5 156312... . .
v. c..."
7

”93353.
. 420%..qu

JUKGVQ.

p 3‘!
3A...
..

7

. .........Puh.rr

.xnwa; ,

. ﬂit- h‘)!. ‘5
. 9233’.

~.....

w .sxx:“?b§
243$: 2011Y

‘1
.31: £\»t. .1,
x.~ovvv'.»1.u.
.533 .‘u:

t

.ﬂ I: . .. . . luxuxwlxz
. ..o u....p...lv\.4.\l»vxa. .. . 11.... a...

Warns}... . v‘

, . . l. T u . . . . . . . . . 2 . .5
..-....§ s-.§.% _ . . , . . ﬁﬁwmwm... .. . V . MW; .-:.Aw

 

THESIS

UNIVERSITY LIBRARI IES

IIIIIIIIIIIIIIIIIIIIIIIII I IIIIII III

3 1293 01405

I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the
thesis entitled
DRV COW DIET, MANAGEMENT AND ENERGY BALANCE AS

RISK FACTORS FOR ABOMASAL DISPLACEMENT IN HIGH
PRODUCING DAIRII HERDS IN MICHIGAN

presented by

Richa/Ld Edmund Bauhwich Came/Lon

has been accepted towards fulﬁllment
of the requirements for

M . S . degree in LACS

%%%

Major professor

Date .%?/¢é

0-7 639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

LIB RAHY
MIChIQan State
University

 

 

 

 

PLACE ﬂ RETURN BOX to remove We checkout tram your record.
TO AVOID FINES return on or before dete due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU le An Manetlve Action/Emu Opportunlty lnetltulon
Wanna-9.1

DRY COW DIET, MANAGEMENT AND ENERGY BALANCE AS RISK FACTORS
FOR ABOMASAL DISPLACEMENT IN HIGH PRODUCING
DAIRY HERDS IN MICHIGAN

By

Richard Edward Borthwick Cameron

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Large Animal Clinical Sciences

1996

ABSTRACT

DRY COW DIET, MANAGEMENT AND ENERGY BALANCE AS RISK FACTORS
FOR ABOMASAL DISPLACEMENT IN HIGH PRODUCING DAIRY HERDS IN
MICHIGAN

By

Richard Edward Borthwick Cameron

A prospective study of 1170 multiparous Holstein cows, in 67 commercial herds,
was conducted to determine risk factors for displaced abomasum (DA), in high producing
dairy herds in Michigan. Each farm was visited four times within a six week period. At
each visit, data on nutrition and management were collected and all multiparous cows
within 35 days of calving were body condition scored (BCS) and a blood sample taken
for plasma nonesteriﬁed fatty acids (NEFA) assay. A multivariate linear regression model
was developed for lactational incidence of DA by herd, and a multivariate random-effects
logistic model was developed for the occurrence of DA by cows. Signiﬁcant risk factors
for DA included a negative energy balance prepartum (as estimated from plasma NEF A),
high BCS, suboptimal bunk management, prepartum diets with NEl > 1.65Mcal/kg dry

matter, winter and summer seasons, high genetic merit and low parity.

To my parents, Jeanette and James Cameron.

iii

ACKNOWLEDGMENTS

This study represents the work of many people, in particular Paul Dyk, Dr. Roy
Emery, Dr. Herb Bucholtz and Jim Liesman. I thank Dr. Tom Herdt for serving as my
major professor. The guidance and assistance he readily provided were much
appreciated. I gratefully acknowledge the help and patience of Jim Liesman, Dr. John
Kaneene, and RoseAnn Miller with the data processing and statistical analysis. I thank
Dr. Roy Emery for his helpful input at all levels of this study. I am grateful to Dr. Herb
Bucholtz and Dr. Kent Ames for their input in making my program a rounded experience.

Lastly but by no means least, thank-you Fiona for all your great support.

TABLE OF CONTENTS

LIST OF TABLES. ......................................................................................................... vi
INTRODUCTION ............................................................................................................ 1
LITERATURE REVIEW ................................................................................................. 4
Displaced abomasum ............................................................................................... 4
Deﬁnitions ............................................................................................................... 4
Incidence .................................................................................................................. 5
Non-nutritional risk factors ...................................................................................... 5
Nutritional risk factors ........................................................................................... 12
Pathogenesis of abomasal displacement ................................................................ 17
Energy Metabolism in Ruminants ......................................................................... 21
Introduction ............................................................................................... 21

Metabolic response to negative energy balance ......................................... 22

Assessment of energy balance ................................................................... 24

Normal pattern of NEFA concentrations in the dry period ........................ 26

Pathological factors that affect plasma NEF A concentrations ................... 27

Temporal factors that affect plasma NEFA concentrations ....................... 30

Hepatic lipidosis and disease sequelae to negative energy balance ........... 3]

Signiﬁcance of hepatic lipidosis to the dairy industry ............................... 31

Prevention of hepatic lipidosis ................................................................... 32

Hepatic lipidosis - a potential risk factor for abomasal displacement ....... 33
MATERIALS AND METHODS .................................................................................... 35
The Data ................................................................................................................. 35
Selection of farms ...................................................................................... 35

Farm visits .................................................................................................. 35

Collection of DA incidence data ................................................................ 38

Statistical analysis .................................................................................................. 39
Cow-model ................................................................................................. 44

Herd-model ................................................................................................ 44

RESULTS .................................................................................................................... 45

Lactational incidence rate for DA .......................................................................... 45
Cow-based analysis ................................................................................................ 48
Herd-based analysis ............................................................................................... 48
DISCUSSION ................................................................................................................. 52
Lactational incidence rate ...................................................................................... 52
Cow-model ............................................................................................................. 53
Herd-model ............................................................................................................ 58
Comparison of ﬁndings between both modeling approaches ................................ 63
Limitations in the study ......................................................................................... 65
SUMMARY and CONCLUSIONS ................................................................................ 67
Practical recommendations .................................................................................... 68
APPENDIX .................................................................................................................... 71
BIBLIOGRAPHY ........................................................................................................... 73

vi

List of Tables

Table 1. Herd-based risk factors used in analysis of DA .................................................. 41
Table 2. Cow-based risk factors used in analysis of DA .................................................. 42
Table 3. Outcome variables used in analysis of DA ......................................................... 43
Table 4. DA incidence for all cows (excluding those cows with twins) ........................... 46
Table 5. DA incidence for cows with twins ...................................................................... 47
Table 6. Breakdown of risk factors based on cows’ DA status ........................................ 49

Table 7. Cow-based multivariate random effects logistic models for the
occurrence/non-occurrence of DA ...................................................................... 50

Table 8. Herd-based multivariate linear regression model using logit transformed
DA incidence rate ................................................................................................ 51

vii

INTRODUCTION

Since the ﬁrst reported case of left displaced abomasum by Ford in 1950,
displaced abomasum (DA) has emerged as a common production disease in high
producing dairy herds (Robertson, 1968; Grymer, 1980). The cost per case associated
with treatment and lost production has been estimated to range from $256 to $406 in
1993 (Bartlett et al., 1994). An additional factor in considering the signiﬁcance of this
disease is that it is often associated with other production diseases such as hypocalcemia
(Curtis et al., 1983), ketosis, and metritis (Curtis et al., 1985).

Epidemiological studies have identiﬁed breed, sex, age, season, concurrent
diseases (Constable, 1992), and nutrition (Robertson, 1968; Coppock, 1974; Correa,
1990) as potential risk factors for DA. Nutrition is considered to have a signiﬁcant impact
on incidence and is perhaps the factor most feasible to control in a dairy. Yet a paucity of
understanding still exists in the area of nutritional risk factors and research in the area has
been relatively sparse and the results controversial.

The speciﬁc objective of this investigation was to identify potential risk factors
for abomasal displacement that may operate on a herd level with special attention to
nutrition and management and compare these results with those of an individual-cow
model based on the same data set. The rationale for this two model approach is that in

practice, the herd is the unit of concern and disease control measures are generally

applied at the farm level. However, as disease affects the individual it is useful to
evaluate factors that are speciﬁc to the individual such as age, the presence or history of
other disease, nutritional status, so that preventative measures can be instituted at this
level.

This study was based on commercial, high producing herds in Michigan so that
ﬁndings from the study would be directly pertinent to these high producing herds. This
investigation offers the following advantages over several previous studies: a prospective
design that allows for greater data accuracy and the establishment of temporal
relationships between hypothesized risk factors and the outcome; a large size so that
conclusions are less likely to be inﬂuenced by the peculiarities of a single herd, and
ﬁnally, a consideration of a wide range of variables; all of which are necessary for the
study of a multifactorial disease. In the setting of a large ﬁeld study involving many
herds, the study of the incidence of DA has a distinct advantage over other disease
incidences. The occurrence of DA is usually heralded by a dramatic change in appetite
and performance that is easily noticed by the farmer. In addition, it has a speciﬁc
diagnosis that is highly repeatable by persons skilled in the clinical evaluation for the
condition. Hence, the diagnosis of DA is not subject to the same owner- and
veterinarian-variation in diagnosis as is hypocalcemia, metritis, mastitis and ketosis. This
variation in diagnosis together with variation in management practices leads to increased
or decreased reporting of disease relative to its true incidence which can have a major
inﬂuence on the results and interpretation of a study (Kaneene and Miller, 1995). With
DA being the primary focus of this study, it is not intended that the importance of these

other periparturient diseases to health and production be ignored. Due to the association

between DA and other periparturient diseases, it seems reasonable to expect that
measures that reduce the incidence of DA may have a beneﬁcial impact on the incidence

of other periparturient disorders.

LITERATURE REVIEW

Wm
Deﬁnitions

The abomasum’s normal position is on the abdominal ﬂoor with the body of the
abomasum positioned between the ventral sac of the rumen and the omasum (Habel,
1970). When it becomes distended, the abomasum can migrate or displace from its
normal position. With distention, the abomasum bends in the middle and forms a loop
between the omasurn and the cranial duodenum. The greater curvature, with the line of
attachment of the greater omentum, is on the outside of the loop. The lesser omentum is
on the inside. The crest of the loop may slip under the rumen and come up along the left
side where it is buoyed up by gas trapped at the top of the loop (left displacement of the
abomasum, LDA), (deLahunta and Habel, 1986; Sack, 1968). Alternatively, the loop of
abomasum may simply dilate and displace caudally in a sagittal plane on the right side
with the pyloric part dorsal (right displacement, RDA). In some cases, the abomasum
may undergo volvulus through 180 to 450 degrees, usually counterclockwise as viewed

from the right side of the cow with the lesser omentum serving as the axis of rotation

(Habel and Smith, 1981).

Incidence

In a New York survey, Coppock (1974) reported a DA incidence rate of 1.16%
among the affected herds and .3 5% when expressed over the total number of lactations
from all herds that reported in the survey. Various authors have reported that the
incidence rate appears to be increasing: Robertson (1968) reported that the number of
cattle with LDA admitted to the Large Animal Hospital of the University of Pennsylvania
increased from 3% of total admissions in 1960 to 30% in 1963; Markusﬁeld (1986)
documented an apparent increase in lactational incidence rate from 0.8% during 1974-
1976 to 1.7% during 1980-1984 in seven Israeli-Friesian herds.

Dohoo and Martin (1984) found a lactational incidence of 1.4% based on 2008
cows in 32 commercial herds in Ontario. Bartlett et a1. (1986) reported a lactational
incidence rate of 2.5% based on 2847 lactations in 21 Michigan dairy herds. Robb et a1.
(1987) reported an incidence rate of 4.36% in 50 dairy herds. Pehrson and Shaver (1992)
in a study involving 71 herds in southern Wisconsin recorded an average incidence rate of
5.0%.

Non-nutritional risk factors
WW

Female cattle are at higher risk of developing DA than are males (Constable,
1992). Abomasal displacement is found more frequently in dairy breeds than beef
breeds (Constable, 1992). Pinsent et al. (1961), observed that Channel Island breeds are
more susceptible than other breeds to the condition. Conversely, Robertson (1968) did
not detect a difference among the breeds when the comparison was based on the

population distribution. Constable et a1. (1992) noted that within breeds, Guernsey cattle

are more frequently affected than Holstein-Friesians, which in turn are more frequently
affected than Brown Swiss. Although the authors have given no explanation for
difference in breed susceptibility, it may be possible that the susceptibility in Channel
Island breeds may be related to their predisposition for hypocalcemia at parturition. The
inﬂuence of hypocalcemia will be discussed later (page 18).

It has been observed that affected cattle are larger than the population average
(Coppock ,1974). Fox (1965) noted that the depth of body of the dairy cow has increased
markedly in the past 20 years, allowing potentially more room for abdominal viscera to
become dislocated. Mahoney et al. (1986) compared health care needs of cows bred for
large or small body size. Two groups were formed from a paired foundation population.
The large group was mated to sires with extreme estimates of transmitting ability for tall
height and deep and wide bodies, while the small group was similarly mated to extreme
sires but to those transmitting short height and shallow and narrow bodies. They found
that large cows required signiﬁcantly more health care than small cows. Digestive
disorders accounted for much of the group difference and displaced abomasums were
more frequent among large cows, ( 6 of 220 large and 0 of 191 small cows had DA
during ﬁrst lactations and 10 of 220 large and 2 of 191 small cows had DA during their
life times).

Martin (1972) cited evidence from a pedigree study which indicated genotype as a
possible predisposing factor. The genetic factors were unknown but shape and size of
abdomen were cited as possibilities. These factors appear plausible although there must
indeed be more factors involved as a signiﬁcant change in anatomical proportions of

dairy cows is unlikely to have occurred so suddenly to explain the emergence of this

disease. Dairy cows do have deeper abdomens than beef breeds which in part may explain
the difference in susceptibility between these two breeds.
Ag:

Many studies have shown an increased risk of DA with increasing parity (Martin,
1972; Erb and Martin, 1978). The incidence of DA increases with age with the highest
risk between the ages of 4 and 7 years (Constable, 1992), however, high incidence rates
in ﬁrst parity cows have been reported (Willeberg et al., 1982). There is evidence that the
disease is affecting younger cattle more commonly than previously reported (Pehrson and
Shaver, 1992). Dohoo and Martin (1984) were unable to detect a trend with age in their
study. A possible explanation for the commonly reported trend of increasing incidence
with age may be that increasing age is associated with an increase in occurrence of other
diseases such as parturient paresis which, as discussed later, may be risk factors for DA.
E . .

The occurrence of abomasal displacement is closely associated with parturition.
Based on 7,695 records of LDA cases from 17 North American veterinary schools,
Constable (1992) reported that 57.0% of cases were diagnosed within 2 weeks and 80.2%
of cases within one month of parturition. Approximately 10% of abomasal displacements
occurred during pregnancy and were distributed equally in the second and third
trimesters. Parturition heralds signiﬁcant change to the dairy cow in terms of
management and dietary changes, metabolic demands, and risk of disease. Thus the
challenge of experimental and observational studies, to elucidate signiﬁcant risk factors
for DA, can be appreciated when it is realized that so many variables of importance are

changing during this period of risk.

