DISSERTATION         EPIDEMIOLOGICAL,  PHYSIOLOGICAL  AND  GENETIC  RISK  FACTORS  ASSOCIATED  WITH   CONGESTIVE  HEART  FAILURE  AND  MEAN  PULMONARY  ARTERIAL  PRESSURE  IN  CATTLE                 Submitted  by     Joseph  Michael  Neary   Department  of  Clinical  Sciences           In  partial  fulfillment  of  the  requirements       For  the  Degree  of  Doctor  of  Philosophy       Colorado  State  University       Fort  Collins,  Colorado       Summer  2014           Doctoral  Committee:       Advisor:  Franklyn  Garry     Co-­‐Advisor:  Milton  Thomas       Christopher  Orton     Mark  Enns   Paul  Morley   Timothy  Holt                                       Copyright  by  Joseph  Michael  Neary  2014   All  Rights  Reserved   ii   ABSTRACT         EPIDEMIOLOGICAL,  PHYSIOLOGICAL  AND  GENETIC  RISK  FACTORS  ASSOCIATED  WITH   CONGESTIVE  HEART  FAILURE  AND  MEAN  PULMONARY  ARTERIAL  PRESSURE  IN  CATTLE           Congestive  heart  failure,  secondary  to  pulmonary  hypertension,  has  historically   been  considered  a  disease  associated  with  high  altitude  exposure.  The  disease  was  first   reported  to  occur  at  altitudes  over  2,440  m  (8,000  ft.)  and  so  became  known  as  “high   altitude  disease”.  One  common  clinical  sign  due  to  congestive  heart  failure  in  cattle  is   swelling  of  the  brisket.  Consequently,  the  disease  also  became  known  as  “brisket  disease”.     In  more  recent  years,  congestive  heart  failure  has  been  reported  to  occur  in  both   beef  and  dairy  cattle  at  a  more  moderate  altitude  of  1,600  m.  Anecdotal  reports  from  cattle   producers  in  Nebraska,  Colorado  and  Texas  suggest  that  the  incidence  of  congestive  heart   failure  may  be  increasing.  This  suggests  that  bovine  congestive  heart  failure  is  not  strictly  a   disease  of  high  altitude  exposure.       Anatomical  studies  of  cattle  indicate  that  cattle  have  a  smaller  lung  volume  and   alveolar  surface  area  available  for  gas  exchange  than  mammals  with  similar  body  masses   and  oxygen  requirements.  This  may  be  because  selection  for  increased  growth  rate,  and   other  traits  of  high  production,  increases  metabolic  oxygen  demand.  The  overarching   hypothesis  of  this  doctoral  dissertation  was  that  congestive  heart  failure  secondary  to   pulmonary  hypertension  is  not  strictly  a  disease  of  high  altitude  but,  a  multifactorial   disease,  that  is  also  associated  with  physiological  traits  that  increase  metabolic  oxygen   demand  relative  to  oxygen  supply  via  the  cardiopulmonary  system.  The  goal  of  this     iii   doctoral  dissertation  was  to  identify  epidemiological,  physiological  and  genetic  risk  factors   associated  with  congestive  heart  failure  and  increased  mean  pulmonary  arterial  pressure   in  cattle.     The  results  of  this  dissertation  indicate  that  pulmonary  arterial  pressures  of  cattle   are  substantially  higher  than  other  mammalian  species.  Among  pre-­‐weaned  calves,  mean   pulmonary  arterial  pressures  increased  significantly  with  age  even  at  the  moderate  altitude   of  1,470  m.  As  hypothesized,  high  oxygen  demand  relative  to  supply  was  positively   associated  with  mean  pulmonary  arterial  pressure  in  both  pre-­‐weaned  calves  at  high   altitude  (2,170  m)  and  feedlot  cattle  at  moderate  altitudes  (1,300  m).  A  study  of  10   Canadian  feedlots  indicated  that  the  risk  of  congestive  heart  failure  increased  from  the  year   2000  to  the  year  2012.  The  risk  of  congestive  heart  failure  increased  more  than  the   underlying  change  in  the  risk  of  digestive  disorders.  Death  from  congestive  heart  failure   occurred  throughout  the  feeding  period  but  typically  occurred  late  in  the  feeding  period,   which  makes  this  disease  particularly  costly  to  the  feedlot  industry.  Treatment  for   respiratory  disease  was  a  significant  risk  factor  for  CHF.    Increased  growth  rate  and   increased  feed  efficiency  were  risk  factors  for  increased  mean  pulmonary  arterial  pressure   in  cattle.  Mean  pulmonary  arterial  pressures  were  significantly  higher  at  the  end  of  the   confined  feeding  period  at  moderate  altitude  (1,300  m)  than  in  pre-­‐weaned  calves  at  high   altitude  (2,170  m).  Growth  promotion  through  a  steroid  implant  containing  estradiol  and   trenbolone  acetate  did  not  significantly  increase  mean  pulmonary  arterial  pressure  as   hypothesized.  However,  diastolic  pulmonary  arterial  pressure  was  significantly  lower  than   non-­‐implanted  controls,  which  suggests  that  one  or  both  of  these  steroid  hormones  has   cardio-­‐pulmonary  protective  effects.  Genome-­‐wide  association  analyses  of  mean     iv   pulmonary  arterial  pressure  and  traits  physiologically  associated  with  mean  pulmonary   arterial  pressures  among  calves  at  4  and  6  months  of  age  did  not  identify  any  concordant   single  nucleotide  polymorphisms  (SNPs).  However,  multiple  SNPs  were  identified  to  be   associated  with  mean  and  systolic  pulmonary  arterial  pressures  that  have  been  associated   with  pulmonary  hypertension  in  humans  or  have  a  plausible  biological  role  in  the   development  of  pulmonary  hypertension.  In  conclusion,  the  results  of  these  investigations   provide  evidence  to  suggest  that  congestive  heart  failure  of  cattle  is  a  multifactorial  disease   that  is  exacerbated  by  high  altitude  exposure.         v   ACKNOWLEDGMENTS         I  offer  my  sincere  appreciation  to  my  committee  members  for  their  guidance  and   hard  work.  In  particular,  I  would  like  to  acknowledge  my  advisor  Dr.  Frank  Garry  who  has   changed  the  course  of  my  life.  I  will  be  forever  grateful.     I  would  also  like  to  thank  my  family  and  friends  for  their  support  and   encouragement.  In  particular,  I  would  like  to  thank:  Dr.  Elizabeth  Fraser,  Bill  Trampe,  Barb   East  and  my  wife  Gretchen.               vi   TABLE  OF  CONTENTS         ABSTRACT……………………………………………………………………………………………………………..     ii   ACKNOWLEDGEMENTS……….…………………………………………………………………………………     iv   CHAPTER  1:  INTRODUCTION…………………………………………………………………………………     1   CHAPTER  2:  BOVINE  PULMONARY  HYPERTENSION:  A  REVIEW  OF  THE   EPIDEMIOLOGY,  GENETICS  AND  PATHOPHYSIOLOGY…………………………………………….       6   CHAPTER  3:  THE  RISK  OF,  AND  RISK  FACTORS  FOR,  CONGESTIVE  HEART  FAILURE   IN  NORTH  AMERICAN  FEEDLOTS…………………………………………………………………………..       29   CHAPTER  4:  PULMONARY  ARTERIAL  PRESSURES  IN  CALVES  AT  4  DIFFERENT   ALTITUDES…………………………………………………………………………………………………………...       48   CHAPTER  5:  CHANGES  IN  PULMONARY  ARTERIAL  PRESSURE  WITH  AGE……………….     64   CHAPTER  6:  SYSTEMIC  OXYGEN  EXTRACTION  IS  POSITIVELY  ASSOCIATED  WITH   MEAN  PULMONARY  ARTERIAL  PRESSURE  AND  THE  ODDS  OF  CALF  MORTALITY   AT  HIGH  ALTITUDE  ………………………………………………………………………………………………         86   CHAPTER  7:  SYSTEMIC  TISSUE  OXYGEN  EXTRACTION  AND  MEAN  PULMONARY   ARTERIAL  PRESSURE  IN  FEEDLOT  CATTLE……………………………………………………………       103   CHAPTER  8:  HIGH  GROWTH  RATE  AND  HIGH  FEED  EFFICIENCY  ARE  POSITIVELY   ASSOCIATED  WITH  MEAN  PULMONARY  ARTERIAL  PRESSURE  AND  SYSTEMIC   OXYGEN  EXTRACTION  IN  CATTLE………………………………………………………………………….         117   CHAPTER  9:  A  SLOW-­‐RELEASE  TRENBOLONE  ACETATE  AND  17-­‐β-­‐ESTRADIOL   GROWTH  IMPLANT  REDUCED  DIASTOLIC  PULMONARY  ARTERIAL  PRESSURE  IN   FEEDLOT  CATTLE  …………………………………………………………………………………………………         132   CHAPTER  10:  GENOME-­‐WIDE  ASSOCIATION  STUDY  OF  PULMONARY  ARTERIAL   PRESSURE  AND  TRAITS  ASSOCIATED  WITH  PULMONARY  ARTERIAL  PRESSURE  IN   BOVINE  CALVES  ……………………………………………………………………………………………………         143   CHAPTER  11:  DISCUSSION……………………………………………………………………………………..     176   REFERENCES…………………………………………………………………………………………………………     184       1   CHAPTER  1:  INTRODUCTION         Background   Historically  congestive  heart  failure,  secondary  to  pulmonary  hypertension,  was   only  considered  to  be  problematic  at  high  altitude  (>  2,130  m)  (Glover  and  Newsom,  1915;   Hecht  et  al.,  1962).  Pulmonary  hypertension,  and  death  from  congestive  heart  failure   secondary  to  pulmonary  hypertension,  is  still  problematic  in  calves  at  high  altitude  (Neary   et  al.,  2013a;  Neary  et  al.,  2013b).  More  notably,  congestive  heart  failure  secondary  to   pulmonary  hypertension  has  also  been  reported  to  occur  at  more  moderate  altitudes  (Hull   and  Anderson,  1978;  Jensen  et  al.,  1976;  Malherbe  et  al.,  2012;  Pringle  et  al.,  1991).     Anatomically,  cattle  have  a  substantially  smaller  lung  volume  and  alveolar  surface   area  (Veit  and  Farrell,  1978)  than  mammals  of  a  similar  body  mass  and  oxygen   requirement.  Functional  maturity  of  the  pulmonary  system  is  not  achieved  until   approximately  1-­‐year  of  age  in  Holstein  calves  (Lekeux  et  al.,  1984;  Neary  et  al.,  2014a).   Therefore,  hypobaric  hypoxia  of  high  altitude  may  exacerbate  any  physiological   inadequacies,  if  present,  in  meeting  the  high  oxygen  requirements  associated  with  the   growth  and  development  of  young  calves.  A  study  conducted  at  altitudes  over  2,130  m   revealed  that  pre-­‐weaned  calves  were  markedly  hypoxic  despite  high  alveolar  ventilation   rates,  as  indicated  by  hypocapnia  (Neary  et  al.,  2013a).    The  calves  also  showed  evidence  of   compromised  oxygen  delivery  to  peripheral  tissues,  as  indicated  by  L-­‐lactate   concentrations  over  1.5  mmol/L  (Neary  et  al.,  2013a).           2   Goals  and  objectives   Studies  of  bovine  cardio-­‐pulmonary  anatomy  and  physiology  in  combination  with   anecdotal  reports  of  congestive  heart  failure  occurring  at  moderate  altitudes  and  genetic   selection  for  increased  productivity  suggest  that  a  physiological  imbalance  between  oxygen   supply  and  demand  may  be  a  leading  risk  factor  for  congestive  heart  failure  secondary  to   pulmonary  hypertension.  The  overarching  hypothesis  of  this  doctoral  dissertation  was  that   bovine  pulmonary  hypertension  is  not  only  a  disease  of  high  altitude  but  also  a  disease  of   high  production;  high  altitude  exacerbates  any  underlying  physiological  imbalance   between  oxygen  supply,  via  the  cardio-­‐pulmonary  system,  and  metabolic  oxygen  demand.   The  rationale  being  that  domestic  cattle  have  little  cardio-­‐pulmonary  reserve,  which  limits   their  ability  to  respond  to  physiological  stressors  that  either  increase  oxygen  demand,  such   as  rapid  growth  or,  decrease  oxygen  supply,  such  as  high  altitude  hypobaric  hypoxia,  or   both.   In  humans  a  positive  diagnosis  of  pulmonary  hypertension  is  made  when  mean   pulmonary  arterial  pressure  is  over  25  mm  Hg  at  rest  and  over  30  mm  Hg  during  exercise.   There  are  no  criteria  in  place  for  establishing  a  positive  diagnosis  of  pulmonary   hypertension  in  cattle.  Mean  pulmonary  arterial  pressure  in  cattle  varies  with  age,  altitude,   environmental  conditions,  diet  and  health-­‐status  (Holt  and  Callan,  2007).  Therefore,  in  the   investigations  that  follow,  pulmonary  hypertension  was  not  defined  based  on  a  fixed  mean   pulmonary  arterial  pressure;  rather,  it  was  more  loosely  defined  to  mean  that  an  individual   had  a  mean  pulmonary  arterial  pressure  that  was  higher  than  the  mean  value  of  that   individual’s  contemporaries  within  the  cohort  studied.  Further  studies,  outside  the  realm  of     3   this  dissertation,  are  required  to  define  breed-­‐specific  reference  intervals  for  mean,  systolic   and  diastolic  pulmonary  arterial  pressures.     It  should  also  be  pointed  out  that  pulmonary  hypertension  is  not  synonymous  with   congestive  heart  failure.  Congestive  heart  failure  is  not  an  inevitable  consequence  of  high   mean  pulmonary  arterial  pressure.  There  are  multiple  factors  that  influence  an  individual’s   susceptibility  to  heart  failure  secondary  to  pulmonary  hypertension.  Such  factors  include:   how  acutely  the  increase  in  pulmonary  arterial  pressure  occurred;  whether  there  is   regurgitation  of  blood  through  the  tricuspid  valve;  how  compliant  the  pulmonary  arterial   walls  are;  and,  if  there  is  one  or  more  concurrent  illnesses.  Again,  further  studies  outside  of   this  dissertation  are  needed  to  evaluate  the  prognostic  value  of  pulmonary  arterial   pressure  measurement  in  cattle.  For  simplicity,  and  for  the  purpose  of  this  dissertation,  an   animal  with  a  higher  mean  pulmonary  arterial  pressure  than  another  animal  within  the   cohort  studied  was  considered  to  be  at  greater  risk  of  developing  congestive  heart  failure.     The  goal  of  this  doctoral  study  was  to  identify  epidemiological,  physiological  and   genetic  risk  factors  for  bovine  pulmonary  hypertension.  This  was  achieved  through  the   following  objectives:   1. An  epidemiological  study  of  congestive  heart  failure  in  US  and  Canadian   feedlots.  It  was  hypothesized  that  the  risk  of  congestive  heart  failure  has   significantly  increased  over  time.  The  rationale  being  that  genetic  selection  practices   within  the  beef  industry  for  increased  rate  of  growth  have  increased  metabolic   oxygen  demand  relative  to  cardio-­‐pulmonary  capacity.  If  there  has  been  a   progressive  reduction  in  bovine  cardio-­‐pulmonary  reserve  it  could  manifest  as  a     4   progressive  increase  in  the  incidence  of  CHF.  Select  risk  factors  for  congestive  heart   failure  relative  to  digestive  disorders  were  also  evaluated.   2. A  study  of  calf  cardio-­‐pulmonary  physiology  in  association  with  altitude.  It  was   hypothesized  that  calves  located  at  higher  altitude  would  show  a  greater  increase  in   mean  pulmonary  arterial  pressure  (mPAP)  with  age  than  calves  located  at  lower   altitudes.       3. An  evaluation  of  pulmonary  arterial  pressures  with  age;  from  calfhood  into   the  confined  feeding  period.  It  was  hypothesized  that  mPAP  would  increase   during  the  feeding  period  and,  that  the  calves  with  the  highest  mPAP  pre-­‐weaning   would  have  the  highest  mPAP  at  the  end  of  the  confined  feeding  period.   4. A  study  to  determine  if  inadequate  oxygen  delivery,  as  indicated  by  increased   systemic  oxygen  extraction  fraction  (sOEF),  is  associated  with  mPAP  and   increased  odds  of  mortality  in  calves  at  high  altitude.  It  was  hypothesized  that   sOEF  would  be  positively  associated  with  mPAP  and  increased  odds  of  mortality.   5. A  study  to  determine  if  inadequate  oxygen  delivery,  as  indicated  by  the  sOEF,   is  associated  with  mPAP  in  feedlot  cattle  at  moderate  altitude.    It  was   hypothesized  that  sOEF  would  be  positively  associated  with  mPAP.  The  rationale   being  that  high  oxygen  demand  relative  to  supply  is  not  only  a  risk  factor  for   congestive  heart  failure  among  calves  at  high  altitude  but  is  also  a  risk  factor  for   congestive  heart  failure  among  feedlot  cattle  at  moderate  altitude.   6. A  study  to  determine  if  the  rate  of  body  mass  gain  is  associated  with  mPAP  and   sOEF  in  pre-­‐weaned  calves  at  high  altitude  (altitude  2,170  m)  and  in  cattle   during  the  confined  feeding  period  (altitude  1,300  m).    It  was  hypothesized  that     5   growth  rate  would  be  positively  associated  with  mean  pulmonary  arterial  pressure   and  sOEF.  The  rationale  being  that  rapidly  growing  animals  have  a  high  oxygen   demand,  which  mediates  an  increase  in  cardiac  output.  Systemic  OEF  increases   when  the  increase  in  cardiac  output  associated  with  an  increase  in  oxygen  demand   is  insufficient  to  meet  oxygen  requirements.     7. A  study  to  determine  if  growth  promotion  induced  by  a  steroid  implant   increased  the  risk  of  pulmonary  hypertension  in  feedlot  cattle.  It  was   hypothesized  that  feedlot  steers  administered  a  growth  hormone  implant  would   develop  a  higher  mPAP  and  sOEF  than  non-­‐implanted  control  steers.     8. A  genome-­‐wide  association  study  of  pulmonary  arterial  pressure  and  traits   associated  with  mPAP  in  calves.  It  was  hypothesized  that  SNPs  associated  with   mPAP  would  also  be  associated  with  physiological  variables  found  to  be   physiologically  associated  with  mPAP.               6   CHAPTER  2:  BOVINE  PULMONARY  HYPERTENSION:  A  REVIEW  OF  THE  EPIDEMIOLOGY,   GENETICS  AND  PATHOPHYSIOLOGY     In  1913  in  South  Park,  Colorado  two  CSU  researchers,  George  Glover  and  Isaac   Newsom,  set  out  to  investigate  a  strange  new  disease  of  beef  cattle.  This  was  the  first   investigation  of  bovine  congestive  heart  failure  (CHF)  secondary  to  pulmonary   hypertension  (BPH)  in  history.  One  hundred  years  later,  CHF  is  still  a  problematic  disease   for  cattle  producers,  perhaps  to  an  even  greater  degree.  The  goal  of  this  chapter  was  to   provide  an  up  to  date  comprehensive  review  of  the  epidemiology,  genetics  and   pathophysiology  of  the  disease.  It  is  not  an  exhaustive  description  of  all  studies  undertaken   on  this  subject.  The  main  findings  and  ideas  that  have  been  advanced  have  been  cited  as   parsimoniously  as  possible.  For  a  more  comparative  review  of  the  physiology  of  hypoxic   pulmonary  hypertension,  see  Rhodes  (2005).     EPIDEMIOLOGY  OF  BOVINE  PULMONARY  HYPERTENSION     The  incidence  and  risk  factors  for  CHF  among  cattle  at  high-­‐altitude   There  are  few  studies  of  the  epidemiology  of  CHF.  The  first  report  of  CHF  contained   perhaps  the  most  detailed  information  regarding  the  epidemiology  of  the  disease  (Glover   and  Newsom,  1915).  At  the  time  of  that  publication  CHF  was  estimated  to  cause  an  annual   death  loss  of  1  to  2  %  of  all  cattle  at  altitudes  over  2,440  m  in  Colorado,  USA.  The  authors   had  not  seen  the  disease  below  an  altitude  of  2,130  m  and  it  was  not  commonly  recognized   in  calves  because  the  clinical  signs  were  similar  to  respiratory  diseases,  such  as  diphtheria.     7   In  addition,  calves  did  not  always  present  with  edema  of  the  brisket  region.  Producers  were   advised  to  take  affected  animals  down  to  an  altitude  of  2,130  to  2,440  m  in  order  to   promote  recovery:    the  effect  being  to  “strengthen  the  heart  or  lessen  its  work”.  There  was   early  evidence  of  a  genetic  predisposition  to  CHF  as  the  authors  reported  that  the  progeny   of  bulls  from  low  altitude  were  particularly  susceptible.  In  order  to  minimize  cardiac   workload  the  authors  suggested  that  exertion  of  cattle  at  high  altitude  should  be   minimized.     Approximately  35  years  later  the  incidence  of  BPH  was  estimated  to  be  0.5  to  2.0  %   of  animals  at  altitudes  over  2,130  m  although  this  varied  from  year  to  year  and  from  herd   to  herd;  in  some  herds,  the  incidence  was  reported  to  reach  as  high  as  10  %  (Alexander  and   Jensen,  1959;  Hecht  et  al.,  1962).  The  authors  reported  that  CHF  primarily  occurred  from   fall  to  spring  with  calves  less  than  1-­‐year  old  most  at  risk.       The  effect  of  climate  on  the  incidence  of  BPH   Climate  is  reported  to  affect  the  incidence  of  CHF;  the  disease  being  more   problematic  in  wet  pastures  than  dry  (Hecht  et  al.,  1959;  Hull  and  Anderson,  1978)  and   during  wet,  cold  summers  and  cold  winters  (Glover  and  Newsom,  1915).  Cold   temperatures  may  affect  pulmonary  arterial  and  venous  vasoconstriction  (see  below).   There  is  currently  no  explanation  for  the  relationship  between  wet  pasture  and  increased   incidence  of  CHF.  However,  tryptophan  in  lush  pastures  is  a  plausible  risk  factor  (see   Discussion  chapter).           8   Trends  over  time   Historically,  CHF  was  considered  to  be  problematic  at  altitudes  over  2,130  m   (Glover  and  Newsom,  1915;  Hecht  et  al.,  1962).  However,  it  has  since  been  reported  to   occur  at  lower  altitudes  among:  feedlot  cattle  at  1,600  m  above  sea-­‐level  (Jensen  et  al.,   1976);  dairy  calves  at  1,600  m  and  289  m  above  sea-­‐level  (Malherbe  et  al.,  2012;  Pringle  et   al.,  1991);  and,  among  cattle  pastured  in  moist  areas  (Hull  and  Anderson,  1978).     BPH  among  feedlot  cattle   There  are  even  fewer  studies  of  the  epidemiology  of  CHF  in  feedlot  cattle.  In  1974,  a   study  conducted  across  4  feedlots  located  at  an  altitude  of  1,600  m  reported  the  attack  risk   of  CHF  to  be  2.85  cases  per  10,000  yearling  cattle  entering  the  feedlot  (Jensen  et  al.,  1976).   Cases  of  CHF  were  also  reported  to  occur  predominantly  during  the  last  half  of  the  feeding   period  (Jensen  et  al.,  1976),  which  would  make  CHF  a  particularly  costly  disease.  Other  risk   factors  reported  to  increase  the  risk  of  CHF  include:  previous  mountain  grazing,   hypoventilation  and  rapid  rate  of  growth  (Jensen  et  al.,  1976).     BPH  among  dairy  cattle   A  study  conducted  from  2007  to  2011  of  Holstein  heifers  at  a  heifer-­‐raising  facility   and  2  dairies  located  at  the  modest  altitude  of  1,600  m  reported  that  CHF  secondary  to   pulmonary  hypertension  was  the  second  leading  cause  of  death  loss,  behind  respiratory   disease  (Malherbe  et  al.,  2012).  Congestive  heart  failure  accounted  for  22  %  of  all  deaths   over  a  7-­‐year  period  at  the  heifer-­‐raising  facility  but  no  risk  factors  were  identified   (Malherbe  et  al.,  2012).  Congestive  heart  failure  has  also  been  reported  at  a  much  lower     9   altitude  in  5  to  6-­‐month  old  dairy  calves  on  one  farm  in  Tennessee  (Pringle  et  al.,  1991).   Again,  no  risk  factor  was  identified.     PATHOPHYSIOLOGY  OF  BOVINE  PULMONARY  HYPERTENSION     The  first  study  of  bovine  CHF,  published  in  1915,  concluded  that  some  cattle  do  not   have  sufficient  cardiac  reserve  to  meet  the  demands  of  high  altitude  resulting  in   “exhaustion  of  the  heart”  (Glover  and  Newsom,  1915).    However,  recovery  could  be   achieved  be  taking  affected  animals  down  to  a  lower  altitude.  Interestingly,  they  report  that   one  yearling  calf  brought  down  to  lower  altitude  was  taken  back  to  the  original  altitude  1   month  after  recovering;  7  months  later  the  calf  was  reported  to  be  in  perfect  health.  This   suggests  that  in  addition  to  a  genotype  by  environment  interaction,  there  may  also  be  a   third  variable  of  importance:  age.  When  correcting  for  carcass  weight,  heart  weights  of   healthy  cattle  slaughtered  in  Denver,  Colorado  were  on  average  0.40  kg  heavier  than  the   heart  weights  of  cattle  raised  and  slaughtered  at  sea  level,  supporting  their  notion  that   cardiac  workload  increased  with  increasing  altitude  (Glover  and  Newsom,  1918).     Pulmonary  hypertension  as  a  cause  of  right-­‐sided  congestive  heart  failure   A  study  conducted  at  1,400  m  above  sea-­‐level  reported  that  calves  from  3-­‐months  to   1-­‐year  of  age  with  BPH  in  the  acute  phase  (n=  27)  had  a  lower  cardiac  output  than  healthy   controls  (n  =  16)  (Hecht  et  al.,  1962).  It  was  also  found  that  calves  with  “brisket  disease”  or   “high  altitude  disease”,  as  CHF  became  known,  had  higher  mean  pulmonary  arterial   pressures  (mPAP)  and  higher  atrial  pressures,  both  left  and  right,  than  control  animals     10   (Hecht  et  al.,  1962).  Therefore,  it  became  apparent  that  the  clinical  signs  and  gross   pathology  of  right-­‐sided  congestive  heart  failure  associated  with  “brisket  disease”  was   secondary  to  pulmonary  hypertension  (Blake,  1965).  The  correlation  (r  =  0.91)  of  mPAP  in   calves  following  acute  and  chronic  exposure  to  a  simulated  altitude  of  4,572  m  (Will  et  al.,   1975b)  suggested  a  common  pathophysiology,  whether  the  hypoxia-­‐induced  pulmonary   hypertension  was  acute  or  chronic  in  nature.     Hypoxia,  hypoventilation  and  mean  pulmonary  arterial  pressure   The  positive  association  between  mPAP  and  altitude  was  first  reported  by  Donald   Will  (Will  et  al.,  1962).  Twenty  Hereford  steers  of  uniform  weight  (300  kg)  and  breeding   were  obtained  from  one  ranch  at  1,100  m  and  then  taken  to  high  altitude  (3,050  m,  n  =  10)   or  moderate  altitude  (1,525  m,  n  =  10).  Those  taken  to  high  altitude  showed  a  significant   increase  in  mPAP  but  they  did  not  show  an  increase  in  minute  ventilation,  cardiac  output   or  hematocrit  relative  to  the  steers  that  remained  at  moderate  altitude.  These  findings   indicate  a  failure  to  adapt  to  the  hypoxic  conditions  of  high  altitude  and  may  explain  why,   over  a  6  month  period,  the  weight  gain  of  the  steers  maintained  at  the  moderate  altitude   (102  kg)  was  over  twice  that  of  the  steers  taken  to  high  altitude  (45  kg).  Robert  Grover  also   reported  that  following  exposure  to  high  altitude  cattle  had  an  increased  mPAP,  were   hypoxemic  due  to  a  failure  to  maintain  effective  ventilation  and  they  showed  poor  weight   gain  (Grover  et  al.,  1963;  Grover  and  Reeves,  1962).  The  potential  role  of  hypoventilation  in   BPH  was  confirmed  by  Bisgard  and  Vogel  (1971);  hypoventilation  induced  by  excision  of   carotid  bodies  resulted  in  increased  mPAP.     11   Interestingly,  an  increase  in  effective  ventilation  was  observed  in  BPH  ‘resistant’   cows  (n  =  4)  during  pregnancy  and  when  exposed  to  acute  hypoxia  but  not  in  BPH   ‘susceptible’  cows  (n  =  4)(Moore  et  al.,  1979).  Hypoxia  during  the  fetal  and  perinatal  period   alters  development  of  the  pulmonary  vasculature  (Gao  and  Raj,  2011;  Papamatheakis  et  al.,   2013).  Therefore,  inadequate  maternal  ventilation  may  have  detrimental  consequences  on   the  development  of  the  pulmonary  vasculature  of  the  in-­‐utero  calf.   Conversely,  hyperventilation  may  also  have  a  deleterious  effect  on  pulmonary   arterial  pressure.  In  vitro  studies,  that  used  pulmonary  tissue  isolated  from  sheep  and  cats,   suggested  that  chronic  alkalosis  may,  for  an  unknown  reason,  result  in  a  more  pronounced   pulmonary  hypoxic  vasoconstrictive  response  (Gordon  et  al.,  1993).     The  increase  in  mPAP  observed  to  occur  on  exposure  to  high  altitude  was   moderately  reduced  by  administration  of  100  %  oxygen  (Will  et  al.,  1962).  Within  5-­‐ minutes  of  the  supplemental  oxygen  being  removed  mPAP  returned  to  baseline  values.  The   rapidity  with  which  changes  in  mPAP  were  brought  about  by  oxygen  administration  was   indicative  of  hypoxia-­‐induced  pulmonary  vasoconstriction.  This  phenomenon  was  first   reported  to  occur  in  the  cat  (Von  Euler  and  Liljestrand,  1946).       The  primary  site  of  bovine  pulmonary  hypertension  development   Pulmonary  arteriographic  studies  showed  that  “pruning”  of  the  distal  pulmonary   arteries  occurred  in  association  with  BPH  (Alexander  and  Jensen,  1963b).  It  was  suggested   by  the  authors  that  this  vascular  remodeling  could  cause  increased  pre-­‐capillary  resistance   to  flow  thereby  increasing  mPAP.  As  a  result  of  this  finding  an  “emphasis  was  placed  upon   the  small  pulmonary  arteries  in  subsequent  histologic  investigations”.  Medial  hypertrophy     12   of  the  pulmonary  arterioles  was  identified  to  be  the  most  consistent  lesion,  found  in  20  of   the  24  calves  that  died  from  CHF  (Alexander  and  Jensen,  1963c).  However,  the  next  most   consistent  lesion  type  was  intimal  fibroelastosis  of  the  elastic  arteries,  found  in  16  of  the  24   calves.  Intimal  lesions,  mineralization  of  the  intima  and  media,  adventitial  proliferation  and   thrombosis  were  more  commonly  observed  in  the  elastic  arteries  than  either  muscular   arteries  or  arterioles.  Therefore,  although  the  most  consistent  vascular  lesion  involved  the   small  pulmonary  arteries,  lesions  were  apparent  throughout  the  pulmonary  vascular  tree.     Perhaps  the  strongest  evidence  for  the  small  pulmonary  arteries  being  the  primary   site  of  pulmonary  hypertension  development  was  that  medial  hypertrophy  of  these   arteries,  as  determined  by  the  ratio  of  the  area  of  the  media  to  the  area  of  the  intima  plus   the  internal  elastic  membrane,  was  highly  correlated  with  mPAP  (r  =  0.98)  (Alexander  and   Jensen,  1963d).  More  supporting  evidence  for  the  small  pulmonary  arteries  as  the  primary   site  of  BPH  development  came  from  the  discovery  that  the  medial  thickness  of  small   pulmonary  arteries  from  7  species  at  1,600  m  was  correlated  with  pulmonary  arterial   pressure  (r  =  0.88)  and  right  ventricular  hypertrophy  (r  =  0.97)  following  chronic  exposure   to  a  simulated  altitude  of  4,500  m  (Tucker  et  al.,  1975).  The  authors  concluded,  “the   amount  of  pulmonary  vascular  smooth  muscle  inherent  within  each  species  may  determine   the  response  of  each  species  to  high-­‐altitude  exposure”.  This  may  be  true  but  it  should  be   pointed  out  that  the  species  with  the  highest  mPAP  and  pulmonary  medial  hypertrophy  at   4,500  m  (calves  and  pigs)  also  had  the  highest  mPAP  and  greatest  medial  hypertrophy  at   1,600  m.  So,  whether  medial  hypertrophy  of  small  pulmonary  arteries  was  a  cause  or   consequence  of  increased  mPAP  remains  to  be  determined.  Medial  hypertrophy  may  reflect     13   work  hypertrophy  due  to  increased  vasomotor  tone  (Naeye,  1961)  or  occur  in  response  to   altered  hemodynamics,  such  as  increased  pulmonary  flow  (Wagenvoort  et  al.,  1969).     Jaenke  and  Alexander  (1973)  exposed  7  Hereford  cattle,  born  and  raised  at  1,524m,   to  a  simulated  altitude  of  4,572m.  A  gradual  increase  in  mPAP  was  recorded  but  marked   ‘contracture’  of  medial  smooth  muscle  cells  and  endothelial  cells  was  not  evident  until  days   32  to  36.  This  ‘contracture’  was  only  evident  at  the  very  distal  arterial  vasculature.   However,  only  the  intermediate  to  distal  pulmonary  arteries  were  examined.  The  temporal   discordance  between  mPAP  and  medial  hypertrophy  apparent  in  this  study  suggests  that   the  early  increase  in  mPAP  was  not  due  to  vascular  remodeling  in  the  small  pulmonary   arteries.       Hypoxia  and  post-­‐natal  pulmonary  development   Neonatal  hypoxia,  like  antenatal  hypoxia,  may  be  detrimental  to  the  development  of   the  pulmonary  vasculature.  Studies  of  calves  exposed  from  birth  to  an  altitude  of  3,355  m   for  several  months  found:  mPAP  to  increase  over  70  mm  Hg  after  just  2  weeks;  formation   of  smaller  diameter  pulmonary  arteries;  and,  a  more  pronounced  hypoxic  pulmonary   vasoconstrictor  response  than  control  calves  maintained  at  an  altitude  of  305  m  (Reeves   and  Leathers,  1967).  In  contrast,  low  altitude  calves  (altitude  305  m)  showed  a  drop  in   mPAP,  a  decrease  in  the  smooth  muscle  content  of  pulmonary  arteries,  an  increase  in  the   diameter  of  larger  pulmonary  arteries  and  an  increase  in  the  number  of  small  pulmonary   arteries  that  could  be  identified  on  radiographs  (Reeves  and  Leathers,  1967).  Another   study  of  healthy  cattle  at  sea-­‐level  found  that  after  1  year  of  age  the  decrease  in  thickness  of   the  media  relative  to  the  external  diameter  of  the  pulmonary  artery  reached  its  minimum     14   of  approximately  11  %  (Wagenvoort  and  Wagenvoort,  1969).  This,  perhaps  by  no   coincidence,  is  the  approximate  age  at  which  the  functional  maturity  of  the  cardio-­‐ pulmonary  system  is  optimal  (Lekeux  et  al.,  1984;  Neary  et  al.,  2014a).       There  is  also  some  evidence  to  suggest  that  chronic  hypoxia  is  detrimental  to  the   development  of  the  pulmonary  airways.  Lung  airflow  resistance  was  increased  in  neonatal   calves  after  2  weeks  of  exposure  to  a  simulated  altitude  of  4,500m  due  to  increased  fibrous   and  smooth  muscle  content  in  the  bronchioles  and  large  airways  (Inscore  et  al.,  1991).     Hypoxia  and  myocardial  depression     There  is  some  evidence  to  suggest  that  a  reduction  in  myocardial  function  occurs  in   calves  exposed  to  hypoxic  conditions.  The  stroke  index  (volume  of  blood  ejected  from  the   heart  in  one  cardiac  cycle  relative  to  body  surface  area)  of  Holstein  calves  exposed  to  an   altitude  of  3,400  m  decreased  substantially  but  recovered  back  towards  normal  after   administration  of  100  %  oxygen  (Will,  1975).  A  reduction  in  cardiac  index  (cardiac  output   per  minute  relative  to  body  surface  area)  was  also  demonstrated  in  Holstein  calves  at  2  and   4  weeks  of  exposure  to  a  simulated  altitude  of  3,400  m  (Ruiz  et  al.,  1973).       Pulmonary  vein  involvement   Marked  medial  muscularization  of  both  pulmonary  arteries  and  veins  down  to   vessels  of  20  μm  in  diameter  is  present  in  cattle  (Alexander,  1965).  The  pulmonary  veins,   unlike  arteries,  have  a  sphincter-­‐like  appearance  (Figures  2.1  and  2.2)  (Alexander  and   Jensen,  1963a).       15     Figure  2.1:  Cross-­‐section  of  a  pulmonary  artery  (right)  and  vein  (left)  from  a  healthy   feedlot  steer  with  a  mean  pulmonary  arterial  pressure  of  44  mm  Hg  at  an  altitude  of   1,440  m.  Magnification:  10-­‐fold       16     Figure  2.2:  Longitudinal-­‐section  of  a  pulmonary  vein  obtained  from  a  healthy  feedlot   steer  with  a  mean  pulmonary  arterial  pressure  of  67  mm  Hg  at  an  altitude  of  1,440   m.  Magnification:  20-­‐fold     Kuida  et  al.  (1963)  provided  evidence  that  some  cattle  with  CHF  had  increased   resistance  downstream  of  the  pulmonary  bed;  these  cattle  had  pulmonary  arterial  wedge   pressures  that  were  over  6  mm  Hg  greater  than  left  ventricular  end-­‐diastolic  pressure.  The   authors  of  this  study  suggested  that  pulmonary  venoconstriction  or  left  ventricular  failure   might  be  contributing  to  the  development  of  BPH.       17   Both  hypoxia  and  cold  exposure  may  lead  to  pulmonary  venoconstriction.  In  pigs,   alveolar  hypoxia  produced  significant  pulmonary  venoconstriction  although  the  magnitude   was  not  as  great  as  arterial  constriction  (Nelin  et  al.,  1994).  Hereford  calves  exposed  to  cold   temperatures  (-­‐  2  to  1  °C)  for  48  hours  at  both  1,524  m  and  3,048  m  showed  a  significant   increase  in  mPAP,  pulmonary  arterial  wedge  pressure  and  pulmonary  vascular  resistance   but  no  change  in  cardiac  output  (Busch  et  al.,  1985).  The  interpretation  of  these  results   indicates  that  vasoconstriction  of  both  pulmonary  arteries  and  veins  were  contributing  to   pulmonary  arterial  pressure.  The  effect  of  altitude  and  cold  exposure  on  pulmonary  arterial   pressure  was  additive.  Minute  ventilation  and  paO2  fell  on  cold  exposure  but  paCO2   increased.  Oxygen  administration  to  restore  paO2  to  control  values  partially  restored   pulmonary  vascular  resistance  and  mPAP  to  control  values,  which  suggested  that  cold-­‐ induced  alveolar  hypoventilation  was  partly  responsible  for  the  pulmonary  hypertensive   response.     Medial  hypertrophy  of  pulmonary  veins  was  evident  in  some  humans  resident  at   high  altitude  (Wagenvoort  and  Wagenvoort,  1982).  It  seems  likely  that  the  pulmonary   venous  system  and  left-­‐ventricular  function  may  play  a  greater  role  in  the  development  of   pulmonary  hypertension  than  is  currently  appreciated.     Aberrant  cardiopulmonary  hemodynamics  and  pulmonary  hypertension   It  has  been  known  for  over  50  years  that  high  pulmonary  arterial  flow  rates  may  be   sufficient  to  induce  pulmonary  hypertension  in  neonatal  calves  (Vogel  et  al.,  1963).   However,  only  recently  has  it  become  apparent  that  aberrant  hemodynamics  and  vascular   injury  may  contribute  to  the  initiation  and  development  of  BPH  (Botney,  1999).  The     18   afterload  on  the  right  ventricle  is  composed  of  two  components:  hydraulic  resistance   primarily  attributable  to  the  pulmonary  arterioles  and  hydraulic  capacitance,  a  dynamic   load,  attributable  to  the  large,  elastic  pulmonary  arteries,  which  act  as  conduits  between   the  right  ventricle  and  sites  of  gaseous  exchange.  In  neonatal  calves  exposed  to  hypoxia,  the   large,  elastic  pulmonary  arteries  were  found,  in  an  ex  vivo  study,  to  increase  in  stiffness   due  to  elastin-­‐based  extracellular  matrix  remodeling  (Lammers  et  al.,  2008).  Stiffening  of   the  large,  elastic  pulmonary  arteries  has  multiple  adverse  consequences:  a  greater   afterload  that  is  independent  of  the  distal  vascular  resistance;  increased  pulsatility  and   energy  transfer  to  the  small  pulmonary  arteries  creating  a  high-­‐stress  environment;  and,   reduced  flow  efficiency  due  to  a  reduction  in  the  Windkessel  effect  (For  a  review  of   hemodynamics  in  pulmonary  hypertension  see  (Lammers  et  al.,  2012)).  Normal  pulmonary   pulse  pressures  in  healthy  humans  are  approximately  17  mm  Hg  (Figure  2.3).           19     Figure  2.3:  In  humans,  systolic,  mean  and  diastolic  pulmonary  arterial  pressures   average  25,  15  and  8  mm  Hg,  respectively.  From  left  to  right,  the  x-­‐axis  represents   the  flow  of  blood  from  the  pulmonary  artery  to  the  left  atrium.  Mean  pressures  in  the   left  atrium  and  pulmonary  veins  average  2  mm  Hg.  Diagram  adapted  from  Guyton   (2006)     The  result  of  pathological  flow  arriving  in  the  distal  pulmonary  vasculature  is  an   exacerbation  of  small  vessel  dysfunction.  An  in  vitro  study  found  that  pathologically  high   and  low  shear  stress  resulted  in  a  reduction  in  the  release  of  vasodilators  (nitric  oxide,   prostacyclin)  and  increased  release  of  vasoconstrictors  (endothelin,  thromboxane)  from   pulmonary  arterial  endothelial  and  smooth  muscle  cells  isolated  from  neonatal  calves  (Li  et   al.,  2009b).  Shear  stress  is  the  frictional  force  imposed  on  a  vessel  wall  caused  by  blood   flowing  rapidly  through  a  stationary  vessel  lumen.    The  expression  of  2  smooth  muscle   contractile  proteins  (α-­‐SM-­‐actin  and  SM-­‐MHC)  was  significantly  increased  at  high  shear   stress    (≥  90  dynes/cm2)  compared  to  physiological  shear  stress  (20  dynes/cm2).  Increased   pulsatility  due  to  upstream  vascular  stiffening  has  also  been  shown  to  increase     20   inflammatory  gene  expression  and  cell  proliferation  in  the  downstream  endothelial  cells  (Li   et  al.,  2009a).     The  consequences  of  proximal  vascular  stiffening  of  the  pulmonary  artery  may  be   particularly  severe  in  cattle  as  pulmonary  blood  flow  is  reported  to  be  markedly  pulsatile   relative  to  other  mammalian  species  (Weekley  and  Veit,  1995).  It  has  also  been  suggested   that  the  density  of  the  bovine  pulmonary  capillary  bed  is  less  than  other  mammalian   species  (Epling,  1964).  Therefore,  there  may  be  less  dampening  of  the  pressure  wave  due   to  decreased,  or  low  dispersion  of  flow  among  the  micro-­‐circulation.       Toxin-­‐induced   Several  reports  have  indicated  that  CHF  may  be  toxin  induced.  Holstein  bull  calves   fed  Locoweed  (Oxytropis  serica  and  Astragalus  lentiginosus),  which  contains  swainsonine,   an  indolizidine  alkaloid  toxin,  at  3,090  m  were  at  greater  risk  of  developing  CHF  secondary   to  BPH  than  control  calves  (James  et  al.,  1991).  The  risk  of  Locoweed  ingestion  is  magnified   for  calves.  In  addition  to  ingesting  swainsonine  directly  from  the  plant,  which  commonly   grows  on  high  altitude  pastures,  the  toxin  also  accumulates  in  milk  (James  and  Hartley,   1977).  The  pathogenesis  of  swainsonine  toxicity  is  unclear.     Ingestion  of  larkspur  (Delphinium  spp.)  can  cause  neuromuscular  paralysis  through   the  action  of  diterpenoid  alkaloids  (Pfister  et  al.,  1999).  Respiratory  distress  and  rumen   bloat,  which  may  both  result  from  intoxication  may  cause  sufficient  pulmonary   hypertension  to  result  in  congestive  heart  failure.     Water  sulfate  concentration  in  yearling  steers  was  found  to  be  positively  correlated   with  mPAP  (Loneragan  et  al.,  2005).  It  is  likely  that  ruminal  hydrogen  sulfide  gas  is  inhaled     21   since  ruminants  inhale  a  substantial  proportion  of  eructated  gas  (Dougherty  et  al.,  1962).   Hydrogen  sulfide  may  play  a  role  in  the  oxygen  sensing  mechanism  of  pulmonary  arteries.   Under  hypoxic  conditions  pulmonary  metabolism  of  H2S  is  reduced  resulting  in  increased   pulmonary  H2S  levels,  which  may  be  responsible  for  mediating  the  vasoconstrictive   response  to  hypoxia  (Olson  et  al.,  2009).  Therefore,  water  with  high  sulfate  content  may   result  in  the  inhalation  of  hydrogen  sulfide  gas,  which  may  be  a  mediator  of  the  hypoxia-­‐ induced  vasoconstrictive  response  of  the  muscular  pulmonary  arteries.       Reactive  oxygen  species   Reactive  oxygen  species  are  being  increasingly  implicated  in  chronic  hypoxic   pulmonary  hypertension  (for  a  review  see  (Nozik-­‐Grayck  and  Stenmark,  2007)).  The   source  of  reactive  oxygen  species  production  is  not  known  but  there  is  accumulating   evidence  that  there  are  various  isoforms  of  vascular  NADPH  oxidases,  which  may  be   activated  by  tissue  injury  and  hypoxia  and  act  as  a  major  source  of  superoxide  radicals   (Keaney,  2005).       Cold-­‐induced  pulmonary  hypertension   Cold  temperatures  have  been  associated  with  an  increase  in  mPAP  (Busch  et  al.,   1985;  Will  et  al.,  1978).  At  an  ambient  temperature  of  25  °C  cattle  significantly  decreased   cardiac  output  and  increased  systemic  and  pulmonary  vascular  resistances  following   cooling  of  the  skin  but  there  was  no  change  in  alveolar  ventilation,  as  measured  by  arterial   blood-­‐gases  (McMurtry  et  al.,  1975).  This  indicated  that  the  increase  in  vascular  resistance     22   occurring  in  response  to  skin  cooling  was  independent  of  changes  in  alveolar  oxygen   tension.         Gender  differences   Female  animals  have  a  smaller  increase  in  pulmonary  arterial  pressure  following   exposure  to  chronic  hypoxia  (Burton  et  al.,  1968;  McMurtry  et  al.,  1973;  Rabinovitch  et  al.,   1981).  However,  the  gender  effect  appears  to  be  age-­‐dependent:  the  significant  difference   was  only  apparent  in  the  adult  rats,  which  implies  that  a  hormonal  influence  in  post-­‐ pubertal  animals  may  be  responsible  (Rabinovitch  et  al.,  1981).  Female  cattle  appear  to   have  lower  mPAP  than  males.  At  6-­‐months  of  age,  at  an  altitude  of  2,730  m,  the  mPAP  of   heifers  was  approximately  4  mm  Hg  lower  than  steer  calves  although  a  statistical   difference  was  not  detected  (Neary  et  al.,  2013a).  In  that  study,  husbandry  was  uniform  for   heifers  and  steers.  However,  the  relationship  between  gender  and  mPAP  may  have  been   confounded  by  other  sex-­‐related  traits,  such  as  growth  rate.       Recovery  from  bovine  pulmonary  hypertension   Cattle  can  recover  from  BPH  when  moved  to  lower  altitude  (Glover  and  Newsom,   1915).  In  one  study,  cattle  with  clinical  signs  of  CHF  were  moved  to  a  moderate  altitude  of   1,600  m  and  serial  mPAP  measurements  and  lung  biopsies  of  small  pulmonary  arteries   obtained  (Alexander  et  al.,  1965).  For  the  first  6  weeks,  the  mPAP  and  the  ratio  of  the   tunica  media  to  intima  in  small  pulmonary  arteries  remained  at  high  altitude  levels.   However,  after  12  weeks  at  the  lower  altitude,  both  mPAP  and  the  ratio  of  media  to  intima   were  significantly  reduced.  This  suggested  that  vascular  lesions  associated  with  pulmonary     23   hypertension  may  be  reversible.  However,  the  ratio  of  the  tunica  media  to  tunica  intima   accounts  for  only  a  limited  part  of  the  vascular  remodeling  process  that  occurs  in   association  with  pulmonary  hypertension.  Further  studies  are  needed  to  determine  if   pulmonary  remodeling  is  reversible.     GENETICS  OF  BOVINE  PULMONARY  HYPERTENSION     The  hypoxic  pulmonary  vasoconstrictive  response  of  cattle  (Bos  taurus)  is  notably   greater  than  other  mammalian  species  (Tucker  et  al.,  1975).  However,  within  the  Bos   taurus  species  and  within  breeds  there  exists  marked  variability  in  the  pulmonary  pressor   response  to  high  altitude  exposure  (Alexander  et  al.,  1960).  Genetics  and  epi-­‐genetic   regulation  of  gene  expression  likely  has  a  leading  role  in  determining  susceptibility  to  BPH.         Genotype  by  environment  interaction  and  bovine  pulmonary  hypertension   The  first  evidence  of  a  genotype  by  environment  interaction  was  provided  by   Alexander  et  al.  (1960).  They  took  20  yearling  Hereford  steers  from  one  ranch  at  an   elevation  of  1,100m  in  eastern  Colorado  and  exposed  them  to  an  altitude  of  either  1,524  m   (n  =  10)  or  3,048  m  (n  =  10).    The  mPAP  of  control  steers  maintained  at  an  altitude  of  1,524   m  for  6  months  remained  consistent  at  approximately  28  mm  Hg.  However,  the  mPAP  of   the  high  altitude  steers  was  not  uniform:  4  steers,  that  developed  severe  pulmonary   hypertension  6  months  later,  had  significantly  higher  mPAP  values  after  7  weeks  at  3,048   m  than  the  6  steers  that  subsequently  developed  mild  pulmonary  hypertension  6  months     24   later.  The  results  of  this  study  indicated  that  genetic  predisposition  to  pulmonary   hypertension  is  not  phenotypically  apparent  at  all  altitudes.     Evidence  for  a  genotype  by  environment  interaction  is  also  provided  by  Weir  et  al.   (1974).  Two  lines  of  Hereford  cattle  selected  to  be  either  resistant  or  susceptible  to  BPH   showed  physiological  differences  after  2  weeks  at  3,420  m  above  sea  level  but  not  at  the   original  altitude  of  1,500  m  (Weir  et  al.,  1974).  In  response  to  an  increase  in  altitude,   susceptible  cattle  (n  =  7)  showed  a  significantly  greater  increase  in  mPAP,  total  pulmonary   vascular  resistance  and  hematocrit  relative  to  resistant  cattle  (n  =  5).  Susceptible  cattle  also   showed  a  greater  increase  in  cardiac  output  suggestive  of  a  greater  hypoxic  physiological   stimulus  but  a  statistically  significant  difference  was  not  detected.     Similar  results  were  reported  one  year  later.  Calves  born  to  susceptible  animals   showed  a  greater  increase  in  mPAP  when  chronically  exposed  to  3,048  m  than  calves  born   to  resistant  parents  (Will  et  al.,  1975a).  Calves  born  to  susceptible  parents  showed  a   greater  increase  in  mPAP  when  exposed  to  acute  hypoxia  than  calves  born  to  resistant   parents  at  both  5  and  9  months  of  age  (Will  et  al.,  1975b).  The  change  in  mPAP  in  response   to  acute  hypoxia  was  highly  correlated  with  the  change  in  mPAP  after  18  days  at  4,572  m   when  1  year  old.  This  suggested  that  the  genetic  risk  factors  for  a  profound  vasopressor   response  to  acute  hypoxia  were  also  responsible  for  increasing  pulmonary  arterial   pressure  in  response  to  chronic  hypoxia.  However,  at  the  moderate  altitude  of  1,524  m  the   mPAP  values  of  BPH  susceptible  and  resistant  cattle  at  5,  9  and  12-­‐months  of  age  did  not   differ,  once  again  indicating  a  genotype  by  environment  interaction.       25   Native  Simien  cattle  of  Ethiopia  were  reported  to  have  mPAP  of  only  32.5  ±  5.4   mmHg  (n  =  32)  at  an  altitude  of  3,500m  (Wuletaw  et  al.,  2011).  One  very  notable   phenotypic  characteristic  of  these  cattle  is  their  small  mature  body  weight.       Pulmonary  arterial  pressure  and  growth  traits   Selection  for  growth  may  have  the  unfavorable  consequence  of  increasing  mPAP.   Direct  genetic  correlations  between  PAP  and  both  body  mass  at  birth  and  weaning  have   been  estimated  to  be  approximately  0.5  (Levalley,  1978;  Shirley  et  al.,  2008).  No  evidence   of  maternal  genetic  effects  on  the  mPAP  of  offspring  has  been  found  (Shirley  et  al.,  2008).   Darling  and  Holt  (1999)  reported  that  the  correlation  between  sire  PAP  and  son  PAP  (r  =   0.2)  in  their  dataset  of  966  calves  differed  from  the  correlation  between  sire  PAP  and   daughter  PAP  (r  =  -­‐0.01).  The  authors  suggested  this  to  be  due  to  an  abnormality  in  the  Y-­‐ chromosome  causing  variable  penetrance  of  an  autosomal  gene.       Comparative  studies  within  the  Bos  family   Within  the  Bos  family  pulmonary  hypertension  may  be  a  problem  unique  to  the  Bos   taurus  species.  Bos  indicus,  or  zebu,  cattle  are  thought  to  be  resistant  (Hecht  et  al.,  1962).   However,  anecdotal  reports  suggest  that  Bos  indicus  cattle  may  also  be  susceptible.   Historically,  CHF  has  been  more  often  reported  in  beef  breeds  than  dairy  breeds.  Beef  cattle   production,  unlike  dairy  cattle  production,  is  possible  on  the  mountainous  terrain   commonly  associated  with  high  altitude.  Therefore,  relatively  more  beef  cattle  are  exposed   to  high  altitude  than  dairy  cattle.  This  likely  explains  the  greater  occurrence  of  CHF  in  beef   cattle  relative  to  dairy  cattle.  Holstein  dairy  cattle  are  highly  susceptible  to  CHF  secondary     26   to  pulmonary  hypertension  (Malherbe  et  al.,  2012;  Stenmark  et  al.,  1987)  and  are   commonly  used  as  an  animal  model  of  human  pulmonary  hypertension  (Lammers  et  al.,   2008;  Zuckerman  et  al.,  1992).   The  germplasm  evaluation  program  conducted  at  the  U.  S.  Meat  Animal  Research   Center,  Clay  Center,  NE  has  evaluated  divergence  among  breeds  of  beef  cattle  for  over  30   years.  The  results  show  that  breed  differences  have  not  remained  static  over  time.  In  fact,   breed  convergence  has  occurred  on  traits  such  as  yearling  weight,  weaning  weight  and   mature  cow  size  (Cundiff  et  al.,  2004).  This  may  mean  that  the  risk  of  congestive  heart   failure  among  breeds  may  have  also  converged  over  time.   Weir  et  al.  (1974)  suggested  that  susceptibility  to  pulmonary  hypertension  is   genetically  transmitted  in  an  autosomal  dominant  manner.  This  mechanism  of   transmission  was  supported  by  breeding  experiments  of  cattle  (Bos  taurus)  and  yaks  (Bos   grunniens)  (Anand  et  al.,  1986),  close  family  members  adapted  to  life  at  high  altitude.  Dzos   (n  =  6),  which  are  progeny  of  cow  x  yak  crosses,  had  mPAPs  closer  to  yak  than  cattle.  Stols   (n  =  7),  progeny  of  female  dzo  x  B.  taurus  bull  crosses,  had  divergent  mPAP  values   suggestive  of  simple  Mendelian  inheritance.  The  authors  suggested  a  single  autosomal   dominant  genetic  transmission  of  mPAP,  but  given  the  limited  number  of  study  animals  it   would  be  prudent  to  interpret  such  results  with  caution.       Yaks  have  much  reduced  muscularization  of  the  tunica  media  of  the  small   pulmonary  arteries  than  cattle  (Durmowicz  et  al.,  1993;  Tucker  et  al.,  1975).  Unlike  cattle,   yaks  do  not  demonstrate  increased  hemodynamic  forces  under  hypoxic  challenge  and  so   may  lack  a  stimulus  for  BPH  lesion  development.  If  so,  hemodynamic  forces,  rather  than   hypoxia  per  se,  may  be  the  principle  stimulus  for  cell  wall  proliferation  and  extracellular     27   matrix  production  (Durmowicz  et  al.,  1993).  Other  adaptations  of  yaks  to  hypoxic   environments  include:  a  shorter,  wider  trachea,  a  larger  thoracic  capacity  and  a  larger   heart  and  lungs  than  cattle  of  equivalent  body  mass  (Zhang  et  al.,  2000).       Molecular  genetics   Alleleic  association  analyses  of  single  nucleotide  polymorphisms  (SNPs)  of  10  cattle   with  severe  pulmonary  hypertension  and  10  cattle  with  ‘low’  was  performed  using  a  10  k   SNP  array  (Newman  et  al.,  2011).  Due  to  the  low  power  of  the  study,  no  SNPs  were  found  to   be  statistically  associated  with  mPAP.  However,  a  follow  up  study  of  3  genes  (Myosin  heavy   chain  15,  NADH  flavoprotein  2  and  FK  Binding  Protein  1A)  identified  by  Newman  et  al.   (2011)  to  be  associated  with  mPAP,  that  may  plausibly  be  associated  with  pulmonary   hypertension,  were  further  evaluated  in  a  cohort  of  166  yearling  bulls  (Neary  et  al.,  2014b).     It  was  found  that  the  T  allele  (rs29016420)  of  myosin  heavy  chain  15  gene  was  linked  to   lower  mPAP  in  a  dominant  manner.  Interestingly,  the  allelic  frequency  of  the  T  allele  at  the   same  loci  in  a  sample  of  24  Himalayan  yaks  (Bos  grunniens)  was  found  to  be  100  %.     Gene  expression     Transcriptome  analysis  of  peripheral  blood  mononuclear  cells  identified  respiratory   disease  to  be  the  top  disease  process  associated  with  mPAP  (Newman  et  al.,  2011).  Genes   associated  with  cell-­‐signaling,  immune  and  endothelial  cell  functions  were  differentially   expressed  according  to  mPAP.  The  findings  of  Frid  et  al.  (2006)  also  indicate  extensive   involvement  of  the  immune  system  in  the  development  of  bovine  pulmonary  hypertension.   Similarly,  in  humans  (Pullamsetti  et  al.,  2011)  and  broiler  chickens  (Wideman  et  al.,  2013),     28   the  immune  system  is  also  reported  to  play  a  critical  role  in  the  etiology  and  progression  of   pulmonary  hypertension.     Comparative  studies   Similar  to  cattle,  pulmonary  hypertension  in  broiler  chickens  has  also  been   estimated  to  be  moderately  to  highly  heritable  (Lubritz  et  al.,  1995;  Pavlidis  et  al.,  2007).   Various  modes  of  inheritance  have  been  proposed  from  a  few  major  genes  to  polygenic   (Wideman  et  al.,  2013).  One  study,  starting  from  baseline  with  a  commercial  pedigree  flock,   found  that  extremes  in  incidence  of  ascites  were  obtained  after  8  generations  of  selection   for  pulmonary  hypertension  susceptibility  (ascites  incidence  95.1  %)  and  after  9   generations  of  selection  for  pulmonary  hypertension  resistance  (ascites  incidence  7.1  %)   when  reared  at  a  simulated  altitude  of  2,900  m  (Pavlidis  et  al.,  2007).  These  results  suggest   that  a  few  major  genes  are  responsible  for  a  large  component  of  the  variation  seen  in  the   ascites  phenotype.  An  additional  finding  from  the  same  study  was  that  broilers  in  the  line   selected  for  resistance  to  pulmonary  hypertension  were  substantially  lighter  at  42  days  of   age  than  the  broilers  in  the  line  selected  for  susceptibility  to  pulmonary  hypertension   (Pavlidis  et  al.,  2007).               29   CHAPTER  3:  THE  RISK  OF,  AND  RISK  FACTORS  FOR,  CONGESTIVE  HEART  FAILURE  IN   NORTH  AMERICAN  FEEDLOTS         INTRODUCTION     Anecdotal  reports  suggest  that  the  incidence  of  right-­‐sided  congestive  heart  failure   (CHF)  in  feedlot  cattle  is  increasing.  In  cattle,  pulmonary  hypertension  is  the  primary  cause   of  right-­‐sided  CHF  (Blake,  1965;  Will  et  al.,  1975).  Historically,  CHF  was  considered  to  be   problematic  at  altitudes  over  2,130  m  (Glover  and  Newsom,  1915;  Hecht  et  al.,  1962),  but  it   has  since  been  reported  to  occur  at  lower  altitudes.  In  1974,  a  study  conducted  across  4   feedlots  located  at  an  altitude  of  1,600  m  reported  the  attack  risk  of  CHF  to  be  2.85  cases   per  10,000  cattle  entering  the  feedlot  (Jensen  et  al.,  1976).  CHF  cases  were  also  reported  to   occur  predominantly  during  the  last  half  of  the  feeding  period  (Jensen  et  al.,  1976),  which   makes  CHF  a  particularly  costly  disease.     The  purposes  of  this  study  were  to  determine:  if  the  risk  of  CHF  among  fed-­‐cattle   has  increased  over  time;  the  distribution  of  CHF  mortality  through  the  feeding  period;  and,   to  evaluate  respiratory  disease  treatment,  season  of  placement  and  gender  as  risk  factors   for  CHF.  The  first  objective  was  to  characterize  the  risk  of  CHF  across  10  Canadian  feedlots   every  4  years  from  the  year  2000  to  the  year  2012.  The  second  objective  was  to  compare   the  risk  of  CHF  in  2012  among  10  Canadian  and  among  5  U.S.  feedlots.  The  third  objective   was  to  evaluate  risk  factors  for  CHF  using  death  from  digestive  disorders  as  a  control.    It   has  been  reported  that  respiratory  disease  is  a  risk  factor  for  CHF  (Holt  and  Callan,  2007).   However,  to  what  extent  respiratory  disease  increases  the  risk  of  CHF  has  not  been     30   previously  reported.    Treatment  for  respiratory  disease  was  one  of  the  risk  factors   evaluated  in  our  study.       MATERIALS  AND  METHODS     Study  overview   Data  from  10  Canadian  feedlots  and  5  U.S.  feedlots  were  obtained  for  this   investigation.  Data  from  the  Canadian  feedlots  were  obtained  for  the  years  2000,  2004  and   2008  and  2012.  Data  from  the  U.S.  feedlots  were  obtained  for  the  year  2012.  The  number  of   animals  entering  a  given  feedlot  within  a  calendar  year  was  stratified  by  season  of   placement,  age,  gender  and  risk  of  undifferentiated  fever/respiratory  disease  (Table  3.1).     A  cause  of  death  was  determined  for  every  animal  that  died  within  the  feedlots  studied.   The  number  of  deaths  due  to  CHF  and  digestive  disorders  was  also  stratified  (Table  3.2).   From  these  data,  the  risk  of  CHF  was  determined  every  4  years  from  the  year  2000  to  the   year  2012.  Additionally,  individual  animal  records  were  obtained  from  animals  that  died  of   CHF  or  a  digestive  disorder  (ruminal  bloat,  enteritis,  peritonitis  or  intestinal  disorder).  This   information  included  the  number  of  days  from  entry  into  feedlot  to  post-­‐mortem   evaluation  and,  treatment  history  for  respiratory  disease.  Risk  factors  for  CHF  were   evaluated  using  death  from  a  digestive  disorder  as  a  control  group.  Digestive  disorders   were  considered  to  be  an  appropriate  control  because  of  the  minimal  risk  of  confounding   due  to  a  common  pathogenesis.           31   Study  population   Feedlot  Health  Management  Services  (Okotoks,  AB,  Canada)  provided  data  for  this   study.  The  study  population  consisted  of  cattle  placed  in  10  feedlots  located  in  western   Canada  during  the  years  2000,  2004  and  2008  and  2012  and  cattle  placed  in  5  feedlots  in   the  western  United  States  during  the  year  2012.  These  feedlots  were  chosen  based  on  the   availability  of  the  data  required  for  the  purpose  of  this  study.  These  data  were  obtainable   from  the  Canadian  feedlots  going  back  to  the  year  2000  and  from  the  U.S.  feedlots  going   back  to  the  year  2012.    The  procurement  and  management  of  cattle  within  the  feedlots   studied  was  typical  of  those  practices  used  at  large  commercial  cattle  feedlots  in  western   Canada  and  U.S.    Black-­‐hided  Angus-­‐based  cattle  were  the  predominant  breeds  of  cattle  in   both  Canada  and  the  U.S   The  same  standardized  health  and  production  procedures  were  used  across  all   feedlots  as  per  the  protocols  developed  by  veterinarians  at  Feedlot  Health  Management   Services.    In  brief,  at  arrival,  all  cattle  received  an  ear  tag  with  a  unique  identification   number,  a  growth  implant,  a  topical  avermectin  anthelmintic  and  vaccines  against  bacterial   and  viral  agents  of  respiratory  disease.  Cattle  determined  to  have  a  high  risk  of  developing   a  fever  of  unknown  origin  (undifferentiated  fever)  or  respiratory  disease  at  the  time  of   arrival  were  administered  a  parenteral  antibiotic  as  a  prophylactic  or  metaphylactic   treatment.  The  risk  of  developing  undifferentiated  fever/bovine  respiratory  disease  for   each  group  of  feedlot  animals  was  determined  by  feedlot  personnel  using  standardized  risk   profiles  that  were  based  on  factors  such  as  age  class  (calf  versus  yearling),  body  weight   (often  a  proxy  for  age),  procurement  method  (sale  barn  versus  private  treaty),  amount  of   commingling  before  and  after  arrival,  and  previous  vaccination  and  management  history.     32   Cattle  were  provided  with  provided  with  water  and  feed  ad  libitum.  The  feed  offered  was   typical  of  a  commercial  feedlot  operation  and  was  formulated  using  guidance  from  The   National  Research  Council  (2000).     Data  collection   The  population  at  risk  of  CHF  was  the  number  of  cattle  placed  within  the  feedlots   studied  during  a  calendar  year  and  was  stratified  according  to  the  variables  listed  in  Table   3.1.  The  cattle  placed  within  that  year  were  followed  until  completion  of  the  feeding  period.   The  buyers  and  sellers  of  the  animals  categorized  cattle  as  calves  or  yearlings  during  the   normal  processes  of  commerce.  Subsequently,  feedlot  personnel  recorded  this   classification  on  an  individual  animal-­‐based  computer  system.         Table  3.1:  The  population  at  risk,  the  number  of  cattle  placed  in  a  given  feedlot   within  a  given  year,  was  stratified  according  to  the  variables  listed       Variable   Category   Feedlot   15  feedlots  identified  by  a  unique  number   Placement  year   2000,  2004,  2008  and  2012   Season  of  placement   January  1st  to  April  30th       May  1st  to  August  31st       September  1st  to  December  31st     Age   Calf  or  yearling   Gender   Male  or  female   Risk  of  undifferentiated   fever/  respiratory  disease   High  or  low           Diagnosis  of  disease   Cattle  health  was  evaluated  daily  by  trained  feedlot  personnel.  Cattle  showing   evidence  of  disease  were  treated  under  the  supervision  of  a  veterinarian  according  to     33   protocols  developed  by  Feedlot  Health  Management  Services.  Cattle  that  died  while  in  the   feedlot  were  examined  post-­‐mortem  by  a  veterinarian  or  trained  feedlot  personnel.  The   primary  cause  of  death  was  recorded  according  to  the  criteria  in  table  3.2.  There  could  only   be  one  cause  of  death.  If  a  specific  cause  of  death  could  not  be  determined  but  peritonitis   was  evident  then  a  diagnosis  of  peritonitis  was  recorded.  Individual  animal  information   from  cattle  that  died  of  congestive  heart  failure  or  a  digestive  disorder  within  the  feedlots   during  the  years  studied  was  included  in  the  dataset.  Digestive  disorders  included:  ruminal   bloat,  enteritis,  intestinal  disorders  and  peritonitis  (Table  3.2).  Cattle  that  died  of  digestive   disorders  served  as  a  control  group  for  identifying  risk  factors  associated  with  death  from   CHF.  In  addition  to  the  information  provided  in  table  3.1  the  individual  animal  information   included:  treatment  for  respiratory  disease  (yes/no),  two  or  more  treatments  for   respiratory  disease  (yes/no)  and  the  number  of  days  from  entry  into  the  feedlot  to  post-­‐ mortem  examination.                                           34   Table  3.2:  The  post-­‐mortem  lesions  used  for  determining  cause  of  death.  One  or   more  of  the  lesions  listed  may  have  been  sufficient  to  determine  the  cause  of  death.   Cattle  did  not  have  to  demonstrate  evidence  of  all  of  the  lesions  listed.  Peritonitis   was  diagnosed  when  evident  and  a  specific  cause  of  death  was  not  determined     Category   Cause  of  death   Post-­‐mortem  lesions   Case   Congestive  heart  failure   Brisket  and  ventral  edema;   hydroperitoneum;  hydrothorax  and   secondary  atelectasis;  hepatomegaly   and  chronic  passive  congestion;   intestinal  and  mesenteric  edema;   hydropericardium;  right-­‐ventricular   hypertrophy  and  dilation     Control   (Digestive   disorder)   Ruminal  bloat   Underinflated  lungs;  anterior  carcass   congestion;  posterior  carcass  pallor;   edema  of  subcutaneous  tissue  and   facial  planes  of  hind  limbs;  rumen   distended  with  gas;  small,  pale  liver;   small,  pale  heart     Enteritis   Hyperemia  and  edema  of  intestinal   muscosa;  fibrinous  mucosa;  luminal   hemorrhage;  dark,  fluid-­‐filled   intestine;  diffuse  or  segmented;       Intestinal  disorder   Intussusception;  mesenteric  rent;   intestinal  parasitism;  lodged   trichobezoar;  stricture;  intestinal   torsion/volvulus     Peritonitis  (specific  cause   of  death  cannot  be   determined)     Hydroperitoneum;  fibrin  deposition;   adhesions;  local  or  diffuse       Statistical  analyses   Statistical  analyses  were  performed  using  STATA  version  12  (Stata  Corporation,   College  Station,  Texas,  USA).  