Feeding Lactating Primiparous Sows to Establish Three Divergent Metabolic States: III. Milk Production and Pig Growth 1

Transcription

1 Feeding Lactating Primiparous Sows to Establish Three Divergent Metabolic States: III. Milk Production and Pig Growth 1 J. R. Pluske*,2, I. H. Williams, L. J. Zak*,3, E. J. Clowes*, A. C. Cegielski*,3, and F. X. Aherne*,3,4 *Department of Agricultural, Food and Nutritional Sciences, University of Alberta, Edmonton AB T6G 2P5, Canada and Animal Science, University of Western Australia, Nedlands W.A. 6907, Australia ABSTRACT: First-litter sows fitted with stomach cannulas were used to test the hypothesis that making gilts anabolic during lactation by providing them with extra nutrition would increase milk production and pig growth. Gilts were allocated to one of three dietary treatments after farrowing: 1) restricted, sows were fed 50% of their estimated ad libitum intake; 2) ad libitum, sows were encouraged to eat as much feed as possible; and 3) superalimented, sows were infused seven times daily through their cannula to achieve a 25 to 30% increase in energy intake in excess of that achieved by sows fed on an ad libitum basis. Milk production was estimated in mid- ( d 10 to 15) and late ( d 21 to 25) lactation by a modification of the isotope dilution technique. Milk production was similar between treatments in mid- and late lactation ( P >.05), and this was reflected in a similarity in weaning litter weight ( P =.238). Milk composition was similar also ( P >.05) between dietary treatments. Superalimentation provided gilts with 38% more energy ( P <.001) than gilts fed on an ad libitum basis, and they accrued live weight (5.1 kg) and backfat (1.8 mm) during lactation ( P <.001). These data provide evidence that, unlike multiparous sows that show an increase in milk yield when made anabolic during lactation, primiparous sows seem to partition extra energy into body growth rather than into milk production. Key Words: Gilts, Lactation, Milk Production, Pigs 1998 American Society of Animal Science. All rights reserved. J. Anim. Sci : Introduction 1 Appreciation is extended to the Pig Improvement Company (Canada) for provision of experimental animals, to Paul Gregory and Steve Melnyk for managerial assistance in coordinating the study, to Janeal Mick and staff at the University of Alberta swine unit for care and assistance of animals, to Brenda Tchir and Charlane Gorsak for expert surgical assistance and animals care, to Clint Lysgaard for care and feeding of gilts, to Yvonne Stepanov, Fred Isla, and Doo-Seok Nam for help with feeding sows and collection of blood samples, to Terry Fenton for assistance with D 2 O analysis, and to Kent Ames and Paul Matzat for helpful discussions. This study was funded by the National Science and Engineering Research Council of Canada Industry-Oriented Grant (IOR ), the Alberta Pork Producers Development Corporation, and the Alberta Agriculture Research Institute. JRP received a Post-Doctoral Fellowship from the Pig Research and Development Corporation (PRDC) of Australia. 2 Present address: Monogastric Research Centre, Massey University, Private Bag , Palmerston North, New Zealand. 3 Present address: Alberta Agriculture, Swine Research group, O.S. Longman Building, Street, Edmonton AB T6H 4P2, Canada. 4 To whom correspondence should be addressed. Received April 11, Accepted October 1, Sows rarely eat sufficient feed during lactation to cover the energy lost in milk and the energy cost of maintenance (Williams et al., 1994). They mobilize body reserves to support milk production and generally become catabolic. Matzat et al. (1990) made sows anabolic during lactation and increased milk production by giving feed via a stomach cannula to increase energy intake over and above that normally consumed. With primiparous sows, the problem of low feed intake during lactation is accentuated because, relative to multiparous sows, they are physiologically younger and need extra energy for body growth (Aherne and Williams, 1992). In view of this, primiparous sows most likely partition energy differently than multiparous sows, and any additional energy goes to body growth rather than milk. Thus, we propose that if primiparous sows are provided with extra energy and maintained in an anabolic state during lactation, their requirements for body growth will be met with the additional energy and milk production and pig growth will increase. We 1165

2 1166 tested this notion by feeding lactating primiparous sows through stomach cannulas similarly to Matzat et al. (1990), so that energy intake was 25 to 30% higher than that in the ad libitum-fed groups. Effects on milk composition were also examined. Materials and Methods All experimental and surgical procedures performed in this study were approved by the University of Alberta Animal Care Committee to ensure adherence to the Canadian Council of Animal Care Guidelines. Experimental Design Primiparous sows were given a high-quality diet (Table 1) during lactation according to one of three feeding regimens: 1) restricted, 2) ad libitum, or 3) superalimented. These treatments were designed so that sows were grossly catabolic, slightly catabolic, or anabolic by the end of a 28-d lactation. Primiparous sows were made grossly catabolic by restricting their feed intake during lactation to about 3 kg/d (fed at 0600, 1330, and 2100). This was calculated to mobilize in excess of 1 kg of BW/d. Ad libitum-fed sows were offered unlimited meals each of 2 kg on a regular basis between 0600 and 2100 daily so that by the end of lactation they would be slightly catabolic. Sows were made anabolic by gastric alimentation of feed (superalimentation) that commenced within 3 d after farrowing. These sows were given their calculated feed allocation via a stomach cannula at (0600, 0830, 1100, 1330, 1600, 1800, and 2100 daily, and they were also allowed free access to feed from a trough. The amount of feed given to superalimented sows daily was based on the results of a pilot trial conducted at the University of Alberta Swine Research Centre. In this trial, it was estimated that superalimented sows would need to receive about 125% of the estimated feed intake of the ad libitum-fed group to be in an anabolic state by the end of a 28-d lactation. Cannulation Pregnant Camborough Canabrid gilts (n = 36; PIC, Acme, Alberta) were individually housed and fed 2 to 2.3 kg/d of a gestation diet (Table 1) according to their live weight. Between d 65 and 75 of gestation, all gilts underwent surgery for the insertion of a gastric cannula (Pluske et al., 1995). Diets and Animal Management At d 109 of gestation, gilts were moved to individual farrowing crates in a room accommodating 12 crates. Room temperature was maintained between 20 and 23 C. A drip-cooling system was activated when the room temperature exceeded 23 C. From d 109 of PLUSKE ET AL. Table 1. Composition of the experimental diets (as-fed basis) as a percentage of the diet Item Gestation Lactation Ingredient Barley Wheat Soybean meal (44% CP) Fish meal 5.5 Sugar 16.0 Canola oil 5.0 Tallow 2.0 Iodized salt.5 Dicalcium phosphate Limestone 1.4 Vitamin/mineral supplement a Analysis Digestible energy, MJ DE/kg Crude protein, % Lysine, % a Supplied the following per kg of complete feed: 10,000 IU vitamin A, 1,000 IU vitamin D, 80 IU vitamin E, 2 mg vitamin K, 30 mg vitamin B 12, 12 mg riboflavin, 25 mg niacin, 25 mg calcium pantothenate, 600 mg choline, 200 mg biotin, 200 mg folic acid, 5 mg ethoxyquin, 150 mg iron, 12 mg manganese, 120 mg zinc, 12 mg copper, 200 mg iodine, and 100 mg selenium. gestation and during the 28-d lactation, all sows were fed a wheat-barley-soybean diet formulated to provide 15.4 MJ DE/kg and 18.6% crude protein (Table 1). All feed troughs were thoroughly cleaned daily, and dry feed refusals were weighed. Water was freely available at all times to the sows and pigs from nipple drinkers. At parturition, animals were randomly allocated to one of the three treatments: restricted ( n = 9), ad libitum ( n = 12), and superalimented ( n = 8). To facilitate gastric alimentation of superalimented sows, a suspension was created by adding water ( 2 parts water:1 part feed) and.5% xanthan gum to the diet (Pluske et al., 1995). Litters were standardized to 8 to 10 pigs within 48 h after farrowing. At 72 h of age, pigs were processed (teeth and tail clipping, and iron and antibiotic injection). All male pigs were castrated between d 15 and 19 of lactation. Creep feed was not available. Sow live weight, backfat depth (65 mm from the midline at the last rib, or P 2 ; Scanoprobe II, Scano, Ithaca, NY), and litter weights were recorded within 24 h after farrowing and at weekly intervals thereafter until weaning. Milk Production Milk production was estimated from d 10 to 15 (mid) and from d 21 to d 25 (late) of lactation. In each measurement period, milk production was calculated from the milk intakes of individual pigs and estimated from their water turnover, as determined by dilution of injected deuterium oxide (D 2 O; Pettigrew et al., 1987; King et al., 1993). Each measurement period commenced by separating pigs from the sow

3 MILK PRODUCTION IN PRIMIPAROUS SOWS 1167 immediately after they were observed to drink milk. Each pig was weighed, and the appropriate volume of D 2 O (injected at the rate of 1.8 g/kg in midlactation and 1.4 g/kg in late lactation) was drawn into a syringe and weighed to four decimal places. After 45 min, each pig was injected i.m. with D 2 O using a 23-gauge needle. Sixty minutes after injection, a 5-mL blood sample was collected from each pig into a sterile vacuum tube via puncture of the anterior vena cava for determination of the equilibrium concentration of D 2 O in body water. At the end of each measurement period and after pigs were observed to suckle, they were weighed again, separated from the sow for 45 min, and then bled for determination of D 2 O concentration in body water. The vacuum tubes were placed at an angle of 45 and allowed to clot overnight at room temperature. The following day, serum was decanted into 10-mL polypropylene tubes and stored at 20 C until analysis. Milk samples were collected from four to eight mammary glands on each sow at the first suckling bout following the conclusion of each measurement period. To assist in milk collection, a small amount of oxytocin (.5 IU) was diluted in 5 ml of sterile, physiological saline and injected i.m. into the sow. Samples from all glands were pooled (20 to 30 ml), and then a sub sample (10 ml) was taken, placed immediately into a prechilled vial, and frozen at 20 C until analysis. Concentrations of fat, protein and lactose were determined by infrared analysis using a MilkoScan (Foss Electrics, Denmark) at the Alberta Central Milk Testing Laboratory (Edmonton, AB, Canada). The appropriateness of the MilkoScan for testing the composition of sows milk was done by first validating a bulked sample of sows milk against the internal standard used for cows milk. Ash content of milk was assumed to be 8.8 g/kg (Elliott et al., 1971), and total solids was calculated as the sum of fat, protein, lactose, and ash. Water turnover ( WTO) was calculated for each pig during each measurement period according to the calculations and assumptions presented by Dove and Freer (1979) and King et al. (1993). The concentration of D 2 O in body water was determined in pig serum using the technique of Pluske et al. (1997). Statistical Analysis Data were subjected to least squares analysis of variance for a randomized complete block design having three treatments (Snedecor and Cochran, 1967) using the GLM procedure of SAS (1990). Because this experiment was conducted in three parts, replicate was included as an independent variable in the initial analysis. Replicate was not significant for any variables analyzed ( P >.05), so the data were reanalyzed with treatment group included as the only independent variable. Variation among the experimental units (sows within block treatment) was used as the estimate of experimental error and for significance testing of treatments. For milk production data, the statistical model included stage of lactation and treatment. Differences between treatment means were tested using Fisher s protected least significant difference test. Results Although 36 gilts were placed on experiment, seven sows were withdrawn before the completion of lactation. One sow gave birth to three pigs, three litters developed severe diarrhea early in lactation and were considered aberrant, and three superalimented sows were withdrawn from the study. One of the superalimented sows was removed from the study because of metritis, another because of a rectal prolapse, and the third sow failed to recover from anesthesia associated with surgery at d 26 of lactation. Thus, total of 29 sows and litters was used, for calculation of sow production and pig growth data. In the estimation of milk production, a total of 24 and 28 sows was used in midlactation and late lactation, respectively. The disparity in sow numbers was caused by technical difficulties associated with blood collection procedures in some litters and the removal of the seven sows from the study. Sow Performance By the end of lactation, the average feed and energy intakes of superalimented sows were 38% higher ( P <.001) than those of sows offered feed on an ad libitum basis (7.2 vs 5.2 kg/d, and vs 80.1 MJ DE/d, respectively). First-litter sows fed on an ad libitum basis during lactation lost 16.3 kg of live weight and 3.7 mm of backfat (P 2 ). In contrast, gilts fed on a restricted basis lost 38.9 kg of live weight and 8.9 mm of backfat. Primiparous sows given feed via a gastric cannula, however, gained body condition during lactation, accreting 5.1 kg of live weight and 1.8 mm of backfat (Table 2). Milk Production and Milk Composition Milk yield was statistically similar ( P >.05) between dietary treatments in mid- and late lactation, with a significant decrease ( P =.018) in milk yield observed with advancing lactation (9.3 vs 8.0 kg/d for mid- and late lactation, respectively). There was no stage of lactation dietary treatment interaction for milk production ( P =.412). Milk yield in gilts offered feed on an ad libitum basis decreased by 5% between mid- and late lactation (Table 3). This decrease was more marked, however, in restricted-fed and superalimented gilts, which showed decreases in milk production of 18 and 15%, respectively. Although statistically similar, milk production in restricted-fed

4 1168 PLUSKE ET AL. Table 2. Nutrient intake and live weight and backfat changes in paimiparous sows that were restricted-fed (Restrict), ad libitum-fed (Ad libitum), or superalimented (Super) during lactation a Treatment Item Restrict (n = 9) Ad libitum (n = 12) Super (n = 8) Lactation length, d 27.9 ± ± ±.33 Feed intake, kg/d d ±.17 x 4.5 ±.16 y 5.6 ±.16 z d ±.13 x 5.4 ±.12 y 7.5 ±.13 z d ±.16 x 5.9 ±.15 y 8.3 ±.16 z d 22-weaning 3.0 ±.11 x 5.1 ±.11 y 7.6 ±.13 z d 1-weaning 2.9 ±.11 x 5.2 ±.11 y 7.2 ±.13 z Energy intake, MJ DE/d d ± 2.62 x 69.3 ± 2.46 y 86.2 ± 2.46 z d ± 2.00 x 83.2 ± 1.85 y ± 2.00 z d ± 2.46 x 90.9 ± 2.31 y ± 2.46 z d 22-weaning 46.2 ± 1.69 x 80.1 ± 1.69 y ± 3.54 z d 1-weaning 44.7 ± 1.76 x 80.1 ± 1.69 y ± 1.97 z Live weight, kg Farrowing 176 ± ± ± 3.0 Weaning 137 ± 3.7 x 163 ± 3.2 y 182 ± 4.2 z Weight change 38.9 ± 3.65 x 16.3 ± 3.48 y +5.1 ± 3.22 z Backfat (P 2 ), mm Farrowing 18.3 ± ± ±.51 Weaning 9.4 ±.84 x 14.4 ±.80 y 18.9 ±.94 z Backfat change 8.9 ±.69 x 3.7 ±.66 y +1.8 ±.77 z a Least squares means ± SE. x,y,z Within each item and between dietary treatments, values not having the same superscript differ (P <.001). sows was between 14 and 20% lower than that in superalimented and ad libitum-fed sows in late lactation, commensurate with a reduced litter weight at weaning (Table 4). Milk composition was unaffected ( P >.05) by dietary feeding regimen in midand late lactation (Table 3). There was no significant stage of lactation treatment interaction for pig growth rate within the two milk production measurement periods ( P =.215) (Table 3). Significant main effects on pig growth rate within periods existed for stage of lactation (269 vs 228 g/d for mid- and late lactation, respectively; P =.013, SED = 43.6) and treatment (229, 275, and 243 g/ d for restricted-fed, ad libitum-fed, or superalimented sows respectively; P =.038, SED = 48.4). When milk production was measured in late lactation, however, one superalimented sow supported a pig growth rate of only 50 g/d. Exclusion of this sow from the data set increased pig growth rate in the superalimentation treatment from 206 to 232 g/d and resulted in superalimented sows supporting a pig growth rate not different from restricted or ad libitum-fed sows ( P =.307). Pig Growth During Lactation Weekly litter weights and pig growth rate over the entire lactation, adjusted for birth weight and number of pigs weaned, were similar ( P >.05) for all treatment groups. Pigs suckling primiparous sows that were restricted-fed during lactation were on average, however, 9% lighter at weaning than pigs suckling sows that were ad libitum-fed or superalimented (Table 4). Discussion Milk Production and Pig Growth Despite receiving 38% more energy and being anabolic at the end of a 28-d lactation compared to animals fed on an ad libitum basis, primiparous sows that were superalimented did not produce more milk or support a superior pig growth rate. The superalimented and ad libitum-fed sows produced the same estimated amount of milk energy (66 MJ/d) and milk N (76 g/d) during lactation; however, superalimented sows retained an additional 31 g/d of nitrogen over and above the sows fed on an ad libitum basis. Almost all of this nitrogen, 96%, appeared in maternal protein (Clowes et al., 1998). These data do not support our hypothesis and provide evidence that, unlike multiparous sows, first-litter sows seem to partition extra energy into body growth rather than into milk production. Matzat et al. (1990) found that multiparous sows made anabolic by gastric infusion of nutrients during lactation produced more milk and grew their litters at a faster rate than sows allowed ad libitum access to

5 MILK PRODUCTION IN PRIMIPAROUS SOWS 1169 Table 3. Milk yield, pig growth rate, and milk composition determined at two stages of lactation in primiparous sows that were restricted-fed (Restrict), ad libitum-fed (Ad libitum), or superalimented (Super) a Treatment Item Restrict Ad libitum Super Midlactation ( d 10 15) n b Milk yield, kg/d cd 8.9 ± ± ±.57 Pig growth, g/d cde 248 ± ± ± 19.2 Milk composition, % Protein 5.2 ± ± ±.15 Fat 7.9 ± ± ±.23 Lactose 5.6 ± ± ±.16 Total solids f Late lactation ( d 21 25) n b Milk yield, g/kg c,d 7.3 ± ± ±.61 Pig growth, g/d c,d,e 210 ± ± ± 20.3 Milk composition, % Protein 5.2 ± ± ±.15 Fat 7.1 ± ± ±.24 Lactose 5.3 ± ± ±.18 Total solids f a Values are least squares means ± SE (unless otherwise stated). b Number of observations. c Interaction least squares means ± SE presented for milk yield and pig growth rate only. d P =.412 for stage of lactation treatment interaction on milk yield, and P =.215 for stage of lactation treatment interaction on pig growth rate. e Pig growth rate as determined within each milk production estimation period (adjusted for litter size at the start of each milk production estimation period). f Assumes an ash concentration of 8.8 g/kg (Elliott et al., 1971). feed. These workers concluded that the sow responds to an increase in nutrient intake by increasing nutrient output in milk. The difference in the way gilts and older sows partition additional energy above that required to maintain zero energy balance is most likely due to gilts being physiologically younger and, therefore, at a lower proportion of their mature body size than older sows. In the present experiment, milk yield was statistically similar between treatments in mid- and late lactation and was within the range reported by King et al. (1993) and Toner at al. (1996) for primiparous sows suckling between 7 and 10 pigs. Although superalimented gilts produced 7 and 10% more milk per day in midlactation than ad libitum-fed and restricted-fed gilts, respectively, a large within-treatment CV for milk production (16 to 24%) resulted in no statistical significance. Sow milk yield was 14% lower ( P =.018) in late lactation than in midlactation; the milk production of restricted-fed and superalimented gilts declined by 18 and 15%, respectively, between these two periods. In contrast, milk production of sows fed on an ad libitum basis decreased by only 5%. Milk yield throughout lactation depends on a variety of factors, including the stage of lactation (King et al., 1993), the number of pigs suckled per sow (King et al., 1989; Auldist and King, 1995; Toner et al., 1996), pig weight (Auldist and King, 1995), the intensity of suckling (Auldist and King, 1995), temperature (Mullan et al., 1992), season (Mullan et al., unpublished data), and dietary protein (lysine) and energy intake (Verstegen et al., 1985; Noblet and Etienne, 1986; Tokach et al., 1992; King et al., 1993). Our aim in this study was to keep as many of these variables as possible constant and vary only nutrient intake, such that the effects observed on milk production, pig growth, and reproduction (see Zak et al., 1998) could be attributed directly to the nutrient load entering the sow. It is unlikely, for example, that differences in litter size would have accounted for the lack of statistical difference we observed in milk yield. During the two periods in which milk production was estimated, the average litter size for sows on all three treatment groups varied by only.6 pigs (range: 8.4 to 9.0 pigs/sow). Nevertheless, a larger nutrient drain (i.e., larger litter size and hence greater suckling stimulus) on superalimented sows may have resulted in a greater transfer of nutrients to the mammary gland with reduced traffic to body reserves. Conversely, it is well recognized that severe restriction of feed intake during lactation reduces milk production and pig growth (see review by Pettigrew, 1995). Decreased milk production by restricted-fed sows observed in this study, especially toward the end of lactation, can most likely be attributed to a lack of nutrients needed to support milk synthesis in the face of increasing milk demand by the pigs. The decrease in milk production of superalimented sows, however, was unexpected. Table 4. Pig performance during lactation a Treatment b Item Restrict Ad libitum Super No. of pigs weaned 8.8 ± ± ±.93 Litter weight, kg Birth 12.7 ± ± ±.55 d ± ± ±.97 d ± ± ± 1.66 d ± ± ± 2.31 Weaning 64.1 ± ± ± 2.85 Litter gain, birth to weaning, kg 51.5 ± ± ± 2.85 Lactation growth rate, g/d 214 ± ± ± 11.5 a Values are least squares means ± SE adjusted for birth weight and number of pigs weaned. b Restrict = restricted-fed; Ad libitum = ad libitum-fed; Super = superalimented.

6 1170 PLUSKE ET AL. Even though we have suggested that superalimented sows did not produce more milk because nutrients were partitioned toward body growth, we observed that in the final week of lactation when most sows were being infused in excess of 9 kg/d, some sows hyperventilated concomitant with an elevated rectal temperature ( > 40 C). Based on this observation, we suggest that these sows reached their evaporative critical temperature ( ECT) despite a room temperature of 21 to 23 C. Infusion was stopped in these cases to allow the sows to recover. Mullan et al. (1992) and Black et al. (1993) showed a decrease in milk production when lactating, first-litter sows were maintained at high temperatures. Mullan et al. (1992) elegantly demonstrated that the decrease in milk production resulting from keeping sows at 30 C was due to a decrease in the efficiency at which milk was produced, because pairfed sows kept at 20 C produced 9% more milk and sustained a higher rate of pig growth. These data suggest that under circumstances in which sows are kept at temperatures approaching their ECT, sows actively divert nutrients away from the mammary gland and to the periphery to assist in metabolic heat disposal. Whether this occurred in the present experiment cannot be determined, but it may help to explain some of the large decrease in milk production with the concomitant decrease in pig growth observed between mid- and late lactation in superalimented gilts. Alternatively, and as suggested by Williams (1995), the lack of any increase in milk production by superalimented sows may reflect nothing more complex than a shortage of milk secretory cells in the mammary tissue of these animals. The resolution of these confounding issues can only be determined with further experimentation. Despite pigs growing at the same rate throughout lactation, pigs suckling restricted-fed sows were, on average,.98 kg lighter at weaning than their counterparts suckling ad libitum-fed and superalimented sows. Similar findings have been reported elsewhere (see Pettigrew, 1995) and confirm that by the end of a 28-d lactation, grossly catabolic primiparous sows support an inferior pig growth rate compared to sows allowed free access to feed. Milk Composition Milk composition was remarkably consistent between experimental treatments and differed very little between mid- and late lactation. This supports the data of numerous authors, including Revell et al. (1995) and Toner et al. (1996). Revell et al. (1995) reported no differences in the average concentrations of fat, protein, and lactose in sows made either lean (18 mm backfat) or fat (25 mm backfat) at farrowing and then fed either a high- (19% CP/kg) or lowprotein (7.9% CP/kg) diet in lactation. In contrast, Noblet and Etienne (1986) found that sows restricted to a low level of energy intake had a higher concentration of fat in milk and produced more milkfat than did sows on a higher energy intake. King et al. (1993) reported significant linear increases in the concentration of total solids and fat in milk during mid- and late lactation in response to an increasing level of dietary protein from 62 to 238 g of CP/kg. The absolute values for milkfat content reported by King et al. (1993) were also considerably higher, in agreement with the work of Atwood and Hartmann (1992). Fat concentrations reported in our study are more in line with values reported by Elliott et al. (1971), Klobasa et al. (1987), and Toner et al. (1996). Variation among reports is most likely attributable to differences in diet, sampling technique, genotype, sow body condition, and(or) analytical methods. Efficiency of Pig Gain We observed differences among the three dietary treatments in the efficiency with which pigs converted milk to body gain. In midlactation, and for restrictedfed, ad libitum-fed, and superalimented sows, respectively, pigs required 4.15, 3.76 and 3.84 g milk/g body gain. In late lactation, and for restricted-fed, ad libitum-fed, and superalimented sows, respectively, pigs required 4.13, 3.90, and 3.92 g milk/g body gain. These data are difficult to explain, especially in the absence of any treatment differences in milk composition. The higher milk conversion efficiency of pigs suckling restricted-fed sows may have arisen from a higher maintenance requirement associated with increased physical activity and(or) an increased number of sucklings as pigs attempted to increase their milk intake, especially in late lactation. However, evidence to support this notion is not evident. In conclusion, primiparous sows made anabolic during lactation seemed to shunt excess nutrients toward their maternal protein reserves rather than toward the mammary gland and milk production. Although we observed no increase in milk production in superalimented sows that received 38% more energy than ad libitum-fed sows, a larger nutrient drain on the sow provided by an increased suckling demand and(or) a cooler ambient temperature may have produced different results with less nitrogen being diverted toward body reserves. Implications This study has demonstrated that first-litter sows made anabolic by the end of lactation by infusing feed directly into the stomach shunt additional nutrients into maternal tissue rather than into increased levels of milk production. Given that milk production and pig growth were similar between superalimented gilts and those fed on an ad libitum basis, first-litter sows should be encouraged to eat as much feed as possible during lactation to promote maximum rates of pig gain.

7 Literature Cited Aherne, F. X., and I. H. Williams Nutrition for optimizing breeding herd performance. Vet. Clin. North Am. Food Anim. Pract. 8: Atwood, C. S., and P. E. Hartmann Collection of fore- and hind-milk from the sow and the changes in milk composition during suckling. J. Dairy Res. 59: Auldist, D. E., and R. H. King Piglets role in determining milk production in the sow. In: P. D. Cranwell and D. P. Hennessy (Ed.) Manipulating Pig Production V. pp Australasian Pig Science Association, Werribee, VIC, Australia. Black, J. L., B. P. Mullan, M. L. Lorschy, and L. R. Giles Lactation in the sow during heat stress. Livest. Prod. Sci. 35: Clowes, E. J., I. H. Williams, V. E. Baracos, J. R. Pluske, A. C. Cegielski, L. J. Zak, and F. X. Aherne Feeding lactating primiparous sows to establish three divergent metabolic states: II. Effect on nitrogen partitioning and skeletal muscle composition. J. Anim. Sci. 76: Dove, H., and M. Freer The accuracy of tritiated water turnover rate as an estimate of milk intake in lambs. Aust. J. Agric. Res. 30: Elliott, R. F., G. W. Vander Noot, R. L. Gilbreath, and H. Fisher Effect of dietary protein level on composition changes in sow colostrum and milk. J. Anim. Sci. 32: King, R. H., M. S. Toner, and H. Dove Pattern of milk production in sows. In: J. L. Barnett and D. P. Hennessy (Ed.) Manipulating Pig Production V. p 98. Australasian Pig Science Association, Werribee, VIC, Australia. King, R. H., M. S. Toner, H. Dove, C. S. Atwood, and W. G. Brown The response of first-litter sows to dietary protein level during lactation. J. Anim. Sci. 71: Klobasa, F., E. Werhahn, and J. E. Butler Composition of sow milk during lactation. J. Anim. Sci. 64: Matzat, P. D., M. G. Hogberg, R. L. Fogwell, and E. R. Miller Lactation performance in high producing gilts fed in excess of ad libitum. In: Report of Swine Research, AS-SW pp Michigan State University, East Lansing. Mullan, B. P., W. Brown, and M. Kerr The response of the lactating sow to ambient temperature. Proc. Nutr. Soc. Aust. 17:215. Noblet, J., and M. Etienne Effect of energy level in lactating sows on yield and composition of milk and nutrient balance of piglets. J. Anim. Sci. 63: MILK PRODUCTION IN PRIMIPAROUS SOWS 1171 Pettigrew, J. E The influence of substrate supply on milk production in the sow. In: P. D. Cranwell and D. P. Hennessy (Ed.) Manipulating Pig Production V. pp Australasian Pig Science Association, Werribee, VIC, Australia. Pettigrew, J. E., S. G. Cornelius, R. L. Moser, and A. F. Sower A refinement and evaluation of the isotope dilution method for estimating milk intake by pigs. Livest. Prod. Sci. 16: Pluske, J. R., T. W. Fenton, M. L. Lorschy, J. E. Pettigrew, A. F. Sower, and F. X. Aherne A modification to the isotopedilution technique for estimating milk intake of pigs using pig serum. J. Anim. Sci. 75: Pluske, J. R., I. H. Williams, A. C. Cegielski, and F. X. Aherne Stomach cannulation of pregnant gilts for nutrition studies during lactation. Can. J. Anim. Sci. 75: Revell, D. K., I. H. Williams, J. L. Ranford, B. P. Mullan, and R. J. Smits A high-protein diet maximizes milk output and minimizes weight loss in lactation. In: P. D. Cranwell and D. P. Hennessy (Ed.) Manipulating Pig Production V. p 136. Australasian Pig Science Association, Werribee, VIC, Australia. SAS SAS/STAT User s Guide (Release 6.06). SAS Inst. Inc., Cary, NC. Snedecor, G. W., and W. G. Cochran Statistical Methods (6th Ed.). Iowa State University Press, Ames. Tokach, M. D., J. E. Pettigrew, B. A. Crooker, G. D. Dial, and A. F. Sower Quantitative influence of lysine and energy intake on yield of milk components in the primiparous sow. J. Anim. Sci. 70: Toner, M. S., R. H. King, F. R. Dunshea, H. Dove, and C. S. Atwood The effect of exogenous somatotropin on lactation performance of first-litter sows. J. Anim. Sci. 74: Verstegen, M.W.A., J. Mesu, G.J.M. van Kempen, and C. Geerse Energy balances of lactating sows in relation to feeding level and stage of lactation. J. Anim. Sci. 60: Williams, I. H Sows milk as a major nutrient source before weaning. In: P. D. Cranwell and D. P. Hennessy (Ed.) Manipulating Pig Production V. pp Australasian Pig Science Association, Werribee, VIC, Australia. Williams, I. H., F. X. Aherne, and D. K. Revell Food intake in lactation: A limit to production? In: G. R. Foxcroft (Ed.) Advances in Pork Production (Vol. 5). pp Proc Banff Pork Seminar, University of Alberta, Edmonton, Canada. Zak, L. J., I. H. Williams, G. R. Foxcroft, J. R. Pluske, A. C. Cegielski, E. J. Clowes, and F. X. Aherne Feeding lactating primiparous sows to establish three divergent metabolic states: I. Associated endocrine changes and postweaning reproductive performance. J. Anim. Sci. 76: