Nutrient requirements of pre-weaned foals
The nutrient requirements of foals before age 4 months are not specifically addressed in the 2007 National Research Council (NRC) guidelines (Anon 2007) and have been estimated based on mare's milk composition and milk intake (Oftedal et al. 1983). There is increasing need for more research in this area because a considerable number of newborn foals are born with less than optimal joint alignment and show signs of developmental orthopaedic disease (DOD) before they are weaned. Leibsle (2005) revealed that during the first week of life straight carpal and fetlock conformations were present in only 3% and 55% of Thoroughbred foals, respectively. Rapid growth rate is one of the reported predisposing factors for certain types of DOD (Thompson et al. 1988a), although some studies failed to demonstrate clinical expression of DOD on a high-energy balanced diet (Thompson et al. 1988b). Normal foals grow from about 10% of mature bodyweight at birth to nearly 50% of mature bodyweight at weaning. These findings illustrate the importance of providing the foal with a balanced diet and serve as indirect evidence that the suckling period is just as susceptible to nutritional influences as later phases of growth.
Unique nutritional qualities of colostrum and milk
Mare colostrum is the first nutrition of the foal. Both colostrum and milk provide energy, nutrients and non-nutritive components such as immunoglobulins, cells, enzymes, hormones, IGF-I, and protective and trophic factors that play a role in immune competency, metabolism, musculoskeletal growth and disease prevention. The colostral period of mares is short and colostrum composition quickly changes within 12 h post partum.
The energy content of whole milk is reported to be quite variable at 430–590 kcal gross energy (GE)/kg (Oftedal et al. 1983; Pagan and Hintz 1986). In one study, the GE of colostrum was 1350 kcal/kg, which decreased to 590 kcal/kg by 8 days of lactation (Ullrey et al. 1966). Energy values were determined by different analytical methods (adiabatic bomb calorimetry vs. calculated) which is probably the main source of this variation. Nevertheless, the general trend is that as milk yield increases and lactation progresses, the energy density decreases.
Throughout lactation the milk's specific gravity, total solids, protein and fat decrease, while lactose concentration increases. Mare colostrum is reported to be about 25% total solids, 2.9% fat, 16.4% protein and 4.6% lactose on an as-fed (AF) basis. These concentrations tend to decrease to 10.5% total solids, 1.2% fat and 2.3% protein on Days 8–45 of lactation (Csapo et al. 1995), whereas lactose concentration tends to increase to 5.9% AF (Ullrey et al. 1966). The lipid component of mare's milk consists primarily of long-chain fatty acids (LCFA). Because there is minimal fatty acid hydrogenation by gut microbes before intestinal absorption in horses, milk fatty acid (FA) composition more closely resembles the FA composition of the feed. The polyunsaturated fatty acid (PUFA) content of mare's milk is generally 5–10 times higher than in cow's milk owing to marked PUFA hydrogenation in the cow's rumen (Marconi and Panfili 1998). The percentage of saturated (SFA), monounsaturated (MUFA) and PUFA is reported to be 48.2%, 27.8% and 18.8%, respectively (Marconi and Panfili 1998). The ratio of total unsaturated to saturated fatty acids was found to be 1:1. The origin of PUFA in milk probably stems from the mare's diet, since forages are rich in α-linolenic acid (C18 : 3 n-3) and cereal grains in linoleic acid (C18 : 2 n-6). Fatty acid composition of the milk of mammals is not only determined by the fatty acid transport from plasma to mammary cell but also by de novo synthesis of fatty acids in the mammary gland (Neville and Picciano 1997). The milk concentration eicosapentaenoic (EPA; C20 : 5; n-3) and docosahexaenoic (DHA; C22 : 6; n-3) fatty acid in mares consuming 4–5 kg cereal grains and ad libitum grass forage was found to be trace, 0.02–0.04% AF and 0.03–0.04% AF, respectively (Orlandi et al. 2003). In one study, the biological value of the protein fraction of colostrum, calculated based on amino acid composition, was higher than that of the mare's milk because colostrum is richer in essential amino acids, especially threonine (Csapo-Kiss et al. 1995). Mare's milk also contains nonprotein nitrogen such as urea, creatinine, allantoin and α-amino acids, which are believed to play a role as taste factors or substrates for milk microbes (Salimei et al. 2002). Disaccharide lactose is recognized to be the major factor controlling milk osmotic pressure and milk yield through increased water output into the milk (Doreau et al. 1992). Foals have active digestive enzyme lactase in their intestinal brush border and are accustomed to getting a large proportion of daily calories from milk sugar (Lawrence and Lawrence 2009).
The ash content of colostrum is reported to be 1.3–1.5 times higher than that of mare's milk (Ullrey et al. 1966; Csapo-Kiss et al. 1995). Calcium and phosphorus concentrations tend to reach maximum at Day 5–8 post partum and then start to decline (Ullrey et al. 1966; Csapo-Kiss et al. 1995; Grace et al. 1999). The Ca:P ratio of colostrum was 1–2.2:1, while the Ca:P ratio in milk was 1.5–3.4:1 (Ullrey et al. 1966; Csapo-Kiss et al. 1995). Zinc and copper concentrations appear to be highest in colostrum and then continually decline. Conversely, manganese concentration was found to increase to Day 5 post partum and then remained constant. In one study, the zinc and copper ratio increased from 4.8:1 in colostrum to 8.3–8.7:1 in milk (Csapo-Kiss et al. 1995).
Colostral concentrations of vitamins A, D3, K3 and C were found to be 1.4–2.6 times the levels in normal mare's milk, but there was no difference found in vitamin E concentration between colostrum and milk (Csapo et al. 1995). The major vitamin E isomer is α-tocopherol (Marconi and Panfili 1998).
Effects of mare's nutritional status and nutrition on milk composition and foal growth
Mare's milk composition and milk yield are affected by a number of factors, including stage of lactation, age, parity, breed differences, body condition score (BCS) and nutrition.
Body condition score of the mare at foaling and the mare's energy intake play a role in the growth rate of foals. Henneke et al. (1981) observed that foals of overconditioned mares had higher bodyweight gains when than foals of thin mares. These trends continued even when obese mares were underfed and thin mares were fed for weight gain during lactation (Henneke et al. 1981). Feeding mares 128% of NRC energy requirements resulted in weight gain in mares and increased weight gain in foals compared to mares fed 103% NRC energy requirements and their foals (Pagan et al. 1984). This indicates that obese or overfed mares potentially have higher milk energy output. Conversely, underfeeding thin mares during early lactation can result in weight loss in mares and may reduce foal growth (Martin-Rosset and Doreau 1980). In one study, foals of mares fed about 85% of NRC requirements had decreased weight gain (Jordan 1979). In some studies, underfeeding energy to lactating mares caused weight loss in mares but no change in foal growth. Collectively, studies indicate that BCS of the mare and energy intake can influence foal growth rate.
Leptin is a hormone that is released by adipose tissue and its plasma concentration increases with increasing adiposity of the mare. Leptin plays a role in coordinating food intake, energy expenditure and metabolic rate. It has recently been discovered that leptin is secreted into the milk of different species, which suggests that it may have an effect on growth and development of the neonate. Leptin was also measured in colostrum and transitional milk in mares of similar BCS and on similar rations (Salimei et al. 2002). The highest concentrations were found in colostrum, but leptin in milk was still measurable 96 h post partum. Leptin was associated with the fat globule fraction of the milk. Further studies are needed to determine whether milk-derived leptin has any effect on the development of the foal's GI tract or a systemic effect on the growing foal.
Diet profile can influence both milk production and composition. There are several studies that have examined the effects of protein, fat, nonstructural carbohydrates, fibre and forage intake on lactation.
Consumption of excellent quality protein is essential for adequate milk production. Underfeeding protein to mares reduces milk production and weight gain of their foals (Martin et al. 1992). Protein in the mare's ration should be provided from protein sources rich in essential amino acids, such as soybean meal, dried milk products or fish meal.
The effect of energy from starch or fibre on milk production was studied by Doreau et al. (1992) who reported that reducing the hay: high-starch concentrate ratio from 95:5 to 50:50 in mares fed ad libitum increases milk production by 10%. Compared to high-fibre intake, high-starch intake resulted in higher weight gain of mares, higher milk lactose concentration, higher milk yield, and lower milk fat and protein concentration. The high-fibre diet contributed to higher milk fat, probably due to increased production of the milk fat precursor acetic acid arising from hindgut fibre fermentation. The production of a larger volume of milk on the high-starch diet did not seem to translate to higher energy output in the milk because the growth of foals was not affected by the diet. This indicates similar milk energy output on these 2 different energy substrates. In conclusion, adding high-starch concentrate to the hay based diet of lactating mares and providing total daily calories above energy requirement can increase mare weight gain and milk yield, without increasing milk energy output and acceleration of foal growth.
While milk energy output does not appear to be different between mares fed a high-starch vs. a high-fibre diet, calories originating from fat in the mare's diet appear to increase body fat stores of the mare, fat content of the milk and milk energy output. Davidson et al. (1991) fed 2 types of isoenergetic diet containing grass hay and concentrate with no added fat (6.2% fat in total diet) and with added fat (8% fat in total diet) to lactating mares in amounts to maintain constant bodyweight and rump fat thickness. Mares consuming the higher-fat diet produced milk with a higher fat concentration but protein and total solids of the milk were not affected. Foals nursing mares fed the higher-fat diet had higher plasma lipid concentrations, tended to gain more weight and had higher rump fat thickness than foals nursing control mares. Foals of mares fed balanced high fat-fibre supplement compared to the foals of mares fed high sugar-starch isocaloric supplement had decreased radiographic bone mineral density between mean ± age 140 ± 5 days and 376 ± 5 days of age, potentially due to the formation of indigestible calcium soaps with fat in the GI tract (Hoffman et al. 1999). Therefore, although it is important to maintain the mare's body condition, feeding her a high-fat diet may increase the foal's growth rate, result in an overconditioned foal and lower bone mineral content in the foal. These factors may be involved in the multifactorial pathogenesis of DOD.
Dietary fibre and fat can positively influence the quality of colostrum. Hoffman et al. (1998) demonstrated improvement of colostral quality in pastured mares fed concentrate high in fibre (41.2% dry matter [DM] neutral detergent fibre [NDF]; beet pulp, soy hulls, oat straw), high in fat (10.4% DM fat; corn oil), and moderate in nonstructural carbohydrate (NSC) (26.5% DM NSC) compared to those fed concentrate low in fibre (15.3% DM NDF), low in fat (2.4% DM fat) and high in NSC (62.4% DM NSC; corn, molasses) (Hoffman et al. 1998). Colostrum of mares fed the high-fibre/high-fat/low-starch diet had higher protein, lower lactose, higher linoleic acid (n-6 FA) and a 4.2-fold increase in IgG concentration. This study implies that fibre and corn oil based concentrates with moderate NSC content affect the quality of colostrum and may enhance passive immunity. Interest is increasing in the health benefits of n-3 fatty acids, particularly EPA and DHA polyunsaturated fatty acids. EPA and DHA are not considered essential fatty acids to horses because they can be synthesised to some degree from the essential α-linolenic acid. However, in one study, horses fed a diet consisting of 80% alfalfa-based pellet supplemented with a source of α-linolenic acid (flaxseed oil, 10% of the pellet) and 20% of timothy grass hay had significantly increased plasma concentration of EPA but not DHA (Hansen et al. 2002). Fish oils and algal products provide a preformed source of EPA and DHA and can be incorporated into the ration. Eicosapentaenoic acid gives rise to the 3-series prostaglandins and thromboxanes and the 5-series of leucotrienes, and DHA is incorporated in large amounts in structural lipids of the developing central nervous system and retina. The relative importance or essentiality of a dietary source of preformed DHA for the growing central nervous system of the foal is uncertain. Studies show that maternal milk EPA and DHA concentration in humans, sows and dogs can be influenced by dietary intake of fish oil. Similarly, mares fed fish oil had increased plasma and milk EPA and DHA concentrations compared to control mares, and their foals also had higher plasma EPA and DHA concentrations (Stelzleni et al. 2006). Another observed effect was that colostrum from mares fed fish oil had lower IgG concentration, but milk and foal serum IgG was not affected. Studies are needed to elucidate the effects of EPA and DHA in milk on the physiological parameters of mares and foals.
Lactation requires increased dietary calcium and phosphorus intake. The peak calcium concentration in milk appears to occur around Day 8 post partum, followed by a slow decline; phosphorus concentration follows similar trend (Ullrey et al. 1966). Dietary deficiencies of calcium were shown to result in increased serum parathyroid hormone (PTH) and bone demineralisation in order to sustain calcium secretion into the milk (Martin et al. 1991). Magnesium, sodium, potassium, copper, zinc and selenium concentrations are typically higher in colostrum than in milk. Copper and zinc concentrations were shown to decrease by 72 and 42%, respectively, by Week 3 post partum, and selenium level decreased 80% by Week 1 post partum (Breedveld et al. 1987). The effect of dietary micromineral intake on milk composition appears to be minimal. A number of studies have demonstrated that milk and foal blood copper concentrations are not influenced by mare dietary copper intake (Baucus et al. 1987). Similarly to copper, dietary intake of zinc had no effect on mare's serum and milk or foal serum zinc concentrations (Breedveld et al. 1987).
Nutrient intake of suckling foals
Average daily intakes of dry matter, fat, protein, sugar and gross energy from mare's milk were measured in suckling foals and these values were proposed as guidelines for nutrient requirements of nursing foals (Oftedal et al. 1983; Martin et al. 1992). Those studies reported daily consumption of one-month-old foals with average daily gain of 1.07 kg/day between age 11 and 39 days as 240–300 kcal GE, 5.6–7.7 g fat, 11.3–11.4 g protein, 34–40.1 g lactose and 54.7–62.2 g DM/metabolic bwt (bwtkg0.75). Starting mean weight of foals at age 11 days was 59.2 kg (Oftedal et al. 1983) and 63.9 kg (Martin et al. 1992). Throughout lactation, concentrations of total solids, protein, fat and caloric density in milk decreased, while lactose concentration increased. When milk intake is expressed as a percentage of bodyweight per day, foals consumed 27, 20 and 19.3% at ages 11, 25 and 39 days, respectively (Oftedal et al. 1983). With changes in milk nutrient composition during lactation and increasing bodyweight associated with growth, foals consume an increasing volume of milk, a constant amount of total solids and gross energy, a constant or decreasing amount of fat and protein, and an increasing amount of lactose from one week until about age 2 months (Oftedal et al. 1983; Doreau et al. 1986; Martin et al. 1992).
Minerals contribute to normal bone growth and are therefore an important component of the suckling foal's diet. However, modification of the lactating mare's diet does not always influence the mineral content of colostrum or milk. Asai et al. (1995) demonstrated that feeding mares different concentrations of dietary zinc, copper and calcium did not affect concentrations of these minerals in colostrum. In addition, it was observed that colostral concentration of zinc and copper decreased with the age of mares regardless of diet. This suggests that foals of older mares could have lower zinc and copper intakes from colostrum. In contrast, supplementing copper in the diets of mares during late pregnancy had a positive impact on foal development after birth. Pearce et al. (1998) demonstrated that supplementation of pregnant mares with copper sulphate in amounts to meet or exceed their copper requirements reduced the incidence of DOD in foals, whereas direct supplementation of weanlings with copper did not. In this study, only foals whose dams were supplemented with copper had a lower incidence of physitis and a lower prevalence of articular cartilage lesions (Pearce et al. 1998).
Another study raised concern with deficient intakes of calcium, phosphorus and copper in 3–4-week-old suckling foals managed on pasture with their mares (Grace et al. 1999). Mare's milk was analysed and the daily mineral requirements of the foal determined using a factorial model approach. Based on those analyses, mare's milk provided a sufficient amount of Mg, Na, K and Zn, but only 78% Ca, 90% P and 49–89% Cu requirements. According to this study, nursing foals should have access to good quality forage or creep feed to correct milk deficits of Ca, P and Cu.
Recommendations for creep feeding
When feeding a growing horse it is important to meet nutrient and energy requirements while avoiding growth spurts and maintaining a smooth growth curve (Fig 1).
Foals receive most of their nutrition from mare's milk during the first 2 months of life. During this time, foals may nibble on the mare's concentrate and forages. Foals consume an estimated 20–25% bodyweight in milk daily during the first 5 weeks of life and 17–20% bodyweight in milk after age 5 weeks (Anon 2007). Mares produce a large volume of milk but its nutritional and energy density decreases over time. The data comparing milk composition, milk consumption and nutrient needs of the foal indicate that at about age 2 months mare's milk will no longer meet all of the foal's nutritional requirements and concentrates (creep feed) should be introduced to the foal (Schryver et al. 1986). Creep feeding should be introduced any time from age 1–2 months in nursing foals and the first week of life in orphaned foals. Creep feeding before weaning helps transition the foal from a liquid to a solid diet, reduces stress at weaning, allows for weight gain at an acceptable rate, maintains body condition after weaning, prevents a growth spurt at weaning and thus promotes adequate bone development (Coleman et al. 1999; Rezende et al. 2000). The foal's concentrate should be accessible only to the foal, not to the mare. This can be managed by feeding the foal separately from the mare or by feeding the foal's concentrate in a creep feeder, a structure made so that there is a narrow opening that allows only the foal to enter (Fig 2). The foal's concentrate should be hung in a light-coloured bucket at the foal's chest height so the foal can see it easily. The foal should be led into the creep feeder initially to teach him how to use it.
Concentrations of protein, minerals and vitamins required in creep feed are driven by the energy density of the feed. Fibre reduces and supplemental fat increases the energy density of the feed. Creep feeds containing not less than 6% crude fibre and not more than 3% crude fat (natural fat content of cereal grains) should provide at least 16% crude protein, 0.7% lysine, 0.8% calcium and 0.5% phosphorus on an as-fed basis. Feeds with lower concentrations of these nutrients and some broodmare feeds can make it hard for foals to consume enough protein, amino acids (especially lysine from a synthetic source) and minerals daily. There are currently some newer properly balanced formulations of commercial creep feeds with increased fibre and fat content. Because supplemental fat increases the energy density of the feed, at a given crude fibre concentration, high-fat feeds need higher percentages of lysine and minerals than creep feeds with no supplemental fat. Suckling foals fed poorly balanced high-energy dense creep feeds deficient in essential amino acids and minerals may consume enough energy to gain weight and become overconditioned, but may receive a nutrient supply inadequate for proper musculoskeletal development.
Creep feed should be introduced slowly and usually should be made available on a free-choice basis to allow the foal to mimic nursing behaviour and eat frequent small meals around the day. If too rapid growth and excessive body condition of the foal are of concern, the creep feed should be offered in a measured amount of 0.25–0.5 kg/100 kg bwt/day. The daily amount can be adjusted to maintain ideal body condition score of 5/9, such that the amount of concentrate can be increased to 0.33–0.66 kg or reduced to 0.12–0.25 kg concentrate per 100 kg bwt if body condition is <5/9 or >5/9, respectively.
Some reports demonstrate that nursing foals rarely drink water before weaning, whereas other reports from hot climates observe foals drinking water frequently with their dams (Martin et al. 1992). Therefore, drinking water should always be available to foals, particularly in hot climates.
Bodyweight and BCS of the foal should be assessed and recorded every 2 weeks (Fig 3). Measured bodyweight values should be compared with published growth curves for the given breed or mature size and should follow a smooth curve without growth spurts. When accurate scales are used, marked day to day variations in daily weight gain are normal in foals. In addition to scales, a weight tape calibrated for horses and ponies can be used to monitor the bodyweight of suckling foals (Fig 4). The average daily gain of the foal will decline slowly with age, but the growth curve should always be smooth (Fig 1).