Sow productivity has developed extensively in recent decades due to the improvement of the high genetic potential in modern sows. According to the Canadian Centre for Swine Improvement (2015) the average litter size is 14.00 in modern sows, mainly due to the increased sow productivity. Efforts to meet the nutrient requirements of high-producing sows have been undertaken by supplementing nutrients to support both the normal reproductive cycle and body maturation (Boyd et al., 2000). Primiparous sows are particularly sensitive to adequate energy during gestation, as they are still growing and utilizing ingested nutrients to support body maturation, growth of the fetus and body maintenance (Jang et al., 2014).

However, prior research in energy supplementation in the diet is still controversial. Almeida (2000) reported that high dietary energy levels during the ovulation period can increase the ovulation rate and promote progesterone secretion to increase fetal survival. In addition, Kongsted (2005) demonstrated that higher energy intake in gestation period may reduce the risk of being culled due to pregnancy failure because of insufficient body weight (BW) gain and backfat deposition. Recently, NRC (2012) suggested that higher energy intake is good for the development and growth of the fetus, corresponding tissues (placenta, uterus, and mammary tissue) and deposition of maternal lipids and proteins. The energy requirement of gestating gilts and sows should be between 6,678 and 8,182 kcal of metabolizable energy (ME)/d, which is 1,650 kcal higher than previous recommendations (NRC, 1998). However, more research is needed to understand the precise energy requirements of gestating gilts.

Therefore, the objective of the current study is to determine the optimal dietary energy level for reproductive performance in high-producing modern sows.

The effects of the energy level on BW and BFT are presented in Table 2. Body weight and BW gain in gestation increased as the dietary energy level increased (linear response, p<0.05). Back fat thickness was not affected by dietary treatments during the gestation period, but back fat gain from breeding to 110 d of gestation increased linearly (p<0.05) as the dietary energy level increased. Body length was not affected by dietary treatments. During the lactation period, the BW and BW gain of sows decreased as energy levels increased (linear response, p<0.05), but back fat thickness was not affected by treatments. Consequently, the BW change of gilts from mating to 3 weeks of lactation tended to decrease as energy levels increased (linear response, p = 0.06). During the lactation period, the voluntary feed intake of sows tended to decrease (linear response, p = 0.08) when dietary energy levels increased (Table 2). The weaning to estrus interval (WEI) was not significantly affected by treatments, but the culling rates of sows were highest in the 3,100 kcal/kg ME treatment (Table 2). The estimated protein and fat masses of gilts were calculated based on the BW and BFT (Dourmad et al., 1997). The protein mass and fat mass increased as energy levels increased (linear response, p<0.01, and p<0.05, respectively) during the gestation period (Table 3). During the lactation period, the protein mass and fat mass decreased (linear and quadratic response, p<0.01, and p<0.05, respectively), while the dietary energy level increased (Table 3). There were no significant differences in the total number of pigs born per litter or litter birth weight (Table 4). There were also no significant differences in the fat, protein, lactose, or solid-not-fat (SNF) content of colostrum and milk (Table 5). There was no significant difference in the fatty acid content of colostrum due to dietary energy levels (Table 6). The BUN concentration in the serum of gilts tended to increase with dietary energy level at d 110 of gestation and 24 h postpartum (linear response, p = 0.06, and p = 0.07, respectively) (Table 7). The glucose concentrations tended to increase with dietary energy level (linear response, p = 0.07) at 24 h postpartum, but the FFA concentration was not affected (Table 7). There were no significant differences in the concentration of IgG or IgA in colostrum (Table 8).

In the current study, increasing energy intake led to a higher BW and BW gain from 42.5 to 50.46 kg, which agreed with the results of Noblet et al. (1990), who observed an increased BW of gilts during gestation with higher energy intake. This observation may be attributed to the higher BFT due to the high dietary energy during gestation (Long et al., 2010). In the present study, although gilts consumed an equal amount of amino acids, fat tissue and protein tissue was increased during gestation, but decreased during lactation, which implies that the energy supply is important to maintain adequate BW and BFT for subsequent reproductive cycles. These results are in agreement with previous studies, which demonstrated that increased feed intake during gestation increased sow weight loss during subsequent lactation (Piao et al., 2010). Moreover, Long (2010) demonstrated that providing high-energy feed during gestation increased BW and BFT loss in the lactation period. The BFT loss during lactation was observed in the highest energy level treatment, which may be attributed to reduced feed intake during lactation due to higher dietary energy levels in gestation.

In this study, WEI was not affected by treatments and the culling rate was highest in the 3,100 kcal/kg ME treatments. Previous studies indicated that increasing body fatness at farrowing is likely to decrease feed intake during lactation (Xue et al., 1997). Low feed intake during lactation could result in higher BW loss and subsequently lead to several common reproductive problems, such as an increased WEI interval (Baidoo et al., 1992), increased anestrus incidence after weaning, and decreased conception rate (Kirkwood and Thacker, 1988). Although feed intake decreased and BW loss increased during lactation with increasing energy levels during gestation, high dietary energy levels did not increase the culling rate in the first parity. However, treatment with 3,100 kcal of ME/kg showed the highest culling rate (38%) due to pregnancy failure and post-weaning anestrus, which suggests that low energy intake during gestation may increase the risks of being culled due to pregnancy failure (Kongsted, 2005). The pregnancy rate and post-weaning anestrus may be correlated with some hormones, and the dietary energy levels differentially affect the release of some reproductive hormones (Kemp et al., 1995). Whitley (2002) reported that insulin targets during reproduction had little connection to nutrition, the body condition, and other management factors. These results indicated that dietary energy levels during pregnancy for optimal longevity are not easy to establish. Further study is needed to clearly demonstrate a possible correlation between dietary energy levels during gestation and reproductive hormones. WEI is delayed by low plasma insulin concentrations during lactation, and the concentration is increased by glucose but lowered by FFA concentrations in the plasma of weaning sows (Armstrong et al., 1986). In this study, the plasma glucose and FFA concentrations of sows at weaning were not affected by treatments, indicating that dietary energy levels did not affect WEI.

NRC (2012) suggested dietary energy levels for 140 kg BW gilts during gestation between 6,678 kcal of ME/kg to 7,932 kcal of ME/kg daily. However, our results suggested that litter size was not affected by increasing dietary energy, which may be considered to be a balance between gestation preparation and early gestation. High energy levels had positive effects on the ovulatory rate, but high energy levels might decrease embryonic survival after breeding (Jindal et al., 1996), while low energy supplies from day 3 after mating until day 15 do not affect embryo survival.

There are many of scientific studies assessing the effects of increased feed or energy intake in gestating sows on piglet birth weight (NRC, 1998). Daily energy requirements during gestation include maintenance for the sow maternal gain as well as uterine growth. Increased energy intake during late gestation can positively affect fetal growth (NRC, 2012). In this experiment, there was no treatment effect on litter birth weight and individual piglet birth weight, which is consistent with the results from Long (2010), who demonstrated that the average piglet BW at farrowing was not affected by energy levels in gestation diets. Increased feed intake during gestation also did not increase the litter weight or individual piglet weight (Piao et al., 2010). This result demonstrates that increased feed intake during gestation could increase BW gain and BFT in gestating gilts. However, increased BW and backfat did not increase litter weight or individual piglet weight.

Yang (2008) reported that there was no change in colostrum composition when the energy levels increased from 13.7 to 14.2 MJ of ME/kg during gestation. Our study suggests that there were no significant differences in the chemical composition (fat, protein, lactose and SNF) of sow colostrum and milk, which is consistent with the results from Willams (1985), who confirmed that the chemical composition of colostrum and milk was not affected by dietary energy levels during gestation because sows mobilized their internal reserves to compensate for deficient nutrients. In the present study, the composition of fatty acids in colostrum did not show any changes due to the dietary energy level of the gestation diets. The fatty acid composition of colostrum is affected by dietary fat levels (Christon et al., 1999). Several studies have investigated how the composition of fatty acids in colostrum is affected by energy levels in gestation diets, and the mechanism is not yet clearly understood. In this experiment, the composition of fatty acids did not show any changes due to the different treatments, but the differences in energy levels among treatments simply may not have been high enough to affect the composition of fatty acids in colostrum.

It is well known that BUN is directly related to protein intake and inversely related to protein quality (Hahn et al., 1995) and the retention of dietary nitrogen in the body (Whang and Easter, 2000). In the present study, the BUN concentration in serum tended to decrease with the increasing energy level at 110 days of gestation and 24 hours post-farrowing. Ruiz (1971) reported that the BUN concentration was higher in swine fed low-energy diets compared to pigs fed high-energy diets, which may suggest that energy intake in sows is affected by protein metabolism during gestation. The glucose concentration influences insulin secretion (Quesnel and Prunier, 1998), and insulin can reduce the oxidation of FFA for the production of energy (Gamble and Cook, 1985). The glucose concentration increased with higher energy levels 24 hours post-farrowing in the current study, whereas FFAs were not affected by treatments. Xue (1997) demonstrated that increased energy levels during gestation could decrease glucose utilization and subsequently decrease feed intake during lactation. Piao (2010) also suggested that increased feed intake in gestating gilts may cause sows to become insensitive to insulin, thereby exhibiting a smaller response in glucose clearance and decreased feed intake during lactation. Therefore, it can be assumed that the effect of insulin on feed intake in lactation might depend on the body condition of sows and metabolites of glucose, which may explain our result that BW and BFT decreased with energy levels during lactation.

A greater intake of immunoglobulins in colostrum may increase immune function in nursing pigs. Some previous research reported that polyunsaturated fatty acids (PUFA) can influence production of immunoglobulins in mammals. Mitre et al. (2005) reported that supply of shark-liver oil to sows from d 80 of gestation to farrowing could be increased IgG concentrations in the colostrum, but not IgA. Mateo et al. (2009) reported increased IgG concentrations in the colostrum and milk of sows fed a fish product. Clearly, dietary n−3 and n−6 FA’s are involved in immune and inflammatory processes.

In the present study, the concentrations IgG and IgA in colostrum was not affected by the tallow in gilt diets, which has lower PUFA than other energy source.

NRC (2012) suggested that the energy requirement of the gestation gilt should be between 6,678 and 7,932 kcal of ME/d. Similarly, our results suggested that 3,100 kcal of ME/kg is not enough to maintain the reproductive performance for gilts during gestation with 2 kg feed daily. Gilts in the treatment 3,400 kcal of ME/kg tended to have a higher weaning number of piglets, but bodyweight and backfat loss were higher than other treatments during lactation. Consequently, an adequate energy requirement of gestating gilts is 6,400 kcal of ME/d.