Estimation of Energy Expenditure from Heart Rate Measurements in Cattle Maintained under Different Conditions

Abstract

We examined whether heart rate (HR) could be used to estimate energy expenditure (EE) in cattle. Six Hereford heifers (345 +/- 10.8 kg BW) 12 mo of age were implanted with HR radio transmitters and maintained in individual pens under the following treatments: 1) shade or sun exposure, 2) high- or low-energy diet, and 3) feeding in morning or afternoon. The HR of animals was measured every .5 h during 3 mo; measurements of oxygen consumption and HR were made simultaneously in the morning and in the afternoon while animals were resting and exercising. Average daily HR (52 +/- 4 beats/min) and average daily EE (380 +/- 9 kJ/kg(.75)) in animals on the low-energy diet were less than values in animals on the high-energy diet (94 +/- 4 beats/min and 653 +/- 9 kJ/ kg(.75), respectively). For each animal and within each diet, linear regressions best described the relationship between HR and EE in resting animals, whereas quadratic regressions best described this relationship for exercising animals. The quadratic equation for the exercising animals could also be used for resting animals. In addition, a constant value of EE per heart beat (EE pulse) for each individual resting animal was found and gave accurate estimations. This method was convenient because 1) no exercise equipment was needed to generate the regression equations and 2) EE pulse was less affected by diet than was EE estimated by regression equations. We conclude that HR, a relatively easy measurement, can be useful and accurate in estimating EE. To increase the accuracy of the estimation of EE by HR, the relationship of HR to EE should be established for each animal. In addition, the nutritional regimen for the animal in which EE is estimated should be used for the animal in establishing the relationship.

Full text

A. Brosh, Y. Aharoni, A. A. Degen, D. Wright and B. Young
under different conditions
Estimation of energy expenditure from heart rate measurements in cattle maintained
1998, 76:3054-3064.J ANIM SCI
http://jas.fass.org/content/76/12/3054
the World Wide Web at:
The online version of this article, along with updated information and services, is located on
www.asas.org
by guest on July 13, 2011jas.fass.orgDownloaded from

3054
1
Contribution from the Agric. Res. Organization, Volcani Center,
Bet Dagan, Israel, No. 2328-E, 1997 series. The authors wish to
thank S. Fennell, G. Beneke, B. Hall, A. Goodwin, K. Rowan, J.
McCosker, M. Josey, F. Gorbacz, R. Englebright, I. Williams, and T.
Schoorl for their contributions.
2
To whom correspondence should be addressed: phone:
972-4-9539523; fax: 04-9836936; E-mail: beefny@netvision.net.il.
Received January 20, 1998.
Accepted June 3, 1998.
Estimation of Energy Expenditure from Heart Rate Measurements
in Cattle Maintained Under Different Conditions
1
A. Brosh*
,2
, Y. Aharoni*, A. A. Degen
†
, D. Wright
‡
, and B. Young
§
*Department of Beef Cattle, Agricultural Research Organization, Institute of Animal Science,
Newe Ya’ar Research Center, P. O. Box 1021, Ramat Yishay 30095, Israel;
†
Desert Animal Adaptations and Husbandry, Jacob Blaustein Institute for Desert Research,
Ben-Gurion University of the Negev, Beer Sheva 84105, Israel;
‡
Department of Companion Animal Medicine and Surgery, University of Queensland,
St. Lucia 4067, Australia; and
§
Department of Animal Production,
University of Queensland, Gatton College 4345, Australia
ABSTRACT: We examined whether heart rate
(HR) could be used to estimate energy expenditure
(EE) in cattle. Six Hereford heifers (345 ± 10.8 kg
BW) 12 mo of age were implanted with HR radio
transmitters and maintained in individual pens under
the following treatments: 1) shade or sun exposure, 2)
high- or low-energy diet, and 3) feeding in morning or
afternoon. The HR of animals was measured every .5
h during 3 mo; measurements of oxygen consumption
and HR were made simultaneously in the morning
and in the afternoon while animals were resting and
exercising. Average daily HR (52 ± 4 beats/min) and
average daily EE (380 ± 9 kJ/kg
.75
) in animals on the
low-energy diet were less than values in animals on
the high-energy diet (94 ± 4 beats/min and 653 ± 9 kJ/
kg
.75
, respectively). For each animal and within each
diet, linear regressions best described the relationship
between HR and EE in resting animals, whereas
quadratic regressions best described this relationship
for exercising animals. The quadratic equation for the
exercising animals could also be used for resting
animals. In addition, a constant value of EE per heart
beat (EE pulse) for each individual resting animal
was found and gave accurate estimations. This
method was convenient because 1) no exercise equip-
ment was needed to generate the regression equations
and 2) EE pulse was less affected by diet than was EE
estimated by regression equations. We conclude that
HR, a relatively easy measurement, can be useful and
accurate in estimating EE. To increase the accuracy of
the estimation of EE by HR, the relationship of HR to
EE should be established for each animal. In addition,
the nutritional regimen for the animal in which EE is
estimated should be used for the animal in establish-
ing the relationship.
Key Words: Cattle, Feeds, Heart Rate, Oxygen Consumption, Energy Expenditure, Solar Radiation
1998 American Society of Animal Science. All rights reserved. J. Anim. Sci. 1998. 76:3054–3064
Introduction
Energy expenditure ( EE) of farm animals has been
determined mostly under controlled, confined condi-
tions. These conditions do not necessarily reflect those
of free-ranging animals, or of commercial cattle in
feedlots. Environmental conditions, feeding level, time
spent eating and digesting, tissue and pelage conduc-
tance, production level, and season of the year may
affect EE of animals (NRC, 1981).
Double-labeled water ( DLW) can be used to
determine CO
2
production, a measure of metabolic
rate of free-living animals, but DLW is expensive and
the method contains errors of ± 8 to 11% (Nagy,
1989). Acute tracheal intubation for measurement of
O
2
uptake (Young and Webster, 1963) and infusion of
[
14
C]bicarbonate (Corbett et al., 1971; Young and
Corbett, 1972) have restricted field application and
are expensive to use with large animals. In addition,
the latter method interferes in the normal behavior of
the animal.
Estimation of EE of free-ranging large animals
based on heart rate ( HR) has been examined by
several workers (Webster, 1967; Yamamoto et al.,
by guest on July 13, 2011jas.fass.orgDownloaded from

HEART RATE AND ENERGY EXPENDITURE IN CATTLE
3055
Table 1. The arrangement of the treatments during the experiment
a
a
The letters represent the treatment conditions according to the following order. 1st: diet of low- (L)
and high- (H) energy concentration, 2nd: exposure to sun (E) or protected by shade (P), and3rd: feeding
in the morning (M) or afternoon (A).
b
Acclimatization period.
Animal
Days 123456Change of
1−4
b
LEA LPA HEA HPA HEA HPA
5−11 LEA LPA HEA HPA HEA HPA
12−18 L E M L P M H E M H P M H E M H P M Time fed
19−25
b
L P M L E M H P M H E M H P M H E M Shade
25−31 LPM LEM HPM HEM HPM HEM
32−38 L P A L E A H P A H E A H P A H E A Time fed
39−52
b
H E M H P M L E M L P M L E M L P M Diets
53−59 HEM HPM LEM LPM LEM LPM
60−66 H E A H P A L E A L P A L E A L P A Time fed
67−73
b
H P A H E A L P A L E A L P A L E A Shade
74−79 HPM HEM LPM LEM LPM LEM
80−85 H P A H E A L P A L E A L P A L E A Time fed
Table 2. Approximate analysis (% dry matter) and ME of the low- (L)
and high- (H) energy diets consumed by the heifers
a
Crude fiber.
Ether ME,
Diet OM CP extract CF
a
NDF MJ/kg
L 89.77 7.68 1.31 52.93 68.53 7.21
H 93.56 16.92 2.23 18.12 29.53 10.63
1979; Richards and Lawrence, 1984; Renecker and
Hudson, 1985; Yamamoto, 1989). This method could
be most useful because recent developments in microe-
lectronics allow the use of small HR radio transmitters
to measure HR of animals in their natural habitat.
The objective of this study was to determine whether
HR in cattle can be used to estimate EE under
different nutritional and environmental conditions.
Materials and Methods
Six trained 12-mo-old Hereford heifers (345 ± 10.8
kg BW) were implanted with heart-rate radio trans-
mitters (Telonics, Mesa, AZ) approximately 1 mo
before commencement of measurements. During the
85-d study, HR for each animal was measured for 5
min every .5 h throughout every day. The six heifers
were studied in a 2 × 2 × 2 arrangement (high- and
low-energy feed; exposure to and protection from solar
radiation; fed in morning and in afternoon) and
underwent all treatments as outlined in Table 1.
Measurements were made during the summer
(January−March) of 1993 in southeast Queensland,
Australia, a subtropical region with summer tempera-
tures commonly over 30°C. Animals were kept in-
dividually in open feedlot pens, each 40 m
2
. Shade was
provided in about half the area of each of the pens:
galvanized iron sheets, at a height of 2.2 m, covering
11.5 m
2
, and 70% shade cloth, at a height of 4 m,
covering 12 m
2
.
The animals were fed either a high-energy diet ( H)
of 80% concentrate and 20% sorghum hay, or a low-
energy diet ( L) of only sorghum hay. The ME of the
concentrate was calculated from its composition and
known energy equivalents. The ME of the sorghum
hay offered and left over was determined by nylon bag
measurements of organic matter digestibility (Setala,
1983) and assuming a value of 15.06 MJ/kg digestible
organic matter (Minson, 1982). The approximate
analysis (AOAC, 1980) of the feed eaten and its
calculated ME, taken as offered minus left over, are
presented in Table 2.
The quantity of feed given was regulated to reduce
the refusals to less than 5% of that offered, and it was
given either in the morning (0800 to 0830) or
afternoon (1630 to 1700). Refusals were collected and
weighed once weekly.
Oxygen uptake was measured by the use of a face
mask open-circuit respiratory system (Taylor et al.,
1982). The accuracy was checked gravimetrically by
injecting nitrogen into the mask (McLean and Tobin,
1990). The EE was calculated assuming 20.47 kJ/L O
2
(Nicol and Young, 1990).
by guest on July 13, 2011jas.fass.orgDownloaded from

