| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
* Institut National de la Recherche Agronomique (INRA), Station de Recherches Zootechniques, 97170 Petit Bourg, Guadeloupe, F.W.I, France;
and
Station de Génétique Quantitative et Appliquée, 78352 Jouy-en-Josas, France; and
and
UMR Systèmes d’Elevage, Nutrition Animale et Humaine, 35590 St-Gilles, France
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: breed • feed intake • lactation • performance • sow • tropical climate
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), the lactating sow reduces metabolic heat production by decreasing ADFI. In tropical areas, ambient temperature is frequently greater than 22°C, and in intensive production systems buildings are usually opened or semi-opened to the outdoor ambient environment. Under these conditions, the lactating sow often suffers from heat stress.
In addition to environmental factors, ADFI is also influenced by sow-related factors (Dourmad, 1988). In comparison with conventional breeds [Large White (LW), Landrace] or their crosses, little information has been published on the performance of high-fat lactating sows. In Caribbean regions, Creole (CR) is the most popular local breed of pig. This genotype is characterized by early sexual maturity, low prolificacy, high adiposity, and its apparently good adaptation to harsh tropical environmental conditions (Canope, 1982). For this reason, the CR breed was introduced into our experimental facilities to study the genetic variability of heat tolerance. Moreover, because of its high adiposity, this breed is an interesting model to study relationships between body composition and performance during lactation.
The effect of temperature on performance of lactating sows is well known (Black et al., 1993), but limited information is available on the effect of temperature on performance of lactating sows of different breeds. This study investigated effects of season on performance and feeding behavior of LW and CR sows in a tropical humid climate. The present paper focuses on performance of sows and their nursing piglets.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
A total of 179 lactations on 71 multiparous sows (30 CR and 41 LW) divided in 24 successive groups of 9 to 11 animals were used in an experiment conducted at the INRA experimental facilities in Guadeloupe, French West Indies (latitude 16°N, longitude 61°W); this area is characterized by a tropical humid climate (Berbigier, 1988). This study covers the period between June 2001 and July 2004.
During gestation, sows were fed a conventional diet formulated with corn, wheat middlings, and soybean meal, and containing 13 MJ of DE, 140 g of CP, and 5.5 g of total lysine per kilogram of feed. Feed allowance during the first 30 d of the postmating period was calculated to standardize body condition at farrowing, according to the model of Dourmad et al. (1997). The feeding level was fixed at 2.7 kg/d from the 30th and the 114th of gestation. Ten days before parturition, sows were moved to open-fronted farrowing pens (2.1 x 2.2 m) on a slatted metal floor. Variations in ambient temperature, relative humidity, and photoperiod closely followed outdoor conditions. Lactation length was approximately 4 wk (27.6 ± 1.8 d).
Postpartum sows received 1 kg of standard gestation diet on farrowing day and controlled amounts of feed increased by 1 kg each day until d 4 of lactation to avoid overconsumption at the beginning of lactation and aga-laxia problems. The proportion of gestation diet decreased progressively over the 3-d postpartum (i.e., 0.75, 0.50, and 0.25 on d 1, 2, and 3, respectively), and sows were fed only the lactation diet on d 4. From d 5 to d 26 postpartum, sows were fed to appetite. The lactation diet (Table 1) contained 14 MJ of DE per kg, 17.1% CP, and 0.92% total lysine and was formulated using corn, wheat middlings, and soybean meal, and met or exceeded AA requirements of lactating sows (NRC, 1998). The feed was distributed once per day between 0700 and 0800 with free access to water provided by a low-pressure nipple drinker. Infrared lights provided supplemental heat for the piglets during the lactation period.
|
The day prior to weaning (i.e., d 27), sows were allocated 3 kg of feed (i.e., at least 1 kg lower than their usual feed intake) to standardize consumption for all sows prior to determination of sow weight at weaning. From weaning to d 14 after weaning, sows were presented to a mature boar twice daily to detect onset of standing estrus. Other signs of estrus such as vulva swelling or reddening, or reaction to human back-pressure, were used. Sows detected in standing estrus were mated twice at 24-h intervals, using either supervised natural mating or artificial insemination. Pregnancy diagnosis was confirmed by ultrasonography 3 wk after insemination.
Measurements
Ambient temperature and relative humidity were continuously recorded (1 measurement every 30 s) from a meteorological station (Campbell Scientific Ltd., Shepshed, UK) within 50 m of the experimental unit. Backfat thickness and BW were measured the day after farrowing and at weaning. Backfat thickness measurements were taken ultrasonically (Agroscan, E.C.M., Angoulême, France) at 65 mm from the midline at the point beside the shoulder and at the last rib on each flank. The total number of piglets born, live, stillborn, and dead during lactation was recorded for each litter. Piglets were individually weighed at birth, at d 7, 14, and 21 of lactation, and at weaning. At d 14, piglets were separated from the sows after suckling, and 30 to 50 min later (i.e., equivalent to the average suckling interval; Renaudeau and Noblet, 2001), the sow was injected with 10 IU of oxytocin (Intervet, Angers, France) in an ear vein and all functional mammary glands were hand milked. Samples (approximately 100 mL) were immediately stored at –20°C and subsequently analyzed for DM, ash, nitrogen, and fat according to AOAC (1990). Each sow’s ADFI was determined as the difference between the amount of feed offered and the amount refused the next morning (determined between 0600 and 0800). Every week, one sample of feed was collected for DM content, and successive samples were pooled for each replicate for DM, ash, fat, and crude fiber content analysis, according to AOAC (1990), for CP (N x 6.25) according to the Dumas method (AOAC, 1990), and for cell wall components (NDF, ADF, and ADL) according to Van Soest and Wine (1967). Rectal temperatures of each sow were measured Monday and Thursday at 0700 and 1200 from the Monday before farrowing to the Monday after weaning.
Statistical Analyses
Daily temperature and relative humidity (maximal, minimal, and mean) were averaged per month. Mean and daily variations of temperature and relative humidity were used in a principal component analysis and a hierarchical classification using SPAD programs (SPAD-TM, Center International de Statistique et d’Informatique Appliquées, Paris) to discriminate seasons. For lactating and reproductive performance data, sow and litter measurements were the experimental units. Because most of the lactation periods occurred over 2 successive seasons, the lactation of a sow was attributed to the season in which the sow spent the greater number of days in lactation. Three parity groups were constructed, namely 2 (n = 54), 3 and 4 (n = 69), and 
5 (n = 56). The ADFI was defined as the ADFI of the sow between d 5 and the last fed-to-appetite day before weaning. Because creep feed was provided to the pigs from d 21 of lactation, milk production was estimated only over the first 3 wk from litter BW at birth (kg), litter ADG (g/d), and the average litter size between d 1 and 21 (Noblet and Etienne, 1989). For performance data in lactation and reproductive data, the variables were interval from weaning to estrus, and interval from weaning to conception, which was defined as the number of days between weaning and a successful insemination. The proportion of stillborn (i.e., number of stillborn over total number of piglets at birth), weaning to estrus, and weaning to conception intervals were assumed to follow a Poisson distribution. Log-linear models were applied to them using the GLIMMIX Macro (Littel et al., 1996). Results are presented as least squares means after back transformation (i.e., on the original scale), and an approximate SE on the original scale was calculated using the delta method (Littel et al., 1996). Effects of breed, season, and their interaction, and the effect of parity groups, on the mean lactation performance of sows and their litter and on milk composition were tested with an analysis of variance using PROC MIXED of SAS/STAT (Version 8.1, SAS Inst., Inc., Cary, NC). The effect of group was tested within season, and a random sow effect was included to account for repeated observations for sows. Complementary statistical analyses were performed with average litter size; the lactation period was a covariate. Residual values were computed from the preceding models (without the random effect of sow), and correlation between residual values of lactating performance of sows and their litter was calculated (CORR Procedure, SAS/STAT). A Fisher’s Z transformation was performed to compare the correlation coefficients between LW and CR sows using the COMPCORR Macro of SAS/STAT, and Bonferroni correction was used to take into account multiple comparison. The effect of stage of lactation on ADFI was also analyzed using SAS PROC MIXED for repeated measures with breed and season as main effects. The least square means procedure (PDIFF option) was used to compare means when a significant F-value was obtained. Significant effects were considered at P < 0.05 and trends at P 
0.10.
|
RESULTS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Two seasons were discriminated: a warm season from November to April (23.8 ± 0.8°C) and a hot season from May to October (26.0 ± 0.5°C; Table 2). Relative humidity averaged 85% and was comparable in both seasons (P > 0.42). Duration of the diurnal period was slightly greater in the hot than in the warm season (12.33 vs. 11.67 h, Météo France source).
|
As presented in Table 3, ADFI was affected (P < 0.01) by season; it was lower in the hot season during the 28-d lactation and during the fed-to-appetite period (3.4 vs. 4.1 kg/d and 3.8 vs. 4.6 kg/d, respectively). The LW sows had greater (P < 0.01) ADFI than CR sows (4.3 vs. 3.1 kg/d, respectively). However, when ADFI was considered with respect to metabolic BW (BW0.75), this difference was no longer apparent (P > 0.10). There was a breed x season interaction for ADFI (P < 0.05) with a lower depressive effect of the hot season in CR than in LW sows (–470 vs. –910 g/d, respectively). The patterns of ADFI in LW and CR sows are presented in Figure 1. Irrespective of breed and season, feed intake increased rapidly during the restricted feeding period (i.e., from d 1 to 4) in connection with the feeding scale. After this initial period, ADFI plateaued in CR sows at around 3.5 and 3.2 kg/d in the warm and the hot seasons, respectively. On the other hand, LW sows continued to increase their feed intake to d 6, and the fed-to-appetite ADFI plateaued at around 5.2 and 4.2 kg/d during warm and hot seasons, respectively. The difference in plateau voluntary feed intake between seasons was then more pronounced in LW than in CR sows (P < 0.05).
|
|
). After farrowing, LW sows were heavier than CR sows (255 vs. 186 kg, respectively; P < 0.01) and had a lower backfat thickness (21.5 vs. 39.8 mm, respectively; P < 0.01). Backfat thickness loss during lactation was greater (P < 0.01) in CR than in LW sows (4.3 vs. 2.3 mm). The breed x season interaction was significant for BW loss during lactation; it was more substantial in the hot season for LW sows (10 kg, P < 0.01), whereas it remained constant in both seasons for CR sows. On average, rectal temperature measured during lactation at 0700 or at 1200 was greater in hot than in warm season (38.5 vs. 38.3°C, and 39.6 vs. 39.3°C, respectively, P < 0.01). It was also affected by breed, being greater in LW than in CR breeds, at both 0700 (38.5 vs. 38.3°C, P < 0.10) and at 1200 (39.5 vs. 39.3°C, P < 0.10).
Litter sizes at birth, at d 1 (i.e., after cross-fostering), and at weaning were influenced by season and breed (P < 0.10; Table 4). The proportion of stillborn, either adjusted or not adjusted for litter size, was greater in LW than in CR sows (8.3 vs. 3.3 %, P < 0.01). Average piglet BW at birth differed markedly between breeds (1.03 and 1.42 kg for CR and LW piglets, respectively) and between seasons (1.26 vs. 1.19 kg, in the warm and the hot season, respectively). Average piglet BW at d 21 and at weaning were also affected by both season and breed (P < 0.05); they were greater during the warm than the hot season (i.e., +364 and +446 g, respectively) and lower in CR than in LW breed (4.78 vs. 5.63 and 6.36 vs. 7.37 kg, respectively). Litter BW gain from birth to weaning did not differ between seasons (P = 0.80) but was greater during the warm season when adjusted for litter size (1,745 vs. 1,645 g/d, P < 0.05). It was greater in LW than in CR breeds (1,819 vs. 1,571 g/d, P < 0.01). Over the entire lactation period, piglet BW gain differed between seasons and breeds (P < 0.05); values were greater during the warm than during the hot season (210 vs. 197 g/d) and in LW compared with CR piglets (215 vs. 192 g/d).
|
). The breed x season interaction was significant for milk fat content; the interaction decreased in the hot season for LW sows (–0.80 g/100 g of milk, P < 0.05), whereas it increased in the hot season for CR sows (0.72 g/100 g milk, P < 0.05). In agreement with results on litter and piglet growth, milk production was greater in LW sows than in CR sows (5.79 vs. 4.27 kg/d and 571 vs. 506 g/d per piglet, P < 0.01).
|
. According to the statistical test of comparison and the Bonferroni correction, there is not strong evidence that the correlations in each breed were estimating different population correlation coefficients. Consequently, the residual correlations are presented with the 2 breeds together. Even if residual correlation values presented in the text were generally low (i.e., < 0.30), they were different from zero (P < 0.05). Average daily feed intake (expressed per kg of BW0.75) had a negative correlation with backfat thickness at farrowing (r = –0.19, P < 0.05) and a positive correlation with litter BW gain (r = 0.56, P < 0.01). A positive correlation was also found between sow ADFI and rectal temperature measured at midday (r = 0.16, P < 0.10) and between litter BW gain and backfat thickness loss (r = 0.26, P < 0.01), but after adjustment for litter size these correlations were not different from zero.
|
|
DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The effect of high ambient temperature on performance of lactating sows is well known in the literature (Black et al., 1993) with a severe reduction of feed intake when ambient temperature rises above the evaporative critical temperature of the sow (i.e., 22°C, Quiniou and Noblet, 1999). Under our tropical humid conditions, the daily ambient temperature frequently exceeded 22°C. Therefore, lactating sows suffered from heat stress most of the time (Gourdine et al., 2004). In the current study, ADFI in LW sows was reduced by about 950 g/d in the hot season, which is consistent with previous results obtained by Renaudeau et al. (2003; i.e., 1,460 g/d) and Gourdine et al. (2004; i.e., 850 g/d) in the same experimental conditions. Expressed per degree increase of ambient temperature, the reduction of ADFI was equivalent to 430 g–1·d–1/°C. This result is critically higher than the value reported also in LW sows by Quiniou and Noblet (1999; i.e., 215 g–1·d–1/°C) between 25 and 27°C for uncontrolled relative humidity ranging between 50 and 60%. In agreement with Renaudeau et al. (2003) and Gourdine et al. (2004), in tropical humid conditions, the high relative humidity emphasizes the negative effect of elevated ambient temperature on sow ADFI. In fact, it is now well established in growing pigs that the evaporative heat loss through increased respiratory rate is limited when relative humidity is high (Renaudeau, 2005). In addition, the decrease of metabolic heat production associated with the reduction of ADFI is insufficient to prevent the increase of rectal temperature. In the current study, the increase of rectal temperature (i.e., 0.3°C) in the hot season is similar to the value reported by Quiniou and Noblet (1999) for multiparous LW sows when the temperature increased from 25 to 27°C (0.3°C).
According to our results, litter size at birth was greater when the lactation was performed during the hot season. Omtvedt et al. (1971) found that high temperatures during the early periods and during the end of gestation had adverse effects on embryo and fetus survival. Consequently, the effect of the season of lactation on litter size is probably not the best criteria to determine the effect of season. When season of mating was considered, sows mated during the hot season had fewer piglets at farrowing (–0.7 and –1.2, for CR and LW sows, respectively; results not presented).
In the warm season, piglet growth rate over the first 3 wk of lactation in LW breed (i.e., 201 g/d) was lower than results reported in temperate conditions (Auldist and King, 1995: 265 g/d; Hulten et al., 2002: 290 g/d; Quiniou, 2005: 253 g/d). As observed for ADFI, this result suggests that LW sows raised in tropical conditions were heat stressed and their milk production was depressed, even in the warm season. According to Mullan (1991) and Quiniou and Noblet (1999), milk yield is reduced at elevated temperatures. Messias de Bragança et al. (1998) suggested a direct influence of high temperature on milk yield. In the current study, the effect of season on milk production or on litter BW gain was not significant due to a greater litter size in the hot season. When litter weight gain or milk production was adjusted for litter size, values differed significantly between seasons with lower values during the hot season (1,745 vs. 1,645 g/d for litter weight gain). Moreover, when milk production was expressed per piglet or when piglet growth rate was considered, the amount of milk available for each piglet decreased in the hot season, which indicated that the negative effect of heat stress on sow milk production was emphasized during the hot season. Irrespective of season, piglet BW gain between d 21 and weaning was greater than piglet gain from the first 3 wk of lactation, in connection with the creep feed allowance during this period. In fact, it can be hypothesized that piglets compensate for the low milk intake by increasing their creep feed consumption with a subsequent attenuated effect of heat stress on performance of the litter (Renaudeau and Noblet, 2001).
The experiment did not show significant effects of season on weaning-to-estrus interval. A high mobilization of body reserves has negative effects on sow fertility after weaning, particularly in primiparous sows. In another study (J. L. Gourdine, unpublished data), primiparous sows lost more body reserves during the hot season and had a higher risk of having a prolonged return to estrus than multiparous sows. This supports the hypothesis that the absence of significant effect of season on the weaning-to-estrus interval of multiparous sows in the current study may be related to a moderate body reserves mobilisation. Similar conclusions were obtained in temperate climatic conditions (Hughes, 1998; Vesseur et al., 1994).
Effect of Breed on Sow and Litter Performance
Little information on the effect of breed on performance of lactating sows is available (Sinclair et al., 1999). As described in the literature review of Eissen et al. (2000), genetic differences in sow lactation performance will, to some extent, reflect differences in BW and body composition at farrowing and in litter size and milk production. In the current study, ADFI was significantly lower in CR than in LW sows, but no difference was found when ADFI was considered with respect to metabolic BW. Consequently, the lower feed intake of CR sows may be due mainly to a between-breed difference in energy requirements for maintenance (i.e., BW). Moreover, it cannot be excluded that, to some extent, the reduced ADFI measured in CR sows would also be related to their high adiposity at farrowing.
In accordance with Canope (1982), the CR sow was less prolific than the LW sow, which is one of the most widely used maternal breeds. Despite the important genetic trend for prolificacy in LW breed that occurred during the last 15 to 20 yr (Tribout et al., 2003), the breed difference is not likely to have increased over the last 23 yr. The lower proportion of stillborn in CR sows can be partly attributed to their lower prolificacy. The remaining difference in favor of CR after adjustment for litter size might be due to a lower vigor or maturity of LW piglets. Indeed, Herpin et al. (1993) and Canario et al. (2005) showed that selection for leanness has resulted in less mature piglets at birth. In agreement with Canope (1982), the average piglet BW gain was significantly lower in CR breed. The between-breed variation for litter BW gain could be related to differences in the sow’s ability to produce milk or in chemical composition of milk, or differences in growth potential of piglets or some combination of these factors. According to equations from Noblet and Etienne (1989), milk production was 1.5 kg/d lower in CR than in LW sows. The reduced milk production in CR sows is partly related to their lower prolificacy, which decreases the nursing demand. However, a 880 g reduction in daily milk production was found for CR sows after adjustment for litter size. When milk production per piglet was adjusted for piglet BW at birth, values did not differ significantly between breeds. From these results, it appears that the reduced milk yield in CR sows seems to be also the result of their lighter piglets. Using a cross-fostering technique, Kanis et al. (1990) found that the reduced milk intake in lighter Meishan than in heavier Dutch piglets was mainly caused by differences in birth BW. In fact, nursing demand is also related to piglet BW; heavier piglets are more efficient for obtaining milk during suckling than lighter piglets (King et al., 1989).
Nutrients for milk production come from feed and/or body reserves. Because the ADFI expressed per kg0.75 was not affected by breed, variation in quantity and/or in composition of lactation BW loss would also explain between-breed difference in piglet BW gain. According to our results, BW loss was not affected by breed, but backfat thickness loss was twice as great in CR than in LW sows, suggesting a higher mobilization of body fat reserves in CR sows. As the chemical composition of BW loss depends to a large extent on the composition of the feed, this result highlights that nutrient requirements may differ between the two breeds. In contrast to the results of O’Grady et al. (1973), the greater body fat mobilization did not induce an increase of DM and fat content of milk in CR sows. Moreover, the protein and fat contents in milk were greater in LW than in CR sows. When Azain et al. (1996) provided piglets with milk substitute with a protein:energy greater than sow milk, they observed a greater piglet BW gain. As reported in a review (Williams, 1995), sow milk is deficient in protein relative to its energy content. These results indicate that the greater protein content of milk in LW sows can also partly explain the better piglet growth rate. To summarize, our results suggest that the effect of breed on piglet BW gain is mainly related to a decrease of milk yield and small changes in milk composition.
Effect of Season and Breed on Sow and Litter Performance
To our knowledge, few studies are available on the season to breed interaction effects on performance of lactating sows. Data from the current study show that the negative effect of season on ADFI was accentuated in LW sows. Indeed, the reduction of ADFI during the hot season represented 14% (–470 g/d) and 20% (–910 g/d) of ADFI during the warm season for CR and LW sows, respectively. In accordance with the results reported by Renaudeau (2005) in growing pigs, our results suggest that in the present experimental conditions, the CR breed seemed to tolerate hot conditions better than the LW breed. From measurement of individual feeding behavior parameters in a subgroup of sows used in this experiment, we showed that unlike LW sows, CR sows were able to eat feed even during the hottest hours of the day, which confirms their greater heat tolerance (Gourdine et al., 2006). This better heat tolerance in CR sows could be related to their lower production level. Nienaber et al. (1997) demonstrated a greater sensitivity to heat stress in pigs from high lean growth potential lines than those from moderate lines.
However, an improved heat tolerance in CR sows could be the result of their better ability to dissipate heat. Indeed, Berbigier (1975) found a greater nonevaporative heat loss in CR x LW crossbred piglets than in LW piglets. Nonevaporative heat loss is partly dependent on cutaneous heat conductivity of the animal. As reviewed by Renaudeau et al. (2004), it appears that an increase in heat conductivity of the animal by changing its blood flow to skin vessels is observed when the ambient temperature exceeds the lower critical temperature. Further investigations are needed to confirm the greater ability of the CR breed to tolerate heat stress and to explain the mechanisms implicated in this adaptation.
|
Footnotes |
|---|
2 Corresponding author: renaudeau{at}antilles.inra.fr
Received for publication April 20, 2005. Accepted for publication September 12, 2005.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. L. Gourdine, J. P. Bidanel, J. Noblet, and D. Renaudeau Effects of season and breed on the feeding behavior of multiparous lactating sows in a tropical humid climate J Anim Sci, February 1, 2006; 84(2): 469 - 480. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) on feed intake during lactation (d 1 to 26). Feed intake was not different from d 1 to 3 (P > 0.05), whereas it differed between seasons and breeds from d 4 to 25 (P < 0.05). Within breed, feed intake during the warm season differed (P < 0.05) from feed intake during the hot season from d 5 to 26 and from d 3 to 18 in Large White and Creole sows, respectively; the pooled SE was 184 g.