| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 7, 3584 CL Utrecht, the Netherlands;
and
Adaptation Physiology Group, Wageningen University, PO Box 338, 6700 AH Wageningen, the Netherlands
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: creep feed • feed intake • growth • performance • pig • weaning
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) and piglets are more vulnerable to developing diarrhea and edema disease (van Beers-Schreurs et al., 1992). Intake of a suffi-cient amount of creep feed during lactation creates a more gradual transition at weaning and can reduce the occurrence of postweaning disorders (English, 1980). However, creep feed consumption during lactation is usually low and is also highly variable among piglets in a litter and between litters (Barnett et al., 1989; Pajor et al., 1991; Kuller et al., 2004b).
Although a review by Matte et al. (1992) showed contradictory results, previous work (Kuller et al., 2004b) demonstrated that intermittent suckling (IS), a management technique in which piglets are separated from the sow during a number of hours every day in the second half of lactation, increased creep feed intake before weaning and increased feed intake and growth shortly after weaning. However, in that study creep feed intake was determined at the level of the litter. Individual variation within the litter in relation to post-weaning performance has not been studied. Also, long-term effects of IS on piglet performance have not been investigated.
The objective of the current study was to determine if the improved feed intake found during intensive (12 h/d) IS is important for postweaning performance of piglets until slaughter. Creep feed intake of the litters was assessed, and within litters, eaters and noneaters were distinguished using chromic oxide as an indigestible marker. These creep feed intake characteristics were related to postweaning performance.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The experiment began on treatment d 0 (T0). To synchronize the beginning of IS within a farrowing batch, T0 was designated as the beginning of data collection. Piglets were born on 2.4 ± 0.3 d (range 1 to 5) before T0 in the control treatment and 2.0 ± 0.4 d (range –1 to 7) before T0 in the IS treatment. Maximum range of age within a farrowing batch was 4 d. Intermittent suckling always begun on d 14 of the treatment (T14), and weaning took place 11 d later (T25). Because weaning took place in the early morning, T25 was considered the first day after weaning. At weaning, the sows were moved to a mating room, and the piglets remained in the pen (weanling pigs). Five days after weaning, the piglets were moved as a litter to the weanling pens with 4 feeding spaces and 1 drinking nipple, with a floor space allowance according to Dutch animal welfare legislation (0.4 m2/pig, until 30 kg of BW).
At 8 wk of age (fattening pigs), 2 barrows and 2 gilts were randomly chosen from each litter and were transported to a finishing facility at another location in the Netherlands. In this facility, 4 to 8 pigs were housed together, depending on the size of the available pens. Each pen consisted of a partly slatted, concrete floor with 2 drinking nipples, and 1 feeding place, with floor space allowance according to Dutch animal welfare legislation (1 m2/pig, until 110 kg of live weight). The average group size was 7.2 ± 0.1 in the control treatment and 7.3 ± 0.1 in the IS treatment. Allotment to pens was based on BW within treatment (control vs. IS), with 1 treatment per pen. Pigs were slaughtered when their BW was approximately 107 ± 0.3 kg.
Treatments
There were 2 treatments, control and IS. Fourteen weekly farrowing batches were alternately allocated to each treatment: 7 control batches (31 sows, total) and 7 IS batches (31 sows, total). Each batch consisted of 4 to 5 sows. The batches were housed in separate rooms during lactation. The rooms (size, lights, flooring), equipment (size of the farrowing crates, feeders, drinking nipples), and environments (controlled climate) were comparable among all farrowing batches, and they were at the same facility. Only 1 treatment was applied per batch to avoid any influence of suckling in control sows because suckling is highly synchronized between animals in a barn. In the control treatment, piglets had access to the sow for 24 h/d. In the IS treatment, piglets were separated from the sow for 12 h each day (0930 until 2130) during T14 to weaning at T25. The piglets were separated by a removable wooden partition (height 65 cm) that was attached on both sides to the farrowing crate of the sow. During separation, IS piglets were allowed extra space at the back of the pen to create comparable floor space with that of the control piglets [IS, 3.75 m2 vs. control, 4.32 m2 (including the sow)]. Although the sow and piglets remained in the same pen, separation did not allow physical contact. During lactation, the lights were on between 0730 and 2330. After weaning, the lights were on between 0730 and 1630.
Feeding
The sows were fed a commercial lactational feed (9.1 MJ of NE/kg, 139 g of CP/kg, 8 g of lysine/kg, as-fed basis) at a level of 1% of BW at farrowing, plus 500 g/ piglet (Dutch feeding tables, 1995), divided over 3 meals daily. Water was available ad libitum. Creep feed, based on milk products (34%, as-fed), soybeans, corn, sugar, vegetable oil, and a premix (12.8 MJ of NE/kg, 21.7% CP, 14.7% crude fat, 2.0% crude fiber, 5.2% ash, 1.46% lysine, 0.86% Ca, 0.58% P, 160 mg of Cu/kg, 500 units of phytase/kg; as-fed basis) was offered to the piglets ad libitum from T7 onward and given in a round piglet trough. From T14 onward, a pig feeder with 4 feeding spaces was used. During T21 to T23, a gradual change (respectively, 40, 60, and 100% replacement) to a weaner diet based on milk products (18.5%), barley, soybeans, corn, sugar, vegetable oil, and a premix (11.4 MJ of NE/kg, 17.9% CP, 10.7% crude fat, 2.9% crude fiber, 5.8% ash, 1.25% lysine, 0.77% Ca, 0.59% P, 160 mg of Cu/kg, 500 units of phytase/kg; as-fed basis) was made, which was given until T31 of the experiment. From T32 onward, a series of 4 finishing diets were given. Drinking nipples provided ad libitum water to the piglets. Feed residuals were assessed at a 7-d interval. Because no food wastage was observed (the feeders were placed on a solid floor), feed that had disappeared was considered eaten.
Measurements
Individual piglet BW was assessed at birth and from T0 onward every week, until 4 wk after weaning (T52). Thereafter and until slaughter, the pigs were weighed every 4 wk. The ADG was calculated. At slaughter, the lean meat percentage was calculated using muscle and backfat thickness, measured on carcasses with the Hen-nessy Grading Probe (Engel and Walstra, 1993). Creep feed residuals were weighed per litter at 7-d intervals. After weaning, feed intake was determined weekly per pen. No food wastage was observed (the feeder was placed on a solid floor). During lactation, creep feed was supplemented with 1% chromic oxide, which colored the piglet feces if the creep feed was eaten (Barnett et al., 1989; Bruininx et al., 2002). At T17, T21, T23, and T24, fecal samples were taken at 1100, using fecal loops (Instruvet, Amerongen, the Netherlands). The color of the samples was determined visually. Piglets with green feces were considered to have eaten creep feed (Barnett et al., 1989; Bruininx et al., 2002). Classifica-tion of the piglets into eaters and noneaters was based on the criteria of Bruininx et al. (2002) and previous experimental data (W. I. Kuller, unpublished data). One point was given to piglets that had green feces at the time of sampling, and 0 points was given when no green color was observed. When the color of the feces was inconclusive, 0.5 points was given. Piglets that had 0 points were noneaters. Piglets that had a sum of 1.5 points or greater for the 4 samples during lactation were designated as eaters. Piglets that had 0.5 or 1 point were nonclassified.
Statistical Analysis
General.
All data were tested for normality using the UNIVARIATE procedure (SAS Inst. Inc., Cary, NC). The data were analyzed using the GLM and MIXED procedures of SAS. Correlations between the variables were calculated using the CORR procedure of SAS. Data are presented as least squares means ± SEM. Differences were considered significant if P < 0.05. Relative 2-way interactions that are not mentioned were not significant.
IS vs. Control During the Suckling and Weanling Periods.
Arcsine-transformed values of mortality rate per litter were normally distributed and analyzed in the following model (PROC MIXED): Yijk = µ + Ti + Bj (Ti) + Lk + eijk, where Ti = treatment, Bj = batch (nested within treatment), and Lk = litter as a random effect. The ADFI and ADG of the piglets and piglet BW were analyzed as litter characteristics using the following linear model (PROC MIXED): Yijk = µ + Ti + Bj (Ti) + Lk + aijk + eijk, where Ti = treatment, Bj = batch (nested within treatment), Lk= litter as a random effect, and aijk = age of the piglets at T0. From T14 onward, feed intake between T7 and T14 and BW of piglets at T14 (beginning of IS) was also used as a covariate in modeling ADFI. Breed and parity of the sow were never sig-nificant and were therefore omitted from the model.
IS vs. Control During the Fattening Period.
The ADG and BW were analyzed on the piglet level using the following linear mixed model (PROC MIXED): Yijkl = µ + Ti + aijk + w14ijk + sexj + penk + eijkl, where Ti = treatment, aijkl = age of the piglets at T0, w14ijkl = BW of the piglets at T14 (beginning of IS), and penk = pen as a random effect. In the analysis of slaughter BW and percentage of lean meat, duration of the fattening period was also included as a covariate. Feed intake during the entire fattening period, on a pen level, was analyzed in a linear model (PROC GLM): Yij = µ + Ti + tsij + eij, where Ti = treatment and ts = time to slaughter. In the analysis of mortality, tsij was omitted from the model. The feed to gain ratio during the entire fattening period was calculated (on a pen level) as the feed disappearance in a pen divided by the total BW gain of the pigs in this pen. Feed to gain ratio was analyzed in a GLM: Yij = µ + Ti + ts + eij with Ti = treatment and ts = time to slaughter. Group size was never sig-nificant and was therefore omitted from the model.
Eaters vs. Noneaters.
The ADG and BW of the piglets were analyzed using the following linear mixed model (PROC MIXED): Yij = µ + Eateri + aijkl + bwijkl + sexj + SPk + eijkl, where Eateri = eater or noneater, aijkl = age of the piglets in relation to T0, bwijkl = BW of piglets at T14 (beginning of IS) or BW at slaughter, and SPk = sow (for the suckling period) or pen (for the weanling and fattening period) as a random effect. Treatment and the interaction of treatment x eater were never significant (P > 0.50) and were therefore omitted from the model. In the analysis of BW at slaughter and the percentage of lean meat, the duration of the fattening period was also included as a covariate. Differences in the number of eaters between the treatments were tested using the 
2 test in SAS.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) in the finishing period was high (IS, 7.4% vs. control, 11.4%) for Dutch standards, but was not affected by treatment (P = 0.30).
|
). Therefore, ADFI was corrected for age at T0 and for creep feed intake before treatment. Average daily feed intake was also corrected for BW at T14 because this was significant in the model. After beginning IS at T14, ADFI was greater in IS litters than in control litters between T14 and T21 and between T21 and T24. Total creep feed intake during the treatment period (T14 to T24) was 284 ± 27 g/piglet in IS litters and 83 ± 28 g/piglet in control litters (P < 0.001; corrected for feed intake during T7 to T13).
|
). No difference was found from T39 to T52 (weanling period) or from T52 to slaughter (fattening period) (IS, 2.2 ± 0.1 vs. control, 2.0 ± 0.1 g·piglet–1·d–1; P = 0.46). Correlation coefficients of creep feed intake between 1 wk and the next week varied from 0.38 to 0.83 (P < 0.01), irrespective of treatment. Therefore, litters with greater feed intake in 1 wk had greater feed intake in the following week. The following relationship between creep feed intake during lactation (ADFIlact) and feed intake in the first week after weaning (ADFIaw) was found: ADFIaw = 136 + 0.26 x ADFIlact (R2 = 0.69). The interaction treatment x ADFIlact was not significant (P = 0.11).
ADG and Weight of Piglets
The ADG of piglets did not differ between treatments before the beginning of IS at T14 (IS, 232 ± 29 vs. 235 ± 29 g/piglet; P = 0.79). The ADG was lower in the IS treatment during T14 to T20 (IS, 198 ± 40 vs. control, 308 ± 40 g/piglet; P < 0.001) and during T21 to T24 (IS, 176 ± 10 vs. control, 246 ± 10 g/piglet; P = 0.005). The first week (T25 to T31) after weaning, however, ADG was greater in IS piglets (IS, 177 ± 14 vs. control, 106 ± 14 g/piglet; P < 0.001). No difference was found between treatments from T31 to T52 (IS, 755 ± 77 vs. control, 765 ± 77 g/piglet; P = 0.97) or thereafter until slaughter (IS, 795 ± 17 vs. control, 789 ± 17 g/piglet; P = 0.80).
Weight of piglets at T14 did not differ between treatments (Table 3). At T21 and at weaning (T25), piglet BW was less in the IS treatment. Within 7 d after weaning, however, this difference in piglet BW had disappeared. No differences in BW were found during the remainder of the weanling or fattening period.
|
).
Eaters vs. Noneaters
The effect of treatment and the interaction between treatment and eater were not significant in the analysis of ADG and BW and were therefore omitted from the model.
In the IS treatment, 85 pigs had more than 1.5 points (eaters), 168 pigs had 0 points (non eaters), and 48 pigs had 0.5 or 1 point (nonclassified). In the control treatment, this distribution was 67, 190, and 40 pigs, respectively. On average, an IS litter consisted of 23.0 ± 4.5% eaters, 62.8 ± 3.9% noneaters, and 13.9 ± 1.9% of nonclassified piglets. A control litter consisted of 19.0 ± 4.1% eaters, 69.8 ± 3.2% noneaters, and 11.3 ± 2.0% nonclassified piglets. Overall distribution of eaters, noneaters, and nonclassified piglets within litters did not differ between treatments (P = 0.15). Overall, more gilts than barrows were designated as eaters (30% of females vs. 21% of males, P = 0.03). From this point onward, focus will be only on eaters and noneaters.
The ADG tended to be greater in noneaters from T0 to T6 (eater, 249 ± 21 vs. noneater, 259 ± 21 g/piglet; P = 0.09) and from T7 to T13 (eater, 341 ± 25 vs. non-eater, 353 ± 25 g/piglet; P = 0.07). From T14 to T20, no difference in ADG was found (eater, 316 ± 38 vs. noneater, 325 ± 37 g/piglet; P = 0.3). During the last days of lactation (T21 to T24) ADG was greater for eaters (eater, 243 ± 18 vs. noneater, 209 ± 17 g/piglet; P = 0.007). After weaning ADG was greater for eaters during T25 to T31 (eater, 199 ± 11 vs. noneater, 106 ± 8 g/piglet; P = 0.001), T32 to T38 (eater, 463 ± 11 vs. noneater, 411 ± 7 g/piglet; P = 0.002), T39 to T45 (eater, 529 ± 19 vs. noneater, 467 ± 14 g/piglet; P = 0.003), and T46 to T52 (eater, 608 ± 21 vs. noneater, 544 ± 17 g/ piglet; P = 0.009). No difference was found in ADG during the fattening period (eater, 792 ± 16 vs. non-eater, 794 ± 13 g/piglet; P = 0.90).
During the suckling period, no difference in BW was found between eaters and noneaters except for T14, where the noneaters where heavier (Table 3). During the weanling period (T25 to T52), however, eaters were at all times heavier than noneaters. From T81 onward, 4 wk after piglets had been moved to the finishing facility, no difference was observed in BW among eaters and noneaters in the IS treatment. No differences were found among eaters and noneaters in lean meat percentage (Table 3).
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). As expected, IS resulted in lower BW at weaning (Thompson et al., 1981; Henderson and Hughes, 1984; Kuller et al., 2004c). This initial growth check during lactation in the IS group was compensated during the first week after weaning. Previous studies did not investigate long-term effects of IS on piglet performance (Grinwich and McKay, 1985; Chapple et al., 1989; Costa and Varley, 1995). Moreover, sometimes piglets were weaned at an age of 49 d (Smith, 1960), which is not compatible with current production methods. One study mentioned that time needed to reach 90 kg of live BW was not influenced by IS (Henderson and Hughes, 1984), but details about ADG and ADFI were not provided. In our experiment, IS improved ADG during the first week and increased ADFI during the first and second week after weaning but had no further positive effects on performance (BW gain and ADFI) during the fattening period and meat percentage at slaughter. The latter might be a result of the overall lower feed intake in the current experiment and the small difference in creep feed intake between the control and the IS treatment.
Intermittent suckling did not increase the percentage of eaters within a litter as was expected; the percentage of eaters in control litters was 19% and was 23% in IS litters. Apparently, IS increased creep feed intake of piglets that were already eating before the period of separation, instead of increasing the number of eating piglets within a litter. Animals within a litter that were eating consumed considerable amounts of creep feed. Considering that in the IS and in the control treatment only 23 and 19% of the animals were classified as eaters, IS piglets that were eaters consumed 1,235 g and control piglets that were eaters ate 437 g, the first being greater than the 600 g reported by English (1980). The question remains whether the percentage of eaters would be increased in experiments in which greater litter creep feed intakes were found.
Creep feed intake was much less in the current experiment than in a previous experiment (Kuller et al., 2004c) in IS (231 vs. 686 g) and control piglets (147 vs. 314 g). The reason for these low feed intakes in the current experiment is not clear. The experiment was carried out 2 yr later in the same season, on the same farm and with the same population of sows with comparable litter size and weaning weights. Weight loss of the sow was on average 3 kg greater in the current than in our previous experiment (Kuller et al., 2004a,c). This suggests that milk intake per piglet might have been greater in the current experiment, which would decrease the need for creep feed. However, such a supposed greater milk intake did not result in heavier piglets at weaning, as might be expected. The lower intakes cannot be attributed to the chromic oxide added to the feed because chromic oxide does not affect palatability (W. I. Kuller, unpublished data).
Piglets designated as eaters during lactation had heavier BW and ADG than noneaters until 28 d after weaning. This agrees with findings of Bruininx et al. (2002) who studied the effect of preweaning creep feed consumption on postweaning performance. In a previous experiment (Kuller et al., 2004b), piglets that were designated as eaters also had greater net absorption in the small intestine at d 4 after weaning. So, stimulation of creep feed intake during lactation may improve performance of newly weaned piglets. The effect of creep feed intake, however, on postweaning diarrhea was dif-ficult to assess in the current experiment because it is rarely observed on this farm.
In a previous experiment (Kuller et al., 2004c), IS litters with little or no creep feed intake during lactation tended to have greater BW gains after weaning than control litters with comparable feed intakes. We therefore speculated that these positive effects might be mediated by reduction in stress at weaning in IS piglets. In the current experiment, however, the relationship between preweaning feed intake and feed intake and growth shortly after weaning were not affected by treatment. Thus, increased performance of IS piglets shortly after weaning (increased feed intake and ADG) may not have been influenced by reduced stress at weaning. Possibly, the combination of the overall lower creep feed intake and the absence of an eater x treatment interaction on performance after weaning may have resulted in the fact that IS did not have positive or negative effects on long-term performance to slaughter.
The percentage of animals classified as eaters is dependent on classification criteria and method of classi-fication. Classification of animals as eaters, noneaters, and nonclassified was based on visual assessment of green coloring of feces, which resulted from intake of creep feed that contained chromic oxide (Barnett et al., 1989; Bruininx et al., 2002; W. I. Kuller, unpublished data). In the latter experiment (W. I. Kuller, unpublished data), piglets were forced fed during 2 successive days of lactation at different levels of creep feed, and then excretion patterns of chromic oxide were assessed by visual examination of the feces. That experiment showed that piglets eating considerable amounts of creep feed could be distinguished accurately as eaters because excretion of green feces was abundant. Animals with a low feed intake, however, often had 1 or more inconclusive samples (W. I. Kuller, unpublished data). To guarantee selection of animals that consumed considerable amounts of creep feed in the present experiment, animals were considered eaters when, out of 4 samples, they had at least 1 positive and 1 inconclusive sample or 3 inconclusive samples. Unfortunately, no alternative or more accurate marker is currently available to identify eaters and noneaters or to measure individual creep feed intake in suckling piglets under farm conditions.
The IS increased ADFI during lactation and shortly after weaning. Piglets consuming substantial amounts of creep feed (designated as eaters) showed greater BW gains and BW during the first 4 wk after weaning. However, IS did not increase the percentage of eaters within a litter during lactation, and no long-term effects on ADG or ADFI were observed throughout the finish-ing period.
|
Footnotes |
|---|
3 Current affiliation: GD-Animal Health Service, Pig Health Department, PO Box 9, 7400 AA Deventer, the Netherlands.
2 Corresponding author: w.i.kuller{at}vet.uu.nl
Received for publication March 24, 2006. Accepted for publication December 22, 2006.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