Season

Pinsent (1961), was one of the ﬁrst authors to note a marked seasonal incidence
with the great majority of cases occurring between October and April. This took into
consideration seasonal calving patterns. A seasonal incidence was also reported by
Robertson (1968) with 60% of cases occurring between November and May, and by
Martin (1972) who reported a high incidence in spring and summer. Constable (1992)
showed that the peak number of LDA cases develop in spring and the lowest number in
autumn. As yet there is no proven explanation for the seasonal occurrence however,
Constable (1992) theorized that seasonal changes in rumen volume secondary to seasonal
variation in roughage intake may be involved with a full rumen constituting a moderately
effective barrier for displacement. This theory has some merit although no documented
proof exists that such a distinct change in forage intake occurs through out the
population. It is possible that there may be factors associated with season that inﬂuence
dry matter intake (DMI) in herds.

Mm

Markusfeld (1986) reported an association between DA and the birth of twins, an
observation that has been repeated by Pehrson and Shaver (1992). The application of
Constable (1992) and Svendsen’s (1969) theory that a ﬁll] rumen constitutes a moderately
effective barrier for displacement would ﬁt well in this situation. Twin fetuses in utero
would severely compromise rumen capacity in the cow such that after calving the rumen
volume may be even lower than usual. In addition cows with twins are more predisposed

to common metabolic diseases due to the extra demands of the additional fetus.

Concurrentﬂiseasrs

Constable (1992) documented that 53.6% of 780 cattle with LDA and 30.4% of
280 cattle with abomasal volvulus had additional diseases.

Hypocalcemia

Hull et al. (1973), Curtis et al. (1983) and Massey et al. (1993) have presented
evidence indicating hypocalcemia to be a potential risk factor for DA. Markusfeld (1986)
could ﬁnd no such association and suggested that previously described associations
might have been the result of a tendency for a rising rate with parity, common to both
diseases. Oetzel (1993) investigated the effects of prophylactic treatment with a calcium
chloride gel on serum calcium concentration at calving, parturient paresis and displaced
abomasum in Holstein cows (n = 204). Treatment decreased (p< .05) the incidence of
parturient paresis (4.9 versus 11.8%), and decreased (p<.01) the incidence of subclinical
hypocalcemia (29.4 versus 52.0%). Displaced abomasum during the ﬁrst 3 weeks after
calving was also decreased (p< .05) by treatment (1.0 versus 7.8%). The role of
hypocalcemia in the etiology of DA will be discussed further (page 18).

Retained placenta and metritis

In both a study using a bivariate analysis (Coppock et al., 1972) and another using
a multivariate analysis (Markusfeld, 1987), retained placenta has been indicated as a risk
factor for DA. However, other multivariate studies did not show an association (Dohoo
and Martin, 1984; Curtis et al., 1985). Erb and Grohn (1988) concluded that parturient
paresis is a risk factor for LDA and that much of the association is mediated by retained

placenta.

10

In the multivariate model put forward by Erb and Grohn (1988), metritis had an
independent direct effect on the occurrence of LDA. An association was also reported
by Markusfeld (1987) and Coppock et al. (1972). An earlier observation by Robertson
(1968) was that 52% of cows with LDA also had concurrent metritis and 24% had
metritis prior to the diagnosis of DA. This rate of metritis, one assumes, was higher than
expected although no control data was listed. Apart from the relationship between
parturient paresis, retained placenta and DA reported by Erb and Grohn, it is also
conceivable that metritis ﬁts into the possible association as well. Hypocalcemia, as
discussed later, may reduce uterine involution which in turn may predispose the uterus to
metritis, especially in the presence of a retained placenta. Apart from the effect of
hypocalcemia causing abomasal atony, endotoxins from a metritis may also inhibit
abomasal motility (Vlaminck et al., 1985). Poor abomasal motility is considered an
important factor in the etiology of DA.

Ketosis

Curtis et al. (1985) in a path analysis based on 1374 lactations of multiparous
cows, showed that clinical ketosis diagnosed prior to LDA had an effect on occurrence,
increasing the odds ratio to 11.9 . Conversely Dohoo and Martin 111 (1984) in a path
analysis of 2711 lactations that included primiparous cows, were unable to detect a
relationship. Curtis and co-workers (1985) did not offer an explanation for these different
results. Possible reasons to be considered include that different populations were studied
and that a large difference in lactational incidence rates for uncomplicated ketosis were
reported, (7.4% in Dohoo and Martin's study versus 1.5% in the study of Curtis et al.)

despite a similar lactational incidence rate for DA (1.2% versus 1.4%). Varden (1979)

11

reported that 30 out of 45 cows were treated for ketosis one to eight days before
displacement could be demonstrated. Robb et al. (1986) in a case control study found
that ketosis prior to diagnosis of DA was associated with an odds ratio of 38.8 (18.6 -
81.1) that was strongest for ﬁrst lactation animals. A potential pathophysiological
explanation for this association will be discussed later (page 19).
Milkﬁmdnstinn

It has long been suggested that higher milk production is associated with
increased risk of DA (Fox , 1965; Martin, 1972; Coppock, 1974; Grymer, 1981). In a
bivariate comparison, Robertson (1968 ) found that herds reporting LDAs (as opposed to
those herds with no LDAs) had a milk yield higher than the average for herds on DHI.
Shanks et al. (1978) found in a pedigree study a non signiﬁcant doubling of the rate of
LDA in cows with a high pedigree for milk production. Martin et al. (1978) in a case
control study found no indication that higher production cows had a higher incidence rate.
Curtis et al. (1985) in a large multivariate analysis (n = 1347) which tested deviation
from herd mate average in previous-lactation, mature-equivalent milk yield, found no
association between milk yield and LDA. Dohoo and Martin (1984) found no association
between previous milk production and the occurrence of DA in a discriminant analysis
applied to 817 cows that had completed two lactations. If one is to interpret the
signiﬁcance of the opposing results in the above mentioned studies, attention has to be
drawn to the design and type of analysis applied to the different studies. The studies by
Dohoo and Martin (1984) and Curtis and workers (1985) are superior to the others in
terms of their prospective design, large sample size and better control for confounding

variables in a multivariate analysis. Hence the weight of evidence seems to indicate that

l2

milk production is not a risk factor for DA. Erb and Grohn (1988) stated that it is unlikely
that high cow milk yield is a risk factor for LDA, however, high previous-lactation herd
yield but not high previous cow milk yield was a signiﬁcant risk factor.
Iiushandeamrs

It has been suggested that lack of exercise associated with tethering or housing in
winter increased the incidence of DA, however Martin (1972) observed that twice as
many affected herds as unaffected herds (28/66 versus 4/26) used loose housing as
opposed to tie stalls or a combination. Furthermore, LDA has been observed in cows that
are exclusively pastured (Jubb et al. 1991) indicating that exercise is unlikely to be
protective or that restricted exercise is a signiﬁcant risk factor for DA.
Nutritional risk factors

A number of nutritional factors have been speculated upon as risk factors for DA
over the past 40 years; however, there are 3 main factors that have featured predominantly
in the literature. These are:-
-the level of concentrate feeding
-dietary ﬁber levels
-dietary calcium levels and more recently, the dietary cation-anion (DCAD) difference
Cnncentratefredina

The issue of concentrate feeding was the ﬁrst to receive attention. It is interesting
historically that after the ﬁrst reported case of LDA by Ford in 1950, the number of clinical
reports showed a steady increase as reviewed by Pinsent (1961). The question had to be
asked what changes were occurring in the dairy industry that may have brought about the

emergence of this disease? Signiﬁcant changes were occurring. Research by Huffman

13

(1961) had shown that conditions in the dairy industry made it economically worthwhile to
feed dairy cows higher levels of grain than had been previously practiced. Thus, the trend in
the dairy industry since the early '608 has been to feed more grain. This has been speculated
to be responsible for the increased occurrence of DA (Pinsent, 1961). Later it was further
speciﬁed that lead feeding (the feeding of grain to cows in the late dry period) was
associated with a high incidence of DA (Robertson, 1968; Coppock, 1974; Correa et al.,
1990)

Lead feeding allows the rumen microorganisms to adapt to a higher grain diet prior
to parturition and, perhaps more importantly, induces elongation of the rumen papillae.
Several researchers including Robertson (1968), Coppock (1974), and Correa et al., (1990)
have concluded that lead feeding is associated with a higher incidence of DA.
Unfortunately, in the reports of Robertson and Correa the level of grain feeding was not
deﬁned. Conceivably, cows could have been receiving anywhere from a pound of grain to
an all grain diet. Robertson used a case-control study involving 30 herds reporting the
occurrence of DA, that were matched with 30 herds without any occurrence of the disease.
He reported a statistically signiﬁcant difference in the two groups in the mean amount of
grain consumed in the last month of pregnancy. No correlation, however, was found
between the level of concentrate feeding and the incidence of LDA within herds. It could
be argued that Robertson’s conclusion that lead feeding predisposes to the occurrence of
DA is not substantiated by his data. It is interesting that, contrary to the ﬁndings of
Robertson and Correa, Curtis and workers (1985) came to the opposite conclusion in a
study of individual cows using the same data set as Correa. In Correa and co-workers'

study, the data were analyzed at the herd level and the association evaluated was between

14

DA incidence and whether or not the farm's stated policy was to lead feed. In Curtis and co-
workers' analysis of individual cows, the speciﬁc diet fed on each farm was classiﬁed with
respect to energy and assigned to one of three energy levels. Curtis concluded that the
higher tercils of estimated energy were equated with the practice of lead feeding, which
decreased the risk of LDA.

Svendson (1969) investigated the link between DA and high concentrate diets and
concluded that high concentrate diets lead to high concentrations of volatile fatty acids
(V FAs) in the abomasum, resulting in abomasal atony - an important prerequisite for
displacement. However, Breukink and De Ruyter (1976) could not demonstrate elevated
abomasal VFAs in cows on high grain diets. Another explanation for the potential effect of
high grain diets on the incidence of DA is that these types of diets lead to ruminal acidosis.
Ruminal acidosis results in the release of histamine and endotoxin (Dunlop, 1972) which
has been shown to decrease abomasal emptying (Vlaminck et al., 1985).

Eibsnchcl

Fiber can be conceived as having two overlapping functions in the prevention of
DA. 1n the dry cow, a high ﬁber diet, as opposed to one high in concentrates, will lead to
greater dry matter intakes (Coppock, 1972; Nocek, 1983). Furthermore, due to the ﬁbrous
nature of the diet, feed has a longer residence time in the rumen, leading to a greater rumen
volume. Both Svendson (1969) and Constable (1992) have hypothesized that the normal
rumen volume constitutes a moderately effective barrier against LDA. Postpartum cows
have a signiﬁcantly reduced rumen volume after calving as the fetus in utero occupies a
large space in the abdomen that restricts rumen capacity. After calving, rumen volume will

increase with time in relation to increasing dry matter intake. Constable (1992) considered

15

that evidence for the importance of normal rumen volume is reﬂected in the ﬁndings that
proportionately fewer cases of abomasal volvulus, compared with LDA, develop in the
immediate postpartum period indicating that rumen volume may directly inﬂuence the
direction of abomasal displacement. Furthermore, Constable considered that seasonal
variation in LDA incidence may in part be explained by a potential seasonal variation in
rumen volume if the high incidence in spring is related to a smaller rumen volume which in
turn may be related to lower forage intakes in late winter or early spring accompanying
depletion of farm hay stores. Constable (1992) surmised that the high incidence of LDA in
lactating dairy cattle results from the additive effects of: decreased rumen volume, increased
abdominal void immediately after parturition and increased exposure to factors that induce
abomasal atony.

Grymer's report (1981) is the only journal publication located in which the
relationship between ﬁber levels and DA was investigated, in an epidemiological manner.
He concluded from a case control study that lactating rations less than 16 to 17% crude
ﬁber were associated with increased risk of DA.

The physical nature of the ﬁber appears important. A ration may have a high ﬁber
content but have little effect on rumination if it is ﬁnely processed. Dawson et al. (1992)
documented that cows (n = 46) fed an experimental diet in which ground alfalfa hay was the
principal roughage source were 10.8 times more likely to develop LDA as compared to
those fed a normal alfalfa hay.

Apart from its physical affect on rumen volume, ﬁber also serves as a stimulus for
the cow to ruminate. When the cow is placed on a high concentrate diet rumination

becomes vital for a supply of bicarbonate rich saliva to prevent ruminal acidosis.

16

Obviously the levels of ﬁber and concentrate feeding are inversely correlated to
each other and are not necessarily separate issues. Robertson (1968) reported that non-
affected herds showed a trend of consistently consuming more nutrients in the form of
roughage and that affected herds consistently fed rations having signiﬁcantly higher grain
content. However, a correlation was not evident between the level of grain feeding and the
incidence of LDA within herds. Nocek et al. (1983) divided 289 cows into 3 groups during
the dry period and fed diets of 100% hay, 50:50 hayzcom silage, or limited corn silage plus
liquid protein supplement. The LDA incidence was 3, 4.3, and 6.3% respectively, which
although not signiﬁcantly different, indicated a trend of increasing incidence as the diets
declined in hay content. In an earlier experiment Coppock (1972) assigned 40 Holsteins to
four complete diets of 75:25, 60:40, 45:55 and 30:70 forage to concentrate ratios 28 days
before parturition. During 25 days following parturition, 0, 2, 4, and 4 cases of LDA
occurred in the four ration groups respectively. Conversely, Johnson and Otterby (1981)
reported that in a trial evaluating 3 dry period diets ranging from all hay to a complete
ration of 47% grain and 53% ensiled forage, there was little effect on feed consumption and
no cases of DA recorded.

Based on the conﬂicting reports in the literature, it would appear that the issue of
concentrate and ﬁber levels is an important area in need of further research. A large number
of references that are often quoted in reviews of this issue have no statistically signiﬁcant
results and weight is placed on mere potential trends.
WWW

The effect of dry cow nutrition on the occurrence of clinical and sub-clinical

hypocalcemia has been well documented (Oetzel et al., 1988; Oetzel, 1991). There have

17

been reports by Hull et al. (1973), Curtis et al. (1983) and Massey et al. (1993) that show
hypocalcemia as signiﬁcant risk factor for DA. Recent research has shown that the dietary
cation-anion difference is more important in preventing hypocalcemia than dietary calcium
concentration (Oetzel, 1991 ).
Pathogenesis of abomasal displacement
McchanicaLIhenry

Beg and Whiteford (1956) postulated that the uterus in late pregnancy elevates the
rumen and pushes the abomasum cranially and to the left of midline. After parturition the
rumen drops back and traps the abomasum between the body wall and the rumen,
according to their theory. Constable (1992) showed that the incidence of LDA did not
increase substantially from the second to third trimester giving little support for the
theory of Beg and Whiteford. The fact that DA is reported in bulls, steers, and calves,
and that a number of nutritional and metabolic risk factors have been identiﬁed, makes it
likely that the mechanical theory plays a minor, if any role in most cases of DA.