BROSH ET AL.
3056
Measurements of HR and O
2
uptake were made at
the same time for each animal on each treatment
during two consecutive days when penned. The
animals stood most of the time and, occasionally, lay
down. Measurements were made over 15 to 20 min
each time: between 0700 and 0830 before morning
feeding and between 1400 and 1530 in the afternoon.
Data were averaged every 5 s, recorded on a data
logger (Mini-Logger

, Mini-Mitter Co., Sunriver,
OR), and transferred to a laptop computer for
processing. For analysis, the data were pooled over
30-s intervals. For each such simultaneous measure-
ment of HR and O
2
uptake, the O
2
pulse was
calculated as the O
2
uptake per heartbeat.
There were four measurements per animal per
treatment and 32 measurements per animal in total,
composed of two measurements for each time (morn-
ing or afternoon) for each treatment. For analysis of
the stability of O
2
pulse, the difference between these
two measurements was tested.
Oxygen uptake and HR during exercise were
measured simultaneously on four occasions for each
animal on each diet. Measurements were made over
20 min on 2 d for each animal, each day, once in the
morning and once in the afternoon. A circular walker
was used to exercise the animals and the rate of
walking was up to 6 km/h.
The EE pulse was calculated as the energy expendi-
ture per heart beat. Daily animal EE was calculated
by multiplication of the total daily heart beats by the
average EE pulse.
Body weights of the heifers were determined every
2 wk and at the beginning and end of each treatment.
Live weight ( LW) and daily LW gain ( LWG) were
calculated from the regression slope of all the meas-
urements of LW on each diet.
Meteorological data (ambient temperature [Ta],
black globe temperature [BG], and relative humidity
[RH]) were collected between 0700 and 0830 and
between 1400 and 1530, and black globe humidity
indices ( BGHI) (Buffington et al., 1981) were
calculated.
Statistical Analysis
Regression analysis was used to test whether a
relationship existed between O
2
uptake and HR and
between O
2
pulse and HR. The regressions were done
separately for the data taken at rest and at exercise.
For exercising animals, averages of O
2
uptake and O
2
pulse for each 5-beats/min interval of HR were taken
for each animal at each state. Regressions were also
done for all the animals, either at rest or exercising,
with a random animal effect, to obtain 1) intercepts
and slopes for each animal in each state and 2) a
common slope that characterized the state. In the
resting animal model, fixed effects were assigned for
radiation conditions, time of feeding, time of measure-
ment, and diet, with either a linear or a quadratic
effect of HR within diet. In the exercising model, a
fixed effect was assigned to diet, and either a linear or
a quadratic effect within diet.
The equations of the model were as follows:
Y = A + R + TF + TM + D/HR + e
[1]
Y = A + R + TF + TM + D/(HR + HR
2
)+e
[2]
The equations of the models for exercising were as
follows:
Y = D/HR + e
[3]
Y = D/(HR + HR
2
)+e
[4]
where Y = O
2
uptake or O
2
pulse, A = random animal
effect, R = radiation (exposed or protected) effect, TF
= time of feeding (morning or afternoon) effect, TM =
time of measurement (morning or afternoon) effect, D
= diet (L or H) effect, HR, HR
2
= linear and quadratic
effects of heart rate, and e = error.
The regression equations for each animal, linear
and quadratic, based on the exercising data, were used
to examine whether its O
2
uptake at rest could be
estimated from its HR. The O
2
pulse was calculated in
the same way except that only a linear regression was
used. These estimated values were compared with the
observed values, which were recorded simultaneously
with the HR, by means of a paired t-test.
The effect of time elapsed between a last given meal
and measurement on the observed values of HR, O
2
uptake, and O
2
pulse during rest was estimated by
linear regression of these variables on the elapsed
time (hours). In this regression, a random effect was
assigned to the animal, and fixed effects to conditions
of radiation, time of feeding, time of measurement,
and diet.
The contribution of the variance between the pair of
measurements for each situation to the residual
variance was tested by means of the residual error
when all the fixed effects and the linear effect of the
elapsed time were accounted for by the regression. The
absolute value of the difference between the residuals
of the two replicates for each state for each animal (n
= 96) was calculated, and the mean value of this
difference was averaged for each variable and com-
pared with the mean value of this variable.
All analyses were made with Genstat 5 Release 3.2
(Lawes Agricultural Trust, 1995).
Results and Discussion
Morning and afternoon air temperatures were 23.5
± .8 and 30.2 ± .7°C, whereas RH were 65 ± 2and45±
3%, respectively. Black globe temperatures were 25.2
by guest on July 13, 2011jas.fass.orgDownloaded from

HEART RATE AND ENERGY EXPENDITURE IN CATTLE
3057
Table 3. Linear regression of O
2
uptake (mL O
2
/(kg
.75
·h)) and of O
2
pulse (mLO
2
/
(kg
.75
·beat) on heart rate (HR), with fixed effects of radiation (exposed or
protected), time of feeding (morning or afternoon), time of measurements
(morning or afternoon), and diet (L or H) at rest (df = 82)
and at exercise (df = 102)
a
Significance: NS, not significant.
b
Value/beat.
c
Difference of slopes between diets L and H (L − H).
†
P < .10.
**P < .01.
***P < .001.
Rest Exercise
Effect Value SE P
a
Value SE P
a
O
2
uptake
Radiation 14.7 20.0 NS —
Time of feeding −39.2 19.8
†
—
Time of measurement −65.2 21.0 ** —
Diet 146 172 NS 145 251 NS
HR slope on L diet
b
18.1 2.5 *** 23.2 1.7 ***
Difference H diet
c
−3.7 2.8 NS −3.4 2.5 NS
r
2
of regression .933 .742
O
2
pulse
Radiation 3.77 4.36 NS —
Time of feeding −8.14 4.33
†
—
Time of measurement −17.98 4.58 *** —
Diet −2.4 37.5 NS −14.3 39.8 NS
HR slope on L diet
b
.34 .55 NS 1.24 .27 ***
Difference H diet
c
−.37 .61 NS −.26 .39 NS
r
2
of regression .314 .202
Figure 1. Average daily heart rates (±SE) of heifers on
high (n = 4; D) and low (n = 2; o) energy diets. Feed was
given at 0830.
± .8 and 30.2 ± .7°C under the shade and 37.6 ± 1.3
and 45.0 ± 1.7°C in the sun in the morning and
afternoon, respectively. Black globe humidity indices
were 72.6 ± 1.1 and 80.1 ± .7 in the shade and 84.9 ±
1.3 and 92.1 ± 1.6 in the sun, in the morning and in
the afternoon, respectively.
The HR was higher in heifers fed the H diet than
the L diet and increased in all heifers during and after
feeding (Figure 1).
Previous studies that measured HR to estimate EE
used linear or logarithmic regression equations to
relate O
2
uptake and HR (Webster, 1967; Yamamoto
et al., 1979; Richards and Lawrence, 1984; Renecker
and Hudson, 1985; Purwanto et al., 1990). In this
study, within each diet, linear regression equations
best described resting animals, whereas quadratic
regression equations best described exercising animals
(Tables 3 and 4). Generally, the slopes of the
equations for exercising animals were steeper than
those of resting ones (Figure 2).
These data indicate that the dependence of O
2
uptake on HR during rest is linear, and, hence, the O
2
uptake per heartbeat (the O
2
pulse) does not depend
on HR under these conditions (Figure 3). In contrast
to the resting conditions, a physical strain such as
that imposed during exercise induced a O
2
uptake
increase that was not linearly related to the increase
in HR. However, if the O
2
uptake could be accurately
described by a quadratic regression on HR, then the
O
2
pulse should be satisfactorily described by a linear
regression, because the O
2
pulse is given by O
2
uptake
divided by HR. The finding that the O
2
pulse
regression on HR was not linear suggests, therefore,
by guest on July 13, 2011jas.fass.orgDownloaded from

BROSH ET AL.
3058
Table 4. Quadratic regression of O
2
uptake (mL O
2
(kg
.75
·h) and of O
2
pulse
(mLO
2
/(kg
.75
·beat) on heart rate (HR), with fixed effects of radiation
(exposed or protected), time of feeding (morning or afternoon), time
of measurements (morning or afternoon), and diets (L or H)
at rest (df = 82) and at exercise (df = 102)
a
Significance: NS, not significant;
†
P < .10, *P < .05, **P < .01, ***P < .001.
b
Value/beat.
c
Value/beat
2
.
d
Difference of slopes between diets L and H (L − H).
Rest Exercise
Effect Value SE P
a
Value SE P
a
O
2
uptake
Radiation 12.3 20.6 NS —
Time of feeding −44.0 20.6 * —
Time of measurement −59.6 21.7 ** —
Diet −286 1,207 NS −2,880 810 ***
HR slope on L diet
b
−21.2 34.6 NS 43.5 7.8 ***
HR
2
slope on L diet
c
.387 .339 NS −.106 .040 **
Difference HR H diet
2
24.0 39.0 NS 44.2 14.6 **
Difference HR
2
H diet
2
−.33 .35 NS −.178 .064 **
r
2
of regression .932 .805
O
2
pulse
Radiation 3.35 4.49 NS —
Time of feeding −.922 4.48 * —
Time of measurement −16.66 4.72 *** —
Diet −69 262 NS –381 130 **
HR slope on L diet −9.01 7.51 NS 5.30 1.25 **
HR
2
slope on L diet .092 .074 NS −.021 .0065 **
Difference HR H diet
d
5.57 8.48 NS 5.12 2.34 *
Difference HR
2
H diet
d
−.075 .077 NS −.018 .010
†
r
2
of regression .319 .382
that the regression of O
2
uptake on HR during
exercise could not be described satisfactorily, even
with a quadratic polynomial.
During exercise, three parameters that contribute
to the O
2
uptake are altered: 1) HR; 2) A-V difference,
the difference between arterial and venous O
2
in
blood; and 3) stroke volume (Eckert et al., 1988;
Jones et al., 1989). The negative slope on HR
2
could
have resulted from a reduction in the rate of increase
of the A-V difference (Jones et al., 1989) and from a
reduction of the stroke volume as the O
2
uptake and
HR increased as a result of exercise. This is probably
why, when a polynomial of the third order was used to
describe the relations of O
2
uptake and O
2
pulse on
HR, neither of the regressions was improved, com-
pared with the second-order polynomial (results of the
third-order polynomial are not presented).
During rest, the effect of time of feeding on the O
2
uptake and the O
2
pulse tended to be significant, and
the effect of time of measurement on these parameters
was significant. These parameters were lower in the
afternoon than the morning feeding and in the
afternoon than the morning measurement. No effect of
radiation was detected on either O
2
uptake or O
2
pulse
on HR.
Individual regression equations for each animal on
each diet (Figure 2) generated when exercising were
used to estimate EE at rest. The difference between
estimated and measured values was significant only
for the L diet, and only when the linear regression
equations were used (Table 5). However, the stan-
dard errors of the mean were much lower when O
2
pulse was used than O
2
uptake for the linear and
quadratic regressions. As can be seen in Figure 2,
regression equations differed among animals and were
affected by dietary ME. Therefore, to estimate EE
from HR, a regression equation should be determined
for each animal when it consumes a diet similar to the
one in which the animal’s EE is to be estimated. In
contrast to this, despite the significant effect of dietary
ME on O
2
pulse, the range of variations among
animals and diets is minor (Figure 3), so O
2
pulse (or
EE pulse) to estimate EE by HR is less affected by
diet than was EE estimated by regression equations.
The HR increased after the meal and decreased
thereafter (Figure 1). The correlation of O
2
uptake
and O
2
pulse with HR (Tables 3 and 4) was affected
by the time of feeding, morning or afternoon, and by
the time of measurement, and it was postulated that
these effects could also be attributed to the time that
elapsed between the meal and the measurement.
Therefore, we tested the effect of elapsed time on HR,
O
2
uptake, and O
2
pulse (Table 6). We used all 192
individual measurements of HR and O
2
uptake,
by guest on July 13, 2011jas.fass.orgDownloaded from

HEART RATE AND ENERGY EXPENDITURE IN CATTLE
3059
Figure 2. Oxygen uptake (mL/(kg
.75
·h) ± SE on the ordinate and heart rate (beats/min) on the abscissa of six
heifers on low- (L) and high- (H) ME diets. Measurements at rest in the pens (R, empty symbols) under all treatments
and at exercise using a walker (W, filled symbols). Linear regression, broken line for resting animals and quadratic
regression, full line when exercising. Note the different axes limits for each animal.
comprising two measurements for each combination of
diet, time of feeding, time of measurement, and
radiation conditions for each heifer. Random effect of
the animal and fixed effects of all the above-mentioned
conditions were accounted for by the regression
equations obtained in the first step, in addition to the
linear effect, the elapsed time. In each of the
regressions of HR, O
2
uptake, and O
2
pulse, all the
fixed effects that were not significant in the first step
were excluded from the regression in the next step.
Thus, only diet and time of measurement effects were
included in the regression of HR, only diet effect in the
regression of O
2
uptake, and all the fixed effects but
radiation in the regression of O
2
pulse on the time
elapsed since the last meal before the measurement.
The HR and O
2
uptake decreased with increasing
time, at rates of .72 and .92% of the mean per hour,
respectively, which left the O
2
pulse relatively cons-
by guest on July 13, 2011jas.fass.orgDownloaded from

BROSH ET AL.
3060
Figure 3. Oxygen pulse (mL/(beat·kg
.75
) ± SE and heart rate (beats/min) of six heifers on low- (L) and on high- (H)
ME diets at rest in the pens, panels A and B, respectively, and at exercise using a walker on diet L and on diet H,
panels C and D, respectively. Note the different axes limits for each panel.
Table 5. The percentage difference between measured O
2
uptake and O
2
pulse of
heifers at rest and those estimated from linear and quadratic regression
equations at exercise. The regressions for exercising animals
were calculated separately for each one (n = 6)
a
NS, not significant; *P < .05.
Low-energy diet High-energy diet
Parameter Calculation % difference SE P
a
% difference SE P
O
2
uptake Linear 25.5 9.8 * −1.0 6.2 NS
O
2
uptake Quadratic 3.3 13.4 NS −.6 7.0 NS
O
2
pulse Average .6 2.5 NS 1.0 4.0 NS
tant with time. Even though the effect of diet was
highly significant for all three parameters, the magni-
tude of this effect was only 10.5% of the mean for the
O
2
pulse, compared with 68.5 and 59.0% of the mean
for HR and O
2
uptake, respectively.
The O
2
pulse was relatively constant for each
animal, compared with O
2
uptake, for the range of
conditions in the pens, provided that the animals were
not under intensive physical strain. We suggest,
therefore, that O
2
pulse is the most appropriate
parameter to use in estimating EE from the HR of
animals in rest conditions. In pens, the percentage
change between daily minimum and maximum HR
was 80 to 175% on the L diet and 70 to 149% on the H
diet, with constant O
2
pulse throughout this range
(Table 3).
Determinations of HR, O
2
uptake, and O
2
pulse
were done twice for each animal on two separate days,
by guest on July 13, 2011jas.fass.orgDownloaded from

HEART RATE AND ENERGY EXPENDITURE IN CATTLE
3061
Table 6. The linear effect of time elapsed (h) from the previous feeding time
a
to measurement, and fixed effects of treatments in the pens,
on HR, O
2
uptake, and O
2
pulse (n = 192)
a
Varied between 3.4 and 30.2 h.
b
Units/h. Evaluated in a common analysis of fixed and linear effects. Only significant fixed effects were
included.
c
Beats/min.
d
NS, not significant; *P < .05, ***P < .001.
e
mL O
2
/kg
.75
.
f
mLO
2
/(kg
.75
·beat).
g
Dash indicates that the fixed effect did not account for a significant amount of variance in an initial
model and thus was dropped from the analysis used to generate the data in this table.
Fixed effects
Time of Time of Time from
Dependent parameter measurement feeding Diet meal, h
b
HR
c
Slope 4.41 — 48.2 .508
% of mean 6.3 — 68.5 .72
SE 1.29 — 1.16 .091
P
d
*** — *** ***
O
2
uptake
e
Slope —
g
— 613.2 −9.54
% of Mean — — 59.0 .92
SE — — .34 .024
P
d
— — *** ***
O
2
pulse
f
Slope −17.81 −8.48 −26.39 −.183
% of Mean 7.1 3.4 10.5 .07
SE 3.67 3.43 3.26 .269
P
d
*** * *** NS
Table 7. The O
2
pulse (mL/(kg
.75
·beat) and CV (SE/mean·100) of cattle at rest under different conditions
a
Average of six measurements of 16 animals.
b
Measurements of animals at rest.
Animals and conditions n BW, kg O
2
pulse CV, % References
Heifers, low-energy diet 6 345 265 2.60 Present study
Heifers, high-energy diet 6 337 238 2.50 Present study
Standing Hinterwaeelder oxen 7 494 347 2.8 Rometsch et al., 1997
Standing Zebu oxen 5 516 300 2.7 Rometsch et al., 1997
Swedish Red and White steers 3 449 377 — Jones et al., 1989
Simmental oxen — 562 338 — Clar, 1991
Boran cows — 475 429 — Zerbini et al., 1992
Heifers of various breeds 8 123−177 252 6.80 Liang et al., 1997
Black-Pied dairy cattle bulls (16
a
)6280−350 286 2.4 Derno et al., 1997
Mean ± SE of the 9 references 315 ± 3.2 315 ± 3.2 1.0 CV of the 9 references
under each set of conditions in the pens. The mean
absolute values of the differences between the residu-
als of the two replicates for each state for each animal
were .52 beat/min, .78 mL O
2
/(kg
.75
·h), and .90 mLO
2
/
(kg
.75
·beat) for HR, O
2
uptake, and O
2
pulse,
respectively. These differences were equal to only .74,
3.90, and .36% of the mean values of HR, O
2
uptake,
and O
2
pulse, respectively. It is suggested, therefore,
that a single determination is sufficient.
The O
2
pulse of a number of cattle breeds under
different conditions at rest is presented in Table 7.
The variance in each breed was small. It is suggested,
therefore, that despite the individual character of the
O
2
pulse and its dependence on diet and environmen-
tal conditions, HR can provide an estimation of their
EE. Moreover, provided that the cattle in pens and
pasture are not exposed to exercise, day-to-day
changes in HR provide a reliable estimation of
changes in EE.
It can be concluded that the EE of the heifers can be
estimated from HR measurements, using quadratic
equations that relate O
2
uptake to HR or by using a
by guest on July 13, 2011jas.fass.orgDownloaded from

BROSH ET AL.
3062
Figure 4. Metabolizable energy intake (A), energy
expenditure (B), and retained energy (C) of six heifers.
Four were fed a high-energy diet (10.62 MJ/kg DM of
ME) during the first 45 d (thin lines) and a low-energy
diet (7.21 MJ/kg DM of ME) subsequently. The other
two (thick lines) were fed the low- and then high-energy
diets.
constant value of O
2
uptake per heartbeat (O
2
pulse),
which is less affected by the nutritional conditions.
The relation of O
2
uptake to HR is dependent on the
causes for changes of HR rather than on HR itself.
Use of HR to estimate energy retention and energy
balance was evaluated. The ME intake ( MEI), EE,
and retained energy ( RE) of the heifers during the
entire study are presented in Figure 4A, B, and C,
respectively. The LW, gain, and the energy balance
parameters on each diet during the entire experiment
are presented in Table 8. The MEI of the heifers was
directly related to the ME concentration of the diet,
and the EE and the RE followed the changes in the
MEI. Before the start of the experiment, the heifers
were fed the L diet; during the first days of the
experiment, the intake of the heifers on the H diet
greatly increased, resulting in similar increases in the
EE and RE. Because the amount of feed given during
each period was constant and the weight gain of the
animals on the H diet was positive, the intake per
metabolic weight decreased during each period. When
the diets were switched, the MEI followed the changes
in the diet energy density and, consequently, so did
the EE and the RE.
Calculation of EE by multiplication of the HR by
the EE pulse, calculated from the O
2
pulse, was less
accurate immediately after the change in feed quality.
For calculation of EE during that period, the EE pulse
of the previous diet was used for the calculation.
Because the difference between the EE pulses on the
two diets was only 10.5% and the time taken for the
heifers to reach stability in intake and EE after the
switch of diets was short, the resulting inaccuracy in
the determination of EE was small.
Predicted EE according to the National Research
Council (NRC, 1984) is 432 ± 47 and 750 ± 46 kJ/
(d·kg
.75
) on the L and H diets, respectively (the
individual MEI and the LWG of the heifers in the
present study were used for the calculation). The NRC
predictions were not significantly different from our
estimation according to the HR and the EE pulse
(Table 8).
The calculated RE per LWG for the H diet (28.4 ±
1.5 MJ/kg LWG) tended to be larger than the value
obtained by use of the observed LW and LWG in the
NRC (1984) equation (23.6 ± 1.0 MJ/kg LWG). The
LWG on the L diet was negligible, as was the RE, and,
therefore, the calculation of RE per LWG from these
values were not done.
We conclude that HR, a relatively easy measure-
ment, can be useful and accurate in estimating EE. To
increase the accuracy of the estimation of EE by HR,
the relationship of HR to EE should be established for
each individual animal. In addition, the nutritional
regimen used in establishing the relationship should
be similar to that for the animal in which EE will be
estimated. Measurements obtained using this method,
combined with MEI, would allow estimations of body
energy changes in animals losing or gaining body
weight.
Implications
Heart rate is relatively easy to measure, and this
measurement seems to be a practicable method for
estimating accurately the short-term and long-term
by guest on July 13, 2011jas.fass.orgDownloaded from

HEART RATE AND ENERGY EXPENDITURE IN CATTLE
3063
Table 8. Body weight, live weight (LW) gain (LWG), energy balance parameters,
and energy content in LW gain of the heifers (mean of six heifers ± SE) that
were fed high- (H) and low- (L) energy diets. The animals
were kept in individual pens during the summer
a
kg.
b
g/d.
c
g/(d·kg
.75
).
d
Metabolizable energy intake, kJ/(d·kg
.75
).
e
Energy expenditure, kJ/(d·kg
.75
).
f
Retained energy, kJ/(d·kg
.75
).
g
NE
g
, energy content in live weight gain, kJ/g.
h
Not calculated because of the negligible LWG.
i
Period of adaptation to the diets and period of diet changeover in the middle of the experiment (about
2 wk for each period) are included.
Diet BW
a
Gain
b
LWG
c
MEI
d
EE
e
RE
f
NE
g
g
H 337 1,460 18.45 1,190 670 520 28.4
±SE 15 120 .88 25 10 22 1.5
CV 4.6 8.0 .8 2.1 1.4 4.14 5.20
L 345 157 2.079 468 394 74
h
± SE 10 97 1.233 26 19 11 —
CV 3.0 61.9 59.3 5.6 4.8 14.6 —
All
i
337 810 10.37 843 541 301 29.6
± SE 13 50 .64 13 8 9 2.1
CV 6.2 6.3 6.2 1.6 1.4 3.0 7.2
variations of energy expenditure in free-range cattle.
When metabolizable energy intake is also measured,
energy retention in animals and the energy content of
the gain or loss of body weight can be estimated. This
would allow the determinations of energy require-
ments of animals and would be more accurate than
simply using changes in body mass as a criterion.
Under certain circumstances, it would also allow the
calculation of the efficiency of utilization of energy
intake for maintenance and for growth.
Literature Cited
AOAC. 1980. Official Methods of Analysis (13th Ed.). Association of
Official Analytical Chemists, Washington, DC.
Buffington, D. E., A. Collazo-Arocho, G. H. Canton, D. Pitt, W. W.
Thatcher, and R. J. Collier. 1981. Black globe humidity index
(BGHI) as comfort equation for dairy cows. Trans. Am. Soc.
Agric. Eng. 24:711−714.
Clar, U. 1991. Development of a field method for measuring energy
expenditure of draught animals. Ph.D. dissertation. Hohen-
heim Univ., Hohenheim, Germany.
Corbett, J. L., D. J. Farrell, R. A. Leng, G. L. McClymont, and B. A.
Young. 1971. Determination of energy expenditure of penned
and grazing sheep from estimates of carbon dioxide entry rate.
Br. J. Nutr. 26:277−291.
Derno, M., B. Lo¨hrke, W. Jentsch, and H.-D. Matthes. 1997. Effect of
a
2
-adrenergic stimulation and feeding on heat production of
growing bulls. In: K. J. McCracken, E. F. Unsworth, and A.R.G.
Wylie (Ed.) Energy Metabolism of Farm Animals. pp 151−154.
CAB International, Wallingford, U.K.
Eckert, R., D. Randall, and G. Augustine. 1988. Animal Physiology:
Mechanisms and Adaptations (3rd Ed.). p 446. W. H. Freeman,
New York.
Jones, J. H., K. E. Longworth, A. Lindholm, K. E. Conley, R. H.
Karas, S. R. Kayar, and C. R. Taylor. 1989. Oxygen transport
during exercise in large mammals. I. Adaptive variation in
oxygen demand. J. Appl. Physiol. 67:862−870.
Lawes Agricultural Trust. 1995. Genstat V, Mark 4.03. Rothamsted
Experimental Station, Harpenden, Hertfordshire, England.
Liang, J. B., J. Roch, M. A. Sharuddin, and S. Shanmugavelu. 1997.
Heat production and body temperature of Bos × Bos indicus
crossbred in the humid tropics. In: K. J. McCracken, E. F.
Unsworth, and A.R.G. Wylie (Ed.) Energy Metabolism of Farm
Animals. pp 303−306. CAB International, Wallingford, U.K.
McLean, J. A., and G. Tobin. 1990. Animal and Human Calorimetry.
p 276. Cambridge University Press, Cambridge, U.K.
Minson, D. J. 1982. Effect of chemical composition on feed digestibil-
ity and metabolizable energy. Nutr. Abstr. Rev. Ser. B 52:
592−615.
Nagy, K. A. 1989. Field bioenergetics: Accuracy of models and
methods. Physiol. Zool. 62:237−251.
Nicol, A. M., and B. A. Young. 1990. Short-term thermal and meta-
bolic responses of sheep to ruminal cooling: Effects of level of
cooling and physiological state. Can. J. Anim. Sci. 70:833−843.
NRC. 1981. Effect of Environment on Nutrient Requirements of
Domestic Animals. National Academy Press, Washington, DC.
NRC. 1984. Requirements of Beef Cattle (6th Ed.). National
Academy Press, Washington, DC.
Purwanto, B. P., Y. Abo, R. Sakamoto, F. Furumoto, and S.
Yamamoto. 1990. Diurnal patterns of heat production and
heart rate under thermoneutral conditions in Holstein Friesian
cows differing in milk production. J. Agric. Sci. 114:139−142.
Renecker, L. A., and R. J. Hudson. 1985. Telemetered heart rate as
an index of energy expenditure in moose ( Alces alces). Comp.
Biochem. Physiol. 82A:161−165.
Richards, J. I., and P. R. Lawrence. 1984. The estimation of energy
expenditure from heart rate measurements in working oxen
and buffalo. J. Agric. Sci. 102:711−717.
Rometsch, M., U. Roser, K. Becker, and A. Susenbeth. 1997. Estima-
tion of the energy expenditure from heart rate measurements
in working oxen. Anim. Sci. 64:409−412.
by guest on July 13, 2011jas.fass.orgDownloaded from

BROSH ET AL.
3064
Setala, J. 1983. The nylon bag technique in determination of rumi-
nal feed protein degradation. J. Sci. Agric. Soc. Finl. 55:
1−78.
Taylor, C. R., N. C. Heglund, and G.M.O. Maloiy. 1982. Energetics
and mechanics of terrestrial locomotion as a function of
speed and body size in birds and mammals. J. Exp. Biol. 97:
1−21.
Webster, A.J.F. 1967. Continuous measurement of heart rate as an
indicator of the energy expenditure of sheep. Br. J. Nutr. 21:
769−785.
Yamamoto, S. 1989. Estimation of heat production from heart rate
measurement of free living farm animals. JARQ (Jpn. Agric.
Res. Q.) 23:134−143.
Yamamoto, S., J. A. McLean, and A. J. Downie. 1979. Estimation of
heat production from heart rate measurements in cattle. Br. J.
Nutr. 42:507−513.
Young, B. A., and J. L. Corbett. 1972. Maintenance energy of
grazing sheep in relation to herbage availability. Aust. J. Agric.
Res. 23:57−76.
Young, B. A., and M.E.D. Webster. 1963. A technique for the estima-
tion of energy expenditure in sheep. Aust. J. Agric. Res. 14:
867−873.
Zerbini, E., T. Gemeda, D. H. O’Neill, P. J. Howell, and R. C.
Schroter. 1992. Relationship between cardio-respiratory
parameters and draught work output in F1 crossbred dairy
cows under field conditions. Anim. Prod. 55:1−10.
by guest on July 13, 2011jas.fass.orgDownloaded from

Citations
http://jas.fass.org/content/76/12/3054#otherarticles
This article has been cited by 7 HighWire-hosted articles:
by guest on July 13, 2011jas.fass.orgDownloaded from