Effects of lactation feed intake and creep feeding on sow and piglet performance

A total of 84 sows (PIC, Line 1050) and their litters were used to determine the effects of lactation and creep feeding on sow and piglet performance. Three groups of sows were blocked according to day of farrowing and parity and allotted to four treatments in a 2 Ã— 2 factorial with lactation feed intake (ad libitum vs. restricted) and creep feeding (none vs. creep) as factors. Piglets were cross-fostered within each block to standardize litter weights and litter size (>11 pigs). A common lactation diet (1,586 kcal ME/lb, 0.97% TID Lys) was used in the study. From d 3 of lactation, ad libitum sows were allowed free access to feed while restricted sows were fed 25% less than those fed ad libitum. A pelleted creep diet (1,585 ME/lb, 1.56% TID Lys) with 1.0% chromium oxide was offered to creep-fed pigs from d 3 to weaning (d 21). Piglets were weighed individually at d 3, 7, 14, and 21. Amount of creep feed consumed was determined daily. Fecal samples from all creep-fed pigs were taken on d 7, 14, and 21 and fecal color was assessed to categorize pigs as eaters or non-eaters. Sow weight and P2 backfat thickness (6.5 cm from the midline over the last rib) were measured after farrowing and at weaning. There was no interaction between lactation feed intake and creep feeding. Ad libitum feeding of sows reduced BW loss (-33.0 vs. -52.9 lb; P<0.01), improved total (P<0.04) and daily (P<0.04) gains of litters, and increased (90 vs. 71%; P<0.03) the percentage of sows returning to estrus by d 14 compared with limit-fed sows. Creep feeding did not affect (P>0.30) sow BW and backfat loss, but increased days to estrus (5.4 vs. 4.9 d; P<0.03) for sows that returned to heat by 14 d. Creep feeding tended to improve litter weaning weights (132.7 vs. 124.9 lb/d; P<0.09) by reducing mortality rate after cross-fostering (3.9 vs. 7.3%; P<0.06). Total creep feed intake of litters did not differ (2.24 vs. 2.28 lb/litter; P<0.93) between ad libitum and limit-fed sows. About 60% of the creep-fed pigs were categorized as eaters. Of those identified as eaters, 23, 20, and 57% began consuming creep feeding from d 3 to 7, 7 to 14, and 14 to 21, respectively. From d 0 to 28 post-weaning, there was no effect of creep feeding on d 28 weights (P<0.93), ADG (P<0.86), ADFI (P<0.93), and F/G (P<0.95) compared to non-creep fed pigs. Eaters tended to be heavier until d 28 postweaning (P<0.16) and had greater (P<0.06) ADG and total gains than non-eaters and no creep pigs. In conclusion, creep feeding improved survivability, but had no effects on pre-weaning gain and sow performance. Low feed intake during lactation negatively affected both sow and litter performance. Creating more eaters in whole litters may be beneficial in improving post-weaning performance.; Swine Day, 2007, Kansas State University, Manhattan, KS, 2007


INTRODUCTION
Preweaning growth is a major determinant of postweaning pig performance (Klindt, 2003), but the biological growth potential of suckling piglets is largely unattained in commercial production (Pluske et al., 1995;Le Dividich and Sève, 2001).As evidenced in artificial rearing studies, greater preweaning growth rates in piglets can be achieved through increased nutrient availability (Hodge, 1974;Harrell et al., 1993).This can be provided by improving the milk output of the sow with high-density lactation diets (Shurson et al., 1986;Yang et al., 2000) or maximizing lactation feed intake (Koketsu et al., 1996;Eissen et al., 2003).Providing a solid, highly digestible creep feed to suckling piglets may also augment milk production especially during late lactation (Lucas and Lodge, 1961;Elsley, 1971), and as a supplemental nutrient source, creep feed may increase nutrient intake of the piglets and improve preweaning daily BW gains and weaning weights (Mavromichalis, 2006).
Effects of lactation feed intake by sows and creep feeding for piglets on preweaning and postweaning performance have been evaluated independently in previous studies (Appleby et al., 1991;Pajor et al., 1991), but no work has been done on the potential interactive effects of these 2 nutritional regimens.In addition, it has been suggested that creep feeding can reduce the nutritional load in lactating sows, especially those nursing large litters, which may reduce lactation BW loss and the weaning-to-estrus interval (Pajor et al., 2002).At present, no studies support this claim.
Therefore, the objectives of this experiment were to evaluate the effect of lactation feed intake and creep feeding on preweaning and postweaning performance and determine the effect of creep feeding on BW loss, backfat thickness, and weaning-to-estrus interval in sows.

MATERIALS AND METHODS
All animal procedures were reviewed and approved by the Kansas State University Institutional Animal Care and Use Committee.

Animals and Treatments
A total of 84 sows (PIC Line 1050, Hendersonville, TN) and their litters were used.The study was conducted at the Kansas State University Swine Research and Teaching Center farrowing facilities.Sows originated from 3 batches of litters farrowed in August, October, and November 2006; 28 experimental sows were included from each batch.Sows were blocked according to parity and date of farrowing and allotted to 4 experimental treatments in a randomized complete block design with treatment factors arranged in a 2 × 2 factorial arrangement.Treatment factors were lactation feed intake [feed offered for ad libitum intake (ad libitumfed) vs. restricted feed intake (restricted-fed)] and creep feeding (no vs. yes).Piglets were cross-fostered within each block to standardize litter weights and litter size (>11 pigs).Cross-fostering was completed within 48 h after farrowing.The sow and litter was the experimental unit; there were 21 replicates per treatment.

Diets and Management
A common lactation diet (3,503 kcal of ME/kg, 0.97% standardized ileal digestible Lys) was fed to sows in the study (Table 1).From d 3 of lactation and on, ad libitum-fed sows were allowed free access to feed, whereas the feeding schedule of restricted-fed sows aimed to provide 25% less feed.The restricted feeding level was estimated by using historical feed consumption data of the farm.In the creep feed treatments, a creep diet (3,495 kcal of ME/kg, 1.56% standardized ileal digestible Lys; Table 1) with 1.0% chromium oxide was offered ad libitum from d 3 until weaning (d 21) in a rotary creep feeder (Rotecna Mini Hopper Pan, Rotecna SA, Agramunt, Spain).This feeder has a 6-L capacity hopper, which is adjustable to 5 different settings to allow ad libitum consumption.The creep diet was in pellet form (2-mm pellets).Sufficient amounts of creep feed were placed in the hopper to ensure that feed was always available.The feeder setting was checked daily to ensure ad libitum access and minimize feed wastage.
The creep feeder was placed in the middle of the side of the farrowing crate opposite to the heat lamp.
Piglets were weighed individually at d 3, 7, 14, and 21 (weaning).Amount of creep feed consumed was determined daily.Daily creep feed consumption per litter was computed as the difference in feeder weights between consecutive days.Each individual pig in creepfed litters were subjectively tested for creep feed consumption using procedures adapted from Barnett et al. (1989) and Bruininx et al. (2002).Fecal samples from creep-fed pigs were taken with sterile swabs once per sampling day on d 7, 14, and 21.The cotton-tipped Sow add pack (Nutra Blend LLC, Neosho, MO) provided the following nutrients per kilogram of complete diet: 22 IU of vitamin E; 0.22 mg of biotin; 1.65 mg of folic acid; 5 mg of pyridoxine (as pyridoxine HCl); 551 mg of choline (as choline Cl); 50 mg of l-carnitine; 0.20 mg of chromium (as chromium picolinate).
swab was inserted in a clockwise motion into the anal opening of the piglet for about 5 cm and was pulled slowly until fecal matter was collected.Fecal color was assessed to categorize piglets as eaters or non-eaters of creep feed.Piglets were categorized as eaters when the fecal sample was colored green at least once on any of the 3 sampling days.Piglets were categorized as non-eaters when all samples collected were negative for green-colored feces.Pigs that were not provided with creep feed were designated as no-creep pigs.
To determine the effects of creep feeding on postweaning pig performance, 624 pigs of the 819 pigs weaned from the 3 farrowing groups were randomly selected and used in 3 nursery trials.These pigs were randomly selected to have the same number of pigs per pen and pens per treatment to collect ADFI and G:F data.These pigs were blocked according to initial BW and creep feed access (with or without creep feed during lactation) and allotted to their respective treatments.All pigs and feeders were weighed weekly until d 28 postweaning to calculate ADG, ADFI, and G:F.The extra 195 pigs were kept in an adjacent room in the same nursery facility and fed the same diets as the 624 pigs, but ADFI and G:F data were not collected.All 819 pigs were used to compare the growth characteristics of eaters vs. non-eaters of pigs offered creep feed with the growth of pigs not offered creep feed.
Weekly feed intake of the sows was recorded to calculate total and ADFI.Sows were weighed, and P2 backfat thickness (6.5 cm from the midline over the last rib) was measured postfarrowing and at weaning.Estrus detection was performed by using the back pressure test twice a day from weaning until 14 d after weaning to determine days to estrus and the percentage of sows returning to estrus within 14 d.In this study, 6 sows were removed from the test because of poor daily feed intake or death of the sow.General health of piglets was checked daily, and use of medication was monitored.Temperature in the farrowing facility was maintained at a minimum of 20°C, and supplementary heat was provided to piglets with heat lamps when needed.
Periodic and cumulative ADG and creep feed intake were calculated for each treatment group.Preweaning mortality was also calculated.The CV for pig BW within each litter was determined at d 3 and 21.Sow BW loss, change in P2 backfat thickness, and weaningto-estrus interval were calculated.

Statistical Analyses
Preweaning data were analyzed as a randomized complete block design with a 2 × 2 factorial treatment structure with sow and litter as the experimental unit.The model included main effects of lactation feed intake and creep access, the interaction of lactation feed intake × creep access as a fixed effect, and block as the random effect.An ANOVA was performed by using PROC MIXED (SAS Inst.Inc., Cary, NC).Least squares means were calculated for each independent variable.Regression models for daily litter creep feed intake were developed by using PROC REG of SAS.Logistic regression curves were also developed by using PROC LOGISTIC of SAS to determine estimated probabilities of changes in the proportion of eaters as determined by BW on d 3 (start of creep feeding) and d 21 (weaning).Chi-squared analysis was used to determine the effect of lactation feed intake on the proportion of pigs consuming creep feed.
For postweaning data, the effects of creep feeding were analyzed by using PROC MIXED of SAS.The model included treatment and block as the fixed and random effects, respectively, with pen as the experimental unit.For the effect of creep feed consumption category, data were analyzed by using PROC MIXED of SAS with pig as the experimental unit.The model included consumption category as the fixed effect and litter and weaning group as random effects.The homogeneity of variances was tested using the HOVTEST option of SAS.Results showed equal (P > 0.23) variances for all of the response variables measured.When treatment effect was a significant (P ≤ 0.05) source of variation, differences were determined by using the preplanned, pairwise comparisons (PDIFF option of SAS).Least squares means were calculated for each independent variable.Statistical significance and tendencies were set at P ≤ 0.05 and P < 0.10 for all statistical tests.

Effects on Lactating Sow Performance
Effects of lactation feeding level and creep feeding on sow performance are shown in Table 2. Sows had an average parity of 1.6 ± 0.7 and lactation length of 21.1 ± 1.9 d (Table 2).There were no significant interactions (P > 0.15) between lactation feeding level and creep feeding for any of the performance variables measured; therefore, only main effects are presented.
Ad libitum-fed sows had 32 and 26% greater total (P = 0.0001) and daily (P = 0.0001) feed intake, respectively, than restricted-fed sows.Daily energy intakes were 12.6 and 17.2 Mcal of ME for the restricted-and ad libitum-fed sows, respectively.There were no differences in postfarrowing (P = 0.37) and weaning (P = 0.23) BW of ad libitum-and restricted-fed sows, but ad libitum-fed sows had less lactation BW loss (−15 vs. −24 kg; P = 0.01) compared with restricted-fed sows.Backfat thickness after farrowing (P = 0.84) and at weaning (P = 0.44) was also similar; likewise, backfat loss throughout lactation did not differ (P = 0.27) between ad libitum-and restricted-fed sows.Numerous studies have evaluated the effects of energy intake during lactation on sow performance (Reese et al., 1982(Reese et al., , 1984;;Verstegen et al., 1985;Armstrong et al., 1986;Brendemuhl et al., 1989;Eissen et al., 2003).These studies compared energy intakes between greater and lesser levels ranging from 14.0 to 18.0 Mcal of ME/d and 6.5 to 10.4 Mcal of ME/d, respectively.Overall, less energy intake during lactation resulted in significant increases in sow BW loss and backfat loss, which is an indication of greater rates of body tissue mobilization in restricted-fed sows (Close and Cole, 2000).Greater lactation BW loss in restricted-fed sows was also observed in the current study; however, backfat loss was unaffected by lactation feeding level.van den Brand et al. (2000) had similar observations, which may be related to the degree of energy restriction.In van den Brand et al. (2000) and the current study, restricted energy intakes were 73 to 75% of ad libitum-fed sows, which is a smaller difference compared with other studies (40 to 60% of ad libitum-fed sows; Reese et al., 1982Reese et al., , 1984;;Verstegen et al., 1985;Armstrong et al., 1986;Brendemuhl et al., 1989;Eissen et al., 2003).
Days to estrus for sows that returned to estrus by d 14 after weaning were similar (P = 0.83) between ad libitum-and restricted-fed sows.Ad libitum-fed sows had greater (P = 0.03) percentage of sows returning to estrus by d 14, which agrees with Reese et al. (1982Reese et al. ( , 1984)).The greater rate of sows failing to exhibit estrus within 14 d demonstrates the detrimental effects of limit-feeding sows during lactation.Decreased feed intake during lactation has previously been shown to depress luteinizing hormone secretion, which is required for the resumption of estrus postweaning (Barb et al., 1991;Tokach et al., 1992).
Providing litters with creep feed did not affect total (P = 0.56) and daily (P = 0.57) feed intake of sows.Likewise, there were no differences (P > 0.27) in sow BW after farrowing or at weaning or in lactation BW loss between sows with litters provided and not provided with creep feed.The same effect (P > 0.14) was observed for backfat thickness after farrowing and at weaning and backfat loss.Creep feeding increased days to estrus (5.4 vs. 4.9 d; P = 0.03) for sows that returned to heat by 14 d.There were no differences (P = 0.77) in the proportion of sows that returned to estrus within 14 d between sows with creep-and non-creep-fed litters.No previous study has evaluated the effects of creep feeding on sow performance, although there have been commercial claims of some potential benefits of the practice (e.g., Van Enckevort, 2001).Creep feeding was thought to increase energy economy in lactating sows, especially those with large litters, which may have corollary effects in reducing lactation BW loss and weaning-to-estrus interval (Pajor et al., 2002).These effects were not observed in this study; creep feeding for 18 d did not have any effect on sow performance except for increasing days to estrus.The amount of litter creep feed intake observed in this study was too small (3.65 Mcal; 1.26% of total energy intake of the sows) to generate any appreciable, nutritional savings to lactating sows that may merit a reduction in mobilized body reserves or improve metabolic state of sows.Close and Cole (2000)  Three groups of sows (total = 84, PIC Line 1050, Hendersonville, TN) were blocked according to day of farrowing and parity and allotted to 4 treatments arranged in a 2 × 2 factorial with lactation feeding level (restricted fed vs. ad libitum fed) and creep feeding (no vs. yes) as factors.A total of 6 sows were removed from the test because of poor daily feed intake or death.
2 There was no significant interaction (P > 0.10) between lactation feeding level and creep feeding on any variable measured; least squares means of main effects are reported.

3
Restricted-fed sows were fed 25% less than sows fed for ad libitum intake.

4
Creep feed with 1.0% chromium oxide was offered for ad libitum intake from d 3 to weaning (21 ± 0.1 d).For sows returning to estrus within 14 d postweaning.
6 Percentage of sows returning to estrus within 14 d postweaning.
feeding was not expected, and more work needs to be performed to confirm this observation.The effect of creep feeding on sow performance may be greater with older weaning ages (Aumaître and Salmon-Legagneur, 1961;Aherne et al., 1982) but does not appear to be beneficial in a 21-d lactation period.

Effects on Preweaning Pig Performance
Effects of lactation feeding level and creep feeding on pig and litter performance are shown in Table 3. Lactation feeding level had no effect (P > 0.76) on litter size at weaning or preweaning mortality rate, which agrees with previous studies (Reese et al., 1984;Nelssen et al., 1985;Armstrong et al., 1986).Ad libitum-fed sows had greater total (P = 0.04) and daily (P = 0.04) BW gains of litters and tended to have greater litter weaning weights (P = 0.10) compared with restricted-fed sows.Likewise, total BW gain (P = 0.04), daily BW gain (P = 0.03), and weaning BW (P = 0.06) of individual pigs were greater in ad libitum-fed than restricted-fed sows.These results agree with other studies that demonstrated the benefits of ad libitum feed intake during lactation on litter growth rates (Mullan and Williams, 1989;Eissen et al., 2003).Nelssen et al. (1985) also observed a linear increase in average pig and litter weights as daily energy intake of lactating sows increased from 10, 12, and 14 Mcal of ME/d for a 28-d lactation period.Numerous studies have shown increased milk production with increased lactation feeding level (Verstegen et al., 1985;van den Brand et al., 2000), which may explain the improvements in litter growth rates.The CV in litters of ad libitum-fed and restricted-fed sows were similar at weaning (P = 0.22); likewise, there were no differences (P = 0.78) in litter CV change between the 2 levels of lactation feeding.This suggests that lactation feeding level had no effects on litter uniformity.
Litter size at weaning in creep-fed litters was greater by 0.4 pig per litter, but this difference was not significant (P > 0.19).Total and daily BW gains of litters were also unaffected (P > 0.16) by creep feeding, but litter weaning weights tended to be greater (P = 0.09) in creep-fed litters.This difference in litter weights at weaning was mainly due to a tendency for decreased (3.9 vs. 7.3%; P = 0.06) preweaning mortality after cross-fostering in creep-fed litters.However, preweaning mortality rates may be confounded by greater (P = 0.04) litter weights for the creep-fed litters at the start of creep feeding (d 3).There were no differences (P > 0.53) between creep-fed and non-creep-fed litters in total BW gains, daily BW gains, and pig BW at weaning.There were no differences in litter CV at weaning (P = 0.25) and CV change throughout lactation (P = 0.49), which indicates a lack of effect of creep feeding in improving litter uniformity.Litters of restricted-fed sows had greater (54.4 vs. 40.8g; P = 0.02) creep feed intake than litters of ad libitum-fed sows from d 3 to 7 (Figure 1), but no differences (P > 0.41) in litter creep feed intake were observed in other periods.Overall, total creep feed intake was highly variable between litters, ranging from 263 to 2,349 g per litter over the 18-d period that creep feed was provided.Previous studies also observed greater variability in creep feed intake during lactation between and within litters (Okai et al., 1976;Barnett et al., 1989;Pajor et al., 1991;Bruininx et al., 2002).Factors that influence the large variation in voluntary creep feed intake between and within litters during lactation are unknown.However, it is speculated that creep diet composition, creep access, and feeder design may contribute to this variability in litter creep feed intake.Creep feed wastage may also be an important source of the variation because there have been differences in how creep feed was presented and measured across studies.
Total creep feed intake of litters did not differ (1,019 vs. 1,034 g per litter; P = 0.93) between ad libitum-fed and restricted-fed sows, which suggests that a limited nutrient supply to both sows and litters did not drive piglets to consume more creep feed.About 72 and 77% of the total creep feed intake of litters of restricted-fed and ad libitum-fed sows was consumed in the last week before weaning.Daily creep feed intake of litters increased quadratically (R 2 = 0.22; P = 0.0001) from d 3 to weaning, but intakes greater than 50 g per litter were attained only at d 13 of lactation (Figure 2).For creepfed piglets, 59 (n = 254) and 41% (n = 173) were categorized as eaters and non-eaters, respectively.Of pigs identified as eaters, 23, 20, and 57% were positive for creep feed consumption on d 7, 14, and 21, respectively.
The greater intake and percentage of eaters in the last week before weaning indicate that piglets more readily accepted and consumed greater amounts of creep feed at an older age.This is plausible because 60 to 80% of the total creep feed intake is usually consumed in the last week before weaning regardless if pigs are weaned at 3 wk (Sulabo et al., 2008) or 4 wk of age (Fraser et al., 1994;Bruininx et al., 2002;Pluske et al., 2007).Thus, individual creep feed consumption seems to be more related to maturity of the piglets rather than age of induction of creep feeding.
Lactation feeding level did not influence (P > 0.78) the proportion of eaters within whole litters (57 vs. 62% for litters of restricted-fed and ad libitum-fed sows, respectively).Kuller et al. (2007a) demonstrated that intermittent suckling (IS), a management technique in which piglets are separated from the sow during several hours every day in the second half of lactation, increased creep feed intake before weaning.However, IS did not increase the percentage of eaters within a litter (19 and 23% in the control and IS litters, respectively).This suggests that IS increased creep feed intake of piglets that were already eating before the period of separation, instead of increasing the number of pigs consuming creep feed within a litter.These findings agree with the results of the current study, in which limited sow milk supply (as a result of restricted feeding of lactating sows) did not stimulate more individual piglets to eat creep feed when it was offered.Therefore, it is suggested that increased nutritional demand or physiological need had no effect on individual creep feed consumption.
It is suggested that smaller pigs might benefit more from creep feeding and have greater tendencies of becoming an eater; thus, there may be some potential BW relationships.For example, piglets that nurse from posterior teats that produce less milk are smaller than those nursing from anterior teats with greater milk production (Pluske and Williams, 1996;Kim et al., 2000).The poorer ability of smaller pigs to compete at the udder and extract milk may predispose these pigs to consume more creep feed when it is offered.However,  logistic regression analyses (data not shown) showed no relationship between the BW of pigs at the start of creep feeding (d 3 of lactation; P = 0.17) and weaning weights (P = 0.94) with the probability of becoming an eater.This suggests that individual creep feed consumption is not related to pig BW.

Postweaning Pig Performance
From d 0 to 28 postweaning, creep feeding did not (P > 0.86) affect ADG, ADFI, G:F, and d-28 BW compared with non-creep-fed pigs (Table 4).Postweaning effects of creep feeding have been inconsistent across studies, with most showing no effects on growth performance (Okai et al., 1976;Aherne et al.,1982;Barnett et al., 1989;Pajor et al., 1991;Appleby et al., 1992;Fraser et al., 1994;Lawlor et al., 2002), whereas others observed positive effects on postweaning BW gains (English et al., 1980;Fraser et al., 1994;Bruininx et al., 2002;Kuller et al., 2004;Berkeveld et al., 2007).Pajor et al. (1991) suggested that the lack of any clear relationship between creep feed intake and postweaning BW gain could be attributed to differences in the absolute amount of creep feed consumed before weaning.Carstensen et al. (2005) showed that piglets with less creep feed contact during the suckling period ate less feed the first 2 d after weaning, compared with piglets with greater creep feed contact and non-creep-fed piglets, respectively.Kuller et al. (2007a) also found a strong positive relationship between litter creep feed intake and feed intake in the first week after weaning and litter ADG during the first week after weaning.
When all individual pigs weaned (including extra pigs) were evaluated and categorized on the basis of creep feed consumption category, pigs categorized as eaters tended to be heavier at d 21 postweaning (Figure 3; P = 0.05) and had greater (P < 0.05) ADG and total BW gains than non-eaters and no-creep pigs (Figure 4).These differences in postweaning BW gain agree with previous studies in which eaters, non-eaters, and noncreep-fed pigs were compared (Bruininx et al., 2002(Bruininx et al., , 2004;;Kuller et al., 2004Kuller et al., , 2007a;;Kim et al., 2005;Pluske et al., 2007).The differences in postweaning BW gains is mainly due to differences in postweaning feed intake between eaters and non-eaters, which has been fairly consistent (Fraser et al., 1994;Delumeau and Meunier-Salaün, 1995;Bruininx et al., 2002;Carstensen et al., 2005), although a few studies did not observe the same findings (Hedemann et al., 2007).Bruininx et al. (2001) attributed these differences in postweaning growth efficiency to shorter latency time (interval between weaning and first feed intake) and greater postweaning feed intake in eaters.Using segment perfusion tests, Kuller et al. (2007b) also showed greater net absorption in the small intestine of eaters compared with non-eaters, though some studies have reported no effect of preweaning eating activity on gut morphology (Bruininx et al., 2004;Hedemann et al., 2007).
Majority of the early studies that evaluated creep feeding used whole-litter values in determining creep feed consumption, where average intake was estimated by dividing the total creep feed disappearance by the number of piglets in the litter (Pajor et al., 1991;Fraser et al., 1994;Pluske et al., 2007).This relies on the assumption that all piglets in a litter consume exactly the same amount of feed.However, recent studies have demonstrated that there is a wide variation in creep feed consumption within litters and certain proportion A total of 624 of the 819 pigs that were weaned from the 3 farrowing groups were randomly selected and used in the nursery studies.These pigs were blocked according to BW and creep-feed access (no = without creep feed; yes = offered creep feed during lactation) and were randomly selected to have the same number of pigs per pen and pens per treatment.Values are means of 52 pens of 6 pigs each. of pigs do not consume creep feed even when it is very accessible or offered for longer durations (Barnett et al., 1989;Bruininx et al., 2002;Sulabo et al., 2008).As with the current study, it is suggested that the lack of differences observed between creep-fed and non-creepfed pigs on postweaning performance may be because pigs were penned according to creep feed access, not individual creep feed consumption.To further investigate postweaning effects of creep feed intake, it seems necessary to know whether individual pigs actually consumed creep feed during lactation.Therefore, determining individual consumption characteristics of creep fed pigs is important to obtain the true response attributable to creep feed consumption.Decreased feed in-take during lactation negatively affected sow and litter performance.Creep feeding for 18 d had no effects on preweaning and lactating sow performance.When pigs were categorized on the basis of creep feed consumption category, eaters had greater postweaning gains and BW than non-eaters and non-creep-fed pigs.Creating more eaters in whole litters may be beneficial in improving postweaning performance.Therefore, dietary and nondietary factors that can enhance the proportion of eaters in litters should be investigated. 5

Figure 1 .
Figure 1.Effects of lactation feeding level [sow intake was restricted (restricted fed) or offered for ad libitum intake (ad libitum fed)] on litter creep feed intake (percent of total litter creep feed intake in parentheses).

Figure 2 .
Figure 2. Daily litter creep feed intake during lactation (d 3 to weaning; 39 litters).Color version available in the online PDF.

Figure 3 .
Figure 3. Postweaning BW trends of 819 piglets according to creep feed consumption category [415 were not offered creep (no-creep); 404 were offered creep for ad libitum intake, of which 243 readily consumed creep (eaters) and 161 did not (non-eaters)].

Figure 4 .
Figure 4. Overall postweaning ADG and total BW gain (d 0 to 28) of 819 piglets according to creep feed consumption category [415 were not offered creep (no-creep); 404 were offered creep for ad libitum intake, of which 243 readily consumed creep (eaters) and 161 did not (non-eaters)].Color version available in the online PDF.

Table 1 .
Composition of the creep and lactation diet (as-fed basis) 6

Table 2 .
calculated that each kilogram of lactation BW loss contributes about 2.87 Mcal of ME to the energy economy of the sow; therefore, creep feeding provided energy equivalent to only 1.27 kg of sow BW loss.The increase in days to estrus observed with creep Effects of lactation feeding level and creep feeding on sow performance (main effects)1,2

Table 3 .
Effects of lactation feeding level and creep feeding on litter and pig performance (main effects)1,2

Table 4 .
Effects of creep feeding on postweaning growth performance of pigs 1