Effects of Adding Medium-Chain Triglycerides to Sow Diets During Late Gestation and Early Lactation on Litter Performance'

Transcription

1 Published December 11, 2014 Effects of Adding Medium-Chain Triglycerides to Sow Diets During Late Gestation and Early Lactation on Litter Performance' M. J. Azain Animal and Dairy Science Department, University of Georgia, Athens ABSTRACT: This experiment was conducted to determine whether feeding medium-chain triglycerides (MCT) to sows during late gestation and early lactation would improve neonatal pig survival. Beginning on d 91 of gestation and continuing through d 7 of lactation, sows were fed isoenergetic (7,000 kcal of ME/d) and isonitrogenous (278 g of CP/d) amounts of either control (19% starch, 2% soybean oil), longchain triglycerides (LCT, soybean oil, 12%), or MCT (10% MCT, 2% soybean oil) diets. Sows (n = 18, 19, and 17, respectively) were induced to farrow on d 112 of gestation. Litters were weighed at birth, before suckling, and on d 1, 3, 7, and 21 of lactation. There was no effect of treatment on average pig weight at any time and no difference in the number of live pigs at birth. Beginning on d 3 ( P <.05) and continuing through weaning (d 21, P <.02) survival was improved in litters from sows fed MCT relative to litters from sows fed the control diet. Overall survival rates were 80, 81, and 90% in control, LCT, and MCT groups, respectively. The greatest improvement in survival was observed in pigs weighing < 900 g at birth. Survival of pigs in this weight range was 32, 53, and 68% in control, LCT, and MCT treatments, respectively. Although feeding MCT resulted in an increase in content of medium-chain fatty acids in milk, these accounted for < 5% of the fatty acids in milk and likely cannot account for the improved survival rate. The observation of increased blood glucose ( P <.05) at birth in pigs from both the LCTand MCT-fed sows is supportive of a prenatal effect of the diets. The results suggest that MCT may provide a practical means to improve survival rates, particularly in herds with preweaning mortality > 15%. Key Words: Sows, Pigs, Survival, Dietary Fat, Medium Chain Triglycerides J. him. Sci : Introduction Neonatal mortality represents a major loss to the swine industry. The majority of these deaths occur during the first 3 d after farrowing with losses typically in the range of one or two pigs per litter (Pomeroy, 1960; England, 1986; USDA, 1992). Ultimately, both stillborn and postfarrowing losses can be attributed to the physiological immaturity and lack of stored energy in the baby pig that result in an overall lack of vigor and inability to survive (Mersmann, 1974). Mortality in pigs weighing c 1 kg at birth is 30 to 40% and mortality increases sharply in pigs 'The author wishes to acknowledge the suggestions of Bob Seerley on the design of the experiment, James Peifer for the development of the gas chromatographic procedures and Robert Dove for improvement of the manuscript, and the technical assistance of Tom Glaze and Claudia Weidwald in conducting these experiments. I would also like to thank P. J. Reilly and F. L. Detrano of Karlshamms Lipid Specialties for supplying the mediumchain triglycerides used in these studies. Received September 8, Accepted June 30, weighing <.7 kg. The most successful means to decrease neonatal mortality has been through labor intensive management procedures such as induced and attended farrowings, crossfostering, and providing supplemental milk to lower weight pigs (England, 1986). Nutritionally, fat feeding during late gestation and into lactation has been shown to result in improved survival (Seerley et al., 1974; Pettigrew, 1981). Improved survival or improved indices of survival have been observed in litters from sows fed ketogenic compounds such as 1,3-butanediol (Stahly et al., 1985, 1986) and medium-chain triglycerides (Rosebrough et al., 198 1). Medium-chain triglycerides ( MCT), because of their rapid metabolism to ketone bodies and the ability of ketone bodies to both stimulate lipogenesis and spare carbohydrate use for nonoxidative functions (Allee et al., 1972; Seccombe et al., 1977; Shambaugh, 19851, have also been used postnatally to improve survival by direct administration to the pig within 24 h after birth (Benevenga et al., 1989; Odle et al., 1989). The objective of this work was to determine whether feeding MCT to the sow during late gestation and early lactation would improve litter performance. 3011

