THE EFFECT OF BLOOD UREA NITROGEN ON REPRODUCTIVE PERFORMANCE OF NITROGEN SUPPLEMENTATION

THE EFFECT OF BLOOD UREA NITROGEN ON REPRODUCTIVE PERFORMANCE OF NITROGEN SUPPLEMENTATION
THE EFFECT OF BLOOD UREA NITROGEN
ON REPRODUCTIVE PERFORMANCE OF
BEEF HEIFERS ON DIFFERENT LEVELS OF
NITROGEN SUPPLEMENTATION
T Tshuma (2013)
© University of Pretoria
THE EFFECT OF BLOOD UREA NITROGEN ON REPRODUCTIVE PERFORMANCE OF BEEF
HEIFERS ON DIFFERENT LEVELS OF NITROGEN SUPPLEMENTATION
BY
TAKULA TSHUMA
Submitted to the Department of Production Animal Studies,
Faculty of Veterinary Science, University of Pretoria,
in partial fulfilment of the requirements for the Master of Veterinary Medicine
(MMedVet) Degree
September 2013
© University of Pretoria
Acknowledgements
I would like to acknowledge and express my sincere gratitude to the following people and
institutions without whom it would have been extremely difficult, if not impossible, for me
to undertake this work:
Dr Dietmar Holm, my promoter, for the friendly and excellent tutorage during the course of
this research work. I did not only learn about research and clinical techniques from him, but
also life skills that I will be able to apply in my career as a scientist.
Professor Dirk Lourens, my co-promoter, for his wisdom and advice. It was very assuring to
learn from his vast experience. His insurmountable patience and ever-present
encouragement is what kept me going even through challenging times.
Professor Geoffrey Fosgate, my co-promoter, for his intelligent and positive way of
analysing data and life situations. I could not help but assimilate his attitude. My outlook
towards research is more positive because of him.
The following farms; Arcadia Bonsmaras, Alhansa Boerdery, Culmpine Boerdery, PGM
Boerdery and Triple B Ranch, for providing their animals, facilities and manpower during
the data collection period. Their excellent management made the work a lot easier.
Musawenkosi, my wife and Brendon, my son, for the moral support and the sacrifices they
made during the course of the study. No one could have played their role the way they did.
The Clinical Pathology Laboratory at the Onderstepoort Veterinary Academic Hospital for
their relentless enthusiasm in analysing the samples and timely reporting of the results.
Above all, I want to thank God Almighty who created all things in great wisdom, and then
gave me the resources and opportunity to pursue this interesting work.
The study was carried out under protocol number v072-810, as approved by the Animal Use
and Care Committee of the University of Pretoria.
i
© University of Pretoria
Summary
Ruminants have a unique ability to acquire protein from non-protein nitrogen (NPN)
sources, and to recycle nitrogen back into the rumen, instead of excreting all of it via the
urine, faeces and milk. However, a high concentration of blood urea nitrogen (BUN) has a
negative influence on conception. Additionally, a high dietary nitrogen intake poses a
challenge to the environment in the form of ammonia emissions, eutrophication and bad
odours. This calls for strategies to reduce the environmental impact of livestock production.
Variation exists in the ability of cattle to recirculate nitrogen between as well as within
cattle breeds. The purpose of this study was to investigate the effects of BUN concentration
on reproductive performance in beef heifers under different management systems in South
Africa. Serum samples from 369 Bonsmara heifers were taken in November and December
2010 to determine the BUN concentrations prior to the onset of the breeding season.
Heifers were from five herds with different levels of protein supplementation during the
weeks before the commencement of the breeding season. Body mass, age, body condition
score (BCS) and reproductive tract score (RTS) were recorded at the same time as BUN
concentration. Trans-rectal ultrasound and/or-palpation was performed four to eight weeks
after the three-month breeding season to detect and estimate the stage of pregnancy. Days
to pregnancy (DTP) was defined as the number of days from the start of the breeding
season until a heifer was successfully mated. Logistic regression and Cox proportional
hazards survival analysis were performed to estimate the effect of BUN concentration on
subsequent pregnancy and DTP respectively, while stratifying by herd and adjusting for
potential confounders. The correlations between BUN concentration, BCS and RTS were
estimated using Spearman’s rho. Pearson correlations were used for the normally
distributed variables of age and body mass. BUN concentration was not a significant
predictor of pregnancy status but was a significant (P = 0.007) and independent predictor of
DTP in heavily and some moderately supplemented herds. As BUN concentration increased,
DTP also increased [hazard ratio (HR) = 0.827; 95% CI: 0.721 – 0.949; P = 0.007], while the
chance of becoming pregnant decreased, although this was not statistically significant [odds
ratio (OR) = 0.882; 95% CI: 0.772 – 1.007; P = 0.063]. Bonsmara heifers with higher BUN
ii
© University of Pretoria
concentration, which suggests a better ability to recirculate nitrogen, might be at a
disadvantage when the production system includes high levels of RDP supplementation
because of this negative impact on reproductive performance. It is proposed that
production systems be adapted to avoid selection against animals with an improved ability
to recirculate nitrogen.
iii
© University of Pretoria
Table of contents
Acknowledgements ................................................................................................................ i
Summary............................................................................................................................... ii
Table of contents ................................................................................................................. iv
List of figures....................................................................................................................... viii
List of tables .......................................................................................................................... x
List of acronyms and abbreviations ..................................................................................... xiii
1
Introduction ................................................................................................................... 1
2
Literature Review ........................................................................................................... 3
2.1
Protein metabolism in the ruminant ....................................................................... 3
2.2
Intraruminal nitrogen recycling ............................................................................... 5
2.3
The biosynthesis of urea nitrogen ........................................................................... 5
2.4
Sources of variation in BUN concentrations ............................................................ 6
2.5
Relationship between MUN and BUN concentrations ............................................. 8
2.6
Effect of BUN concentration on reproductive performance .................................... 9
2.7
Fate of nitrogen in the environment ..................................................................... 11
2.8
Other factors affecting the reproductive performance of heifers .......................... 12
2.9
Measuring reproductive performance ................................................................... 14
2.10 Practical uses of BUN concentration data .............................................................. 14
iv
© University of Pretoria
3
Research Questions...................................................................................................... 16
4
Hypotheses .................................................................................................................. 17
5
Objectives .................................................................................................................... 18
6
Materials and Methods ................................................................................................ 19
7
6.1
Model system and justification of the model ........................................................ 19
6.2
Experimental design .............................................................................................. 22
6.3
Experimental procedures ...................................................................................... 22
6.4
Observations ......................................................................................................... 26
6.5
Data analysis ......................................................................................................... 26
6.6
Experimental animals ............................................................................................ 27
Results ......................................................................................................................... 29
7.1
All herds combined ............................................................................................... 29
7.1.1
Descriptive statistics ...................................................................................... 29
7.1.2
Logistic regression analysis ............................................................................ 30
7.1.3
Proportional hazards survival analysis ............................................................ 32
7.1.4
Correlation analysis........................................................................................ 33
7.2
Herd A ................................................................................................................... 34
7.2.1
Descriptive statistics ...................................................................................... 34
7.2.2
Logistic regression analysis ............................................................................ 36
v
© University of Pretoria
7.2.3
Proportional hazards survival analysis ............................................................ 37
7.2.4
Correlation analysis........................................................................................ 38
7.3
Herd B ................................................................................................................... 39
7.3.1
Descriptive statistics ...................................................................................... 39
7.3.2
Logistic regression analysis ............................................................................ 40
7.3.3
Proportional hazards survival analysis ............................................................ 41
7.3.4
Correlation analysis........................................................................................ 41
7.4
Herd C ................................................................................................................... 42
7.4.1
Descriptive statistics ...................................................................................... 42
7.4.2
Logistic regression analysis ............................................................................ 44
7.4.3
Proportional hazards survival analysis ............................................................ 44
7.4.4
Correlation analysis........................................................................................ 44
7.5
Herd D .................................................................................................................. 45
7.5.1
Descriptive statistics ...................................................................................... 45
7.5.2
Logistic regression analysis ............................................................................ 47
7.5.3
Proportional hazards survival analysis ............................................................ 48
7.5.4
Correlation analysis........................................................................................ 49
7.6
Herd E ................................................................................................................... 50
7.6.1
Descriptive statistics ...................................................................................... 50
vi
© University of Pretoria
8
7.6.2
Logistic regression analysis ............................................................................ 54
7.6.3
Proportional hazards survival analysis ............................................................ 55
7.6.4
Correlation analysis........................................................................................ 57
Discussion .................................................................................................................... 59
8.1
Introduction .......................................................................................................... 59
8.2
Herd A ................................................................................................................... 62
8.3
Herd B ................................................................................................................... 63
8.4
Herd C ................................................................................................................... 63
8.5
Herd D .................................................................................................................. 64
8.6
Herd E ................................................................................................................... 65
8.7
Effect of BUN concentration on reproductive performance .................................. 66
8.8
Potential weaknesses of the study ........................................................................ 69
9
10
Conclusion ................................................................................................................... 71
References ............................................................................................................... 72
vii
© University of Pretoria
List of figures
Figure 2.1: Fate of dietary crude CP in the ruminant animal (Adapted from McDonald et al.,
1995) ............................................................................................................................. 4
Figure 2.2: Regression of MUN on BUN concentration (Adapted from Broderick and Clayton,
1997). ............................................................................................................................ 9
Figure 2.3: Factors determining the age at puberty (Adapted from Holm et al., 2009)......... 13
Figure 7.1: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in all herds combined. ........................................................................... 34
Figure 7.2: BUN concentration in relation to sampling order in Herd A ................................ 35
Figure 7.3: Pearson’s correlation for BUN with age, and Spearman’s correlation for BCS with
age and BUN concentration, in Herd A ......................................................................... 39
Figure 7.4: BUN concentration in relation to sampling order in Herd B ................................ 40
Figure 7.5: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd B ............................................................................................... 42
Figure 7.6: BUN concentration in relation to sampling order in Herd C ................................ 43
Figure 7.7: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd C ............................................................................................... 45
Figure 7.8: BUN concentration in relation to sampling order in Herd D ................................ 46
viii
© University of Pretoria
Figure 7.9: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd D .............................................................................................. 50
Figure 7.10: BUN concentration in relation to sampling order in Herd E on day 1 of sampling
.................................................................................................................................... 52
Figure 7.11: BUN concentration in relation to sampling order in Herd E on day 2 of sampling
.................................................................................................................................... 52
Figure 7.12: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd E day 1...................................................................................... 57
Figure 7.13: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd E day 2...................................................................................... 58
ix
© University of Pretoria
List of tables
Table 6.1: BCS system (Wildman et al., 1982). ..................................................................... 23
Table 6.2: RTS system (Adapted from Andersen et al., 1991; Holm et al., 2009). ................. 24
Table 6.3: Determination of stage of pregnancy by rectal palpation (adapted from Sheldon
and Noakes, 2002; Youngquist, 2007). ......................................................................... 25
Table 7.1: Descriptive statistics for all herds (n = 369) ......................................................... 29
Table 7.2: Descriptive statistics by pregnancy status for all herds, indicating pregnant (n =
233), non-pregnant (n = 102) and lost to follow up (n = 31) heifers............................... 30
Table 7.3: Effects of BUN concentration, age, body mass and RTS on pregnancy proportion in
all herds combined (single variable logistic regression)................................................. 31
Table 7.4: The combined effects of BUN concentration and age on pregnancy proportion in
all herds (multivariable logistic regression) .................................................................. 32
Table 7.5: Effects of BUN concentration, age, body mass and RTS on DTP in all herds
combined (single variable survival analysis) ................................................................. 32
Table 7.6: The combined effects of BUN concentration and age on DTP in all herds
(multivariable survival analysis) ................................................................................... 33
Table 7.7: Descriptive statistics for Herd A (n = 115) ............................................................ 35
Table 7.8: Descriptive statistics by pregnancy status for Herd A, indicating pregnant (n = 82),
non-pregnant (n = 24) and lost to follow up (n = 9) heifers ........................................... 36
Table 7.9: Effects of BUN concentration, age, body mass and RTS on pregnancy proportion in
Herd A (single variable logistic regression) ................................................................... 37
x
© University of Pretoria
Table 7.10: Effects of BUN concentration, age, body mass and RTS on DTP in Herd A (single
variable survival analysis)............................................................................................. 38
Table 7.11: Descriptive statistics for Herd B (n = 33) ............................................................ 40
Table 7.12: Effects of BUN concentration, age, body mass and RTS on DTP in Herd B (single
variable survival analysis)............................................................................................. 41
Table 7.13: Descriptive statistics for Herd C (n=34) .............................................................. 43
Table 7.14: Effects of BUN concentration, age, body mass and RTS on DTP in Herd C (single
variable survival analysis)............................................................................................. 44
Table 7.15: Descriptive statistics for Herd D (n = 29)............................................................ 46
Table 7.16: Descriptive statistics by pregnancy status for Herd D, indicating pregnant (n =
15), non-pregnant (n = 7) and lost to follow up (n = 7) heifers ...................................... 47
Table 7.17: Effects of BUN concentration, age, body mass and RTS on pregnancy proportion
in Herd D (single variable logistic regression) ............................................................... 48
Table 7.18: Effects of BUN concentration, age, body mass and RTS on DTP in Herd D (single
variable survival analysis)............................................................................................. 49
Table 7.19: The combined effects of BUN concentration and RTS on DTP in Herd D
(multivariable survival analysis) ................................................................................... 49
Table 7.20: Descriptive statistics for Herd E on day 1 of sampling (n = 88) ........................... 51
Table 7.21: Descriptive statistics for Herd E on day 2 of sampling (n =70) ............................ 51
Table 7.22: Descriptive statistics by pregnancy status for Herd E day 1, indicating pregnant
(n = 43), non-pregnant (n = 37) and lost to follow up (n = 8) heifers .............................. 53
xi
© University of Pretoria
Table 7.23: Descriptive statistics by pregnancy status for Herd E day 2, indicating pregnant
(n = 26), non-pregnant (n = 37) and lost to follow up (n = 7) heifers .............................. 54
Table 7.24: Effects of BUN concentration, age, body mass and RTS on pregnancy proportion
in Herd E (single variable survival analysis) ................................................................... 55
Table 7.25: The combined effects of BUN concentration and body mass on pregnancy
proportion in Herd E (multivariable logistic regression) ................................................ 55
Table 7.26: Effects of BUN concentration, age, body mass and RTS on DTP in Herd E (single
variable survival analysis)............................................................................................. 56
Table 7.27: The combined effects of BUN concentration and body mass on DTP in Herd E
(multivariable survival analysis) ................................................................................... 57
Table 8.1: Summary of the effect of BUN concentration on pregnancy proportion and DTP in
the different herds ....................................................................................................... 68
xii
© University of Pretoria
List of acronyms and abbreviations
BCS
Body condition score
BUN
Blood urea nitrogen
CI
Confidence interval
CIDR
Controlled Internal Drug Release
CP
Crude protein
CRL
Crown rump length
DMI
Dry matter intake
DTC
Days to calving
DTP
Days to pregnancy
HR
Hazard ratio
LH
Luteinizing hormone
MUN
Milk urea nitrogen
NEB
Negative energy balance
NEFA
Non-esterified fatty acids
NPN
Non protein nitrogen
NSC
Non-structural carbohydrates
OR
Odds ratio
PE
Pregnancy examination
xiii
© University of Pretoria
PER
Protein to energy ratio
PR
Pregnancy rate
RDP
Rumen degradable protein
RTS
Reproductive tract score
RUP
Rumen undegradable protein
SAWS
South African weather services
SC
Structural carbohydrates
SD
Standard deviation
xiv
© University of Pretoria
1 Introduction
Ruminants are unique in that they are capable of recycling nitrogen back into the rumen,
instead of excreting all of it in the urine (Erickson and Klopfenstein, 2010), faeces or milk
(Dijkstra et al., 2011), thus supplying rumen microbes with their need for ammonia (Marini
and van Amburgh, 2003). In light of the increasing public health and global warming
concerns that have been focused on animal production systems as a source of
environmental pollution (Marini and van Amburgh, 2005), there is a need for more research
aimed at reducing nitrogen excretion into the environment. It is estimated that the
proportion of dietary nitrogen that is retained in feedlot cattle is less than 20% (Bierman et
al., 1999), implying that more than 80% of it is excreted. Most of this nitrogen (up to 97%) is
excreted in the form of urea in urine and organic nitrogen in faeces (Varel et al., 1999;
McCrory and Hobbs, 2001; Dijkstra et al., 2011). It is therefore logical to assume that even in
less intensive beef production systems where the levels of dietary nitrogen supplementation
are relatively lower, dietary nitrogen is still excreted into the environment.
Blood urea, which is synthesized in the liver in cattle, can be found in varying concentrations
without causing any adverse effects to the animal. However if present at very high levels, it
may be associated with reproductive problems (Larson et al., 1997; Kauffman and St-Pierre,
2001).
BUN concentration can be used to indicate the nitrogen recycling efficiency of cattle. The
efficiency of ruminants in the utilisation of dietary nitrogen depends on the availability of
dietary energy for the conversion of ammonia to microbial protein. In the presence of
adequate amounts of energy, less ammonia is converted to urea for excretion
(Ipharraguerre et al., 2005).
BUN concentration is known to vary with the dietary protein levels, hydration status of the
animal, breed and time of blood sample collection (Godden et al., 2001). The dietary
nitrogen content is the main determinant of BUN concentration and nitrogen excretion in
cattle (Roseler et al., 1993; Kauffman and St-Pierre, 2001). Casper et al. (1994) suggested
1
© University of Pretoria
that a balance in the protein to energy ratio (PER) is very critical in growing heifers because
they have a limited dry matter intake (DMI) and fermentation capacity. Several studies have
reported genetic variation in milk urea nitrogen (MUN) concentration between cows,
suggesting that genetic differences in nitrogen recirculation efficiency do exist. The reported
heritability estimates of MUN ranged between 0.14 and 0.44 (Mitchell et al., 2005; Stoop et
al., 2007; Bouwman et al., 2010; Hossein-Zadeh and Ardalan, 2011). Selection of animals
with the ability to optimally recirculate nitrogen could be useful to reduce environmental
pollution from livestock production by reducing the need for dietary nitrogen
supplementation.
2
© University of Pretoria
2 Literature Review
2.1 Protein metabolism in the ruminant
Ruminants derive most of their energy and protein from microorganisms, which live
symbiotically in the rumen. Hoover and Strokes (1991) identified carbohydrates and
proteins as the major nutrients that are required to support microbial growth. It is logical to
deduce that for ruminant diets to meet the requirements of the animal, they should first
meet the microbial needs for growth and multiplication.
Ruminants use carbohydrates and fats for energy. During Negative energy balance (NEB),
they will also utilise protein. Complex carbohydrates in the diet undergo microbial
fermentation and enzyme breakdown in the rumen. The microbial fermentation process
yields volatile fatty acids (VFA) which provide a large portion of the energy requirement in
the ruminant (Demeyer, 1981; Fondevila and Dehority, 1994). In a study done by Leedle et
al. (1986), it was shown that easily solubilized carbohydrates like sugars, starches, and
pectins undergo the most rapid fermentation, while that of the less soluble polysaccharides
(hemicellulose and cellulose) was slower.
The ruminant acquires its protein when the undegraded true protein (amino acids and
peptides) fraction and the microbial protein, passes from the rumen to the abomasum and
then to the small intestines, where it is digested and absorbed. The nitrogen for the process
of microbial growth is obtained from protein nitrogen and non-protein nitrogen (NPN). The
rumen degradable protein (RDP) fraction consists of NPN, soluble intake protein (SIP) and
some more slowly degraded proteins. A proportion of the dietary true protein passes from
the rumen into the abomasum and small intestine and this fraction is described as the
rumen undegradable protein (RUP) (Schwab et al., 2003).
Bacteria acting on the structural carbohydrate (SC) fraction (cellulose and hemicellulose) of
the diet require only ammonia for growth. Whereas bacteria acting on the non-structural
carbohydrate (NSC) fraction (sugars, starches and pectins) derive about 65% of their
3
© University of Pretoria
nitrogen from amino acids and peptides and the remainder from ammonia (Russell et al.,
1992; McDonald et al., 1995).
Urea is quantitatively the most important end-product of nitrogen metabolism in ruminants,
with at least 70% of dietary nitrogen passing through the urea pool of goats daily (Harmeyer
and Martens, 1980). Urea is not only a waste product of nitrogen metabolism in ruminants,
but it also serves the important functions of buffering the blood pH and providing an
important precursor of protein biosynthesis (Harmeyer and Martens, 1980). The
detoxification of ammonia into urea occurs in the liver and this is an energy dependant
process, which may aggravate an existing energy shortage. A schematic summary of the fate
of dietary crude protein (CP) in ruminants follows (Figure 2.1):
Crude Protein
Saliva
RUMEN
Rumen Degradable Protein (RDP)
Urea
Ammonia
Peptides
Rumen Undegradable Protein (RUP)
Amino Acids
Microbial Protein synthesis
Digestible Microbial Protein
Urine
Digestible Undegraded Protein
Undigestible Protein
Faeces
Amino acids
ABOMASUM AND SMALL INTESTINE
Tissue Protein
Figure 2.1: Fate of dietary crude CP in the ruminant animal (Adapted from McDonald et al.,
1995)
The microbial degradation of the RDP fraction usually releases ammonia at a faster rate
than its uptake by microorganisms. Excess ammonia gets absorbed through the rumen wall
4
© University of Pretoria
into the portal vein and is transported to the liver where it is converted into urea (Roseler et
al., 1993; Tamminga, 2006). Urea is a water-soluble molecule, which readily enters the
blood circulation, and is distributed to all body fluids. Some of the urea recirculates back
into the rumen via the saliva, or diffusion across the rumen wall. While most of the urea is
excreted through urine and milk, the kidneys may also recirculate a fraction of the urea back
into the blood. When high levels of ammonia exist in the rumen, the ruminal pH is elevated
and this increases the rate of absorption through the rumen wall. This causes a rise in BUN
concentration (Roseler et al., 1990; Elrod and Buetler, 1993).
Microbial protein is generally produced proportional to the amount of carbohydrate
fermented in the rumen (Ørskov, 1994). Ruminal microbes utilise fermentable
carbohydrates when metabolising dietary nutrients into microbial protein. It is therefore
essential to ensure that the ruminant receives balanced proportions of fermentable
carbohydrate and RDP; otherwise, most of the dietary RDP will be degraded into ammonia
in the rumen (Chalupa and Sniffen, 1996; Schroeder and Titgemeyer, 2008).
2.2 Intraruminal nitrogen recycling
Recycling of nitrogen also occurs in the rumen. The outflow of nitrogen from the rumen is
reduced by proteolytic bacteria and protozoa, which digest other rumen bacteria. Changing
the microbial population of the rumen through antibiotics or some plant products including
saponins and essential oils can have substantial effects on the anabolic nitrogen flow and
hence the BUN concentration (Lapierre and Lobley, 2001).
2.3 The biosynthesis of urea nitrogen
Urea or carbamide is an organic chemical compound with the formula CO(NH2)2. It is mainly
formed from the detoxification of ammonia in the liver, after which it equilibrates into the
bloodstream and other body fluids (Harmeyer and Martens, 1980). The quantity of
ammonia that is available for detoxification is a direct reflection of both dietary RDP and the
availability of fermentable carbohydrates that support microbial growth and protein
synthesis (Butler, 1998).
5
© University of Pretoria
Other sources of urea in the body include the deamination of amino acids that occurs in the
liver. Circulating amino acids that are not assimilated by the body are deaminated to
produce urea and energy substrates (Butler, 1998). It has also been demonstrated in an in
vitro study that ruminant gut tissues (ruminal epithelial and duodenal mucosal cells) have
the capacity to produce urea in vitro (Oba et al., 2004). This production by gut tissues is
thought to occur in vivo as well, thus contributing smaller amounts of urea to the circulating
pool. Arginine catabolism in the mammary gland can also produce small amounts of urea
that make up part of the MUN (Nousiainen et al., 2004).
The urea from the various sources circulates in the blood and equilibrates with other body
tissues like milk and urine (Gustafsson and Palmquist, 1993). The urea is easily measured in
plasma or serum by the nitrogen content (i.e., the urea nitrogen concentration) (Butler,
1998).
2.4 Sources of variation in BUN concentrations
Many studies in dairy cattle have shown that BUN concentration is directly related to the
amount of CP in the diet, the proportions of RDP and RUP as well as the PER of the diet,
especially the fermentable carbohydrates (Roseler et al., 1990; Roseler et al., 1993; Chalupa
and Sniffen, 1996; Schroeder and Titgemeyer, 2008). The accepted target range for BUN
concentrations in dairy cattle is between 8 and 18 mg/dL (Jonker et al., 1999), as obtained
from directly converting the MUN range of between 10 and 16 mg/dL to BUN
concentrations.
Previous studies have also demonstrated that several other factors in addition to feed
intake and dietary composition are involved in the determination of BUN concentrations in
cattle. These include factors like the time of sample collection, the mass of the animal, the
method used to measure BUN concentration, parity, breed, the hydration status, the DMI
and the method of analysis (Godden et al., 2001; Kauffman and St-Pierre, 2001; RajalaSchultz and Saville, 2003; Hossein-Zadeh and Ardalan, 2011). Mitchell et al. (2005) clearly
demonstrated that variation in BUN concentration is also genetically determined with a
6
© University of Pretoria
moderate heritability. Most studies on nitrogen metabolism focus on urea because it is
stable and easily measured (Butler et al., 1996).
BUN concentration is known to vary throughout the day in relation to the time of feeding in
dairy cattle. Gustafsson and Palmquist (1993) determined that ruminal ammonia peaked
one hour after feeding and returned to baseline levels after six hours. BUN concentrations
peaked at 2.5 to 3 hours after feeding. This pattern is important to consider when collecting
blood samples from a large group of animals as it may introduce an unintended variation
among animals due to sampling order. It is yet to be determined how this variation interacts
with other factors such as breed, dietary composition and the feeding schedule in
determining the measured BUN concentration (Rodriguez et al., 1997).
The level of nitrogen in the diet also affects the efficiency of BUN recirculation in ruminants.
Those animals that are fed low nitrogen diets tend to be more efficient at recycling nitrogen
when compared to those that are fed high levels of nitrogen (Marini and van Amburgh,
2003; Marini et al., 2004). The mechanism that regulates the recycling of BUN back into the
gastrointestinal tract (GIT) of ruminants remains unknown (Røjen et al., 2011). Some studies
have shown that BUN concentration is related to the rate at which BUN is transferred back
into the GIT of ruminants (Sunny et al., 2007; Kristensen et al., 2010). Kristensen et al.
(2010) showed that the transport of urea nitrogen across gut epithelia is regulated by mass
action and adaptive changes in their permeability in ruminants. Other studies have
suggested that the transport system tends to adapt to dietary induced changes causing
changes in the permeability of the gut to BUN and hence the rate of influx of BUN into the
GIT (Calsamiglia et al., 2010). A factor named urea transporter B is expressed in the
epithelial cells of the rumen (Stewart et al., 2005) confirming the involvement of
transporters in the regulation of BUN recycling. However, in a recent study, Røjen et al.
(2011) could not find a correlation between the expression of urea transporter B factor and
changes in the arterial supply of nitrogen.
It is generally accepted that BUN concentration will be lower in heifers than in adult cows
(Oltner et al., 1985; Canfield et al., 1990; Arunvipas et al., 2003), although other studies
found no effect of age on BUN concentration (Eicher et al., 1999). Others even suggested
7
© University of Pretoria
that BUN concentration is higher in younger cows and decreases with age (Doska et al.,
2012). The proposed explanation for the increase in BUN concentration is that growing
animals utilise amino acids more efficiently. This is thought to cause reduced deamination
and urea formation in the liver, leading to lower BUN concentrations in younger animals
(Oltner et al., 1985).
The influence of gender on BUN concentration in cattle has not received adequate attention
in literature, but a study involving camels showed a significantly higher BUN concentration
in females than in males (Patodkar et al., 2010).
Barton et al. (1996) demonstrated that breed significantly affects measured BUN
concentrations in Holstein and Jersey cattle, but others reported that it has no effect
(Miettinen and Juvonen, 1990).
2.5 Relationship between MUN and BUN concentrations
High dietary protein supplementation that is aimed at increasing production, leads to
elevated concentrations of urea and ammonia, which impairs fertility in dairy cattle (Elrod
and Butler, 1993). MUN concentration can be used to estimate BUN concentration
(Ferguson et al., 1993) because of the strong linear correlation between the two (Roseler et
al., 1993; Harris, 1996) in dairy cattle. In one study involving dairy cows, a strong correlation
was observed between BUN and MUN concentrations (r² = 0.73; P < 0.001) (Gonda and
Lindberg, 1994). This correlation is thought to be caused by rapid diffusion of urea from the
blood compartment into the milk through the epithelium of the mammary gland after a bit
of a time lag (Gustafsson and Palmquist, 1993). Broderick and Clayton et al. (1997) proposed
the following equation indicating the relationship between MUN and BUN concentration:
= 0.620 + 4.75 ( ² = 0.842),
where
= MUN concentration and
= BUN concentration.
The regression of MUN on BUN concentration they obtained in a study involving 2231 dairy
cows (Figure 2.2):
8
© University of Pretoria
Figure 2.2: Regression of MUN on BUN concentration (Adapted from Broderick and Clayton,
1997).
2.6 Effect of BUN concentration on reproductive performance
The association between excess protein and fertility is controversial (Ward, 2000). Many
studies have reported that high protein intakes or high BUN or MUN concentrations have
negative effects on reproductive performance of dairy cattle (Butler et al., 1996; Larson et
al., 1997; Rajala-Schultz et al., 2001; Arunvipas et al., 2007). Ward (2000) suggested that
some of these negative effects might have been confused with those of a concurrent energy
deficit. Several other studies found no association between protein intake or plasma urea
levels at the time of service and reproductive performance in cattle (Whitaker et al., 1993;
Kenny et al., 2002a; Kenny et al., 2002b).
Other studies have reported that reduced reproductive performance only occurs when
MUN was either too low (<7 mg/dL) or too high (>17.6 mg/dL) (Pehrson et al., 1992;
Carlsson and Pehrson, 1993). Butler et al. (1996) demonstrated that MUN concentrations in
excess of 19 mg/dL have a negative effect on conception rates while Guo et al. (2004)
reported that MUN concentrations had minimal effects on the rate of conception.
9
© University of Pretoria
Many of the studies on the effects of protein and nitrogen supplementation on the
reproductive performance were performed in dairy cattle (Elrod and Butler, 1993; GarciaBojalil et al., 1994; Barton et al., 1996; Smith et al., 2001; Kenny et al., 2002a; Babashahi et
al., 2004; Rhoads et al., 2006) rather than beef heifers. Literature review failed to identify
such studies within South Africa. Very few studies, to the knowledge of the researcher, have
been published seeking to measure the association between genetic determinants of BUN
or MUN concentrations and its reproductive performance. These studies identified that
MUN concentrations are genetically determined with moderate heritability but the genetic
correlations were too weak to justify inclusion of MUN concentration as an indicator trait
for reproductive performance in a breeding program (Wood et al., 2003; Mitchell et al.,
2005; König et al., 2008).
Although Schoeman (1989) reported on the existence of significant breed differences (P <
0.01) in BUN concentrations of Hereford, Bonsmara and Nguni breeds, Ndlovu et al. (2009)
found no significant differences (P > 0.05) in a study involving the Nguni, Bonsmara and
Angus breeds.
Several mechanisms by which high dietary CP may reduce cow fertility have been proposed.
Degradation of protein in the rumen or its metabolism in the body for energy releases
ammonia and urea (Tamminga, 2006). Ammonia is believed to play a negative role prior to
ovulation, whereas urea mainly exerts its effects during cleavage and blastocyst formation
of the embryo after fertilisation (Jorritsma et al., 2003). Elrod and Butler (1993) reported
that high concentrations of BUN lowered the uterine pH. It is not known how urea causes
this drop in uterine pH, but Zhu et al. (2000) suggested that ureagenesis lowers the pH by
removing bicarbonate from the blood.
Other authors reported a direct effect of ammonia and BUN on reproductive performance.
They suggested that urea acts directly on the oocyte and through altering the composition
and pH of follicular, oviductal and uterine environments (Jordan and Swanson, 1979; Jordan
et al., 1983; Ocon and Hansen, 2003). Sinclair et al. (2000) demonstrated a detrimental
effect of ammonia on cleavage rates and blastocyst formation. However, another recent
study demonstrated that the embryo survival rate is not affected by dietary urea
10
© University of Pretoria
supplementation (Kenny et al., 2002a). The idea of disruptions to the oviductal environment
as a cause of impaired fertility has been disputed by Kenny et al. (2002b).
Butler (1998) suggested that high levels of dietary RDP exert its effects through exacerbating
the NEB and its negative effect on reproductive performance. The exacerbating effect is
caused by the additional energy cost of detoxifying ammonia from the rumen and protein
catabolism (Garcia-Bojalil et al., 1998). These authors also showed a negative effect of RDP
on plasma progesterone, which could be rectified by dietary supplementation of fat (GarciaBojalil et al., 1998).
In an extensive review Leroy et al. (2008) suggested that NEB acts through a complex
pathway involving the endocrine system causing a disruption in luteinizing hormone (LH)
pulse frequency and amplitude, and this is responsible for compromising embryo survival.
The same review reported that NEB also affects fertility by increasing the concentration of
non-esterified fatty acids (NEFA), which have direct toxic effects on the developing oocyte.
In the presence of NEB, protein catabolism causes elevated BUN concentrations. Leroy et al.
(2008) further clarified that the detrimental effects of high blood urea and ammonia
concentrations are at the level of both the embryo (especially through ammonia) and the
oocyte (particularly through urea).
From current knowledge the interactions and potential confounding between the effects of
NEB and increased BUN on reproductive performance has not been completely clarified.
2.7 Fate of nitrogen in the environment
Sixty-nine per cent of the nitrogen in urine of cows is in the form of urea (Bristow et al.,
1992), and upon excretion, the urea is rapidly converted to ammonia by urease enzymes
(Powell and Russelle, 2009). These enzymes are produced by bacteria that are present in
faeces and the soil (Béline et al., 1998). In contrast, the degradation of organic nitrogen in
faeces occurs more slowly and may require months or years to complete (Ndegwa et al.,
2008).
11
© University of Pretoria
Between 20 to 55% of nitrogen in the manure-urine mixture volatilises into the atmosphere,
leading to air pollution (Varel et al., 1999; Powell and Russelle, 2009). Some of the nitrogen
in the manure enters rivers and other surface water bodies and subsequently causes
eutrophication. Over the past 15 years, strategies of reducing this environmental impact
were aimed at manure management to mitigate runoff (Powell et al., 2008). Other
strategies include reduction of protein in cattle diets (Wu and Satter, 2000), segregation of
urine from faeces to reduce contact between urease and urine, use of urease inhibitors,
lowering manure pH, use of chemical additives that bind to ammonia and biological agents
that convert ammonium into non-volatile nitrogen such as nitrite, nitrate or gaseous
nitrogen (Ndegwa et al., 2008).
2.8 Other factors affecting the reproductive performance of heifers
It is important to note that other factors such as age at puberty, body condition score (BCS),
bull factors, farm management and environment have an effect on the reproductive
performance of heifers. Heifers require a high plane of nutrition to attain puberty at an early
age. Age at puberty can be defined as the age at which a heifer shows the first visual signs of
oestrus (Pineda and Dooley, 2003).
Age at puberty is determined in individual heifers by a genetically determined body mass
that has to be reached before the heifer will attain puberty. The age at which this mass will
be reached is determined by the growth rate. Growth rate is partially determined by
genetics but is mostly influenced by environmental factors, in particular nutrition (Figure
2.3) (Short and Bellows, 1971; Hall et al., 1997). Reproductive tract scoring (RTS) by
transrectal palpation of the uterus and ovaries provides an indirect measure of age at
puberty, and has been shown to have a good correlation with pregnancy proportion and
days to calving (DTC) when applied before the first breeding season in heifers (Andersen et
al., 1991; Holm et al., 2009).
12
© University of Pretoria
Environmental
effects
Genetic
effects
Nutrition
Climate
Critical weight at
which puberty
will be reached
GROWTH
RATE
Season of birth
Biostimulation
Age at which the
critical weight
will be reached
Age at Puberty
Figure 2.3: Factors determining the age at puberty (Adapted from Holm et al., 2009)
A balance in the PER of heifer rations is important because it determines how well the
rumen microbes can synthesize microbial protein from dietary protein, thus affecting BUN
concentrations (Ørskov, 1994). Fermentable carbohydrates and roughages play an
important role in ensuring a healthy rumen environment (Chalupa and Sniffen, 1996;
Schroeder and Titgemeyer, 2008).