A  descriptive  analysis  of  the  number  of  cattle  entering  the   feedlots  during  the  calendar  years  studied  was  performed.    The  number  of  cattle  that  died     35   from  CHF  and  digestive  disorders  (bloat,  enteritis,  intestinal  disorder  and  peritonitis)  was   summated  across  all  feedlots  for  all  of  the  years  studied.  The  mean  ±  standard  deviation  of   the  number  of  days  to  death  from  arrival  at  the  feedlot  was  calculated  for  CHF  and  each   digestive  disorder.     Risk  factors  for  CHF  were  evaluated  separately  for  U.S.  and  Canadian  cattle.  The   primary  risk  factor  of  interest  was  year  of  placement,  which  was  evaluated  using  the   Canadian  dataset  only.  Other  risk  factors  of  interest  were  season  of  placement,  risk   category,  gender,  age  and  feedlot  effects.  These  risk  factors  were  evaluated  separately  for   U.S.  and  Canadian  populations.  The  number  of  cattle  entering  the  feedlots  during  the  study   periods  was  the  exposure  of  interest.  The  duration  of  exposure  was  not  available  for  all   animals  therefore,  attack  risk  per  1,000  cattle  entering  the  feedlot  and  attack  risk  ratio   were  the  outcome  measures  of  interest  rather  than  incidence  risk  and  incidence  risk  ratio.   Feedlot  effects  were  controlled  in  the  analyses  as  fixed-­‐effects  so  that  separate  CHF  attack   risk  estimates  could  be  obtained  for  each  feedlot.  Likelihood  ratio  tests  were  performed  to   determine  the  statistical  significance  of  the  following  categorical  variables:  season  of   placement,  year  of  placement  and  feedlot  effects.  A  zero-­‐inflated  negative  binomial  model   was  used.  A  Poisson  model  was  not  suitable  as  a  likelihood  ratio  test  of  the  over-­‐dispersion   parameter  alpha  =  0  was  statistically  significant  (p  <  0.001).    This  indicates  that  the  mean  is   not  equal  to  the  variance,  which  is  an  assumption  of  the  Poisson  model.  An  excess  of  zero   counts,  as  indicated  by  a  Vuong  statistic  of  7.9,  meant  that  a  zero-­‐inflated  negative  binomial   model  was  superior  to  a  standard  negative  binomial  model  (p  <  0.001).       Linear  regression  was  used  to  compare  the  attack  risk  of  CHF  between  U.S.  and   Canadian  feedlots  in  the  year  2012.  The  outcome  of  interest  was  risk  of  CHF  per  1,000     36   cattle  entering  the  feedlot.  The  explanatory  variable  of  interest  was  country  when   controlling  for  season  of  placement,  risk  category,  gender,  age  and  feedlot  effects.  The  least   squares  mean  estimate  of  the  risk  of  CHF  and  95  %  confidence  interval  (95  %  CI)  were   calculated  for  all  feedlots  when  controlling  for  country,  season  of  placement,  risk  category,   gender  and  age.   The  third  objective  was  to  identify  risk  factors  associated  with  mortality  due  to  CHF   (cases)  using  mortality  due  to  digestive  disorders  as  a  baseline  reference  (control).     Digestive  disorders  were  considered  to  be  an  appropriate  control  because  of  the  minimal   risk  of  confounding  due  to  a  common  pathogenesis.  The  etiologies  of  the  various  digestive   disorders  were  likely  to  be  distinct  from  the  etiology  of  CHF.  Separate  logistic  regression   models  were  evaluated  for  Canadian  and  U.S.  feedlots.  Only  cattle  that  died  of  either  CHF  or   a  digestive  disorder  were  included  in  the  logistic  regression  model.  Year  of  placement   (Canadian  model  only),  season  of  placement,  feedlot,  gender,  risk  of  undifferentiated   fever/respiratory  and  treatment  for  bovine  respiratory  disease  were  included  in  the  model   as  explanatory  variables.  Animals  were  categorized  according  to  the  number  of  treatments   for  BRD  received:  no  treatment,  1  treatment  or  at  least  2  treatments.       RESULTS     Cattle  numbers  and  feedlot  altitudes   During  the  4  calendar  years  studied  a  total  of  1.28  million  cattle  entered  into  the   Canadian  feedlots  (Table  3.3).  In  2012,  a  total  of  273,319  cattle  entered  the  5  U.S.  feedlots   studied.  Feedlots  in  Canada  were  located  at  altitudes  ranging  from  657  m  (Feedlot  10)  to     37   1,145  m  (Feedlot  14).  Feedlots  in  the  U.S.  were  located  at  altitudes  ranging  from  596  m   (Feedlot  185)  to  1,282  m  (Feedlot  171).     Table  3.3:  The  total  number  of  cattle  entering  a  feedlot  within  a  calendar  year             Year   Total  all   years  Country   Feedlot   Altitude,   m   2000   2004   2008   2012   Canada   1   1,006   30,933   29,107   64,926   44,740   169,706   3   837   3,453   3,045   1,738   2,646   10,882   5   1,018   20,933   23,682   25,538   13,368   83,521   6   934   63,817   60,212   104,364   57,247   285,640   9   917   23,281   34,761   51,394   23,597   133,033   10   657   55,489   53,865   60,071   27,735   197,160   14   1,145   9,858   9,716   6,464   5,252   31,290   19   887   2,779   1,529   2,234   1,984   8,526   20   1,102   65,582   60,148   86,947   49,380   262,057   31   1,005   10,941   25,613   48,016   17,809   102,379   Total     287,066   301,678   451,692   243,758   1,284,194                   USA   169   1,161   -­‐   -­‐   -­‐   91,088   -­‐   170   1,242   -­‐   -­‐   -­‐   34,736   -­‐   171   1,282   -­‐   -­‐   -­‐   44,164   -­‐   174   1,142   -­‐   -­‐   -­‐   28,590   -­‐   185   596   -­‐   -­‐   -­‐   74,741   -­‐   Total           273,319   273,319     Time  from  feedlot  entry  to  death  for  CHF  and  digestive  disorders   Death  from  CHF  generally  occurred  later  in  the  feeding  period  than  death  from   digestive  disorders  (Table  3.4).  However,  deaths  from  CHF  and  digestive  disorders   occurred  throughout  the  feeding  period.  Death  from  CHF  occurred  throughout  the  feeding   period  but  tended  to  occur  approximately  5  months  after  arrival  (Table  3.4).  Death  from   bloat  occurred  throughout  the  feeding  period  but  tended  to  occur  approximately  4  months   post-­‐arrival.  Ruminal  bloat  accounted  for  72  %  and  83  %  of  deaths  due  to  digestive   disorders  in  Canadian  and  US  feedlots,  respectively.       38   Table  3.4:  The  days  on  feed  to  postmortem  for  congestive  heart  failure  (CHF)  and   digestive  disorders  and  the  total  number  of  cases  in  Canada  for  all  years  (2000,   2004,  2008  and  2012)  and  in  the  USA  in  2012         Disease   Days  in  feedlot,  mean  ±  SD   n Canada n   USA CHF   537 137  ±  79 261   137  ±  71 Bloat     2,324   129  ±  72   954   121  ±  83   Enteritis   398   59  ±  59   29   88  ±  61   Intestinal   disorder   174   91  ±  73   8   118  ±  75   Peritonitis   314   99  ±  78   148   149  ±  106     Risk  factors  for  time  from  feedlot  entry  to  death  from  CHF  (days  on  feed)  in  Canadian  feedlots   Only  age  (p  <  0.001)  and  gender  (p  =  0.03)  were  significantly  associated  with  days   from  feedlot  entry  to  death  from  CHF  when  controlling  for  season  of  placement  (p  =  0.72),   risk  category  (p  =  0.97),  feedlot  effects  (p  <  0.001)  and  year  of  placement  (p  <  0.001).  On   average,  yearlings  died  26  days  (95  %  CI:  34,  17  days)  earlier  in  the  feeding  period  than   calves  when  controlling  for  gender,  season  and  year  of  placement,  and  feedlot  effects.    On   average,  males  died  6  days  (95  %  CI:  11,  1  days)  earlier  than  females.         Risk  of  CHF  from  the  year  2000  to  the  year  2012   The  risk  of  CHF  differed  among  years,  with  a  trend  to  increase  from  2000  to  2012,  in   Canadian  feedlots  when  controlling  for  season  of  placement,  risk  category,  gender,  age  and   feedlot  effects  (Table  3.5).  The  odds  of  death  from  CHF  relative  to  digestive  disorders  also   differed  significantly  in  these  years  with  an  increasing  trend  from  the  year  2000  (Table   3.5).  The  mean  attack  risk  for  digestive  disorders  was  almost  5  times  higher  than  the  attack   risk  for  CHF  in  U.S.  feedlots  (Risk  of  digestive  disorder  =  4.66  per  1,000  cattle,  95  %  CI  =   3.61,  5.72;  Risk  of  CHF  =  1.04  per  1,000  cattle,  95  %  CI=  0.82,  1.26).     39   Table  3.5:  The  estimated  attack  risk  of  congestive  heart  failure  (CHF)  and  digestive   disorders  (DD)  per  1,000  cattle  entering  the  feedlot  by  year  of  placement  while   controlling  for  age,  risk  of  undifferentiated  fever  and  respiratory  disease,  gender   and  season  of  placement;  and,  the  odds  of  CHF  relative  to  the  year  2000  using   digestive  disorders  as  a  baseline  reference  while  controlling  for  age,  risk  of   undifferentiated  fever  and  respiratory  disease,  gender,  season  of  placement  and  the   number  of