2 ~ 3012 AZAIN Table 1. Composition of basal and lactation diets Table 2. Treatment diet composition Ingredient, 9% Basal Lactation Corn Soybean meal (49% CP) Soyhulls Dicalcium phosphate Limestone salt Vitamin premixa Trace mineral premixb Lysine Aurofac aprovided the following per kilogram of premix: vitamin A, 1.98 x lo6 IU; vitamin D3, 3.52 x lo5 IU; vitamin E, 8.8 x lo3 IU; menadione, 1.32 g; pantothenic acid, 8.8 g; niacin, g; ribofla- Vinb1.84.g; choline chloride, 44 g; BIZ, 8.8 mg; and biotin, 22 mg. Provided the following per kilogram of premix: Mn, 150 g; Zn, 120 g; Fe, 50 g; Cu, 6 g; I, 2 g; and Se,.6 g. Materials and Methods The study was conducted in three separate farrowing groups (October 1989, February 1990, and April 1990) of approximately 20 crossbred (Yorkshire x Landrace or Yorkshire x Duroc) sows each. Treatments were assigned randomly within parity (38 primiparous, 16 multiparous) to sows in each farrowing group. Beginning on d 91 of gestation, sows were assigned to one of three dietary treatments: control, long-chain triglycerides (soybean oil, LCT), or MCT. The composition of the basal diet is shown in Table 1. Cornstarch, soybean oil, or MCT were mixed with the basal diet to yield the experimental diets presented in Table 2. The ME value of MCT was assumed to be similar to that of soybean oil (Newcomb et al., 1991). Sows were fed a preweighed amount of diet once daily (0800) such that the total intake of crude protein (278 g/d) and energy (7,000 kcal of ME/d) was similar across diets. Thus, only the source of energy was altered. The control diet contained approximately 2% soybean oil and 19% cornstarch. The LCT diet contained approximately 12% soybean oil and the MCT diet contained 2% soybean oil and 10% MCT (Capital Cities Products/Karlshamms Lipid Specialties, Columbus, OH). To have attended farrowings and obtain birth weights before suckling, sows were induced to farrow on d 112 of gestation (Lutalyse, Upjohn, Kalamazoo, MI). Body weight of pigs in each litter was obtained at birth, and on d 1, 3, 7, and 21 of lactation. Sows were allowed to consume the experimental diets ad libitum up to d 7 of lactation, at which time all sows were allowed ad libitum access to a standard lactation diet (Table 1). Blood and milk samples were obtained from animals in the first two farrowing groups (n = 11 to 1Ytreatment). Jugular blood samples were obtained from sows on d 110 of gestation and d 7 of lactation at 1100 (approximately 3 h after feeding). Blood samples were obtained by heart puncture from a pig in Treatment Ingredient Control LCP MCT~ kg/d Basal dietc Cornstarch Soybean oil MCT~ Total fed/dd alct = Long-chain triglycerides (soybean oil). bmct = Medium-chain triglycerides (Captex 300, Karlshamns Lipid Specialties, Columbus, OH 43216). CComposition of basal diet shown in Table 1. ddiets were formulated to provide a similar level of energy (7,000 kcal of ME/d) and protein (278 g of CP/d) when fed as shown. each litter at birth, before suckling. Selection of the pig from which a sample was obtained was based on a ranking of birth weights within the litter into heavy, medium, and light categories. Blood was obtained from a randomly selected pig in the lighter third of the litter. If treatment had an effect on survival, it was reasoned that pigs in this category would express a metabolic change, such as elevated glucose. Serum was obtained from sow and pig blood samples by centrifugation and was stored at -20 C for determination of glucose and triglycerides using commercially available kits (Sigma, St. Louis, MO). Milk samples were obtained, after injection of 10 IU of oxytocin, during the first 24 h after farrowing and on d 7 of lactation. Samples were stored at -20 C for determination of fatty acid profile and total lipid content. Because of the expected presence of medium-chain fatty acids (volatile) in milk from MCT-fed sows, modified extraction, saponification, and methylation procedures that did not require solvent evaporation were developed and validated. Milk samples were prepared for fatty acid analysis by first saponifying a 1-mL aliquot of milk in 1 ml of 4 N NaOH and 1 ml of methanol. One milligram of a C17:O internal standard (1 mg/ml in methanol) was added before saponification. Samples were placed in tightly capped, 20-mL screw cap vials and heated at 75 C for 6 h. Vial contents were cooled on ice for 5 min and acidified by the addition of 2 ml of 3 N HC1. Fatty acids were extracted into 3 ml of hexane and transferred to a second vial for methylation. Three milliliters of methanol and 1 ml of methanolic boron triflouride (Sigma Chemical, St. Louis, MO) were added and samples were heated at 95 C for 1 h. Samples were cooled to room temperature (approximately 30 min) and 4 ml of water and 2 ml of hexane were added. Upon separation of the aqueous and organic phases, an aliquot of the hexane phase was transferred to small teflon-capped vials for analysis by gas chromatography ( GC). Completeness of saponification and methylation steps was confirmed by TLC using

3 MEDIUM-CHAIN TRIGLYCERIDES AND PIG SURVIVAL 3013 appropriate triglyceride, free fatty acid, and fatty acid methyl ester standards. Fatty acid methyl esters were separated using a temperature-programmed procedure on a 1.83-m x 2-mm i.d. glass column packed with GP 10% SP-2330 on 100/120 Chromosorb WAW (Supelco, Bellefonte, PA) on a model 6000 Varian (Varian Analytical Instruments, Sugarland, TX) gas chromatograph equipped with an auto sampler and integrator. A 1-pL injection of the methylated fatty acids was used. Initial column temperature was 100 C and was maintained for 1 min followed by a 7.5"C/min increase to 200 C (He flow 20 mumin; flame ionization detector, detector temp 215"C, injector temp 210 C). Peaks were identified using a known standard mix containing C6:O through C18:3 methyl esters (Reference Standard #86, Nu-Chek-Prep, Elsian, MN). Quantification of peak areas was based on the reference standard, with correction for recovery of the internal standard. Total lipid content of the milk was determined gravimetrically by chlorofordmethanol extraction of a 1-mL aliquot of milk (Folch et al., 1957). Statistical Analysis. Results were analyzed using the GLM procedure of SAS (1985). The effects of treatment, parity (primiparous vs multiparous), and farrowing group, and their interactions were included in the model. Differences in litter performance, blood metabolites, and milk fatty acid profile were determined using orthogonal (Control vs LCT and MCT, LCT vs MCT) and preplanned, nonorthogonal (Control vs MCT) contrasts. Preweaning survival was adjusted for the number of live pigs at birth by covariance. A graphic presentation of the differences in survival rate based on birth weight was made for each treatment by plotting survival rate (percentage) vs birth weight for pigs in 100-g birth weight increments from 550 to 2,050 g. A second-order regression line was then drawn to better visualize these differences (Cieslak et al., 1983). However, because this line is influenced by pigs in the lighter birth weight categories and these pigs represent only 10 to 15% of the total pigs, a statistical test of these lines was not performed. Rather, significance of differences in survival of pigs weighing < 900 g at birth were determined using chi-square analysis. Results During gestation, sows adapted readily to the high fat diets (LCT or MCT). During the 1st wk of lactation, sows consumed slightly more of the LCT and MCT diets (22.7 and 23.0 kg) than the control diet (20.7 kg, Table 3). Diet had no effect on sow blood glucose, although there was a trend (P <.1) for decreased glucose in sows fed the MCT diet. Both the LCT and MCT diets increased circulating triglycerides (P <.05) in late gestation and lactation, with the effect of MCT being less evident in lactation. Relative to the control group, pigs from both LCT- and MCT-fed sows had increased blood glucose ( P <.05) at birth. There was no effect of treatment on average pig weights at birth or at any time through weaning (Table 4). There was no difference among treatments in number of live pigs at birth or in stillborn pigs. However, beginning on d 1 (P <.05) of lactation and continuing through weaning ( P <.Ol), the number of pigs per litter was greater in the MCT group relative to the control. Although the contrast of control vs LCT and MCT indicates an overall benefit of fat feeding ( P <.l), this response is entirely accounted for by the MCT group. Table 3. Sow weights, feed intake, parity, and circulating glucose and triglycerides Item Significance, P < Control LCT Control Control LCTa MCT~ SEM vs fat vs MCT vs MCT No. of sows Panty, primiparous/multiparous 13/ / Initial wt, kg NSC NS NS Feed intake Day 0 to 7 lactation, kg NS.1 Blood glucose, mg/dl Day 110 gestation NS.2.1 Day 7 lactation NS.1.1 Blood triglycerides, mg/dl Day 110 gestation Day 7 lactation NS Pig blood glucose at birth, mg/dl NS.1 *LCT = long-chain triglycerides (soybean oil). bmct = medium-chain triglycerides. CNS = not significant.

4 ~ ~ 3014 AZAIN Table 4. Litter characteristics Item No. of sows Significance, P < Treatment Control LCT Control Control LCTa MCT~ SEM vs Fat vs MCT vs MCT Avg pig wt, kg Birth (live pigs) NSC NS NS Day NS NS NS Day NS NS NS Day ll NS NS NS Day NS NS NS Stillborn, pigsflitter NS NS NS Live pigsnitter Birth NS.1.2 Day NS Day Day Day ,005 Total litter wt, kg Birth NS NS NS Day o 1 Survival, 9% of pigs born alived Day NS NS.2 Day Day Day Preweaning mortality, no. of pigflitter' alct = long-chain triglycerides (soybean oil). bmct = medium-chain triglycerides. CNS = not significant. dsurvival and preweaning mortality adjusted for the number of live pigs at birth. There was no difference in the total litter weight at birth. At weaning, the total litter weight in sows fed the MCT diet was 22 and 18% greater than that in litters from sows fed either the control (P <.01) or LCT diets (P <.05), respectively. Percentage of survival (birth to weaning) was improved with MCT, whether compared with the control ( P <.01) or the LCT (P <.05) groups (Table 4). The results are presented as the percentage of pigs that survived to d 1, 3, 7, and 21, and have been adjusted by covariance for the number of live pigs at birth. The overall 10% improvement in survival in litters from MCT-fed sows resulted in a > 1 pig per litter reduction in mortality ( P <.05). There was no significant benefit of the LCT diet on survival. However, further analysis of survival rates in various birth weight categories revealed an improved survival of lighter weight pigs in litters from both fat diets. Survival of pigs weighing < 900 g at birth was 32% (6 of 19) in control litters and was increased to 53% (16 Table 5. Effect of late gestation diet on survival of pigs within birth weight categories Birth wt, g < good 901-1,200 1,201-1,500 1,501-1, ,801 Control LCTa MCT~ % of pigs SurvivalC % of pigs Survival % of pigs Survival alct = long-chain triglycerides (soybean oil). bmct = medium-chain triglycerides. CPercentage of live-born pigs in each birth weight category surviving to weaning (d 21). dbased on chi-square analysis with survival rate in control litters used as the expected, survival was improved in bath LCT (P <,011 and MCT (P <,001) groups.

5 MEDIUM-CHAIN TRIGLYCERIDES AND PIG SURVIVAL 3015 A. CONTROL B. LCT Birth Weight, g Birth Weight. g Birth Weight, g I zo t/ Figure 1. Comparison of percentage of survival to 21 d as a function of birth weight within 100-g increments of birth weight (bar graphs) for pigs from sows fed A] control, BJ long-chain triglycerides (LCT], or C) medium-chain triglycerides (MCT). Lines in each panel represent a second-order regression line of the survival data, with the composite shown in D). See Table 5 for distribution of pigs in broader birth weight categories. of 30) and 68% (15 of 22) in litters from LCT and MCT fed sows, respectively (Table 5). Pigs in this weight range represented approximately 10% of the total pigs. Survival rates increased with birth weight and were not markedly different between treatments for pigs weighing > 900 g at birth. A regression line for each treatment (Figure 1 ) emphasizes the improved survival of pigs in the LCT and MCT groups. The shape of these lines is influenced by differences in survival of low birth weight pigs. Because this weight range (< 900 g) represented < 10% of the total pigs and all litters did not have pigs in this category, we chose to analyze the significance of the apparent improvement in survival by chi-square ( x2) analysis rather than by the procedures described by Cieslak et al. (1983). If it is assumed that the average survival across treatments is the expected, the probability that the treatments are different is P <.1 (x2 = 5.51; df = 2). Using a less conservative test, in which the survival rate in the control group is taken to be the expected, the survival in both the LCT ( P <.01; x2 = 7.00, df = 1) and MCT ( P <.001; x2 = 14.32, df = 1) groups were significantly different from the expected. Changes in milk composition are a potential explanation for the improved survival rates. Milk collected during the first 24 h after parturition was only marginally effected by sow diet (data not presented). Levels of C8:0, ClO:O, and C12:O fatty acids were undetectable in milk from control or LCT-fed sows, but averaged.28 f.01,.08 f.01, and.02 k.01 mg/ml in milk from sows in the MCT group. The C8:O was detected in all samples from the MCT group, whereas C1O:O and C12:O were found in 8 and 2 of the 12 sows in this treatment, respectively. There were no other significant changes due to MCT at this time. Total lipid content of d-1 milk was slightly greater in the LCT group, whether measured by ch1oroform:methano1 extraction (Control, 96; LCT, 114; MCT, 83 mg/ ml; SE = 15, NS) or by gas chromatography (Control, 65; LCT, 85; MCT, 57 mg/ml; SE = 10, P <.2). The correlation coefficient between the two methods of determining total lipid was.97 (P <.0001). The increase in total lipid in the LCT group was accounted for largely by increases in C18:2 (Control, 14.2; LCT, 23.9; MCT, 11.4 mg/ml; SE, 3.4, P <.05) and C18:3 (Control,.98; LCT, 2.77; MCT,.72 mg/ml; SE,.40, P <.05) fatty acids. Dietary induced changes in d-7 milk composition (Table 6) were greater than those in d-1 milk. In particular, feeding LCT resulted in an increase in

6 3016 AZAIN Table 6. Effect of late gestation and early lactation diet on the fatty acid profile of sow milka Item No. of sows C8:Od c1o:o c12:o C14:O C16:O C16:l C18:O C18:l C18:2 C18:3 c20:o Total, mg/ml Medium chain, % Long chain saturated, % Monounsaturated, % Polyunsaturated, % P:S ratio Total lipid, mg/ml Significance, P < Treatment Control LCT Control Control LCT~ MCT' SEM vs Fat vs MCT vs MCT ,002,0001, , ll.1,0001 NSe NS a NS NS NS NS NS NS NS NS , , , NS ,0001,0001 NS ,0001,0001 NS NS.01.1 amilk samples obtained on d 7 of lactation. blct = long-chain triglycerides (soybean oil). CMCT = medium-chain triglycerides. dresults for individual fatty acids are presented as milligrams/milliliter of milk. ens = not significant. total lipid and a shift in the ratio of polyunsaturated: saturated fatty acids (P:S ratio). In contrast, feeding MCT resulted in a decrease in milk lipid ( P <. 1) and a slight decrease in P:S ratio (not significant). Relative to the control sows, feeding MCT caused a fivefold increase (P <.0001) in content of mediumchain fatty acids in the milk. Discussion Oral administration of MCT to newborn pigs has been used as a means of improving survival (Benevenga et al., 1986, 1989; Odle et al., 1989; Lepine et al., 1989). In this application, MCT are administered as a bolus to the smaller pigs in a litter shortly after birth. Although this has been shown to improve metabolic indices that are predictive of survival (liver glycogen, fasting blood glucose), there are no actual demonstrations of improved survival. In the present studies, MCT was feed to sows during the last 3 wk of gestation at a level of approximately 10% of the diet. The total amount of MCT fed during this period was approximately 4 kg. An additional 2.3 kg was consumed during the 1st wk of lactation. The importance of the continued feeding of test diets into the 1st wk of lactation was not established. The total amount of MCT used in the present studies is considerably more than the 10 to 20 mupig volume suggested for newborn pigs (Benevenga et al., 1986, 1989; Odle et al., 1989; Lepine et al., 1989). The rationale for feeding MCT to the sow in the present study, was that given the high rate of mortality during the first few days postpartum, it may be important to improve neonatal energy stores before birth. Oral administration to the pigs at birth may be too late to save pigs that have depleted energy reserves in the farrowing process. Various nutritional and endocrine manipulations have been employed to improve neonatal survival. Manipulations of the endocrine system, such as inducing a diabetic state with insulin deficiency (Ezekwe, 1986) or insulin resistance with somatotropin (Kveragas et al., 1986; Baile et al., 1989; Harkins et al., 1989; Cromwell et al., 1992), although not of practical significance at the present time, have the potential to improve neonatal survival. The feeding of fat during late gestation and lactation has been shown to result in improved survival, particularly, in litters with low-birth weight pigs and herds with survival rates of < 80% (Seerley et al., 1974; Pettigrew, 1981). Typically, the source of fat is of plant or animal origin, which represents primarily long-chain fatty acids esterified in the form of triglycerides. Providing additional feed to the sow during the last 3 to 4 wk of gestation has also been shown to improve the number of pigs per litter and increase total litter weight at weaning (Cromwell et

7 MEDIUM-CHAIN TRIGLYCERIDES AND PIG SURVIVAL 3017 al., 1989). Fasting the sow during late gestation has also been shown to improve indices of survival such as fasting blood glucose in the neonate or carcass fat at birth (Kasser et al., 1981, 1982). As with the endocrine manipulations (diabetes or somatotropin treatment), the negative effects of fasting during late gestation on subsequent lactation and reproductive performance of the sow make it impractical. The basis for the improved survival in response to dietary fat, increased total energy or fasting is likely related to the increased nutrients (glucose, fatty acids, ketone bodies) in maternal circulation, which result in an increase in carcass fat of the pig at birth (Kasser et al., 1981). The mechanism and repeatability of this increase is not certain (Ruwe et al., 1991). In the case of dietary fat, improved survival may also be related to the increase in the level of fat in colostrum and milk of the sow that has been observed (Seerley et al., 1974, Pettigrew, 1981). Feeding ketogenic agents such as MCT or 1,3-butanediol during late gestation have been suggested as means to improve survival (Rosebrough et al., 1981; Stahly et al., 1985; Newcomb et al., 1991). The results of various studies using 1,3-butanediol during late gestation have been summarized by Hansen (199 1). With the exception of Rosebrough et al. (19811, who reported an increase of > 2 pigs per litter in response to feeding 1,3-butanediol to sows during the last 4 wk of gestation, most studies have reported improved survival in the range of.5 pig per litter (Stahly et al., 1985, 1986; Maddock et al., 1986). Because ketone bodies can readily cross the placenta and are used in the developing fetus for the synthesis of lipid and to spare glucose (Seccombe et al., 1977; Shambaugh, 19851, ketogenic agents have the potential to improve fetal energy stores at birth. The present study represents both a confirmation and an extension of earlier investigations. Previously, Rosebrough et al. (1981) examined the effects of feeding the ketogenic agents, 1,3-butanediol or MCT on neonatal metabolism, chemical composition, and survival. Beginning on d 60 of gestation, sows were fed diets containing 20% of the dietary energy as 1,3-butanediol, MCT, or a combination of the two. Fetuses were removed from a portion of the sows on each diet on d 105 of gestation. There was no effect of ketogenic diets on carcass fat of the fetuses. However, ketogenic diets increased brain weight, liver glycogen content, and liver glycogen synthetase activity. In general, 1,3-butanediol was more effective than MCT. The remaining sows in each group were allowed to farrow and litter performance was monitored. Survival was 75% in the control group and was increased to > 90% in all ketogenic diet treatments. It was suggested that maternal ketosis in response to feeding MCT and 1,3-butanediol resulted in transfer of ketone bodies to the fetus, which in turn stimulated brain growth and spared carbohydrate for glycogen synthesis, both of which likely contribute to improved survival. Although the glycogen status of the pigs in the present study was not determined, the survival benefit of MCT was confirmed. In the present studies, feeding MCT at a similar portion of energy intake (approximately 21% j but for a shorter time (23 vs 54 d) resulted in an additional 1.4 pigs per litter at weaning (Table 4). More recently, Newcomb et al. (1991) examined the effects of MCT relative to starch and LCT in a design similar to that in the present study. Beginning on d 100 of gestation, sows were fed isoenergetic test diets with 15.7% soybean oil or MCT substituted for starch. Pigs were removed before suckling and were fasted for 36 h. As in the present study (Table 31, blood glucose was increased in pigs from both LCTand MCT-fed sows. No other changes in circulating metabolites were observed. There was no effect of dietary treatment on liver glycogen or on carcass fat. Changes in fatty acid profile of colostrum were similar to those observed in the present study, with increased total lipid and increased C18:2 and C18:3 content in milk from the sows in the LCT group. Milk from MCTfed sows was similar in composition to that of the control group. It was concluded that the increase in fasting blood glucose in pigs from lipid-fed sows may provide a survival advantage (Newcomb et al., 1991). The results of the present study confirm the earlier results of Rosebrough et al. ( 198 1) on survival of pigs as well as those of Newcomb et al. (1991) on improved glucose homeostasis in response to feeding MCT during late gestation. The observation of improved survival in pigs weighing < 900 g at birth in litters from the sows fed diets with added fat is in agreement with the results of Cieslak et al. (1983). The presentation of data in Figure 1 is also similar to the previous study (Cieslak et al., 1983). However, this weight range represented < 10% of the total pigs and all litters did not have pigs in this category. Because of these factors, we chose to analyze the significance of the apparent improvement in survival by chi-square analysis rather than by the procedures described by Cieslak et al. (1983). It is suggested that the overall improvement in survival in pigs from MCT-fed sows is due largely to the doubling of the survival rate in the lighter birth weight pigs. The mechanism for the improved survival in response to feeding MCT is not clear. The effects of diet on milk composition indicate that although LCT caused changes in milk composition that may be of biological significance, those in response to MCT feeding were minimal (Table 6). For comparison, the fatty acid profile of lard, soybean oil (LCT), and MCT was determined under the same conditions described for analysis of milk. Lard has a P:S ratio of.4 and contained C16:O (23%) and C18:l (42%) as its predominant fatty acid species. Milk samples from the control sows had a fatty acid profile similar to that of lard. Soybean oil has a P:S ratio of 4.6 and contained

8 3018 AZAIN C18:l (24%) and C18:2 (54%) as its predominant fatty acids. Feeding soybean oil (LCT) to the sow shifted the fatty acid profile of the milk toward that of the soybean oil sample and increased the total fat content of the milk. Similar changes in total lipid content and fatty acid profile have been reported previously in studies in which plant oils were fed during late gestation or lactation (Seerley et al., 1974; Stahly et al., 1981; Newcomb et al., 1991). The MCT sample was composed exclusively of C6:O (2%), C8:O (63%), and C1O:O (36%) fatty acids. Dietary MCT altered milk composition slightly, particularly at d 7 of lactation. Nevertheless, the presence of C8:O and C1O:O fatty acids accounted for < 5% of the total fatty acids. This, coupled with the overall decrease in milk lipid in response to MCT ( P <.1), makes it unlikely that changes in milk composition could account for the increase in survival. An effect of the diet on the fetus during late gestation rather than on milk composition during lactation is more likely the basis for improved survival. In the present study, blood glucose was elevated in the pigs from both LCT- and MCT-fed sows. This is in agreement with the results of Newcomb et al. ( ) in fasting pigs. Liver glycogen or carcass lipid were not determined in the present study, although in the previous work there were mixed results on the effects of MCT on liver glycogen and no effect on carcass lipid (Rosebrough et al., 1981; Newcomb et al., 1991). The elevated blood glucose should be predictive of improved survival in these animals. In particular, blood samples in the present study were obtained from pigs in the lighter third of the litter and thus represent the pigs most at risk. Although there was no overall survival benefit of LCT feeding observed in this study (Table 41, there was increased blood glucose and improved survival of the pigs from this treatment group weighing < 900 g at birth (Table 5, Figure 1). The basis for the failure to detect an overall increase in survival of the pigs from the LCT group is not known. Taken together the results of the present study, as well as previous investigations (Rosebrough et al., 1981; Newcomb et al., 1991), are supportive of the potential application of MCT as a cost-effective means to improve neonatal survival. At the present time the regulatory status of MCT as a feed additive is not clear. Because MCT are derived from coconut oil it is likely that regulatory approval will not be an issue. Alternatively, coconut oil itself, which is composed of > 50% medium-chain fatty acids (i C12:O) may substitute for the more refined MCT. Implications Litters from sows fed diets containing 10% mediumchain triglycerides for the last 3 wk of gestation had decreased preweaning mortality. The overall benefit of this dietary treatment was an increase of one additional pig weaned per litter. In particular, the survival rate in pigs weighing < 900 g at birth was doubled relative to that in the sows fed a control diet. These results suggest that feeding medium-chain triglycerides or a natural oil, such as coconut oil that has a high percentage of medium-chain fatty acids, may provide a cost-effective means of improving neonatal survival. Literature Cited Allee, G. L., D. R. Romsos, G. A. Leveille, and D. H. Baker Metabolic consequences of dietary medium-chain triglycerides in the pig. Proc. SOC. Exp. Biol. Med. 139:422. Baile, C. A., M. J. Azain, F. C. Buonomo, and T. R. Kasser Effect of somatotropin treatment in sows during late gestation on birthweight and performance of pigs. J. Anim. Sci. 67(Suppl. 