It has been shown in previous studies that BCS, body mass and age at puberty of the heifers
at the start of the breeding season are associated with reproductive performance (Buckley
et al., 2003; Berry et al., 2003).
Reproductive performance in a production system that utilises natural service is a
combination of heifer and bull fertility. The negative effects of an infertile bull can be
exacerbated in single-sire herds where herd fertility is compromised by the bull’s fertility. It
is recommended that a bull breeding soundness examination be performed prior to the
breeding season to reduce problem of infertile bulls (Irons et al., 2007; Alexander, 2008).
13
© University of Pretoria
2.9 Measuring reproductive performance
Several methods for measuring and recording reproductive performance of beef heifers are
available. These include, but are not limited to DTC, age at first calving, first insemination
conception and pregnancy proportion (Cammack et al., 2009). As written by MacGregor and
Casey (1999), the official National Beef Performance and Progeny Testing Scheme in South
Africa uses age at first calving as the criteria for evaluating reproductive performance in
beef herds. However, most of these methods may not be applicable to heifers that are bred
in a restricted breeding season. Pregnancy proportion and DTC are considered the most
useful criteria for beef heifers (MacGregor and Casey, 1999; Eler et al., 2002).
The DTC is estimated as the length of time from the onset of the breeding season to calving,
and it is similar to measurements of the calving date (Meyer et al., 1990). It is easy to
measure (Buddenberg et al., 1990), and is considered a strong and practical measure of
reproductive performance in beef heifers because it is of great economic importance.
Longer DTC will cause lighter weaning weights (Bourdon and Brinks, 1983), and the heifers
that calve late in the season do not have adequate time to recuperate prior to the onset of
the next breeding season (MacGregor and Casey, 1999).
The pregnancy proportion in heifers is an indirect measure of sexual maturity at the onset of
the breeding season. It is a binary measure (1 = pregnant; 0 = non-pregnant) defined as the
probability of a heifer that was exposed to the bull at the onset of the breeding season
becoming pregnant by the end of the breeding season and remaining pregnant to the time
of examination for pregnancy (Evans et al., 1999; Eler et al., 2002).
2.10 Practical uses of BUN concentration data
BUN concentration data, when available can be useful for monitoring dietary CP and energy
intake relative to the heifer’s requirements (Rajala-Schultz and Saville, 2003). Monitoring
BUN concentration in a beef herd can serve as an important management tool because
excess dietary nitrogen increases the energy requirements of the animal and the producer
has to spend money on feed to sustain the excess nitrogen in the diet. Besides, protein
14
© University of Pretoria
supplementation is relatively expensive and BUN concentration data will help optimise
protein supplementation. In addition to the negative effects of excess nitrogen on
reproductive performance, excessive nitrogen excretion into the environment should be
avoided (Broderick and Clayton, 1997; Frank and Swensson, 2002).
In Nguni cattle, it is believed that animals with high BUN concentrations are more capable of
maintaining body condition (Schoeman, 1989), hence higher growth rates and should
therefore have a lower age at puberty and better reproductive performance. BUN
concentration data can be used as a management strategy to select for those animals that
are better adapted for efficient utilisation of nitrogen resources.
15
© University of Pretoria
3 Research Questions
1. Is BUN concentration associated with the reproductive performance of beef heifers,
and if so, how is this influenced by nitrogen supplementation?
2. What is the correlation between BUN concentration, body mass, age, BCS and RTS in
beef heifers?
16
© University of Pretoria
4 Hypotheses
1. BUN concentration or its interaction with nitrogen supplementation has an effect on
the reproductive performance of Bonsmara heifers.
2. A correlation exists between BUN concentration, body mass, age, BCS and RTS in
Bonsmara heifers.
17
© University of Pretoria
5 Objectives
1. To estimate the association between BUN concentration and reproductive
performance of Bonsmara heifers at different levels of nitrogen supplementation.
2. To estimate the correlation between BUN concentration, body mass, age, BCS and
RTS in Bonsmara heifers.
18
© University of Pretoria
6 Materials and Methods
6.1 Model system and justification of the model
Three hundred and sixty-nine nulliparous Bonsmara heifers from five herds were used. All
the herds were located between the latitudes 23˚ 21̕’ 53” and 27˚29’ 9” south. Herds were
identified by convenience sampling and they all practised a restricted breeding for a period
of 3 months, starting on 1 December. Herds were subsequently classified according to the
level of dietary nitrogen supplementation that they practised during the month prior to
sampling into none, low, moderate and high. Information on the weather elements were
obtained from the South African Weather Services (SAWS). The SAWS stations nearest (less
than 80 km) to the farms were used.
The first herd (Herd A) was a commercial Bonsmara herd in the sourish mixed bushveld of
the Limpopo province. This herd was managed on natural rangeland (veldt) defined as a low
input nutritional system. A commercial protein lick supplement with 45% CP was provided
throughout the dormant season (winter) until approximately two months before the
beginning of the breeding season when they were changed to a mineral lick, which lasted
throughout the breeding season. Due to that, the herd was defined as having no
supplementation of nitrogen for two months prior to the breeding season. In this herd, 106
heifers aged 22 to 26 months were bred by natural service in multisire groups of four bulls
per 100 heifers. A breeding soundness examination was performed on all the bulls two
months prior to the onset of the breeding season. The exact ages of the individual animals
in this herd was not known. On the day of sampling, the heifers had access to drinking water
whilst in the holding pens but feed was not available. The weather was cool with a minimum
and maximum temperature of 17.8 ˚C and 31.4 ˚C respectively (SAWS). It rained a total of
2.8 mm during sampling and the humidity for the day was 55% (SAWS).
Herd B was a stud Bonsmara herd in the sweet mixed bushveld of the Limpopo province.
The heifers in this herd were managed on veldt in a low input nutritional system. A
commercial energy lick supplement (23% CP) was provided throughout the dormant season
(winter and spring) until the beginning of the breeding season, when they were changed
19
© University of Pretoria
onto a higher energy lick supplement (18% CP). Due to this, the herd was defined as having
a low level of nitrogen supplementation prior to the breeding season. Thirty-three heifers
aged 20 to 26 months were bred by natural service in a single sire group. The bull was
examined for breeding soundness two months prior to the onset of the breeding season. On
the day of sampling, these heifers had access to drinking water whilst they were in the
holding pens. Sampling was done on a hot day with a humidity of 69%, minimum and
maximum temperature of 18 ˚C and 35.5 ˚C respectively (SAWS).
Herd C was a Bonsmara herd in the sourish mixed bushveld of the Limpopo province. This
herd was managed on irrigated oats pastures during the winter, defined as a medium input
nutritional system. A commercial energy lick supplement (18% CP) was provided throughout
the dormant season (winter and spring) until the beginning of the breeding season when
they were moved to normal veldt with a mineral lick supplement, which lasted throughout
the breeding season. Due to the high RDP content of the irrigated pasture, this herd was
defined as having a moderate level of nitrogen supplementation. Thirty-four heifers aged 23
to 27 months were artificially inseminated after synchronisation with progesterone
impregnated intravaginal devices (CIDR Easy Breed, Pfizer Animal Health), followed by
natural mating with one bull. The bull was examined for breeding soundness two months
prior to the onset of the breeding season. On the day of sampling, the heifers had no access
to feed or water while in the holding pens. Sampling was done on a hot day with a humidity
of 75%, minimum and maximum temperatures were 17.5 ˚C and 32.8 ˚C respectively
(SAWS).
Herd D was a stud Bonsmara herd in the sourish mixed bushveld of Limpopo province.
Heifers were managed on veldt in a low input nutritional system. A commercial protein lick
supplement (45% CP) was provided during the dormant season until one month after the
first significant spring rains have occurred, when they were changed onto a mineral lick for
the duration of the rainy season (summer). Due to the high CP content in the protein lick
supplement, this herd was defined as having a moderate level of nitrogen supplementation
prior to the onset of the breeding season. These heifers were kept in two separate camps
with differing grazing quality in a low input system. Twenty-two heifers aged 15 to 26
20
© University of Pretoria
months were bred by natural service using a single sire group. The bull was examined for
breeding soundness two months prior to the onset of the breeding season. On the day of
sampling, the heifers had access to water, but not feed while in the holding pens. Sampling
was done on a cool day with a humidity of 92%, minimum and maximum temperatures were
13.7 ˚C and 22.7 ˚C respectively (SAWS).
Herd E was a stud Bonsmara herd in the sour veldt of the Free State province. Heifers in this
herd were managed on irrigated rye grass pastures, defined as a high input nutritional
system. A commercial energy lick supplement (18% CP) was supplied for five months prior
to the onset of the breeding season. On the first day of the breeding season, they were
moved to natural pasture (over sown with Themeda triandra), and received a mineral lick
for the duration of the breeding season. Due to the high RDP content of pasture and the
energy lick supplement combined, this herd was defined as having a high level of nitrogen
supplementation. One hundred and forty-three heifers aged 12 to 20 months were bred by
natural mating in multisire groups of 25 to 30 heifers per bull. A breeding soundness
examination of the bulls was performed two months prior to the onset of the breeding
season. Sampling in this herd was done over two consecutive days. The weather on the first
sampling day was cool, with intermittent rain and the sampling lasted the whole day until
the evening. On the second day, sampling lasted until midday and it was a warm day. On
both sampling days, the heifers did not have access to feed or water whilst in the holding
pens. The heifers that were sampled on the second day spent the first day in the holding
pen with the other group. It was cold and raining with a humidity of 76%, total daily rainfall
of 1.4 mm, minimum and maximum temperatures were 16.1 ˚C and 26.7 ˚C respectively
(SAWS) on the first day of sampling, but very hot on the second day with a humidity of 72 %,
minimum and maximum temperatures were 14.4 ˚C and 28.4 ˚C respectively (SAWS).
Informed consent was obtained from all herd owners before commencement of the study.
21
© University of Pretoria
6.2 Experimental design
The study was a prospective cohort study. The sample size was estimated at 438 heifers
using the formula below, based on the normal approximation to the binomial assuming
equal group sizes at a power of 80% and allowable alpha error of 5% (Fosgate, 2009):
,
where P1 and P2 are the expected proportions in each group, and
is the average of the
expected proportions. Variables Z1-α/2 and Zβ are the standard normal Z values
corresponding to the selected alpha (2-sided test) and beta, respectively.
All farmers were blinded to RTS, BCS and BUN concentration data.
6.3 Experimental procedures
The first visit to all the farms was performed within one week prior to the commencement
of the breeding season. All heifers were driven into a holding pen, and from there they
would enter the crush in batches averaging 20 animals and blood samples were collected by
venepuncture from the coccygeal vein or artery into evacuated serum tubes. Immediately
after sampling, the blood was centrifuged at 4000 rpm for 8 minutes. Serum was separated
into labelled micro centrifuge tubes (2 ml) and then immediately frozen in a portable freezer
at -18 °C. Delivery of the serum samples to the clinical pathology laboratory at the Faculty of
Veterinary Science of the University of Pretoria was done on the first or second day after
sampling, where the serum was frozen at -80 °C for a maximum of 30 days until analysed.
Analysis for all samples was done using an auto analyser machine (Cobas Integra 400 plus,
Roche, Switzerland).
After blood sampling, BCS and RTS were performed and recorded. BCS was assigned based
on a 1 to 5 scale whereby score 1 represents emaciated animals and score 5 represents
obese animals (Wildman et al., 1982) with scores further subdivided into halves. The
22
© University of Pretoria
technique used for the scoring combined visual assessment of the whole animal and tactile
assessment of the loin area and the ischiorectal fossa (Table 6.1).
Table 6.1: BCS system (Wildman et al., 1982).
Score
1
2
3
4
5
Description
Emaciated cow, distinct, sharp spinous processes, very prominent hooks,
pins and tail head, line between hook & pin bone is V shaped, deep pigeon
holes next to tail head and deep sunken area around hip joint (thurl area).
Less prominent spinous processes that feel rounded rather than sharp,
half of the short rib covered with fat, hook and pin bones still prominent,
line between hook and pin bone less V shaped, thurl area sunken and
pigeon holes still deep.
Backbone forms straight line, individual processes still palpable, two thirds
of short ribs covered with fat, hook and pin bones round and smooth, thurl
area slightly depressed, pigeon holes have some fat and sacral ligaments
less distinct.
Spinous processes of backbone not visible or palpable anymore, short ribs
totally covered with fat, hook and pin bone rounded, span between
backbone and hook and pin bones is flat and pigeon holes nearly filled
with fat.
Over conditioned cow, bone structure of backbone, short ribs and hook
and pin bones not visible, subcutaneous fat deposits very evident, whole
back area can be compared with a rounded table top and tail head buried
in fat.
The RTS score (Table 6.2) was determined using rectal palpation of the reproductive tract
and ovarian structures and a score from 1 to 5 was assigned (Andersen et al., 1991; Holm et
al., 2009).
23
© University of Pretoria
Table 6.2: RTS system (Adapted from Andersen et al., 1991; Holm et al., 2009).
RTS Uterine horn
1
2
3
4
5
Immature < 20 mm
diameter, no tone
20- to 25 mm
diameter, no tone
25- to 30 mm
diameter, slight tone
30 mm diameter,
good tone
>30 mm diameter,
good tone, erect
Ovary
Length Height Width
Ovarian structures
(mm) (mm) (mm)
15
10
8
No palpable structures
18
12
10
8 mm follicles
22
15
10
8 to 10 mm follicles
30
16
12
>32
20
15
>10 mm follicles, corpus
luteum possible
>10 mm follicles, corpus
luteum present
Four weeks after the breeding season, a combination of transrectal palpation and
ultrasonography was performed to determine pregnancy status and stage. All examinations
were performed by the researcher using a portable ultrasound machine and a 3.5 - 5 MHz
linear transducer (CTS900V, Shantou Institute for Ultrasonic Instruments, China).
Rectal palpation was performed first, following the basic technique for pregnancy
examination (PE) in cattle (Sheldon and Noakes, 2002; Youngquist, 2007). The stage of
pregnancy was recorded in weeks. In cases where the pregnancy was judged to be more
than 8 weeks, the stage of pregnancy was determined solely by rectal palpation (Table 6.3).
24
© University of Pretoria
Table 6.3: Determination of stage of pregnancy by rectal palpation (adapted from Sheldon
and Noakes, 2002; Youngquist, 2007).
Palpable structures
CL ipsilateral to pregnant horn
Asymmetry and fluctuation of pregnant
horn (Ipsilateral to CL)
Amniotic vesicle
Allantochorion
Foetus
Placentomes
Fremitus on pregnant horn
Fremitus on non-pregnant horn
4
6
Pregnancy stage in weeks
8 10 12 14 16 18 20 22 24
The green bars represent periods during which different structures can be clearly palpated
while the light green bar delineates the period when structures can be palpated with some
difficulty. Absence of a green bar indicates that the structure cannot be palpated during that
period.
In cases where the pregnancy was determined to be less than 8 weeks on palpation,
ultrasonography was performed to differentiate an early pregnancy from a non-gravid
uterus. A presumptive diagnosis of an early pregnancy was made if nonechodense fluid was
seen in the uterine lumen and a corpus luteum was present on the ipsilateral ovary. This
was classified as a 4 week old pregnancy. Visualization of the allantochorion or embryo in
the uterine lumen was also attempted, and where either was seen, it was used as a
confirmation of pregnancy (Romano et al., 2006). For estimation of the age of the foetus,
the crown rump length formula (CRL) was used (Riding et al., 2008):
= −0.0009 ² + 0.5509 + 29.184,
where = CRL (mm) and
= estimated foetal age (days).
The farmers were requested to observe and record the days that the heifers were seen to
be mated. This data, when available, was used to verify the accuracy of the estimated foetal
age.
25
© University of Pretoria
6.4 Observations
The exposure of interest was BUN concentration while the other covariates were BCS, RTS,
age and body mass. The outcomes of interest were pregnancy status per three-month
breeding season and DTP. DTP was calculated as the number of days from the onset of the
breeding season to the fixed time of PE.
The units for BUN concentration in most literature is given in mg/dL, whilst in this study BUN
concentration was measured in mmolL-1. BUN concentration is converted from mg/dL to
mmol/L by multiplying the value in mg/dL by 0.357 (Tresley and Sheean, 2008).
6.5 Data analysis
The normality assumption was assessed by plotting histograms, calculating descriptive
statistics, and performing the Anderson-Darling Test. Data satisfying the normality
assumption were presented as mean +/- standard deviation (SD) and non-normal data were
presented as the median and interquartile range (IQR). Data were compared between days
when herd sampling required multiple days to complete. Normally distributed data were
compared using Student t tests and non-normal data using Mann-Whitney U tests.
Conditional logistic regression analysis was performed to measure the association between
BUN concentration and subsequent pregnancy while adjusting for herd as the grouping
factor and other potential confounders by including them as main effects in the models.
Confounding was assessed by measuring the change in the odds ratio (OR) for BUN
concentration in models with and without the covariate. Variables that caused 15% or
greater change were considered important confounders.
Stratified Cox proportional hazards survival analysis was performed to investigate the effect
of BUN concentration on the DTP. Herd was included as the stratifying factor and other
potential confounders were evaluated as main effects. Sampling day was forced into all
models in herds where sampling required two days. The odds ratio (OR) in all analyses was
interpreted as the odds of becoming pregnant. The hazards is a rate defined as the number
26
© University of Pretoria
of pregnant animals divided by the time that these animals were at risk of becoming
pregnant. Therefore, the hazards ratio is an inverse of the average DTP. The hazards ratio
(HR) is the estimate of effect analogous to the OR calculated from logistic regression. This
means that when the HR is less than one, the relationship between the variable and the DTP
is positive.
The ordinal scales of RTS and BCS were screened as ordinal variables in statistical models
and dichotomized when significant associations (P < 0.2) with BUN were identified.
Categorization was performed based on the relative frequencies within each category.
Specifically, RTS was grouped as 1 to 3 versus 4 and 5. Results for the ordinal coding were
reported when categorization did not suggest violation of the assumption of being linear in
the natural logarithm on the odds or hazard scale. In addition to confounding variables, all
variables with P < 0.2 were entered into all multivariable models and removed one by one in
a backward elimination process based on Wald P values.
Pearson’s correlation coefficient was used to evaluate the relationship between normally
distributed data and Spearman’s rho was calculated for non-normal data. Results for
variables that were not significantly correlated to any other variable in any herd were not
presented. The trend in BUN concentrations over sampling order was described using
scatter plots and simple linear regression. Data were analysed using commercially available
software (IBM SPSS Statistics Version 20, International Business Machines Corp., Armonk,
NY, USA; MINITAB Statistical Software, Release 13.32, Minitab Inc., State College,
Pennsylvania, USA).
P-values less than 0.05 were defined as being significant, values between 0.05 and 0.1 as
being close to significant and values greater than 0.1 as being not significant.
6.6 Experimental animals
All study animals were managed at their respective farms as described for the different
herds (Section 6.1). On the day of sample collection and PE, all the heifers were moved into
a holding pen and then into a crush in batches averaging 20 animals at a time. Farm
27
© University of Pretoria
management was responsible for routine daily care of the animals and normal farming
practices continued throughout the study. The researcher had no control over the loss of
animals due to death, sales or loss of identification tags, which occurred between sample
collection and the subsequent follow up at the time of PE.
28
© University of Pretoria
7 Results
7.1 All herds combined
7.1.1 Descriptive statistics
Three hundred and sixty-nine heifers were sampled from five different farms at the start of
the study and 338 were available at the time of examination for pregnancy (Table 7.1).
Heifers that did not become pregnant were considered right censored, whilst those that
were lost to follow up were excluded from the analysis. The mean breeding age ± the
standard deviation (SD) for all herds was 19.03 months ± 4.54 (assuming that the heifers in
herd A were born on the 15th of the month) and the overall pregnancy proportion was
63.1%. The mean BUN concentration for all herds was 5.27 mmolL-1 ± 1.80 at the onset of
the breeding season. Heifers had a median BCS of 3.00 [Interquartile range (IQR): 3.00, 3.00]
and RTS of 4.00 (IQR: 3.00, 4.00). The median BCS increased to 3.50 (IQR: 3.00, 3.50) at the
time of PE. The earliest heifers became pregnant 4 days after the onset of the breeding
season and the median DTP was 40 (IQR: 19, 54) days.
Table 7.1: Descriptive statistics for all herds (n = 369)
Variable
Age
Mass
BUN
BCS
RTS
DTP
BCS at PE
Q3
Max
N Mean SE Mean
SD
Min
Q1 Median
333 19.03
0.25 4.54 11.93 14.03
18.77 22.57 30.57
221 283.00
3.24 48.19 190.00 244.00 271.00 305.00 440.00
368
5.27
0.09 1.80
1.20
4.30
5.20
6.50
9.70
367
3.00
0.02 0.30
2.00
3.00
3.00
3.00
4.00
367
3.78
0.04 0.71
1.00
3.00
4.00
4.00
5.00
233 38.00
1.43 21.75
4.00 19.00
40.00 54.00 120.00
328
3.34
0.02 0.36
2.50
3.00
3.50
3.50
4.50
The mean breeding age ± SD for heifers that became pregnant was 20.53 months ± 4.23
whilst for those non-pregnant was 16.30 months ± 3.77 (Table 7.2). The mean body mass for
pregnant heifers was 297 kg ± 45.12 whilst for those non-pregnant ones was 259 kg ± 44.88.
The mean BUN concentration for heifers that became pregnant was 4.87 mmolL-1 ± 1.73
whilst that for those non-pregnant was 6.04 mmolL-1 ± 1.64. The median BCS at the onset of
the breeding season was 3.00 (IQR: 3.00, 3.00) for pregnant and non-pregnant heifers. The
29
© University of Pretoria
median RTS for pregnant heifers was 4.00 (IQR: 3.00 – 4.00) whilst for those non-pregnant
was 4.00 (IQR: 3.50 – 4.00).
Table 7.2: Descriptive statistics by pregnancy status for all herds, indicating pregnant (n =
233), non-pregnant (n = 102) and lost to follow up (n = 31) heifers
Min
Q1
Median Q3
Max
217 20.53
SE
SD
Mean
0.29
4.23
12.17
18.77
19.77
24.53
30.57
95
0.39
11.93
13.43
14.03
18.77
29.57
Var
PE status
N
Age
preg
non-preg
lost
Mass preg
Mean
16.30
3.77
21 16.07 0.76
138 297.00 3.84
3.47 12.53 13.30 13.67
45.12 220.00 262.00 295.00
18.77 24.03
324.00 435.00
non-preg
72
259.00 5.29
44.88 190.00 234.00 244.00
265.00 440.00
lost
preg
12 256.00 9.12
233 4.87
0.11
31.59 229.00 233.00 244.00
1.73 1.20
3.70
5.00
267.00 321.00
6.10
9.50
non-preg
102 6.04
0.16
1.64
2.20
4.90
5.80
7.20
9.70
lost
preg
30 5.67
233 3.03
0.35
0.02
1.93
0.33
2.50
2.00
4.50
3.00
5.30
3.00
6.85
3.00
9.70
4.00
non-preg
102 2.95
0.02
0.22
2.50
3.00
3.00
3.00
3.50
lost
preg
32 3.00
233 3.76
0.06
0.05
0.36
0.78
2.50
1.00
3.00
3.00
3.00
4.00
3.00
4.00
4.00
5.00
non-preg
102 3.83
0.06
0.58
2.00
3.50
4.00
4.00
5.00
lost
29 3.72
0.08
0.45 3.00
3.00
4.00
4.00
Var = variable; preg = pregnant; lost = animals that were missing at the time of PE
4.00
BUN
BCS
RTS
7.1.2 Logistic regression analysis
7.1.2.1 Single variable logistic regression
When all herds were combined, and Herd E modelled as two separate herds based on
sampling day, age (P = 0.010) and body mass (P = 0.037) were the only significant predictors
of pregnancy outcome (Table 7.3). The older (OR = 1.082; 95% CI: 1.019 – 1.149) and heavier
30
© University of Pretoria
(OR = 1.006; 95% CI: 1.000 – 1.012) heifers were more likely to become pregnant when
compared to the lighter and younger ones. Herd was included as the grouping variable in
the conditional logistic regression model.
Table 7.3: Effects of BUN concentration, age, body mass and RTS on pregnancy status in all
herds combined (single variable logistic regression)
Variable
B
OR
95% CI of OR P value
Lower Upper
BUN
-0.126 0.882 0.772 1.007
0.063
Age
0.079 1.082 1.019 1.149
0.010
Mass
0.006 1.006 1.000 1.012
0.037
RTS
-0.102 0.903 0.746 1.093
0.297
B = beta; OR = odds ratio and CI = confidence interval
7.1.2.2 Multivariable logistic regression
When combined with potential confounding variables, BUN concentration was still not
significant (P = 0.068; Table 7.4).
31
© University of Pretoria
Table 7.4: The combined effects of BUN concentration and age on pregnancy status in all
herds (multivariable logistic regression
Variable
B
OR
95% CI of OR P value
Lower Upper
BUN
-0.127 0.881 0.768 1.009
0.068
Age
0.079 1.082 1.019 1.149
0.011
B = beta; OR =odds ratio and CI = confidence interval
7.1.3 Proportional hazards survival analysis
7.1.3.1 Single variable survival analysis
BUN concentration (P = 0.007), age (P < 0.001), body mass (P = 0.001) and RTS (P = 0.023)
were significant predictors of DTP (Table 7.5). Heifers with a higher BUN concentration (HR =
0.827; 95% CI: 0.721 – 0.949) and RTS (HR = 0.797; 95% CI: 0.656 – 0.969) had a higher
number of DTP. The older (HR = 1.130; 95% CI: 1.061 – 1.202) and heavier (HR = 1.010; 95%
CI: 1.004 – 1.016) heifers had a smaller number of DTP.
Table 7.5: Effects of BUN concentration, age, body mass and RTS on DTP in all herds
combined (single variable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.190 0.827 0.721 0.949
0.007
Age
0.122 1.130 1.061 1.202 <0.001
Mass
0.010 1.010 1.004 1.016
0.001
RTS
-0.226 0.797 0.656 0.969
0.023
B = beta; HR = hazard ratio and CI = confidence interval
7.1.3.2 Multivariable survival analysis
BUN concentration was a significant predictor of DTP (P = 0.012) when adjusting for age
(Table 7.6). Other evaluated covariates were not significant predictors and did not cause
substantial confounding (< 15 %). Herd was included in the conditional logistic regression
model as the grouping variable.
32
© University of Pretoria
Table 7.6: The combined effects of BUN concentration and age on DTP in all herds
(multivariable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.184 0.832 0.722 0.958
0.011
Age
0.122 1.130 1.061 1.203 <0.001
B = beta; HR = hazard ratio and CI = confidence interval
7.1.4 Correlation analysis
A significant positive correlation was estimated between RTS with age (P = 0.004) and BCS (P
< 0.001; Figure 7.1). A significant negative correlation was estimated between BUN
concentration with age (P <0.001) and body mass (P < 0.001). Age was significantly
correlated to body mass (P < 0.001) and BCS (P = 0.021). All the other variables were not
significantly correlated (P >0.05).
33
© University of Pretoria
Figure 7.1: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in all herds combined.
7.2 Herd A
7.2.1 Descriptive statistics
One hundred and fifteen heifers were sampled from this herd at the commencement of the
study and 106 were available at the time of PE (Table 7.7). Heifers in this herd were bred at
a mean age ± SD of 18.6 months ± 1.27 and had a pregnancy proportion of 71.3%. However,
exact age was not known as only the month in which they were born was recorded. In this
herd, it was assumed that all heifers were born on the 15th of the given month. Heifers that
were sampled had a mean BUN concentration of 5.37 mmolL-1 ± 0.81 at the onset of the
breeding season. Heifers were considered mature and ready for breeding based on their
RTS. Median BCS was 3.00 (IQR: 3.00 – 3.00) before breeding and increased to 3.50 (IQR:
3.00 – 3.50) at the time of PE. The earliest heifer conceived 15 days after the onset of the
breeding season and the median DTP was 43 (IQR: 29, 71) days. Body mass of the heifers
were not available in this herd.
34
© University of Pretoria
Table 7.7: Descriptive statistics for Herd A (n = 115)
Variable
n Mean SE Mean
SD
Min
Q1 Median
Q3
Max
Age
110
18.6
0.1
1.3
14
19
19
19
24
BUN
115
5.37
0.08
0.81
2.70
4.90
5.30
5.90
7.50
BCS
115
2.90
0.02
0.20
2.50
3.00
3.00
3.00
3.00
RTS
115
3.77
0.05
0.50
2.00
4.00
4.00
4.00
5.00
106
BCS at PE 106
67.00
3.40
DTP
5.00 50.00 15.00 29.00
0.03 0.28 2.50 3.00
43.00 71.00 155.00
3.50 3.50
4.00
There was no significant linear trend between BUN concentration and sampling order (P =
0.548; Figure 7.2). The least squares regression line for this herd was
= 0.0004 + 5.3451,
where
is the BUN concentration and
is the sampling order.
8
7
BUN (mmolL-1)
6
5
4
3
2
1
0
0
20
40
60
80
100
120
Sampling order
Figure 7.2: BUN concentration in relation to sampling order in Herd A
Mean BUN concentration ± SD for heifers that became pregnant was 5.37 mmolL-1 ± 0.77
whilst for those non-pregnant was 5.45 mmolL-1 ± 0.91 (Table 7.8). The median BCS at the
35
© University of Pretoria
onset of the breeding season was 3.00 (IQR: 3.00 – 3.00 for both pregnant and nonpregnant heifers. The median RTS for the pregnant heifers was 4.00 (IQR: 3.00 – 4.00) whilst
for those non-pregnant was 4.00 (IQR: 4.00 – 4.00).
Table 7.8: Descriptive statistics by pregnancy status for Herd A, indicating pregnant (n = 82),
non-pregnant (n = 24) and lost to follow up (n = 9) heifers
Var
Age
PE status n Mean SE Mean SD
Min
Q1
Median Q3
Max
preg
78 18.80 0.12
1.03 14.67 18.77 18.77
18.77 23.8
non-preg 23 18.37
0.27
1.27 14.67 18.77 18.77
18.77 19.77
0.73
0.09
2.20 13.67 15.70 18.77
0.77 2.70 4.90 5.35
18.77 18.77
5.80 7.50
non-preg 24 5.45
0.19
0.91 3.90
4.90
5.35
6.20
7.50
lost
preg
9 5.18
82 2.92
0.33
0.02
0.98 4.20
0.19 2.50
4.35
3.00
5.10
3.00
5.90
3.00
7.00
3.00
non-preg 24 2.90
0.04
0.21 2.50
3.00
3.00
3.00
3.00
lost
preg
9 2.83
82 3.70
0.08
0.06
0.25 2.50
0.51 2.00
2.50
3.00
3.00
4.00
3.00
4.00
3.00
4.00
non-preg 24 4.04
0.07
0.36 3.00
4.00
4.00
4.00
5.00
9 3.67
0.17
0.50 3.00 3.00 4.00
4.00
lost
Var = variable; preg = pregnant; lost = animals that were missing at the time of PE
4.00
lost
BUN preg
BCS
RTS
9 17.53
82 5.37
7.2.2 Logistic regression analysis
7.2.2.1 Single variable logistic regression
RTS was the only significant predictor of pregnancy status in this herd (P = 0.017; Table 7.9).
Those heifers which had a higher RTS had lower chances of becoming pregnant (OR = 0.083;
95% CI: 0.011 – 0.638). When using the dichotomised RTS data, RTS was a significant
predictor of pregnancy status (B = -2.149, OR = 0.268 and P = 0.041).
36
© University of Pretoria
Table 7.9: Effects of BUN concentration, age, body mass and RTS on pregnancy status in
Herd A (single variable logistic regression)
Variable
B
OR
95% CI of OR P value
Lower Upper
BUN
-0.182 0.833 0.460 1.511
0.548
Age
0.331 1.392 0.923 2.099
0.115
Mass
RTS
-2.494 0.083 0.011 0.638 0.017
B = beta; OR = odds ratio and CI = confidence interval
7.2.2.2 Multivariable logistic regression
Multivariable analysis did not identify a model that was an improvement over models with
only a single predictor.
7.2.3 Proportional hazards survival analysis
7.2.3.1 Single variable survival analysis
RTS was the only significant predictor of DTP (P = 0.023) but age was close to the
significance threshold (P = 0.051; Table 7.10). Heifers with a higher RTS took longer to
become pregnant than those with a lower RTS (HR = 0.636; 95% CI: 0.431 - 0.939). Using the
dichotomised RTS data, RTS was close to significance in predicting DTP (B = -0.487, HR =
0.614 and P = 0.053).
37
© University of Pretoria
Table 7.10: Effects of BUN concentration, age, body mass and RTS on DTP in Herd A (single
variable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.031 0.969 0.737 1.276
0.824
Age
0.266 1.305 0.999 1.704
0.051
Mass
RTS
-0.453 0.636 0.431 0.939
0.023
B = beta; HR = hazard ratio and CI = confidence interval
7.2.3.2 Multivariable survival analysis
Multivariable survival analysis did not identify a model that was an improvement over
models with only a single predictor.
7.2.4 Correlation analysis
RTS and age were positively correlated (P = 0.043; Figure 7.3). No other significant
correlations were identified (P > 0.05). Correlations between body mass and other variables
could not be computed since the body masses were not available.
38
© University of Pretoria
Figure 7.3: Pearson’s correlation for BUN with age, and Spearman’s correlation for BCS with
age and BUN concentration, in Herd A
7.3 Herd B
7.3.1 Descriptive statistics
Thirty-three heifers were sampled at the commencement of the study and all of them were
pregnant at the time of PE. Heifers were bred at a mean age ± SD of 24.87 months ± 1.50
and had a 100% pregnancy proportion (Table 7.11). Sampled heifers had a mean mass of
318 kg ± 32, and a mean BUN concentration of 2.20 mmolL-1 ± 0.67. Median BCS and RTS
were 3.00 (IQR: 2.50 – 3.00) and 3.00 (IQR: 2.75 – 4.00) respectively. The earliest heifer
became pregnant 10 days after the start of the breeding season and the median DTP was 24
(IQR: 17, 31) days.
39
© University of Pretoria
Table 7.11: Descriptive statistics for Herd B (n = 33)
Variable
n
Mean SE Mean
24.87
0.27
SD
Min
1.50
20.43
Q1 Median
26.13
26.37
33
Mass
33 318.00
6.00 32.00 242.00 290.00
BUN
33
2.20
0.11
0.67
1.20
1.70
2.00
2.80
4.40
BCS
33
2.80
0.05
0.27
2.00
2.50
3.00
3.00
3.00
RTS
33
3.27
0.18
1.02
1.00
2.75
3.00
4.00
5.00
DTP
33
26.00
2.00 14.00
10.00
17.00
24.00
31.00
59.00
BCS at PE 33
3.92
3.50
4.00
4.00
4.00
4.50
0.25
25.23
Max
Age
0.04
24.13
Q3
323.00 342.00 371.00
There was no significant linear trend between BUN concentration and sampling order (P =
0.205; Figure 7.4). The least squares regression line for this herd was:
= −0.0158 + 2.4875,
where
is the BUN concentration and
is the sampling order.
5.0
4.5
BUN (mmolL-1)
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
5
10
15
20
Sampling order
25
30
35
Figure 7.4: BUN concentration in relation to sampling order in Herd B
7.3.2 Logistic regression analysis
Logistic regression was not possible because the pregnancy proportion was 100%.
40
© University of Pretoria
7.3.3 Proportional hazards survival analysis
7.3.3.1 Single variable survival analysis
In this herd, none of the evaluated variables significantly predicted DTP (P > 0.05; Table
7.12).
Table 7.12: Effects of BUN concentration, age, body mass and RTS on DTP in Herd B (single
variable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
0.107 1.113 0.644 1.926
0.701
Age
-0.139 0.870 0.663 1.142
0.317
Mass
0.002 1.002 0.990 1.013
0.793
RTS
-0.051 0.951 0.671 1.347
0.776
B = beta; HR = hazard ratio and CI = confidence interval
7.3.3.2 Multivariable survival analysis
Multivariable survival analysis was not possible because none of the variables were
significant predictors of DTP.
7.3.4 Correlation analysis
There was a positive correlation between age and body mass (P = 0.002) and BCS and body
mass (P = 0.009; Figure 7.5). No other variables were significantly correlated (P > 0.05).
41
© University of Pretoria
Figure 7.5: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd B
7.4 Herd C
7.4.1 Descriptive statistics
Thirty-four heifers were sampled at the beginning of the study, and all were pregnant at the
time of PE (Table 7.13). Mean age ± SD at breeding was 25.73 months ± 0.87 and the
pregnancy proportion was 100%. Sampled heifers had a mean mass of 297 kg ± 29 and a
mean BUN concentration of 4.14 mmolL-1 ± 0.92. Their median BCS and RTS was 3.50 (IQR:
3.50 – 3.50) and 5.00 (4.00 – 5.00) respectively. The earliest heifer became pregnant 4 days
after the onset of the breeding season and the median DTP was 4 (IQR: 4, 25).
42
© University of Pretoria
Table 7.13: Descriptive statistics for Herd C (n=34)
Variable
n
Mean SE Mean
25.73
0.17
SD
Min
0.87
23.87
Q1 Median
26.13
Max
26.47
27.17
Age
31
Mass
31 297.00
5.00 29.00 220.00 287.00
BUN
34
4.14
0.15
0.92
2.60
3.45
4.20
4.68
6.90
BCS
34
3.60
0.04
0.23
3.00
3.50
3.50
3.50
4.00
RTS
34
4.69
0.09
0.52
3.00
4.00
5.00
5.00
5.00
DTP
34
18.00
4.00 24.00
4.00
4.00
4.00
25.00
81.00
BCS at PE 26
3.38
0.05
3.00
3.00
3.50
3.50
4.00
0.26
24.83
Q3
300.00 304.00 361.00
There was no significant linear trend between BUN concentration and sampling order (P =
0.438; Figure 7.6). The least squares regression line for this herd was:
= 0.0105 + 3.9775,
where
is the BUN concentration and
is the sampling order.
8
7
BUN mmolL-1
6
5
4
3
2
1
0
0
5
10
15
20
25
30
35
Sampling order
Figure 7.6: BUN concentration in relation to sampling order in Herd C
43
© University of Pretoria
7.4.2 Logistic regression analysis
Logistic regression was not possible in this herd because the pregnancy proportion was
100%.
7.4.3 Proportional hazards survival analysis
7.4.3.1 Single variable survival analysis
In this herd, none of the evaluated variables significantly predicted DTP (P > 0.05; Table
7.14).
Table 7.14: Effects of BUN concentration, age, body mass and RTS on DTP in Herd C (single
variable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.003 0.997 0.647 1.535
0.989
Age
-0.157 0.855 0.580 1.259
0.427
Mass
0.002 1.002 0.989 1.014
0.796
RTS
-0.244 0.784 0.407 1.510
0.466
B = beta; HR = hazard ratio and CI = confidence interval
7.4.3.2 Multivariable logistic regression
Multivariable logistic regression was not possible because none of the evaluated variables
were significant.
7.4.4 Correlation analysis
Age and body mass were positively correlated (P < 0.001; Figure 7.7). No other variables
were significantly correlated (P > 0.05).
44
© University of Pretoria
Figure 7.7: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd C
7.5 Herd D
7.5.1 Descriptive statistics
Twenty-nine heifers were sampled at the beginning of the study, and 22 were later
examined at the time of PE (Table 7.15). Heifers were bred at a mean age of 24.7 months ±
2.27 and their pregnancy proportion was 68.2%. Sampled heifers had a mean mass of 380 kg
± 38 and a mean BUN concentration of 4.50 mmolL-1 ± 1.80. The median BCS and RTS were
3.00 (IQR: 3.00 – 3.00) and 4.00 (IQR: 3.00 – 4.00) respectively. At the time of PE, the
median BCS had increased to 3.50 (IQR: 3.50 – 3.50). The earliest heifer became pregnant 12
days after the onset of the breeding season and the median DTP was 47 (IQR: 40, 61) days.
45
© University of Pretoria
Table 7.15: Descriptive statistics for Herd D (n = 29)
Variable
Age
Mass
BUN
BCS
RTS
DTP
BCS at PE
n Mean SE Mean
SD
Min
Q1 Median
Q3
Max
23 24.70
0.47 2.27 21.73 23.00
24.13 24.90 30.57
20 380.00
8.00 38.00 320.00 343.00 385.00 410.00 440.00
29
4.50
0.33 1.80
1.50
2.75
4.70
6.00
7.00
29
3.00
0.05 0.27
2.50
3.00
3.00
3.00
3.50
29
3.62
0.17 0.90
1.00
3.00
4.00
4.00
5.00
15 46.00
5.00 18.00 12.00 40.00
47.00 61.00 75.00
22
3.52
0.06 0.29
3.00
3.50
3.50
3.50
4.00
There was a significant linear trend between BUN concentration and sampling order (P <
0.001; Figure 7.8). The least squares regression line for this herd was
= 0.1721 + 1.9155,
where
is the BUN concentration and
is the sampling order.
8
7
BUN (mmolL-1)
6
5
4
3
2
1
0
0
5
10
15
Sampling order
20
25
30
Figure 7.8: BUN concentration in relation to sampling order in Herd D
The mean body mass ± SD for heifers that became pregnant was 382 kg ± 37 whilst that for
those non-pregnant was 375 kg ± 40 (Table 7.16). The mean BUN concentration for the
pregnant heifers was 3.98 mmolL-1 ± 1.77 whilst that for those non-pregnant was 5.84
mmolL-1 ± 1.43. Median BCS for pregnant heifers was 3.00 (IQR: 3.00 – 3.00) whilst that for
46
© University of Pretoria
those non-pregnant was 3.00 (IQR: 3.00 – 3.00). The median RTS for pregnant heifers was
3.00 (IQR: 3.00 – 4.00) whilst that for those non-pregnant was and 4.00 (IQR: 3.00 – 5.00).
Table 7.16: Descriptive statistics by pregnancy status for Herd D, indicating pregnant (n =
15), non-pregnant (n = 7) and lost to follow up (n = 7) heifers
Var
Age
PE status n Mean
preg
15 24.73
SEMean SD
0.57
2.20
Min
22.77
Q1
23.73
Median Q3
24.13
24.70
Max
30.57
non-preg
1.03
21.73
22.40
24.67
29.57
lost
Mass preg
7
24.80
2.73
26.57
1 24.03 13 382.00 11.00
24.03 24.03
37.00 320.00 343.00 390.00
24.03
413.00 435.00
non-preg
7
40.00 325.00 340.00 375.00
400.00 440.00
lost
preg
0 15 3.98
0.46
1.77
1.50
2.50
3.70
5.90
7.00
non-preg
7
5.84
0.54
1.43
2.80
5.60
6.50
6.70
6.90
lost
preg
7 4.29
15 3.00
0.64
0.05
1.70
0.19
2.50
2.50
2.60
3.00
4.70
3.00
5.40
3.00
6.80
3.50
non-preg
7
3.00
0.11
0.29
2.50
3.00
3.00
3.00
3.50
lost
preg
7 3.00
15 3.33
0.15
0.25
0.41
0.98
2.50
1.00
2.50
3.00
3.00
3.00
3.50
4.00
3.50
5.00
non-preg
7
0.34
0.90
3.00
3.00
4.00
5.00
5.00
lost
7 3.71
0.18
0.49 3.00
3.00
4.00
4.00
Var = variable; preg = pregnant; lost = animals that were missing at the time of PE
4.00
BUN
BCS
RTS
375.00 15.00
4.14
7.5.2 Logistic regression analysis
7.5.2.1 Single variable logistic regression
BUN concentration was the only significant predictor of pregnancy status (P = 0.046; Table
7.17). Heifers which had a higher BUN concentration had reduced chances of becoming
pregnant (OR = 0.478; 95% CI: 0.232 – 0.987). When using the dichotomised RTS data, RTS
47
© University of Pretoria
was still not a significant predictor of pregnancy status (B = -1.050, OR = 0.350 and P =
0.286).
Table 7.17: Effects of BUN concentration, age, body mass and RTS on pregnancy status in
Herd D (single variable logistic regression)
Variable
B
OR
95% CI of OR P value
Lower Upper
BUN
-0.737 0.478 0.232 0.987
0.046
Age
-0.014 0.986 0.665 1.460
0.943
Mass
0.005 1.005 0.980 1.031
0.689
RTS
-1.079 0.340 0.097 1.187
0.091
B = beta; OR = hazard ratio and CI = confidence interval
7.5.2.2 Multivariable logistic regression
A multivariable model was not identified that improved the prediction of pregnancy status
over the models with only single variables.
7.5.3 Proportional hazards survival analysis
7.5.3.1 Single variable survival analysis
BUN concentration (P = 0.033) and RTS (P = 0.039) were significant predictors of DTP, while
age and mass were not (P > 0.05; Table 7.18). Heifers with a higher BUN concentration took
longer to become pregnant (HR = 0.719; 95% CI: 0.531 – 0.974). In addition, heifers that had
a higher RTS took longer to become pregnant (HR = 0.545; 95% CI: 0.306 – 0.970). When
using the dichotomised RTS data, RTS was still not a significant predictor of DTP (B = -0.516,
HR = 0.597 and P = 0.323).
48
© University of Pretoria
Table 7.18: Effects of BUN concentration, age, body mass and RTS on DTP in Herd D (single
variable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.330 0.719 0.531 0.974
0.033
Age
0.009 1.009 0.813 1.252
0.934
Mass
0.002 1.002 0.986 1.017
0.844
RTS
-0.607 0.545 0.306 0.970
0.039
B = beta; HR = hazard ratio and CI = confidence interval
7.5.3.2 Multivariable survival analysis
Neither BUN concentration (P = 0.207) nor RTS (P = 0.282) were significant predictors of DTP
when combined together in a multivariable model (Table 7.19).
Table 7.19: The combined effects of BUN concentration and RTS on DTP in Herd D
(multivariable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.225 0.799 0.564 1.132
0.207
RTS
-0.373 0.689 0.349 1.358
0.282
B = beta; HR = hazard ratio and CI = confidence interval
7.5.4 Correlation analysis
Age and body mass (P = 0.024), and BUN concentration and RTS (P = 0.001) were positively
correlated (Figure 7.9). No other variables were significantly correlated (P > 0.05).
49
© University of Pretoria
Figure 7.9: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd D
7.6 Herd E
7.6.1 Descriptive statistics
One hundred and fifty-eight heifers were sampled at the commencement of the study and
143 were available at the time of PE (Table 7.20 and Table 7.21). Sampled heifers were bred
at a mean age ± SD of 15.6 months ± 3.37 and had a pregnancy proportion of 47.5%. The
mean body mass was 256.50 kg ± 27.00. The mean BUN concentration for the first and
second day of sampling was 7.40 mmolL-1 ± 1.14 and 4.77 mmolL-1 ± 1.16 respectively. The
median BCS and RTS for these heifers were 3.00 (IQR: 3.00 – 3.00) and 4.00 (IQR: 3.00 –
4.00) respectively. In this herd, the earliest heifer became pregnant 12 days after the onset
of the breeding season while the median DTP was 47 (IQR: 33, 61) days.
50
© University of Pretoria
Table 7.20: Descriptive statistics for Herd E on day 1 of sampling (n = 88)
Variable
Age
Mass
BUN
BCS
RTS
DTP
BCS
n Mean SE Mean
SD
Min
Q1 Median
Q3
Max
79 15.16
0.37 3.27 12.13 13.17
13.60 14.70 22.50
79 256.00
2.92 26.00 199.00 239.00 250.00 270.00 335.00
87
7.40
0.12 1.14
4.60
6.60
7.30
8.20
9.70
87
2.97
0.03 0.24
2.50
3.00
3.00
3.00
3.50
87
3.71
0.06 0.57
3.00
3.00
4.00
4.00
5.00
43 46.84
3.19 20.92 12.00 33.00
40.00 68.00 89.00
80
3.11
0.03 0.27
2.50
3.00
3.00
3.50
3.50
Table 7.21: Descriptive statistics for Herd E on day 2 of sampling (n =70)
Variable
Age
Mass
BUN
BCS
RTS
DTP
BCS at PE
n Mean SE Mean
SD
Min
Q1 Median
Q3
Max
58 16.03
0.47 3.47 11.93 13.53
14.10 20.67 21.50
58 257.00
4.00 28.00 190.00 238.00 251.00 275.00 340.00
70
4.77
0.14 1.16
2.20
4.00
4.65
5.45
8.90
69
2.98
0.03 0.25
2.50
3.00
3.00
3.00
3.50
69
3.74
0.07 0.61
1.00
3.50
4.00
4.00
5.00
26 46.00
4.00 20.00 12.00 31.00
47.00 56.00 89.00
63
3.17
0.03 0.25
3.00
3.00
3.00
3.50
4.00
There was no significant liner trend between BUN concentration and sampling order on the
first day of sampling (P = 0.099; r² = 0.032; Figure 7.10). The least squares regression line for
this group was
= −0.008
where
is the BUN concentration and
+ 7.7478,
is the sampling order.
51
© University of Pretoria
12
BUN (mmolL-1)
10
8
6
4
2
0
0
20
40
60
80
100
Sampling order
Figure 7.10: BUN concentration in relation to sampling order in Herd E on day 1 of sampling
On the second day of sampling, there was no significant linear trend between BUN
concentration and sampling order (P = 0.957; Figure 7.11). The least squares regression line
for this group was
= 9 − 06
+ 4.7725.
10
9
BUN (mmolL-1)
8
7
6
5
4
3
2
1
0
0
10
20
30
40
50
Sampling order
60
70
80
Figure 7.11: BUN concentration in relation to sampling order in Herd E on day 2 of sampling
52
© University of Pretoria
Mean breeding age ± SD for heifers that became pregnant was 16.99 months ± 3.64 whilst
those not pregnant had a mean breeding age of 14.65 months ± 2.57 (Table 7.22 and 7.23).
The mean mass for the heifers that became pregnant was 268.00 kg ± 27.00 whilst those not
pregnant had a mean breeding mass of 246.50 kg ± 21.00. On the first day of sampling, the
mean BUN concentration for heifers that became pregnant was 7.07 mmolL-1 ± 1.02 whilst
for those non-pregnant was 7.65 mmolL-1 ± 1.12. On the second day of sampling, the mean
BUN concentration for heifers that became pregnant was 4.52 mmolL-1 ± 1.06 whilst for
those non-pregnant was 4.86 mmolL-1 ± 1.15. Median BCS and RTS were 3.00 (IQR: 3.00,
3.00) and 4.00 (IQR: 3.00, 4.00) respectively for both pregnant and non-pregnant heifers.
Table 7.22: Descriptive statistics by pregnancy status for Herd E day 1, indicating pregnant
(n = 43), non-pregnant (n = 37) and lost to follow up (n = 8) heifers
Var
Age
PE Status n Mean
preg
40 16.27
SEMean SD
0.60
3.77
Min
12.13
Q1
13.27
Median Q3
13.90
20.97
Max
21.93
non-preg
0.33
12.30
13.10
13.47
21.27
lost
Mass preg
33 13.90
1.83
13.90
6 14.93 1.53
40 266.00 4.00
3.73 12.87 12.93 13.47
27.00 220.00 244.00 258.00
16.30 22.50
289.00 335.00
non-preg
33 245.00 3.00
18.00 199.00 234.00 242.00
258.00 288.00
lost
preg
6 256.00 13.00
43 7.07
0.16
32.00 230.00 231.00 247.00
1.02 4.60
6.40
7.00
278.00 315.00
7.60
9.50
non-preg
37 7.65
0.18
1.12
5.40
6.60
7.80
8.40
9.70
lost
preg
7 8.09
43 2.95
0.54
0.04
1.42
0.26
5.80
2.50
6.60
3.00
8.20
3.00
9.20
3.00
9.70
3.50
non-preg
37 2.97
0.03
0.20
2.50
3.00
3.00
3.00
3.50
lost
preg
7 3.00
43 3.63
0.11
0.09
0.29
0.58
2.50
3.00
3.00
3.00
3.00
4.00
3.00
4.00
3.50
5.00
non-preg
37 3.76
0.10
0.60
3.00
3.00
4.00
4.00
5.00
lost
7 4.00
0
0
4.00
4.00
4.00
4.00
Var = variable; preg = pregnant; lost = animals that were missing at the time of PE
4.00
BUN
BCS
RTS
53
© University of Pretoria
Table 7.23: Descriptive statistics by pregnancy status for Herd E day 2, indicating pregnant
(n = 26), non-pregnant (n = 37) and lost to follow up (n = 7) heifers
Var
Age
PE status n Mean
preg
21 17.70
SEMean SD
0.76
3.50
Min
13.17
Q1
14.23
Median Q3
20.03
21.03
Max
21.43
non-preg 32 15.40
0.57
11.93
13.43
13.83
21.50
lost
Mass preg
BUN
3.30
18.70
5 13.20 0.20
21 270.00 6.00
0.43 12.53 12.87 13.20
27.00 220.00 250.00 265.00
13.57 13.63
284.00 340.00
non-preg 32 248.00 4.00
24.00 190.00 233.00 244.00
263.00 310.00
lost
preg
37.00 229.00 233.00 246.00
1.06 2.40
3.70
4.40
295.00 321.00
5.18
6.80
5 260.00 17.00
26 4.52
0.21
non-preg 37 4.86
0.19
1.15
2.20
4.20
4.80
5.50
8.90
lost
preg
7 5.26
26 3.00
0.56
0.06
1.47
0.28
3.50
2.50
4.50
3.00
4.80
3.00
5.80
3.00
8.10
3.50
non-preg 37 2.95
0.04
0.23
2.50
3.00
3.00
3.00
3.50
lost
preg
6 3.08
26 3.85
0.08
0.13
0.20
0.68
3.00
1.00
3.00
4.00
3.00
4.00
3.13
4.00
3.50
5.00
non-preg 37 3.70
0.09
0.57
2.00
3.00
4.00
4.00
5.00
lost
6 3.50
0.22
0.55 3.00
3.00
3.50
4.00
Var = variable; preg = pregnant; lost = animals that were missing at the time of PE
4.00
BCS
RTS
7.6.2 Logistic regression analysis
Sampling day was forced into all logistic regression models to account for potential
confounding.
7.6.2.1 Single variable logistic regression
BUN concentration (P = 0.012), age (P < 0.001) and body mass (P < 0.001) were significant
predictors of pregnancy status (
54
© University of Pretoria
Table 7.24). Animals with a higher BUN concentration were less likely to become pregnant
(OR = 0.656; 95% CI: 0.472 – 0.912). Older (OR = 1.264; 95% CI: 1.117 – 1.431) and heavier
(OR = 1.041; 95% CI: 1.021 – 1.061) animals were more likely to become pregnant.
