Effect of starter feeding program on growth performance and gains of body components from weaning to market weight in swine 1,2,3

Transcription

1 Effect of starter feeding program on growth performance and gains of body components from weaning to market weight in swine 1,2,3 K. Y. Whang 4, F. K. McKeith, S. W. Kim, and R. A. Easter 5 Department of Animal Sciences, University of Illinois, Urbana ABSTRACT: Three experiments were conducted to test the hypothesis that different starter feeding programs (High, high quality; Low, low quality) will affect growth performance and body composition of pigs from weaning to market weight and that this effect may be influenced by gender (barrows or gilts) and breed (F, Yorkshire-Duroc Hampshire; P, PIC Camborough 15 PIC line 405). In Exp. 1, 21 ± 4-d-weaned F pigs (n = 90) were used in a 2 (High or Low) 2 (barrows or gilts) factorial design. In Exp. 2, 21 ± 3-d-weaned pigs (n = 184) were used in a 2 (F or P) 2 (High or Low) 2 (barrows or gilts) factorial design. In Exp. 3, 21 F pigs from each gender and feeding program treatment were killed at d 0, 3, 7, 14, 42, 82, or 152 postweaning for evaluating body composition. Two starter feeding programs (High or Low) were applied to pigs for 6 wk postweaning. Pigs from both High and Low treatments were provided the same corn-soybean meal-based diets for the growing and finishing periods. Although the ADG of all pigs receiving the High treatment during the early starter period were higher (P <.01) than those of the Low, the terminal BW of F barrows were similar between High and Low (Exp. 1 and 2) and those of gilts were similar between High and Low (Exp. 2) (P >.80). However, the BW of P pigs receiving the High treatment, regardless of gender, tended to be heavier than those receiving the Low (Exp. 2) and F barrows receiving the High treatment tended to be heavier than those receiving the Low (Exp. 1). For the first 7 d postweaning, the High-fed pigs gained more protein (P <.05) and lost less fat (P <.05) than Low-fed pigs. During the growing-finishing period, the Low-fed pigs exhibited compensatory protein gain and achieved a body protein content similar (P <.60) to High-fed pigs by termination. Protein gains from weaning to termination between High- and Low-fed pigs were not different in Exp. 2 and 3. The protein gain of gilts was higher (P <.05) than that of barrows. Similarly, fat gain within genders was not affected by starter feeding program. Fat gain of gilts, however, was lower (P <.08) than that of barrows in Exp. 3. In conclusion, the nutritional quality of the starter feeding program affected growth performance immediately after weaning but did not affect protein gain over the entire production period. Key Words: Body Composition, Body Protein, Feed Rations, Growth Rate, Pigs 2000 American Society of Animal Science. All rights reserved. J. Anim. Sci : Introduction Following weaning, pigs usually experience reduced weight gains due to environmental stress, reduced feed intake, and abrupt changes in dietary composition and physical state (Leibbrandt et al., 1975). Growth retar- 1 The study reported herein was approved by the University of Illinois Laboratory Animal Care Advisory Committee (#A2B472). 2 Presented in part at the ASAS Midwestern Section Meetings, Des Moines, IA, 1994, Abstr. 116, and 1995, Abstr. 174; and at the ADSA/ ASAS Annual Meeting, Minneapolis, MN, 1994, Abstr The authors wish to acknowledge financial support provided by the National Pork Producers Council, Des Moines, IA. 4 Current address: Department of Animal Science, Korea University, 1, 5-ka, Anam-dong, Sungbuk-ku, Seoul , Korea. 5 Correspondence: 116 Animal Sciences Laboratory, 1207 West Gregory Drive (phone: (217) ; fax: (217) ; r-easter@uiuc.edu). Received July 29, Accepted July 12, dation can be minimized by feeding diets rich in milk products, processed carbohydrates, and animal-source proteins. These diets are costly in comparison to simple cereal-soy diets that support modest, but somewhat reduced, weight gain. However, the relationship between the early postweaning growth of healthy pigs and subsequent growth performance to maintain weight and carcass composition is not clearly established. Ultimately, the growth component of interest is lean muscle; however, body weight gain includes both muscle (lean) and fat. In a key experiment, Le Dividich et al. (1980) demonstrated that the pig is able to use energy from fat reserves for protein synthesis. Campbell and Biden (1978) showed that pigs offered a weaning diet with a suboptimal protein content exhibited compensatory growth in the subsequent period when fed adequately. These findings were confirmed in subsequent work by Campbell and Dunkin (1983) and Campbell et al. (1983). However, since that time, multiphase 2885

