The beta-adrenergic agonist zilpaterol hydrochloride may predispose feedlot cattle to cardiac remodeling and dysfunction

The aim of this study was to address producer concerns that the Background: β -adrenergic agonist zilpaterol hydrochloride, a bovine growth promotant, predisposes cattle to cardiac disease and death. Our objectives were to evaluate the effect of zilpaterol on cardiac function, morphology, and risk of myocardial injury. A prospective, case-control study was conducted on one feedlot in Methods: northern Colorado using convenience sampling of Angus-based steers (n = 80). Pulmonary arterial pressures (mean, systolic, and diastolic) were measured. Plasma cardiac troponin I was measured in a sub-sample of steers that were followed to slaughter (n = 31). The carcass, left ventricle plus inter-ventricular septum, and right ventricle were weighed. Relative to controls, steers fed zilpaterol hydrochloride had an Results: adjusted left ventricle and septum that was 185 g heavier (95 % CI: 19, 350 g; P = 0.03), a diastolic pulmonary arterial pressure that was 10 mm Hg greater (95 % CI: 3, 17 mm Hg; = 0.004), and a greater concentration of cardiac troponin P I ( = 0.01), a biomarker of myocardial injury. Furthermore, left ventricular mass P tended to be positively and deleteriously associated with diastolic pulmonary arterial pressure in steers fed zilpaterol ( = 0.08) but not controls ( = 0.28). P P Our findings suggest that zilpaterol hydrochloride induced Conclusions: sufficient left ventricular hypertrophy to cause impaired left ventricular relaxation or diastolic dysfunction and myocardial injury. In conclusion, these results support concerns that, in the feedlot studied, zilpaterol hydrochloride predisposes cattle to cardiac disease. 1 2 3 2


Introduction
In the United States, two β-adrenergic agonists (βAA) are approved for the purpose of growth promotion in cattle fed in confinement for slaughter: ractopamine hydrochloride 1 , a predominantly β 1 AA, and zilpaterol hydrochloride 2 , a predominantly β 2 AA.In 2011, approximately 57% of cattle in U.S. feedlots were fed a βAA 3 .Beta-adrenergic agonists are synthetic, structural, and functional analogs of the endogenously produced compound epinephrine (adrenaline).They have the favorable effect of inducing skeletal muscle hypertrophy while decreasing adiposity 4 .
Chronic β-adrenoreceptor stimulation, however, may also cause left ventricular hypertrophy 5 , a strong, independent risk factor for death in humans [6][7][8] .Moreover, a large observational study concluded that 40 to 50% of deaths among feedlot cattle treated with the β 2 adrenergic receptor-selective agonist zilpaterol hydrochloride (Zilmax®, Merck Animal Health, Summit, NJ, USA) were attributable to the drug 9 .The etiology of this increased mortality risk has not been investigated, but anecdotal reports suggest that cattle fed β 2 AA are at increased risk of heart failure.This is biologically plausible; mortality associated with chronic use of β 2 AA use in humans has been primarily attributed to myocardial disease 10,11 .
The aim of this study was to address the concerns that zilpaterol hydrochloride may predispose cattle to heart failure.An in-situ, prospective, case-control study was performed at one feedlot in northern Colorado, U.S.A., to evaluate the effect of zilpaterol hydrochloride on bovine cardiopulmonary physiology, cardiac morphology, and the risk of myocardial injury.

Ethical considerations
The Colorado State University Animal Care and Use Committee approved of this project (Protocol 13-4451A).Permission to conduct the study at the feedlot and abattoir was granted by the owners prior to the start of sampling.All efforts were taken to ameliorate animal suffering by minimizing the time that steers were restrained for the purposes of this study.

Study site and sampling
Black-hided, Angus-based steers that were within 30 days of slaughter were studied in one feedlot located 1,440 m above sea level in northern Colorado (Table 1).The sample size was limited to 80 due to voluntary market withdrawal of the drug during the course of the study.Cattle were procured from the western U.S. through the auction market system and directly from producers.On arrival at the feedlot the cattle weighed approximately 300 kg.They were vaccinated against bacterial and viral pathogens, injected with an anthelmintic, implanted with a hormone growth promotant, and given an ear tag with a unique identification number.CS that had been treated for respiratory disease, were studied on that day.Following the withdrawal of the drug from the market, only CS were available for further study.

Pulmonary arterial pressure (PAP) measurement
A full description of the equipment, materials, and facilities required for PAP testing is provided elsewhere 12 .In brief, an 8.9 cm (3.5 inch) 12-gauge needle was inserted into the jugular vein through which flexible, saline-filled catheter/polyethylene tubing was threaded.The catheter tip was then fed through the right atrium and ventricle and into the pulmonary artery.A pressure transducer connected the catheter to an oscilloscope (BM5Vet, Bionet America Inc., Tustin, CA, USA).The position of the catheter was determined from the pressure waveform on the oscilloscope.The jugular vein, right atrium, right ventricle, and pulmonary artery have distinct pressure waveforms 12 .Mean, systolic, and diastolic pressures were recorded after the pulmonary pressure waveforms had stabilized.Diastolic PAP is hemodynamically influenced by left ventricular chamber pressures shortly prior to contraction, and therefore provides an indirect measure of end-diastolic left ventricular filling pressure 13 .Pulmonary arteriolar wedge pressure is a more direct measure of left ventricular end-diastolic pressure.Wedge pressures are, however, more difficult and time-consuming to obtain in cattle because the fluid-filled catheters used are not balloontipped; consequently, to minimize the time that cattle spent in the chute, arteriolar wedge pressures were not collected.

Blood analysis
Cardiac troponin I analyses were performed on a single day using plasma samples that had been obtained from all steers that were followed to slaughter.The plasma had been stored at -80° C and thawed at room temperature prior to analysis (cTnI, i-STAT 1, Abaxis, Union City, CA).

Hot carcass weight (HCW) and ventricular weights
In the original study design, all cattle at the feedlot that died within the late feeding period were to be evaluated postmortem throughout the duration of the study.Due to an insufficient number of feedlot personnel, however, postmortem data were not collected.Cardiac dimension data were, therefore, collected at slaughter from the remaining 2 cohorts of steers: the 11 ZS studied on August 29 th , 2013, and slaughtered on September 5 th ; and the 20 CS studied on October 21 st and slaughtered on October 29 th , 2013.Only steers that were healthy at the time of sampling and had not been treated for disease while at the feedlot were followed to slaughter.
The steers were slaughtered at a nearby USDA-approved facility.The individual animal identification number of every animal in the study remained with the carcass and heart throughout the slaughter process.This allowed the hot carcass weight (HCW, mass of the unchilled carcass after removal of the head, hide, and viscera) to be paired with cardiac mass.After inspection by a USDA veterinarian, the heart was removed from the processing line, placed in a plastic bag, and stored in ice.
The atria were dissected from the ventricles at the level of the atrioventricular valves and the ventricles subsequently divided into a left ventricle plus inter-ventricular septum (LVS) and right ventricular free wall (RV).The ventricles were weighed using a scale with precision to 0.045 kg (0.1 lb).All cardiac tissue was dissected and weighed within 3-hours of slaughter.

Statistical analyses
Data were analyzed using statistical software (STATA version 12.1, Stata Corporation, College Station, Texas, USA).
Statistics are provided as an adjusted average (mean or median) ± 95% confidence interval of the average.The animal served as the experimental unit.Only steers that were apparently healthy at the time of testing and had not been treated for any disease while at the feedlot were included in statistical analyses.Steers that had been treated, or needed treatment, for disease were not included in any statistical analyses due to the small sample size.
Between-group statistical differences in PAP (3 variables) were assessed using quantile regression for nonparametric data while controlling for pen-level clustering ('qreg2' command) 14 .In subsequent analyses, pen-level clustering was not a concern because ZS and CS that were followed to slaughter belonged to the same 2-pens.Differences in right and left ventricular masses were assessed using linear regression analyses.A Wilcoxon rank-sum test (Mann-Whitney U test) was performed for evaluation of cardiac troponin I due to the non-parametric distribution of residual error about the mean ('ranksum' command).Finally, the effect of left ventricular mass per unit of HCW on diastolic PAP was evaluated, stratified by the feeding of zilpaterol.Left ventricular mass per unit of HCW was chosen as the dependent variable rather than LVS because of the potential collinearity between LVS and HCW.A linear regression analysis was performed for ZS and, due to lack of normality of the standardized residuals, quantile regression was performed for CS ('qreg' command).Evaluations of model fit included graphical and statistical assessments of residual error normality (Shapiro-Wilk test), heteroskedasticity (Breusch-Pagan/Cook-Weisberg test), and linearity.

Results
The results of this study indicate that, in the feedlot studied, steers fed the βAA zilpaterol hydrochloride had greater left ventricular hypertrophy and plasma cardiac troponin I, a biomarker of myocardial injury, than controls.The mass of the left ventricle tended to be positively associated with diastolic PAP in ZS, but there was no association between left ventricular mass and diastolic PAP in CS.The mass of the LVS per unit of HCW tended to be associated with diastolic PAP in ZS (P = 0.08) but not CS (P = 0.28) (Figure 1A).In ZS, a 0.1 kg increase in LVS per 100 kg of HCW was associated with a 9 mm Hg (95% CI: -1, 19 mm Hg) increase in diastolic PAP.

Cardiac troponin I (cTnI)
Plasma cardiac troponin I (cTnI) concentrations were significantly greater in ZS than CS (P = 0.01).The greatest concentrations were observed in the two steers with the most hypertrophied left ventricles (Figure 1B).

Discussion
The findings of this study indicate that, in the feedlot studied, the β 2 AA zilpaterol hydrochloride is associated with left ventricular hypertrophy and myocardial injury.Furthermore, the tendency in our study for left ventricular mass to be positively and deleteriously associated with diastolic PAP in steers fed zilpaterol hydrochloride suggests that βAA-induced left ventricular hypertrophy likely caused diastolic dysfunction.
Left ventricular hypertrophy is a product of chronic β-adrenoreceptor activation and is a major risk factor for cardiovascular morbidity and mortality in humans and animals 5,15 .Two major sequelae of ventricular hypertrophy pertinent to our study are diastolic dysfunction 16 and myocardial ischemia 17 .Diastolic dysfunction is defined as failure of the myofibrils to return to their resting length and ventricular filling is slow or incomplete unless atrial pressure increases 16 .A hallmark of diastolic dysfunction is an increase in left ventricular end-diastolic pressure 16 .
End-diastolic pressure is positively correlated with diastolic PAP, even when pulmonary vascular resistance or heart rate is elevated 13,18 ; consequently, the positive association between left ventricular mass and diastolic pulmonary arterial pressure in steers fed zilpaterol, but not controls, suggests that the myocardial hypertrophic effects of zilpaterol induced diastolic dysfunction.This is not unreasonable: the myocardial hypertrophic action of βAA, such as isoprenaline, are used for the study of diastolic dysfunction in animal models of heart failure [19][20][21] .
Hypertrophy also predisposes to myocardial ischemia 17 .This may explain why steers fed zilpaterol had significantly greater cardiac troponin I concentrations than controls, the latter having concentrations similar to those reported in healthy cows 22 .Intriguingly, the two steers with the greatest cardiac troponin I concentrations also had the greatest left ventricular hypertrophy.This may have been coincidental, but the research literature suggests otherwise.Left ventricular hypertrophy in humans is a strong predictor of death from cardiovascular disease 6 .Without histological evaluation of the myocardial tissue, however, we cannot say if the statistical difference observed was biologically relevant.
Myocardial injury has, to our knowledge, not been previously associated with zilpaterol in cattle.A recent study of 30-Angus steers found no histologic evidence of apoptosis associated with the βAA zilpaterol and ractopamine 23 .Myocardial necrosis and fibrosis have, however, been reported in association with ßAA in other species 24,25 .The myocardial injury observed in association with zilpaterol in our study could have been exacerbated by poor air quality since the inhalation of particulate matter may exacerbate cardiopulmonary injury induced by ßAA 26,27 .This may, in part, explain the seasonal variation observed in mortality risk associated with zilpaterol hydrochloride 9 .
The health implications of diastolic dysfunction in cattle fed βAA may be extensive: diastolic dysfunction in humans increases the risk of pulmonary edema 28 and all-cause mortality 29,30 .Given that the findings of our study suggest that β 2 AA-induced left ventricular hypertrophy caused diastolic dysfunction, and that an increased in all-cause mortality has been associated with the feeding of βAA to cattle 9 , studies to evaluate the effect of ßAA on cardiac function and morphology are needed.Suitable echocardiographic measures of diastolic ventricular function include transmitral filling velocity, ventricular relaxation velocity, and pulmonary vein flow.Pulmonary arteriolar wedge pressures would provide a more accurate measure of left ventricular end-diastolic pressure than diastolic PAP.Balloon floatation catheters are superior to the saline-filled polyethylene tubing typically used for PAP measurement 12 because arteriolar occlusion is necessary for accurate wedge pressure measurement.
A limitation of our study design is that it was necessary to forfeit external validity to maintain internal validity.This allowed us to successfully address the aim of this study.To achieve internal validity, cattle were not randomized into treatment groups because, in the feedlot studied, the feeding of zilpaterol was at the discretion of the feedlot manager.Concerned that zilpaterol may increase the risk of death, particularly among cattle with a history of disease, the feedlot manager only fed zilpaterol to cattle in pens with a low incidence of morbidity and mortality.Given that respiratory disease predisposes cattle to heart failure 31 , any confounding, if present, should have biased the results towards the null.Large, randomized studies are necessary to validate our findings because our results cannot be extrapolated to other feedlots and the number of cattle studied was small.
In conclusion, our results raise concerns that, in the feedlot studied, zilpaterol hydrochloride predisposes cattle to cardiac disease.Healthy cattle fed the βAA zilpaterol hydrochloride had left ventricular hypertrophy that was positively associated with diastolic PAP and myocardial injury.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com

Figure 1 .
Figure 1.The relationship between A) diastolic pulmonary arterial pressure (PAP) and B) plasma cardiac troponin I with mass of the left ventricle plus inter-ventricular septum (LVS) per 100 kg of hot carcass weight (HCW).Left ventricular hypertrophy in steers fed zilpaterol hydrochloride tended to be positively associated with diastolic PAP.Cardiac troponin I concentrations were significantly greater in steers fed zilpaterol hydrochloride.The two steers with the greatest left ventricular hypertrophy had the greatest diastolic pulmonary arterial pressures (A) and the greatest cardiac troponin I concentrations (B).

Table 1 . The number of cattle tested according to the date of testing in the year 2013.
Cardiac hypertrophySteers fed zilpaterol hydrochloride had hypertrophy of the LVS.Steers fed zilpaterol had an adjusted LVS mass that was 185 g (95% CI: 19, 350 g) heavier than CS (P = 0.03).Right ventricular mass did not differ between ZS and CS (P = 0.33).95% CI: 450, 619 g) in ZS.The adjusted mean mass of the LVS was 1.512 kg (95% CI: 1.419, 1.605 kg) in CS and 1.697 kg (95% CI: 1.570, 1.824 kg) in ZS.