! l l H 'l' I

There is general agreement that abomasal hypomotility is an important
prerequisite for abomasal displacement to occur (Pinsent, 1962; Svendsen, 1969;
Breukink and de Ruyter, 1976). The hypomotility results in the accumulation of ingesta
and gas leading to distention and displacement. In looking at etiologies for abomasal
hypomotility several factors have been considered:-

Volatile Fatty Acids (VFAS)

Svendson (1969) considered that the possible role of large amounts of

concentrates in the etiology a DA may be mediated by the effect of VFAs on abomasal

l8

motility and gas production. In experiments involving the infusion of ruminal contents
from an animal fed a high concentrate diet into the abomasum of other cows, Svendson
was able to demonstrate a reduction in abomasal motility. This could be repeated using
solutions of VFAs similar in composition to what was found in ruminal ﬂuid from a cow
on a high concentrate diet. However, Breukink and De Ruyter (1976) compared a high
grain diet with an all hay ration and found that there was no difference in abomasal VFA
concentrations between the two diets. Madison et a1. (1993) studied the effects of a
change from a diet composed primarily of forage to a diet composed primarily of
concentrate on abomasal antral and duodenal motility of 5 multiparous lactating Holstein
cows. Their results suggest that rapid change from a low to a high concentrate diet alone
is not sufﬁcient to cause disruption of abomasal antral or duodenal motility that would
induce abomasal displacement.

Hypocalcemia

The vital importance of calcium for muscle function, especially smooth muscle
function, is well known. Hypocalcemia can effect the motility of the gastrointestinal tract
including the abomasum resulting in hypomotility or atony. Hull and Wass (1973)
postulated on the possible role hypocalcemia played in the etiology of DA. This was
based on observations that many cases of DA can be temporarily corrected by
administration of calcium preparations, that some cows with DA also had a history of
milk fever, and that the role of hypocalcemia in impaired uterine involution may be the
link between the association between metritis and DA. Reports by Curtis et al. (1983)
and Massey et al. (1993) indicate hypocalcemia as a signiﬁcant risk factor for DA.

However, the fact that DA can be a signiﬁcant disease in primiparous cows, which

l9

reportedly are less affected by hypocalcemia than older cows, means that there are likely
additional factors involved in the etiology of abomasal hypomotility.

Endotoxin

The association of DA with concurrent disease has stimulated research into
possible roles mediators and products of inﬂammation might play in the etiology of
abomasal hypomotility. Vlaminck et al. (1985) in an experimental study with three
duodenal ﬁstulated cows, showed a marked inhibition of abomasal emptying after the
intravenous administration of endotoxin.

Normal periods of inactivity

Kuiper et al. (1988) studied the myo-electric pattern of the abomasum and found
that there were long periods of inactivity which may play a role in the pathogenesis of
DA. These periods may last for up to several hours and can occur less than once a day or
even once a week and are independent of the cow's position.

Negative-energy—balance associated metabolic disease

Speculation has been raised over the possible association between disorders of
energy metabolism (hepatic lipidosis and ketosis) and the occurrence of abomasal
displacement (Holtenius and Niskanen, 1990; Breukink, 1991). As discussed, an
association between abomasal displacement and ketosis diagnosed prior to displacement
has been indicated by some workers (Curtis et al., 1985; Varden, 1979). Although a
pathophysiological explanation for this association has yet to be proven, there is evidence
for the possible explanations that have been advanced.

Meirhaeghe et al. (1988) found that animals with abomasal displacement showed

higher basal blood glucose levels than control animals, independent of their blood ketone

20

body concentration. They also observed a delay in glucose clearance from the blood of
cows with DA, despite elevated insulin levels that point towards the possibility of an
insensitivity of body tissues to insulin. The same authors found in a study of four
ﬁstulated heifers, that endogenous insulin, secreted after intra venous glucose inﬁision, as
well as exogenously injected insulin, caused a marked decrease of abomasal emptying
rate lasting several hours. They suggested that insulin itself, independent of hyper- or
hypoglycemia, might be the major cause of inhibition of abomasal emptying. They noted
that infusions of some volatile fatty acids (butyrate, iso-valerate, valerate) induced an
insulin response greater than that produced by the infusion of glucose. It is possible that
these and other insulinogenic substances are involved in the regulations of abomasal
emptying and that disturbances in volatile fatty acid metabolism may play a role in the
pathogenesis of abomasal displacement. DeCupere et al. (1991) reported that cows (11 =
38) with left displaced abomasrun exhibited insulin resistance which differed both with
non-ketotic (n = 33) and ketotic (n = 7) control animals. This is further supported by
Holtenius and Traven (1990) who showed impaired glucose tolerance and heterogeneity
of insulin responses in studies with 36 cows with abomasal displacement, indicating
insulin resistance and signs of non-insulin-dependent diabetes mellitus. Causes for the
insulin resistance are not clear but lipolysis and ketone bodies may play a role.

The potential importance of disorders of energy metabolism in the pathogenesis of
DA makes it pertinent to review energy metabolism and associated disorders in the

ruminant and this will comprise the second section of this review.

21

Introduction

An understanding of metabolic disease in ruminants related to disorders of energy
metabolism would be facilitated by a brief review of ruminant energy metabolism. This
can be achieved by highlighting the fundamental differences in energy metabolism
between ruminants and the commonly understood models based on humans and rats.

The ﬁrst point is that the majority of carbohydrate (CHO) digestion in ruminants
occurs in the rumen by fermentation to yield 3 major volatile fatty acids - acetate,
propionate and butyrate. Of these products of CHO digestion only propionate can be
metabolically converted to glucose (Bruss, 1993). Other sources of glucose can be
obtained from certain amino acids entering the Krebs cycle. Thus, an important point to
appreciate is that the ruminant is in a constant state of gluconeogenesis (Herdt, 1992,
1988a).

In monogastric species such as the rat, CHO digestion results in glucose
absorption. After digestion in these species, the liver has to cope with rising levels of
blood glucose. In order to prevent excessive blood glucose levels, the liver has to convert
the glucose into alternative forms. Glucose can be converted to glycogen (which can
reach a concentration limit) and to fatty acids. Hence, rat and human livers are important
sites of fatty acid synthesis, in contrast to ruminants which synthesize essentially no fatty
acids in their liver. It follows on that in species where fatty acid synthesis is a signiﬁcant
feature of hepatic metabolism, that they are able to secrete the fatty acids from
hepatocytes, in the form of trigylceride (TG), in an efﬁcient manner. The TGs that are

produced are packaged into very low density lipoprotein (V LDL), which is TG

22

surrounded by an envelope of protein, phospholipid, and cholesterol. The VLDL are
secreted into the circulation (Herdt, 1992). Pullen et al. (1990) compared liver slices
from 8 different species and concluded that the ability of liver to secrete TG as VLDL is
proportional to the livers capacity for lipogenic fatty acid synthesis. Thus, it is thought
that because ruminant livers are not a site of fatty acid synthesis that in turn the metabolic
machinery to package TG into VLDL for secretion out of the hepatocyte is limited. It is
probable that it is the secretory process itself, not the synthesis of triacyglycerol, which is
responsible for the limited hepatic secretion (Emery et al., 1992). Fatty acid synthesis in
the ruminant occurs in adipocytes from acetate ~ the most abundant energy source for
ruminants (Herdt, 1992).
I I 1 1° . l l

The term negative energy balance implies that the animal is expending more
energy in a day than is being supplied from the diet. The result is a decline in blood
glucose which causes a decline in the insulin/glucagon ratio. This stimulates
glycogenolysis from glycogen stores and gluconeogenesis from glucogenic amino acids.
However, the major energy source, especially in prolonged periods of energy deﬁcit, is
fatty acids. Fatty acids can be utilized directly by the tissues for energy or can be
converted to ketone bodies in the liver. Ketone bodies are water soluble molecules that
serve as a substitute for glucose in nearly every tissue. During prolonged negative energy
balance, there is a shifting of peripheral fuel utilization from glucose and amino acids to
ketone bodies and fatty acids (Herdt, 1992).

Fatty acids that are released from lipolysis circulate in the blood bound to albumin

and are referred to as plasma nonesteriﬁed fatty acids (NEFAs) to distinguish them from

23

triglyceride fatty acids in lipoproteins and chylomicrons. NEF As can be utilized by
tissue for energy. However, as NEFAs circulate bound to albumin, a more effective
transport system is necessary as albumin can become saturated with NEFA. Hence, the
liver actively takes up NEFAs. The latter can be metabolized via 3 possible pathways -
complete oxidation, incomplete oxidation or re-esteriﬁcation. Complete oxidation of the
fatty acids to carbon dioxide and water is possible but of minor signiﬁcance. Some of the
fatty acids are transported into the mitochondria and undergo [3 oxidation to ultimately
produce ketone bodies. The passage of fatty acids into the mitochondria is a point of
regulation. This prevents excess ketone body formation or hyperketonemia. In cases of
ketosis there is compromise of this regulatory process (Herdt, 1988a).

Hepatic uptake of NEFAs is proportional to plasma levels. When high levels of
fatty acids accumulate in the hepatocyte they are re- esteriﬁed to form TG. The TG thus
formed has to be packaged into VLDL in order to be secreted into the circulation. The
secretion of VLDL in ruminants is limited and may be even further compromised by high
levels of accumulated TG. Large amounts of hepatic TG lead to fatty liver (Grummer,
1993)

What is relatively new in the understanding of fatty liver is the appreciation of the
chronology of the condition. Fatty liver has often been considered a disorder developing
in the postpartum. Results from investigations by Gerloff (1986), Bertics (1992) and
others have shown that cattle with severe hepatic lipidosis may have high hepatic TG
concentrations accumulating during late gestation and may be more susceptible to severe

TG accumulation developing near the time of parturition (Grummer, 1993).

24

A note on normality - there is a normal rise in NEFA levels in the last week or
more of gestation and especially so around parturition (Grummer, 1993; Vazquez-Anon
et al., 1994). As mentioned, some degree, possibly less than 20% triglyceride (dry matter
basis) inﬁltration of the liver is normal (Grummer, 1993) with no adverse results
occurring. However, in certain circumstances the mobilization of NEFAs is excessive or
occurring earlier than anticipated. The result is an overload of NEFAs to the liver and to
a greater extent than normal, with respect to parturition.

Two potential means for the control of hepatic lipidosis are evident:-
1) Reduction of adipose mobilization
2) Enhanced hepatic VLDL output
As there is currently not much that can be done about the latter, attention needs to be
focused on the former point and this is the direction in which the review will proceed.
Assessment of energy balance

Appreciating that a negative energy balance is the starting point in the
pathogenesis of disorders of energy metabolism, variables that could be used to asses
energy balance would be of great value. Furthermore, they would allow an evaluation
between possible associations between negative energy balance and the occurrence of
disease such as DA.

There are several variables that can be used in assessing the energy balance in

ruminants.

25

Body condition score

This is widely used in dairy herds. A potential disadvantage of the system is that it
may not be sensitive enough. One body condition score unit approximates 54 kgs of body
weight (Garrnsworthy, 1988). Detecting a loss of half a point on the frame of a 600 kg
cow that is expanding with advancing pregnancy and perhaps complicated with
prepartum edema can be difﬁcult. Body condition scoring may detect signiﬁcant weight
loss occurring through out the dry period, but might not be sensitive enough to pick up
condition loss occurring in the late dry period.

Plasma glucose concentrations

This variable was employed in metabolic proﬁle testing widely used in the United
Kingdom and elsewhere (Cote and Hoff, 1991). The disadvantage of this is that glucose
concentrations are maintained via homeostatic mechanisms that could hide mild or early
stages of energy imbalance. In addition this variable is inﬂuenced by stress (Lindsay,
1977).

Plasma ketone bodies

This variable is not well correlated with energy balance (Kunz and Blum, 1985).
It may fail to indicate mild degrees of energy imbalance and is possibly more indicative
of the starch concentration of the ration than its caloric sufﬁciency. Ketone bodies also
exhibit a marked diurnal variation in concentration, possibly linked to feeding, which
comprises a major proportion of total variance (Gaal et al., 1983).

Plasma nonesteriﬁed fatty acids (NEFA)

This is possibly the most sensitive indicator of the caloric sufﬁciency of a diet

(Kunz and Blum, 1985) but there are potential sources of variation in concentrations in

26

addition to caloric sufﬁciency that will be considered later. Annison in 1960 (cited in
Bowden, 1971) was the ﬁrst to suggest that plasma NEFA levels could serve as an
indicator of energy balance in ruminants. A lot of the research in the '605 and '7OS
involved with NEFAs was concerned with nutritional studies in sheep, looking at the
effect of dietary energy levels on a wide range of body metabolites including plasma
NEFA concentrations. These nutritional studies were extended to cattle and in the '8OS
attention was given to lactating dairy cows. It should be appreciated that there is
relatively limited information in the literature as to normal values in dairy cattle. Gerloff
et al. (1986) were one of the ﬁrst groups to investigate NEFA levels in the development
of fatty liver. Further studies have been published by Studer et al. (1993) and Grummer
(1993)

Plasma NEFA assays have been typically a research tool and one of the questions
to be addressed in this ﬁeld investigation is whether this variable has potential for use in
general ﬁeld investigations of negative energy balance, hepatic lipidosis and their
sequelae of periparturient problems including abomasal displacement.

Based on the limited data available there appears to be a standard pattern in the
NEFA proﬁle for cows in late gestation and the periparturient period. Typically cows
have levels that are under 300 qu/l through the dry period. In the last week of gestation
NEFA levels start to rise possibly indicating the increased caloric demands of the fetus
and ampliﬁed by the usual decline in DMI that occurs prior to calving. In the 24 to 48

hours prior to calving there is a dramatic rise in NEF A levels up to 800 qu/l followed

27

by a precipitous drop over the next 48 hours. This trend is documented in different
studies (Blum et al., 1983; Vazquez-Anon, 1994; Simmons, 1993; Yousif, 1995).

More research is needed to clarify what levels are associated with abnormalities.
NEFA levels above 300 qu/l in the dry period are considered to be indicative of a
negative energy balance. This is based on data from different studies (Blum et al., 1983;
Vazquez-Anon, 1994; Simmons, 1993; Yousif, 1995; Herdt, personal communication) in
which cows generally have NEFA concentrations below 300 qu/l until approximately 48
hours before parturition when concentrations start to increase. Gerloff et al. (1986)
showed that cows that were classiﬁed as having severe hepatic lipidosis had NEFA levels
above 400 qu/l greater than 3 weeks from calving.
WWW

The three factors of importance include: (1) mass of fat stores, (2) energy
imbalance (starting sooner with respect to calving or of greater magnitude than normal)
and (3) stress.

Mass of fat stores

Generally, fatty liver is associated with fat cows and hence the name fat cow
syndrome. It is known that epinephrine is the most potent lipolytic agent in ruminants
while norepinephrine is about 80% as potent as epinephrine (Brockman and Laarveld,
1986). It has been demonstrated that increased beta-adrenergic receptors per adipose cell
appear to be associated with the mobilization of greater amounts of glycerol and free fatty
acids to meet energy needs associated early lactation (J aster and Wegner, 1981). It seems
possible that the amount of fat mobilized may be proportional to the mass of fat present

(Bruss, 1993). Another possible reason for this syndrome being particularly associated

28

with fat cows is the fact that fat cows (BCS >3.5) have been shown to eat less in the post
partum period than thinner animals (Garnsworthy and Topps, 1982), but research is still
required to see if this applies to the prepartum period. It is interesting that Zamet et al.
(1979) indicated that either predisposition or possibly pre-clinical signs for peripartum
disease may be reﬂected in reduced dry matter intake (DMI) prepartum.

The degree of energy imbalance

Fat cow syndrome, although rare in thin animals, has been observed in animals
not necessarily classiﬁed as obese. An explanation for high NEF A levels in these
animals is that they may have been in a greater degree of negative energy balance or for
longer than is normal (Bruss ,1993).

Explanations are - inadequate rations or rations that do not get into the animal e. g.
competition at the feed bunk, illnesses - lameness. The signiﬁcance of adequate dietary
energy is highlighted in a study that showed that prepartum diets even more nutrient
dense than recommended by National Research Council (NRC) reduced fat mobilization
prepartum (VandeHaar, 1995). Other factors not related to disease include - twin
pregnanciesz— there is greater energy demand in late pregnancy and these cows experience
an earlier and greater drop in DMI as calving approaches (Van Saun and Sniffen, 1995)
possibly related to reduced rumen capacity. Season - e. g. severe heat stress in June/July
is associated with signiﬁcant fall off in DMI (Zamet et al., 1979).

Finally there is a large variation in the degree of DM1 depression seen amongst
cows in the last 2 weeks or so of gestation (Grummer, 1993). Generally a gradual decline
in feed intake starts about 3 weeks prepartum becoming pronounced in the ﬁnal week of

gestation (Bertics et al., 1992; Zamet et al., 1979). An experimental study reported by

29

Bertics et al. (1992) illustrates the impact variation in DMI can have on the degree of
hepatic lipidosis. In this study there were 2 groups of cows - 11 cows which had ad lib
intake (control), and 11 cows maintained at the same level of DM1 recorded during day
21 to day 17 prior to calving by force feeding refusals via rumen cannulas. In the control
cows DMI decreased by 28% over ﬁnal 17 days and liver lipid increased 227% in the
control cows versus 75% in the force fed group when comparing levels on day l
postcalving to levels 17 days prior to calving.

Stress

Finally, the third reason for elevated NEFAs is stress. Understanding the critical
role of the sympathetic nervous system in fat mobilization (Frohli and Blum, 1988)
allows an appreciation of the impact of stress on NEFA levels. Gerloff et al. (1986)
mentioned the importance of stress in the periparturient period and common examples are
dystocia, new grouping, lack of calving pen, high environmental heat stress. In a genetics
experiment that studied the lipolytic adaptations of bovine adipose tissue during late
pregnancy, lactation, and dry period in Holstein heifers, there was indication of a genetic
component in adrenergic regulation of lipolysis in adipose tissue, independent of energy
balance in supporting lactation. Basal glycerol release in high genetic merit heifers was
consistently higher than in low genetic merit heifers. In addition the same results were
obtained in response to norepinephrine and epinephrine (McNamara and Hillers, 1986).
This would indicate that stress would have a greater effect on fat mobilization in cows of

higher PTAmilk.

30

hmpnmLfaatQLﬂhaLaffcaLnlasmal‘lEEAmntraticns

Time period relative to calving

As discussed earlier, cows sampled very close to calving will have dramatically
elevated plasma NEF A concentrations which may be normal.

Diurnal ﬂuctuations

Earlier research has demonstrated the signiﬁcant effect feeding time has on
plasma NEFA concentrations. This effect is more pronounced when animals are fed once
a day, compared animals having continual access to feed. What is seen is that NEF A
levels may drop by more than 50% in 2 to 4 hours after feeding and then rise steadily
back over six hours or more (Radloff et al., 1966; Bines and Hart, 1977). In a study
looking at the effect of three rations of different energy density on plasma NEFA levels it
was found that diurnal variation in plasma concentrations was of greater importance than
the variation resulting from a 76% difference in DM intake and a 71% difference in
energy status (Coggins and Field, 1976). Hence cows should not be sampled soon after
feeding; this should rather be delayed by several hours and taken at a ﬁxed time after
feeding (Coggins and Field, 1977).

Eﬂect of stress

Researches have pointed out that NEFA levels are very stress sensitive (Lindsay,
1977). A 50 to 150% increase in NEF A levels has been associated with excitement from
handling (Holmes and Lamboume, 1976).

Sample handling

Blood samples taken for plasma NEF A assay have to be handled correctly. This

involves placing the samples on ice for transport to the lab. At warm temperatures

31

signiﬁcant hydrolysis of TG in VLDL can occur elevating NEFA levels (Herdt, personal
communication).
in. . H. . a“ . ‘1‘. ., f .3 . .11. . .11., u ..

Hepatic lipidosis is commonly called fatty liver or fat cow syndrome. As the
latter name states, this is a clinical syndrome usually observed in the immediate
postpartum period and is associated with severe fat inﬁltration of the liver ( Morrow,
1976). As a syndrome it can present in a number of different ways. Typically an obese
cow is presented for one or more of the typical periparttuient disorders that are common
in dairy cows - milk fever, mastitis, metritis, DA. However, in these cows there is
typically a poor response to treatment, and the cow is more depressed and anorectic than
expected for the given disorder. There is always ketosis present, and the prognosis is
guarded to poor (Morrow ,1976; Herdt, 1988).

The significance of hepatic lipidosis to the dairy industry

This can be addressed by studying the disease incidence and outcome in dairy
herds. There are limited references in the literature on this point and reference will be
made to three in this area. Reid (1980) looked at the incidence of fatty liver in a 109 high
yielding Freisian cows in the United Kingdom (UK) and Gerloff et al. (1986) examined
80 cows in 9 herds that were investigated for possible fatty liver problems in Michigan.

Reid evaluated, by quantitative morphology, the percentage (%) fat in liver cells
one week after calving. Gerloff et al.(1986), using gravimetric methods, looked at
maximum triglyceride concentrations in liver samples between one month before to one
month after calving. In both studies they classiﬁed the cows into 3 groups - mild,

moderate and severe based on the degree of hepatic lipidosis. The three classiﬁcations of

32

mild, moderate and severe hepatic lipidosis are fortunately comparable across the two
studies as liver fat content determined histologically as done by Reid is correlated with
liver TG content measured chemically as done by Gerloff. Of the 109 high-yielding cows
Reid examined, 37% were classiﬁed with mild hepatic lipidosis, 48% were moderate and
15% had severe fatty inﬁltration. In Gerloffs study of 80 cows 65% were mild, 20%
moderate and 15% were severe. Gerloff went further in his study and looked at death and
cull rates due to disease in the 3 groups of cows. In the mild group 15%, in the moderate
group 31%, and in the severe group 42% died or were culled due to disease. In a study
involving 49 pluriparous cows van Dijk et al. (1989) found 16 cows (33%) with severe
hepatic lipidosis, 13 cows (26%) moderate, and 20 cows (41%) with mild hepatic
lipidosis. They found the severe group to be related to impaired fertility and a high
incidence of periparturient disease. Reid and Roberts (1983) estimate that about one-
third of high yielding cows in the UK are affected by subclinical fatty liver.

Thus, based on these and other studies in the literature, there is a relatively high
incidence of the disease in dairy herds and that the risk of morbidity in general, including
DA, appears proportional to liver TG levels.

Prevention of Hepatic Lipidosis

The treatment of hepatic lipidosis is aimed at decreasing fat mobilization,
decreasing ketone formation, stimulating appetite and treatment of concurrent disease. As
mentioned earlier, cows with clinical fatty liver generally have a poor response to therapy
which is usually prolonged, labor intensive and expensive. Hence it is sensible to focus
ones attention more on prevention than treatment. From the above review, a number of

preventative measures can be drawn and are listed as follows:-

33

1) Maximize DMI throughout late gestation
2) Proper ration formulation
3) Close monitoring of health
4) Avoid over-conditioned animals
5) Avoid stressful environmental conditions
6) Propylene glycol (PG) administration in late gestation
7) Assessment of energy balance for early correction of a negative balance

A study by Studer et al. (1993) illustrates the potential beneﬁt of one therapy that
could be instituted in the dry period. In this experiment there were 2 groups of cows - one
group of 13 cows were given 1 liter oral drench of PG once daily starting approximately 2
weeks prepartum until the day of calving; the other group of 11 cows formed the control
group and received a 1 liter water drench. The average NEFA concentrations (qu/l) in
the week prepartum were 403 for the control group and 234 for the treated group. At day
one after calving the liver TG% in the control group was 14.5 and for the propylene
glycol group 5.0 (p <.01). In conclusion, this study illustrates how an understanding of
the pathogenesis of hepatic lipidosis with an appreciation of the chronology and the
importance of plasma NEFA levels, opens up opportunity for early intervention that may
avoid or reduce the severity of the condition and its sequelae.

Hepatic lipidosis - a potential risk factor for abomasal displacement

The literature suggests an association between hepatic lipidosis and abomasal
displacement (Holtenius and Niskanen, 1985; Breukink, 1990). It is now understood that
a negative energy balance is the initiating factor in the development of both ketosis and

fatty liver disease and that this imbalance may frequently start in the prepartum period

34

(Gerloff et al., 1986; Grummer, 1993). It seems logical to consider that a negative energy
balance in the prepartum period, as determined by elevated plasma NEFA levels, may
show an association with the occurrence of abomasal displacement. Further investigation
of this association in the ﬁeld may determine its signiﬁcance as a risk factor and
underscore the importance of the practical, preventative measures discussed for hepatic

lipidosis when considering preventative measures for abomasal displacement.

MATERIALS AND METHODS

The Data
Selactinnnffanns

Selection of farms was based on two criteria - being a member of the Michigan
Dairy Herd Improvement Association (DHIA) and having a current herd average above
the Michigan DHIA average of 8760 kg (19300 lb) of milk per cow per year. Three
hundred farms met these criteria of which 118 responded to a questionnaire sent out
soliciting involvement in the study. Of the 118 farms that responded, 104 farms were
visited between October 27, 1993, and January 25, 1995.

A total of 67 farms (1170 cows) were used in this study. The reasons for deleting
the farms (3 7) included incomplete data (20) (failure to complete the health sheets (6),
inadequately deﬁned rations (7), incomplete management data (7) ) and inadequate cow
numbers (13). A total of 1737 cows (465 primiparous and 1272 multiparous) were
sampled in the 67 herds. Primiparous cows and cows that gave birth to twins were
deleted from the analysis.
Eannxisits

Responsibility for collecting and recording data on the assigned farms was done
by 4 investigators among which the 104 farms were divided. Each farm was visited 4

times within a 6 week period with a minimum of 6 days between each visit.

35

36

Herd data

On the ﬁrst visit, nutrition and management data were recorded on standardized
sheets and at the second, third and fourth visit, changes in nutrition or management were
recorded.

A TMR is deﬁned as a total mixed ration such that all the dietary components
being fed are mixed together before being fed. In this study, if any dietary component
was fed separately then the ration was not classiﬁed as a TMR A transition diet was
deﬁned as a ration that is fed to dry cows for at least the last week of gestation and which
differed in content from what the cows were fed in the earlier part of the dry period.

It was not feasible in this study to conduct feed analysis of the rations on each
farm. Hence reliance was placed on ration data supplied by the farmer. In cases where
there was no feed analysis available, book values (NRC) were used to describe the feed
examined. Estimates of feed intake were available on some farms but in the majority of
cases this was difﬁcult to predict with accuracy. In cases where intakes were unknown it
was assumed that the average cow's dry matter intake (DMI) in late gestation was 10kg
per day. When the amount of concentrate fed was known, the amount of forage
consumed was estimated by the difference between the concentrate dry matter and the
potential DMI.

Bunk management with respect to late gestation cows was evaluated. Length of
bunk space, how frequently the bunks were cleaned and for how long feed was available
was recorded. A penalty score (0-2) was devised for bunk management and is described

in the Appendix.

37

An estimate of herd ME milk production was obtained from the mean of herd ME
milk production on DHIA test days between 1 and 150 days after the ﬁrst visit to the
farm. This variable was collected as an indicator variable as opposed to a risk factor.

Cow data

On the ﬁrst visit and subsequent visits the following was collected from all
Holstein cows and heifers within 5 weeks of calving:- identiﬁcation number, body
condition score and a blood sample collected in an EDTA tube from a coccygeal vessel.
No effort to classify the animal with regards to parity was done at this stage as these data
along with the animal's PTAmilk index were to be obtained from DHIA records once the
animal had calved. PTAmilk is a value derived from the animal model which predicts
genetic merit of each animal in a population from the animal's own performance (if
available) and the production records of all related animals. The Predicted Transmitting
Ability (PTA) is a one-half breeding value, adjusted so that cows born in 1985 average 0
(Wiggans and van Raden, 1989).

At the second, third and fourth visit all animals within 5 weeks of expected date
of calving were identiﬁed, body condition scored and blood sampled, including animals
from the previous visit that had not yet calved.

Before the start of the ﬁeld investigation, the investigators standardized their body
condition scoring technique using a chart from Elanco Products Company (Indianapolis,
IN, USA). Body condition scoring was done on a scale of 1 to 5 (1/2 point increments),

with 1 being emaciated to 5 being severely obese (Wildman et al., 1982).

38

The blood samples for plasma NEFA analysis were put into ice within 15
minutes, transported back to the laboratory, centrifuged at 3000 rpm for 15 minutes and
the plasma then placed into labeled plastic tubes and stored at - 20 °C. The plasma
samples obtained were analyzed for NEF A using a commercial kit (N EF A-C kit, Wako
Chemicals USA, Richmond, VA) with modiﬁcations by McCutcheon and Bauman
(1986), Sechen et al. (1993), Johnson and Peters (1993), Dyk (1995).

Heifers were excluded from the analysis because they represented another
population which in addition often differed in management from the other cows in the
herd. Cows that gave birth to twins were excluded from the analysis as in terms of NEFA
levels, abomasal displacement and other diseases, they represented another subset of the
population.

Cow model

A health sheet was left with the farmer at the end of the ﬁrst visit to record health
information on the cows sampled after they had calved. From this sheet the occurrence of
DA in each animal sampled was determined.

Herd model

In this model a lactational incidence rate for DA at each farm was calculated
based on the animals sampled at the farm visits. However, as heifers were excluded from
the analysis in both models, the number of multiparous cows initially identiﬁed was
limiting in the smaller herds and in herds that had few cows over the study interval. One

approach would be to eliminate herds with small nrunbers but this would bias the study in

39

favor of larger herds. The minimum number of multiparous cows that had to be
considered in assessing the occurrence of DA was chosen after considering that the
National Animal Health Monitoring System (NAHMS) uses a minimum of 10 animals
for their herd disease estimates (Kaneene and Hurd, 1990). The inclusion criteria set for
this study were that a minimum of 20 consecutive multiparous cows with no DA had to
be considered for a zero occurrence to be assigned to the herd. A herd could be included
if one or more DAs were reported for less than 20 cows (9 cows was the lowest number)
but not if no DAs had been reported. This was done to improve the conﬁdence in
classifying herds as low incidence. It is appreciated that the incidence estimate between
herds may be affected in some cases by the difference in population size but this concern
diminishes when differences in incidence between herds is large. It is accepted that the
estimate of incidence in this dry cow population sampled may differ signiﬁcantly from
the annual incidence rate obtained for the same herd, however, the estimate obtained is
not being used for this inference.

In order to increase the number of cows on which the DA frequency estimate for
the herd was based, 44 of the herds in this study were approached retrospectively to
create an expanded data base on which DA incidence rates could be calculated. This will
be referred to as the expanded data set as opposed to the original data set. Using DHIA
records, additional cows that calved directly following those that were sampled, were
identiﬁed. The farmer supplied data on whether the cows experienced a DA during the

lactation and, in addition, if the cow had a DA, whether the cow had had twins.

40

Statistical analysis

The analysis was done at two levels - a herd level and a cow level. The cow-
model considered risk factors for abomasal displacement speciﬁc to the individual and
used a logistic regression analysis. The herd model considered potential risk factors
operating at the herd level and involved a multiple regression analysis.

Herd level variables

The risk factors considered in the herd model are listed in Table 1. Where
applicable a further description of these variables can be found in the Appendix.

The outcome variable was the herd incidence rate of abomasal displacement in the
expanded data set. Each herd's DA incidence rate (DAIR) was calculated as in equation
1.

Equation 1:

DAIR = No. affected with DA
No. recorded calvings

 

Cow level variables
The risk factors considered in the cow model are listed in Table 2. The outcome
variable was categorical (1/0) - the presence or absence of DA - in each cow. Table 3

shows the outcome variables for both models and their coding.

41

Table l
Herd-based risk factors used in analysis of DA

 

Risk Factor Coding Description

 

season of visit 1-4 months (1-12) categorized: 1) 3 4 5 2) 6 7 8 3) 9 10 11
4) 12 1 2
bcs 1 1-5 herd average body condition score 35 to 1 day prepartum
N3 001 0-.93 herd incidence of NEFA concentrations above 300 qu/ l
in cows sampled 35 to 3 days prepartum
hdmilk 20351- mean of herd ME milk from ﬁrst visit to 150 days (lb)
30296 -
lac 2-4.6 average lactation number of cows sampled
ptamilk 4303- average PTAmilk index (PTAmilk < -200 coded as
1795.8 missing values) for herd, based on cows sampled

Feed Management Risk Factors

bunk 0-2 combined score for bunk space and feed availability and
freshness (0-2)

d_trat 0-30 days transition ration fed - continuous

trrrr 0/1 total mixed ration fed precalving (l = yes, 0 = other)

trat 0/ 1 transition ration fed (1 = yes, 0 = no)

Feed Risk Factors

cmcpr O/ 1 precalving ration energy density categorized: if >1 .65

NEl Meal/kg DM cmcpr = 1, if = or <1 .65 NEl Meal/kg
DM then cmcpr = 0

denerg 1-5 change in concentrate feeding before & around calving,
categorical : 1(mild) - 5(severe)

dforage 1-5 change in forage type & relative amount around calving,
categorical: 1(mild) - 5(severe)

dnfcb -5-20 difference in non ﬁber carbohydrate (NFC) % between
ration precalving and postcalving

hay 0-79 % hay in ration precalving

mcd 1.21-1.61 NEl Meal/kg DM of dry cow ration

ndﬂ 15-46 forage - neutral detergent ﬁber % of lactating ration

ndft 16-53 forage - neutral detergent ﬁber % of prepartum diet

Disease risk factor
mf 0-.33 herd incidence of milk fever based on cows sampled

 

42

 

Description

 

average body condition score 35 to 1 day prepartum for
individual cow

lactation number

pta milk index (pta < -200 coded as missing values)

Table 2

Cow -based risk factors used in analysis of DA
Risk Factor Coding

Descriptive and NEFA Factors

bcs 1 1-5

lac 2-9

ptamilk -151-2770

N300] O/ 1

Disease risk factors

dyst
mf
1'13

0/1
0/1
0/1

if cow had a NEF A concentration above 300 qu/ l
sampled 35 to 3 days prepartum then N300] = 1

case of dystocia
case of milk fever
case of retained placenta

43

 

 

Table 3
Outcome variables used in analysis of DA
Variable Coding Contents
DA 0/1 Displaced abomasum in observed cow

DAIR -2.94 - -0.49 Logit transformed DA incidence rate (herd)

 

44

Cow model

This was approached by ﬁrst generating univariate random effects logistic
regression models for each of the risk factors using the variable DA (coded as yes or no)
as the outcome (Statistics and Epidemiology Research Corporation and Cytel Software
Corporation, 1991). All risk factors were included in the multivariable logistic regression
model with random effects. This approach of using the random effects accounts for
possible lack of independence of individual animals in the same herd. A backwards
model building procedure was used to develop the ﬁnal reduced model in which risk
factors had to have a signiﬁcant association with the outcome at p <.05. The correlation
matrices for both the initial data set and the ﬁnal model were checked for potential
collinearity among the independent variables by means of evaluating the correlation
coefﬁcients between the independent variables. Based on an acceptable limit of .6 (p
<.05) for a correlation coefﬁcient, no collinearity was detected.

Herd model

The risk factors that were initially selected from the complete data set were
chosen to reﬂect factors which had previously been associated with the occurrence of
DA. In addition, management factors were also considered. As the outcome variable
DAIR was not normally distributed, a logit transformation was performed to improve the

distribution. The formula for the logit transformation is seen in equation 2.

45

Equation 2:

DAIR + 0.05
l — (DAIR + 0.05)

 

logit transformation 2 log

Univariate regression models were developed for each of the risk factors.
Collinearity diagnostics were run during the regression process ( Statistical Analysis
Systems Institute Inc., 1985) by means of evaluating the correlation coefﬁcients between
the independent variables. Based on an acceptable limit of .6 (p <.05) for a correlation
coefficient, no collinearity was detected. A multivariable linear regression model was
developed using a backwards model building procedure in which the remaining variables
had p <.05. Collinearity within this model was evaluated by inspection of variance
inﬂation factors and Eigenvalues. No collinearity was detected. The plot of residuals for
the ﬁnal model showed the residuals to be normally distributed, hence indicating that the
normality requirement for linear modeling was upheld.

Results

Lactational incidence rate for DA based on lactation number for all cows sampled

in the study is shown in Table 4. The DA incidence rate for individuals with twins is

shown in Table 5.

46

Table 4
DA incidence for all cows (excluding those cows with twins)

 

Lactation No. Number of cows Number with DA (%)

 

1 456 29 (6)
2 509 37 (7)
3 325 26 (8)
4 171 10 (6)
5 82 5 (6)
6 53 2 (4)
2 and greater 1 170 80 (7)

 

47

Table 5
DA incidence for cows with twins

 

Lactation No. Number of cows Number with DA (%)

 

1 9 1(11)
2 and greater 102 12 (12)

 

48

Wrists

Table 6 shows a break down of risk factors based on cows” DA status and Table 7
shows the ﬁnal cow-based multivariate model.

Signiﬁcant factors in the cow-based model associated with an increased risk of
DA included increased body condition score, winter season and a plasma NEFA level
greater than 300 qu/l between 35 to 3 days prepartum. An increasing lactation number
brought about a reduction in risk of DA.
Herdiasedanalxsis

Table 8 shows the herd-based multivariate model for the logit transformed
incidence rate of DA. Signiﬁcant factors which were associated with increased risk of
DA in herds included increased PTAmilk, increased body condition score, winter,
summer, and rations with energy densities higher than 1.65 Mcal/kg NEl DM fed before
calving. Having adequate bunk space together with fresh feed always available was

signiﬁcantly associated with a reduced risk of DA.

49

Table 6

Breakdown of risk factors based on cows’ DA status

 

 

 

Risk Factor Mean observations for cows
grouped by DA status
With DA Without DA
herd size 260.67 253.96
lactation number 2.9] 3.11
body condition score 3.54 3.32
NEFA concentration above 300 qu/l 35 to 3 days .53 .35
prepartum
PTAmilk index 1 141 1 1 14
bunk management penalty score (0-2) 1.07 .93
days transition ration fed 13.05 13.09
total mixed ration (0/1) .68 .59
transition ration (0/1) .83 .85
change in concentrate feeding (1-5) 2.75 2.78
change in forage type and amount (1-5) 2.99 2.77
difference in non ﬁber carbohydrate (NFC %) 2.40 3.25
between ration precalving and post-calving
% hay in ration precalving 11.85 14.75
NE] Meal/kg DM of dry cow ration 1.44 1.4]
forage - neutral detergent ﬁber % of transition diet 27.41 29.30
forage - neutral detergent ﬁber % of lactating ration 20.14 20.57
post calving
occurrence of dystocia .037 .034
occurrence of milk fever .099 .093
occurrence of retained placenta .123 .113

 

50

Table 7
Cow-based multivariate random effects logistic models for the occurrence / non-
occurrence of DA

 

 

Variable Coefficient P Odds ratio 95% CI
Grand Mean -5.694 <.001 .3365E - 02 .497E - 03 - .228E-01
Body condition score .8777 .001 2.405 1.41 - 4.10
Winter season 1.087 .002 2.967 1.48 - 5.96
Elevated NEFAs .7139 .007 2.042 1.22 - 3.42
Lactation number -.2l38 .045 .8075 .66 - 1.0
%SCL .6524

 

Overall model deviance with 1077 degrees of freedom: 519.530
Likelihood ratio statistic for random effects term: 10.9 (p <.283 )

51

Table 8
Herd-based multivariate linear regression model using logit transformed DA incidence
rate

 

 

Variable Coefﬁcient F P
Intercept -5.30 25.06 .0001
PTAmilk .0006 7.27 .0091
Optimal bunk management -.343 4.69 .0343
Summer season .339 4.67 .0347
Winter season .632 10.28 .0022
NE] > 1.65Mcal / Kg DM .388 4.13 .0465
Body condition score .650 5.21 .0260

 

Model R2 = .344, F = 5.25 with 66 d.f. (p <.002)

DISCUSSION

The lactational incidence rate for DA in this study was 7% for the 1170
multiparous cows. Although primiparous cows were excluded from the data analysis, it is
interesting to note that the incidence rate in this parity based on 465 animals was 6.4%.
These rates are higher than what is reported in the literature. The highest lactational
incidence rate previously reported in the observational studies reviewed was 5.0%. This
was also the most recent report (Pehrson and Shaver, 1992).

Care must be exercised in interpreting these rates relative to those in the literature
as in the majority of literature reports it is the lactational incidence rate reported over a
year or more. As has been reported and observed in this study, there is a distinct
association between season and incidence of DA. Hence it is quite probable that the
incidence rate reported here is higher partly as a result of the time of year when the
incidence rates were recorded. Furthermore, the herds studied in this investigation had
the state average or higher level of milk production which will likely contrast from the
average population on which some studies were based. However, the trend in the
literature of increasing incidence with each new report is supported by these results. The

trend of increasing incidence with increasing parity that has been reported (Martin, 1972;

52

53

Erb and Martin, 1978; Constable, 1992) is not supported by this study and in fact an
opposite trend is observed. This issue is addressed later in the discussion.
w e]

Elevated plasma NEFA concentrations prepartum

In the cow-based model increased plasma NEFA levels prepartum were
signiﬁcantly associated with the occurrence of abomasal displacement. Increased plasma
NEFA concentrations in the prepartum period have been demonstrated to be important in
the development of hepatic lipidosis (Gerloff et al., 1986; van Dijk et al., 1989;
Grummer, 1993). There are in turn studies that have shown the degree of hepatic
lipidosis to be associated with the occurrence of periparturient problems (Gerloff et al.,
1986; van Dijk et al., 1989), including DA (Holtenius and Niskanen, 1990). Research
indicates an association between the occurrence of ketosis and the presence of hepatic
lipidosis (Grummer, 1993). Ketosis (diagnosed prior to DA occurrence) has been
implicated as a risk factor for DA in several studies (Curtis et al., 1985; Varden, 1979;
Robb et al., 1987). These ﬁndings support the opinion of Holtenius and Niskanen (1985)
and Breukink (1991) that hepatic lipidosis may have a signiﬁcant role in the pathogenesis
of DA.

Two different reports in the literature, one by Meirhaeghe and co-workers (1988)
the other by Holtenius (1992), put forward another possible explanation that may link the
occurrence of DA to hepatic lipidosis. Meirhaeghe and co-workers (1988) found that
animals with abomasal displacement showed higher basal blood glucose levels than

control animals, independent of their blood ketone body concentration. They also

54

observed a delay in glucose clearance from the blood of cows with DA, despite elevated
insulin levels, pointing towards the possibility of an insensitivity of body tissues to
insulin. The same authors found in a study of four ﬁstulated heifers, that endogenous
insulin, secreted after intravenous glucose infusion, as well as exogenously injected
insulin, caused a marked decrease of abomasal emptying rate lasting several hours. They
suggested that insulin itself, independent of hyper- or hypoglycemia, might be a major
cause of reduced abomasal motility and inhibition of abomasal emptying. DeCupere et al.
(1991) reported that cows (n = 38) with left displaced abomasum exhibited insulin
resistance which differed from both non-ketotic (n = 33) and ketotic (n = 7) control
animals. This is further supported by Holtenius and Traven (1990) who showed impaired
glucose tolerance and heterogeneity of insulin responses in studies with 36 cows with
abomasal displacement, indicating insulin resistance and signs of non-insulin-dependent
diabetes mellitus. Causes for the insulin resistance are not clear but lipolysis, ketone
bodies and the degree of body fatness may play a role. Holtenius (1992) made the
observation that various degrees of insulin resistance are associated with moderate to
severe fatty livers.

