT readily locate and ID calf as ill. Mild depression; slower in movement but no signs of weakness; small amount of serous nasal discharge; slight ocular discharge.

Moderate BRD
Calf is obviously ill with BRD; the calf shows no change in appearance when approached by observer (does not brighten up and moves slowly or reluctantly). If observer loses sight of calf, upon re-exam of pen, he/she CAN easily locate and ID calf as ill. Moderate depression; signs of weakness or "knuckling" and calf may be reluctant to stand or move about pen; some shallowness apparent in left flank; considerable serous nasal discharge or moderate amount of mucopurulent nasal discharge; dyspnea or respiratory rate is increased; cough or coughing episodes are present.

Severe BRD
Calf is severely ill with BRD; the calf shows no change in appearance when approached by observer; abnormal respiration and/or depression present. Severe depression; stumbling or moves with extreme prodding; obvious lack of fill in left flank signaling anorexia; head lowered or extended to facilitate breathing; may be open-mouthed breathing with considerable noise (expiratory grunts, moans); copious mucopurulent to purulent nasal discharge; cough or coughing episodes are present. Administer emergency intervention therapy.

(Moribund)
Calf is moribund and near death -calf in general is not ambulatory, cannot rise from recumbency, and cannot acquire food or water. Very likely cannot be removed from pen for treatment without mechanical transport. Euthanize humanely. 156 Power calculation was based on the expected incidence of tulathromycin resistant M. haemolytica 158 isolates to show 85% agreement with treatment failures. The calculation was based on 1000 head 159 sampled at arrival and a 30% incidence of M. haemolytica isolation and 1% incidence of 160 tulathromycin resistance (3 head) and 275 head sampled at first pull with an incidence of 50% M. 161 haemolytica isolation and 30% incidence of tulathromycin resistance (21 head) for a total of 24 162 animals with tulathromycin resistant M. haemolytica isolates. Due to low or no tulathromycin 163 resistance reported in the literature, it was not expected to have enough tulathromycin resistant 164 isolates from arrival samples and therefore the protocol was designed to combine all samples for 165 susceptibility analysis. 166 Randomization of individual cattle or lots of cattle was not necessary because of the study design, 167 i.e., all cattle received the same arrival protocol (Table 2) administered by individual body weight 168 and label instructions. Certified scales were calibrated each day before enrollment of cattle. 169 Laboratory personnel were masked to the origin of samples and research personnel were masked to 170 the laboratory results until the completion of the study. This study complied with all applicable 171 animal welfare regulations related to the humane care and use of animals and all procedures and 172 products were reviewed and approved by an animal use and welfare committee. Any animal 173 identified with critically severe clinical signs (CAS 3), was given immediate emergency therapy and 174 removed from the study. Any animal found to be moribund (CAS 4), was removed from the study 175 and humanely euthanized per the American Veterinary Medical Association "Guidelines for 176 Euthanasia of Animals," 2013 Edition. 24 Products in the arrival and treatment protocols were 177 administered per Beef Quality Assurance Guidelines (BQA). (Table 2) A timeline of the study 178 design is illustrated in Figure 1. 179 Table 3.

ANALYSIS EQUATIONS
294 295 2x2 Contingency Table  True   363 Three heifers were found to be persistently infected with BVDV, with agreement of both labs, and 364 were removed from the study within the first 5 days. After the initial 7-day treatment suspension 365 period (PMI) following the metaphylaxis administration of tulathromycin, 402 heifers (38.9%) were 366 identified by the investigating veterinarian as having clinical signs of BRD, thus classified as 367 treatment failures of tulathromycin metaphylaxis. Mean day of first pull was 13 days with a range 368 per load of 8-16 days. Morbidity per load ranged from 23% to 44% and treatment failure ranged 369 from 23% to 49% after tulathromycin metaphylaxis and 38% to 69% after first pull tulathromycin 370 therapy. There were three instances where animals were given BRD treatments that deviated from 371 the protocol resulting in removal of two animals from the study, after initial processing, and the          Another limitation of this study was the low prevalence of resistant isolates to tulathromycin in 586 arrival cattle. The population of high-risk feeder heifers was selected to mitigate bias from variable 587 proportions of steers and bulls and the risk of BRD associated with castration. This population was 588 also selected because tulathromycin resistance was found in samples from previous cattle from 589 these sources that also had poor TSR to tulathromycin, which was assumed to be from high 590 prevalence of tulathromycin resistance. Because tulathromycin resistance at arrival was expected to 591 be the limiting factor for sample size, the arrival and first pull susceptibility tests were combined. 592 The authors admit that this is a limitation to the internal validity of the study because arrival and 593 first pull cattle can be considered different populations. However, considering the low prevalence 594 of tulathromycin resistance at arrival, in this population of high-risk feeder heifers, the authors feel,