1):214 (Abstr.). Benevenga, M. J.. J. K. Steinman, and T. D. Crenshaw Medium-chain triglycerides-a source of energy for the newborn piglet. Proc. Georgia Nutr. Conf. pp Athens. Benevenga, N. J., J. K. Steinman-Goldsworthy, T. D. Crenshaw, and J. Odle Utilization of medium-chain triglycerides by neonatal piglets: I. Effects on milk consumption and body fuel utilization. J. Anim. Sci. 67:3331. Cieslak, D. G., V. D. Leibbrandt, and N. J. Benevenga Effect of a high fat supplement in late gestation and lactation on piglet survival and performance. J. Anim. Sci. 57:954. Cromwell, G. L., D. D. Hall, A, J. Clawson, G. E. Combs, D. A. Knabe, C. V. Maxwell, P. R. Noland, D. E. Orr, Jr., and T. J. Prince Effects of additional feed during late gestation on reproductive performance of sows: A cooperative study. J. him. Sci. 67:3. Cromwell, G. L.. T. S. Stahly, L. A. Edgerton, H. J. Monegue, T. W. Burnell, B. C. Schenck, and B. R. Schricker Recombinant porcine somatotropin for sows during late gestation and throughout lactation. J. Anim. Sci. 70:1404. England, D. C Improving sow efficiency by management to enhance opportunity for nutritional intake by neonatal piglets. J. Anim. Sci. 63:1297. Ezekwe, M Effect of Streptozotocin-induced diabetes in primiparous swine on subsequent reproductive performance. J. him. Sci. 62:1005. Folch, J., M. M. Lees, and G. H. Sloane-Stanley A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:497. Hansen, B. C Improving sow productivity: A role for 1,3-butanediol. Feedstuffs 63( 17):19. Harkins, M., R. D. Boyd, and D. E. Bauman Effect of recombinant porcine somatotropin on lactational performance and metabolite patterns in sows and growth of nursing pigs. J. Anim. Sci. 67:1997. Kasser, T. R., J. H. Gahagan, and R. J. Martin Fetal hormones and neonatal survival in response to altered maternal serum glucose and free fatty acid concentrations in pigs. J. Anim. Sci. 55:1351. Kasser, T. R., R. J. Martin, and C. E. Allen Effect of gestational alloxan diabetes and fasting on fetal lipogenesis and lipid deposition in pigs. Biol. Neonate 40:105. Kveragas, C. L., R. W. Seerley, R. J. Martin, and W. L. Vandergrift Influence of exogenous growth hormone and gestational diet on sow blood and milk characteristics and on baby pig food, body composition and performance J. Anim. Sci. 63:1877. Lepine, A. J., R. D. Boyd, J. A. Welch, and K. R. Roneker Effect of colostrum or medium-chain triglyceride supplementation on the pattern of plasma glucose, non-esterified fatty acids and survival of neonatal pigs. J. Anim. Sci. 67:983.

9 MEDIUM-CHAIN TRIGLYCERIDES AND PIG SURVIVAL 3019 Maddock, H. M., T. S. Stahly, V. C. Speer, K. L. Simkins, D. D. Johnson, G. R. Parker, and W. M. Coleman, Jr Effect of different levels of 1,3 butanediol fed prepartum to sows on baby pig survival and sow rebreeding. J. him. Sci. 63(Suppl. 1):275 (Abstr.). Mersmann, H. J Metabolic patterns in the neonatal swine. J. Anim. Sci. 38:1022. Newcomb, M. D., D. L. Harmon, J. L. Nelssen, A. J. Thulin, and G. L. Allee Effect of energy source fed to sows during late gestation on neonatal blood metabolite homeostasis, energy stores and composition. J. Anim. Sci. 69:230. Odle, J., N. J. Benevenga, and T. D. Crenshaw Utilization of medium-chain triglycerides by neonatal piglets: 11. Effects of even- and odd-chain triglyceride consumption over the first 2 days of life on blood metabolites and urinary nitrogen excretion. J. Anim. Sci. 67:3340. Pettigrew, J. E., Jr Supplemental dietary fat for peripartal sows: A Review. J. him. Sci. 53:107. Pomeroy, R. W Infertility and neonatal mortality in the sow Neonatal mortality and foetal development. J. Agric. Sci. (Camb.) 54:31. Rosebrough, R. W., N. C. Steele, and L. T. Frobish Effect of ketogenic diets in gestation on some characteristics of carbohydrate metabolism in fetal pig brain and liver. Growth 45:42. Ruwe, P. J., C. K. Wolverton, M. E. White, and T. G. Ramsay Effect of maternal fasting on fetal and placental lipid metabo- lism in swine. J. him. Sci. 69:1935. SAS SAS User s Guide: Statistics. SAS Inst. Inc., Cary, NC. Seccombe, D. W., P. G. Harding, and F. Possmayer Fetal utilization of maternally derived ketone bodies for lipogenesis in the rat. Biochem. Biophys. Acta 488:402. Seerley, R. W., T. A. Pace, C. W. Foley, and R. D. Scarth Effect of energy intake prior to parturition on milk lipids and survival rate, thermostability and carcass composition of piglets. J. Anim. Sci. 38:64. Shambaugh, G. E Ketone body metabolism in the mother and fetus. Fed. Proc Stahly, T. S., G. L. Cromwell, and H. J. Monegue Effects of prepartum administration of 1,3-butanediol to sows on growth and survival of neonatal pigs. J. h im. Sci. 61:1485. Stahly, T. S., G. L. Cromwell, and H. J. Monegue Effects of dietary additions of 1,3-butanediol or lard for sows on survival of neonatal pigs. J. Anim. Sci. 63:1156. Stahly, T. S., G. L. Cromwell, and W. S. Simpson Effects of level and source of supplemental fat in the lactation diet of sows on the performance of pigs from birth to market weight. J. Anim. Sci. 51:352. USDA National Swine Survey: MorbidityMortality and Health Management of Swine in the United States. Animal and Plant Health Inspection Service: Veterinary Services. Washington, DC.