Table 7.24: Effects of BUN concentration, age, body mass and RTS on pregnancy status in
Herd E (single variable survival analysis)
Variable
B
OR
95% CI of OR P value
Lower Upper
BUN
-0.421 0.656 0.472 0.912
0.012
Age
0.235 1.264 1.117 1.431 <0.001
Mass
0.040 1.041 1.021 1.061 <0.001
RTS
-0.030 0.971 0.557 1.691
0.916
Sample day 0.503 1.654 0.849 3.222
0.139
B = beta; OR = hazard ratio and CI = confidence interval
7.6.2.2 Multivariable logistic regression
BUN concentration (P = 0.017) was a significant predictor of pregnancy status when
adjusting for body mass (Table 7.25). Other evaluated covariates were not significant
predictors and did not cause substantial confounding (< 15 %).
Table 7.25: The combined effects of BUN concentration and body mass on pregnancy status
in Herd E (multivariable logistic regression)
Variable
B
OR
95% CI of OR P value
Lower Upper
BUN
-0.482 0.617 0.422 0.903
0.013
Mass
0.041 1.042 1.022 1.062 <0.001
Sample day 1.969 7.163 1.983 25.877
0.003
B = beta; OR = hazard ratio and CI = confidence interval
7.6.3 Proportional hazards survival analysis
All models were stratified by sampling day to account for potential confounding.
55
© University of Pretoria
7.6.3.1 Single variable survival analysis
BUN concentration (P = 0.008), age (P < 0.001) and mass (P < 0.001) were significant
predictors of DTP (Table 7.26). Heifers with a higher BUN concentration took a longer time
to conceive (HR = 0.736; 95% CI: 0.588 – 0.922). Older (HR = 0.1.159; 95% CI: 1.081 – 1.242)
and heavier (HR = 1.024; 95% CI: 1.015 – 1.033) heifers had shorter DTP.
Table 7.26: Effects of BUN concentration, age, body mass and RTS on DTP in Herd E (single
variable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.306 0.736 0.588 0.922
0.008
Age
0.148 1.159 1.081 1.242 <0.001
Mass
0.024 1.024 1.015 1.033 <0.001
RTS
-0.052 0.950 0.641 1.406
0.797
Sample day -0.341 0.711 0.437 1.157
0.170
B = beta; HR = hazard ratio and CI = confidence interval
7.6.3.2 Multivariable survival analysis
BUN concentration (P = 0.017) was a significant predictor of DTP when adjusting for mass
(Table 7.27). Other evaluated covariates were not significant predictors and did not cause
substantial confounding (<15%).
56
© University of Pretoria
Table 7.27: The combined effects of BUN concentration and body mass on DTP in Herd E
(multivariable survival analysis)
Variable
B
HR
95% CI of HR P value
Lower Upper
BUN
-0.290 0.748 0.589 0.950
0.017
Mass
0.023 1.023 1.014 1.032 <0.001
Sample day 1.221 3.390 1.542 7.452
0.002
B = beta; HR = hazard ratio and CI = confidence interval
7.6.4 Correlation analysis
Age and body mass were positively correlated (P < 0.001) in the groups that were sampled
on both days (Figure 7.12 and Figure 7.13). Age and BCS were negatively correlated in the
group of heifers that were sampled on the first (P = 0.015) and second days (P = 0.012). No
other variables were significantly correlated (P > 0.05).
Figure 7.12: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd E day 1
57
© University of Pretoria
Figure 7.13: Pearson’s correlation for BUN concentration with age and body mass, and age
with body mass, and Spearman’s correlation for BCS to age, body mass and BUN
concentration, in Herd E day 2
58
© University of Pretoria
8 Discussion
8.1 Introduction
The hypothesis of the study was that BUN concentration or its interaction with nitrogen
supplementation would have a significant effect on the reproductive performance of
Bonsmara heifers. The correlations between BUN concentration, body mass, age, BCS and
RTS in these heifers were also estimated. Potential classification bias was controlled by
blinding farmers to RTS, BCS, and BUN concentration data, and adjusting for herd and
evaluating other potential confounders in the statistical models.
As a scientific study, this prospective cohort study offered the advantage of being able to
demonstrate an appropriate temporal sequence between BUN concentration and
reproductive outcome. Heifers were raised under typical South African commercial beef
cow-calf enterprise conditions. The researcher had no control over farming practices
including the sale of animals, provision of dietary supplements and the use of reproductive
technologies such as synchronisation and artificial insemination that were potential sources
of bias. Despite this limitation, the results from this study have an important practical
application for the evaluation of the hypothesis because they were obtained from real,
commercially viable operations.
Comparison of the descriptive statistics of different herds revealed that herd origin was an
important determinant of the measured variables as well as the reproductive performance
(Table 7.7; Table 7.8; Table 7.11; Table 7.13; Table 7.15; Table 7.16; Table 7.20; Table 7.21;
Table 7.22; Table 7.23). The observed between-herd differences in these parameters clearly
indicate that differences in environmental factors, including herd management, had an
important role to play in reproductive performance. A large number of factors, both known
and unknown could have caused the differences in reproductive performance between
herds (Section 2.8, page 12); hence, the statistical significance of the observed differences
between herds was not tested.
59
© University of Pretoria
Sampling from different herds could have contributed towards the observed between-herd
differences in BUN concentration because sampling was performed on different days, at
different times of the day, and under different weather conditions. The effects of these
conditions on BUN concentration were reviewed (Section 2.8, page 12). Stratifying data by
herd and sampling day was performed to control for the variation due to different herds in
this study.
Most authors have advocated the use of pregnancy proportion and DTC as the most
appropriate measure of reproductive performance (Meyer et al., 1990; MacGregor and
Casey, 1999; Eler et al., 2002). In this study, DTP was used instead of DTC in order to exclude
abortions and variations in gestational length (Andersen and Plum, 1965; Foote, 1981;
Norman et al., 2009), which are not related to BUN concentration. Using DTP also reduced
the necessary follow up for the animals. The longer follow up period that is associated with
DTC would have inevitably increased the loss of study animals, considering that 31 animals
were lost within the first five months of the study period.
In a restricted breeding season, DTP is a better measure of reproductive performance than
pregnancy proportion because if the breeding season had been long enough, most heifers
would eventually become pregnant and pregnancy proportion alone would not differentiate
between animals with good and poor reproductive performance.
Assuming that cycling occurs randomly in heifers, except in herds where oestrous
synchronisation is practised, it follows that heifers were at random stages of their oestrus
cycles when the breeding season commenced. Due to this, those heifers that were at
oestrus were likely to become pregnant sooner (hence a shorter DTP) than those that were
at dioestrous. The major drawback of using DTP as a parameter for measuring reproductive
performance is that it makes those heifers that became pregnant earlier, to appear as if
they were more fertile than those that became pregnant later during the first 21-day period.
However, because of the large sample size, it was assumed that this random bias did not
affect the usefulness of DTP.
60
© University of Pretoria
The mean BUN concentration ± SD of 5.27 mmolL-1 ± 1.80 (14.76 mg/dl ± 5.04) obtained in
this study was higher and more variable than the 3.17 mmolL-1 ± 0.11 in 8-month old
Bonsmara steers raised on sweetveld, reported by Ndlovu et al. (2009). The difference in
reported BUN concentration was most likely because steers in Ndlovu et al.’s (2009) study
were raised on pasture and received no nitrogen supplementation. Previous studies have
demonstrated a direct relationship between urea nitrogen concentration and the amount of
dietary CP (Roseler et al., 1993; Chalupa and Sniffen, 1996; Schroeder and Titgemeyer,
2008).
BUN concentration showed a significant negative correlation with age (P < 0.001) and body
mass (P < 0.001) in the combined data set, although the correlation was not present within
all herds. The correlation with age is in agreement with findings of recent work, which
showed that BUN concentration decreases with age (Doska et al., 2012) in cows but is
contrary to findings from earlier work (Oltner et al., 1985). It is logical to assume that the
correlation between BUN concentration and age is a weak one hence the conflicting results
reported both in literature and in this study. BUN concentration did not show any significant
correlations with BCS in the beef heifers of this study. In a study of lactating dairy cows,
animals in a lower BCS were likely to have higher BUN concentrations (Ward, 2000; Guo et
al., 2004; Tamminga, 2006), most likely because mature cows calving in a poor body
condition have less adipose tissue available for milk production, leading to protein
mobilisation to support gluconeogenesis.
It was expected that those heifers that were sampled on hot days with no access to drinking
water would get dehydrated causing an increase in BUN concentration with increasing
sampling order (Weeth and Lesperance, 1965; Burgos et al., 2001). However, this study
found no association between BUN concentration and the sampling order except in Herd D
where the heifers were managed as two separate groups. It is likely that the apparent linear
trend between BUN concentration and sampling order was caused by differences in the way
the two groups were managed (Figure 7.8).
61
© University of Pretoria
8.2 Herd A
The mean BUN concentration ± SD of 5.37 mmolL-1 ± 0.81 obtained from this herd was
higher than the 2.20 mmolL-1 ± 0.67 obtained in Herd B, which practised low levels of
nitrogen supplementation. The reason for the high concentration cannot be explained by
the level of dietary nitrogen supplementation at the time of sampling alone, because this
herd was classified as having no supplementation of nitrogen (Section 6.1, page 19). Other
farm specific factors likely caused the high BUN concentration. It is also thought that the
energy lick that was supplemented in Herd B probably helped to reduce BUN
concentrations, as reviewed in the literature section (Section 2.1, page 3).
The pregnancy proportion (71.3%) in this herd was lower than the benchmark for beef cow calf operations (> 90% pregnancy rate) with a 62-day breeding season (Chenoweth and
Sanderson, 2001). The reason for the lower pregnancy proportion in this herd is not clear,
but is thought to be due to some heifers being immature at the onset of the breeding
season and probably not cycling, considering that the age range at breeding was 14 to 24
months with a median breeding age of 18.60 months. The high number of DTP in this herd
further strengthens the suspicion of immaturity at the onset of the breeding season.
This would also explain why RTS was the only significant predictor of pregnancy status and
DTP, whilst age was close to significance in predicting DTP. The positive correlation between
age and RTS also indicates that maturity of the reproductive tract was the main determinant
of reproductive performance. This is in agreement with Holm et al. (2009), who observed
that RTS was associated more strongly with age than with the body weight of the heifer.
However, it is important to note that only the month of birth was known and not the exact
birth dates of the heifers. Using the 15th of the month as the birth date led to either
overestimating or underestimating the age of heifers.
62
© University of Pretoria
8.3 Herd B
The mean BUN concentration ± SD of 2.20 mmolL-1 ± 0.67 obtained in this herd was lower
than that of the first herd. This low BUN concentration could be due to heifer management.
It is logical to assume that the production lick supplement that was provided in this herd
resulted in lower BUN concentration by supplying the rumen microbes with fermentable
carbohydrates leading to less ammonia and urea production (Ørskov, 1994; Chalupa and
Sniffen, 1996; Schroeder and Titgemeyer, 2008).
The 100% pregnancy proportion obtained in this herd was probably because all the heifers
were mature and cycling before the onset of the breeding season. The suspicion that most
heifers were already cycling before the onset of the breeding season is strengthened by the
low median DTP of 24 days. This suggests that most heifers became pregnant within the first
oestrus cycle after the onset of the breeding season.
Since all the heifers became pregnant, it was not possible to perform logistic regression.
Multivariable survival analysis was also not possible because none of the measured
variables was a significant predictor of DTP. In this herd, as anticipated, there was a
significant positive correlation between body mass and age because healthy growing heifers
are expected to gain weight.
8.4 Herd C
The mean BUN concentration ± SD of 4.14 mmolL-1 ± 0.92 obtained from this herd is higher
than the 2.20 mmolL-1 ± 0.67 seen in Herd B. This may be partly due to the moderate level
of dietary nitrogen supplementation practised in Herd C. This finding is in agreement with
that of other workers who found that increasing the levels of dietary CP leads to higher BUN
concentration (Roseler et al., 1993; Chalupa and Sniffen, 1996; Schroeder and Titgemeyer,
2008).
Similar to Herd B, this herd had a 100% pregnancy proportion. It is believed that this was
because all the heifers were mature and cycling before the onset of the breeding season.
63
© University of Pretoria
This was confirmed by the median RTS of 5 obtained in this herd. It is likely that the heifers
with a RTS of 3 were actually above 3 but they were scored down because they were not
distinguishable from a score of 4 because their corpora lutea were not discernible by
palpation. The suspicion that most heifers were already cycling before the onset of breeding
is further strengthened by the fact that 22 out of 34 (64.7%) heifers became pregnant after
the first synchronised artificial insemination. However, it is important to note that oestrus
synchronisation and artificial insemination is known to affect both the pregnancy proportion
and DTP (Xu and Burton, 1999). Although it seemed proper to exclude this herd from the
study, because of its different management style, it was decided to retain it in order for the
study to be relevant to commercial operations where the choice of management practices is
driven by economics. Herd effects were controlled by adjusting for herd as the stratifying
factor in the statistical models and thus prevented confounding.
Similar to Herd B, it was not possible to perform logistic regression because all heifers
became pregnant. Multivariable survival analysis was also not possible because none of the
measured variables was a significant predictor of DTP. As anticipated, there was a significant
positive correlation between body mass and age because healthy growing heifers are
expected to gain weight. BCS did not correlate with any other variable because there was no
variation in BCS data in this herd.
8.5 Herd D
The mean BUN concentration ± SD of 4.50 mmolL-1 ± 1.80 obtained from this herd was
higher than that of Herd B. This is thought to be at least partly due to the moderate level of
dietary nitrogen supplementation practised in this herd (Section 8.4, page 63). The two
subgroups in this herd seem to be different (Figure 7.8), and this caused the overall mean
not to be a good representation of the entire herd. It was anticipated that the larger group
with a higher mean would determine the overall effect of BUN concentration on
reproductive performance.
The pregnancy proportion (68.2%) was lower than the industry benchmark of more than
90% (Chenoweth Radostis 2001). It is thought that the high BUN concentration played a
64
© University of Pretoria
significant role in determining the pregnancy proportion considering that the heifers in this
herd were mature, as indicated by their age and RTS (Table 7.15). However, the BCS of these
heifers were low compared to other herds, suggesting an energy deficiency. Other factors
that might have played important roles in determining the overall pregnancy proportion
were the sale of some study animals and the subgrouping that occurred in this herd. It is
assumed that the animals in good condition (most likely to be pregnant) are the ones that
were sold. This herd could have been excluded from the study because of the unexpected
changes in the management of the heifers but the researcher decided to retain the herd and
adjust for herd in the statistical models.
BUN concentration was the only predictor of pregnancy status and DTP in this herd. It is
assumed that this was because all the heifers were mature, and the other measured
variables ceased to be important in determining the reproductive outcome.
8.6 Herd E
In this herd, the mean BUN concentration ± SD for the heifers that were sampled on the
second day was significantly lower (P < 0.001) than for those sampled on the first day (Table
7.20; Table 7.21). The reason for this is not clear but it is thought that this was due to
reduced DMI of the group that was sampled on the second day after spending the previous
day in the holding pen. Sampling day was forced into all regression models to account for
potential confounding. The mean BUN concentration of 6.23 mmolL-1 ± 1.74 obtained from
this herd was higher than that obtained in Herd B. This is thought to be due to the high level
of dietary nitrogen supplementation practised in this herd, as discussed for Herd C (Section
8.4, page 63).
This herd had a pregnancy proportion of 47.5% and median DTP of 44 days. BUN
concentration, age and body mass were significant predictors of both pregnancy status and
DTP. Age and body mass are thought to have been significant in this herd because some of
the heifers in this herd were young with low body mass (Table 7.20; Table 7.21). This
suggests that some had not attained puberty. Older heifers, and thus more mature, had a
higher chance of becoming pregnant and at an increased rate. This is in agreement with the
65
© University of Pretoria
work done by other researchers that have reported that heifers that mature and undergo a
few complete oestrous cycles before the breeding season tend to have better reproductive
performance (Yelich et al., 1996; Berry et al., 2003; Buckley et al., 2003; Whittier et al.,
2008).
The significant positive correlation between mass and age was anticipated. The negative
correlation that was seen on the second day of sampling between BUN concentration and
age is in agreement with Doska et al. (2012), who suggested that BUN concentration is
higher in young cows and tends to decrease with age. RTS was significantly correlated with
age (P = 0.004) and BCS (P < 0.001). This is consistent with the view that age, BCS and frame
size of the heifer has an influence on the RTS that is assigned (Rosenkrans and Hardin,
2003).
Although other studies (Patterson et al., 2000; Rosenkrans and Hardin, 2003; Holm et al.,
2009) have shown that increasing RTS predicts reproductive performance, the reason for
the failure of RTS to predict reproductive performance in this herd was likely due to a lack of
variability in RTS data in this herd. Regarding RTS as categorical data through
dichotomisation did not lead to different conclusions in the significance of its prediction of
reproductive performance.
8.7 Effect of BUN concentration on reproductive performance
BUN concentration was a significant independent predictor of DTP (Table 7.6) but was not
significant in predicting pregnancy status in all herds combined (Table 7.4). This suggests
that high BUN concentration before the onset of the breeding season negatively affects the
chances of becoming pregnant only for a short period. Another possibility is that some
animals adapted to the increased BUN concentration as suggested by Calsamiglia et al.
(2010), or their BUN levels decreased during the course of the breeding season. The latter
option is highly possible because RDP supplementation was stopped in all herds at the
beginning of the breeding season. The results of this study are in agreement with those of
Guo et al. (2004), who showed that in among-herd analyses, MUN concentration had
minimal effect on conception rate but was associated with greater days open. Ferguson et
66
© University of Pretoria
al. (1993) also showed that within dairy herds with mean MUN concentrations above 20
mg/dL, cows with higher MUN levels were associated with poorer conception rates at first
service, but not at subsequent services. The assumption that BUN concentration reduced
the chances of becoming pregnant for a short period after the onset of the breeding season
is consistent with the hypothesis that urea affects cleavage and blastocyst formation but not
necessarily the early development of the oocyte (Jorritsma et al., 2003).
This is in contrast with the theory, which states that the negative effect of high BUN
concentration is exerted through the exacerbation of an underlying NEB by the energy costs
of detoxifying large quantities of ammonia in post-partal dairy cattle (Staples et al., 1990;
Garcia-Bojalil et al., 1998; Overton et al., 1999). It is logical to assume that beef heifers will
not suffer severely from NEB like lactating dairy cattle. In this study, BCS was used as an
indicator of the animal’s energy reserves. This is widely supported in literature for both beef
and dairy cattle (Wildman et al., 1982; Edmonson et al., 1989; Houghton et al., 1990).
However, BCS has been shown to have lower accuracy in young growing cows because
growing animals tend to have less fat deposits (Nicholson and Butterworth, 1986). The
heifers in this study were in a good BCS with little variation in the data. The lack of variation
made it impossible to estimate the effect of BCS on reproductive performance. As
recommended by Ndlovu et al. (2007) in their review, measuring of NEFAs would have been
a more reliable way to assess the energy status of the heifers.
BUN concentration was only determined once prior to the onset of the breeding season.
However, it is likely that BUN concentration in the heifers continued to vary during the
breeding season because of dietary and environmental changes that occurred (Rodriguez et
al., 1997; Godden et al., 2001; Schroeder and Titgemeyer, 2008). Nevertheless, since the
environmental changes are likely to affect all the heifers in a herd in a similar way, and that
urea concentration is genetically determined (Mitchell et al., 2005; Hossein-Zadeh and
Ardalan, 2011), it is assumed that those heifers with a higher BUN concentration would
remain high relative to the rest of the herd throughout the breeding season.
67
© University of Pretoria
BUN concentration was a significant independent predictor of both DTP and pregnancy
status only in Herds D and E (Table 8.1). These herds practised moderate to high levels of
nitrogen supplementation in order to achieve early breeding.
Table 8.1: Summary of the effect of BUN concentration on pregnancy status and DTP in the
different herds
Age
Nitrogen
Range
supplement
level
Herd
n
(months)
All
338
12 - 31
A
106
14 - 24
B
33
C
BUN as
BUN as
predictor of
predictor
pregnancy
of DTP
Odds
BUN ± SD
PP (%) Ratio
Hazard
DTP
Ratio
5.27 ± 1.80
63.1 0.882*
40
0.827**
None
5.37 ± 0.81
71.3 0.833
43
0.969
20 - 26
Low
2.20 ± 0.67
100.0 N.A.
24
1.113
34
24 - 27
Moderate
4.14 ± 0.92
100.0 N.A.
4
0.997
D
22
22 - 31
Moderate
4.50 ± 1.80
68.2 0.478**
47
0.719**
E
143
12 - 23
High
6.23 ± 1.74
47.5 0.656**
44
0.736**
* = P - value < 0.1
** = P - value < 0.05
DTP = median DTP
PP = pregnancy proportion
BUN = mean BUN concentration
In this study, those heifers that were fed high levels of nitrogen supplementation had the
highest BUN concentrations and the lowest pregnancy proportion. Although the current
study was not designed to investigate the causal relationship between nitrogen
supplementation and BUN concentration, this finding is in agreement with the findings of
other workers who demonstrated that high dietary NPN levels lead to high BUN
concentration (Canfield et al., 1990; Kenny et al., 2002a). This is known to lead to a decrease
in reproductive performance of cattle (Rhoads et al., 2006).
68
© University of Pretoria
Variation in the genetic ability to recirculate nitrogen has been proposed, with animals with
improved nitrogen recirculating ability having higher BUN levels (Schoeman, 1989). Feeding
high levels of RDP is known to down regulate the efficiency of nitrogen recirculation (Marini
and van Amburgh, 2003; Marini et al., 2004), leading to higher BUN concentration and renal
loss. This suggests that those heifers with better abilities to recirculate nitrogen within herds
that were over supplied with RDP probably lost their advantage in the presence of an
oversupply of RDP. These heifers might have suffered more negative consequences from the
effects of high BUN concentrations. Since BUN concentration affected reproductive
performance only in herds where the mean BUN concentration was high (similar to findings
by Ferguson et al. (1993)), one could reason that the potential exists that heifers with an
improved ability to recirculate nitrogen within herds heavily supplemented with RDP are at
risk of being culled for poor fertility in a restricted breeding system.
8.8 Potential weaknesses of the study
Due to unavailability of herds that met the study criteria, only 369 heifers were enrolled for
the study instead of the calculated minimum sample size of 438. Of these, only 338 heifers
finished the study.
The hydration status of the heifers in the study was not recorded. It is well known that the
hydration status of ruminants has an influence on measured BUN concentration (Weeth and
Lesperance, 1965; Utley et al., 1970; Mousa et al., 1983; Aganga et al., 1989; Maloiy et al.,
2000). It was not practical to assess the hydration status of the heifers in this study as that
would have increased the time spent on data collection, leading to more variation in the
BUN concentration data (Gustafsson and Palmquist, 1993).
The CRL formula was accurate in estimating the stage of pregnancy when checked against
the 20 recorded mating dates. Although this was a small number, the accuracy of the
formula satisfied the researcher that it is useful for estimating the age of early pregnancies.
According to Riding et al. (2008), this formula is known to be most accurate between day 36
and 103 of gestation. In the current study, some animals fell outside this range because the
breeding season was longer than 67 days.
69
© University of Pretoria
Time spent collecting additional data during BUN sampling potentially exacerbated the
effect of sampling order in this study. There is evidence in the literature that MUN
concentration in dairy cows is associated with the time of sampling in relation to feeding
(Gustafsson and Palmquist, 1993), the hydration status and the DMI of the animal (Godden
et al., 2001). Since BUN concentration is also known to vary with season and diurnal
patterns, one should attempt to collect blood for BUN concentration analysis at the same
time of the day and make all known factors as similar as possible (Wattiaux et al., 2005).
70
© University of Pretoria
9 Conclusion
In this study, blood urea nitrogen (BUN) concentration at the start of breeding was
independently associated with reproductive performance of Bonsmara heifers, especially in
those herds where management included heavy supplementation of dietary protein to
achieve early breeding.
It is recommended that production systems designed to achieve early breeding in beef
heifers investigate whether oversupplying rumen degradable protein (RDP) selects against
animals with an improved ability to recirculate nitrogen.
71
© University of Pretoria
10 References
Aganga, A. A., N. N. Umunna, E. O. Oyedipe, and P. N. Okoh. 1989. Influence of water
restriction on some serum components in Yankasa ewes. Small Ruminant Research.
2(1): 19-26.
Alexander, J. H. 2008. Bull breeding soundness evaluation: A practitioner's perspective.
Theriogenology. 70(3): 469-472.
Andersen, H., and M. Plum. 1965. Gestation length and birth weight in cattle and buffaloes:
A review. Journal of Dairy Science. 48(9): 1224-1235.
Andersen, K. J., D. G. LeFever, J. S. Brinks, and K. G. Odde. 1991. The use of reproductive
tract scoring system in beef heifers. Agri-Practice. 12(4): 19-26.
Arunvipas, P., I. R. Dohoo, J. A. van Leeuwen, and G. P. Keefe. 2003. The effect of nonnutritional factors on milk urea nitrogen levels in dairy cows in Prince Edward Island,
Canada. Preventative Veterinary Medicine. 59(1–2): 83-93.
Arunvipas, P., J. A. van Leeuwen, I. R. Dohoo, E. R. Leger, G. P. Keefe, A. S. Burton et al.
2007. Milk urea-nitrogen negatively affected first-service breeding success in
commercial dairy cows in Prince Edward Island, Canada. Preventative Veterinary
Medicine. 82(1–2): 42-50.
Babashahi, M., G. R. Ghorbani, and H. R. Rahmani. 2004. Relationship between blood and
milk urea nitrogen and fertility in dairy cows. Journal of Science and Technology of
Agriculture and Natural Resources. 8(3): 171-179.
Barton, B. A., H. A. Rosario, G. W. Anderson, B. P. Grindle, and D. J. Carroll. 1996. Effects
of dietary crude protein, breed, parity and health status on the fertility of dairy cows.
Journal of Dairy Science. 79(12): 2225-2236.
Béline, F., J. Martinez, C. Marol, and G. Guiraud. 1998. Nitrogen transformations during
anaerobically stored 15N-labelled pig slurry. Bioresource Technology. 64(2): 83-88.
72
© University of Pretoria
Berry, D. P., F. Buckley, P. Dillon, R. D. Evans, M. Rath, and R. F. Veerkamp. 2003. Genetic
relationships among body condition score, body weight, milk yield, and fertility in dairy
cows. Journal of Dairy Science. 86(6): 2193-2204.
Bierman, S., G. E. Erickson, T. J. Klopfenstein, R. A. Stock, and D. H. Shain. 1999.
Evaluation of nitrogen and organic matter balance in the feedlot as affected by level and
source of dietary fiber. Journal of Animal Science. 77(7): 1645-1653.
Bourdon, R. M., and J. S. Brinks. 1983. Calving date versus calving interval as a reproductive
measure in beef cattle. Journal of Animal Science. 57(6): 1412-1417.
Bouwman, A. C., G. C. B. Schopen, H. Bovenhuis, M. H. P. W. Visker, and J. A. M. van
Arendonk. 2010. Genome-wide scan to detect quantitative trait loci for milk urea
nitrogen in Dutch Holstein-Friesian cows. Journal of Dairy Science. 93(7): 3310-3319.
Bristow, A. W., D. C. Whitehead, and J. E. Cockburn. 1992. Nitrogenous constituents in the
urine of cattle, sheep and goats. Journal of the Science of Food and Agriculture. 59(3):
387-394.
Broderick, G. A., and M. K. Clayton. 1997. A statistical evaluation of animal and nutritional
factors influencing concentrations of milk urea nitrogen. Journal of Dairy Science.
80(11): 2964-2971.
Buckley, F., K. O'Sullivan, J. F. Mee, R. D. Evans, and P. Dillon. 2003. Relationships among
milk yield, body condition, cow weight, and reproduction in spring-calved HolsteinFriesian. Journal of Dairy Science. 86(7): 2308-2319.
Buddenberg, B. J., C. J. Brown, and A. H. Brown. 1990. Heritability estimates of calving date
in Hereford cattle maintained on range under natural mating. Journal of Animal Science.
68(1): 70-74.
Burgos, M. S., M. Senn, F. Sutter, M. Kreuzer, and W. Langhans. 2001. Effect of water
restriction on feeding and metabolism in dairy cows. American Journal of Physiology Regulatory, Integrative and Comparative Physiology. 280(2): R418-R427.
73
© University of Pretoria
Butler, W. R., J. J. Calaman, and S. W. Beam. 1996. Plasma and milk urea nitrogen in
relation to pregnancy rate in lactating dairy cattle. Journal of Animal Science. 74(4):
858-865.
Butler, W. R. 1998. Review: Effect of protein nutrition on ovarian and uterine physiology in
dairy cattle. Journal of Dairy Science. 81(9): 2533-2539.
Calsamiglia, S., A. Ferret, C. K. Reynolds, N. B. Kristensen, and A. M. v. Vuuren. 2010.
Strategies for optimizing nitrogen use by ruminants. Animal. 4(7): 1184-1196.
Cammack, K. M., M. G. Thomas, and R. M. Enns. 2009. Reproductive traits and their
heritabilities in beef cattle. Professional Animal Scientist. 25(5): 517-528.
Canfield, R. W., C. J. Sniffen, and W. R. Butler. 1990. Effects of excess degradable protein
on postpartum reproduction and energy balance in dairy cattle. Journal of Dairy Science.
73(9): 2342-2349.
Carlsson, J., and B. Pehrson. 1993. The relationships between seasonal variations in the
concentration of urea in bulk milk and the production and fertility of dairy herds. Journal
of Veterinary Medicine Series A. 40(1-10): 205-212.
Casper, D. P., D. J. Schingoethe, M. J. Brouk, and H. A. Maiga. 1994. Non-structural
carbohydrate and undegradable protein sources in the diet: Growth responses of dairy
heifers. Journal of Dairy Science. 77(9): 2595-2604.
Chalupa, W., and C. J. Sniffen. 1996. Protein and amino acid nutrition of lactating dairy
cattle—today and tomorrow. Animal Feed Science and Technology. 58(1-2): 65-75.
Chenoweth, P. J., and M. W. Sanderson. 2001. Health and Production Management in Beef
Cattle Breeding Herds. Pages 530-531 in Herd Health Food Animal Production
Medicine. Duncan, L. L., ed. 3rd ed. W.B. Saunders Company, New York, London.
Demeyer, D. I. 1981. Rumen microbes and digestion of plant cell walls. Agriculture and
Environment. 6(2–3): 295-337.
74
© University of Pretoria
Dijkstra, J., O. Oenema, and A. Bannink. 2011. Dietary strategies to reducing N excretion
from cattle: Implications for methane emissions. Current Opinion in Environmental
Sustainability. 3(5): 414-422.
Doska, M. C., Silva, Delma Fabíola Ferreira da, J. A. Horst, A. A. Valloto, P. Rossi Junior,
and R. d. Almeida. 2012. Sources of variation in milk urea nitrogen in Paraná dairy
cows. Revista Brasileira De Zootecnia. 41(3): 692-697.
Edmonson, A. J., I. J. Lean, L. D. Weaver, T. Farver, and G. Webster. 1989. A body
condition scoring chart for Holstein dairy cows. Journal of Dairy Science. 72(1): 68-78.
Eicher, R., E. Bouchard, and M. Bigras-Poulin. 1999. Factors affecting milk urea nitrogen
and protein concentrations in Québec dairy cows. Preventative Veterinary Medicine.
39(1): 53-63.
Eler, J. P., J. A. I. I. V. Silva, J. B. S. Ferraz, F. Dias, H. N. Oliveira, J. L. Evans et al. 2002.
Genetic evaluation of the probability of pregnancy at 14 months for Nellore heifers.
Journal of Animal Science. 80(4): 951-954.
Elrod, C. C., and W. R. Butler. 1993. Reduction of fertility and alteration of uterine pH in
heifers fed excess ruminally degradable protein. Journal of Animal Science. 71(3): 694701.
Erickson, G., and T. Klopfenstein. 2010. Nutritional and management methods to decrease
nitrogen losses from beef feedlots. Journal of Animal Science. 88(13): E172-E180.
Evans, J. L., B. L. Golden, R. M. Bourdon, and K. L. Long. 1999. Additive genetic
relationships between heifer pregnancy and scrotal circumference in Hereford cattle.
Journal of Animal Science. 77(10): 2621-2628.
Ferguson, J. D., D. T. Galligan, T. Blanchard, and M. Reeves. 1993. Serum urea nitrogen and
conception rate: The usefulness of test information. Journal of Dairy Science. 76(12):
3742-3746.
75
© University of Pretoria
Fondevila, M., and B. A. Dehority. 1994. Degradation and utilization of forage hemicellulose
by rumen bacteria, singly in coculture or added sequentially. Journal of Applied
Microbiology. 77(5): 541-548.
Foote, R. H. 1981. Factors affecting gestation length in dairy cattle. Theriogenology. 15(6):
553-559.
Fosgate, G. T. 2009. Practical sample size calculations for surveillance and diagnostic
investigations. Journal of Veterinary Diagnostic Investigation. 21(1): 3-14.
Frank, B., and C. Swensson. 2002. Relationship between content of crude protein in rations
for dairy cows and milk yield, concentration of urea in milk and ammonia emissions.
Journal of Dairy Science. 85(7): 1829-1838.
Garcia-Bojalil, C. M., C. R. Staples, C. A. Risco, J. D. Savio, and W. W. Thatcher. 1998.
Protein degradability and calcium salts of long-chain fatty acids in the diets of lactating
dairy cows: Productive responses. Journal of Dairy Science. 81(5): 1374-1384.
Garcia-Bojalil, C. M., C. R. Staples, W. W. Thatcher, and M. Drost. 1994. Protein intake and
development of ovarian follicles and embryos of superovulated nonlactating dairy cows.
Journal of Dairy Science. 77(9): 2537-2548.
Godden, S. M., K. D. Lissemore, D. F. Kelton, K. E. Leslie, J. S. Walton, and J. H. Lumsden.
2001. Factors associated with milk urea concentrations in Ontario dairy cows. Journal of
Dairy Science. 84(1): 107-114.
Gonda, H. L., and J. E. Lindberg. 1994. Evaluation of dietary nitrogen utilization in dairy
cows based on urea concentrations in blood, urine and milk, and on urinary
concentration of purine derivatives. Acta Agricultura Scandinavica. Section A, Animal
Science. 44(4): 236-245.
Guo, K., E. Russek-Cohen, M. A.Varner, and R. A. Kohn. 2004. Effects of milk urea
nitrogen and other factors on probability of conception of dairy cows. Journal of Dairy
Science. 87(6): 1878-1885.
76
© University of Pretoria
Gustafsson, A. H., and D. L. Palmquist. 1993. Diurnal variation of rumen ammonia, serum
urea, and milk urea in dairy cows at high and low yields. Journal of Dairy Science.
76(2): 475-484.
Hall, J. B., R. B. Staigmiller, R. E. Short, R. A. Bellows, M. D. MacNeil, and S. E. Bellows.
1997. Effect of age and pattern of gain on induction of puberty with a progestin in beef
heifers. Journal of Animal Science. 75(6): 1606-1611.
Harmeyer, J., and H. Martens. 1980. Aspects of urea metabolism in ruminants with reference
to the goat. Journal of Dairy Science. 63(10): 1707-1728.
Harris, B., J. 1996. Using Milk Urea Nitrogen and Blood Urea Values as Management Tools.
Pages 75-81 Alltech UK, Stamford.
Holm, D. E., P. N. Thompson, and P. C. Irons. 2009. The value of reproductive tract scoring
as a predictor of fertility and production outcomes in beef heifers. Journal of Animal
Science. 87(6): 1934-1940.
Hoover, W. H., and S. R. Stokes. 1991. Balancing carbohydrates and proteins for optimum
rumen microbial yield. Journal of Dairy Science. 74(10): 3630-3644.
Hossein-Zadeh, N. G., and M. Ardalan. 2011. Estimation of genetic parameters for milk urea
nitrogen and its relationship with milk constituents in Iranian Holsteins. Livestock
Science. 135(2–3): 274-281.
Houghton, P. L., R. P. Lemenager, G. E. Moss, and K. S. Hendrix. 1990. Prediction of
postpartum beef cow body composition using weight to height ratio and visual body
condition score. Journal of Animal Science. 68(5): 1428-1437.
Ipharraguerre, I. R., J. H. Clark, and D. E. Freeman. 2005. Varying protein and starch in the
diet of dairy cows. I. effects on ruminal fermentation and intestinal supply of nutrients.
Journal of Dairy Science. 88(7): 2537-2555.
Irons, P. C., J. O. Nöthling, and H. J. Bertschinger. 2007. Bull breeding soundness evaluation
in southern Africa. Theriogenology. 68(6): 842-847.
77
© University of Pretoria
Jonker, J. S., R. A. Kohn, and R. A. Erdman. 1999. Milk urea nitrogen target concentrations
for lactating dairy cows fed according to national research council recommendations.
Journal of Dairy Science. 82(6): 1261-1273.
Jordan, E. R., and L. V. Swanson. 1979. Effect of crude protein on reproductive efficiency,
serum total protein, and albumin in the high-producing dairy cow. Journal of Dairy
Science. 62(1): 58-63.
Jordan, E. R., T. E. Chapman, D. W. Holtan, and L. V. Swanson. 1983. Relationship of
dietary crude protein to composition of uterine secretions and blood in high-producing
postpartum dairy cows. Journal of Dairy Science. 66(9): 1854-1862.
Jorritsma, R., T. Wensing, A. M. K. Theo, L. A. M. V. Peter, and P. T. M. N. Jos. 2003.
Metabolic changes in early lactation and impaired reproductive performance in dairy
cows. Veterinary Research. 34(1): 11-26.
Kauffman, A. J., and N. R. St-Pierre. 2001. The relationship of milk urea nitrogen to urine
nitrogen excretion in Holstein and Jersey cows. Journal of Dairy Science. 84(10): 22842294.
Kenny, D. A., M. P. Boland, M. G. Diskin, and J. M. Sreenan. 2002a. Effect of rumen
degradable protein with or without fermentable carbohydrate supplementation on blood
metabolites and embryo survival in cattle. Animal Science. 74(3): 529-537.
Kenny, D. A., P. G. Humpherson, H. J. Leese, D. G. Morris, A. D. Tomos, M. G. Diskin et al.
2002b. Effect of elevated systemic concentrations of ammonia and urea on the
metabolite and ionic composition of oviductal fluid in cattle. Biology of Reproduction.
66(6): 1797-1804.
König, S., Y. M. Chang, U. U. Borstel, D. Gianola, and H. Simianer. 2008. Genetic and
phenotypic relationships among milk urea nitrogen, fertility, and milk yield in Holstein
cows. Journal of Dairy Science. 91(11): 4372-4382.
78
© University of Pretoria
Kristensen, N. B., A. C. Storm, and M. Larsen. 2010. Effect of dietary nitrogen content and
intravenous urea infusion on ruminal and portal-drained visceral extraction of arterial
urea in lactating Holstein cows. Journal of Dairy Science. 93(6): 2670-2683.
Lapierre, H., and G. E. Lobley. 2001. Nitrogen recycling in the ruminant: A review. Journal
of Dairy Science. 84: 223-E236.
Larson, S. F., W. R. Butler, and W. B. Currie. 1997. Reduced fertility associated with low
progesterone postbreeding and increased milk urea nitrogen in lactating cows. Journal of
Dairy Science. 80(7): 1288-1295.
Leedle, J. A. Z., K. Barsuhn, and R. B. Hespell. 1986. Postprandial trends in estimated
ruminal digesta polysaccharides and their relation to changes in bacterial groups and
ruminal fluid characteristics. Journal of Animal Science. 62(3): 789-803.
Leroy, J., A. Van Soom, G. Opsomer, I. Goovaerts, and P. Bols. 2008. Reduced fertility in
high-yielding dairy cows: Are the oocyte and embryo in danger? Part II mechanisms
linking nutrition and reduced oocyte and embryo quality in high-yielding dairy cows*.
Reproduction in Domestic Animals. 43(5): 623-632.
MacGregor, R. G., and N. H. Casey. 1999. Evaluation of calving interval and calving date as
measures of reproductive performance in a beef herd. Livestock Production Science.
57(2): 181-191.
Maloiy, G. M. O., B. M. Rugangazi, and E. T. Clemens. 2000. Nitrogen metabolism and
renal function in the dik-dik antelope (Rhynchotragus kirkii). Small Ruminant Research.
37(3): 243-248.
Marini, J. C., J. D. Klein, J. M. Sands, and M. E. van Amburgh. 2004. Effect of nitrogen
intake on nitrogen recycling and urea transporter abundance in lambs. Journal of Animal
Science. 82(4): 1157-1164.
Marini, J. C., and M. E. van Amburgh. 2005. Partition of nitrogen excretion in urine and the
feces of Holstein replacement heifers. Journal of Dairy Science. 88(5): 1778-1784.
79
© University of Pretoria
Marini, J. C., and M. E. van Amburgh. 2003. Nitrogen metabolism and recycling in Holstein
heifers. Journal of Animal Science. 81(2): 545-552.
McCrory, D. F., and P. J. Hobbs. 2001. Additives to reduce ammonia and odor emissions
from livestock wastes: A review. Journal of Environmental Quality. 30(2): 345-355.
McDonald, P., R. A. Edwards, J. F. D. Greenhalgh, and C. A. Morgan. 1995. Animal
nutrition. 5th ed. Essex, England, Longman Scientific and Technical.
Meyer, K., K. Hammond, P. F. Parnell, M. J. MacKinnon, and S. Sivarajasingam. 1990.
Estimates of heritability and repeatability for reproductive traits in Australian beef cattle.
Livestock Production Science. 25(1-2): 15-30.
Miettinen, P. V. A., and R. O. Juvonen. 1990. Diurnal variations of serum and milk urea
levels in dairy cows. Acta Agricultura Scandinavica. 40(3): 289-296.
Mitchell, R. G., G. W. Rogers, C. D. Dechow, J. E. Vallimont, J. B. Cooper, U. SanderNielsen et al. 2005. Milk urea nitrogen concentration: Heritability and genetic
correlations with reproductive performance and disease. Journal of Dairy Science.
88(12): 4434-4440.
Mousa, H. M., K. E. Ali, and I. D. Hume. 1983. Effects of water deprivation on urea
metabolism in camels, desert sheep and desert goats fed dry desert grass. Comparative
Biochemistry and Physiology Part A: Physiology. 74(3): 715-720.
Ndegwa, P. M., A. N. Hristov, J. Arogo, and R. E. Sheffield. 2008. A review of ammonia
emission mitigation techniques for concentrated animal feeding operations. Biosystems
Engineering. 100(4): 453-469.
Ndlovu, T., M. Chimonyo, A. I. Okoh, V. Muchenje, K. Dzama, S. Dube et al. 2009. A
comparison of nutritionally related blood metabolites among Nguni, Bonsmara and
Angus steers raised on sweetveld. The Veterinary Journal. 179(2): 273-281.
80
© University of Pretoria
Ndlovu, T., M. Chimonyo, A. I. Okoh, V. Muchenje, K. Dzama, and J. G. Raats. 2007.
Assessing the nutritional status of beef cattle: Current practices and future prospects.
African Journal of Biotechnology. 6(24): 2727-2734.
Nicholson, M. J., and M. H. Butterworth. 1986. A guide to condition scoring of zebu cattle.
1st ed. Addis Ababa, Ethiopia, International Livestock Centre for Africa.
Norman, H. D., J. R. Wright, M. T. Kuhn, S. M. Hubbard, J. B. Cole, and P. M. van Raden.
2009. Genetic and environmental factors that affect gestation length in dairy cattle.
Journal of Dairy Science. 92(5): 2259-2269.
Nousiainen, J., K. J. Shingfield, and P. Huhtanen. 2004. Evaluation of milk urea nitrogen as a
diagnostic of protein feeding. Journal of Dairy Science. 87(2): 386-398.
Oba, M., R. L. Baldwin VI, S. L. Owens, and B. J. Bequette. 2004. Urea synthesis by ruminal
epithelial and duodenal mucosal cells from growing sheep. Journal of Dairy Science.
87(6): 1803-1805.
Ocon, O. M., and P. J. Hansen. 2003. Disruption of bovine oocytes and preimplantation
embryos by urea and acidic pH. Journal of Dairy Science. 86(4): 1194-1200.
Oltner, R., M. Emanuelson, and H. Wiktorsson. 1985. Urea concentrations in milk in relation
to milk yield, live weight, lactation number and amount and composition of feed given to
dairy cows. Livestock Production Science. 12(1): 47-57.
Ørskov, E. R. 1994. Recent advances in understanding of microbial transformation in
ruminants. Livestock Production Science. 39(1): 53-60.
Overton, T. R., J. K. Drackley, C. J. Ottemann-Abbamonte, A. D. Beaulieu, L. S. Emmert,
and J. H. Clark. 1999. Substrate utilization for hepatic gluconeogenesis is altered by
increased glucose demand in ruminants. Journal of Animal Science. 77(7): 1940-1951.
Patodkar, V. R., A. P. Somkuwar, S. Parekar, and N. Khade. 2010. Influence of sex on certain
biochemical parameters in nomadic camels (Camelus dromedarius) nearby Pune, in
Maharashtra. Veterinary World. 3(3): 115-117.
81
© University of Pretoria
Patterson, D. J., S. L. Wood, and R. F. Randle. 2000. Procedures that support reproductive
management of replacement beef heifers. Journal of Animal Science. 77(E-Supplement):
1-15.
Pehrson, B., K. P. Forshell, and J. Carlsson. 1992. The effect of additional feeding on the
fertility of high-yielding dairy cows. Journal of Veterinary Medicine Series A. 39(1-10):
187-192.
Pineda, M. H., and M. P. Dooley. 2003. McDonald's veterinary endocrinology and
reproduction. 5th ed. Ames, Iowa State Press.
Powell, J. M., G. A. Broderick, and T. H. Misselbrook. 2008. Seasonal diet affects ammonia
emissions from tie-stall dairy barns. Journal of Dairy Science. 91(2): 857-869.
Powell, J. M., and M. Russelle P. 2009. Dairy heifer management impacts manure N
collection and cycling through crops in Wisconsin, USA. Agriculture, Ecosystems &
Environment. 131(3–4): 170-177.
Rajala-Schultz, P. J., and W. J. A. Saville. 2003. Sources of variation in milk urea nitrogen in
Ohio dairy herds. Journal of Dairy Science. 86(5): 1653-1661.
Rajala-Schultz, P. J., W. J. A. Saville, G. S. Frazer, and T. E. Wittum. 2001. Association
between milk urea nitrogen and fertility in Ohio dairy cows. Journal of Dairy Science.
84(2): 482-489.
Rhoads, M. L., R. P. Rhoads, R. O. Gilbert, R. Toole, and W. R. Butler. 2006. Detrimental
effects of high plasma urea nitrogen levels on viability of embryos from lactating dairy
cows. Animal Reproduction Science. 91(1–2): 1-10.
Riding, G. A., S. A. Lehnert, A. J. French, and J. R. Hill. 2008. Conceptus-related
measurements during the first trimester of bovine pregnancy. The Veterinary Journal.
175(2): 266-272.
82
© University of Pretoria
Rodriguez, L. A., C. C. Stallings, J. H. Herbein, and M. L. Mcgilliard. 1997. Diurnal
variation in milk and plasma urea nitrogen in Holstein and Jersey cows in response to
degradable dietary protein and added fat. Journal of Dairy Science. 80(12): 3368-3376.
Røjen, B. A., P. K. Theil, and N. B. Kristensen. 2011. Effects of nitrogen supply on interorgan fluxes of urea-N and renal urea-N kinetics in lactating Holstein cows. Journal of
Dairy Science. 94(5): 2532-2544.
Romano, J. E., J. A. Thompson, D. W. Forrest, M. E. Westhusin, M. A. Tomaszweski, and D.
C. Kraemer. 2006. Early pregnancy diagnosis by transrectal ultrasonography in dairy
cattle. Theriogenology. 66(4): 1034-1041.
Roseler, D. K., J. D. Ferguson, and C. J. Sniffen. 1990. The effects of dietary protein
degradability/undegradability on milk urea, milk NPN and blood urea in lactating dairy
cows.73: 168.
Roseler, D. K., J. D. Ferguson, C. J. Sniffen, and J. Herrema. 1993. Dietary protein
degradability effects on plasma and milk urea nitrogen and milk nonprotein nitrogen in
Holstein cows. Journal of Dairy Science. 76(2): 525-534.
Rosenkrans, K. S., and D. K. Hardin. 2003. Repeatability and accuracy of reproductive tract
scoring to determine pubertal status in beef heifers. Theriogenology. 59(5–6): 10871092.
Russell, J. B., J. D. O'Connor, D. G. Fox, P. J. Van Soest, and C. J. Sniffen. 1992. A net
carbohydrate and protein system for evaluating cattle diets: I. ruminal fermentation.
Journal of Animal Science. 70(11): 3551-3561.
Schoeman, S. J. 1989. Recent research into the production potential of indigenous cattle with
special reference to the Sanga. South African Journal of Animal Science. 19(2): 55-61.
Schroeder, G. F., and E. C. Titgemeyer. 2008. Interaction between protein and energy supply
on protein utilization in growing cattle: A review. Livestock Science. 114(1): 1-10.
83
© University of Pretoria
Schwab, C. G., T. P. Tylutki, R. S. Ordway, C. Sheaffer, and M. D. Stern. 2003.
Characterization of proteins in feeds. Journal of Dairy Science. 86, Supplement(0): E88E103.
Sheldon, M., and D. Noakes. 2002. Pregnancy diagnosis in cattle. In Practice. 24(6): 310-317.
Short, R. E., and R. A. Bellows. 1971. Relationships among weight gains, age at puberty and
reproductive performance in heifers. Journal of Animal Science. 32(1): 127-131.
Sinclair, K. D., M. Kuran, F. E. Gebbie, R. Webb, and T. G. McEvoy. 2000. Nitrogen
metabolism and fertility in cattle: II. Development of oocytes recovered from heifers
offered diets differing in their rate of nitrogen release in the rumen. Journal of Animal
Science. 78(10): 2670-2680.
Smith, J. F., S. Beaumont, L. Hagemann, R. M. McDonald, A. J. Peterson, Z. Z. Xu et al.
2001. Relationship between bulk milk urea nitrogen and reproductive performance of
New Zealand dairy herds.61: 192-194.
Staples, C. R., W. W. Thatcher, and J. H. Clark. 1990. Relationship between ovarian activity
and energy status during the early postpartum period of high producing dairy Cows.
Journal of Dairy Science. 73(4): 938-947.
Stewart, G. S., C. Graham, S. Cattell, T. P. L. Smith, N. L. Simmons, and C. P. Smith. 2005.
UT-B is expressed in bovine rumen: Potential role in ruminal urea transport. American
Journal of Physiology. 289(2(2)): R605-R612.
Stoop, W. M., H. Bovenhuis, and J. A. M. van Arendonk. 2007. Genetic parameters for milk
urea nitrogen in relation to milk production traits. Journal of Dairy Science. 90(4): 19811986.
Sunny, N. E., S. L. Owens, R. L. Baldwin, S. W. El-Kadi, R. A. Kohn, and B. J. Bequette.
2007. Salvage of blood urea nitrogen in sheep is highly dependent on plasma urea
concentration and the efficiency of capture within the digestive tract. Journal of Animal
Science. 85(4): 1006-1013.
84
© University of Pretoria
Tamminga, S. 2006. The effect of the supply of rumen degradable protein and metabolisable
protein on negative energy balance and fertility in dairy cows. Animal Reproduction
Science. 96(3–4): 227-239.
Tresley, J., and P. M. Sheean. 2008. Refeeding syndrome: Recognition is the key to
prevention and management. Journal of the American Dietetic Association. 108(12):
2105-2108.
Utley, P. R., N. W. Bradley, and J. A. Boling. 1970. Effect of restricted water intake on feed
intake, nutrient digestibility and nitrogen metabolism in steers. Journal of Animal
Science. 31(1): 130-135.
Varel, V. H., J. A. Nienaber, and H. C. Freetly. 1999. Conservation of nitrogen in cattle
feedlot waste with urease inhibitors. Journal of Animal Science. 77(5): 1162-1168.
Ward, W. R. 2000. Fertility and dietary energy and protein in dairy cows. British Cattle
Veterinary Association, BCVA. 8: 383-389.
Wattiaux, M. A., E. V. Nordheim, and P. Crump. 2005. Statistical evaluation of factors and
interactions affecting dairy herd improvement milk urea nitrogen in commercial
Midwest dairy herds. Journal of Dairy Science. 88(8): 3020-3035.
Weeth, H. J., and A. L. Lesperance. 1965. Renal function of cattle under various water and
salt loads. Journal of Animal Science. 24(2): 441-447.
Whitaker, D., E. Smith, G. da Rosa, and J. Kelly. 1993. Some effects of nutrition and
management on the fertility of dairy cattle. Veterinary Record. 133(3): 61-64.
Whittier, J. C., J. Berardinelli, and L. Anderson. 2008. Understanding puberty and
postpartum anoestrus. Applied reproductive strategies in beef cattle, Fort Collins.
Wildman, E. E., G. M. Jones, P. E. Wagner, R. L. Boman, H. F. Troutt Jr., and T. N. Lesch.
1982. A dairy cow body condition scoring system and its relationship to selected
production characteristics. Journal of Dairy Science. 65(3): 495-501.
85
© University of Pretoria
Wood, G. M., P. J. Boettcher, J. Jamrozik, G. B. Jansen, and D. F. Kelton. 2003. Estimation
of genetic parameters for concentrations of milk urea nitrogen. Journal of Dairy Science.
86(7): 2462-2469.
Wu, Z., and L. D. Satter. 2000. Milk production during the complete lactation of dairy cows
fed diets containing different amounts of protein. Journal of Dairy Science. 83(5): 10421051.
Xu, Z. Z., and L. J. Burton. 1999. Effects of oestrus synchronisation and fixed-time artificial
insemination on the reproductive performance of dairy heifers. New Zealand Veterinary
Journal. 47(3): 101-104.
Yelich, J. V., R. P. Wettemann, T. T. Marston, and L. J. Spicer. 1996. Luteinizing hormone,
growth hormone, insulin-like growth factor-I, insulin and metabolites before puberty in
heifers fed to gain at two rates. Domestic Animal Endocrinology. 13(4): 325-338.
Youngquist, R. S. 2007. Pregnancy Diagnosis. Pages 294-303 in Current Therapy in Large
Animal Theriogenology (Second Edition). Robert S. Youngquist et al., ed. W.B.
Saunders, Saint Louis.
Zhu, L. H., L. E. Armentano, D. R. Bremmer, R. R. Grummer, and S. J. Bertics. 2000.
Plasma concentration of urea, ammonia, glutamine around calving, and the relation of
hepatic triglyceride, to plasma ammonia removal and blood acid-base balance. Journal of
Dairy Science. 83(4): 734-740.
86
© University of Pretoria
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement