- Email: [email protected]
Evaporative Cooling of Ventral Regions of the Skin in Heat-Stressed Laying Hens
Evaporative Cooling of Ventral Regions of the Skin in Heat-Stressed Laying Hens D. Wolfenson,1 D. Bachrach,2 M. Maman, Y. Graber, and I. Rozenboim Department of Animal Science, Faculty of Agricultural, Food and Environmental Quality Sciences, The Hebrew University, Rehovot 76100, Israel temperature and respiration rate were higher and skin temperatures were lower than in ventrally cooled hens. During 10 d of heat exposure, ventrally cooled hens maintained egg weight and shell index (mg/cm2), whereas their food intake decreased moderately. In contrast, egg weight, shell index, and food intake all decreased markedly in uncooled or dorsally cooled hens. Transient alterations in plasma concentrations of corticosterone, progesterone, and estradiol were noted in uncooled and dorsally cooled hens but not in ventrally cooled hens. Results indicate that ventral cooling is an efficient method to alleviate heat stress in laying hens during summer. Successful implementation of ventral cooling in poultry houses will depend on optimal installation of sprinklers and on minimal wetting of manure.
(Key words: hen, heat stress, cooling, hyperthermia, egg production) 2001 Poultry Science 80:958–964
a result of the high rate of egg production (Blem, 2000). Second, there is little heat dissipation by convection and radiation, because of the very effective insulation of the body surface by their feathers. Third, the hens lack sweat glands, and their respiratory water evaporation rate is not high enough to maintain normothermia at high ambient temperatures (Etches et al., 1995; Dawson and Whittow, 2000). Several approaches have been examined over the years to alleviate heat stress in domestic fowl (Smith, 1981; Carr and Carter, 1985) and include roof insulation, orientation of buildings to maximize natural ventilation, and installation of fans to increase ventilation. Measures to facilitate heat conduction, which are not used commercially in poultry farms, include provision of cool drinking water (Beker and Teeter, 1994; Van Kampen, 1988), immersion of the hens’ legs in cool water (Van Kampen, 1988), and provision of water-cooled floor-level perches inside the cages (Muiruri and Harrison, 1991; Reilly et al., 1991). Evaporation of water could be used to cool hens directly or indirectly. Indirect cooling of the microclimate surrounding the birds is achieved by means of a highpressure fogging system: water evaporates rapidly before it reaches the birds, from a fine mist of minute droplets, which cools the air in the poultry house (Carr and Carter,
INTRODUCTION Laying hens held in naturally ventilated poultry houses in hot countries often develop hyperthermia during the summer (Etches et al., 1995). As a result, layer performance deteriorates during heat stress. The deterioration is manifested as reduced feed intake and altered nutrient absorption, respiratory alkalosis, decreased blood flow to several organs, and endocrine changes affecting steroid hormones secretion, etc. (Hillman et al. 1985; Etches et al., 1995). Examples of low productive and reproductive performance of laying hens in summer include low egg weight, low egg production, and high eggshell breakage (De-Andrade et al., 1976, 1977; Wolfenson et al., 1979; Tanor et al., 1984; Marsden and Morris, 1987; Sauveur and Picard, 1987; Roland, 1988). Laying hens are susceptible to heat stress for several reasons. First, their metabolic heat production is high as
2001 Poultry Science Association, Inc. Received for publication September 25, 2000. Accepted for publication March 13, 2001. 1 To whom correspondence should be addressed: [email protected]. 2 Present address: Poultry Department, Extension Services, The Ministry of Agriculture, Israel.
958
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
ABSTRACT Laying hens held in battery cages in naturally ventilated poultry houses in hot countries usually develop hyperthermia, which adversely affects their performance. The present means of cooling alleviate to some degree, but cannot eliminate, the stress imposed by heat. A new approach to cooling of laying hens was developed, based on wetting the skin and promoting evaporation of water from the ventral regions of the bird. The type of plumage in the ventral regions and the exposed skin of the apteria enable more efficient wetting than is possible with dorsal cooling. A ventral cooling regime, comprising an initial period of frequent wettings followed by intermittent wetting for 10 s every 30 min was able to maintain normothermia of laying hens subjected to a 10-h period of heat exposure. Dorsal cooling was less efficient; body
VENTRAL COOLING OF HENS DURING HEAT STRESS
MATERIALS AND METHODS Experimental Birds The experiments were conducted on White Leghorn (Loman) laying hens that had been obtained from a commercial hatchery3 and that were held in individual battery cages in two identical, adjacent, temperature-controlled chambers. Hens brought from the experimental poultry house to the chambers were allowed a 1-wk adaptation period before experiments started. The hens were kept at 24 C and 40% RH, were exposed to 17 h of light, and were fed ad libitum with a commercial layer feed containing 2,800 kcal/kg, 18% protein, and 3.7% calcium. In the first set of short-term experiments, the optimal conditions for ventral cooling were determined through a series of 10-h heat exposures. Hens (n = 24) were in their second production period and had an average laying rate of 75%. In a second set of long-term experiments, the productive and endocrine responses of hens exposed to ventral cooling, under the optimal protocol established in the short-term experiments, were examined in two 1-
3
Kibbutz Hafetz-Haim, 76817 Israel. Naan, model 7955, Kibbutz Naan, 76829 Israel. 5 YSI, model 4000, Yellow Springs, OH 45487. 6 Novasina G Hygrometer, MIK 3000, CH 8808 Pfaffikon, Switzerland. 4
mo experiments. The hens (n = 24 in each experiment) were in their first production period and had an average laying rate of 92%. The experiments were conducted according to the guidelines of the local committee for animal care and welfare.
Developing the Ventral Cooling System In two, six-cage batteries, one of which was located in each chamber, static sprinklers4 were installed below the food trays; each sprinkler covered two cages. Various water outputs of the sprinklers were examined: 30, 43, 54, and 85 L/h, with a covering angle of 90°. The positioning of the sprinklers was adjusted to obtain maximal wetting of the hens (n = 12). The dorsally positioned sprinklers were installed 50 cm above the hens (n = 12) in the other two, six-cage batteries, one of which was located in each chamber, on the side remote from the ventrally cooled batteries. Care was taken to eliminate any cross wetting of hens in either treatment by the sprinklers in the other treatment. A set of short-term experiments was performed twice weekly, in each of which the hens were exposed to 38 C and 40% RH for 10 h. The temperature in the chambers was gradually elevated from 24 C at a rate of 5 C/h, starting at 0700. Intervals between heat exposures served as recovery periods. To determine the optimal cooling procedure, various water capacities and duration and frequency of wettings were examined. In addition, wetting regimes with or without more frequent wetting at the start of heat exposure were examined. In each of the 10h heat exposures, body temperatures were monitored,5 at an accuracy of 0.1 C, by means of a thermistor inserted into the cloaca. Skin temperatures were recorded with unilateral flat thermistors attached to two feathered skin areas on the breast and back and to two featherless skin areas of the comb and metatarsus. Respiration rates were monitored, and air temperature and relative humidity6 were recorded frequently.
Performance and Endocrine Responses Two long-term experiments were conducted. In the first, the responses of hens exposed to ventral cooling were compared with those of uncooled control hens. In the second, they were compared with those of hens exposed to dorsal cooling. The experimental protocol was the same in both experiments. Responses during 10 d of heat exposure (37 C and 40% RH) were compared with those during 10 d before and 10 d after, when the hens were under normothermic conditions (24 C and 40% RH). During the 10 d of heat stress, air temperature was cycled daily—high from 0700 to 1700 and normothermic thereafter—to mimic the fluctuation of air temperatures during a natural 24-h cycle. Individual feed intake was recorded daily to an accuracy of ± 1 g. Egg weight was recorded daily to an accuracy of ± 0.1 g. Eggshell weight was recorded after drying the shell at 70 C for 18 h. The shell index, which relates
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
1985; Ernst, 1995). Direct cooling is achieved by means of a sprinkling system that wets the skin of the hens with large droplets so that the evaporating water cools the birds (Ernst, 1995). The latter system is used in many commercial, naturally ventilated poultry houses in hot countries. Because the sprinklers are located above the cages, the hens are wetted over their dorsal hemisphere and, therefore, experience dorsal cooling. Although this approach alleviates heat stress to some extent and lowers hen mortality during hot spells, it has several limitations: 1) the decrease in productivity during summer is not eliminated, 2) excess water wets the manure and the feed mixture in the trays, and 3) the dense and effectively insulating feathers on the birds’ wings and backs limit penetration of water to the skin over the dorsal hemisphere. Another possible approach that has not been examined before involves cooling laying hens by wetting the ventral hemisphere of the body thus providing them with ventral cooling. The main advantage of ventral cooling, compared with dorsal cooling comes from the type of feathers in the ventral areas. The plumage there is less dense than in the dorsal regions, and several areas are apteria (Lucas and Stettenheim, 1972), which eases the access of water to the skin. The aim of the present study was to test the efficacy of ventral cooling applied to laying hens held in naturally ventilated battery cages. We compared thermal responses, performances, and circulating hormone concentrations of ventrally cooled, uncooled, and dorsally cooled hens.
959
960
WOLFENSON ET AL.
shell weight and egg surface area, was calculated from egg weight by means of the equation I = a/3.9782 × b0.7056, where I = shell index (mg/cm2), a = egg shell weight (mg), and b = egg weight (g; Abdallah et al., 1993). Blood samples were collected from the brachial vein 2 d before heat stress started, 2 and 10 d after it started, and 4 d after it ended. Plasma was separated and stored at −20 C. Plasma osmolality7 was determined with a calibrated osmometer.
Hormone Analyses
Data Analysis Data were analyzed by ANOVA, by means of the general linear models procedure of the Statistical Analysis System software (SAS, 1987). Data are presented as means and standard errors. Data were analyzed in models based on within-hen repeated measurements. The statistical models included experimental groups (ventrally cooled vs. dorsally cooled or uncooled groups), hens (nested within group), time (hours or days in short- or long-term experiments, respectively), period (in long-term experiments, the 10-d periods before, during, and after heat stress), and interactions. Data were analyzed separately for Experiments 1 and 2. Hens (within group) served as an error term for the effect of experimental group. Time or period effects and their interactions were tested against the residual error. Skin temperatures data were analyzed by one-way ANOVA.
7
Osmette, 2007, Precision System, Sudbury, MA 01776. Donated by F. Kohen, Department of Hormone Research, Weizmann Institute, Rehovot, 76100 Israel. 9 Sigma Chemical Co., St. Louis, MO 63178-9916. 8
FIGURE 1. Body temperatures of dorsally and ventrally cooled laying hens during heat exposure. n = 12 in each group, pooled SEM −0.2 C. Dorsal and ventral temperatures differ; P < 0.01.
RESULTS The Ventral Cooling System The optimal protocol of ventral cooling with the minimal amount of water necessary to prevent hyperthermia was determined in a series of 10-h experiments. The following is a brief description of the short-term experiment, to the point at which the optimal protocol was established (data not shown). We initially found that although hyperthermia was prevented, excessive water was used when 54- or 85-L/h sprinklers were operated for 30 s every 60 min. Therefore, we increased the interval between sprinklings to 90 min and used low-capacity sprinklers of 30L/h for 120 s each time. However, the 90-min interval was found to be too long and an application rate of 30 L/h seemed to be inadequate. Consequently, we introduced a new protocol, which began with three 30-s wettings via 43-L/h sprinklers at 5-min intervals, to achieve pronounced wetting of the feathers, which was followed by a 10-s wetting every 20 min for 10 h. This protocol prevented hyperthermia in ventrally cooled hens, whereas body temperatures of uncooled hens increased by 1.5 C and of dorsally cooled hens increased by 0.5 to 1.0 C. To reduce the amount of water used, the duration of the initial wetting was shortened from 30 to 10 s and the interval between sprinklings was increased from 20 to 45 min. However, this protocol did not maintain normothermia in ventrally cooled hens. Finally, the optimal protocol for ventral cooling was found to be three initial 30-s wettings of 43 L/h at 5-min intervals followed by a 10-s wetting every 30 min for 10 h. A difference of 0.4 to 0.7 C in body temperature between ventral and dorsal cooling was consistently achieved (P < 0.01; Figure 1). This ventral cooling protocol was later used in the long-term production experiments. Figure 2 presents the average of three measurement days in which body temperatures and respiration rates of ventrally and dorsally cooled hens were recorded between 5 and 7 h after the start of heating, during the 10-d heat exposure of Experiment 2. Ventrally cooled hens maintained body temperatures below 42 C
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
Concentrations of estradiol, progesterone, and corticosterone in plasma were analyzed by previously described and validated dextran-coated charcoal RIA (Meidan et al., 1990; Wolfenson et al., 1997, Heiblum et al. 2000), using specific antibodies for estradiol and progesterone8 and for corticosterone.9 Plasma samples were extracted by diethyl ether according to the procedure described by Badinga et al. (1992). Recovery rates were 80% for progesterone and 75% for estradiol and corticosterone. Assay sensitivities were 1.0 pg/tube for estradiol, 1.95 pg/tube for progesterone, and 7.8 pg/tube for corticosterone. Intra- and interassay CV for estradiol were 6.4 and 13.9%, for progesterone were 6.8 and 11.5%, and for corticosterone were 8.8 and 11.6%. Antibody for estradiol-17β cross-reacted with estrone (2.5%), estriol (0.95%), progesterone (< 0.2%), and estradiol-17α (1.5%). Antibody for progesterone cross-reacted with estradiol, testosterone, and corticosterone (< 0.3%); 17α-progesterone (4%); and 20α-progesterone (1.5%). Antibody for corticosterone cross-reacted with progesterone (7.5%), testosterone (3.8%), estradiol (< 0.1%), and 17αprogesterone (1.2%).
VENTRAL COOLING OF HENS DURING HEAT STRESS
FIGURE 2. Mean values of body temperatures (empty bars) and respiration rates (dotted bars) of dorsally and ventrally cooled hens, measured between 5 and 7 h after start of heating, on three different days during the 10-d heat exposure of experiment 2. n = 12 in each group, pooled SEM −0.1 C and 15 breaths per min. Dorsal and ventral values differ; P < 0.01.
Production Responses to Ventral Cooling Experiment 1. This experiment compared ventrally cooled and uncooled birds. During 10 d of heat stress, egg weight in uncooled hens decreased by about 8% (P < 0.01) and did not return to normal values after heat stress ended (Table 1). In contrast, ventrally cooled hens maintained stable egg weights during heat stress, with a
FIGURE 3. Skin temperatures in four regions of dorsally and ventrally cooled hens and uncooled hens, 5 h after start of heating (n = 6, 12, and 6, respectively). Pooled SEM −0.4 C. Mean values among cooling treatments, of a specific skin region with no common lowercase letter differ significantly (P < 0.05).
slightly elevated egg weight after heat stress ended (P < 0.01). Feed intake during heat stress was 15 and 33% lower in ventrally cooled and uncooled hens, respectively (P < 0.01; Table 1). After heat stress ended, intake returned to normal in both groups, but the ventrally cooled group attained a higher recovery value (P < 0.01). Shell index was unaffected during heat stress in ventrally cooled hens but was lowered (P < 0.01) in uncooled hens, in which it returned to normal after heat stress ended (Figure 4). Plasma osmolality was unaffected by heat stress in both groups (303.1 ± 4.3 mOsm/kg). Experiment 2. In contrast to the first experiment, this experiment compared ventrally and dorsally cooled hens. In Experiment 2, as occurred in Experiment 1, egg weight was not affected by heat stress in ventrally cooled hens (Table 1). In contrast, it was lowered by 11% during heat stress in dorsally cooled hens (P < 0.01); it was slightly improved after heat stress ended but did not return to the level of preheat stress (P < 0.01). As in Experiment 1, food intake was 18 and 32% lower in ventrally cooled and dorsally cooled hens, respectively (P < 0.01, within group compared with preheating values and between groups during heat stress; Table 1). After heat stress ended, both groups returned to preheat values. Shell index was lowered during heat stress in the ventrally cooled and dorsally cooled groups (P < 0.01), but the decrease was minor in the former and pronounced in the latter group (P < 0.01; Figure 4). In both groups, the shell index returned to normal value after heat stress ended. As in Experiment 1, plasma osmolality was not affect by heat stress, in either group, and was within the normal range for laying hens (287 ± 2.9 mOsom/kg). This result indicated that the hens were not dehydrated during heat exposure.
Steroid Concentrations in Blood Plasma Generally speaking, alterations in plasma concentrations of corticosterone, progesterone, and estradiol during heat stress were minor. Plasma concentration of corticosterone was not affected by heat stress in ventrally cooled hens in Experiment 1 or 2 (Table 2); it increased in uncooled hens in the early stages of heat stress (P < 0.05; Experiment 1) and decreased in dorsally cooled hens in the late stages of heat stress (P < 0.05; Experiment 2). Plasma estradiol concentrations were not altered during heat stress in ventrally cooled hens; in uncooled and dorsally cooled hens, it decreased in the early stage of heat stress (P < 0.01) and later returned to normal. Selected plasma samples were included in the analysis of progesterone concentration. Because time of oviposition was not recorded in the present study, some blood samples that were taken routinely during the morning hours could have been drawn during the preovulatory progesterone surge. Therefore, in accordance with Bahr and Johnson (1991) and others, all samples with more than 1 ng/mL were excluded from the data set. Elimination of those samples facilitated a reliable comparison of the effect of heat stress on basal concentrations of the hormone.
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
and respiration rates below 100 breaths/min, much lower than those of their dorsally cooled counterparts (P < 0.01). Skin temperatures of ventrally cooled hens (n = 12) were compared with those of dorsally cooled (n = 6) and uncooled hens (n = 6), 5 h after the start of heating (Figure 3). In ventrally cooled hens, the temperatures of feathered skin of the breast and back were 1.0 to 1.5 C higher than those of dorsally cooled hens (P < 0.05); the temperatures of the featherless skin of the comb and metatarsus did not differ between the two groups. Skin temperatures of uncooled hens were higher at all four skin points than those of dorsally cooled hens (P < 0.05) and were higher in the two feathered regions than in the ventrally cooled hens (P < 0.05).
961
962
WOLFENSON ET AL. TABLE 1. Egg weight and feed intake during 10-d periods before, during, and after heat stress1 Experiment 1
Egg weight Before During After Feed intake Before During After
Experiment 2
Vental cooling
No cooling
SEM
Ventral cooling
Dorsal cooling
SEM
63.9b 63.1b 65.1a
62.3a 57.2b 57.2b
0.5* 0.4** 0.5**
65.2 64.8 65.7
66.2a 58.7c 61.8b
0.4 0.5** 0.6**
114a 97b 119a
113a 76b 110a
3 2** 3**
106a 88b 106a
106a 73b 101a
2 3** 3
Means within columns with no common superscript differ significantly (P < 0.01). Responses of ventrally-cooled hens were compared with uncooled hens in Experiment 1 and with dorsally cooled hens in Experiment 2 (n = 12 hens per group). *P < 0.05, **P < 0.01 indicate significant differences between groups at a specific time. a–c 1
DISCUSSION This study presents, for the first time, a new approach to cooling laying hens held in naturally ventilated cages. Wetting of the ventral regions, and evaporation of water from them, was found to be highly effective in alleviating
FIGURE 4. Shell index (mg/cm2) of ventrally vs. uncooled hens (upper panel, Experiment 1) and of ventrally vs. dorsally cooled hens (lower panel, Experiment 2), before, during, and after a 10-d heat exposure. n = 12 in each group, pooled SEM in both experiments, 2 mg/ cm2. For significance detail, see text.
heat stress. The initial wetting at the start of heat exposure effectively saturated the feathered layer with water. It created a relatively large reservoir of water in the feathers of the breast, thigh, rump, abdomen, and legs, enabling a high rate of water evaporation from the beginning of heat exposure. The type of plumage on the ventral regions of the skin, unlike the large feathers of the wings and tail, facilitates water accumulation. In addition, wetting interrupted the arrangement of the barbs and barbules of the feather vanes and of the plumulae. This feather treatment decreased the thermal insulation of the body covering and narrowed the boundary layer of warm air surrounding the body, resulting, probably, in facilitated convective, as well as evaporative, heat loss. After the initial wetting, a minimal period of 10 s sprinkling every 30 min was sufficient to maintain the wetness of the feather coat and to ensure a high cooling rate, which is in agreement with the study by McArthur and Ousey (1994), who showed a stable rate of evaporative heat loss over a period of 100 min after wetting of a physical model maintained at 38 C. Such a short period of sprinkling is crucial to minimize wetting of the manure. The results clearly demonstrated that laying hens exposed to severe heat stress, and which were ventrally cooled, were able to maintain their body temperature within the normothermic range most of the time, 0.4 to 0.7 C lower than that of dorsally cooled hens. Cooling of the skin by water evaporation may cause some degree of vasoconstriction, which may subsequently lower the skin temperature and, consequently, attenuate convective and radiative heat losses. Interestingly, all four skin temperatures of dorsally cooled hens were 2 C lower than those of uncooled hens. Those temperatures of ventrally cooled hens were lower, by only 1 C in only two out of four skin points, than those of uncooled hens. Local cooling of thermoregulatory centers in the brain and spinal cord may occur during dorsal cooling and could be associated with the lower skin temperatures recorded in this group. This hypothesis is consistent with the finding that local cooling of the hypothalamus and spinal cord depressed the panting responses in domestic fowls under heat stress (Darre and Harrison. 1981; Hill-
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
In Experiments 1 and 2, basal values of progesterone were not altered during heat stress in ventrally cooled hens but were lower (P < 0.01) in uncooled and dorsally cooled hens in the late stage (Day 10) of heat stress.
963
VENTRAL COOLING OF HENS DURING HEAT STRESS TABLE 2. Plasma concentrations (ng/mL) of steroid hormones at 2 d before, at 2 d and 10 d during, and at 4 d after heat stress1 Experiment 1
Corticosterone 2 d before 2 d during 10 d during 4 d after Progesterone2 2 d before 2 d during 10 d during 4 d after
Ventral cooling
No cooling
SEM
Ventral cooling
Dorsal cooling
SEM
1.61 1.34 1.38 1.53
1.7 2.19 1.05 1.19
0.28 0.26* 0.26 0.27
1.03 1.59 1.51 1.47
1.75 2.15 1.10 1.52
0.24 0.27 0.24* 0.25
0.61 0.44 0.50 0.44
0.54 0.33 0.33 0.33
0.07 0.06 0.06* 0.06
0.32 0.26 0.44 0.18
0.38 0.31 0.17 0.19
0.04 0.05 0.05** 0.05
0.18 0.17 0.20 0.17
0.19 0.09 0.16 0.19
0.01 0.02** 0.02 0.02
0.17 0.13 0.18 0.23
0.18 0.08 0.13 0.22
0.01 0.02** 0.02 0.02
1 Responses of ventrally cooled hens were compared with uncooled hens in Experiment 1 and with dorsally cooled hens in Experiment 2. 2 Progesterone concentrations of basal secretion only are presented (see text for detail). *P < 0.05, **P < 0.01 indicate significant differences between groups at a specific time.
man et al., 1985). Low skin temperatures may affect the hens’ ability to maintain thermal balance. In the present study, the skin temperatures of dorsally cooled hens, i.e., about 38 C, equaled the air temperatures in the chambers, indicating that nonevaporative heat loss was nil in this group (Dawson and Whittow, 2000). In contrast, ventrally cooled hens maintained a 1 to 1.5 C difference between skin and air temperatures, enabling some degree of convective and radiative heat dissipation. The stability of skin temperatures in ventrally cooled hens during heat stress could be associated with two unique characteristics of the vascular system of domestic fowls. First, the majority of the rise in metatarsal and toe blood flow in hyperthermic hens is due primarily to a rise in arteriovenous anastomoses flow and not to an increase in capillary blood flow (Wolfenson, 1983). This rise, and the fact that arteriovenous anastomoses blood flow is regulated by central stimuli of thermoregulatory neurones and not by peripheral stimuli (Hales et al., 1978), may explain the high blood supply even during local evaporative cooling of the skin. Second, capillary blood flow in the ventral skin regions was previously found to increase sevenfold, compared with a twofold increase in the dorsal skin regions in heat-stressed hens (Wolfenson et al., 1981). The relatively higher blood flow to the ventral skin regions contributes to the stability of skin temperatures during ventral cooling. The production responses of the ventrally cooled hens were unaffected by heat stress in Experiments 1 and 2. In contrast, the uncooled and dorsally cooled hens showed decreased egg weights and reduced eggshell quality; the decrease being more pronounced in the former, as expected. Stability in the responses of ventrally cooled hens was evident, even though their feed intake was lowered during the 10-d heat exposure. Decreased egg weight could be associated, at least in part, with decreased feed
intake under heat stress (De-Andrade et al., 1977). However, alterations in the shell index cannot be directly related to calcium deficiency induced by decreased food intake, because, according to our calculations, calcium was supplied in excess in the layer diet, and hens may cope with a 30% decline in food intake. In agreement, De-Andrade et al. (1977) pointed out that low shell quality under heat stress is not related to nutritional deficits. Others have indicated that low shell quality is associated with the decrease in the plasma concentration of bicarbonate (Odom et al., 1985) or with depressed blood supply to the shell gland under heat stress (Wolfenson et al., 1981). The endocrine responses in this study were less consistent than the thermal and the production responses. A transient rise in plasma concentration of corticosterone was observed in early stages of heat exposure in uncooled hens, and concentration returned to a normal range later, even though hens were still under stress conditions. This pattern of corticosterone secretion was described before (Etches et al., 1995; Siegel, 1995). Decreases in basal concentrations of plasma progesterone were noted at the end of the 10-d heat exposure in dorsally cooled and uncooled hens but not in ventrally cooled hens. It has been suggested (Novero et al., 1991) that a decrease in progesterone secretion was related to a heat-stress-induced alteration in the steroidogenic capacity of granulosa cells. The fact that ventral cooling successfully alleviated heat stress corresponded well with the stability of plasma progesterone concentrations. The decrease in plasma concentration of estradiol was transient, which was noted in the earlier stages of heat stress (Day 2), whereas the decrease in plasma progesterone concentration was noted later during heat exposure (Day 10). The above indicates that a lack of progesterone secretion was not the reason for the decline in estradiol secretion.
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
Estradiol 2 d before 2 d during 10 d during 4 d after
Experiment 2
964
WOLFENSON ET AL.
In conclusion, ventral cooling was found to be an efficient means of alleviating the heat stress of laying hens housed in battery cages. The results clearly indicate the advantages of ventral cooling over dorsal cooling in terms of the hen abilities to maintain normothermia, a high level of productivity and normal steroids secretion. The type of plumage and the unfeathered areas in the ventral regions and the relatively high blood supply to the ventral skin all contribute to ventral cooling being superior to dorsal cooling. Another advantage of ventral cooling over dorsal cooling is that it hardly wets the feed mixture. Successful implementation of ventral cooling in commercial, naturally ventilated poultry houses will depend on optimal installation of the sprinklers and on the ability to minimize wetting of the manure.
REFERENCES
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
Abdallah, A. G., R. H. Harms, and O. El-Husseiny, 1993. Various methods of measuring shell quality in relation to percentage of cracked eggs. Poultry Sci. 72:2038–2043. Badinga, L., M. A. Driancourt, J. D. Savio, D. Wolfenson, M. Drost, R. L. de la Sota, and W. W. Thatcher, 1992. Endocrine and ovarian responses associated with the first-wave dominant follicle in cattle. Biol. Reprod. 47:871–883. Bahr, J. M., and P. A. Johnson, 1991. Reproduction in poultry. Pages 517–575 in: Reproduction in Domestic Animals. P. T. Cupps, ed. Academic Press, New York, NY. Beker, A., and R. G. Teeter, 1994. Drinking water temperature and potassium chloride supplementation effects on broiler body temperature and performance during heat stress. J. Appl. Poult. Res. 3:87–92. Blem, C. R., 2000. Energy balance. Pages 327–341 in: Sturkie’s Avian Physiology. 5th ed. G. C. Whittow, ed. Academic Press, New York, NY. Carr, L., and T. Carter, 1985. Housing and management of poultry in hot and cold climates. Pages 73–110 in: Stress Physiology in Livestock. Poultry. Vol. 3. M. K. Yousef, ed. CRC Press, Boca Raton, FL. Darre, M. J., and P. C. Harrison, 1981. The effects of heating and cooling localized areas of the spinal cord and brain stem on thermoregulatory responses of the domestic fowl. Poultry Sci. 60(Suppl. 1):1644. (Abstr.). Dawson, W. R., and G. C. Whittow, 2000. Regulation of body temperature. Pages 343–390 in: Sturkie’s Avian Physiology. 5th ed. G. C. Whittow, ed. Academic Press, New York, NY. De-Andrade, A. N., J. C. Rogler, and W. R. Featherston, 1976. Influence of constant elevated temperature and diet on egg production and shell quality. Poultry Sci. 55:685–693. De-Andrade, A. N., J. C. Rogler, W. R. Featherston, and C. W. Alliston, 1977. Interrelationships between diet and elevated temperatures (cyclic and constant) on egg production and shell quality. Poultry Sci. 56:1178–1188. Ernst, R. A., 1995 Housing for improved performance in hot climates. Pages 67–99 in: Poultry Production in Hot Climates. N. J. Daghir, ed. CAB International, Cambridge, UK. Etches, R. J., T. M. John, and G.A.M. Verrinder-Gibbins, 1995. Behavioural, physiological, neuroendocrine and molecular responses to heat stress. Pages 31–66 in: Poultry Production in Hot Climates. N. J. Daghir, ed. CAB International, Cambridge, UK. Hales, J.R.S., M. Iriki, K. Tsuchiya, and E. Kozawa, 1978. Thermally-induced cutaneous sympathetic activity related to blood flow through capillaries and arteriovenous anastomoses. Eur. J. Physiol. 375:17–24. Heiblum, R., E. Arnon, G. Gvaryahu, B. Robinzon, and N. Snapir, 2000. Short-term stress increases testosterone secre-
tion from testes in male domestic fowl. Gen. Comp. Endocrinol. 120:55–66. Hillman, P. E., N. R. Scott, and A. van Tienhoven, 1985. Physiological responses and adaptations to hot and cold environments. Pages 1–71 in: Stress Physiology in Livestock. Poultry. Vol 3. M. K. Yousef, ed. CRC Press, Boca-Raton, FL. Lucas, A. M., and P. R. Stettenheim, 1972. Pages 467–552 in: Avian Anatomy. Integument. Part 2. US Government Print Office, Washington, DC. Marsden, A., and T. R. Morris, 1987. Quantitative review of the effects of environmental temperature on food intake, egg output and energy balance in laying pullets. Br. Poult. Sci. 28:693–704. McArthur, A. J., and J. C. Ousey, 1994. Heat loss from a wet animal: changes with time in the heat balance of a physical model representing a newborn homeotherm. J. Therm. Biol. 19:81–89. Meidan, R., E. Girsh, O. Blum, and E. Aberdam, 1990. In vitro differentiation of bovine theca and granulosa cells into small and large luteal-like cells: Morphological and functional characteristics. Biol. Reprod. 43:913–921. Muiruri, H. K., and P. C. Harrison, 1991. Effect of peripheral foot cooling on metabolic rate and thermoregulation of fed and fasted chicken hens in a hot environment. Poultry Sci. 70:74–79. Novero, R. P., M. M. Beck, and E. W. Gleaves, 1991. Plasma progesterone, luteinizing hormone concentrations, and granulosa cell responsiveness in heat-stressed hens. Poultry Sci. 70:23. Odom, T. W., P. C. Harrison, and M. J. Darre, 1985. The effects of drinking carbonated water on the egg shell quality of Single Comb White Leghorn hens exposed to high environmental temperature. Poultry Sci. 64:594–596. Reilly, W. M., K. W. Koelkebeck, and P. C. Harrison, 1991. Performance evaluation of heat-stressed commercial broilers provided water-cooled floor perches. Poultry Sci. 70:1699– 1703. Roland, D. A., 1988. Research note: egg shell problems: estimates of incidence and economic impact. Poultry Sci. 67:1801–1803. SAS Institute, 1987. SAS/STAT User’s Guide. 6th edition. SAS Institute Inc, Cary, NC. Sauveur, B., and M. Picard, 1987. Environmental effects on egg quality. Pages 219–234 in: Egg quality—Current Problems and Recent Advances. R. G. Wells and C. G. Belyavin, ed. Butterworths, London, UK. Siegel, H. S., 1995. Stress, strain and resistance. Br. Poult. Sci. 36:3–22. Smith, W. K., 1981. Poultry housing problems in the tropics and subtropics. Pages 235–258 in: Environmental Aspects of Housing for Animal Production. J. A. Clark, ed. Butterworths, London, UK. Tanor, M. A., S. Leeson and J. D. Summers, 1984. Effect of heat stress and diet composition on performance of white leghorn hens. Poultry Sci. 63:304–310. van Kampen, M., 1988. Effects of drinking water temperature and leg cooling on heat stress of laying hens (Gallus domesticus). J. Therm. Biol. 13:43–47. Wolfenson, D., 1983. Blood flow through arteriovenous anastomoses and its thermal function in the laying hen. J. Physiol. 334:395–407. Wolfenson, D., Y. F. Frei, N. Snapir, and A. Berman, 1979. Effect of diurnal or nocturnal heat stress on egg formation. Br. Poult. Sci. 20:167–174. Wolfenson, D., Y. F. Frei, N. Snapir, and A. Berman, 1981. Heat stress effects on capillary blood flow and its redistribution in the laying hen. Eur. J. Physiol. 390:89–93. Wolfenson, D., B. J. Lew, W. W. Thatcher, Y. Graber, and R. Meidan, 1997. Seasonal and acute heat stress effects on steroid production by dominant follicles in cows Anim. Reprod. Sci. 47:9–19.
(Key words: hen, heat stress, cooling, hyperthermia, egg production) 2001 Poultry Science 80:958–964
a result of the high rate of egg production (Blem, 2000). Second, there is little heat dissipation by convection and radiation, because of the very effective insulation of the body surface by their feathers. Third, the hens lack sweat glands, and their respiratory water evaporation rate is not high enough to maintain normothermia at high ambient temperatures (Etches et al., 1995; Dawson and Whittow, 2000). Several approaches have been examined over the years to alleviate heat stress in domestic fowl (Smith, 1981; Carr and Carter, 1985) and include roof insulation, orientation of buildings to maximize natural ventilation, and installation of fans to increase ventilation. Measures to facilitate heat conduction, which are not used commercially in poultry farms, include provision of cool drinking water (Beker and Teeter, 1994; Van Kampen, 1988), immersion of the hens’ legs in cool water (Van Kampen, 1988), and provision of water-cooled floor-level perches inside the cages (Muiruri and Harrison, 1991; Reilly et al., 1991). Evaporation of water could be used to cool hens directly or indirectly. Indirect cooling of the microclimate surrounding the birds is achieved by means of a highpressure fogging system: water evaporates rapidly before it reaches the birds, from a fine mist of minute droplets, which cools the air in the poultry house (Carr and Carter,
INTRODUCTION Laying hens held in naturally ventilated poultry houses in hot countries often develop hyperthermia during the summer (Etches et al., 1995). As a result, layer performance deteriorates during heat stress. The deterioration is manifested as reduced feed intake and altered nutrient absorption, respiratory alkalosis, decreased blood flow to several organs, and endocrine changes affecting steroid hormones secretion, etc. (Hillman et al. 1985; Etches et al., 1995). Examples of low productive and reproductive performance of laying hens in summer include low egg weight, low egg production, and high eggshell breakage (De-Andrade et al., 1976, 1977; Wolfenson et al., 1979; Tanor et al., 1984; Marsden and Morris, 1987; Sauveur and Picard, 1987; Roland, 1988). Laying hens are susceptible to heat stress for several reasons. First, their metabolic heat production is high as
2001 Poultry Science Association, Inc. Received for publication September 25, 2000. Accepted for publication March 13, 2001. 1 To whom correspondence should be addressed: [email protected]. 2 Present address: Poultry Department, Extension Services, The Ministry of Agriculture, Israel.
958
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
ABSTRACT Laying hens held in battery cages in naturally ventilated poultry houses in hot countries usually develop hyperthermia, which adversely affects their performance. The present means of cooling alleviate to some degree, but cannot eliminate, the stress imposed by heat. A new approach to cooling of laying hens was developed, based on wetting the skin and promoting evaporation of water from the ventral regions of the bird. The type of plumage in the ventral regions and the exposed skin of the apteria enable more efficient wetting than is possible with dorsal cooling. A ventral cooling regime, comprising an initial period of frequent wettings followed by intermittent wetting for 10 s every 30 min was able to maintain normothermia of laying hens subjected to a 10-h period of heat exposure. Dorsal cooling was less efficient; body
VENTRAL COOLING OF HENS DURING HEAT STRESS
MATERIALS AND METHODS Experimental Birds The experiments were conducted on White Leghorn (Loman) laying hens that had been obtained from a commercial hatchery3 and that were held in individual battery cages in two identical, adjacent, temperature-controlled chambers. Hens brought from the experimental poultry house to the chambers were allowed a 1-wk adaptation period before experiments started. The hens were kept at 24 C and 40% RH, were exposed to 17 h of light, and were fed ad libitum with a commercial layer feed containing 2,800 kcal/kg, 18% protein, and 3.7% calcium. In the first set of short-term experiments, the optimal conditions for ventral cooling were determined through a series of 10-h heat exposures. Hens (n = 24) were in their second production period and had an average laying rate of 75%. In a second set of long-term experiments, the productive and endocrine responses of hens exposed to ventral cooling, under the optimal protocol established in the short-term experiments, were examined in two 1-
3
Kibbutz Hafetz-Haim, 76817 Israel. Naan, model 7955, Kibbutz Naan, 76829 Israel. 5 YSI, model 4000, Yellow Springs, OH 45487. 6 Novasina G Hygrometer, MIK 3000, CH 8808 Pfaffikon, Switzerland. 4
mo experiments. The hens (n = 24 in each experiment) were in their first production period and had an average laying rate of 92%. The experiments were conducted according to the guidelines of the local committee for animal care and welfare.
Developing the Ventral Cooling System In two, six-cage batteries, one of which was located in each chamber, static sprinklers4 were installed below the food trays; each sprinkler covered two cages. Various water outputs of the sprinklers were examined: 30, 43, 54, and 85 L/h, with a covering angle of 90°. The positioning of the sprinklers was adjusted to obtain maximal wetting of the hens (n = 12). The dorsally positioned sprinklers were installed 50 cm above the hens (n = 12) in the other two, six-cage batteries, one of which was located in each chamber, on the side remote from the ventrally cooled batteries. Care was taken to eliminate any cross wetting of hens in either treatment by the sprinklers in the other treatment. A set of short-term experiments was performed twice weekly, in each of which the hens were exposed to 38 C and 40% RH for 10 h. The temperature in the chambers was gradually elevated from 24 C at a rate of 5 C/h, starting at 0700. Intervals between heat exposures served as recovery periods. To determine the optimal cooling procedure, various water capacities and duration and frequency of wettings were examined. In addition, wetting regimes with or without more frequent wetting at the start of heat exposure were examined. In each of the 10h heat exposures, body temperatures were monitored,5 at an accuracy of 0.1 C, by means of a thermistor inserted into the cloaca. Skin temperatures were recorded with unilateral flat thermistors attached to two feathered skin areas on the breast and back and to two featherless skin areas of the comb and metatarsus. Respiration rates were monitored, and air temperature and relative humidity6 were recorded frequently.
Performance and Endocrine Responses Two long-term experiments were conducted. In the first, the responses of hens exposed to ventral cooling were compared with those of uncooled control hens. In the second, they were compared with those of hens exposed to dorsal cooling. The experimental protocol was the same in both experiments. Responses during 10 d of heat exposure (37 C and 40% RH) were compared with those during 10 d before and 10 d after, when the hens were under normothermic conditions (24 C and 40% RH). During the 10 d of heat stress, air temperature was cycled daily—high from 0700 to 1700 and normothermic thereafter—to mimic the fluctuation of air temperatures during a natural 24-h cycle. Individual feed intake was recorded daily to an accuracy of ± 1 g. Egg weight was recorded daily to an accuracy of ± 0.1 g. Eggshell weight was recorded after drying the shell at 70 C for 18 h. The shell index, which relates
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
1985; Ernst, 1995). Direct cooling is achieved by means of a sprinkling system that wets the skin of the hens with large droplets so that the evaporating water cools the birds (Ernst, 1995). The latter system is used in many commercial, naturally ventilated poultry houses in hot countries. Because the sprinklers are located above the cages, the hens are wetted over their dorsal hemisphere and, therefore, experience dorsal cooling. Although this approach alleviates heat stress to some extent and lowers hen mortality during hot spells, it has several limitations: 1) the decrease in productivity during summer is not eliminated, 2) excess water wets the manure and the feed mixture in the trays, and 3) the dense and effectively insulating feathers on the birds’ wings and backs limit penetration of water to the skin over the dorsal hemisphere. Another possible approach that has not been examined before involves cooling laying hens by wetting the ventral hemisphere of the body thus providing them with ventral cooling. The main advantage of ventral cooling, compared with dorsal cooling comes from the type of feathers in the ventral areas. The plumage there is less dense than in the dorsal regions, and several areas are apteria (Lucas and Stettenheim, 1972), which eases the access of water to the skin. The aim of the present study was to test the efficacy of ventral cooling applied to laying hens held in naturally ventilated battery cages. We compared thermal responses, performances, and circulating hormone concentrations of ventrally cooled, uncooled, and dorsally cooled hens.
959
960
WOLFENSON ET AL.
shell weight and egg surface area, was calculated from egg weight by means of the equation I = a/3.9782 × b0.7056, where I = shell index (mg/cm2), a = egg shell weight (mg), and b = egg weight (g; Abdallah et al., 1993). Blood samples were collected from the brachial vein 2 d before heat stress started, 2 and 10 d after it started, and 4 d after it ended. Plasma was separated and stored at −20 C. Plasma osmolality7 was determined with a calibrated osmometer.
Hormone Analyses
Data Analysis Data were analyzed by ANOVA, by means of the general linear models procedure of the Statistical Analysis System software (SAS, 1987). Data are presented as means and standard errors. Data were analyzed in models based on within-hen repeated measurements. The statistical models included experimental groups (ventrally cooled vs. dorsally cooled or uncooled groups), hens (nested within group), time (hours or days in short- or long-term experiments, respectively), period (in long-term experiments, the 10-d periods before, during, and after heat stress), and interactions. Data were analyzed separately for Experiments 1 and 2. Hens (within group) served as an error term for the effect of experimental group. Time or period effects and their interactions were tested against the residual error. Skin temperatures data were analyzed by one-way ANOVA.
7
Osmette, 2007, Precision System, Sudbury, MA 01776. Donated by F. Kohen, Department of Hormone Research, Weizmann Institute, Rehovot, 76100 Israel. 9 Sigma Chemical Co., St. Louis, MO 63178-9916. 8
FIGURE 1. Body temperatures of dorsally and ventrally cooled laying hens during heat exposure. n = 12 in each group, pooled SEM −0.2 C. Dorsal and ventral temperatures differ; P < 0.01.
RESULTS The Ventral Cooling System The optimal protocol of ventral cooling with the minimal amount of water necessary to prevent hyperthermia was determined in a series of 10-h experiments. The following is a brief description of the short-term experiment, to the point at which the optimal protocol was established (data not shown). We initially found that although hyperthermia was prevented, excessive water was used when 54- or 85-L/h sprinklers were operated for 30 s every 60 min. Therefore, we increased the interval between sprinklings to 90 min and used low-capacity sprinklers of 30L/h for 120 s each time. However, the 90-min interval was found to be too long and an application rate of 30 L/h seemed to be inadequate. Consequently, we introduced a new protocol, which began with three 30-s wettings via 43-L/h sprinklers at 5-min intervals, to achieve pronounced wetting of the feathers, which was followed by a 10-s wetting every 20 min for 10 h. This protocol prevented hyperthermia in ventrally cooled hens, whereas body temperatures of uncooled hens increased by 1.5 C and of dorsally cooled hens increased by 0.5 to 1.0 C. To reduce the amount of water used, the duration of the initial wetting was shortened from 30 to 10 s and the interval between sprinklings was increased from 20 to 45 min. However, this protocol did not maintain normothermia in ventrally cooled hens. Finally, the optimal protocol for ventral cooling was found to be three initial 30-s wettings of 43 L/h at 5-min intervals followed by a 10-s wetting every 30 min for 10 h. A difference of 0.4 to 0.7 C in body temperature between ventral and dorsal cooling was consistently achieved (P < 0.01; Figure 1). This ventral cooling protocol was later used in the long-term production experiments. Figure 2 presents the average of three measurement days in which body temperatures and respiration rates of ventrally and dorsally cooled hens were recorded between 5 and 7 h after the start of heating, during the 10-d heat exposure of Experiment 2. Ventrally cooled hens maintained body temperatures below 42 C
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
Concentrations of estradiol, progesterone, and corticosterone in plasma were analyzed by previously described and validated dextran-coated charcoal RIA (Meidan et al., 1990; Wolfenson et al., 1997, Heiblum et al. 2000), using specific antibodies for estradiol and progesterone8 and for corticosterone.9 Plasma samples were extracted by diethyl ether according to the procedure described by Badinga et al. (1992). Recovery rates were 80% for progesterone and 75% for estradiol and corticosterone. Assay sensitivities were 1.0 pg/tube for estradiol, 1.95 pg/tube for progesterone, and 7.8 pg/tube for corticosterone. Intra- and interassay CV for estradiol were 6.4 and 13.9%, for progesterone were 6.8 and 11.5%, and for corticosterone were 8.8 and 11.6%. Antibody for estradiol-17β cross-reacted with estrone (2.5%), estriol (0.95%), progesterone (< 0.2%), and estradiol-17α (1.5%). Antibody for progesterone cross-reacted with estradiol, testosterone, and corticosterone (< 0.3%); 17α-progesterone (4%); and 20α-progesterone (1.5%). Antibody for corticosterone cross-reacted with progesterone (7.5%), testosterone (3.8%), estradiol (< 0.1%), and 17αprogesterone (1.2%).
VENTRAL COOLING OF HENS DURING HEAT STRESS
FIGURE 2. Mean values of body temperatures (empty bars) and respiration rates (dotted bars) of dorsally and ventrally cooled hens, measured between 5 and 7 h after start of heating, on three different days during the 10-d heat exposure of experiment 2. n = 12 in each group, pooled SEM −0.1 C and 15 breaths per min. Dorsal and ventral values differ; P < 0.01.
Production Responses to Ventral Cooling Experiment 1. This experiment compared ventrally cooled and uncooled birds. During 10 d of heat stress, egg weight in uncooled hens decreased by about 8% (P < 0.01) and did not return to normal values after heat stress ended (Table 1). In contrast, ventrally cooled hens maintained stable egg weights during heat stress, with a
FIGURE 3. Skin temperatures in four regions of dorsally and ventrally cooled hens and uncooled hens, 5 h after start of heating (n = 6, 12, and 6, respectively). Pooled SEM −0.4 C. Mean values among cooling treatments, of a specific skin region with no common lowercase letter differ significantly (P < 0.05).
slightly elevated egg weight after heat stress ended (P < 0.01). Feed intake during heat stress was 15 and 33% lower in ventrally cooled and uncooled hens, respectively (P < 0.01; Table 1). After heat stress ended, intake returned to normal in both groups, but the ventrally cooled group attained a higher recovery value (P < 0.01). Shell index was unaffected during heat stress in ventrally cooled hens but was lowered (P < 0.01) in uncooled hens, in which it returned to normal after heat stress ended (Figure 4). Plasma osmolality was unaffected by heat stress in both groups (303.1 ± 4.3 mOsm/kg). Experiment 2. In contrast to the first experiment, this experiment compared ventrally and dorsally cooled hens. In Experiment 2, as occurred in Experiment 1, egg weight was not affected by heat stress in ventrally cooled hens (Table 1). In contrast, it was lowered by 11% during heat stress in dorsally cooled hens (P < 0.01); it was slightly improved after heat stress ended but did not return to the level of preheat stress (P < 0.01). As in Experiment 1, food intake was 18 and 32% lower in ventrally cooled and dorsally cooled hens, respectively (P < 0.01, within group compared with preheating values and between groups during heat stress; Table 1). After heat stress ended, both groups returned to preheat values. Shell index was lowered during heat stress in the ventrally cooled and dorsally cooled groups (P < 0.01), but the decrease was minor in the former and pronounced in the latter group (P < 0.01; Figure 4). In both groups, the shell index returned to normal value after heat stress ended. As in Experiment 1, plasma osmolality was not affect by heat stress, in either group, and was within the normal range for laying hens (287 ± 2.9 mOsom/kg). This result indicated that the hens were not dehydrated during heat exposure.
Steroid Concentrations in Blood Plasma Generally speaking, alterations in plasma concentrations of corticosterone, progesterone, and estradiol during heat stress were minor. Plasma concentration of corticosterone was not affected by heat stress in ventrally cooled hens in Experiment 1 or 2 (Table 2); it increased in uncooled hens in the early stages of heat stress (P < 0.05; Experiment 1) and decreased in dorsally cooled hens in the late stages of heat stress (P < 0.05; Experiment 2). Plasma estradiol concentrations were not altered during heat stress in ventrally cooled hens; in uncooled and dorsally cooled hens, it decreased in the early stage of heat stress (P < 0.01) and later returned to normal. Selected plasma samples were included in the analysis of progesterone concentration. Because time of oviposition was not recorded in the present study, some blood samples that were taken routinely during the morning hours could have been drawn during the preovulatory progesterone surge. Therefore, in accordance with Bahr and Johnson (1991) and others, all samples with more than 1 ng/mL were excluded from the data set. Elimination of those samples facilitated a reliable comparison of the effect of heat stress on basal concentrations of the hormone.
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
and respiration rates below 100 breaths/min, much lower than those of their dorsally cooled counterparts (P < 0.01). Skin temperatures of ventrally cooled hens (n = 12) were compared with those of dorsally cooled (n = 6) and uncooled hens (n = 6), 5 h after the start of heating (Figure 3). In ventrally cooled hens, the temperatures of feathered skin of the breast and back were 1.0 to 1.5 C higher than those of dorsally cooled hens (P < 0.05); the temperatures of the featherless skin of the comb and metatarsus did not differ between the two groups. Skin temperatures of uncooled hens were higher at all four skin points than those of dorsally cooled hens (P < 0.05) and were higher in the two feathered regions than in the ventrally cooled hens (P < 0.05).
961
962
WOLFENSON ET AL. TABLE 1. Egg weight and feed intake during 10-d periods before, during, and after heat stress1 Experiment 1
Egg weight Before During After Feed intake Before During After
Experiment 2
Vental cooling
No cooling
SEM
Ventral cooling
Dorsal cooling
SEM
63.9b 63.1b 65.1a
62.3a 57.2b 57.2b
0.5* 0.4** 0.5**
65.2 64.8 65.7
66.2a 58.7c 61.8b
0.4 0.5** 0.6**
114a 97b 119a
113a 76b 110a
3 2** 3**
106a 88b 106a
106a 73b 101a
2 3** 3
Means within columns with no common superscript differ significantly (P < 0.01). Responses of ventrally-cooled hens were compared with uncooled hens in Experiment 1 and with dorsally cooled hens in Experiment 2 (n = 12 hens per group). *P < 0.05, **P < 0.01 indicate significant differences between groups at a specific time. a–c 1
DISCUSSION This study presents, for the first time, a new approach to cooling laying hens held in naturally ventilated cages. Wetting of the ventral regions, and evaporation of water from them, was found to be highly effective in alleviating
FIGURE 4. Shell index (mg/cm2) of ventrally vs. uncooled hens (upper panel, Experiment 1) and of ventrally vs. dorsally cooled hens (lower panel, Experiment 2), before, during, and after a 10-d heat exposure. n = 12 in each group, pooled SEM in both experiments, 2 mg/ cm2. For significance detail, see text.
heat stress. The initial wetting at the start of heat exposure effectively saturated the feathered layer with water. It created a relatively large reservoir of water in the feathers of the breast, thigh, rump, abdomen, and legs, enabling a high rate of water evaporation from the beginning of heat exposure. The type of plumage on the ventral regions of the skin, unlike the large feathers of the wings and tail, facilitates water accumulation. In addition, wetting interrupted the arrangement of the barbs and barbules of the feather vanes and of the plumulae. This feather treatment decreased the thermal insulation of the body covering and narrowed the boundary layer of warm air surrounding the body, resulting, probably, in facilitated convective, as well as evaporative, heat loss. After the initial wetting, a minimal period of 10 s sprinkling every 30 min was sufficient to maintain the wetness of the feather coat and to ensure a high cooling rate, which is in agreement with the study by McArthur and Ousey (1994), who showed a stable rate of evaporative heat loss over a period of 100 min after wetting of a physical model maintained at 38 C. Such a short period of sprinkling is crucial to minimize wetting of the manure. The results clearly demonstrated that laying hens exposed to severe heat stress, and which were ventrally cooled, were able to maintain their body temperature within the normothermic range most of the time, 0.4 to 0.7 C lower than that of dorsally cooled hens. Cooling of the skin by water evaporation may cause some degree of vasoconstriction, which may subsequently lower the skin temperature and, consequently, attenuate convective and radiative heat losses. Interestingly, all four skin temperatures of dorsally cooled hens were 2 C lower than those of uncooled hens. Those temperatures of ventrally cooled hens were lower, by only 1 C in only two out of four skin points, than those of uncooled hens. Local cooling of thermoregulatory centers in the brain and spinal cord may occur during dorsal cooling and could be associated with the lower skin temperatures recorded in this group. This hypothesis is consistent with the finding that local cooling of the hypothalamus and spinal cord depressed the panting responses in domestic fowls under heat stress (Darre and Harrison. 1981; Hill-
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
In Experiments 1 and 2, basal values of progesterone were not altered during heat stress in ventrally cooled hens but were lower (P < 0.01) in uncooled and dorsally cooled hens in the late stage (Day 10) of heat stress.
963
VENTRAL COOLING OF HENS DURING HEAT STRESS TABLE 2. Plasma concentrations (ng/mL) of steroid hormones at 2 d before, at 2 d and 10 d during, and at 4 d after heat stress1 Experiment 1
Corticosterone 2 d before 2 d during 10 d during 4 d after Progesterone2 2 d before 2 d during 10 d during 4 d after
Ventral cooling
No cooling
SEM
Ventral cooling
Dorsal cooling
SEM
1.61 1.34 1.38 1.53
1.7 2.19 1.05 1.19
0.28 0.26* 0.26 0.27
1.03 1.59 1.51 1.47
1.75 2.15 1.10 1.52
0.24 0.27 0.24* 0.25
0.61 0.44 0.50 0.44
0.54 0.33 0.33 0.33
0.07 0.06 0.06* 0.06
0.32 0.26 0.44 0.18
0.38 0.31 0.17 0.19
0.04 0.05 0.05** 0.05
0.18 0.17 0.20 0.17
0.19 0.09 0.16 0.19
0.01 0.02** 0.02 0.02
0.17 0.13 0.18 0.23
0.18 0.08 0.13 0.22
0.01 0.02** 0.02 0.02
1 Responses of ventrally cooled hens were compared with uncooled hens in Experiment 1 and with dorsally cooled hens in Experiment 2. 2 Progesterone concentrations of basal secretion only are presented (see text for detail). *P < 0.05, **P < 0.01 indicate significant differences between groups at a specific time.
man et al., 1985). Low skin temperatures may affect the hens’ ability to maintain thermal balance. In the present study, the skin temperatures of dorsally cooled hens, i.e., about 38 C, equaled the air temperatures in the chambers, indicating that nonevaporative heat loss was nil in this group (Dawson and Whittow, 2000). In contrast, ventrally cooled hens maintained a 1 to 1.5 C difference between skin and air temperatures, enabling some degree of convective and radiative heat dissipation. The stability of skin temperatures in ventrally cooled hens during heat stress could be associated with two unique characteristics of the vascular system of domestic fowls. First, the majority of the rise in metatarsal and toe blood flow in hyperthermic hens is due primarily to a rise in arteriovenous anastomoses flow and not to an increase in capillary blood flow (Wolfenson, 1983). This rise, and the fact that arteriovenous anastomoses blood flow is regulated by central stimuli of thermoregulatory neurones and not by peripheral stimuli (Hales et al., 1978), may explain the high blood supply even during local evaporative cooling of the skin. Second, capillary blood flow in the ventral skin regions was previously found to increase sevenfold, compared with a twofold increase in the dorsal skin regions in heat-stressed hens (Wolfenson et al., 1981). The relatively higher blood flow to the ventral skin regions contributes to the stability of skin temperatures during ventral cooling. The production responses of the ventrally cooled hens were unaffected by heat stress in Experiments 1 and 2. In contrast, the uncooled and dorsally cooled hens showed decreased egg weights and reduced eggshell quality; the decrease being more pronounced in the former, as expected. Stability in the responses of ventrally cooled hens was evident, even though their feed intake was lowered during the 10-d heat exposure. Decreased egg weight could be associated, at least in part, with decreased feed
intake under heat stress (De-Andrade et al., 1977). However, alterations in the shell index cannot be directly related to calcium deficiency induced by decreased food intake, because, according to our calculations, calcium was supplied in excess in the layer diet, and hens may cope with a 30% decline in food intake. In agreement, De-Andrade et al. (1977) pointed out that low shell quality under heat stress is not related to nutritional deficits. Others have indicated that low shell quality is associated with the decrease in the plasma concentration of bicarbonate (Odom et al., 1985) or with depressed blood supply to the shell gland under heat stress (Wolfenson et al., 1981). The endocrine responses in this study were less consistent than the thermal and the production responses. A transient rise in plasma concentration of corticosterone was observed in early stages of heat exposure in uncooled hens, and concentration returned to a normal range later, even though hens were still under stress conditions. This pattern of corticosterone secretion was described before (Etches et al., 1995; Siegel, 1995). Decreases in basal concentrations of plasma progesterone were noted at the end of the 10-d heat exposure in dorsally cooled and uncooled hens but not in ventrally cooled hens. It has been suggested (Novero et al., 1991) that a decrease in progesterone secretion was related to a heat-stress-induced alteration in the steroidogenic capacity of granulosa cells. The fact that ventral cooling successfully alleviated heat stress corresponded well with the stability of plasma progesterone concentrations. The decrease in plasma concentration of estradiol was transient, which was noted in the earlier stages of heat stress (Day 2), whereas the decrease in plasma progesterone concentration was noted later during heat exposure (Day 10). The above indicates that a lack of progesterone secretion was not the reason for the decline in estradiol secretion.
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
Estradiol 2 d before 2 d during 10 d during 4 d after
Experiment 2
964
WOLFENSON ET AL.
In conclusion, ventral cooling was found to be an efficient means of alleviating the heat stress of laying hens housed in battery cages. The results clearly indicate the advantages of ventral cooling over dorsal cooling in terms of the hen abilities to maintain normothermia, a high level of productivity and normal steroids secretion. The type of plumage and the unfeathered areas in the ventral regions and the relatively high blood supply to the ventral skin all contribute to ventral cooling being superior to dorsal cooling. Another advantage of ventral cooling over dorsal cooling is that it hardly wets the feed mixture. Successful implementation of ventral cooling in commercial, naturally ventilated poultry houses will depend on optimal installation of the sprinklers and on the ability to minimize wetting of the manure.
REFERENCES
Downloaded from http://ps.oxfordjournals.org/ at Serials Acq (Law) on October 6, 2014
Abdallah, A. G., R. H. Harms, and O. El-Husseiny, 1993. Various methods of measuring shell quality in relation to percentage of cracked eggs. Poultry Sci. 72:2038–2043. Badinga, L., M. A. Driancourt, J. D. Savio, D. Wolfenson, M. Drost, R. L. de la Sota, and W. W. Thatcher, 1992. Endocrine and ovarian responses associated with the first-wave dominant follicle in cattle. Biol. Reprod. 47:871–883. Bahr, J. M., and P. A. Johnson, 1991. Reproduction in poultry. Pages 517–575 in: Reproduction in Domestic Animals. P. T. Cupps, ed. Academic Press, New York, NY. Beker, A., and R. G. Teeter, 1994. Drinking water temperature and potassium chloride supplementation effects on broiler body temperature and performance during heat stress. J. Appl. Poult. Res. 3:87–92. Blem, C. R., 2000. Energy balance. Pages 327–341 in: Sturkie’s Avian Physiology. 5th ed. G. C. Whittow, ed. Academic Press, New York, NY. Carr, L., and T. Carter, 1985. Housing and management of poultry in hot and cold climates. Pages 73–110 in: Stress Physiology in Livestock. Poultry. Vol. 3. M. K. Yousef, ed. CRC Press, Boca Raton, FL. Darre, M. J., and P. C. Harrison, 1981. The effects of heating and cooling localized areas of the spinal cord and brain stem on thermoregulatory responses of the domestic fowl. Poultry Sci. 60(Suppl. 1):1644. (Abstr.). Dawson, W. R., and G. C. Whittow, 2000. Regulation of body temperature. Pages 343–390 in: Sturkie’s Avian Physiology. 5th ed. G. C. Whittow, ed. Academic Press, New York, NY. De-Andrade, A. N., J. C. Rogler, and W. R. Featherston, 1976. Influence of constant elevated temperature and diet on egg production and shell quality. Poultry Sci. 55:685–693. De-Andrade, A. N., J. C. Rogler, W. R. Featherston, and C. W. Alliston, 1977. Interrelationships between diet and elevated temperatures (cyclic and constant) on egg production and shell quality. Poultry Sci. 56:1178–1188. Ernst, R. A., 1995 Housing for improved performance in hot climates. Pages 67–99 in: Poultry Production in Hot Climates. N. J. Daghir, ed. CAB International, Cambridge, UK. Etches, R. J., T. M. John, and G.A.M. Verrinder-Gibbins, 1995. Behavioural, physiological, neuroendocrine and molecular responses to heat stress. Pages 31–66 in: Poultry Production in Hot Climates. N. J. Daghir, ed. CAB International, Cambridge, UK. Hales, J.R.S., M. Iriki, K. Tsuchiya, and E. Kozawa, 1978. Thermally-induced cutaneous sympathetic activity related to blood flow through capillaries and arteriovenous anastomoses. Eur. J. Physiol. 375:17–24. Heiblum, R., E. Arnon, G. Gvaryahu, B. Robinzon, and N. Snapir, 2000. Short-term stress increases testosterone secre-
tion from testes in male domestic fowl. Gen. Comp. Endocrinol. 120:55–66. Hillman, P. E., N. R. Scott, and A. van Tienhoven, 1985. Physiological responses and adaptations to hot and cold environments. Pages 1–71 in: Stress Physiology in Livestock. Poultry. Vol 3. M. K. Yousef, ed. CRC Press, Boca-Raton, FL. Lucas, A. M., and P. R. Stettenheim, 1972. Pages 467–552 in: Avian Anatomy. Integument. Part 2. US Government Print Office, Washington, DC. Marsden, A., and T. R. Morris, 1987. Quantitative review of the effects of environmental temperature on food intake, egg output and energy balance in laying pullets. Br. Poult. Sci. 28:693–704. McArthur, A. J., and J. C. Ousey, 1994. Heat loss from a wet animal: changes with time in the heat balance of a physical model representing a newborn homeotherm. J. Therm. Biol. 19:81–89. Meidan, R., E. Girsh, O. Blum, and E. Aberdam, 1990. In vitro differentiation of bovine theca and granulosa cells into small and large luteal-like cells: Morphological and functional characteristics. Biol. Reprod. 43:913–921. Muiruri, H. K., and P. C. Harrison, 1991. Effect of peripheral foot cooling on metabolic rate and thermoregulation of fed and fasted chicken hens in a hot environment. Poultry Sci. 70:74–79. Novero, R. P., M. M. Beck, and E. W. Gleaves, 1991. Plasma progesterone, luteinizing hormone concentrations, and granulosa cell responsiveness in heat-stressed hens. Poultry Sci. 70:23. Odom, T. W., P. C. Harrison, and M. J. Darre, 1985. The effects of drinking carbonated water on the egg shell quality of Single Comb White Leghorn hens exposed to high environmental temperature. Poultry Sci. 64:594–596. Reilly, W. M., K. W. Koelkebeck, and P. C. Harrison, 1991. Performance evaluation of heat-stressed commercial broilers provided water-cooled floor perches. Poultry Sci. 70:1699– 1703. Roland, D. A., 1988. Research note: egg shell problems: estimates of incidence and economic impact. Poultry Sci. 67:1801–1803. SAS Institute, 1987. SAS/STAT User’s Guide. 6th edition. SAS Institute Inc, Cary, NC. Sauveur, B., and M. Picard, 1987. Environmental effects on egg quality. Pages 219–234 in: Egg quality—Current Problems and Recent Advances. R. G. Wells and C. G. Belyavin, ed. Butterworths, London, UK. Siegel, H. S., 1995. Stress, strain and resistance. Br. Poult. Sci. 36:3–22. Smith, W. K., 1981. Poultry housing problems in the tropics and subtropics. Pages 235–258 in: Environmental Aspects of Housing for Animal Production. J. A. Clark, ed. Butterworths, London, UK. Tanor, M. A., S. Leeson and J. D. Summers, 1984. Effect of heat stress and diet composition on performance of white leghorn hens. Poultry Sci. 63:304–310. van Kampen, M., 1988. Effects of drinking water temperature and leg cooling on heat stress of laying hens (Gallus domesticus). J. Therm. Biol. 13:43–47. Wolfenson, D., 1983. Blood flow through arteriovenous anastomoses and its thermal function in the laying hen. J. Physiol. 334:395–407. Wolfenson, D., Y. F. Frei, N. Snapir, and A. Berman, 1979. Effect of diurnal or nocturnal heat stress on egg formation. Br. Poult. Sci. 20:167–174. Wolfenson, D., Y. F. Frei, N. Snapir, and A. Berman, 1981. Heat stress effects on capillary blood flow and its redistribution in the laying hen. Eur. J. Physiol. 390:89–93. Wolfenson, D., B. J. Lew, W. W. Thatcher, Y. Graber, and R. Meidan, 1997. Seasonal and acute heat stress effects on steroid production by dominant follicles in cows Anim. Reprod. Sci. 47:9–19.