J. Dairy Sci.

96:6355–6365
http://dx.doi.org/10.3168/jds.2013-6665
© American Dairy Science Association®, 2013.

Physiological responses and lactational performances of late-lactation

dairy goats under heat stress conditions
S. Hamzaoui,* A. A. K. Salama,*†1 E. Albanell,* X. Such,* and G. Caja*
*Grup de Recerca en Remugants (G2R), Departament de Ciència Animal i dels Aliments, Universitat Autònoma de Barcelona,
08193 Bellaterra, Spain
†Sheep and Goat Research Department, Animal Production Research Institute, 12311 Dokki, Giza, Egypt

ABSTRACT yield by losing body mass, but milk protein content and
protein yield were depressed. Further research is needed
Eight Murciano-Granadina dairy goats in late lacta- to assess the response of dairy goats to HS at earlier
tion were exposed to different ambient conditions, us- stages of lactation.
ing metabolic cages in a climatic chamber. The experi- Key words: heat stress, lactation, digestibility, dairy
mental design was a crossover (2 periods of 35 d and goat
4 goats each) and conditions were (1) thermal neutral
(TN; 15 to 20°C day-night) and (2) heat stress (HS;
INTRODUCTION
12-h day at 37°C and 12-h night at 30.5°C). Humidity
was maintained at 40% and light-dark was constant Heat stress (HS) decreases milk production of dairy
(12–12 h). The forage:concentrate ratio was adjusted animals, and half of this reduction in milk yield is due
daily for maintaining similar value in TN and HS goats to reduced DMI (Rhoads et al., 2009). The other half
(70:30). Water was freely available at ambient tempera- of milk yield losses could be explained by the increase
ture. Rectal temperature and respiratory rate (0800, in maintenance requirements (NRC, 2007), decreas-
1200 and 1700 h) and milk yield were recorded daily, ing secretion of growth hormone (Mitra et al., 1972),
whereas milk composition, nonesterified fatty acids lowering blood flow to the udder (Lough et al., 1990),
and haptoglobin in blood were analyzed weekly. At d downregulating milk protein genes, and upregulating
25, additional blood samples were taken for analysis of apoptosis genes in the mammary gland (Collier et al.,
metabolites and indicators of the acid-base balance. Di- 2006). Cows under HS had greater levels of insulin with
gestibility coefficients and N balance were determined improved insulin sensitivity and lacked the ability of fat
(d 31 to 35) and body weight was recorded (d 35). mobilization from adipose tissue to face the decreased
Compared with TN goats, HS goats experienced great- DMI (Baumgard and Rhoads, 2013). The reduction
er rectal temperature (+0.58°C), respiratory rate (+48 in dairy farm profit associated with HS when the
breaths/min), water intake (+77%) and water evapora- temperature-humidity index (THI) is extremely high
tion (+207%). Intake of HS goats rapidly declined until is not only a result of decreased milk yield, but also
d 7 (−40%), partially recovered from d 7 to 19, and includes impaired milk quality, reproduction problems,
steadied thereafter (−14%). No changes in digestibility increased health care costs, and even animal death.
or N balance were detected. Blood nonesterified fatty Despite the large number of studies carried out in
acids and haptoglobin peaked at d 7 in HS goats but dairy cows, little is known about the effects of HS in
did not vary thereafter. Although milk yield did not dairy goats. Goats are considered more tolerant to high
vary by treatment, milk of HS goats contained −12.5% THI values compared with dairy cows because of their
protein and −11.5% casein than TN goats. Panting re- metabolic size and high water-conservation capacity
duced concentration and pressure of CO2 in the blood (Silanikove, 2000). When environmental temperatures
of HS goats, but they were able to maintain their blood increased from 20 to 40°C, respiration rate increased
pH similar to the TN group by lowering HCO3− and from 30 to over 200 breaths/min in East African goats
increasing Cl− concentrations in their blood. In con- (Maloiy and Taylor, 1971) and domestic Swedish goats
clusion, HS dairy goats showed dramatic physiological (Olsson et al., 1995), indicating that water evaporation
changes during the first week of treatment and partially by respiration plays an important role in heat dissipa-
recovered thereafter. They were able to maintain milk tion in goats.
Lactating Saanen goats exposed to moderate or se-
vere HS for 4 d (THI = 81 or 89) lost milk yield by
Received February 4, 2013.
Accepted July 2, 2013. 3 or 13%, respectively (Sano et al., 1985). Brown et
1
Corresponding author: [email protected] al. (1988) reported that the exposure of dairy goats
6355
6356 HAMZAOUI ET AL.

to moderate HS conditions for 5 wk (34°C and 25% 45 ± 5% (THI = 61) for the TN goats. The HS goats
humidity; THI = 79) depressed milk yield in Alpine but were kept in a 4 × 6 × 2.3-m climatic chamber (Eu-
not in Nubian goats, indicating that the response to HS roshield; ETS Lindgren-Euroshield Oy, Eura, Finland)
varies according to breed. provided with a temperature and humidity controlling
The objective of the current study was to measure system (Carel Controls Ibérica, S.L., Barcelona, Spain).
the physiological, lactational, and nutritional responses A continuous 90 m3/h air turnover was maintained
to extreme heat stress conditions in Spanish Murciano- throughout the experiment.
Granadina dairy goats in late lactation. Moreover, Goats had a 4-wk preexperimental period under TN
blood acid-base status and stress indicators were also conditions for adaptation to the diet and to metabolic
evaluated. No information is available on the effects of cages. When goats were switched from TN to HS con-
HS on this dairy breed, which is widely spread in the ditions, a transition period of 2 d was allowed (1 d
Mediterranean area. at 25°C and 1 d at 30°C), but no transition was ap-
plied for the change from HS to TN. Photoperiod was
MATERIALS AND METHODS maintained constant at 12-h light:12-h dark (0900 to
2100 h) and data of environmental temperature and
Animal and Management Conditions humidity were recorded every 10 min by using 2 data
loggers (Opus 10; Lufft Mess- und Regeltechnik GmbH,
Animal care conditions and management practices Fellbach, Germany).
agreed with the procedures stated by the Ethical Com- Daily ration of the goats consisted of (as-fed) dehy-
mittee of Animal and Human Experimentation of the drated fescue hay ad libitum (20% daily refusal), 0.65
Universitat Autònoma de Barcelona (Bellaterra, Spain; kg of alfalfa pellets, and 0.8 kg of concentrate mixture
CEEAH reference 09/771) and the codes of recommen- (30% corn, 25.8% barley, 25% soybean meal, 8.5%
dations for the welfare of livestock of the Ministry of sunflower meal, 5% FA sodium salts, 2.5% dicalcium
Agriculture, Food and Environment of Spain (Madrid). phosphate, 2% calcium carbonate, 1% sodium chloride,
Eight open multiparous Murciano-Granadina dairy and 0.2% vitamins A, E, and D3; as fed). Mineralized
goats (43.5 ± 2.6 kg of BW) with healthy and sym- salt blocks were freely available in each metabolic
metrical udders, from the herd of the experimental cage (composition: 36.74% Na, 0.32% Ca, 1.09% Mg,
farm of the Universitat Autònoma de Barcelona were 5 g of Zn/kg, 1.5 g of Mn/kg. 912 mg of S/kg, 304
blocked in 2 balanced groups and used at late lactation mg of Fe/kg, 75 mg of I/kg, 50 mg of Co/kg, and 25
(194 ± 3 DIM; 1.53 ± 0.04 L/d). The experimental mg of Se/kg; Ovi Bloc; Sal Cupido, Terrassa, Spain).
design was a crossover with 2 treatments in 2 periods, The concentrate mixture was offered in 2 daily por-
lasting 35 d, and 4 goats each. Goats were switched to tions at 0900 and 1600 h. Changes in the forage intake
the opposite treatment in the second period. Climatic of the HS goats were taken into account throughout
conditions were (1) thermal neutral (TN; 15 to 20°C the experiment and the amount of concentrate offered
and 45% relative humidity; THI = 59 to 65) and (2) HS was daily modified to maintain a constant and similar
(12-h day at 37°C and 40% relative humidity; THI = forage:concentrate ratio to that of the TN goats. Clean
85; and 12-h night at 30.5°C and 40% relative humid- water was permanently available at ambient tempera-
ity; THI = 77). The order of treatments on each goat ture, according to treatment.
was recorded and taken into account in the statistical Goats were milked once daily (0800 h) with a por-
analyses. The THI values were calculated according to table milking machine (Westfalia Separator Ibérica SA,
NRC (1971) as follows: Granollers, Spain) set at 42 kPa, 90 pulses/min, and
66% pulsation ratio, provided with recording jars (2
THI = (1.8 × Tdb + 32) − [(0.55 − 0.0055 × RH) L ± 5%). Milking routine included cluster attachment
× (1.8 × Tdb − 26.8)], without udder preparation or teat cleaning, machine
milking, machine stripping before cluster removal, and
where Tdb is the dry bulb temperature (°C) and RH is teat dipping in an iodine solution (P3-io shield; Ecolab
the relative humidity (%). Hispano-Portuguesa S. L., Barcelona, Spain).
Throughout the experiment, (mid-January to mid-
April), the TN goats were kept indoors and the tem- Sample Collection, Analyses, and Measurements
perature was maintained at 15 to 20°C with the help
of an electric heather equipped with a thermostat (3.5 Body Temperature and Respiration Rate. Rec-
kW; General Electric, Barcelona, Spain). The tempera- tal temperatures and respiration rates were recorded at
ture and relative humidity averaged 16.7 ± 0.3°C and 0800, 1200, and 1700 h. Rectal temperature was mea-

Journal of Dairy Science Vol. 96 No. 10, 2013

 HEAT STRESS EFFECTS ON PHYSIOLOGY AND LACTATION OF GOATS 6357

sured by a digital clinical thermometer (model “mini lated by the difference between true protein and CN,
color”; ICO Technology, Barcelona, Spain; range, 32 and NPN was calculated by the difference between
to 43.9°C; accuracy, ± 0.1°C), whereas the number of total protein and true protein.
inhalations and exhalations during 60 s indicated the Blood Measures. Blood samples were taken weekly
respiration rate. from the jugular vein into 10-mL plastic lavender Va-
Feed Intake and Water Consumption. Feed in- cutainers with spray-coated K2-EDTA (BD Diagnos-
take and water consumption (accuracy: ± 20 g) were tics, Franklin Lakes, NJ) before the morning feeding.
recorded daily throughout the experiment. Trays with Plasma was obtained by centrifugation of whole blood
saw dust were put below the drinking troughs and for 15 min at 1,500 × g at 4°C, and stored at −20°C
weighted twice daily to take into account water wastes. for the NEFA and haptoglobin analyses. The NEFA
Feed samples were collected before the beginning of were determined by the colorimetric enzymatic test
each experimental period and were ground through a ACS-ACOD method using a commercial kit (Wako
1-mm stainless steel screen, and then analyzed for DM, Chemicals GmbH, Neuss, Germany). Haptoglobin
ADF, NDF, and ash content according to analytical concentration was determined colorimetrically by the
standard methods (AOAC International, 2003). The hemoglobin-binding method using a commercial hap-
Dumas method (AOAC International, 2003) with a toglobin assay (Assay phase Range; Tridelta Develop-
Leco analyzer (Leco Corp., St. Joseph, MI) was used for ment Ltd., Maynooth, Ireland) and an Olympus AU400
N determinations and CP was calculated as percentage analyzer (Olympus Europa Holding GmbH, Hamburg,
of N × 6.25. The chemical composition and nutritive Germany).
value of ration ingredients are shown in Table 1. At d 25, blood samples (approximately 0.3 mL) were
Milk Yield and Milk Composition. Milk yield of collected using insulin syringes (1 mL; BD Micro-Fine;
individual goats was recorded daily throughout the ex- BD Medical-Diabetes Care, Franklin Lakes, NJ) at 0800
periment and milk composition was evaluated weekly. and at 1700 h and immediately analyzed for major ions
A milk sample of approximately 100 mL was collected and metabolites. A single drop of blood was applied to
and preserved with an antimicrobial tablet (bronopol, disposable cartridges containing biochemical and silicon
Broad Spectrum Microtabs II; D & F Control Systems chip technology (i-STAT EC8+; Abbott Point of Care
Inc., San Ramon, CA) at 4°C until analysis. Milk sam- Inc., Princeton, NJ). Then, the cartridge was inserted
ples were analyzed with a near-infrared spectrometer into an i-STAT hand-held analyzer, and the results of
(Foss NIRSystems 5000; Foss Electric A/S, Hillerød, glucose, urea, Cl, Na, K, total CO2 concentration, an-
Denmark) for contents of TS, fat, total protein (N × ion gap, hematocrit, hemoglobin, pH, partial pressure
6.38), true protein, and CN. Whey protein was calcu- of CO2, HCO3−, and base excess were obtained.

Table 1. Chemical composition and nutritive value (DM basis) of the ration ingredients used for dairy goats

Fescue Alfalfa
Item hay pellets Concentrate
Component
DM, % 89.99 93.04 90.30
OM, % of DM 89.70 86.50 88.60
CP, % of DM 10.60 12.60 17.90
NDF, % of DM 48.10 44.90 12.60
ADF, % of DM 23.30 27.00 6.19
Nutritive value1
UEm,2 /kg 1.57 — —
UFL,3 /kg 0.57 0.68 1.17
NEL, Mcal/kg 0.97 1.16 1.99
PDIE,4 g/kg 66 83 129
PDIN,5 g/kg 57 95 153
PDIA,6 g/kg 24 47 79
Ca, g/kg 3.5 16.5 18.8
P, g/kg 2.5 2.5 9.1
1
Calculated according to Institut National de la Recherche Agronomique (INRA, 2007).
2
Fill units for sheep (1 UEm = 1 kg of reference grass DM).
3
Feed units for lactation (1 UFL = 1.7 Mcal of NEL).
4
Protein digested in the small intestine supplied by microbial protein from rumen-fermented OM.
5
Protein digested in the small intestine supplied by microbial protein from RDP.
6
Protein digested in the small intestine supplied by RUP.

Journal of Dairy Science Vol. 96 No. 10, 2013

6358 HAMZAOUI ET AL.

Digestibility Coefficients, and Water and N and nutrient balance were analyzed using PROC GLM
Balances. Feed orts were daily collected (d 31 to 35), of SAS. The model contained the effect of treatment
weighed, and composted for analysis. Feces of each goat and period, the interaction treatment × period, and
were daily collected and 10% of fresh feces were dried the residual error.
at 60°C for 48 h. Then, a composted sample for each Data were tested for the normality of distribution,
goat was stored at room temperature until analysis. and a logarithmic transformation (log10) was applied
Urine was collected in containers with 20 mL of H2SO4 to haptoglobin concentration in blood. Differences be-
(96%) and urine volume was daily measured (accuracy: tween least squares means were determined with the
± 2 mL). Urine samples (5% of total volume) were PDIFF test of SAS. Significance was declared at P <
composted and stored at −25°C for N content analysis. 0.05 unless otherwise indicated.
Samples of urine without H2SO4 were collected at 0800
and at 1700 h during the last 2 experimental days to RESULTS AND DISCUSSION
measure urine pH. Orts and feces samples were ground
through a 1-mm stainless steel screen and then ana- Rectal Temperature and Respiration Rate
lyzed for DM, CP, CF, ADF, NDF, and ash content,
Rectal temperatures and respiration rates increased
as previously indicated. Water balance was also done
from 0800 to 1700 h in both goat groups, but were
during the digestibility period.
greater in HS than in TN goats at all time points (Fig-
Corticosterone in Feces. Samples of fresh feces
ure 1A and 1B; P < 0.001). The increased rectal tem-
were collected on the last day of each experimental
peratures and respiration rates in TN goats from 0800
period and stored at −25°C for corticosterone analysis.
to 1700 h were in accordance with the increment in
Fecal samples were first lyophilized, then extracted
ambient temperature throughout the day (15 to 20°C).
with methanol, and finally diluted 1:10 with the as-
Comparing the HS goats at 0800 h (after being exposed
say buffer of the kit. Analyses were performed in the
to 30.5°C during the night; THI = 77) with TN goats
Clinical Biochemistry Service of the Veterinary Faculty
at 1700 h (THI = 65), we observed similar rectal tem-
of the Universitat Autònoma de Barcelona using the
peratures (+0.04°C; P = 0.390) but greater respiration
commercially available 125I RIA kit (rats and mice
rates (+13 breaths/min; P < 0.01), indicating that HS
corticosterone kit; ICN Pharmaceuticals, Orangeburg,
goats were under HS throughout the day, but to a lower
NY), as described by Morrow et al. (2002). Recovery
extent during the night. Reference respiratory rate for
of known amounts of corticosterone added to the pro-
adult goats ranges between 15 and 30 breaths/min ac-
cessed samples was 59.8%. The relationship between
cording to Pugh and Baird (2012), but it was greater
theoretical and true values of corticosterone in goat
in our TN goats at 1700 h, probably because of breed
feces was linear (y = 0.9919x − 2.863; R2 = 0.997).
and physiological state differences. Maximum rectal
The inter- and intra-assay coefficients of variation were
temperature difference (+0.70°C; P < 0.001) between
16.8 and 9.5%, respectively.
HS and TN goats occurred at 1200 h, whereas the larg-
est difference in respiration rate (+65 breaths/min; P
Statistical Analyses < 0.001) occurred at 1700 h. Increased respiration rate
under HS conditions is a known mechanism for dis-
Data were analyzed by the PROC MIXED for re- sipating heat load by evaporation. Rectal temperature
peated measurements of SAS version 9.1.3 (SAS Insti- and respiration rate values peaked in HS goats during
tute Inc., Cary, NC). The statistical mixed model con- the first week and then gradually decreased, which indi-
tained the fixed effects of the treatment (HS vs. TN), cates a partial adaptation to the HS conditions.
day, and period; the random effect of the animal; the
interactions treatment × day and treatment × period; Feed Intake
and the residual error. The model took into account
the possible carryover effects of previous HS periods On average, DMI decreased by 21% throughout the
through the treatment × period interaction. Data of 35-d experimental period but showed a marked effect
performances (i.e., intake, water, and milk yield) and of time elapsed after the start of the HS treatment
physiological indicators (i.e., rectal temperature and (Figure 2; P < 0.001). Feed intake under HS conditions
respiratory rate) were analyzed on a daily basis. gradually decreased during wk 1, partially recovered
For blood parameters measured at 0800 and 1700 during wk 2 and 3, and remained constant thereafter
h, the model included the effects of treatment, sam- (Figure 2). Heat stress caused a 27% reduction in DMI
pling hour, and period, and the interaction treatment from d 1 to 19 (1.47 ± 0.05 vs. 2.00 ± 0.06 kg/d; P <
× period and treatment × hour. Data on digestibility 0.001) and 14% from d 20 to 35 (1.75 ± 0.06 vs. 2.03

Journal of Dairy Science Vol. 96 No. 10, 2013

 HEAT STRESS EFFECTS ON PHYSIOLOGY AND LACTATION OF GOATS 6359

Figure 1. Rectal temperature (a) and respiratory rate (b) at different hours during the day (0800, 1200, and 1700 h) in dairy goats under
thermal neutral (TN; dashed lines; n = 8) or heat stress (HS; solid lines; n = 8) conditions at late lactation.

± 0.06 kg/d; P < 0.001). The partial recovery of DMI HS chamber conditions under controlled conditions in
from d 19 onwards indicates the adaptation of goats dairy cows (Rhoads et al., 2009; Shwartz et al., 2009)
to HS conditions. Previous studies comparing TN and did not show such an adaptation.
Journal of Dairy Science Vol. 96 No. 10, 2013
6360 HAMZAOUI ET AL.

Figure 2. Milk yield ( and ), DMI (Δ and ), and water consumption (□ and ) of dairy goats under thermal neutral (, Δ, and □ with
dashed lines; n = 8) or heat stress (, , and  with solid lines; n = 8) conditions at late lactation. The SEM values of milk yield, DMI, and
water consumption are 0.05 L, 0.06 kg, and 1.14 L, respectively.

Feed intake reduction due to HS has been previ- intake would not be enough to cover the daily require-
ously reported in dairy goats (Sano et al., 1985), ewes ments and resulted in an apparent BW loss of 1.5 kg
(Abdalla et al., 1993), and cows (Lough et al., 1990; (−41 g/d). The apparent BW changes in TN and HS
Rhoads et al., 2009). Heat-stressed animals decreased goats included the inevitable variations in the digestive
feed intake in an attempt to create less metabolic heat tract content, which were unknown in our data.
because the heat increment of feeding, especially in
ruminants, is an important source of heat production Water Consumption and Water Balance
(Kadzere et al., 2002). Moreover, the gut fill by water
observed in the current study for HS goats (see later) Results of water consumption throughout the experi-
might also be related to the reduced DMI. ment and water balance measured from d 30 to 35 are
According to the lactational performances of our shown in Tables 2 and 3, respectively. The HS goats
goats (Table 2) and the requirements estimated accord- had greater water consumption compared with TN
ing to Institut National de la Recherche Agronomique goats (+5.50 L/d; P < 0.001). The greatest values of
(INRA, 2007), TN goats showed a greater feed intake water intake were recorded during wk 1 when DMI was
(2.03 kg of DM/d or 4.6% of BW) than predicted (1.72 at its lowest value. However, water intake values stabi-
kg of DM/d or 3.9% of BW), which allowed them to lized earlier than DMI in the HS goats (Figure 2) and
cover their daily requirements [1.16 UFL, where UFL remained greater (P < 0.01) than in TN goats through-
= feed units for lactation (1 UFL = 1.7 Mcal of NEL) out the experiment. Increased water intake was mainly
and 102 g of PDI, where PDI = protein digested in the used by HS goats for boosting heat loss by evaporation
small intestine] and to have positive energy and pro- from the skin (sweating) and by respiration (panting).
tein balances (+0.29 Mcal of NEL and +80 g of PDI). The total water evaporation calculated by subtract-
Moreover, TN goats gained 1.8 kg (+51 g/d) during the ing water losses in milk, urine, and feces from water in-
experiment, as can be calculated from data shown in put (water intake + water in food) was 3 times greater
Table 2. On the other hand, feed intake of HS goats was (Table 3) in HS goats than TN goats (+2.23 L/d; P <
lower (1.60 kg of DM/d or 3.7% of BW) than predicted 0.01). We were unable to distinguish between evapora-
(1.72 kg of DM/d or 3.9% of BW). If we also take into tion by sweating and panting, but increased sweating
account a 30% increase in maintenance requirements rates have been reported in heat-stressed dairy cows
because of HS, as indicated by NRC (2007), the energy (Shwartz et al., 2009) and goats (Baker, 1989).

Journal of Dairy Science Vol. 96 No. 10, 2013

 HEAT STRESS EFFECTS ON PHYSIOLOGY AND LACTATION OF GOATS 6361
Table 2. Lactational performance of Murciano-Granadina dairy goats under thermal neutral (TN; n = 8) and heat stress (HS; n = 8) conditions
in late lactation [values are LSM and SE of the difference (SED)]

Treatment Effect (P-value)

Item TN HS SED Treatment Period T × P1

Initial BW, kg 43.6 44.1 1.3 0.72 0.58 0.83
Final BW, kg 45.4 42.6 2.6 0.034 0.69 0.047
DMI, kg/d 2.03 1.60 0.08 0.001 0.010 0.15
Water consumption, L/d 5.5 11.1 1.20 0.001 0.32 0.47
Milk yield, L/d 1.24 1.21 0.02 0.20 0.001 0.67
3.5% FCM,2 L/d 1.38 1.35 0.07 0.53 0.004 0.23
Milk composition, %
TS 12.89 12.41 0.29 0.26 0.62 0.73
Fat 4.21 4.22 0.19 0.96 0.90 0.12
Protein 3.84 3.36 0.15 0.030 0.79 0.99
True protein 3.62 3.12 0.14 0.022 0.98 0.96
Casein 3.21 2.84 0.12 0.034 0.18 0.71
Casein, % of protein 84.1 84.9 0.64 0.20 0.001 0.091
Whey protein 0.63 0.53 0.04 0.029 0.001 0.33
NPN 0.22 0.24 0.01 0.16 0.001 0.80
Fat yield, g/d 52 51 2.9 0.68 0.021 0.21
Protein yield, g/d 48 40 1.8 0.001 0.22 0.18
NEFA,3 mmol/L plasma 0.192 0.137 0.031 0.081 0.67 0.34
1
Interaction of treatment (T) × period (P).
2
3.5% FCM = L of milk yield × [0.432 + 0.162 × (fat %)].
3
Average values of blood samples collected weekly from d 0 to d 35.

Milk Yield NEFA levels were similar between groups, but our late-
lactation HS goats were probably able to maintain milk
Despite the reduced DMI, increased body tempera- yield because they partially recovered DMI, tended
ture, and the known negative effect of HS on milk to have greater digestibility (see later), and had lower
production in dairy cows (West, 2003), milk yield and metabolic demands compared with early-lactation
FCM did not vary between HS and TN goats (Table 2 goats. Milk yield response of dairy goats to HS in early
and Figure 2; P > 0.05). Brown et al. (1988) reported lactation needs further research.
that the exposure of dairy goats to constant conditions Our results of NEFA concentrations from d 14 to 28
of 34°C and 25% relative humidity (THI = 79) for 5 wk (but not at d 7) agreed with findings obtained in dairy
depressed milk yield in Alpine but not in Nubian goats. cows, where HS did not cause an increase in blood
Even when DMI was at its lowest value during the NEFA despite the reduced feed intake (Rhoads et al.,
first week, milk yield was not affected in HS goats. 2009; Baumgard and Rhoads, 2013). Thus, it seems HS
During wk 1, HS goats may have been able to partially cows are more sensitive to insulin than TN cows, al-
cover their lactation requirements by body fat mobili- lowing a potent antilipolytic action of insulin that will
zation, as indicated by the greater values of NEFA in prevent body fat mobilization. Consequently, lactating
plasma at d 7 of HS (Figure 3). From d 14 to 28 blood HS cows fail to have sufficient glucose for milk synthesis

Table 3. Water input and water losses of dairy goats under thermal neutral (TN; n = 8) and heat stress (HS; n = 8) conditions in late lactation
[values are LSM and SE of the difference (SED)]

Treatment Effect (P-value )

Item TN HS SED Treatment Period T × P1

Water intake,2 mL 5,504 9,728 1,863 0.035 0.86 0.64
Water in food, mL 143 127 7 0.034 0.010 0.083
Water in milk,2 mL 969 1,004 62 0.55 0.16 0.29
Urine volume,2 mL 2,143 4,757 1,737 0.41 0.99 0.80
Water in feces, mL 1,426 825 144 0.002 0.33 0.18
Evaporation water,2,3 mL 1,074 3,304 1,430 0.007 0.44 0.48
1
Interaction of treatment (T) × period (P).
2
P-values were extracted after logarithmic transformation because of the nonnormal distribution of data.
3
Calculated by the difference between water input (water intake + water in food) and water losses in milk, urine, and feces, without taking into
account the water produced metabolically.

Journal of Dairy Science Vol. 96 No. 10, 2013

6362 HAMZAOUI ET AL.

feed intake conditions, increasing the percentage of

RUP and supplementing with lysine increased milk
protein content under hot conditions in dairy cows. On
the other hand, Bernabucci et al. (2002) suggested that
decreased mammary synthesis of milk protein (rather
than the reduction in AA intake) is the reason for the
low milk protein during the hot season in dairy cows.

Digestibility and Nitrogen Balance

Goats under HS showed numerically greater values

of digestibility and tended to have greater ADF digest-
ibility than under TN conditions (+3.1 points; P <
Figure 3. Plasma NEFA concentrations of dairy goats under ther- 0.10), as shown in Table 4. To our knowledge, no data
mal neutral (TN; ; n = 8) or heat stress (HS; ; n = 8) conditions are available comparing digestibility under TN and
at late lactation. Values are means, with SE indicated by vertical bars. HS ambient conditions in lactating dairy goats. Our
* indicates a difference at P < 0.001 between TN and HS treatments.
results agree with those of previous research carried out
under climatic chamber conditions with male goats (Hi-
and, therefore, milk yield was depressed. We did not rayama et al., 2004), dairy cows (McDowell et al., 1969)
observe such a decrease in milk yield in our HS goats, and heifers (Bernabucci et al., 1999). The increased di-
probably because the glucose was sufficient for the milk gestibility in the HS treatments in the aforementioned
yield level at late lactation (see later). studies might be partially due to the reduction of DMI
as also observed for HS goats in our results (Table 2).
Milk Composition Another reason for the enhanced digestibility under
HS conditions could be a depressed passage rate of the
Milk TS and fat contents did not vary (P > 0.05) solid phase of digesta as reported by Bernabucci et al.
between HS and TN goats (Table 2). In short-term (1999). On the other hand, a greater rate of passage of
studies carried out using climate-controlled HS, milk the liquid phase of the digesta may be expected as a
fat was also not affected in cows (Rhoads et al., 2009; consequence of the dramatic increase in water intake.
Shwartz et al., 2009) despite the known negative effect This could have an effect on the availability of rumen
of summer on milk fat (Kadzere et al., 2002). This usual soluble fermentable compounds, although this state-
decrease in milk fat content of heat-stressed dairy cows ment needs experimental confirmation. The tendencies
during summer might be related to a reduction in for- observed for an increased digestibility of nutrients (i.e.,
age intake, which could decrease the forage-to-concen- DM, OM, and ADF; Table 4) in the HS goats might
trate ratio in the diet. In the current study, concentrate have compensated the reduction in DMI, and could
amount was adjusted according to the forage intake in partially explain the lack of effects of HS on milk yield
HS goats to maintain a constant forage-to-concentrate in our results.
ratio throughout the experiment and similar to that of Although HS goats had lower N intake (Table 4), as
TN goats. a consequence of the DMI reduction, they experienced
With the exception of milk NPN, the content of lower N losses in feces and urine, which resulted in simi-
protein and protein fractions in milk (true protein, lar daily N retention in TN goats (13.9 ± 1.0 g/d, on
CN, and whey protein) were reduced by HS (Table 2). average). Despite the similar N retention, milk protein
Nevertheless, the CN-to-protein ratio was not affected. content was lower in HS than in TN goats, indicat-
Similarly, dairy cows under controlled conditions of HS ing that ingested N would have been directed to other
decreased their milk protein content (Rhoads et al., metabolic functions rather than milk protein synthesis
2009; Shwartz et al., 2009). Decreased protein intake in HS goats. It is possible that a portion of N intake
and increased sweat secretion that contains protein and was lost in sweat in the form of urea, as previously
urea (Joshi et al., 1968) might have limited the avail- proposed by Joshi et al. (1968). Moreover, the possibil-
ability of AA for milk protein synthesis. Moreover, the ity of a lower supply of essential AA for milk protein
decrease in milk protein content under HS may be due synthesis could not be excluded.
to a lowered microbial protein synthesis in the rumen Treatment × period interactions were detected for
because of changes in rumen environment (dilution or DM and CP digestibilities, and for fecal excretion and
clearance of soluble substances) by the high water in- apparent absorption of N (Table 4). These significant
take. Huber et al. (1994) indicated that under decreased interactions could indicate some carryover effect when
Journal of Dairy Science Vol. 96 No. 10, 2013
 HEAT STRESS EFFECTS ON PHYSIOLOGY AND LACTATION OF GOATS 6363
Table 4. Digestibility coefficients and nitrogen balance of dairy goats under thermal neutral (TN; n = 8) and heat stress (HS; n = 8) conditions
in late lactation [values are LSM and SE of the difference (SED)]

Treatment Effect (P-value )

Item TN HS SED Treatment Period T × P1

Digestibility
DM, % 56.6 58.8 2.3 0.12 0.57 0.047
OM, % of DM 58.8 61.2 1.4 0.11 0.41 0.058
CP, % of DM 70.5 72.1 1.5 0.31 0.72 0.021
NDF, % of DM 36.0 38.8 1.9 0.16 0.98 0.76
ADF, % of DM 35.1 38.2 1.7 0.094 0.14 0.60
N balance
Intake, g/d 48.4 42.1 0.8 0.001 0.012 0.059
Fecal excretion, g/d 14.3 11.8 0.8 0.010 0.35 0.017
Urinary excretion, g/d 20.3 16.4 1.3 0.012 0.96 0.27
Apparent absorption, % 70.5 72.1 1.5 0.31 0.72 0.021
Retention, g/d 13.9 13.9 1.0 0.95 0.16 0.057
1
Interaction of treatment (T) × period (P).

HS goats went to the TN treatment in the second pe- greater respiration rate observed in panting HS goats
riod. However, this seems unlikely in our case, as these contributed to a greater loss of CO2, lowering the
interactions might result from the effect of period on carbonic acid content of the blood. As a consequence,
DMI (P < 0.01; Table 2), where the DMI reduction in HCO3− was transferred from the blood to urine by the
HS goats was more pronounced during period 2 com- kidney to maintain a constant blood pH. Heat stress
pared with period 1 (data not shown). had no effect on blood Na and K concentrations in
accordance with results previously reported in HS dairy
Blood Indicators and Urinary pH cows (Schneider et al., 1988). On the other hand, Cl
concentration was greater at both time points in HS
Heat stress had no effect on blood glucose, urea, he- than in TN goats (Table 5; P < 0.05). Calamari et al.
matocrit, and hemoglobin concentrations (Table 5; P > (2007) reported an inverse relationship between blood
0.05). A tendency in the treatment × hour interaction HCO3− and Cl in dairy cows under TN and HS condi-
was detected for blood glucose (P < 0.10). At 0800 h, tions. Due to the greater Cl concentrations of the HS
HS goats showed lower glucose concentration than TN goats, they also have greater anion gap values, compared
goats (P < 0.05), whereas the concentration at 1700 with TN goats (Table 5; P < 0.05). It was expected
h increased (P < 0.001) and was similar to that of that the increased HCO3− secretion in the urine of HS
TN goats. It must be stressed that the morning blood goats should increase the urine pH, but we observed the
samples were taken 1 h before offering the diet, whereas opposite, as the urine pH of HS goats tended (P = 0.11)
the afternoon sampling was done 1 h after distribution to be lower than that of TN goats. We speculated that
of the second portion of concentrate, which may have HS goats were able to increase their renal excretion of
contributed to the similar blood glucose levels between H+, which resulted in a partial reabsorption of HCO3−
groups at 1700 h. into the blood, as previously observed by Masero and
Blood pH is regulated by a complex system of buffers Siegel (1977). In fact, the decrease in pCO2 at 1700 h
that continuously work to maintain it slightly basic in due to HS was more marked (−20%) than the decrease
a range of 7.35 to 7.45 in most mammals (Constable, in HCO3− (−12%). The secretion of HCO3− in urine
1999). Measured blood pH was similar at 0800 in TN and its reabsorption suggests a large requirement and
and HS goats, but slightly decreased at 1700 h (P < turnover of body bicarbonate to maintain blood pH
0.05) in the TN goats. Despite the importance of blood during heat stress.
pH for understanding the mechanism of respiratory
evaporative heat loss, changes observed in our goats Haptoglobin Concentration in Blood
were marginally relevant and varied within the normal
range in both TN and HS groups. Haptoglobin (an acute-phase protein linked to meta-
Values of total CO2, partial pressure of CO2 (pCO2), bolic stress) concentration in blood plasma was greater
HCO3−, and base excess were lower in HS compared in HS than TN goats at d 7 (P < 0.05), when effects of
with TN goats (Table 5; P < 0.01). The deceased pCO2 HS were more marked, but differences between treat-
and HCO3− under HS conditions agree with the results ment groups disappeared at d 28 (Figure 4; P > 0.05).
reported in dairy cows by Schneider et al. (1988). The Ametaj et al. (2005) indicated an association between
Journal of Dairy Science Vol. 96 No. 10, 2013
6364 HAMZAOUI ET AL.

Table 5. Metabolic and acid-base balance indicators of dairy goats under thermal neutral (TN; n = 8) and heat stress (HS; n = 8) conditions
at different daily hours in late lactation (values are least squares means and SEM)

TN HS Effect (P-value )

Item 0800 h1 1700 h2 0800 h3 1700 h3 SEM Treatment Hour T × H4

Glucose, g/L 3.21 3.41 3.04 3.44 0.05 0.197 0.001 0.057
Urea, mg/dL 40.4 30.0 36.5 28.3 2.5 0.331 0.001 0.498
Na, mmol/L 139.5 140.4 142.3 141.0 1.2 0.161 0.881 0.403
K, mmol/L 3.83 4.16 3.70 4.25 0.10 0.793 0.003 0.399
Cl, mmol/L 105.0 108.9 108.3 110.4 0.9 0.031 0.004 0.324
Hematocrit, % PCV5 18.63 17.75 18.25 16.63 0.63 0.372 0.003 0.296
Hemoglobin,6 mmol/L 6.33 6.03 6.23 5.64 0.21 0.363 0.002 0.187
pH 7.42 7.38 7.42 7.42 0.01 0.156 0.142 0.056
Total CO2,6 mmol/L 26.9 24.4 22.3 21.5 0.8 0.002 0.036 0.231
Anion gap,6 mmol/L 12.50 12.00 16.50 14.00 0.73 0.001 0.099 0.428
pCO2,7 mm of Hg 38.9 39.8 33.1 31.5 1.7 0.006 0.769 0.300
HCO3−,6 mmol/L 25.71 23.41 21.29 20.51 0.83 0.003 0.041 0.283
Base excess6 1.38 −1.75 −3.00 −4.00 0.87 0.005 0.019 0.193
Urine pH 9.09 8.94 8.91 8.85 0.08 0.108 0.215 0.559
1
Before changing from night (30.5°C and 40% humidity; THI = 77) to day (37°C and 40% humidity; THI = 85) conditions.
2
During the day (37°C and 40% humidity; THI = 85) conditions.
3
Indoor daily variation from 15 (night) to 20°C (day) at 45% relative humidity (THI = 59 to 65).
4
Treatment (T) × hour (H) interaction.
5
PCV = packed cell volume.
6
Calculated values by the i-STAT device software (Abbott Point of Care Inc., Princeton, NJ).
7
Partial pressure of CO2.

increased serum haptoglobin concentration and hepatic and HS goats would have been detected if fecal corti-
lipidosis in periparturient dairy cows. Hiss et al. (2009) costerone was measured earlier (i.e., at d 7 when HS
found that elevated haptoglobin concentrations in milk goats were suffering greater metabolic stress).
of dairy cows were associated with high NEFA values
in early lactation. This relationship was also observed CONCLUSIONS
in our HS goats.
It seems that HS goats at d 7 responded by increased Despite the reduction observed in feed intake of HS
circulating haptoglobin when they were metabolically goats, they produced similar milk yield to goats under
challenged, as evidenced by reduced feed intake (Fig- TN conditions in late lactation. However, milk protein
ure 2) and greater NEFA concentrations (Figure 3). content decreased in the HS goats, with no change
However, when goats were more adapted to heat stress in milk fat content. Heat-stressed goats had similar
conditions (i.e., d 28), haptoglobin levels returned to N retention to goats under TN conditions, indicating
values similar to TN goats. that the ingested N might have been directed to other

Corticosterone in Feces

Fecal corticosterone has been used to evaluate stress

in cows (Morrow et al., 2002). This approach is based
on the fact that glucocorticoids are secreted by the
adrenal gland after the activation of the hypothalamic-
pituitary-adrenal axis by a stressor. Circulating gluco-
corticoids are metabolized (conjugated) in the liver and
excreted via the urine and feces.
Corticosterone concentrations in feces did not vary
between TN and HS goats and averaged 4.28 ± 0.55
ng/g of DM. Similarly, HS did not increase plasma
cortisol concentrations in cows (Ei-Nouty et al., 1978)
and goats (Olsson and Dahlborn, 1989). Nevertheless, Figure 4. Haptoglobin concentrations in plasma of dairy goats
under thermal neutral (TN; □; n = 8) or heat stress (HS; ; n = 8)
it should be stressed that fecal samples were collected conditions at late lactation. Values are means, with SE indicated by
at d 35, and it is possible that differences between TN vertical bars. Means with different letters (a and b) differ (P < 0.05).

Journal of Dairy Science Vol. 96 No. 10, 2013

 HEAT STRESS EFFECTS ON PHYSIOLOGY AND LACTATION OF GOATS 6365

metabolic functions rather than milk protein synthesis. tidiuretic hormone of cattle exposed to 20 and 33°C. J. Dairy Sci.
61:189–196.
Further studies are needed to test the HS effects dur- Hirayama, T., K. Katoh, and Y. Obara. 2004. Effects of heat exposure
ing early lactation and to clarify whether reduced milk on nutrient digestibility, rumen contraction and hormone secretion
protein content is related to a nutrient-limiting factor of goats. Anim. Sci. J. 75:237–243.
Hiss, S., C. Weinkauf, S. Hachenberg, and H. Sauerwein. 2009. Rela-
or reduced mammary protein synthesis, as well as to tionship between metabolic status and the milk concentrations of
evaluate the effects on the coagulation properties of haptoglobin and lactoferrin in dairy cows during early lactation.
the milk. J. Dairy Sci. 92:4439–4443.
Huber, J. T., G. Higginbotham, R. A. Gomez-Alarcon, R. B. Taylor,
K. H. Chen, S. C. Chan, and Z. Wu. 1994. Heat stress, interactions
ACKNOWLEDGMENTS with protein, supplemental fat and fungal cultures. J. Dairy Sci.
77:2080–2090.
INRA (Institut National de la Recherche Agronomique). 2007. Alimen-
This work is part of a research project funded by tation des bovins, ovins et caprins. Quae, Paris, France.
the Spanish Ministry of Economy and Finance (Ma- Joshi, B. C., R. E. McDowell, and D. P. Sadhu. 1968. Effect of drugs
drid; Plan Nacional I+D+i; Project AGL2009-09376) on sweating rates in Hariana cattle. J. Dairy Sci. 51:905–909.
Kadzere, C. T., M. R. Murphy, N. Silanikove, and E. Maltz. 2002.
and was also supported by a research scholarship Heat stress in lactating dairy cows: A review. Livest. Prod. Sci.
to S. Hamzaoui from the International Centre for 77:59–91.
Advanced Mediterranean Agronomic Studies (Insti- Lough, D. S., D. K. Beede, and C. J. Wilcox. 1990. Effects of feed
intake and thermal stress on mammary blood flow and other
tuto Agronómico Mediterráneo de Zaragoza, Zaragoza, physiological measurements in lactating dairy cows. J. Dairy Sci.
Spain). The authors are also grateful to the team of 73:325–332.
Servei de Granges i Camps Experimentals (SGCE) of Maloiy, M. O., and C. R. Taylor. 1971. Water requirements of African
goats and haired-sheep. J. Agric. Sci. 77:203–208.
the Universitat Autònoma de Barcelona for the care of Masero, E. J., and P. D. Siegel. 1977. Acid-Base Regulation: Its Physi-
the animals. ology and Pathophysiology and the Interpretation of Blood-Gas
Analysis. 2nd ed. W. B. Saunders, Philadelphia, PA.
McDowell, R. E., E. G. Moody, P. J. Van Soest, R. P. Lehmann, and
REFERENCES G. L. Ford. 1969. Effect of heat stress on energy and water utiliza-
tion of lactating cows. J. Dairy Sci. 52:188–194.
Abdalla, E. B., E. A. Kotby, and H. D. Johnson. 1993. Physiological Mitra, R. G., G. I. Christison, and H. D. Johnson. 1972. Effect of
responses to heat-induced hyperthermia of pregnant and lactating prolonged thermal exposure on growth hormone (GH) secretion in
ewes. Small Rumin. Res. 11:125–134. cattle. J. Anim. Sci. 34:776–779.
Ametaj, B. N., B. J. Bradford, G. Bobe, R. A. Nafikov, Y. Lu, J. W. Morrow, C. J., E. S. Kolver, G. A. Verkerk, and L. R. Mathews. 2002.
Young, and D. C. Beitz. 2005. Strong relationships between media- Fecal glucocorticoid metabolites as a measure of adrenal activity in
tors of the acute phase response and fatty liver in dairy cows. Can. dairy cattle. Gen. Comp. Endocrinol. 126:229–241.
J. Anim. Sci. 85:165–175. NRC. 1971. A Guide to Environmental Research on Animals. National
AOAC International. 2003. Official Methods of Analysis of AOAC In- Academies Press, Washington, DC.
ternational. Vol. I. 17th ed. AOAC International, Gaithersburg, NRC. 2007. Nutrient Requirements of Small Ruminants, Sheep, Goats,
MD. Cervids, and New World Camelids. National Academies Press,
Baker, M. A. 1989. Effects of dehydration and rehydration on thermo- Washington, DC.
regulatory sweating in goats. J. Physiol. 417:421–435. Olsson, K., and K. Dahlborn. 1989. Fluid balance during heat stress in
Baumgard, L. H., and R. P. Rhoads Jr. 2013. Effects of heat stress lactating goats. Q. J. Exp. Physiol. 74:645–659.
on postabsorptive metabolism and energetics. Annu. Rev. Anim. Olsson, K., M. Josäter-Hermelin, J. Hossaini-Hilali, E. Hydbring,
Biosci. 1:311–337. and K. Dahlborn. 1995. Heat stress causes excessive drinking in
Bernabucci, U., P. Bani, B. Ronchi, N. Lacetera, and A. Nardone. fed and food deprived pregnant goats. Camp. Biochem. Physiol.
1999. Influence of short- and long-term exposure to a hot environ- 110:309–317.
ment on rumen passage rate and diet digestibility by Friesian heif- Pugh, D. G., and N. Baird. 2012. Sheep and Goat Medicine. Elsevier
ers. J. Dairy Sci. 82:967–973. Saunders, Maryland Heights, MO.
Bernabucci, U., N. Lacetera, and B. Ronchi. 2002. Effects of the hot Rhoads, M. L., R. P. Rhoads, M. J. VanBaale, R. J. Collier, S. R.
season on milk protein fractions in Holstein cows. Anim. Res. Sanders, W. J. Weber, B. A. Crooker, and L. H. Baumgard. 2009.
51:25–33. Effects of heat stress and plane of nutrition on lactating Holstein
Brown, D. L., S. R. Morrison, and G. E. Bradford. 1988. Effects of cows: I. Production, metabolism, and aspects of circulating so-
ambient temperature on milk production of Nubian and Alpine matotropin. J. Dairy Sci. 92:1986–1997.
goats. J. Dairy Sci. 71:2486–2490. Sano, H., K. Ambo, and T. Tsuda. 1985. Blood glucose kinetics in
Calamari, L., F. Abeni, F. Calegari, and L. Stefanini. 2007. Metabolic whole body and mammary gland of lactating goats exposed to
conditions of lactating Friesian dairy cows during the hot season heat stress. J. Dairy Sci. 68:2557–2564.
in the Po valley. 2. Blood minerals and acid-base chemistry. Int. Schneider, P. L., D. K. Beede, and C. J. Wilcox. 1988. Nycterohemeral
J. Biometeorol. 52:97–107. patterns of acid-base status, mineral concentrations and digestive
Collier, R. J., C. M. Stiening, B. C. Pollard, M. J. VanBaale, L. H. function of lactating cows in natural or chamber heat stress envi-
Baumgard, P. C. Gentry, and P. M. Coussens. 2006. Use of gene ex- ronments. J. Anim. Sci. 66:112–125.
pression microarrays for evaluating environmental stress tolerance Shwartz, G., M. L. Rhoads, M. J. VanBaale, R. P. Rhoads, and L. H.
at the cellular level in cattle. J. Anim. Sci. 84(E. Suppl.):E1–E13. Baumgard. 2009. Effects of a supplemental yeast culture on heat-
Constable, P. D. 1999. Clinical assessment of acid-base status: Strong stressed lactating Holstein cows. J. Dairy Sci. 92:935–942.
ion difference theory. Vet. Clin. North Am. Food Anim. Pract. Silanikove, N. 2000. The physiological basis of adaptation in goats to
15:447–471. harsh environments. Small Rumin. Res. 35:181–193.
Ei-Nouty, F. D., I. M. Elbanna, and H. D. Johnson. 1978. Effect of West, J. W. 2003. Effects of heat-stress on production in dairy cattle.
adrenocorticotropic hormone on plasma glucocorticoids and an- J. Dairy Sci. 86:2131–2144.

Journal of Dairy Science Vol. 96 No. 10, 2013