Another factor to consider is that ketosis is recognized as having a negative effect
on DMI. It is possible that poor DMI may reduce motility of the forestomachs due to
reduced stimulation of stretch receptors and in turn lead to a reduction in abomasal

motility.

55

Body condition score

In both the cow-based and herd-based models there was a signiﬁcant association
between body condition score and displaced abomasum with increased body condition
score associated with increased risk. To my knowledge, this is the ﬁrst time this
association has been observed in a ﬁeld study. In a study of 429 Holstein cows, body
condition at calving was not associated with an increased incidence of disease (Ruegg
and Milton, 1995). However, there are other reports that indicate an association between
high body condition score and periparturient disease in general. In trials involving cows
that belonged to either a control (n = 22) or overconditioned (n = 22) group, the total
number of cases of periparturient diseases were 31 in the overconditioned and 21 in the
control during the ﬁrst 11 weeks postpartum (Fronk et al., 1980). Treacher et al. (1981)
reported that cows calving in very good condition (Body condition score 4 or greater),
developed more severe fatty liver, lost more weight and were more susceptible to disease
and produced less milk than leaner cows. Shirley (1994) reported that in 35 heifers fed to
a body condition score of 4.0, on a 1 to 5 scale, and maintained at that condition for the
last 60 days before calving experienced a high incidence of subclinical ketosis and
abomasal displacement. Of the 35 primiparous cows, 17 experienced a displaced
abomasum within the ﬁrst 30 days after calving.

A pathophysiological explanation for this association between BCS and DA is
likely based on a negative energy balance. There is evidence that fatter animals eat less
(Garnsworthy and Topps, 1982), and have greater negative energy balance in early

lactation that is supported by high NEF A concentrations (Fronk et al., 1980). Elevated

56

NEFA concentrations increase the degree of hepatic lipidosis (Bertics et al., 1992; Studer
et al., 1993; Grummer, 1993). Researchers (Gerloff et al., 1986; van Dijk et al., 1989)
have shown the impact of hepatic lipidosis on increased periparturient disease. Further
support for this association is shown in this study by the signiﬁcant association between
elevated NEFA levels prepartum and the incidence of DA in the cow-model. The large
sample size and controlling for many variables may have allowed the detection of an ’

association between BCS and DA, in contrast to previous studies.

.: em--m!l. I 'A‘ ..

Season
Season was a signiﬁcant risk factor in both models and their results will be I

discussed together. Winter (December, January, February) in the cow-model and winter

and summer (June, July, August) in the herd model were associated with an increased

occurrence of DA. Similar frndings for a seasonal association with DA incidence have

been reported (Robertson, 1968; Pinsent, 1961; Constable, 1992). Although no theory to

explain this association has been proven it is interesting to consider whether a negative

energy balance may be involved. It is known that high temperatures reduce dry matter

intake in summer (Zamet et al., 1979), which may possibly lead to a greater degree of

hepatic lipidosis and a smaller rumen volume at calving. Cold temperatures in winter

increase the energy maintenance requirements of cows and extreme temperatures are

likely a period of increased stress. A negative energy balance in the cow model is

associated with DA occurrence. Management and dietary factors which restrict energy

intake in late gestation will have an even more profound affect in winter when

maintenance energy requirements are increased. In addition feed intake may be further

57

compromised especially in herds with exposed bunks due to feed being frozen or covered
in snow and the cows may have been less willing to stand to eat in ﬁill exposure to the
elements.

Increasing lactation number

Increasing lactation number was associated with a reduction in risk for the
occurrence of DA. This ﬁnding is opposite those reported in the literature (Martin,1972;
Erb and Martin, 1978; Constable, 1992), while Dohoo and Martin (1984) were unable to
detect a consistent trend. There are several aspects to consider for the different trend seen
in this study. First of all, primiparous cows were excluded from this analysis and had
they been included a different result may have been obtained. Secondly, although heifers
were not included in either model, a very high incidence of DA in the primiparous cows
was present in this study which contrasts sharply with the low incidence in most of the
previous studies. A cow that has had a DA surgically corrected is generally unlikely to
redisplace in subsequent lactations. Thus, the population at risk in higher parities is
reduced by the presence of previously treated cows and this change in population at risk,
not accounted for in this investigation, may have had a signiﬁcant impact especially since
differences in incidence rate between parities is small. In addition it would be interesting
to compare the age distribution of the population in this study with those in previous
observational studies. The impression from looking at the distribution of cows among
parities in our study, compared to other published reports, is that herds in our study were
made up of younger animals. Thus, it is possible that there was a higher degree of culling

in our herds which may also have a signiﬁcant effect on the population at risk, when one

”731'

58

considers that cows predisposed to problems are removed earlier leaving behind a more
‘resistant’ population.
Herdmodcl

PTAmilk

PTAmilk is a value derived from the animal model which predicts genetic merit
of each animal in a population from the animal's own performance (if available) and the
production records of all related animals. In the herd-based model, a higher average
PTAmilk was found to be positively associated with the incidence of DA. This
association has not been reported in other observational studies. However, support for this
association in the literature can be found in Shanks et al., (1978), and Jones et al., (1994).

Shanks et al. (1978) in a study with 43 pairs of heifers found that high pedigree
cows produced more milk but had a higher incidence of disease compared to low
pedigree cows. It is interesting to note that, although not statistically signiﬁcant, the high
pedigree group in their study had twice the number (6 vs. 3) displaced abomasums as did
the low group. Similar ﬁndings were reported in a selection experiment conducted over
16 years that compared one group of cows that were bred to the highest PTAmilk sires to
another unselected control group. Single-trait selection for milk yield resulted in
increases of health expenses as a correlated response. Increases in total health costs
during the 16 year study were primarily due to reproduction, digestion and ketosis (Jones
et al., 1994).

Care should be taken in the interpretation of the association between PTAmilk

and DA occurrence. Although herds with higher average PTAmilk were associated with

59

a higher incidence of DA it may not necessarily mean that the higher PTAmilk cows with
in herds were the ones that experienced DA. As herd PTAmilk is inﬂuenced by
management selection of sires and dams with greater milking potential, it is possible that
the occurrence of DA may be associated with some correlated aspects of management,
possibly concerned with maximizing milk production, undeﬁned in this study.

The signiﬁcance of PTAmilk may indicate that increased milk production is a risk
factor for DA. However, no association between DA occurrence and the mean herd ME
milk production (between the ﬁrst visit and 150 days later) considered in the initial herd
model was found. A similar lack of association was found in multivariate analysis by
Curtis and co-workers (1985) and Dohoo and Martin (1984). Caution must be given to
the lack of association between DA incidence and herd ME milk in our study as a much
larger population is considered in the herd ME milk production compared to the
population on which the DA incidence estimate was based in this study. Given all the
numerous factors that can potentially inﬂuence herd ME milk production, perhaps mean
herd PTAmilk is a more legitimate estimate of a herd’s potential milk production and, as
this study reveals, this might be where a signiﬁcant association between DA occmrence
and milk production is present. Further evaluation of this association is indicated. It is
interesting to consider that, with the extensive use of Al, the average herd PTAmilk is
likely to have shown a steady increase over time and this in part may explain the
comparable trend of increasing incidence of DA observable in the literature.

One has to be cautious of applying PTAmilk at the individual cow level as our

models give no support for this application. However, if PTAmilk operates at the

60

individual cow level there is research that may give an explanation for this. A report by
McNamara and Hillers (1986), indicated a genetic component in adrenergic regulation of
lipolysis in adipose tissue, independent of energy balance, in supporting lactation. Cows
with high genetic merit showed higher rates of lipolysis measured at 30 and 15 days
before parturition and this higher rate was maintained independent of energy balance
during lactation. Apart from a genetic component, it also seems probable that the amount
of fat mobilized may be relative to the mass of fat stores present (Bruss, 1993).

If having a higher PTAmilk is associated with a greater ability to mobilize fat in
the prepartum, then an association between PTAmilk and elevated NEFA levels seems
logical. Furthermore, if higher PTAmilk is combined with heavier body condition score
and suboptimal bunk management there is reason to expect even higher prepartum NEFA
levels. This is not supported by the results of this study but it is an area worthy of ﬁrrther
research.

The aim of most dairy herds is to produce replacement animals of higher genetic
merit than the parent herd. Thus it seems likely that PTAmilk would be increasing in the
direction of lower parity numbers and in turn earlier parities may be more predisposed to
metabolic disorders initiated by a negative energy balance especially if they are well
conditioned. PTAmilk is associated with age, with younger parites having a higher index.
In the cow-model lower lactation numbers were associated with greater risk of DA. If
indeed lower parites are at greater risk this may give support to the association between
DA incidence and PTAmilk. Prirrriparous cows were excluded from the models

constructed in this investigation so applying the results of our models to this group of

'h
I _LI

6]

animals is not legitimate. However, if they do in fact share factors in common with
multiparous cows, a negative energy balance will have a greater effect on NEFA
concentrations in these animals.

Optimum bunk management

In the herd model there was a signiﬁcant association between having optimum
bunk management, and a reduced incidence of DA. These features of bunk management
(adequate bunk space and continuously available fresh feed) would be conducive to
optimum dry matter intake (DM1) and even consumption of feed throughout the day,
allowing for a better nutritional status and greater rumen ﬁll at calving. The signiﬁcant
positive association between elevated NEF A levels and the occurrence of abomasal
displacement shown in the cow-based model gives added weight to the importance of
nutritional status. Svendson (1969) and Constable (1992) discuss the importance of
rumen ﬁll as a moderately effective barrier against displacement of the abomasum. It
seems likely that optimum bunk management would be conducive to maximizing rumen
volume.

It has been shown that cows exhibit a decline in DMI in the ﬁnal days of gestation
that may be the result of the effect of placental estrogens (Forbes, 1987). In an
experiment (unpublished) cited by the later author in which the feeding behavior of four
ewes was monitored from midpregnancy, it was found that meal size decreased and meal
frequency increased with advancing pregnancy. This change in feeding behavior may be a
consequence of decreasing functional rumen capacity as the uterus enlarges. If this is

applicable to cows, then the importance of having fresh feed readily available throughout

 

62

the day can be appreciated. Management of herds that feed restricted amounts to dry
cows, especially in late gestation, may seriously reduce the DMI of cows in advanced
pregnancy. This situation could be further compounded by competition for limited bunk
space.

Prepartum diets greater than 1. 65Mcal NEl per kg

In this investigation it was found that prepartum diets greater than 1.65Mcal NE] r
per kg were associated with increased risk of abomasal displacement. A possible
explanation for this association may be that cows on a high concentrate diet, compared to

a low concentrate diet have a greater decline in DMI before calving as has been observed

 

 

in other studies (Coppock et al., 1972). A possible reason for this decline may be
palatability. There is indication that estrogen, which increases in the ﬁnal days of
gestation, affects dietary selection in ruminants causing a reduction in concentrate intake
but not in forage intake (Forbes, 1987). Hence this might explain the lower intake of
rations high in concentrate and the beneﬁt of rations high in forage. A decreased DMI
compounded by a low forage component may have resulted in a reduced rumen volume
post calving. As discussed, a small rumen volume offers less resistance for abomasal
displacement to occur (Constable, 1992).

There is another explanation to be considered for the relationship between
excessively high energy density and DA occurrence. Cows on a very high plane of
nutrition through out the dry period may become overconditioned. Such animals have
been shown to have a greater degree of negative energy balance in early lactation

possibly as a result of lower DMI. The higher postpartum plasma NEFA levels in these

63

animals can result in hepatic lipidosis (Van Den Top et al., 1995). The association

between hepatic lipidosis and the occurrence of DA is discussed later on.

 

The two models complement each other as the cow-based study allows for risk
factors within herds to be studied which may explain the variation in herd incidence rate
more clearly than between-herd risk factors. This may be particularly true of variables E
that apply to the individual, such as plasma NEFA concentrations. The cow-model treats H

each animal as a separate observation and hence will be more sensitive to these risk

 

wxhﬂ‘b

factors than the herd-based model. However, the herd-based model complements the
cow-based model as it is better at evaluating herd-level management practices. This is an
area of weakness in the cow-level model because even though it groups observations by
herd, the larger herds with more observations will have a greater inﬂuence in the model.

The impact of optimum bunk management, herd average PTAmilk, prepartum
rations greater than 1.65 Mcal NE] per kg dry matter, and summer on the occurrence of
DA would have been missed if only a cow model was studied. Similarly the cow model
added the importance of negative energy balance to the occurrence of DA. The combined
ﬁndings of these two models appear to reinforce each other as discussed, with a common
line through several of the factors of signiﬁcance.

Other dietary factors

No signiﬁcant association was found between prepartum ﬁber levels and the
incidence of DA in both the herd and cow based models. A signiﬁcant shortcoming in a

ﬁeld investigation of this magnitude is the inability to accurately estimate the DMI on the

64

farms investigated. Hence, as cows eat kilograms of forage as opposed to proportions,
which were estimated in this study, variations in DMI between and within herds likely
introduced signiﬁcant error in the estimate of forage NDF consumed. Furthermore,
selective feeding by cows in herds in which ration components were fed seperately would
further increase the potential error of the DMI estimate. In as much as dietary energy
density is negatively correlated with ﬁber levels, it could be stated that this study shows E
that dry-period diets that are extremely high in concentrates (and hence low in forage) are

associated with increased risk of DA.

 

No other signiﬁcant associations between diet and the occurrence of DA were
found in this investigation. No association with dietary factors was found in studies by
Dohoo and workers (1984), and Martin (1972). The diffrculty of estimating DMI in ﬁeld
based investigations makes comparisons between herds, with respect to diet, potentially
inaccurate. This, in part, is likely responsible for the absence and, in some cases, the
presence of controversial signiﬁcant ﬁndings.

Twins

Twins were excluded from the analysis as, in terms of NEF A levels, abomasal
displacement and other diseases, they represented another subset of the population.
Reasons for this may be a feature of a greater and earlier drop in DMI together with the
greater prepartum energy needs that they experience before parturition that lead to
elevated NEFA concentrations (Van Saun and Sniffen, 1995). These authors suggest that
much of the observed health and production problems associated with twin pregnancy

may result from inadequate nutrition in the prepartum period to meet fetal needs resulting

65

in severe maternal nutrient depletion. The high ME requirement for twin pregnancy and
the weight loss experienced by dams is also reported by Koong et al. (1982). It seems
logical to state that the tendency of twin pregnancies to be associated with the occurrence
of DA gives support to both the association of negative energy balance prepartum and
the occurrence of DA and also the theory advanced by Constable (1992) and Svendson
(1969) that rumen volume constitutes a moderately effective barrier to displacement.

However, it cannot be stated if one or both factors are involved in this tendency for DA. Ia

li"'ll

In the herd-based model it has to be appreciated that in some herds the average J

 

body condition score was based on a smaller number of cows than used to estimate DA
incidence. This may have led to an inaccurate assessment of the herd average. However,
the fact that the same association is detected in the cow model, where only cows with
BCS were used, indicates the signiﬁcant inﬂuence of this risk factor. The same concern
is applicable for the herd average PTAmilk and the average incidence of elevated plasma
NEF A concentrations.

Cows that had had a DA previously and were surgically corrected were not
accounted for in the analysis. This may have had an impact on incidence especially if the
sample for the herd was small.

Although special effort was given to the accurate recording of feeding practices
on the farms, a potential short fall was an accurate analysis of the rations themselves. Due
to the size of the investigation it was not feasible to chemically analyze individual

feedstuffs. Reliance for these data was placed on the farmer. No doubt this information

66

was accurate in some cases, but not in others. In retrospect, additional inquires on the
questionnaire may have given a greater appreciation of the potential accuracy of the
ration evaluation. Such questions could include:- Which ration components have been
analyzed and when was this done? How were samples collected? How often is the DM
of ensiled feeds checked and by what method? How are ingredients weighed and

amounts of orts obtained?

An effort was made to evaluate prepartum rations with respect to their DCAD

value. However, as this factor was not assessed accurately in a number of herds it was left

._ “h...“ .“-_—~h

out to maintain the sample size. A trend of higher DCAD diets being associated with

ﬁn-

higher incidence of DA was observed. This is a potential risk factor worthy of further

study.

Summary and Conclusions

Signiﬁcant risk factors for DA included a negative energy balance prepartum (as
estimated from plasma NEFA), high BCS, suboptimal bunk management, prepartum

diets with NE] > 1.65Mcal/kg dry matter, winter and summer seasons, high genetic merit

~ ._. =1. manna-J . r..-

and low parity. The results of this study give support to the theory that hepatic lipidosis ,
may be an important risk factor for DA given the signiﬁcance of a negative energy
balance in the prepartum. The signiﬁcance of suboptimal bunk management and
overconditioning give further support to this theory. The importance of ensuring
optimum nutrition is highlighted in both models. The potential affect of seasonal
inﬂuence on the energy balance of cows in late gestation offers, in part, an explanation
for the seasonal incidence that is commonly observed.

Herds with higher average PTAmilk are associated with a higher occurrence of
DA. This association is a new ﬁnding and is worthy of further investigation. Although
this may be a factor that has to be accepted in the drive to improve herd genetics, its
importance, if indeed signiﬁcant, is that it will need to be considerd in ﬁrrther
investigations into risk factors of DA. Furthermore, as improving PTAmilk is desirable,
the trend for a continued increase in DA incidence is apt to continue, making the need for

effective control measures more pressing.

67

68

Bracncdreconunendnticns

DA occurrence in a herd is likely to be reduced by the avoidance of a negative
energy balance prepartum, overconditioned animals and by providing optimal bunk
management to cows in late gestation. Prepartum rations excessively high in energy (N E]
>1 .65 Meal/kg DM) are also to be avoided. In addition, if it were feasible, a seasonal
calving pattern in either the spring or fall may avoid the higher risk periods of the year. If
this is not feasible, at least an effort could be made to avoid calving cows in the hot
summer months when it is known that dry matter intakes can be signiﬁcantly reduced.

The importance of ensuring optimum nutrition is highlighted in both models.
Improving bunk management is a recommendation that can be readily implemented on
most farms and the beneﬁcial impact of maximizing dry matter intake in late pregnancy
can only help improve the energy balance of cows in late pregnancy. Negative aspects of
bunk management that were frequently encountered in the analysis included restricted
bunk space, infrequent cleaning of the bunk and restricting feed such that feed was not
continuously available. As has been reported by VandeHaar and co-workers (1995), a
prepartum diet more nutrient dense than recommended by NRC may be a feasible way of
improving the nutritional status of peripartum cows. However, there must be caution that
the energy density is not raised excessively (>1 .65 Mcal NEl/kg DM).

The results of this study give support to the theory that hepatic lipidosis may be
an important risk factor for DA given the signiﬁcance of negative energy balance in the

preparttun as assessed by plasma NEFA concentrations. Thus, it may be worthwhile to

Tu an,“ nah-“mu.”
:~ 1- l» ‘3 '
i

69

consider a number of preventative measures discussed for the prevention of hepatic
lipidosis and are listed as follows:-

1) Maximize DMI throughout late gestation - this may best be achieved by having a 'close
up' or ‘transition group’ into which cows are moved 3 weeks before anticipated calving

date. This group of animals can be fed a more nutrient dense ration and together with

optimum bunk management will reduce the likelihood of a negative energy balance I
occurring in the prepartum period. I
2) Proper ration formulation to avoid potential errors !
3) Close monitoring of health including conditions such as lameness that may impact on I
nutrient intake
4) Avoid over-conditioned animals
5) Avoid stressful environmental conditions. The hot summer months are at least easy to
predict and calving animals at this time should be avoided if possible.
6) Propylene glycol administration in late gestation. This could be considered as a ﬁnal
measure to reduce fat mobilization in late gestation especially in individual animals that
are off feed.
7) Assessment of energy balance for early correction of a negative balance. The use of
plasma NEFA assays may prove useful in ﬁeld investigations where problems in energy
balance are considered.

The inﬂuence of negative energy balance, overconditioning and suboptimal bunk

management in late gestation on the occurrence of DA is an important concept to be

taken from this study when considering preventative measures for DA. This is an

70

important contribution to the understanding of this disease and is a facet absent in the

common literature when preventative recommendations are considered.

 

‘3...

APPENDIX

 

APPENDIX

Description of variables
ChangeinEnergLLeielIdenerg)

This categorical variable was formulated to describe the temporal change in dietary
energy density in the periparturient period (3 weeks before to 3 weeks after parturition)
on a scale of 1 (mild) to 5 (severe).

1) more than 20 days on a transition ration (or a gradual change over from basal ration to
high ration) and an intermediate ration post calving for 7 days or more prior to high
ration

2) 20 days or more on a transition ration or a gradual change over from basal to high
ration
or:
14 days transition ration and then intermediate ration 7 days post calving

3) 14 days transition ration then high ration
or:
7-10 days transition ration then intermediate ration 7 days post calving
or:
14 days intermediate ration post calving

4) 7-10 day intermediate ration straight to high ration
or:
basal ration to intermediate ration postcalving or basal diet with exposure to some
concentrate prior to calving

5) Basal ration to high ration

v-4 av-.. “‘2..M“¢tl
)-

72

 

This variable was formulated to consider the change in forage types and amounts in the
periparturient period (3 weeks before to 3 weeks after calving) and to classify this on a
scale of 1 (mild) to 5 (severe)

1) little change in type and proportion

2) no change in type, minor change in relative amounts

3) a greater than 50% changes in relative amounts

4) a greater than 50% changes in relative amount when corn silage decreases relative to
haylage
or:
introduction of a new forage type

5) severe changes in types (excluding hay) and amounts

I)Bunk Space (b sp) -
a) if < ft/cow
or 1-2 ﬂ/cow and TMR limited or cows clean up entire TMR
Then b sp = 1

b) if >2ft/cow

or 1-2ft/cow and feed available all the time
Then b sp = 0

2)Feed availability and freshness (f&f)
a) if bunks not cleaned at least once a day

or if cows clean it up
Then f&f = 1

b) if cleaned at least once a day
Then f&f = 0

Brmkmanagementpenalmscnretﬁllhﬂs)
Penalty score (BUNK) is calculated in equation 3.
Equation 3:

BUNK = b Sp + f&f

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Bartlett PC, Kaneene JB, Kirk JH, Wilke MA, Martenuik JV: Development of a
computerized dairy herd health data base for epidemiologic research. Prev. Vet. Med., 4:
3-14, 1986.

Bartlett PC, Kopcha M, Coe PH, Ames NK, Ruegg PL, Erskine RJ: Economic comparison
of the pyloro-omentopexy vs the roll- and toggle procedure for treatment of left
displacement of the abomasum in dairy cattle. JAVMA 206, 1156-1162, 1995.

Beg H, Whiteford WA: Displacement of the abomasum in the cow. Vet Rec 68: 122-125,
1956.

Bertics SJ, Grummer RR, Cadoriniga-Valino C, Stoddard E: Effect of prepartum dry
matter intake on liver triglyceride concentration and early lactation. J Dairy Sci 75: 1914-
1922, 1992.

Bines JA, Hart IC: Milk or meat. Blood proﬁles in animal production. Occasional
publication No.1 British Society of Animal Production, 1977.

Blum JW, Kunz P, Leuenverger H, Gautschi K, Keller M: Thyroid hormones, blood
plasma metabolites and hematological parameters in relationship to milk yields in dairy
cows. Anim Prod 36: 93-104, 1983.

Bowden DM: Non-esteriﬁed fatty acids and ketone bodies in blood as indicators of
nutritional status inruminants: A review. Can. J. Anim. Sci. 51: 1-13, 197].

Breukink HJ: Abomasal displacement, etiology, pathogenesis, treatment and prevention.
The Bovine Practitioner 26: 148-153, 1991.

Breukink HJ, De Ruyter T: Abomasal Displacement in Cattle: Inﬂuence of Concentrates in
the Ration on Fatty Acid Concentrations in Ruminal, Abomasal, and Duodenal Contents.
Am J Vet Res 37: 1181-1184, 1976.

Brockman RP, Laarveld B: Hormonal regulation of metabolism in ruminants; A review.
Livest. Prod. Sci., 14: 313-334, 1986.

73

 

 

74

Bruss lVﬂn Metabolic fatty liver of ruminants. Advances in veterinary science and
comparative medicine. 37. 417-449, 1993.

Coggins CR, Field AC: Diurnal variation in the chemical compostion of plasma from
lactating beef cows on 3 dietary energy intakes]. J. agric. Sci., Camb. 86: 595-602, 1976.

Coggins CR, Field AC: Changes in plasma concentrations of glucose, free fatty acids,
ketone bodies, thyroxine and insulin of lactating beef cows in relation to time of feeding
and energy status. Blood proﬁles in Animal production. Occassional publication No. 1
British Society of Anim. Prod. 1977.

Collier RJ, Doelger SG, Head HH, Thatcher WW, Wilcox CJ: Effects of heat stres during
pregnancy on maternal hormone concentrations, calf birth weight and postpartum rrrilk yield
of holstein cows. J. Anim. Sci. Vol 54, No. 2: 309—-3 18, 1982.

Constable PD, Miller GY, Hoffsis GF, Hull BL, Rings DM: Risk factors for abomasal
volvulus and left abomasal displacement in cattle. AM J Vet Res, 53: 1184-1192, 1992.

Coppock CE: Displaced abomasum in dairy cattle: Etiological factors. J. Dairy Sci., 57:
926-933, 1974.

Coppock CE, Noller CH, Wolfe SA, Callahan CJ and Baker JS: Effect of forage-
concentrate ratio in complete feeds fed ad libitum on feed intake prepartum and the
occurence of abomasal displacement in dairy cows. J Dairy Sci 55: 783-789, 1972.

Correa MT, Curtis CR, Erb HN, Scarlett JM, Smith RD: An ecological analysis of risk
factors for postpartum disorders of Holstein-Friesian cows from thirty-two New York
farms. J Dairy Sci 73: 1515-1524, 1990.

Cote JF, Hoff B: Interpretation of blood proﬁles in problem dairy herds. The Bovine
Practitioner 26: 7-11, 199].

Curtis CR, Erb HN, Sniffen C], Smith RD, Powers PA, Smith MC, White ME, Hillrnan
RB, Pearson EJ: Association of parturient hypocalcaemia with eight periparturient disorders
in Holstein cows. JAVMA 183: 559-561, 1983.

Curtis CR, Erb HN, Sniffen C], Smith RD, Kronfeld DS: Path analysis of dry period
nutrition, post partum metabolic and reproductive disorders, and mastitis in Holtstein cows.
J. Dairy Sc 68: 2347-2360, 1985.

Dawson LJ, Aalseth EP, Rice LE, Adams GD: Inﬂuence of ﬁber form in a complete mixed
ration on incidence of left displaced abomasum in postpartum dairy cows. JAVMA 200:
1989—1992, 1992.

 

75

DeCupere F, Muglle E, Van Den Hende C, Oyaert W: Metabolic proﬁle tests on high
yielding normal cows and cows suffering from abomasal displacement. The Bovine
Practitioner 26: 129-130, 1991.

deLahunta A, Habel RE : Applied Veterinary Anatomy, WB Saunders Company, 1986.

Dohoo IR, Martin SW: Disease, production and culling in Holstein-Friesian cows. 11].
Disease and production as determinants of disease. Prev. Vet. Med. 2: 671-690, 1984.

Dohoo IR, Martin SW, Meek AH: Disease, production and culling in Holstein-Friesian
cows. VI. Effects of management on disease rates. Prev. Vet. Med., 3: 15-28, 1984.

Dohoo IR, Martin SW, McMillan 1, Kennedy BW: Disease, production and culling in
Holstein-Friesian cows. 1]. Age, season and sire effects. Prev. Vet. Med., 2: 655-670,
1984.

Dohoo IR, Martin SW, Meek AH, Sandals WCD: Disease, production and culling in
Holstein-Friesian cow. 1. The data Prev. Vet. Med., 1: 321-334, 1983.

Dunlop RH: Pathogenesis of ruminant lactic acidosis. Advances in veterinary science and
comparative medicine. Academic Press, New York, London pp 259-302, 1972.

Durand D, Chilliard Y, Bauchart D: Effects of lysine and methionine on in vivo hepatic
secretion of VLDL in the high yielding dairy cow. J. Dairy Sci. 75 (Suppl. 1) 279, 1992.

Dyk PB: The association of prepartum non-esteriﬁed fatty acids and body condition with
peripartum health problems on 95 Michigan dairy farms. MSU Master’s Thesis 1995.

Dyk PB, Emery RS, Liesman JL, Bucholtz HF, VandeHaar MJ: Prepartum non esteriﬁed
fatty acids in plasma are higher in cows developing peripartum health problems, J. Dairy
Sc. Suppl] 78 264, 1995.

Emery RS, Liesman J S, Herdt TH: Metabolism of long chain fatty acids by ruminant liver.
American Institute of Nutrition, 832-837, 1992.

Erb HN, Grohn YT: Epidemiology of metabolic disorders in the periparturient dairy cow. J
Dairy Sci 7]: 2557-2571, 1988.

Erb HN, Martin SW: Age, breed and seasonal patterns in the occurrence of ten dairy cow
diseases: A case control study. Can J Comp Med 44: 1-9, 1978.

Ford JH: A case of displacement of the bovine abomasum. Vet Rec 49: 763-764,1950.

Forbes JM: Voluntary food intake and reproduction. Proceedings of the nutrition society 46,
l93-201,1987.

and-z“ ...
.
.v' -

 

76

Fox F H: Abomasal Disorders. JAVMA,147, 4, 383-388, 1965.

Frohli DM, Blum JW: Non esteriﬁed fatty acids and glucose in dairy cows: Diurnal
variations and changes in responsiveness during fasting to epinephrine and effects of beta-
adrenergic blockade. J. Dairy Sci. 71: 1170-1177, 1988.

F ronk TJ, Schultz LH. Effect of dry period overconditioning on subsequent metabolic
disorders and performance of dairy cows. J. Dairy Sci. 63: 1080-1090, 1980.

Gaal T, Roberts CJ, Reid IM, Dew AM, Copp CM: Blood compostion and liver fat in
postparturient dairy cows. Vet Rec, 113: 53-54, 1983.

Garnsworthy PC, Topps JH: The effect of body condition of dairy cows at calving on their
food intake and performance when given complete diets. Anim. Prod. 35: 113-119, 1982.

Garnsworthy PC: The effect of energy reserves at calving on performance of dairy cows.
Garnsworthy PC, ed. Nutrition and lactation in the dairy cow. London Butterworth, 157,
1988.

Gerloff BJ, Herdt TH, Emery RS: Relationship of hepatic lipidosis to health and
performance in dairy cattle. JAVMA 188: 845-850, 1986.

Grummer RR: Etiology of lipid-related metabolic disorders in periparturient dairy cows. J.
Dairy Sci. 76: 3882-3896, 1993.

Grymer J: Displaced abomasum - a disease often associated with concurrent disease.
Comp. Cont. Ed. 11: 290-295, 1980.

Grymer H, Hesselholt M, Willeberg P: Feed composition and left abomasal displacement
in dairy cattle. Nord. Vet. Med. 33: 306-309, 1981.

Habel RE : Guide to the dissection of domestic ruminants. Ithaca, New York 1970

Habel RE, Smith DF: Volvulus of the bovine abomasum and omasum. JAVMA 179,
447-455, 1981.

Herdt T: Text book of Veterinary Physiology. Editor JG Cunningham, W.B. Saunders Co.
Philadelphia, 1992.

Herdt T (a): Fuel homeostasis in the ruminant. Veterinary Clinics of North America: Food
Animal Practice, 4, 213-231,1988.

Herdt T (b): Fatty liver in dairy cows. Veterinary Clinics of North America: Food Animal
Practice, 4, 269-287, 1988.

77

Holmes J, Lamboume LJ: The relation between plasma free fatty acid concentration and
the digestible energy intake of cattle. Res. vet. Sci 1]: 27-35, 1970.

Holtenius P: Distributed insulin and glucagon function - etiologic factors in ketosis. Proc.
Eight Int. Conf. on production diseases in farm animals, Beme, Switzerland, pg. 199, 1992.

Holtenius P, Traven M: Impaired glucose tolerance and heterogenecity of insulin responses
in cows with abomasal displacement. J. Vet. Med. A37, 445-451, 1990.

Holtenius P, Niskanen R: Fatty changes of the liver cells in cows with abomasal
displacement. Dtsch. tierartztl. Wschr. 92, 398-400, 1985.

Huffman CF: High- level grain feeding for dairy cows, J. Dairy Sci 44: 2113- 2122, 1961.

Hull BL, Wass WM: Abomasal displacement. 2: Hypocalcemia as a contributing causative
factor. VM SAC 68: 412-417, 1973.

Jaster EH, Wegner TN: Beta-adrenergic receptor involvement in lipolysis of dairy cattle
subcutaneous adipose tissue during dry and lactating state. J. Dairy Sci., 64: 1655-1663,
1981.

Johnson DG, Otterby DE: Inﬂuence of dry period diet on early post-partum health, feed
intake, milk production and reproductive efﬁciency of Holstein cows. J. Dairy Sci 64: 290-
295, 1981.

Johnson MJ, Peters JP: Technical note: an improved method to quantify nonesteriﬁed fatty
acids in bovine plasma. J. Anim. Sci. 71: 753-756, 1993.

Jones WP, Hansen LB, Chester-Jones H: Response of health care to selection for milk
yield of dairy cattle. J. Dairy Sci. 77: 3137-3152, 1994.

Jubb TF, Malmo J, Davis GN, Vauser AS: Left side displacement of the abomasum in
dairy cows at pasture in Australia. Austral. Vet J. 68: 140-142, 1991.

Kaneene JB, Hurd HS: The national animal health monitoring system in Michigan. 1.
Design, data and frequencies of selected dairy cattle diseases. Prev. Vet. Med., 8: 103-114,
1990.

Kaneene JB, Miller R: Risk factors for metritis in Michigan dairy cattle using herd- and
cow-based modelling approaches. Prev. Vet. Med. 23, 183-200, 1995.

Koong LJ, Anderson GB, Garret WN: Maternal energy status of beef cattle during single
and twin pregnancy. J. Dairy Sci 54, 480-484, 1982.

78

Kuiper R, Breukink HJ: Myo-electric activity patterns on the abomasal body in the adult
cow recorded with stainless steel electrodes. J. Vet Med A 35: 340-346,1988.

Kunz PL, Blum JW, Hart IC, Bickel H, Landis J: Effects of different energy intakes before
and after calving on food intake, performance and blood hormones and metabolites in dairy
cows. Anim. Prod. 40: 219-231, 1985.

Lindsay DB: The effect of feeding pattern and sampling procedure on blood parameters.
Blood proﬁles in animal production. Occassional publication No. 1 British Society of
Anim Production, 1977.

Madison JB, Menitt AM, Rice B, Donovan GA, Lester GD, Tran T: Inﬂuence of an abrupt
change in diet on antroduodenal myoelectric activity in lactating cattle. Am. J. Vet. Res. 54,
793-796, 1993.

Mahoney CB, Hansen LB, Young CW, Marx GD, Reneau JK: Health care of holsteins
selected for large or small body size. J. Dairy Sci 69: 3131-3139, 1986.

Markusfeld O: The association of displaced abomasum with various periparturient factors
in dairy cows. A retrospective study. Prev. Vet Med 4: 173-183, 1986.

Markusfeld O: Periparturient traits in seven high dairy herds. Incidence rates, association
with parity, and interrelationships among traits. J Dairy Sci 70:158-166, 1987.

Martin W: Left abomasal displacement: An epidemiological study. Can Vet J 13:61-
68,1972.

Martin SW, Kirby KL, Curtis RA: Left abomasal displacement in dairy cows: its
relationship to production. Can. Vet. J. 19: 250-253, 1978.

Massey CD, Wang C, Donovan GA, Beede DK: Hypocalcemia at parturition as a risk
factor for left displacement of the abomasum in dairy cows. JAVMA 203: 852-853, 1993.

McCutcheon SN, Bauman DE: Effect of chronic growth growth hormone treatment on
responses in epinephrine and thyrotropin-releasing hormone in lactating cows. J .Dairy Sci.
69:44, 1986.

McNamara JP, Hillers JK: Regulation of bovine adipose tissue metabolism during
lactation. 2. Lipolysis response to milk production and energy intake. J. Dairy Sci. 69:
3042-3050, 1986.

Meirhaeghe van H, Deprez P, Hende van den C, Muylle E: The inﬂuence of insulin on
abomasal emptying in cattle. J. Vet. Med. A 35, 213-220, 1988

 

79

Meirhaeghe van H, Deprez P, Hende van den C, Muylle E: Plasma glucose clearance and
insulin response in cows with abomasal displacement. J. Vet. Med. A. 35, 221-228, 1988.

Mills SE, Beitz DC, Young JW: Evidence for ipaired metabolism in Liver during induced
lactation ketosis of dairy cows. J. Dairy Sci. 69: 362-370, 1986,

Moore GR, Riley WF, Westcott RW, Conner GH: Displacement of the bovine abomasum.
Vet Med 49: 49-51, 1954.

Morrow DA: Fat cow syndrome. J. Dairy Sci., 59: 1625-1629, 1976.

Nocek JE, English JE, Braund DG: Effects of various forage feeding programs during the
dry period on body condition and subsequent lactation, health, production and reproduction.
J. Dairy Sci. 66: 1108-1118,1983.

Oetzel GR, Olson JD, Curtis CR, Fettman MJ: Ammonium chloride and ammonium
sulfate for prevention of parturient paresis in dairy cows. J. Dairy Sci 71: 3302-3309, 1988.

Oetzel GR: Meta-analysis of nutritional risk factors for milk fever in dairy cattle. J. Dairy
Sci 74: 3900-3912, 1991.

Oetzel GR, Barmore JA: Intake of concentrate mixture containing various anionic salts fed
to pregnant, non-lactating dairy cows. J. Dairy Sci 76: 1617-1623, 1993.

Pehrson BG, Shaver RD: Displaced abomasum: clinical data and effects of peripartal
feeding and management on incidence. XXV AA of Bovine Practioners Conference 1: 116-
121, 1992.

Pinsent PJN, Neal PA, Ritchie HE: Displacement of the bovine abomasum: a review of 80
clinical cases. Vet Rec 73: 729-735, 1961.

Pullen DL, Liesman JS, Emery RS: A species comparison of liver slice synthesis and
secretion of triacylglycerol from nonesteriﬁed fatty acids in media. J. Anim. Sci. 68: 1395-
1399, 1990.

Radloff HD, Schultz LH, Hoekstra WG: Relationship of plasma free fatty acids to other
blood components in ruminants under various physiological conditions. J. Dairy Sci. 49:
179-182, 1966.

Reid IM: Incidence and severity of fatty liver in dairy cows. Vet Rec 107: 281-284, 1980.

Reid 1, Roberts J: Fatty liver in dairy cows. In Practice 164-168, November, 1982.

 

80

Reid IM, Roberts CJ, Teacher RJ, Williams LA: Effect of body condition at calving on
tissue mobilization, devolpment of fatty liver and blood chemistry of dairy cows. Anim.
Prod. 43: 7-15, 1986.

Reid IM, Roberts CJ: Subclinical fatty liver in dairy cows - current research and future
prospects. Irish Vet J. 37: 104-110, 1983.

Robb EJ, Johnstone C: Cohort investigation of risk factors for bovine abomasal
displacement. J. Dairy Sci. 70 (Suppl) 227, 1987.

Robertson JM: Left displacement of the bovine abomasum: Epizootiologic factors. Am J
Vet Res 29: 421—434, 1968.

Ruegg PL, Milton RL: Body condition scores of Holstein cows on Prince Edward Island,
Canada: Relationships with yield, reproductive performance, and disease. J. Dairy Sci 78:
552-564, 1995.

Sack WO: Abdominal topography of a cow with left abomasal displacement. Am. J. Vet
Res. 8, 1567- 1576, 1968.

Sechen SJ, Dunshea FR, Bauman DE: Somatotropin in lactating cows: Effect on
response to epinephrine and insulin. Am. J. Physiol. 258 (Endocrinol. Metab. 21): E582,
1993.

Shanks RD, Freeman AE, Berger PJ, Kelly DH: Effect of selection for milk production on
reproductive and general health of the dairy cow. J. Dairy Sci. 61, 1765-1772, 1978.

Shirley JE: Preparing the cow for a successful lactation. Proceedings of the Southwest
nutrition and management conference. Phoenix, Arizona, 110-113, 1994.

Simmons CR: Administration of Somatotropin before parturition elicits a short term
increase in subsequent milk protein synthesis. MS thesis, MSU 1993.

Statistical Analysis Systems Institute Inc., 1985. SAS User's guide: Statistics, version 5
edition. Statistical analysis systems institute inc., Cary, NC.

Statistics and epidemiology research corporation and cytel software corporation, 1991.
Egret reference manual, revision 2. Statistics and epidemiology research corporation and
cytel software corporation, Seattle, WA.

Studer VA, Grummer RR, Bertics SJ, Reynolds CK: Effect of prepartum propylene glycol
administration on periparturient fatty liver in dairy cows. J. Dairy Sci.76: 2931-2939, 1993.

Svendsen P: Etiology and pathogenesis of abomasal displacement in cattle. Nord Vet Med
21 (suppl 1): 1-61, 1969.

 

.lI—

 

81

Treacher RJ, Reid 1, Roberts CJ: The effect of body condition at calving on the
development of fatty liver and metabolic disease. Animal Production 32: 363, 1981.

VandeHaar MJ, Sharma BK, Yousif G, Herdt TH, Emery RS, Allen MS,Liesman JS:
Prepartum diets more nutrient-dense than recommended by NRC improve nutritional status
of peripartum cows. J. Dairy Sc. Suppl] 78 264, 1995.

Van den Top AM, Wensing T, Wentink GH, Van't Klooster AT, Beynen AC: Time trends
of plasma lipids and hepatic triacylglycerol synthesizing enzymes during post-parturn fatty
liver development in dairy cows with unlimited access to feed during the dry period. J.
Dairy Sci. 78: 2208-2220, 1995.

Van Dijk S, Wensing T, Wentink G, Joma T: Hepatic lipidosis in dairy cows related to
health and fertility. Proc. V Int Conference Production Disease in Farm Animals, New
York, America. 289-292, 1989.

Van Saun RJ, Sniffen CJ: Comparison of pre- and postpartum performance in dairy cattle
having either a single or twin pregnancy, J. Dairy Sc. Suppl] 78 266, 1995.

Varden SA: Displacement of the abomasum in the cow. Incidence, etiological factors and
results of treatment. Nord Vet Med 31: 106-113, 1979.

Vazquez-Anon N, Bertics S, Luck M, Grummer RR, Pinheiro J: Peripartum liver
triglyceride and plasma metabolites in dairy cows. J. Dairy Sci. 77: 1521-1528, 1994.

Vlaminck VK, Meirhaeghe van H, Hende van den C, Oyaert W, Muylle E: Inﬂuence of
endotoxins on abomasal emptying in cattle. DTW 92, 392-395, 1985.

Wiggans GR, VanRaden PM: USDA-DHIA Animal Model Genetic evaluations. National
Cooperative Dairy Herd Improvement Program. Fact Sheet H-2 1-8, 1989.

Wildman EE, Jones GM, Wagner PE, Boman RL, Trout HF Jr, Lesch TN: A dairy cow
body condition scoring system and its relationship to selected production characteristics. J.
Dairy Sci., 65: 495-501, 1982.

Willeberg P, Grymer J, Hesselholt M: Left displacement of the abomasum: relationship to
age and medical history. Nord Vet Med. 34: 404-411, 1982.

Yousif GM: Effect of prepartum dietary energy density and protein content on prepartum
body fat mobilization, milk yield, health and performance and protein turnover in Holstein
cows. MS. Thesis, Michigan State University, 1995.

"4&1

 

in

82

Zamet CN, Colenbrander VF, Callahan CJ, Chew VP, Erb RE, Moeller NJ: Variables
associated with peripartum traits in dairy cows. 1. Effect of dietary forages and disorders on
volantary intake of feed, body weight and milk yield. Theriogenology 1]: 229-243, 1979.

 

 

"I1111111111111“