Serviços customizados
texto em 
Inglês (pdf)
Artigo em XML
Referências do artigo
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
The existing native nuclei have a high degree of genetic diversity, which indicates their great genetic potential for selection (Borge et al. 2019) and, therefore, their ability to survive under very precarious and inhospitable conditions (Valderrama 2003 and Lesage-Padilla et al. 2019). These nuclei should be conserved as an alternative for crossbreeding with foreign breeds for meat and milk production (Restrepo et al. 2016).
It is expected that as a consequence of climate change, the large area occupied by ecosystems with adverse environments for animal production will increase (Eugenia et al. 2018). This constitutes an opportunity to develop food production systems with the use of own zoogenetic resources for the production in the Amazonian piedmont (Vera and Riera 2004). Currently, these systems are based on extensive regimes, with a mean milk production of 3.3 L cow-1d-1, which can be increased with the implementation of a livestock reconversion model that uses local trees and shrubs. In this way, it would contribute to the conservation of the Amazon and to the subsistence of livestock sector (Fedegán 2016).
The particular conditions of soils in the Colombian Amazon mean that the soil-plant-animal relationship is limited (Forero et al. 2018), especially when it is necessary to meet the minimum requirements of bovines to ensure that productive indicators and body condition remain appropriate and do not interfere with the fertility of bovine females (Calderón et al. 2017). In the same way, the Amazon has a diverse inventory of local species susceptible of use. In this context, to study the effects of supplementation with the inclusion of these species for feeding creole cattle is an alternative to improve the nutritional status of animals (Kubovičová et al. 2013). Nutritional imbalances in terms of energy, protein and minerals have been confirmed in other breeds, caused by irrational management practices of introduced pastures. This has provoked pastures with low yields of dry matter, which can condition forage intake by the animal and favor the nutritional deficit (Gallego-Castro and Mahecha-Ledesma 2017 and Sotelo et al. 2017).
One of the practices to evaluate the nutritional status of animals is the implementation of blood metabolic profile. Through this test, it is possible to study and demonstrate the response of each animal to a particular diet, as well as the possible problems of metabolic origin that are or may be present (Sena et al. 2018) without the farmer being aware of it. Although the animals may appear to be in good health, the metabolic profile is a useful tool to consider changes that should take place in feeding management (Omidi et al. 2018).
From this perspective, the objective of this study was to evaluate the effect of forage supplementation on blood metabolic indicators of Hartón del Valle heifers in the Amazonian piedmont.
Materials and Methods
Location and study area. The research was carried out in Villa Lucero farm, at 0°35'25.6"N and 76°32'05.3"W, Putumayo department, in the southwest of the Republic of Colombia. This region is located at an altitude of 256 m.a.s.l. It has an average temperature of 25.3 ºC, with 85% of relative humidity and annual precipitation of 3,355 mm (IDEAM 2017). These characteristics correspond to a tropical humid forest life zone (Holdridge 1982). Its soils are clayey-loam and clayey, acid (pH 4.6), low in phosphorus (<1.7 mg kg-1) and with high aluminum contents (>3.2 cmol kg-1) and iron (Landínez-Torres 2017). During the sampling period, in the experimental stage, soil received no irrigation and plants were not fertilized.
Trichantera gigantea (Acanthaceae) and Piptocoma discolor (Asteraceae) species were used as forage, the latter native to the Amazon. They were harvested from a forage bank established on the farm with one-year-old plants, which underwent agronomic cutting management, and insect and weed control. An establishment cut was carried out and the regrowth of plants was used at 60 d. From each plot, 15 plants were taken in zigzag as a sample. Their leaves and young stems were used from 20 cm from the soil in the summer season, in June (Fick et al. 1976). Samples of fresh material were dried in an oven at 60°C for 48 hours and ground to reach 1 mm in a hammer mill. Later, three samples of 200 g were taken for analysis. Preparation, drying and weighing was carried out in the biotechnology, soil and water laboratory of SENA, Putumayo regional.
Animals. According to their genotypic evaluation, 12 Hartón del Valle pure cattle were selected, with an average weight of 340 ± 20 kg live weight. Body weights were recorded according to those estimated from the measurement of the thoracic perimeter with a tape measure, according to the methodology proposed by Álvarez (1997). For sanitary management, they were dewormed every three months and an annual vaccination plan was carried out to control reproductive diseases (IBR, DVB, and brucellosis), and others of mandatory registration in Colombia (foot-and-mouth disease and rabies).
Feeding. For the forage base diet, 6.0 ha of Brachiaria decumbens pasture were used, divided into 14 paddocks of 4.285 m2. The pasture was divided with electric fence according to average intake of 15% of live weight, according to capacity data, with a production of 0.450 kg/m2. Each paddock was divided in a uniform manner, according to the four-day occupation period, so that offered forage had the same regrowth age, of 56 d of rest, typical for this area. They underwent an adaptation period of 15 d. During the study, the animals remained in the assigned paddocks from 11:00 a.m. until 7:00 a.m. of the next day. From 7:00 a.m. to 11:00 a.m., they were taken to individual pens, in which they were offered the supplement in equal parts once a day, at 8:00 a.m.
The animals were fed a B. decumbens grass base diet and were supplemented at a rate of 1.8 kg animal-1 d-1. The proportion of raw materials of the supplement are shown in table 1. The daily diet allowed animals to meet their requirements and obtain a body weight gain in the order of 0.5 kg animal day-1 (Ruiz and Menchaca 1990).
Table 1 Proportion of ingredients in supplements
| Ingredients | 20 % T. gigantea | 20 % P. discolor |
|---|---|---|
| T. gigantea | 20.00 | |
| P. discolor | 20.00 | |
| Corn meal | 35.03 | 43.00 |
| Soy bean cake | 0.84 | 0.20 |
| Wheat bran | 37.12 | 30.30 |
| Palm oil | 1.00 | 0.50 |
| Molasses | 5.00 | 5.00 |
| Microminerals * | 1.00 | 1.00 |
* Content micromineral mix: magnesium 10 %, zinc 10 %, iron 10 %, copper 2 %, iodine 0.12 %, selenium 0.06 %, cobalt 0.02 %
Treatments. T1 control (conventional supplement), T2 (2 kg of supplement with 20% of inclusion of T. gigantea) and T3 (2 kg of supplement with 20% of inclusion of P. discolor).
Chemical analysis. Samples per species were weighed and subsequently dried in an oven at 60 °C for 48 h and ground in a hammer mill until reaching 1 mm. Later, three samples (200 g sample-1) were taken for laboratory analysis. Preparation, drying and weighing of samples was carried out in the biotechnology, soil and water laboratory of SENA, Putumayo regional.
Proximal chemical analyzes were carried out in the laboratory of AGROSAVIA (Cundinamarca). They were calculated according to procedures and recommendations established by AOAC (2016): humidity content (method 930.04), crude protein (CP) according to Kjeldahl (N 6.25) (method 955.04), ashes by calcination at 6,000 (method 930.05), ether extract (EE) (method 962.09) and crude fiber (CF) (method 920.39). The metabolizable energy (ME) was determined from the percent of total digestible nutrients (TDN) (Yglesias et al. 2015) and NDF and FDA were determined by the method of Goering and van Soest (1970) (table 2).
Table 2 Chemical composition of treatments and forage plants
| Ingredients | Control | 20 % of T. gigantea | 20 % of P. discolor | AB* | P. discolor | T. gigantea |
|---|---|---|---|---|---|---|
| Dry matter, % | 88.7 | 89.08 | 89.81 | 27.99 | 28.86 | 24.23 |
| Ashes % | 3.13 | 3.70 | 3.80 | 7.31 | 8.17 | 10.03 |
| Ether extract, % | 4.96 | 6.44 | 6.98 | 1.95 | 3.95 | 2.78 |
| Crude protein, % | 11.53 | 11.50 | 11.78 | 6.33 | 21.51 | 19.00 |
| Crude fiber, % | 3.32 | 2.95 | 4.38 | 33.97 | 8.63 | 10.94 |
| Metabolizable energyMJ/kgDM | 11.79 | 11.50 | 11.76 | 6.63 | 10.26 | 9.63 |
* base food with Brachiaria decumbens forage
Food balance was calculated in both treatments using CALRAC® computer program, version 1.0 of 1996, developed by the Institute of Animal Science (ICA) of the Republic of Cuba. They received an isoenergetic and isoprotein balanced to meet the requirements of heifers according to their weight (Trujillo and Pedroso 1989).
Blood sampling. Samples were collected by a professional in veterinary medicine from the University of Nariño, from the private laboratory, certified by the Instituto Colombiano Agropecuario (ICA). Scientific and administrative techniques for research in animals were considered for sample collection, management and conservation procedures (Jagos et al. 1982). A sample was taken from each animal (four per treatment), at the beginning of supplementation and, later, every 15 d (days 0, 15 and 30) for a total of 36 samples. Sampling was always carried out in the morning, between 6:00 and 7:00 a.m., during fasting. Samples were collected from the coccygeal vein in Vacutainer tubes, without anticoagulant. Once collected, they were taken to the laboratory, where they were centrifuged (3500 rpm) for serum extraction, which was divided into aliquots and frozen (-20°C).
Response variables and equipment. The metabolites of urea nitrogen, creatinine, cholesterol, triglycerides, magnesium, calcium and phosphorus were determined for analysis by blood chemistry tests. A BA88A vet MINDRAY® semi-automated equipment was used, and for the Se and Cu tests, one of the Termo Scientific® brand was used. Atomic absorption method was applied. Biochemical tests were carried out immediately after the collection with blood glucose determination using the automated digital Glucometer system (Bayer®, Germany).
Statistic analysis. For data processing, the theoretical assumptions of analysis of variance were tested. The normality of errors was determined by Shapiro and Wilk (1965) test. Pearson correlation analysis and Mauchly (1940) sphericity test were applied.
For urea, triglycerides, magnesium, selenium, copper and phosphorus, these assumptions were fulfilled, so a linear mixed model was used, with repeated measures over time, with the help of Proc. MIXED. However, for creatinine, cholesterol, calcium and glucose, these assumptions were not fulfilled and a mixed generalized linear model was applied, with Proc. GLIMMIX. Treatments, samples and the interaction treatment per sampling were considered as fixed effects, and the intercept as random effect. The methodology proposed by Gómez et al. (2019) was used for data analysis. Toeplitz (Toep) structure was fitted for all variables. For those that did not meet the assumptions, the distribution of best fit to data was Gamma, with a link function (log). To compare means, Tuckey-Kramer (Kramer 1956) fixed range test was used for P <0.05. For data processing, the statistical package SAS (2013), version 9.3, was used.
Results and Discussion
The analysis of blood metabolites showed interaction (P = 0.0009) among treatments and sampling days (table 3). In the case of cholesterol, values were within the normal range for this species (2.07 to 3.11 mmol L-1) (Kaneko et al. 2008). Results indicated that, at 30 d of supplementation, values of control and 20% of T. gigantea treatments increased by 1.23 and 1.20 times, respectively, when taking the begining as a reference. Meanwhile, the treatment with 20% of P. discolor did not differ between periods. This treatment decreased its concentration by 19.01% with respect to control, at 30 d of supplementation.
Table 3 Cholesterol levels in Hartón del Valle bovines, with different supplements and sampling periods
| Variables | Treatments | SE and Sign. | |||
|---|---|---|---|---|---|
| Days | Control | 20 % of T. gigantea | 20 % of P. discolor | ||
| Cholesterol | 0 | 1.14 bcd (3.13) | 1.08 cd (2.93) | 1.24 abc (3.45) | ± 0.0374 P = 0.0009 |
| 15 | 1.17 bcd (3.21) | 1.05 d (2.85) | 1.13 bcd (3.10) | ||
| 30 | 1.35 a (3.84) | 1.25 ab (3.51) | 1.14 bcd (3.11) | ||
a,b,c,d: different letters indicate significant differences for P < 0.05
( ) Means adjusted by link function
Results are comparable with the reports of the evaluation of the development of cattle for meat production, in which cholesterol levels of 1.55 - 2.85 mmol L-1 are indicated (Cordeiro et al. 2015) in animals with a mean weight of 405.1 kg, supplemented with glycerol.
Wehrman et al. (1991) found that progesterone synthesis (P4) improves in female cattle after supplementing them with fat for 30 d, which is associated with an increase of blood cholesterol concentration. In this regard, results of the present study could be related to the process of supplementation with forage resources, which can keep cholesterol concentrations stable.
For Hawkins et al. (1995), cholesterol is a metabolite that is directly related to ovarian function (mandatory precursor of progesterone). Therefore, greater or lesser changes in its concentrations in blood or follicular fluid, could regulate the biosynthesis of this hormone. Similarly, intake decrease influence on the low glucose production and affects cholesterol synthesis and the production of estrogens, since there will be no energy surpluses that can synthesize the precursor of the steroidal hormone (Campos and Hernández 2008).
Hernández et al. (2019) reported that higher values can affect homeostasis of the animal, which is explained by an increase of energy. Fatty acids are transformed into acetyl-CoA, and the liver turns them into ketone bodies that some peripheral tissues (brain) use as an energy source. According to Suksombat et al. (2017), excessive amounts of ketone bodies produce toxicity, and can undergo condensation in the liver and become β-hydroxy-β-methyl-glutaryl-CoA, which serves as a source of mevalonate (precursor of cholesterol formed in excess). This explains the results of this research, probably due to the energy imbalance that occurs at the beginning of supplementation.
Results for the rest of metabolites (urea, creatinine, glucose, triglycerides, calcium, phosphorus, copper, selenium and magnesium) showed no interaction. When analyzing the three used supplements, no differences were observed in their concentration (table 4). This is positive, as there are no marked effects on metabolites from the supply of the supplements.
Table 4 Metabolic indicators in Hartón del Valle bovines (mmol L-1), supplemented with 20 % of T. gigantea and P. discolor
| Variables | Treatments | SE and Sign. | ||
|---|---|---|---|---|
| Control | 20 % of T. gigantea | 20 % of P. discolor | ||
| Urea | 5.76 | 5.48 | 5.25 | ± 0.2378 P = 0.3275 |
| Creatinine | 4.90 (133.79) | 4.84 (126.74) | 4.90 (134.67) | ± 0.0425 P = 0.5627 |
| Glucose | 1.33 (3.79) | 1.32 (3.74) | 1.32 (3.75) | ± 0.0422 P = 0.9782 |
| Triglycerides | 0.70 | 0.71 | 0.70 | ± 0.0272 P = 0.9482 |
| Calcium | 1.04 (2.82) | 0.99 (2.70) | 1.02 (2.78) | ± 0.0343 P = 0.6799 |
| Phosphorous | 1.66 | 1.51 | 1.50 | ± 0.1058 P = 0.5265 |
| Copper (µmol L-1) | 8.50 | 9.41 | 8.78 | ± 0.3524 P = 0.1954 |
| Selenium (µmol L-1) | 0.77 | 0.75 | 0.74 | ± 0.0198 P = 0.5239 |
| Magnesium | 0.78 | 0.81 | 0.86 | ± 0.0599 P = 0.5977 |
() Means adjusted by link function
Blood urea values ranged between 5.25 and 5.76 mmol L-1, comparable with those reported by Xuan et al. (2018), who evaluated different protein contents in the diet and referred 4.46 mmol L-1 in Bos indicus cattle, with a body weight of 178 ± 12.5 kg.
Wang et al. (2018), in B. taurus cattle, with a conventional diet and weight of 356.4±2.6kg, tested a diet with corn silage that provided 10.66 MJ kg-1 and 12.2% of CP. These authors obtained 4.35 mmol L-1 of urea, inferior value to that achieved in this study, when supplementing with 20% of P. discolor.
The creatinine content was among the reference values indicated by Campos et al. (2012) for the species (70.80 to 176.99 mmol L-1). There was a decrease of 27.17% at 15 and 30 d with respect to the beginning (P <0.0001).
Creatinine is an indicator of kidney function and muscle catabolism (Zanferari et al. 2015). The found values were similar to those reported by Kim et al. (2018), who reported 82.30 mmol L-1 in B. taurus cattle that consumed a diet with 17.2% of CP. Likewise, they were similar to the 97.35 mmol L-1 registered in Holstein cows with 256kg of body weight, which consumed a whole diet of corn silage and ryegrass with 13.4% of CP.
Regarding glucose, values were between 3.79 and 3.74 mmol L-1. This interval is in the range of 2.5-4.16 mmol L-1 indicated by Campos et al. (2004) for B. taurus breed. Figures of triglycerides are also in the reference range (from 0 to 1.6 mmol L-1), according to Roa et al. (2017).
Mean values of studied minerals were within the normal ranges for the species: 2.42 to 3.09 mmol L-1 for Ca (Kaneko et al. 2008), 1.83 ± 0.50 mmol L-1 for P (Cedeño et al. 2011) and 9.45> µg dL-1 for Cu (Rosa 2015).
When analyzing the sampling periods (0, 15 and 30 d), differences were registered in the concentration of metabolites urea, creatinine, glucose, triglycerides, calcium, copper, selenium and magnesium (table 5).
Table 5 Metabolic indicators in Hartón del Valle bovines, according to supplementation time (mmol L-1)
| Variables | Days | SE and Sign. | ||
|---|---|---|---|---|
| 0 | 15 | 30 | ||
| Urea | 7.54 a | 5.12 b | 3.83 c | ± 0.2378 P < 0.0001 |
| Creatinine | 5.09 a (163.02) | 4.77 b (118.47) | 4.77 b (118.23) | ± 0.0294 P < 0.0001 |
| Glucose | 1.25 b (3.48) | 1.26 b (3.54) | 1.46 a (4.32) | ± 0.0418 P = 0.0025 |
| Triglycerides | 0.81 a | 0.62 b | 0.69 b | ± 0.0272 P = 0.0001 |
| Calcium | 1.21 a (3.37) | 0.90 b (2.46) | 0.94 b (2.55) | ± 0.0330 P < 0.0001 |
| Phosphorous | 1.72 | 1.51 | 1.45 | ± 0.1058 P = 0.2134 |
| Copper (µmol L-1) | 9.33 a | 9.53 a | 7.83 b | ± 0.3524 P = 0.0038 |
| Selenium (µmol L-1) | 0.65 b | 0.82 a | 0.78 a | ± 0.0198 P < 0.0001 |
| Magnesium | 0.93 a | 0.58 b | 0.94 a | ± 0.0599 P = 0.0032 |
a.b.c: different letters indicate significant differences for P < 0.05
( ) Means adjusted by link function
Regarding the results obtained in the metabolites that were influenced by sampling time, it could be appreciated that, at 30 d, urea concentration decreased (P <0.0001) by 39.5% on average, with respect to 0 and 15 d (table 5). The value of urea is within the reference figures (3.83 - 9.66 mmol L-1) cited by Guerra et al. (2018).
Urea level at the beginning was high in all treatments. However, it began to decrease (P<0.0001) from day 15 in 32.2% with respect to the beginning, and the lowest value was shown at 30 d, with 25.2 and 49.2% less compared to the 15 d after the beginning, respectively. It is possible that there was no adequate protein/energy relation in the base diet, since animals were grazing with B. decumbens, which reported a low energy value (6.36 MJ kg DM-1). This causes ammonia levels to increase in the rumen due to low energy availability and, therefore, blood urea levels increase (Portilla et al. 2019). Likewise, diets with low levels of non-protein nitrogen and degradable protein in the rumen show reduced digestibility, which decreases the flow of microbial protein (Carrillo et al. 2018).
According to Sun et al. (2018), the asynchronism in the fermentation of nitrogen and energy sources results in increased absorption of ruminal ammonia in the bloodstream and its conversion to urea in the liver. This probably caused a high biological value at the beginning of this experiment. Hepatic saturation of excess ruminal ammonia requires a caloric expenditure for ruminants of 0.2 Mcal NL/100 g (Blanchard 1990).
When comparing the periods evaluated for glucose (table 5), differences were found with an increase (P=0.0025) in 1.22 times, on day 30 of starting the supplementation, in relation to the 15 days of initiation, which did not differ from each other. On day 30, it was 3.54, superior to the 3.28 mmol L-1 reported by Akbarian-Tefaghi et al. (2018) in cattle with 229.9 kg of body weight, fed with native forages, with 12.0% of CP and 72.0% of TDN, which is lower than the 4.75 mmol L-1 indicated by Amirifard et al. (2016). These authors studied essential oil supplements in a diet containing 19.6% of urea and 23 g kg-1 of thyme essential oil. Likewise, they are similar to the 3.73 mmol L-1 obtained in prepartum cows, supplemented in the transition period with a diet with 14.6% of CP and 1.55 ENL (Mcal kg-1) (Kekana et al. 2018).
The results found for glucose in this study, favored by the treatment with P. discolo, indicate that it is essential for production and release of GnRH at hypothalamic level. It has been demonstrated that with the inhibition of glycolysis at neuronal level, the frequency of LH pulses is lower, due to the inhibition of GnRH synthesis and levels of IGF-1 (Meléndez and Bartolomé 2017).
The value of triglycerides decreased (P=0.0001) in 19.14% on average, at 15 and 30 d, compared to the beginning. This could be corroborated in the treatment with T. gigantea on day 30. The found results are similar to those reported by Nemati et al. (2015), with triglyceride values of 0.21 to 0.30 mmol L-1, when measuring the effect of the diet with 19.2% of CP, 2.82 of ME (Mcal kg-1) and 12.5% of alfalfa (Medicago sativa). They are also similar to those of Roa et al. (2017), who indicated 0.68 mmol L-1 in bovines supplemented with 3.5 kg of Cratylia argentea, and the 0.52 mmol L-1 reported by Galvis et al. (2017) in a supplementation with 16.8% of CP and EE of 5.72% for cows before parturition.
In this experiment, the values were higher at the beginning of supplementation. It is possible that the increase of nutrients as energy sources caused the mobilization of triglycerides, due to fat storage in the animal, with a decrease of values (Ntallaris et al. 2017).
The values found for the microminerals copper and selenium and magnesium were influenced by the contribution of forage, according to their chemical composition, and by the addition of 1% of vitamin and mineral premix in the supplement, which could explain the obtained results.
In the case of minerals like phosphorus, no variation was demonstrated with the supplementation time. Calcium decreased 25.7% (P <0.0001) on average, in days 15 and 30, which did not differ from each other compared to the beginning. Copper did so by 17%, on days 15 and 30, which did differ. Normocalcemia is related to the adequate entry of the cation and proper functioning of the homeostatic mechanism. In a study carried out with supplementation of 9.7% of CP and 58.2% of TDN, Ca values of 2.61 mmol L-1 were announced (Asano et al. 2017).
Results for copper were below those reported by Campos et al. (2012) in crossbred Taurus heifers, with values of 9.38 mmol L-1. Therefore, on days 0 and 15, the values were in the range of normocupremia, which is superior to 9.45 mmol L-1 (Fazzio 2010). It is important to state that, although no differences were found among treatments, this may indicate that there was a slight copper deficiency.
The previous elements could explain copper insufficiency because the effect of protein in diets accelerates the action of ruminal microorganisms, which originates the formation of sulfates. They are bonded to the iron of soils, which is abundant in this area of the Amazonian piedmont, and forms the iron sulfure, which, according to reports of Arthington and Brown (2005), decreases absorption of copper at ruminal level.
This element has been indicated to be related to synthesis and secretion of gonadotropins, by modulating the ability to release luteinizing hormone (Garrik et al. 2003 and Michaluk and Kochman 2007). In this way, nutrition strategies with oligoelements for growing heifers can ultimately determine yield during their productive and reproductive performance (Yatoo et al. 2013). This last aspect must be taken into account to include this mineral in supplementation programs and incorporate tree systems to improve soil conditions of the Amazon (Ramírez-Iglesias et al. 2020), which justifies the low levels found in this study.
Selenium values increased, on average, by 18.75%, at 15 and 30 d compared to the begining. In all cases, they were in the range of reference values reported by Ceballos et al. (1999), between 0.06 and 1.24 µmol L-1. Regarding magnesium, a decrease of 37.97% was shown on day 15, in relation to the values of the beginning and day 30, which did not differ from each other, and are included in the reference values, from 0.74 to 0.95 mmol L-1 (Kaneko et al. 2008).
These results could be compared with those reported by Kang et al. (2017), who declared values from 0.82 to 1.03 mmol L-1 of Mg, when supplementing for 90 days before service, which causes higher pregnancy rates, according to Irabuena et al. (2016). Indeed, the values found for this mineral may lead to a decrease of fertility in female bovines due to the effect of embryonic mortality in the first third of gestation (Abbott et al. 2019).
Conclusions
The substitution of 20% of the conventional supplement for P. discolor or T. gigantea did not affect the studied blood indicators. However, the increase of supplementation over time improved blood indicators of triglycerides, urea, creatinine and selenium. Thus, supplementation with 20% of inclusion of P. discolor mantained stable cholesterol levels and improved glucose levels in the 30 d of supplementation. It is a viable option to keep a nutritional status in Harton del Valle heifers, under grazing conditions in the Amazon piedemont.
