BLOOD PARAMETERS AND APPARENT DIGESTIBILITY OF CONCENTRATE WITH RICE OIL FOR HORSES

Apparent digestibility coefficients and serum parameters were measured to evaluate the effect of supplementing feed concentrates with rice bran oil in horses. Twelve horses (6 males and 6 females) with a mean age of 18 ± 4 months old and mean live weight of 306 ± 22.6 kg were used. Treatments consisted of increasing rice bran oil concentrate levels of 0, 3.5, 7.0, 10.5, 14.0 and 17.5%, considering a daily intake of 2.25% live weight on a dry matter basis. A dietary effect of supplementation on the apparent digestibility of gross energy (y = 64.55 – 0.58x) was observed (P<0.05). No effect on the other nutrients was observed (P>0.05). Supplementation did not affect serum glucose levels (P>0.05), but cholesterol was affected (P<0.01) based on the interaction of the treatment x gender parameter in females (y = 73.92 + 2.08x). However, no effect was observed in males (P>0.05). A dietary effect on the triglyceride (y = 15.73 – 0.96x + 0.0524x2) and HDL (high-density lipoprotein) (y = 45.24 + 1.0499x) parameters was observed (P<0.01). While the use of rice bran oil does affect blood parameters associated with lipid metabolism, rice bran oil levels up to 17.5% concentrate do not negatively affect the apparent digestibility of dietary nutrients.


INTRODUCTION
Horses are classified as non-ruminant herbivores with functional cecum and colons.The anatomical and physiological features of horses enable them to effectively absorb dietary carbohydrates and lipids before the action of microbiota in the large intestine (FRAPE, 2008).
High dietary starch contents and limited digestive capacity can hinder the digestion of soluble carbohydrates in the small intestine, thereby increasing the supply of readily fermentable carbohydrates to the cecum and colon, which contributes to the onset of metabolic and digestive problems (such as colic, endotoxemia and laminitis).The use of lipids in horse diet is a feasible alternative to increase the energy density of food without correspondingly increasing the level of dry matter intake.Therefore, the addition of lipids enables a decrease in the amount of concentrate ingested (HINEY; POTTER, 1996).
The use of oils and fats in horse feed and their effects on the digestibility of fiber and its components are controversial.Some authors (BUSH et al., 2001;KRONFELD et al., 2004) have found no correlation between dietary supplementation with lipids and fiber digestibility, but other authors (JANSEN et al., 2000) found that dietary supplementation with high amounts of oil reduces the digestibility of fiber components.Kronfeld et al. (2004) studied fat digestibility in Equus caballus and found that ether extract digestibility varies among forages, grains and dietary supplementation with oil based on the levels of triglycerides.
In a study of the effect of corn oil supplementation in horse diets on nutrient digestibility, Resende Junior et al. (2004) found no significant effect of oil supplementation on dry matter, crude protein, fiber or neutral and acid detergent fiber digestibility.However, the authors found significant differences in gross energy and ether extract digestibility.
Studies of rice bran oil in horse feed are scarce.Frank et al. (2005) found that serum levels of unsaturated fatty acids and triglycerides decrease upon diet supplementation with rice bran oil in mares.Godoi et al. (2009) assessed oil supplementation in horse diets and found that dietary supplementation with 19.5% soybean oil increased serum triglyceride levels compared to a diet without lipid supplementation.
The aim of the present study was to evaluate the effect of supplementation with increasing levels of rice bran oil on the apparent digestibility coefficient of nutrients and on serum glucose, cholesterol, triglyceride and highdensity lipoprotein (HDL) levels in horses.

MATERIALS AND METHODS
The experiment was conducted at Iguatemi Experimental Farm, Marin gá St ate Un iversity (Universidade Estadual de Maringá -UEM), from January to March 2010.
Twelve crossbred (English Thoroughbreds crossed with Mangalarga) foals (6 males and 6 females) with a mean age of 18 ± 4 months old and live weight of 306 ± 22.6 kg were used.The horses were housed in individual stalls.Two animals were used per treatment -one male and one female for each treatment.
During a period of 90 days, the animals were fed the experimental diets formulated as recommended by the NATIONAL RESEARCH COUNCIL -NRC (2007) for growing foals (18 months), considering an intake of 2.25% live weight on a dry matter basis.Treatment consisted of concentrate supplementation with 0, 3.5, 7.0, 10.5, 14.0 and 17.5% rice bran oil (Tables1 and 2), with a defined 50:50 roughage (Tifton 85)-to-concentrate ratio.The diets were partitioned into three daily meals, supplied at 7 hours 30 minutes, 13 hours and 17 hours.
The animals were exercised during the morning, twice a week throughout the experiment.The exercise protocol was divided into two stages.In the first stage (a 7-week period), the animals walked for 3 minutes, trotted for 3 minutes and walked again for 3 minutes, for a total of 18 minutes of exercise with equal training on both the left and right sides.In the second stage (a 4-week period), this protocol was maintained, but the trotting time was increased by 4 minutes, for a total of 22 minutes of exercise.
For the digestibility assay, feces were collected in totality, for four days in the last experimental week.
Table 1 -Percent composition of experimental concentrates.

Ingredient
Percent inclusion of rice oil 0 3.5 Ciênc.agrotec., Lavras, v. 37, n. 5, p. 435-442, set./out., 2013 The feces were collected immediately following animal defecation and stored in properly identified plastic containers to be weighed 24 hours after collection.The samples were homogenized, and an aliquot of 10% total feces was removed, weighed and pre-dried in a convection oven at 55 ºC for 72 hours at the end of each collection day.Samples were sent for analysis after predrying.
The bromatological analyses of food and feces were performed at the Laboratory of Animal Nutrition (Laboratório de Nutrição Animal -LANA) of the Animal Science Department (Departamento de Zootecnia -DZO) of UEM according to the method suggested by Silva (2002).
Blood samples were collected (following animal adaptation to exercise) for analysis of blood parameters on the 66 th , 78 th and 90 th day after beginning the experimental protocol and before the first dairy feed of the foals.Blood was collected by venipuncture of the left jugular vein into Vacutainer tubes to measure glucose, cholesterol, triglyceride and HDL levels.A fraction of the total blood collected was kept cooled and sent for glucose and triglyceride analysis, while the remainder was immediately centrifuged for 15 minutes at 3,000 rpm, and the plasma was collected and stored at -20 °C for subsequent cholesterol and HDL analysis.
Serum parameters, including glucose, cholesterol, triglycerides and HDL, were assessed using biochemical kits and spectrophotometry.
The data from the completely randomized design used for the apparent digestibility trial were statistically analyzed using Bayesian inference at a 5% significance level, analyzed by Saeg program.
All the results were analyzed using the STATISTICAL ANALYSIS SYSTEM SOFTWARE -SAS (2001).The experimental design used for the blood parameters was a split-plot with repeated measurements over the time.The serum parameter data were assessed using a mixed model considering the animal as the random effect and the following parameters as fixed effects: time, treatment, gender and the interaction of gender x treatment, in addition to the residuals.Because the treatment is a quantitative variable, regression analyses were performed upon the identification of significant results, at 5% significance level.

RESULTS AND DISCUSSION
Supplementation with rice bran oil did not affect DM, CP, EE, NDF or ADF digestibility (P>0.05)(Table 3).Supplementation did affect GE digestibility (P<0.05).In contrast, Brandi et al. ( 2007) assessed supplementation with increasing oil levels (up to 24% in concentrate) and detected a difference in the DM apparent digestibility coefficient (P<0.05) and the digestibility observed using the equation y = 76.927+ 0.006858x + 0.00665861x 2 , with the highest digestibility found with 24% oil levels.These authors, also reported differences in the digestibility coefficients of NDF and ADF (P<0.05).represented by the equations y = 69.9201+ 0.0610912x + 0.005868x 2 and y= 6.3 + 6.21074x -0.138393x 2 , respectively, and differences in the EE digestibility coefficient (y = 78.365-0.1018x + 0.03594x 2 ) (P<0.05) between diets supplemented with increasing levels of soybean oil.These differences can be attributed to the class of animal used, type and intensity of exercise and amount and source of oil.
No effect of oil level on EE digestibility coefficients was observed in this study, unlike the findings by Brandi et al. (2007).Diets with 17.5% oil may have high dietary utilization without negatively affecting the digestibility of the fiber fraction, as reported by Jansen et al. (2000).The maintenance of a roughage-to-concentrate ratio of 50:50 may have greatly contributed to the maintenance of NDF (51.62%) and ADF (41.86%) digestibility coefficients.as previously reported by the NRC (2007).
Figure 1 shows the apparent digestibility values of digestible energy in horse diets supplemented with increasing levels of rice bran oil in the present experiment.
A negative effect of increasing oil inclusion levels was observed on the apparent digestibility of GE; the control diet displayed the highest apparent digestibility coefficient (65.71%), while the diet with the highest level of rice bran oil (17.5%) displayed the lowest apparent digestibility coefficient (53.89%).A noticeable variation between the estimated and experimental values was observed (Figure 1), suggesting that variations between animals can affect energy digestibility based on repeated evaluations of the same animals over a short time period.Additionally, the animal class (foals) and training protocol may have affected the dietary utilization of this source.Other studies that have assessed the effects of oil levels in horse diets on GE digestibility coefficients have reported results that contrast with those in the present study.Brandi et al. (2007) found a GE digestibility coefficient value of y = 7.60363 + 0.400505 x -0.177206x2, with the 12% oil level as the inflection point, when using adult horses and assessing moderate energy-demand horse diets supplemented with different levels of soybean oil.Moreover, Godoi et al. (2009) found no significant difference between treatments when assessing soybean oil levels (0, 8.5 and 19.5%) in animals submitted daily to moderate physical activity, reporting GE digestibility coefficient values of 63.0, 66.2 and 67.1%.respectively.
Dietary oil supplementation and digestibility of the lipid fraction affected the triglyceride and HDL blood levels (Table 4).
The interaction of treatment x gender was found to affect cholesterol (P<0.01)(Figure 2).Manzano et al. (1995) found significant increases in serum cholesterol levels upon diet supplementation with 0.5% soybean oil and 5.5% animal fat, but no gender effects were reported.Similar to the present study, Ropp et al. (2003) reported increased serum cholesterol levels in horses fed lipid-rich foods compared with those fed conventional foods.
According to Bertan et al. (2006), the gender effect observed in the present study may result from the process of corpus luteum (CL) formation, wherein cholesterol is the main substrate for progesterone production.Cholesterol is primarily synthesized in the liver, then transported in the form of lipoproteins to all steroidogenic tissues, including the adrenal cortex, ovarian follicles and CL.The increase in serum cholesterol levels found in the present study may also be linked to the onset of the estrous cycle in the fillies during the experiment because these animals were, on average.18 months old.
Table 4 -Regression equations of blood parameters with treatment effect and of interaction treatment x sex evaluated in equines undergoing diet containing rice oil.A treatment effect was observed for the triglyceride and HDL variables (P<0.05)(Figure 3).
A dietary effect was observed on the triglyceride parameter (P<0.01).Godoi et al. (2009) assessed soybean oil levels of 0, 8.5 and 19.5% in horse diets and found a significantly higher triglyceride value in the diet with 19.5% oil (P <0.05), with increasing values of 39.5.46.2 and 57.0 mg/dl in the respective diets.These values are higher than those found in the present study; the differences may be attributed to the animal class and source of oil used in the studies.
The values measured in the present study and those reported in the literature suggest that the performance of triglycerides is related to the lipid levels in the diets and that a still undefined maximum value of supplementation with lipids exists and can cause a significant increase in serum triglyceride levels.According to Dutra-de Oliveira and Marchini (1998), non-oxidized fatty acids form triglycerides, and excess triglycerides are stored or used by other tissues as an energy source when their synthesis exceeds the requirements for lipoprotein formation.These observations suggest a possible mechanism for the increase in triglyceride concentrations resulting from a given intake level.
A significant treatment effect was observed on HDL (P<0.01)(Figure 4).which may be explained by Marchello et al. (2000).These authors reported that increasing dietary fat increases bile production, which consequent ly increases endogenous cholest erol absorption, thereby increasing the HDL levels.The same effect was reported by Geelen (1999), who found a correlation between increased soybean oil levels and increased lipoprotein lipase (LPL) activity.The author explained that the increase in serum HDL in horses fed hyperlipidic diets resulted from increased LPL activity; this increased activity caused increased triglyceride removal from VLDL (very low-density lipoprotein), and the surface material (including cholesterol) of VLDL transferred to HDL, resulting in an increased fraction of cholesterol in HDL.
Conversely, Ribeiro et al. ( 2009) assessed different lipid sources in horse diets and found no significant difference between a control diet and diets supplemented with animal fat and vegetable oil, with values of 76.0.73.2 and 74.75 mg/dl, respectively.Similarly, Gonzaga et al. ( 2012) compared soybean and rice bran oils in stallion diets and found no significant difference in the HDL parameter with gender between treatments.
Supplementation with rice bran oil did not affect glucose levels (P>0.05), which is consistent with the literature on the subject.Oliveira et al. ( 2010) assessed the effect of rice bran and soybean oils in horse diets and failed to measure a significant difference in glucose concentrations when the animals were assessed at rest or following exercise.Marqueze Kessler and Bernadi (2001) found no significant differences in glucose levels using horse diets with 0 and 7.8% soybean oil levels; the glucose concentrations were 106.0 and 107.8 mg/dl in soybean oil-supplemented and control diets, respectively.Dutra-de-Oliveira and Marchini (1998) corroborated the  data shown in the present study, reporting that lipid intake does not significantly affect glucose or insulin blood levels, and these parameters directly correlate with carbohydrate intake.Conversely, Harkins et al. (1992) found a significant increase in serum glucose levels in horses fed corn oil-supplemented diets equivalent to 10% digestible energy.According to the authors, the results may be affected by the amount and form of supply of soybean oil and especially the period of diet intake.
indicating that the values may be related to animal adaptation to the diet.

CONCLUSION
The use of rice bran oil at levels up to 17.5% concentrate affect lipid metabolism-related blood parameters but does not negatively affect the apparent digestibility of dietary nutrients.GONZAGA, I. V. F. et al.Qualidade espermática e perfil plasmático de lipídeos em garanhões suplementados com óleo de arroz semi-refinado com alto teor de gama-orizanol.Brazilian Journal of Veterinary Research and Animal Science.São Paulo, v. 49, p. 215-220, 2012.

Figure 1 -
Figure 1 -Apparent Digestibility Coefficient of Gross Energy (GE) of diets for equines with different levels of rice oil.

Figure 2 -
Figure 2 -Effect of variable interaction diet x sex on the values of plasma cholesterol (mg / dl).

Figure 3 -
Figure 3 -Effect of diets (inclusion levels of rice oil) on the values of the concentration of triglycerides.

Figure 4 -
Figure 4 -Effect of diets (inclusion levels of rice oil) on the values of the concentration of triglycerides.

Table 2 -
Chemical composition of the Tifton Hay 85 and of the experimental concentrates.DM -Dry Matter. 2 CP -Crude Protein, 3 NDF -Neutral Detergent Fiber, 4 ADF-Acid Detergent Fiber, 5 EE -Ether extract, 2% rice bran oil in horse diets, Gonzaga et al. (2012) reported digestibility coefficient values of 64.64, 71.95, 83.37, 62.15 and 55.05 for DM, CP, EE, NDF and ADF.respectively; these values are similar to those measured in the present study. 1

Table 3 -
Apparent nutrient digestibility coefficients in diets for equines with different levels of rice oil in the treatments (% oil level in the concentrate).