Early weaning to reduce tissue mobilization in lactating sows and milk supplementation to enhance pig weaning weight during extreme heat stress 1,2

Early weaning to reduce tissue mobilization in lactating sows and milk supplementation to enhance pig weaning weight during extreme heat stress 1,2 J. D. Spencer* 3, R. D. Boyd 4, R. Cabrera, and G. L. Allee* 5 *Department of Animal Sciences, University of Missouri, Columbia 65211 and Pig Improvement Co., Franklin, KY 42135 ABSTRACT: This study was conducted to determine the effect of reduced lactation length and supplemental milk replacer (MR) during high ambient temperatures. Thirty nine primiparous and 100 multiparous sows (PIC, Franklin, KY, C-22) were used in a 2 2 2 factorial arrangement of treatments. Treatments consisted of two lactation room temperatures (21 C [TN] and 32 C [HOT]), two lactation lengths (14 or 19 d), and two parity groups (primiparous, multiparous). Pigs were either: 1) sow-reared to 19 d or 2) sow-reared to 14 d, and then reared to 19 d with MR after sow removal. All sows were fed the same diet (1.07% lysine, 3,366 kcal of ME/kg). Sows were weighed and ultrasound for backfat thickness (BF) and longissimus muscle area (LMA) within 6 h after farrowing and at the time of sow removal (d 14 or 19). Pigs were individually weighed at weaning (d 19) and after a 47-d nursery period (d 66). Heat stress increased sow weight loss ( 13.35 kg, P < 0.01) and decreased sow feed intake (4.63 kg/d, P < 0.01) during lactation compared with sows in TN (+4.5 kg and 7.5 kg/d, respectively). Early weaning (d 14) during heat stress decreased maternal weight loss ( 10.1 vs. 16.6 kg, P < 0.01). Primiparous sows lost more BF in both environments ( 2.60 vs. 1.56 mm, P < 0.05), and both parity groups lost more BF ( 3.35 vs. 2.3 mm, P < 0.10) and LMA ( 1.82 vs. 0.77 cm 2, P < 0.05) when lactating for 19 d in the HOT environment than those lactating for 14 d. Pigs nursing primiparous and multiparous sows in the HOT environment and provided MR had heavier individual 19-d weights (7.37 and 8.12 kg/ pig, respectively) than those nursing to 19 d (5.57 and 6.04 kg/pig, P < 0.01). Milk replacer decreased the difference normally observed in 19-d weights between primiparous and multiparous sow-reared pigs in TN. Pigs fed MR in both environments and nursing multiparous sows had improved weight gains in the nursery compared with pigs nursing sows to 19 d (428 vs. 406 g/d, respectively; P < 0.01), or reared by primiparous sows (444 vs. 390 g/d, respectively; P < 0.01). Sow weaning on d 14 in the HOT environment decreased the weanto-estrus interval in primiparous sows (22.8 vs. 9.2 d, P < 0.10). This study shows the benefit of early weaning in combination with milk replacer to preserve the sow and to restore pig weaning weights and nursery end weights under heat stress. Key Words: Heat Stress, Lactation, Milk Composition, Piglet Feeding, Supplementary Feeding 2003 American Society of Animal Science. All rights reserved. J. Anim. Sci. 2003. 81:2041 2052 Introduction High environmental temperatures during lactation decrease milk yield, litter growth, and reproductive 1 The authors thank M. Johnston, R. Graves, and S. Coffey, Franklin, KY, for technical assistance and J. Laurenz, Texas A&M at Kingsville, for milk immunoglobin analysis. 2 Partial support for this study was provided by Advanced Birth- Right Nutrition and Ralco Nutrition, Marshall, MN. 3 Current address: United Feeds, P.O. Box 108, Sheridan, IN 46069. 4 Current address: The Hanor Company, P.O. Box 881, Franklin, KY 42134. 5 Correspondence: Lab 111 Animal Science Research. Received July 15, 2002. Accepted April 29, 2003 performance. To decrease heat production, daily sow feed intake can decrease by 0.17 kg/ C when temperatures rise above 16 C (Black et al., 1993). This decrease in intake results in increased tissue loss, reduced weaning weight, and reduced subsequent litter size, especially in primiparous sows. Reducing dietary heat increment by reducing protein levels was reported to have marginal benefit in reducing tissue loss and showed no benefit for litter growth rate in a hot environment (Johnston et al., 1999; Renaudeau and Noblet, 2001). Mullan et al. (1992) reported no beneficial effect of a nutrient-dense diet on feed intake or milk yield in first-parity sows lactating in heat stress (30 C). Conserving whole-body protein mass during lactation minimizes the wean-to-estrus interval and increases second litter size (Boyd et al., 2000). Since 2041

2042 Spencer et al. diet modifications to increase energy intake (reduced dietary CP content or elevated fat supplementation) have limited benefits in terms of reducing tissue loss or improving litter gain during heat stress, producers are dependent on changes in management to alleviate production inefficiencies. Split-weaning is beneficial during high temperatures, but a slightly shorter lactation period may prove to be more beneficial in reducing body tissue loss and improving subsequent reproductive performance. Early removal of the sow and feeding milk replacer (MR) to piglets would provide the opportunity for piglets to reclaim lost growth potential. The objective of this study was to determine whether sow tissue loss could be minimized, and subsequent reproductive performance improved by reducing sow lactation length under extreme heat stress. Further, the effect of heat stress on milk composition and the value of MR to recover progeny growth before weaning were evaluated in a hot and thermoneutral environment. Materials and Methods Thirty-nine primiparous and 100 multiparous lactating sows (PIC, Franklin, KY, Camborough-22) were used in a 2 2 2 factorial arrangement of treatments with two farrowing room temperatures (21 C[TN] and 32 C [HOT]), two sow lactation lengths (14 or 19 d), and two parity groups (primiparous and multiparous). All procedures followed established standards for the humane care and use of animals. Pigs were either: 1) sow-reared to 19 d or 2) sow-reared to 14 d, and then reared to 19 d using sow MR (SMR, Advanced Birthright Nutrition) via a circulating pressurized system (Supp-Le-Mate C.S., Soppe Systems, Inc., Manchester, IA). Twelve sows were brought into one of four farrowing rooms each week from mid-october through March (21 total farrowing groups). Sows were randomly assigned to treatment within temperature based on sow parity. Sows were brought into the farrowing house on d 110 of pregnancy, and temperature was maintained at 21 C or gradually increased to 28 C by d 113 of pregnancy. Room temperature was increased to 32 C 1 d after farrowing. Temperature was continuously measured in each room (H8 Pro Series, Onset Computer Corp., Pocasset, MA). The two farrowing rooms that were set at a constant 32 C were quite stable (31.4 ± 1.7 C). The same was true for rooms set to be maintained at 21 C (21.7 ± 1.4 C). Lights were on in all farrowing rooms from 0600 to 1500. Litters were standardized to 10 to 11 pigs per litter and weighed within 6hoffarrowing. Sows were fed 1.8 kg/d for 1 d before farrowing, 1.4 kg the day of farrowing, and 2.3 or 2.7 kg/d for primiparous and multiparous sows, respectively, 1 d after farrowing. Two days postfarrowing, all sows were fed ad libitum three times per day for the remainder of lactation. Animals were fed the same meal diet (1.07% lysine, Table 1. Ingredient and chemical composition of the lactation diet a Ingredient Concentration, % Corn 61.83 Soybean meal, 48% CP 26.50 Wheat midds 5.00 Fat b 2.50 Dicalcium phosphate 1.97 Limestone 0.94 Salt 0.61 Vitamin/trace mineral premix c 0.60 L-Lysine HCl 0.05 Calculated composition, % ME, kcal/kg d 3,366 Lysine 1.07 Threonine 0.72 Tryptophan 0.24 Methionine + Cystine 0.63 Valine 0.94 Calcium 0.95 Phosphorus 0.77 Chemical analyses, % Lysine 1.03 Threonine 0.69 Tryptophan 0.22 Methionine + Cystine 0.64 Valine 0.89 Calcium 1.04 Phosphorus 0.85 a As-fed basis. b Animal/vegetable blend. c Premix supplied per kilogram of diet: Cu, 12 mg; I, 0.32 mg; Fe, 80 mg; Mn, 44 mg; Se, 0.24 mg; Zn, 96 mg; vitamin A, 10,310 IU; vitamin D, 2,115 IU; vitamin E, 63.4 IU; biotin, 518.2 g; menadione, 5.3 mg; riboflavin, 15.9 mg; pantothenic acid, 47.6 mg; folic acid, 1.4 mg; niacin, 81 mg; pyridoxine, 8.8 mg; thiamine, 5.3 mg; vitamin B 12, 81.1 g. d Calculated from NRC (1998). 3,366 kcal of ME/kg) (Table 1) for the entire lactation period. Samples of feed were collected throughout the study, pooled together, and analyzed for protein, fat, fiber, calcium, phosphorus, and amino acid content (AOAC, 1995). If the sow was to be weaned at d 14 of lactation, MR (24% CP, 18% crude fat) was made available to piglets in the farrowing crate on d 10 of lactation to acclimate pigs to MR before removing the sow. Milk replacer was mixed fresh daily at a rate of 132 g of MR per liter of water. The circulating milkdelivery system was cleaned and disinfected with bleach and acid every 7 d. Creep feed was not supplied to any pigs during lactation. Sow weight, real-time ultrasound (Aloka model SSD-500 V, Wallingford, CT) backfat thickness 10 cm from the midline (BF) and longissimus muscle area (LMA) were measured within 6 h postfarrowing and at sow weaning (d 14 or 19). Measurements were taken at the 10th and last rib and averaged. Total feed intake during lactation was determined for sows. Litter weights were obtained on d 14 if the sow was weaned on d 14. For all treatments, piglets were individually weighed on d 19 before being transported to an offsite nursery.

Early weaning during heat stress 2043 Pigs were weaned on d 19 and transported to an offsite nursery, sorted by BW, and placed in pens (1.83 2.44 m), with 15 pigs/pen. Pigs were placed on a fourphase nursery feed program based on BW phases for 47 d. Individual pig weights were measured at 66 d of age for determination of daily gain. Sows were removed from their litters (d 14 or 19), weighed, ultrasound measurements were taken, and placed in gestation crates. Environmental temperature was maintained between 19 and 22 C. Sows were boar exposed daily and artificially inseminated with the same commercial boar line when detected to be in estrus. Sows were artificially inseminated daily for the duration of estrus. If not detected to be in estrus by d 12 after weaning, sows were given an injection of PG 600 (Intervet; Millsboro, DE) to induce ovulation. Days to estrus and subsequent litter size were recorded on all animals maintained in the breeding herd. Sows were removed from the herd if they had farrowed over seven litters or if structural problems resulted. Milk Collection Litter weights and milk samples were obtained from 10 and 8 multiparous sows in the TN and HOT environments, respectively, on d 1, 5, 10, 15, and 19 of lactation to evaluate the effect of environmental temperature on colostrum and milk composition. Litter weight gain was used to estimate sow milk yield assuming a 4:1 ratio of milk yield to litter gain (Patience et al., 1995). Litter weight gain for estimating sow milk yield was only determined on litters without access to MR. Sows were given 20 IU of oxytocin i.m. to induce milk release. Two middle teats were completely stripped of colostrum or milk. Until analysis, whole milk samples were frozen at 10 C in vials containing a milk preservative, broad-spectrum micro tab (bronopol; 2-bromo- 2-nitropropane-1,3-diol, D&F Control Systems, Inc., Dublin, CA) until analysis. Milk Analysis Milk samples were homogenized and subjected to infrared analysis of fat, protein, lactose, and solids (Milko Scan 300, Foss Food Tech., Eden Prairie, MN). Fat was quantified using a 3.5- m wavelength filter, whereas protein and lactose were measured at wavelengths of 6.5 and 9.5 m, respectively. Homogenized milk samples were ultracentrifuged (105,000 g) for 60 min at 5 C for removal of casein and free fat. After centrifugation, milk serum was removed by pipette and frozen at 80 C until analysis for cortisol and immunoglobin (Ig) content. Concentrations of cortisol in milk serum were determined by RIA (Diagnostic Products Corp., Los Angeles, CA). A standard curve was prepared using increasing amounts of milk serum and compared with the standard curve generated using the provided cortisol standards to validate the assay for quantification of cortisol content in milk samples. The standard curve generated with increasing amounts of isolated milk serum had a statistically similar slope compared with the standard curve generated with the aqueous standards. Total concentrations of immunoglobins (IgA, IgG, and IgM) in isolated milk serum were determined using a double-antibody sandwich ELISA specific for porcine immunoglobins. Preliminary assays were conducted to determine appropriate sample dilutions (IgA d 1 = 1/100,000, d 5 to 19 = 1/25,000; IgG d 1 = 1/150,000, d5to19= 1/100,000; IgM d 1 = 1/50,000, d 5 to 19 = 1/16,000). Samples were diluted in PBS containing 0.05% Tween-20 and 1% BSA (diluent buffer). Immunoplates (Nalge Nunc International, Denver, CO) were incubated overnight with 100 ml/well of diluted, affinity-purified goat antipig IgG, IgM, or IgA per manufacturer s directions (1:100 in 0.1 M NaCO3, ph 8.2; Bethyl Laboratories, Montgomery, TX). Plates were washed three times with PBS containing 0.05% Tween-20 (wash buffer). To block nonspecific binding, 200 ml/well of blocking buffer (PBS containing 1% BSA) was added, and the plates were incubated for 30 min. Plates were then washed three times with wash buffer. Appropriately diluted samples were added (100 ml/well) to wells and plates incubated for 2hatroom temperature. Following three washes with wash buffer, 100 ml/well of the appropriate detecting antibody (goat anti-pig IgG, IgM, or IgA-Fc conjugated with horseradish peroxidase; diluted 1:2,000 in PBS containing 1.0% BSA and 0.05% Tween-20; Bethyl Laboratories) was added, and plates were incubated for1hatroom temperature. Plates were then washed four times and 100 ml/well of enzyme-substrate buffer (2,2 -azino-bis[3-thylbenzthiazoline-6-sulfonic acid] + 0.05% H 2 O 2 ; ph 4.5; Sigma Chemical Co.; St. Louis, MO) was added. Plates were incubated at room temperature for 1 h and optical densities were determined at 562 nm on an EL311 Bio-Tek Microplate Autoreader (Bio-Tek Instruments Inc., Winooski, VT). Sample IgG, IgM, and IgA concentrations were determined by comparison to standard curves generated with purified swine IgA, IgM, and IgG (Bethyl Laboratories). Respiration Rates and Rectal Temperatures Respiration rates (breaths/minute) and rectal temperature were measured on 20 and 19 multiparous sows in the TN and HOT environments, respectively. Measurements were collected on d 1, 5, 10, 15, and 19 of exposure to the desired environmental temperature. Respiration rates were determined by measuring the number of flank movements per minute. Rectal temperatures were recorded by inserting a digital thermometer approximately 11.5 cm into the rectum of each sow until a constant temperature was observed. Statistical Analysis Sow and piglet performance data were analyzed as a split-plot with the main plot comprising temperature.

2044 Spencer et al. Figure 1. Respiration rate ( ) and rectal temperature ( ) of sows lactating in a constant 21 C, ( ) or a constant 32 C ( ) environment. Points represent least squares means of 20 and 19 sows in the 21 C and 32 C environments, respectively. Respiration rate: main effect of temperature and day, P < 0.01 and P < 0.01, respectively; temperature day, P = 0.19. Rectal temperature: main effect of temperature and day, P < 0.01 and P < 0.01, respectively; temperature day, P < 0.01. Each subplot contained combinations of two parities (primiparous and multiparous) and two sow lactation lengths (14 or 19 d). The linear statistical model contained the effect of temperature, farrowing group within temperature, parity, lactation length, and all interactions among temperature, parity, and lactation length. Temperature was tested using farrowing group within temperature as the error term. The main effects of parity, lactation length, and all possible interactions used the residual mean square as the error term. When interactions of the main effects were found, treatment means were separated using the least significant difference test. Ultrasound measurements of BF and LMA at weaning were adjusted for the covariant of farrowing BF and LMA. Additionally, 19-d litter weights were also adjusted by covariant analysis for differences in litter weight after at farrowing. All performance analysis was performed using the GLM procedure of SAS (SAS Inst., Inc., Cary, NC). Milk yield, milk composition, rectal temperature, and respiration rates were analyzed as a split-plot in time as outlined by Gill and Hafs (1971). Variance and covariance matrix over time was tested for structure, which indicated compound symmetry. The linear statistical model included temperature, sow within temperature, day of lactation, and temperature lactation day interaction. Effect of temperature was tested using sow within temperature as the error term. The main effect of lactation length and possible interaction used the residual mean square as the error term. When interactions of the main effects were found, treatment means were separated using the least significant difference test. Split-plot in time analysis was analyzed using the PROC MIXED procedure of SAS. Results Respiration rate and rectal temperature was elevated (P < 0.01) in sows subjected to heat stress (Figure 1); however, they exhibited some adaptation to the elevated temperatures during lactation (temperature day, P < 0.01). The HOT temperature reduced sow feed intake by approximately 38%. Sows lactating to 19 d consumed more feed/d than those weaned at 14 d(p < 0.01) (Table 2). Sows lactating for 19 d (P < 0.01) and those lactating in the HOT environment (P < 0.01) lost more BW during lactation. Sows in the TN environment gained weight during lactation. Sows lost more BF and LMA when lactating in the HOT environment (P < 0.05, Table 3). Additionally, first-litter females lost more BF than multiparous sows in both environments and lactation lengths (P < 0.05, Table 3). Primiparous and multiparous sows lost more BF (P < 0.10) and more than twice as much LMA (P < 0.02) when lactating in a HOT environment for 19 d compared with those lactating for 14 d (Table 3). Pigs nursing sows for 19 d in the HOT environment had lower weaning weights than all other treatments Table 2. As-fed feed intake and body weight change of sows in a thermoneutral (21 C) or hot (32 C) environment and lactating for 14 or 19 d a 21 C 32 C P < b Item 14 19 14 19 SE T L T L Number of sows 36 34 35 34 Feed intake, kg/d 7.03 7.97 4.43 4.83 0.07 0.01 0.01 0.08 Postfarrowing weight, kg 210 202 212 209 2.95 0.55 0.42 0.70 Wean weight, kg 217 204 201 192 2.93 0.13 0.08 0.76 Weight change, kg +7.21 +1.80 10.1 16.6 0.98 0.01 0.01 0.81 a Fourteen-day lactation involved weaning the sow on d 14. Effect of parity on intake or change in body weight was not significant (P = 0.30 and 0.76, respectively). Values represent least squares means. b T = main effect of temperature, L = main effect of lactation length, T L = temperature and lactation length interaction.

Early weaning during heat stress 2045 Table 3. Effect of temperature, lactation length (14 or 19 d), and parity (1 or 2+) on sow tissue loss during lactation a 21 C 32 C 14 19 14 19 P < b Item 1 2+ 1 2+ 1 2+ 1 2+ SE T L P T L Number of sows 11 25 12 22 8 27 8 26 Fat depth at weaning, mm c 16.5 17.7 16.6 18.5 15.8 17.0 15.3 5.3 0.2 0.01 0.5 0.02 0.10 Change in fat depth, mm ce 2.2 1.0 2.1.2 2.9 1.7 3.3 3.4 0.2 0.01 0.50 0.02 0.10 Change in fat depth, % ce 10.6 6.5 11.7 1.1 15.0 10.8 16.9 17.8 1.0 0.01 0.81 0.05 0.11 Longissimus muscle area at weaning, cm 2d 36.0 36.4 36.2 36.6 36.2 36.0 34.9 35.2 0.3 0.02 0.09 0.43 0.02 Change in longissimus muscle area, cm 2de 0.80 0.50 0.65 0.26 0.69 0.85 1.95 1.68 0.3 0.02 0.08 0.43 0.02 Change in longissimus muscle area, % de 2.0 1.4 1.8 0.6 1.8 2.4 5.3 4.5 0.3 0.02 0.08 0.43 0.02 a Values are least squares means. b T = main effect of temperature, L = main effect of lactation length, P = main effect of parity. Interactions between L P, T P, and T L P were not significant (P > 0.10). c Fat thickness measured ultrasonically at the 10th and last rib. Fat thickness at parturition used as a covariate. d Longissimus muscle area measured ultrasonically at the 10th and last rib. Longissimus muscle area at parturition used as a covariate. e Change fat depth and longissimus muscle area is the difference in tissue measurement from farrowing to weaning. (temperature length; P < 0.01) (Table 4). Pigs nursing first-litter females weighed less at d 19 than pigs nursing multiparous sows for the same time (P < 0.01). Early weaning at d 14 and feeding MR increased weaning weights of piglets nursing parity-one females in both environments (P < 0.01). This is in contrast with pigs nursing multiparous sows, in which weaning weights of pigs were only increased with early weaning and feeding MR in the HOT environment (temperature length parity; P < 0.12;) (Table 4). Pigs reared in the HOT environment were lighter than those in 21 C (P < 0.10) (Table 5) at d 14. However, after sow removal and supplementation with MR, there was no difference in d 19 weight between environments (P < 0.80) (Table 5). There was no difference in preweaning mortality among treatments. Estimated milk yield was reduced in the HOT environment by approximately 30% from d 5 to the end of lactation (P < 0.01) (Figure 2). Milk solids, protein, fat, and lactose content were not affected by lactation temperature (Table 6). However, protein content declined, whereas fat and lactose content increased after d 1 of lactation. Milk serum immunoglobin (IgG, IgA, IgM), and cortisol content decreased after d1oflactation (P < 0.01) (Table 6). Sows in the HOT environment had higher levels of cortisol and IgA in their colostrum on d 1 of lactation than those in TN (temperature day; P < 0.03). During the nursery period, pigs from multiparous sows exhibited greater growth rates compared to those derived from primiparous sows (P < 0.01, Table 7). Piglets nursing the sow to 14 d and supplied MR had greater weight gain during the nursery period (P < 0.01) compared with those nursing the sow to 19 d and displayed heavier weights at 66 d of age (Figure 3). Primiparous sows weaned on d 14 returned to estrus sooner (9.2 vs. 22.8 d) than those lactating for 19 d in the HOT environment (Table 8, P < 0.10). The HOT environment decreased subsequent litter size, litter weight, and number born alive for both first-litter and multiparous sows (P < 0.05). Discussion The zone of thermal comfort has been estimated to be between 12 and 22 C for the lactating sow (Black et al., 1993). However, ambient temperatures in the order of 30 to 35 C are not uncommon environments for lactating sows. These elevated temperatures result in reduced sow productivity and economic return. The present results demonstrate that reducing the length of lactation from 19 to 14 d reduced the negative effects of extreme heat stress on sow tissue loss and subsequent reproductive performance. The effect was greatest for the primiparous female. When sows were early weaned and pigs were fed an unlimited supply of supplemental milk, litters reared in the 32 C environment achieved 19-d weaning weights that were similar to pigs reared in the TN environment.

2046 Spencer et al. Table 4. Effect of temperature, lactation length (14 or 19 d), sow parity (1 or 2+), and supplemental milk on piglet weaning weight on d 19 a 21 C 32 C 14 b 19 14 b 19 P < c Item 1 2+ 1 2+ 1 2+ 1 2+ SE T L P L P T L T P T L P Number of sows 11 25 12 22 8 27 8 26 Day 0 Litter size 10.40 10.59 10.33 10.48 10.76 10.68 10.94 10.54 0.04 Avg. piglet weight, kg 1.79 1.85 1.62 1.75 1.56 1.66 1.42 1.84 0.02 0.01 0.26 0.01 0.11 0.19 0.14 0.29 Day 19 Litter size 9.80 9.40 9.82 9.28 9.89 9.76 9.54 9.32 0.09 0.85 0.30 0.16 0.81 0.42 0.51 0.96 Avg. piglet weight, kg d 7.71 8.07 6.28 7.72 7.37 8.12 5.57 6.04 0.09 0.05 0.01 0.01 0.36 0.01 0.47 0.12 a Values are least squares means. b Fourteen-day lactation involved weaning the sow on d 14. Pigs remained in the crate and received supplemental milk through d 19. c T = main effect of temperature, L = main effect of lactation length, P = main effect of parity. d Adjusted for the covariate of starting litter weight. The negative effects of heat stress on sow feed intake, sow tissue loss, subsequent reproductive performance, milk yield, and litter gain are well documented (Black et al., 1993; Renaudeau et al., 2001; Renaudeau and Noblet, 2001). The negative impact of heat stress on sow productivity is a direct result of increased body temperature. Sows in the 32 C environment had a higher rectal temperature than sows in the 21 C environment but displayed some long-term adaptation to the high environmental temperature. Respiration rate declined throughout the 19-d exposure to 32 C. This is in contrast to adaptations noted by Quiniou and Noblet (1999) and Renaudeau et al. (2001) where sows adapted the first 3dinahotenvironment (29 C) and showed no further adaptations for the rest of lactation. The report of Schoenherr et al. (1989) agrees with our finding of increased evaporative heat loss activity of sows throughout lactation in a hot environment. The amount of heat generated from digestion and metabolism is influenced by the total quantity of feed ingested (Verstegen et al., 1985). Sows lactating in the constant 32 C environment reduced daily feed intake by approximately 38% when compared with sows lactating in the 21 C environment (Table 2). These results are similar to the 36% reductions in total sow feed intake reported by Johnston et al. (1999) (18 vs. 29 C), 42% by Renaudeau et al. (2001) (20 vs. 29 C), and 43% by Quiniou and Noblet (1999) (22 vs. 29 C). Summarizing results from nine experiments, Black et al. (1993) predicted that sows reduce voluntary feed intake by 0.17 kg for every 1 C increase in temperature above 16 C. The decrease in intake may depend on the extent to which the ambient temperature exceeds the individual animal s evaporative critical temperature (Black et al., 1993). Nonetheless, these reductions in feed intake result in increased tissue mobilization in an attempt to maintain an adequate nutrient supply to the mammary gland for milk production. Sows lactating in the HOT environment mobilized a greater amount of body tissue than sows lactating in the TN environment. Sows lactating in the TN environment gained weight through lactation regardless of lactation length. In the HOT environment, a 19-d lactation increased BW loss 64% compared with a 14- d lactation (Table 2). The extreme BW losses under heat stress with the 19-d lactation were mediated through losses in both backfat and protein as measured by real-time ultrasound. Decreasing lactation length from 19 to 14 d reduced the amount of BF and LMA loss only when sows were lactating in the HOT environment. Single-parity and multiparous sows lost 1.26 and 0.83 cm 2 more LMA when lactating for five more days in the 32 C environment, respectively. This increase in protein mobilization during lactation can have an impact on future reproductive performance (Boyd et al., 2000). The thermic effect of the diet is also influenced by nutrient content (Forbes and Swift, 1944; LeBellego et al., 2001). Diets with lower protein and higher fat

Early weaning during heat stress 2047 Table 5. Growth rate piglets weaned from the sow on d 14 and provided milk replacer (MR) from d 14 to 19 in a thermoneutral (21 C) or hot (32 C) environment a Item 21 C 32 C SE Temperature Number of sows 36 35 Litter size, d 14 9.7 9.8 0.14 0.90 Litter size, d 19 9.6 9.8 0.14 0.80 Litter weight on d 14, kg b 49.4 44.6 0.79 0.10 Litter weight on d 19, kg b 78.5 79.7 1.62 0.80 MR gain: from d 14 to 19 Litter gain, kg b 29.1 35.1 1.41 0.16 Gain rate, g pig 1 d 1b 603 716 27.2 0.21 a All sows were pooled within temperature (parity, P > 0.10). Milk replacer was available on d 10 of lactation to acclimate pigs to milk supplementation. Milk replacer was the only source of nutrients for piglets from d 14 to 19. b Least squares means adjusted for the covariate of d-0 litter weight. P < content have been shown to produce the lowest amount of heat during digestion and hence the lowest thermic effect. Researchers have investigated the effect of reduced dietary protein content (Johnston et al., 1999; Renaudeau et al., 2001; Renaudeau and Noblet, 2001) or higher energy density (Black et al., 1993; Renaudeau et al., 2001; Renaudeau and Noblet, 2001) on sow feed intake and lactation performance in hot environments. Improvements in sow performance by these dietary modifications have been limited at best. Johnston et al. (1999) reported no improvement in sow performance during heat stress when dietary protein concentrations were reduced 3.4% during lactation. Sows consuming lower protein diets lost more BW as protein, as indicated by higher total BW loss Figure 2. Estimated milk yield of multiparous sows lactating in a thermoneutral (TN, 21 C) or hot (32 C) environment (P < 0.01). Values represent least squares means of 8 and 10 sows housed in a hot or TN environment, respectively. Milk yield was estimated by weighing litters on specified days and assuming 4 kg of milk yield = 1 kg of litter gain (Patience et al., 1995). and no difference in backfat loss during lactation when compared with sows consuming the higher protein diet. However, the low-protein diets may have been slightly deficient in threonine and valine (58 and 82% of dietary lysine supply, respectively), creating AA deficiencies that would depress sow performance and increase sow protein loss. In contrast, Renaudeau et al. (2001) reported less BW and backfat loss during lactation in a hot environment when sows were fed a reduced protein (3.4% reduction in CP) or lower protein, fat-supplemented diet (2.4% reduction in CP and 4% soy oil addition). However, the benefit of reduced protein or fat supplementation did not improve milk yield or litter growth rate in the 29 C lactation environment (Renaudeau and Noblet, 2001). Evaluating increased nutrient density of the diet for sows lactating in a hot environment, Mullan et al. (1992) reported no deleterious or beneficial effects of a high-energy, high-protein diet (215 kcal/kg increase in DE and 5% increase in CP content) on sow lactation performance or BW loss. Shurson et al. (1986) reported no benefit with high fat inclusion (10% added fat) to lactation diets for sows farrowing during the summer months. Schoenherr et al. (1989) also reported no benefit of fat supplementation (32 C temperature) on minimizing sow BW loss or improving litter gain. Therefore, removing the sow from the stress of lactation in a hot environment or improving the sow s rate of heat loss (McGlone et al., 1988; Black et al., 1993) during lactation are better alternatives. A direct effect of heat stress on sow milk yield that is independent of the reduction in feed intake has been reported (Mullan et al., 1992; Black et al., 1993; Messias de Braganca et al., 1998). Reduced lactation performance during heat stress may also be due to alterations in blood flow away from the mammary gland to increase heat dissipation (Black et al., 1993) and/or to alterations in the sow s endocrine responses during

2048 Spencer et al. Table 6. Effect of environmental temperature and day of lactation (1, 5, 10, 15, or 19) on milk composition from multiparous sows a 21 C 32 C P < b Item 1 5 10 15 19 1 5 10 15 19 SE T D T D Nutrient, % Crude protein 13.48 5.46 4.90 5.07 5.14 13.79 5.40 4.89 4.88 4.77 0.69 0.65 0.001 0.62 Fat 5.66 7.46 8.37 7.57 8.15 5.84 8.10 7.69 7.08 7.81 0.11 0.87 0.001 0.32 Lactose 3.28 5.50 5.62 5.69 5.62 3.03 5.42 5.61 5.83 5.79 0.03 0.92 0.001 0.134 Residual solids 4.19 6.41 6.53 6.60 6.53 3.94 6.33 6.52 6.74 6.70 0.03 0.92 0.001 0.14 Immunoglobulins IgA, mg/ml 23.20 13.58 13.38 14.81 11.92 36.81 14.61 13.35 10.85 11.59 0.81 0.33 0.001 0.02 IgG, mg/ml 92.87 2.88 2.66 3.32 2.51 98.47 2.91 2.71 2.07 2.01 1.4 0.79 0.001 0.95 IgM, mg/ml 33.19 4.29 3.28 3.51 2.17 31.66 4.38 2.50 1.82 2.35 1.0 0.72 0.001 0.99 Cortisol, ng/ml 23.72 4.36 5.02 5.08 4.88 37.5 4.45 5.61 4.94 4.30 0.80 0.12 0.001 0.03 a Values represent least squares means of 8 and 10 sows lactating in a hot (32 C) or thermoneutral (21 C) environment, respectively. b Tested for the main effects of environmental temperature (T), day of lactation (D), and their interaction (T D). high temperatures (Messias de Braganca et al., 1998). It is possible that high environmental temperatures during lactation may inhibit the rate of mammary growth directly, thereby reducing milk output and litter BW gain. No reports have confirmed the direct effects of temperatures outside the thermal comfort zone on the growth of sow mammary tissue. However, Kim et al. (1999) reported that the mammary gland continues to grow after farrowing, at least through d 21 of lactation. Heat stress reduced piglet daily weight gain 11 and 30% when pigs nursed primiparous and multiparous sows in the HOT environment for the full 19-d lactation, respectively. These results are in agreement with studies in gilts (18% reduction, Messias de Braganca et al., 1998; 11% reduction, Mullan et al., 1992) and multiparous sows (21% reduction, Johnston et al., 1999; 17% reduction, Renaudeau and Noblet, 2001). Variations in the litter gain response in hot environments were probably due to differences in experimental temperatures. In our study, milk output was estimated from litter growth rates of multiparous sows that were collected for milk analysis. Milk production was reduced by approximately 20% the first 5 d of lactation and by approximately 30% from d5to19 compared with sows in the TN environment. Using the equations of Noblet and Etienne (1989) to estimate sow milk output for the entire 19-d lactation period, milk output was reduced approximately 23% in sows lactating in the HOT environment. There were no effects of environmental temperature on protein, fat, lactose, or solids content of milk. Renaudeau and Noblet (2001) reported no effects of temperature on milk protein, fat, or lactose, but high environmental temperature increased milk dry matter and energy content. Similar results were reported by Schoenherr et al. (1989). An increase in milk energy content with no increase in protein content would alter the protein content of the weaned piglet, and suggests an alteration in piglet weight gain due to differences in the composition of gain. Using the equations of Noblet and Etienne (1989) to estimate milk nutrient intake (pig per day) and assuming 1 g of milk nitrogen is equal to 6.25 g of milk CP, then grams of CP per kilocalorie was similar (0.04 g of CP/kcal) between environments. Therefore, it is not expected that milk composition altered piglet gain or composition of gain in this study and that the decreases in litter growth Table 7. Subsequent nursery performance (47 d) of pigs previously sow reared throughout lactation (19 d) or sow reared to 14 d and reared solely on milk replacer from d 14 to 19 (14 d) by parity 1 or 2+ sows in a thermoneutral (21 C) or hot (32 C) environment a 21 C 32 C 14 d 19 d 14 d 19 d b P < Item 1 2+ 1 2+ 1 2+ 1 2+ SE T L P L P T L T P T L P Number of pigs 94 211 115 188 78 177 61 216 Day 19, kg/pig 8.4 8.2 6.5 7.5 7.3 8.2 5.3 6.0 0.08 0.09 0.01 0.01 0.20 0.05 0.22 0.10 Day 66 weight, kg/pig 27.2 29.9 24.6 28.5 26.5 28.8 22.6 25.7 0.15 0.57 0.01 0.01 0.22 0.03 0.36 0.77 Day 19 to 66 gain, g pig 1 d 1b 399 463 386 449 408 440 367 422 3.1 0.83 0.01 0.01 0.59 0.27 0.11 0.54 a Values are least squares means. b T = main effect of temperature, L = main effect of lactation length, P = main effect of parity.

Early weaning during heat stress 2049 Table 8. Subsequent reproductive performance of parity 1 or 2+ sows previously lactating for 14 or 19 d in a thermoneutral (21 C) or hot (32 C) environment a 21 C 32 C 14 d 19 d 14 d 19 d P < b Item 1 2+ 1 2+ 1 2+ 1 2+ SE T L P L P T P T L P Number of sows 9 22 8 16 7 18 6 22 Wean to estrus, days 7.7 6.6 9.3 7.2 9.2 6.8 22.8 5.3 0.75 0.05 0.06 0.01 0.06 0.03 0.10 Total litter size c 10.58 10.68 10.53 11.53 9.23 10.45 8.18 9.75 0.29 0.03 0.75 0.19 0.71 0.57 0.87 Litter weight, kg c 17.41 17.66 17.36 18.51 15.58 16.65 13.67 15.42 0.45 0.05 0.61 0.35 0.76 0.75 0.97 Number of pigs born alive 9.9 10.08 10.26 10.68 8.5 9.82 7.67 8.99 0.28 0.01 0.81 0.26 0.94 0.47 0.94 Born alive litter weight, kg 16.66 17.05 16.61 17.59 14.83 15.89 13.04 14.51 0.46 0.04 0.56 0.40 0.85 0.81 0.98 Live pig weight, kg 1.76 1.74 1.62 1.66 1.76 1.65 1.75 1.68 0.03 0.90 0.51 0.62 0.74 0.49 0.96 a Values represent least squares means. b T = main effect of temperature, L = main effect of lactation length, P = main effect of parity. Interaction of T L was not significant (P > 0.10). c Value includes born alive and stillbirths. were in direct relation to depressions in milk output in the HOT environment. A significant day of lactation effect was observed for milk protein content which dropped to approximately 5% by d 5 of lactation, and for milk fat and lactose contents, which increased to 7.7 and 5.5%, respectively. These results are similar to reports of milk composition throughout lactation (Boyd et al., 1982; Klobasa et al., 1987). However, a lactation day temperature interaction occurred for milk cortisol and IgA (P < 0.05). Sows lactating in the HOT environment had elevated levels of cortisol and IgA in their colostrum compared with sows lactating in TN. There were no differences in these parameters in milk samples taken for the rest of lactation. Previous studies have shown an increase in sow colostrum cortisol content due to elevated environmental Figure 3. Pig weight distribution after a 47-d nursery period (d 66 of age). Pigs were weaned at 19 d of age from sows housed in a constant 32 C (hot) or 21 C (TN) lactation environment. Pigs nursed the sow for 19 d (19) or nursed the sow to 14 d and then were reared to 19 d using sow milk replacer (14). temperatures before farrowing (Machado-Neto et al., 1987). Elevated levels of glucocorticoids in the neonate are believed to cause early gut maturation (Sangild et al., 1993) and have been shown to reduce the absorption of IgG in rats (Morris and Morris, 1976) and pigs (Machado-Neto et al., 1987). The depression observed in IgG absorption would depress the passive immunity transfer from the sow to the piglet, and this may compromise piglet viability to pathogens before and after weaning. In the present study, we observed no differences in morbidity or mortality between lactation environments in the preweaning (<19 d) or nursery phase (d 19 to 66). In agreement with Machado-Neto et al. (1987), there were no effects of lactation environmental temperature on milk IgG content. Colostrum IgA content was elevated 59% when sows were exposed to elevated temperature before and after farrowing (temperature day; P < 0.02). The mechanism elevating IgA content in the colostrum of heat-stressed animals is unclear. When sows were removed on d 14 of lactation and piglets were offered an unlimited supply of liquid milk via a semi-automated delivery system in the farrowing crate, the growth rate of piglets in the TN and HOT environments increased to 603 and 717 g/d, respectively. These elevated growth rates show the piglet s potential for growth and the inability of the sow to maximize this potential when nursing large litters. The benefits of milk replacer on increasing weaning weights (Azain et al., 1996) and on improving postweaning growth performance (Zijlstra et al., 1996) have been well documented. Liquid feeding postweaning also has been shown to increase nursery gains (Kim et al., 2001) and to reduce the total days to market (Kim et al., 2001). When the sow was removed on d 14, piglets in the HOT environment were lighter than piglets nursing sows in the TN environment (4.6 vs. 5.1 kg of BW; P < 0.10). However, after pigs were fed MR as the sole source of nutrients from d 14 to 19, those in both environments had similar weaning weights (8.17 vs. 8.15 kg of BW) (Table 5). Azain et

2050 Spencer et al. al. (1996) reported that MR intake was higher during the summer months and that any seasonal effects of piglet gain when fed MR were explained by differences in MR intake. Milk replacer also reduced the difference normally observed in piglet weaning weight between primiparous and multiparous sows at 21 C (Table 4). The milk yield of primiparous sows is generally lower than that of multiparous sows and can be partially explained by the fact that primiparous sows are more physiologically immature (Pluske et al., 1998), with fewer milk secretory cells in the mammary tissue. Pluske et al. (1998) showed that when primiparous sows were forced to a higher feed intake through cannula infusion multiple times a day, they partitioned extra energy differently than multiparous sows, directing more to body growth than to increasing milk production. These results support the hypothesis that MR supplementation of piglets nursing parity-one females is effective in both TN and HOT conditions. Furthermore, the uniformity of piglet weaning weights could be improved in herds since primiparous females comprise a substantial percentage of farrowing groups. Pigs that nursed sows to d 14 and that were provided supplemental MR in both environments and parity groups improved their weight advantage through the 47-d nursery period. This was especially evident for pigs previously nursing primiparous sows in the HOT environment, where an advantage of 1.8 kg of BW/ piglet at d 19 increased to a 3.9 kg of BW advantage at the end of the 47-d period on d 66. At 66 d of age, 67 and 54% of the pigs fed MR in the TN and HOT environments during lactation weighed over 27 kg of BW vs. 52 and 34% of the pigs nursing the sow for 19 d, respectively (Figure 3). These results are consistent with previous reports showing the importance of heavier weaning weights (Mahan and Lepine, 1991) on reducing the total days to market. Piglets consuming MR in the HOT environment presumably had a higher DMI at the time of weaning and may have had less adjustment time during weaning looking for a source of nutrients in the nursery. Renaudeau and Noblet (2001) reported that pigs nursing sows in the 29 C environment consumed more creep feed during lactation, which led to an attenuation of the postweaning lag on feed intake. Zijlstra et al. (1996) reported the importance of milk supplementation and DMI on intestinal morphology, as well as its impact on improving the piglet s insulin:glucagon ratio and reducing the postweaning lag. Reducing sow lactation length below 19 d has been shown to result in a reduction in the sow s subsequent litter size (Aumaitre et al., 2000). This relationship may not hold for the sow or especially the gilt subjected to extreme heat stress. Reducing the lactation length of primiparous sows in the HOT environment improved subsequent reproductive performance by reducing the days to next estrus by approximately 14 d (temperature lactation length parity, P < 0.09). Primiparous sows lactating for the full 19 d in the HOT environment lost approximately three times more LMA than primiparous sows lactating for 14 d in the HOT environment. King and Martin (1989) and King and Williams (1984a,b) showed that increased body protein mobilization in primiparous sows reduced the mean concentration of plasma LH at the time of weaning and increased the occurrence of postweaning anestrus. Similar findings were reported by Reese et al. (1982), Armstrong et al. (1986), and Zak et al. (1998). The impact of reducing the amount of tissue loss by early weaning on d 14 in the HOT environment on the wean-to-estrus interval was not significant for the multiparous sow. This is probably due to the fact that the more mature sow loses a lower percentage of its total body protein mass compared with the primiparous sow. Subsequent litter size in both parity groups was affected by lactation temperature. Heat stress during lactation reduced litter size and the total number of pigs born alive in the subsequent farrowing (P < 0.05). The negative impact of weaning sows earlier than 19 d on wean-to-estrus interval was not apparent in the TN environment in this study, even though its negative influence is found in the literature (Xue et al., 1993; Aumaitre et al., 2000). The lack of effect may have been due to the relatively high feed intakes (approximately 7 and 8 kg/sow for primiparous and multiparous sows, respectively) in the TN environment. This is supported by the data of Koketsu et al. (1997) using PigCHAMP records for farms utilizing early weaning (less than 19 d). They found that the negative effects of reduced lactation length on farrowing rate could be ameliorated by optimizing sow feed intake (>5.7 kg/d) during lactation. In the present study, the negative effect of decreasing lactation length below 19 d on subsequent litter size was not significant. In the TN envirnoment, early weaning numerically depressed the subsequent litter size and number born alive for multiparous sows, but not for primiparous sows. Koketsu and Dial (1998) reported no differences in subsequent litter size in sows of parities one and two as lactation length increased from approximately 15 to 19 d, but they did find improvements in next litter size in sows of parities three to six when lactation length was increased. Reducing the lactation length in the HOT environment numerically increased the subsequent litter size for primiparous and multiparous sows (increase of 1.05 and 0.70 pigs/litter, respectively) compared to both parities lactating for 19 d in the HOT environment, but these effects were not significant. These results suggests however, that reduced lactation length of primiparous and multiparous sows in hot environments reduces the negative impact of heat stress on subsequent litter size, and reduces the production inefficiencies associated with high environmental temperatures.

Early weaning during heat stress 2051 Implications This study shows the benefit of decreasing sow lactation length and supplying milk replacer to piglets during periods of heat stress to conserve sow tissue loss, improve subsequent reproductive efficiency, and regain piglet growth lost due to decreased sow milk yield. By reducing sow tissue loss, producers can improve reproductive efficiency lost to high environmental temperatures, especially in herds containing of a high percentage of first-litter females. By providing milk replacer to piglets previously nursing primiparous sows in thermoneutral and hot environments, the difference in piglet weaning weight was decreased between firstparity and multiparous sows. Supplementing milk replacer to litters nursing first-litter females may be an effective method to decrease weaning weight variation, especially in hot environments. Early weaning combined with brief milk replacer may be most applicable in systems in which there is parity separation or where first-parity females are weaned at 15 to 16 d. Literature Cited AOAC, 1995. Official Methods of Analysis. 16th ed. Assoc. of Offic. Anal. Chem., Arlington, VA. Armstrong, J. D., J. H. Britt, and N. M. Cox. 1986. Seasonal differences in function of the hypothalamic-hypophysial-ovarian axis in weaned primiparous sows. J. Reprod. Fertil. 78:11 20. Aumaitre, A. L., J. Y. Dourmad, and J. Dagorn. 2000. Management systems for high productivity of sows in Europe. Pig News Info. 21:89N 97N. Azain, M. J., T. Tomkins, J. S. Sowinski, R. A. Arentson, and D. E. Jewell. 1996. Effect of supplemental pig milk replacer on litter performance: Seasonal variation in response. J. Anim. Sci. 74:2195 2202. Black, J. L., B. P. Mullan, M. L. Lorschy, and L. R. Giles. 1993. Lactation in the sow during heat stress. Livestock Prod. Sci. 35:153 170. Boyd, R. D., B. D. Moser, E. R. Peo, Jr., A. J. Lewis, and R. K. Johnson. 1982. Effect of tallow and choline chloride addition to the diet of sows on milk composition, milk yield and preweaning pig performance. J. Anim. Sci. 54:1 7. Boyd, R. D., K. J. Touchette, G. C. Castro, M. E. Johnston, K.U. Lee, and I. K. Han. 2000. Recent advances in amino acid and energy nutrition of prolific sows. Asian-Aus. J. Anim. Sci. 13:1638 1652. Forbes, E. B., and R. W. Swift. 1944. Associative dynamic effects of protein, carbohydrate and fat. J. Nutr. 27:453 468. Gill, J. L., and H. D. Hafs. 1971. Analysis of repeated measurements of animals. J. Anim. Sci. 33:331 336. Johnston, L. J., M. Ellis, G. W. Libal, V. B. Mayrose, W. C. Weldon, and NCR-89 Committee on Swine Management. 1999. Effect of room temperature and dietary amino acid concentration on performance of lactating sows. J. Anim. Sci. 77:1638 1644. Kim, J. H., K. N. Heo, J. Odle, I. K. Han, and R. J. Harrell. 2001. Liquid diets accelerate the growth of early-weaned pigs and the effects are maintained to market weight. J. Anim. Sci. 79:427 434. Kim, S. W., W. L. Hurley, I. K. Han, and R. A. Easter. 1999. Changes in tissue composition associated with mammary gland growth during lactation in sows. J. Anim. Sci. 77:2510 2516. King, R. H., and G. B. Martin. 1989. Relationships between protein intake during lactation, LH levels and oestrous activity in firstlitter sows. Anim. Repro. Sci. 19:283 292. King, R. H., and I. H. Williams. 1984a. The effect of nutrition on the reproductive performance of first-litter sows. 1. Feeding level during lactation, and between weaning and mating. Anim. Prod. 38:241 247. King, R. H., and I. H. Williams. 1984b. The effect of nutrition on the reproductive performance of first-litter sows. 2. Protein and energy intakes during lactation. Anim. Prod. 38:249 256. Klobasa, F., E. Werhahn, and J. E. Butler. 1987. Composition of sow milk during lactation. J. Anim. Sci. 64:1458 1466. Koketsu, Y., and G. D. Dial. 1998. Interactions between the associations of parity, lactation length, and weaning-to-conception interval with subsequent litter size in swine herds using early weaning. Prevent. Vet. Med. 37:113 120. Koketsu, Y., G. D. Dial, and V. L. King. 1997. Influence of various factors on farrowing rate on farms using early weaning. J. Anim. Sci. 75:2580 2587. Le Bellego, L., J. van Milgen, S. Dubois, and J. Noblet. 2001. Energy utilization of low-protein diets in growing pigs. J. Anim. Sci. 79:1259 1271. Machado-Neto, R., C. N. Graves, and S. E. Curtis. 1987. Immunoglobulins in piglets from sows heat-stressed preparatum. J. Anim. Sci. 65:445 455. Mahan, D. C., and A. J. Lepine. 1991. Effect of pig weaning weight and associated nursery feeding programs on subsequent performance to 105 kilograms body weight. J. Anim. Sci. 69:1370 1378. McGlone, J. J., W. F. Stansbury, and L. F. Tribble. 1988. Management of lactating sows during heat stress: effects of water drip, snout coolers, floor type and a high energy-density diet. J. Anim. Sci. 66:885 891. Messias de Braganca, M., A. M. Mounier, and A. Prunier. 1998. Does feed restriction mimic the effects of increased ambient temperature in lactating sows? J. Anim. Sci. 76:2017 2024. Morris, B., and R. Morris. 1976. The effects of corticosterone and cortisone on the uptake of polyvinyl pyrrolidone and the transmission of immunoglobulin G by the small intestine in young rats. J. Physiol. 254:389 403. Mullan, B. P., W. Brown, and M. Kerr. 1992. The response of the lactating sow to ambient temperature. Page 17 in Proc. Nutr. Soc. Aust. Noblet, J., and M. Etienne. 1989. Estimation of sow milk nutrient output. J. Anim. Sci. 67:3352 3359. Patience, J. F., P. A. Thacker, and C. F. M. de Lange. 1995. Swine Nutrition Guide. 2nd ed. Prairie Swine Centre Inc., Saskatoon, Saskatchewan, Canada. Pluske, J. R., I. H. Williams, L. J. Zak, E. J. Clowes, A. C. Cegielski, and F. X. Aherne. 1998. Feeding lactating primiparous sows to establish three divergent metabolic states: III. Milk production and pig growth. J. Anim. Sci. 76:1165 1171. Quiniou, N., and J. Noblet. 1999. Influence of high ambient temperatures on performance of multiparous lactating sows. J. Anim. Sci. 77:2124 2134. Reese, D. E., B. D. Moser, E. R. Peo, Jr., A. J. Lewis, D. R. Zimmerman, J. E. Kinder and W. W. Stroup. 1982. Influence of energy intake during lactation on the interval from weaning to first estrus in sows. J. Anim. Sci. 55:590 598. Renaudeau, D., and J. Noblet. 2001. Effects of exposure to high ambient temperature and dietary protein level on sow milk production and performance of piglets. J. Anim. Sci. 79: 1540 1548. Renaudeau, D., N. Quiniou, and J. Noblet. 2001. Effects of exposure to high ambient temperature and dietary protein level on performance of multiparous lactating sows. J. Anim. Sci. 79:1240 1249. Sangild, P. T., L. Diernaes, I. J. Christiansen, and E. Skadhauge. 1993. Intestinal transport of sodium, glucose and immunoglobin in neonatal pigs. Effect of glucocorticoids. Exp. Physiol. 78:485 497. Schoenherr, W. D., T. S. Stahly, and G. L. Cromwell. 1989. The effects of dietary fat or fiber addition on yield and composition of milk from sows housed in a warm or hot environment. J. Anim. Sci. 67:482 495.

2052 Spencer et al. Shurson, G. G., M. G. Hogberg, N. DeFever, S. V. Radecki, and E.R. Miller. 1986. Effects of adding fat to the sow lactation diet on lactation and rebreeding performance. J. Anim. Sci. 62:672 680. Verstegen, M. W. A., J. Mesu, G. J. M. van Kempen, and C. Geerse. 1985. Energy balances of lactating sows in relation to feeding level and stage of lactation. J. Anim. Sci. 60:731 740. Xue, J. L., G. D. Dial, W. E. Marsh, P. R. Davies, and H. W. Momont. 1993. Influence of lactation length on sow productivity. Livest. Prod. Sci. 34:253 265. Zak, L. J., I. H. Williams, G. R. Foxcroft, J. R. Pluske, A. C. Cegielski, E. J. Clowes, and F. X. Aherne. 1998. Feeding lactating primiparous sows to establish three divergent metabolic states: I. Associated endocrine changes and postweaning reproductive performance. J. Anim. Sci. 76:1145 1153. Zijlstra, R. T., K. Whang, R. A. Easter, and J. Odle. 1996. Effect of feeding a milk replacer to early-weaned pigs on growth, body composition, and small intestinal morphology, compared with suckled littermates. J. Anim. Sci. 74:2948 2959.