2 2886 Whang et al. early-weaning programs have been introduced that employ diets rich in simple, highly digestible carbohydrates and proteins that are quite different from the wheat-soybean meal or meat-and-bone meal diets used in the earlier investigations. Additionally, modern genotypes have greater potential for lean growth (Gu et al., 1991; Yen et al., 1991; Cameron et al., 1999) and reduced voluntary feed intake during the growing-finishing phase (Ewan, 1986). This experiment was conducted to test the hypothesis that qualitative differences in the starter feeding program would alter growth performance and body composition from weaning to market weight of pigs of different genders and breeds. Materials and Methods In Exp. 1, 45 barrows (B) and 45 gilts (G) (age = 21 ± 4 d, weight = 5.29 ±.14 kg), derived from mating crossbred females of Yorkshire-Duroc origin to Hampshire sires (F) were used. A 2 2 factorial design was applied to test the effects of gender and different starter feeding programs. Pigs were randomly assigned to experimental treatments. Animals were divided according to gender (gilt, G; barrow, B). A high- (High, or H) or low- (Low, or L) quality starter feeding program was provided for six wk postweaning. The High program was a three-phase starter feeding program, and the Low was a single-phase feeding program. The Phase 1, 2, and 3 diets of High were provided for wk 1, wk 2 to 3, and wk 4 to 6, respectively. Pigs consumed ad libitum. Each treatment was labeled according to feeding program-gender combination (FHB, FHG, FLB, and FLG). In Exp. 2, a total of 184 (age = 21 ± 3 d, weight = 4.93 ±.09 kg) from two different breeds (F, as above, or P, PIC Camborough 15 PIC line 405) and gender (gilts or barrows) were used. Pigs were randomly assigned to a 2 (breed) 2 (feeding program) 2 (gender) factorial arrangement of eight treatments. Pigs consumed ad libitum. Each treatment was labeled according to the breed, feeding program, and gender combination (FHB, FHG, FLB, and FLG, as above; and PHB, PHG, PLB, and PLG). Ten pigs from each treatment (10 pigs 2 feeding programs 2 genders 2 genotypes) were slaughtered at termination. Carcass measurements were taken to estimate the lean body mass by using the NPPC (1991) formula. In Exp. 3, a total of 84 pigs from F breed (age = 21 ± 3 d, weight = 5.18 ±.25 kg) were grouped in a 2 (feeding program) 2 (gender) factorial arrangement of four treatments. Pigs consumed fed ad libitum. Each treatment was labeled according to feeding programgender combination (FHB, FHG, FLB, and FLG). Twelve pigs (three barrows and three gilts for High and Low treatments) were slaughtered each at d 0 (weaning), 3, 7, 14, 42, 82, and 152 (termination) of the experiment by electrical stunning followed by exsanguination (Newcomb et al., 1993) to obtain composition data. Heads were separated from the body and the head, body, and emptied viscera were separately ground three times through a 1.6-cm die. Between grinds, the material was thoroughly mixed. A composite, whole-body sample was prepared by mixing proportional subsamples of the ground head, body, viscera and blood. Samples were frozen at 20 C until analyzed for dry matter, ash, protein (N 6.25) and fat (AOAC, 1990). The water content was calculated by weight loss after drying at 105 C for 15 h in a forced-air oven. Samples were then extracted in 87% chloroform/13% methanol (vol/vol) in a Soxhlet apparatus for 24 h, and fat content was calculated by weight difference. Dry-ashing was accomplished by heating at 550 C for 24 h. A sample mass balance of 99% to 101% was accepted. If this was not achieved, the analysis was repeated. After the 6-wk starter period, pigs from both the High and Low treatments were provided the same corn-soybean meal-based diets for the growing (42 d to 82 d postweaning in Exp. 1 and 3; 41 d to 85 d postweaning in Exp. 2) and for the finishing (82 d to 152 d postweaning in Exp. 1 and 3; 85 d to 151 d postweaning in Exp. 2). Pigs consumed ad libitum. The percentage composition and calculated analysis of experimental diets for the starter and grower-finisher periods in the experiments are shown in Tables 1, 2, and 3, respectively. The nursery pigs were housed on elevated-deck pens in a nursery room equipped with solid partitions, slotted metal floors, self-feeders, and automatic waterers for the first 6 wk of the experiment. The temperature of the nursery room was approximately 26 C, with adequate ventilation provided by a mechanical fan system. The grower and finisher pigs were housed in partially slotted, concrete-floored pens with self feeders and automatic waterers. The temperatures of the growing and finishing houses were approximately 23 C and 21 C, respectively. During the experiments, individual animals were observed each day, and body weight and feed intake were measured weekly. The experiments were statistically treated as a randomized complete block with initial BW, ancestry, gender, and genotype as the blocking criteria. Pigs were the experimental unit for analysis of composition data. Statistical analyses were performed using the General Linear Models procedure (PROC GLM) in SAS/STAT software (SAS, 1989) with treatments as the main effects. Mean values were separated using Student s t- test with the following comparisons: HB vs LB, HG vs LG, H vs L, B vs G, and diet treatment gender in Exp. 1 and Exp. 3, and HB vs LB, HG vs LG, H vs L, B vs G, F vs P, diet treatment gender, diet treatment genotype, and diet treatment gender genotype in Exp. 2. Body weight data were plotted against days postweaning using a third-order polynomial equation to describe the line. Results and Discussion Pigs fed a simple cereal-soy-based starter diet had inferior growth performance (P <.05) in comparison

3 Compensatory growth in swine 2887 Table 1. Percentage composition and calculated analysis of experimental diets for weanling pigs in Exp. 1, Exp. 2, and Exp. 3 Treatment: High Low Ingredient Period: d 0 to d 7 d 7 to d 21 d 21 to d 42 a d0tod42 b Corn Dehulled soybean meal Dried skim milk Dried whey Lactose Fish meal Plasma protein (AP-820) c Soybean oil Lysine-HCl Dicalcium phosphate Ground limestone Vitamin premix d Trace-mineral salt e ASP-250 f Calculated analysis Crude protein, % Lysine, % Ca, % P, % DE, Mcal/kg ME, Mcal/kg a d21tod42inexp.1andexp.3andd21tod41inexp.2. b d0tod42inexp.1andexp.3andd0tod41inexp.2. c Contains 70.00% crude protein, 13.00% ash, 2.00% fat,.14% calcium,.13% phosphorus, 3.01% alanine, 6.36% aspartate, 4.79% arginine, 2.24% cystine, 3.70% phenylalanine, 2.44% glycine, 8.85% glutamate, 2.50% histidine, 1.96% isoleucine, 5.56% leucine, 6.10% lysine,.53% methionine, 4.09% proline, 3.86% serine, 4.13% threonine, 1.33% tryptophan, 3.50% tyrosine and 4.12% valine (American Protein Corporation, Ames, IA). d Provided the following amounts of vitamins per kilogram of complete diet: vitamin A, 6,600 IU; vitamin D 3, 660 IU; vitamin E, 88 IU; vitamin K, 4.4 mg; vitamin B 12,.0352 mg; riboflavin, 8.8 mg; D-pantothenic acid, 24.2 mg; niacin, 33.0 mg; choline chloride, 330 mg (University of Illinois, Urbana, IL). e Provided the following amounts of minerals per kilogram of complete diet: Mn, mg; Fe, mg; Zn, mg; Cu, 8.09 mg; I,.35 mg; Se,.30 mg (University of Illinois, Urbana, IL). f Supplied the following amounts of chemicals per kilogram of complete diet: chlortetracycline, 110 mg; sulfamethazine, 110 mg; penicillin, 55 mg (American Cyanamid Co., Princeton, NJ). to pigs fed high-quality diets in a multiphase starter feeding program immediately postweaning. However, the growth retardation in pigs fed a low-quality diet was compensated when adequate feeding was provided in the growing-finishing periods. The protein gain for the entire production period was not affected (P >.63) by feeding and growth performance during the starter period. Nursery Period Average daily gain of F barrows fed the high-quality diet sequence (FHB) was greater (P <.01) than that of F barrows fed the low-quality diet (FLB) for starter period in both Exp. 1 and Exp. 2. Feed intake by pigs was not affected by feeding program or gender. However, the F group had greater (P <.05) feed intake than the P group during the 6-wk postweaning period in Exp. 2. During the first 7 d, the HB group (136 g/d) and HS group (238 g/d) gained BW although the LB group ( 49 g/d) and LS group ( 24 g/d) lost BW (P <.05) (Figure 1). Pigs from all treatments, however, gained body protein and lost body fat. The H group gained more protein (P <.05) and mobilized less fat (P <.05) than the L group. The body protein:fat ratio of the L group (2.20) was higher (P <.05) at d 7 than that of the H group (1.79). Regardless of dietary treatment, pigs suffered postweaning growth retardation and used body fat as an energy source during the first week after weaning. Growth priority was given to protein deposition. In previous studies, growth rate was reduced immediately after weaning and protein deposition postweaning was not affected by level of feed offered (Whittemore et al., 1981) or environmental temperature (Le Dividich et al., 1980). In our experiment, the protein gains were not affected by level of feed as long as pigs gained some BW, but the protein gains were reduced (P <.05) in the LB and LG groups when pigs failed to maintain positive BW gain. The pigs are able to protect protein deposition even during a period of some BW loss and to increase levels of free fatty acids, indicating mobilization of fat reserves during weaning period. The levels of free fatty acids at d 2 postweaning were 800 to 1,500 mol/l which were similar to those observed in 50-kg BW pigs deprived of feed (Wood et al., 1977). This indicates that lean gain is independent of

4 2888 Whang et al. Table 2. Percentage composition and calculated analysis of experimental diets for grower and finisher pigs in Exp. 1 and Exp. 3 Ingredient Period: 42 to 82 d 82 to 152 d Grower Finisher Corn Dehulled soybean meal Soybean oil 1.50 Lysine-HCl Dicalcium phosphate Calcium carbonate Vitamin premix a Trace-mineral salt b Calculated analysis Crude protein, % Lysine, % Ca, % P, % DE, Mcal/kg ME, Mcal/kg a Provided the following amounts of vitamins per kilogram of complete diet: vitamin A, 6,600 IU; vitamin D 3, 660 IU; vitamin E, 88 IU; vitamin K, 4.4 mg; vitamin B 12,.0352 mg; riboflavin, 8.8 mg; D- pantothenic acid, 24.2 mg; niacin, 33.0 mg; choline chloride, 330 mg (University of Illinois, Urbana, IL). b Provided the following amounts of minerals per kilogram of complete grower diet: Mn, mg; Fe, mg; Zn, mg; Cu, 8.09 mg; I,.35 mg; Se,.30 mg; Provided the following amounts of minerals per kilogram of complete finisher diet: Mn, mg; Fe, mg; Zn, mg; Cu, 6.93 mg; I,.30 mg; Se,.26 mg (University of Illinois, Urbana, IL). fat gain (Kielanowski, 1976; Le Dividich and Noblet, 1982). Fat reserves may be totally depleted and protein catabolism could be started by d 10 or d 12 postweaning if pigs are constantly in negative energy balance (Whittemore et al., 1981). Water gains by the HB group ( g/d) and HG group ( g/d) were higher (P <.01) than those of LB group ( g/d) and LG group ( 4.22 g/d) from d 0 to d 7 postweaning. The water gains of HB and HG groups can explain the maintenance of body weight gains during a period of fat loss. The H group retained extra water as fat losses occurred. Our results suggest that the concept of a constant relationship between body protein and body water may not be valid for animals undergoing fat depletion. Whittemore et al. (1978) has reported that weaning pigs lost body fat and gained more body water than needed for body protein. It was also indicated that percentage of body protein was not affected, but body lipid was markedly affected by level of nutrient intake. The present results also agree with a previous study demonstrating that weanling pigs are able to use body fat as an energy source and to support gains of body protein (Le Dividich et al., 1982). The body protein:fat ratio was dramatically changed from an average 1.32 (d 0) to 1.79 in the H group and 2.20 in the L group (d 7). The weight of viscera was greater in the H group on d 7 and 14 (P <.01) than in the L group (Table 4). There was no effect of gender on viscera weight during 6 wk postweaning, and there was no interaction between feeding program and gender (Table 5). Viscera:empty Table 3. Percentage composition and calculated analysis of experimental diets for grower and finisher pigs in Exp. 2 Sex: Barrows Gilts Period: d41tod85 d85tod151 d41tod85 d85tod151 Ingredient Grower Finisher Grower Finisher Corn Dehulled soybean meal Soybean oil Lysine-HCl Dicalcium phosphate Ground limestone Vitamin premix a Trace-mineral salt b Calculated analysis Crude protein, % Lysine, % Ca, % P, % DE, Mcal/kg ME, Mcal/kg a Provided the following amounts of vitamins per kilogram of complete diet: vitamin A, 6,600 IU; vitamin D 3, 660 IU; vitamin E, 88 IU; vitamin K, 4.4 mg; vitamin B 12,.0352 mg; riboflavin, 8.8 mg; D-pantothenic acid, 24.2 mg; niacin, 33.0 mg; choline chloride, 330 mg (University of Illinois, Urbana, IL). b Provided the following amounts of minerals per kilogram of complete barrow grower and gilt diet: Mn, mg; Fe, mg; Zn, mg; Cu, 8.09 mg; I,.35 mg; Se,.30 mg. Provided the following amounts of minerals per kilogram of complete barrow finisher diet: Mn, mg; Fe, mg; Zn, mg; Cu, 6.93 mg; I,.30 mg; Se,.26 mg (University of Illinois, Urbana, IL).

5 Compensatory growth in swine 2889 Figure 1. Gains of body weight, protein, fat, and water for the first 7 d postweaning in Exp. 3. Different letters on the bars mean significant difference at P <.05. FHB = Barrows fed the high-quality diet sequence during the starter period; FLB = Barrows fed the low-quality diet during the starter period; FHG = Gilts fed the high-quality diet sequence during the starter period; FLG = Gilts fed the low-quality diet during the starter period. body weight ratio was decreased (P <.01) from 11.4% at d 0 to 10.2% at d 3 and then increased (P <.01) to 12.6% at d 7. The highest viscera:empty body weight ratio was 15.9% at d 14 (P <.01). The ratio, however, was not affected by dietary treatment or gender at this stage. The viscera of animals at d 0 had a lower (P <.05) protein:fat ratio than those at d 3. During d 0 to d 3, pigs used visceral fat for energy and had a relatively smaller portion of viscera at d 3. Although pigs lost visceral weight immediately after weaning, they continued to increase the portion of viscera until d 14. This suggests that pigs could accrete the essential body parts (viscera) with higher priority at the expense of nonessential parts (body fat). Our experiment showed that pigs increased (P <.01) viscera as a portion of body weight from d 0 to d 7, although body fat was rapidly diminishing from 11.52% at d 0 to 8.10% at d 7 (P <.01). During the second week (d 7 to 14), pigs from all treatments exhibited positive gains in BW, body protein, and body fat. The H group gained more (P <.05) BW than the L group. The depositions of protein and fat were not different among treatments during this period. From d 14 to 42, the protein gains of HB group (51.09 g/d) and HG group (73.03 g/d) were similar to Table 4. Viscera mass (kilograms) after removal of digesta from gastrointestinal track from weaning to termination (d 152) in Exp. 3 Days postweaning Treatment a HB LB HG LG Diet NS b NS ** b ** NS NS NS Gender NS NS NS NS NS NS NS Diet gender NS NS NS NS NS NS NS a HB = Barrows fed the high-quality diet sequence during the starter period; LB = Barrows fed the lowquality diet during the starter period; HG = Gilts fed the high-quality diet sequence during the starter period; LG = Gilts fed the low-quality diet during the starter period. b NS, not significant; **P <.01

6 2890 Whang et al. Figure 2. Polynomial plot of body weight (mean ± SEM) as a function of days after weaning in Exp. 1. Animals fed the high-quality diet sequence (empty symbols) and animals fed the low-quality diet sequence (filled symbols) during the starter period. Data points with an asterisk are significantly different (***P <.001; **P <.01; *P <.05). a) Duroc-Yorkshire Hampshire barrows (FHB and FLB). b) Duroc-Yorkshire Hampshire gilts (FHG and FLG). those of LB group (53.97 g/d) and LG group (80.42 g/ d). The fat gains were also similar among treatments: HB, g/d; LB, g/d; HG, g/d; LG, g/d. Average BW of the LB and LG groups at d 14 (6.24 kg) and at d 42 (19.92 kg) were lower (P <.01) than those of the HB and HG groups that were 8.65 kg at d 14 and kg at d 42. Considering the lower BW of the L group, the similar average daily gain (H, 517 g/ d; L, 491 g/d), protein gain, and fat gain from d 14 to d 42 indicate that the pigs from the L group used nutrients more efficiently and grew relatively faster than those from the H group. Efficiencies of protein and energy utilization (protein or energy deposited/protein or energy consumed) of previously restricted pigs have been shown to be higher than those of control pigs (Tullis and Whittemore, 1986). Growing-Finishing Period and Compensatory Growth Although the starter feeding program affected growth performance significantly during the starter period, it did not affect the growth rates (P >.63) during the subsequent grower period (approximately 20 to 50 kg BW). The FLB group tended to grow faster (P <.09) than the FHB group during finisher phase. The BW of FHB and FLB groups at termination in Exp. 1 (152 d postweaning) and Exp. 2 (151 d postweaning) were ± 2.76 and ± 3.08, and ± 2.48, and ± 2.60 kg, respectively (P >.81). Although feeding the low-quality starter diet (FLB) resulted in inferior initial growth performance in comparison to feeding the high-quality starter diet sequence, the pigs exhibited compensatory body weight gain during the late grower and finisher phases (Figures 2a and 3a). The terminal BW were ±.02 kg for PHB group and ± 4.78 kg for PLB group (Figure 3c). Average daily gain of gilts fed the high-quality starter diets (FHG) was greater (P <.01) than the gain of gilts fed the low-quality diets (FLG) during the early starter phase in Exp. 1 and Exp. 2. The FLG group did (Exp. 2) and did not (Exp. 1) show compensatory gain during the grower and finisher periods. Body weights of FHG and FLG groups at termination in Exp. 1 and Exp. 2 were ± 1.88 kg and ± 1.62 kg (P <.08), and ± 2.15 and ± 2.58 kg (P >.76), respectively. The growth curves for FHG and FLG groups from weaning to termination are shown in Figures 2b and 3b. In PIC gilts, the LG group did not exceed the growth rate of the HG group for starter or grower phases, but the HG and LG groups had similar ADG for the finisher phases (Figure 3d). The body weights of the HG and LG groups at termination (151 d postweaning) in Exp. 2 were ± 4.00 and ± 1.86 kg, respectively (P >.31). In previous studies, pigs have been shown to exhibit reduced growth rate by lowering crude protein (Wyllie et al., 1969; Zimmerman and Khajarern, 1973; Campbell and Biden, 1978) or feed intake (Mersmann et al., 1987; Stamatarist et al., 1991) during early stages of development. These pigs have shown better subsequent

7 Compensatory growth in swine 2891 growth performance than pigs given adequate protein and permitted unlimited feed intake. In the present experiment, growth in the early starter period (i.e., weaning to d 21 postweaning) was affected by diet quality. The reduced growth rate during the starter period contributed to body weight differences between the H and L groups from early starter phase to early grower phase. This agrees with Meade et al. (1969), who reported that pigs given a complex diet or a high-quality diet were heavier at d 63 postweaning than pigs given a simple diet, but the difference diminished with time. After the immediate reduction in growth rate postweaning, the ADG of the L group was not different from the H group until termination. The FLB group grew more rapidly than the FHB group during the finisher phase. This is direct evidence of compensatory growth. As a result, the live weights of the FHB and FLB groups were not different at termination in Exp. 1 and Exp. 2. This is similar to a previous study (Kornegay et al., 1990) that demonstrated that the growth performance Figure 3. Polynomial plot of body weight (mean ± SEM) as a function of days after weaning in Exp. 2. Animals fed the high-quality diet sequence (empty symbols) and animals fed the low-quality diet sequence (filled symbols) during the starter period. Data points with an asterisk are significantly different (***P <.001; **P <.01; *P <.05). a) Duroc-Yorkshire Hampshire barrows (FHB and FLB). b) Duroc-Yorkshire Hampshire gilts (FHG and FLG). c) PIC barrows (PHB and PLB). d) PIC gilts (PHG and PLG).

8 2892 Whang et al. Table 5. Body content (kilograms) of water, ash, protein, and fat of pigs at termination (d 152) in Exp. 3 Treatment a Water Ash Protein Fat HB LB HG LG Diet NS b NS NS * b Gender * NS * ** b Diet gender NS NS NS NS a HB = Barrows fed the high-quality diet sequence during the starter period; LB = Barrows fed the low-quality diet during the starter period; HG = Gilts fed the high-quality diet sequence during the starter period; LG = Gilts fed the low-quality diet during the starter period. b NS, not significant; **P <.01; *P <.05. was affected by diet quality during the starter period but the effect did not persist for the grower-finisher periods. The FLG group showed contrasting compensatory growth in two experiments. The FLG group of Exp. 1 did not achieve the growth rate of the FHG group for the whole period of the experiment. It is possible that this was a consequence of lower appetite in gilts than in barrows (Blair and English, 1965; Friend and Mac- Intyre, 1970; Skitsko and Bowland, 1970), and the quantitative lysine intake (23.25 g/d) for the LG group during the finisher period may have been insufficient to support compensatory responses. However, this is unlikely given recent work by Hahn (1995), who demonstrated that gilts in our facility require 22 g of lysine per day. In Exp. 2, with diets for gilts containing 1% higher crude protein and.1% higher lysine during grower and finisher periods than diets for barrows, the FLG group regained the BW shortly after the starter period. The response of PIC pigs was quite different. The PLB and PLG groups never achieved the growth rates of the PHB and PHG groups over the whole experimental period. Because of reduced feed intake of PLB and PLG groups during the finisher phase, daily lysine intakes were lower in PLB group (22.93 g/d) than in the PHB group (23.21 g/d) and lower in the PLG group (21.02 g/ d) than in the PHG group (22.03 g/d). The lower lysine intakes of PLB and PLG groups were probably not sufficient to support compensatory growth responses during the finisher phase. This agrees with Stairs et al. (1991), who reported that milk products in a starter diet improved growth performance for the starter period as well as grower and finisher periods but the difference diminished with time. Figure 4. Polynomial plot of body protein and fat (mean SEM) as a function of days after weaning. H-Protein = Protein gain of barrows and gilts fed the high-quality diet sequence during the starter period; L-Protein = Protein gain of barrows and gilts fed the low-quality diet during the starter period; H-Fat = Fat gain of barrows and gilts fed the high-quality diet sequence during the starter period; L-Fat = Fat gain of barrows and gilts fed the low-quality diet during the starter period.

9 Compensatory growth in swine 2893 Figure 5. Percentage composition in whole empty bodies of FHB, FLB, FHG, and FLG at weaning a) and at termination b) of the experiment in Exp. 1. FHB = Duroc-Yorkshire Hampshire barrows fed the high-quality diet sequence during the starter period; FLB = Duroc-Yorkshire Hampshire barrows fed the low-quality diet during the starter period; FHG = Duroc-Yorkshire Hampshire gilts fed the high-quality diet during the starter period. Average daily protein gain of H group ( g/d) and L group ( g/d) during the growing period (d 42 to d 82) was similar, although average daily fat gain of the H group ( g/d) was higher (P <.05) than that of L group ( g/d). The body protein:fat ratios of the H (1.53) and L (1.52) groups at d 42 were similar and tended (P <.06) to be higher for the L group (.865) than for the H (.745) group at d 82. This is clear evidence that protein deposition was given priority over fat deposition. This is similar to results reported by Elsley (1963) and Nielsen (1964), who demonstrated that pigs limited in their growth during the starter period were leaner at 90 kg BW than pigs that grew faster during the starter period. Feed intake was not different either between feeding programs or between genders during growing-finishing period. However, feed intake of the F group was greater (P <.05) than that of the P group during growing period in Exp. 2, and feed intake of barrows was greater (P <.01) than that of gilts during the finishing period in Exp. 2. The visceral portion of the body at d 42 (approximately 20 kg BW) was 15.2%. The highest viscera:empty BW ratio, greater than 15%, was maintained from d 14 to 42 in this study. The ratio was decreased (P <.01) to 13.3% at d 82 (approximately 50 kg BW) and then further decreased (P <.01) to 11.6% at termination (d 152). The weight of viscera was not different between the H and L groups during the growing-finishing period (Table 4). The ratio was not affected by dietary treatment, but was affected by gender at d 42 and 82. The barrows (16.3%) had a higher (P <.01) ratio than the gilts (14.0%) at d 42 and 82 (B, 13.9%; G, 12.6%; P <.05). This is likely due to the fact that barrows had a higher feed intake than gilts during the late growing and finishing periods. During the finishing period (d 82 to 152), the protein gain of the L group ( g/d) exceeded (P <.05) that of the H group ( g/d). The fat gains of the H group ( g/d) and the L group ( g/d) were similar. The L group exhibited compensatory protein gain during late growing and finishing periods and achieved a body protein content similar (P >.60) to the H group by termination (d 152). Unlike protein gain, the fat gain by the L group was not completely compensated and the body fat of the L group tended to be lower than that of the H group. Final body contents of water, ash, protein, and fat are shown in Table 5. The protein gains

10 2894 Whang et al. of barrows ( g/d) and gilts ( g/d) were not different, but barrows ( g/d) deposited more (P <.01) fat than gilts ( g/d) during the finishing period. The growth curves of protein and fat in the H and L groups for the entire experimental period, from weaning to termination, are shown in Figure 4 and are similar to those reported by Shields et al. (1983). In Exp. 3, the initial compositions of the whole empty bodies at weaning were similar regardless of treatments or gender (Figure 5a), but body composition at termination was affected by diet and gender. The FLB and FLG groups tended to have a higher protein but lower fat content than the FHB and FHG groups (Figure 5b). Daily gains of water, protein, fat, and ash from weaning to termination were not affected (P >.25) by starter feeding program. However, the gains of water (P <.08) and protein (P <.05) of barrows were less than those of gilts. Fat gain in barrows was higher (P <.08) than in gilts (Figure 6). In Exp. 2, the starter feeding program did not affect lean body mass at termination within gender or genotype. Although the gilts (51.66% protein) were leaner (P <.01) than the barrows (48.22% protein), the total lean mass was not different (barrows, kg; gilts, kg) at termination because the overall ADG of the barrows (709 g/d) was higher (P <.01) than that of the gilts (664 g/d). The P pigs (50.80%) were leaner (P <.05) than the F pigs (49.07%), and the P pigs (41.43 kg) had more (P <.01) lean body mass than the F pigs (38.72 kg). The ADG of the P (684 g/ d) and F (686 g/d) groups were similar from weaning to termination. Classical studies by Elsley (1963) and Nielsen (1964) demonstrated that pigs limited in their growth during the starter period were leaner at 90 kg BW than pigs that grew faster during the starter period. In addition, Campbell and Biden (1978) found that pigs fed a lowprotein diet between 5.5 and 20 kg BW were not different in carcass and ham composition at 70 kg BW from pigs fed more protein during the starter period. At slaughter in Exp. 2, our L pigs tended to be leaner than the H pigs. Total lean body mass and the gains of body components, water, protein, fat, and ash were not affected by starter feeding program. It could be interpreted that lean gain of L group was not impaired by reduced growth during the early growth period after weaning and that lean gain is independent of the fat gain (Kielanowski, 1976; Le Dividich and Noblet, 1982). The data also indicate that fat gain of the L group did not exceed that of the H group during the subsequent period. It might be that lean gain has priority over fat gain during compensatory growth. Figure 6. Average daily gains of water, fat, protein, and ash from weaning to termination of the experiment (d 152 postweaning) by starter feeding program. Means with different superscripts are significantly different (P <.05). FHB = Duroc-Yorkshire Hampshire barrows fed the high-quality diet sequence during the starter period; FLB = Duroc- Yorkshire Hampshire barrows fed the low-quality diet during the starter period; FHG = Duroc-Yorkshire Hampshire gilts fed the high-quality diet sequence during the starter period; FLG = Duroc-Yorkshire Hampshire gilts fed the low-quality diet during the starter period.

11 Compensatory growth in swine 2895 Implications The simple cereal-soy-based starter diet supported inferior growth performance immediately after weaning. However, the growth retardation was compensated when adequate feeding was provided in the growingfinishing periods. This experiment also demonstrated that the lean gain for the entire production period was not affected by feeding and growth performance during the starter period. Therefore, a high-quality multiphase starter feeding program may not be necessary for highlean muscle production as long as an adequate growingfinishing feeding program is applied. Literature Cited AOAC Official Methods of Analysis. 15th ed. Association of Official Analytical Chemists, Arlington, VA. Blair, R., and P. R. English The effect of sex on growth and carcass quality in the bacon pig. J. Agric. Sci. 64: Cameron, N. D., G. R. Nute, S. N. Brown, M. Enser, and J. D. Wood Meat quality of Large White pig genotypes selected for components of efficient lean growth rate. Anim. Sci. 68: Campbell, R. G., and R. S. Biden The effect of protein nutrition between 5.5 and 20 kg live weight on the subsequent performance and carcass quality of pigs. Anim. Prod. 27: Campbell, R. G., and A. C. Dunkin The influence of nutrition in early life on growth and development of the pig: 3. Effects of energy intake prior and subsequent to 10 kg on growth and development to 30 kg live weight. Anim. Prod. 36: Campbell, R. G., M. R. Taverner, and D. M. Curic The influence of feeding level from 20 to 45 kg live weight on the performance and body composition of female and entire male pigs. Anim. Prod. 36: Elsley, F. W. H Studies of growth and development in the young pig. II. A comparison of the performance to 200 lb of pigs reared along different growth curves to 56 days of age. J. Agric. Sci. 61: Ewan, R. C Voluntary feed intake of swine. In: Proc. Cornell Nutr. Conf Feed Manuf. Ithaca, New York. pp Friend, D. W., and T. M. MacIntyre Paired feeding and metabolism trials comparing barrows with gilts. J. Anim. Sci. 30: Gu, Y., A. P. Schinckel, J. C. Forrest, C. H. Kuei, and L. E. Watkins Effects of ractopamine, genotype, and growth phase on finishing performance and carcass value in swine. II. Estimation of lean growth rate and lean feed efficiency. J. Anim. Sci. 69: Hahn, J. D., R. R. Biehl, and D. H. Baker Ideal digestible lysine level for early- and late-finishing swine. J. Anim. Sci. 73: Kielanowski, J The chemical composition of the live-weight gain and the performance of growing pigs. Livest. Prod. Sci. 3: Kornegay, E. T., J. W. G. M. Swinkels, G. Ball, and C. M. Wood Comparison of three nutritional feeding regimens for weanling pigs and subsequent performance to market weight. J. Anim. Sci. 68(Suppl. 1):378 (Abstr.). Le Dividich, J., and J. Noblet Growth rate and protein and fat gain in early-weaned piglets housed below thermoneutrality. Livest. Prod. Sci. 9: Le Dividich, J., M. Vermorel, J. Noblet, J. C. Bouvier, and A. Aumaitre Effects of environmental temperature on heat production, energy retention, protein and fat gain in early weaned piglets. Br. J. Nutr. 44: Leibbrandt, V. D., R. C. Ewan, V. C. Speer, and D. R. Zimmerman Effect of age and calorie:protein ratio on performance and body composition of baby pigs. J. Anim. Sci. 40: Meade, R. J., J. W. Rust, K. P. Miller, H. E. Hanke, R. S. Grant, L. D. Vermedahl, D. F. Wass, and L. E. Hanson Effects of protein level sequence and kind of starter on rate and efficiency of gain of growing swine, and on carcass characteristics. J. Anim. Sci. 29: Mersmann, H. J., M. D. MacNeil, S. C. Seideman, and W. G. Pond Compensatory growth in finishing pigs after feed restriction. J. Anim. Sci. 64: Newcomb, M. D., R. S. Ott, T. van Kempen, Y. H. Lan, F. K. McKeith, J. E. Novakofski, P. J. Bechtel, and R. A. Easter Effect of hyperalimentation on body composition in swine. J. Anim. Sci. 71: Nielsen, H. E Effects in bacon pigs of differing levels of nutrition to 20 kg body weight. Anim. Prod. 6: NPPC Procedures to Evaluate Market Hogs. 3rd ed. National Pork Producers Council, Des Moines, IA. SAS SAS/STAT User s Guide (Version 6, 4th Ed.). SAS Institute Inc., Cary, NC. Shields, R. G., Jr., D. C. Mahan, and P. L. Graham Changes in swine body composition from birth to 145 kg. J. Anim. Sci. 57: Skitsko, P. J., and J. P. Bowland Performance of gilts and barrows from three breeding groups marketed at three liveweights when offered diets containing two levels of digestible energy for a limited period per day. J. Anim. Sci. 50: Stamatarist, C., I. Kyriazakis, and G. C. Emmans The performance and body composition of young pigs following a period of growth retardation by food restriction. Anim. Prod. 53: Stairs, J. T. F., M. D. Tokach, J. E. Pettigrew, and M. E. Wilson Milk products in starter diets improve subsequent pig performance. J. Anim. Sci. 69(Suppl. 1):116 (Abstr.). Tullis, J. B., and C. T. Whittemore Body composition and feed intake of young pigs postweaning. J. Sci. Food Agric. 37: Whittemore, C. T., A. Aumaitre, and I. H. Williams Growth of body components in young weaned pigs. J. Agric. Sci. 91: Whittemore, C. T., H. M. Taylor, R. Henderson, J. D. Wood, and D. C. Brock Chemical and dissected composition changes in weaned piglets. Anim. Prod. 32: Wood, J. D., N. G. Gregory, G. M. Hall, and D. Lister Fat mobilization in Pietrain and Large White pigs. Br. J. Nutr. 37: Wyllie, D., V. C. Speer, R. C. Ewan, and V. W. Hays Effect of starter protein level on performance and body composition of pigs. J. Anim. Sci. 29: Yen, J. T., J. A. Nienaber, J. Klindt, and J. D. Crouse Effect of ractopamine on growth, carcass traits, and fasting heat production of U.S. contemporary crossbred and Chinese Meishan pure- and crossbred pigs. J. Anim. Sci. 69: Zimmerman, D. R., and S. Khajarern Starter protein nutrition and compensatory responses in swine. J. Anim. Sci. 36: