1. Introduction
The determination of digestibility represents the first step for the evaluation of feed net energy; in horses, digestibility can be determined in vivo by the ingesta–excreta balance or by the marker method [
1]. In addition, feed digestibility can be estimated by chemical composition parameters [
2], by NIRS method [
3,
4] by the “in sacco” method [
5], and by the in vitro pepsin-cellulase technique [
6]. These in vitro techniques arouse particular interest because they are not expensive and are easy to perform. Among these, the cumulative gas production technique (GPT) allows us to study the fermentation kinetics and the digestibility of the organic matter in feed. GPT is based on the anaerobic degradation of carbohydrates by the micro-population of the digestive tract with the production of volatile fatty acids [
7,
8,
9], carbon dioxide, and methane [
10].
Menke and Steingass [
10] found a close correlation between gas production measured after 24 h of incubation using rumen fluid as inoculum, and in vivo digestibility. Khazaal et al. [
11] reported that, in sheep, the relationship between in vivo dry matter digestibility and volume of gas produced was very close (r = from 0.84 to 0.81;
p < 0.01) after 3 and 6 h, respectively. In horses, the GPT is performed using feces as the source of inoculum [
12].
Diet digestibility is influenced by several factors, one the most important being the forage/concentrate ratio [
13,
14], which has also been reported to affect animal health [
15], feeding behavior [
16] and healthy characteristics of foods of animal origin [
17,
18,
19,
20,
21].
In horses, the forage:concentrate ratio is strictly linked with the animals’ attitude and workload. Forage is the most important feed in a horse’s diet, often providing most of the nutrients fed to horses [
21]. Administering the right quantity and type of forage is critical, and hay can also give rise to health problems, depending on the quality and quantity of its components. Health issues can arise either when the level of forage fed is inadequate, the quality of forage is not good enough, or the digestibility is not appropriate for the life-stage or activity of the horses [
22]. Forage contributes to the overall energy and nutrient content of a horse’s ration, but also helps to maintain digestive health through its physical effect on the movement of food through the gut, as well as through the retention of fluid within the digestive tract.
In horses, forage should not be seen as a ‘filler feed’, or just something to keep a horse occupied between hard feeds, as it makes a very positive contribution to the overall ration. Making good choices with regards to forage will help to maintain digestive function. In order to provide an energy source for horses, rations often include starch rather than fiber. This can result in health issues related to the gastrointestinal tract (GIT) in the horse [
23]. In fact, forage contributes to the overall energy and nutrient content of a horse’s ration, but also helps to maintain digestive health through its physical effect on the movement of feed through the gut, as well as through the retention of fluid within the digestive tract.
By contrast, concentrates in the horse diet should only be considered as a good-quality hay supplement. In general, a mature horse does not require the energy that would be provided by concentrated feeds (cereals/sweet feeds, pellet feeds, etc.) unless the horse is used for more than light work and/or production, such as a nursing mare or a breeding stallion [
24]. Horses are more frequently overfed rather than underfed, and this is often due to an excess or an improper use of concentrates in the diet [
25]. Concentrates, however, play an important part of the growing foal’s diet through maturity, even contributing up to 50% of the ration in the first 2–3 years of growth. Thereafter, unless there are high energy and/or growth needs given current age and work level, slight increases in hay can provide the extra energy to balance dietary needs.
In order to make a contribution to this topic, this trial has been performed to study the correlations between the in vivo digestibility and in vitro degradability of five diets with different forage/concentrate ratios (F:C) in horses. The in vitro degradability was determined with GPT, using as an inoculum source the feces of the same subjects used for the in vivo test [
26].
2. Materials and Methods
2.1. Animals and Diets
Four six-year-old horses with a live weight of 500 ± 22 kg were included in the trial. Animals were kept in individual stalls to facilitate the control of feed intake and feces collection.
Five diets consisting of polyphyte hay, straw and grains of barley and oats with a different forage:concentrate ratios (F:C) of 90/10 (Diet 1); 78/22 (Diet 2); 68/32 (Diet 3); 60/40 (Diet 4); and 50/50 (Diet 5) were formulated and administered in succession from the highest to the lowest amount of forage. Diet 1 was the one administered to animals before the onset of the experiment.
2.2. Chemical Composition
The chemical composition was determined on the obtained samples according to the protocol suggested by AOAC [
26]. In particular, the ingredients of the diets were ground through a 1 mm grid with a mill (Brabender Wiley mill, Braebender OHG, Duisburg, Germany) and mixed in the same proportion present in the diets. Feces were ground with the same technique, and the organic substance content was determined [
27].
Acid-insoluble ash in diets and feces was determined by the method of Bergero et al. [
28]. Such a method allows one to determine the content of mineral substances insoluble in hydrochloric acid. Briefly, the sample is deposited in a 500 mL flask to which 100 mL of 4N hydrochloric acid are added. The flask is then brought to a boil for 30 min. The hot solution is filtered (Wathman filters No. 41), and the residue is washed with hot water until the acid reaction disappears. Subsequently, the filter is transferred into a pre-weighed porcelain capsule which, after drying, is placed in a muffle at 650 °C for the determination of the ashes, which are related to the quantity of weighed dry substance.
2.3. In Vivo Digestibility
An adaptation period of 14 days was foreseen for each diet. During this time, individual voluntary intake was evaluated in two daily meals (at 8:00 and 16:00). Therefore, a 6-day trial period started, during which each animal received 90% of the amount of dry matter previously ingested to avoid residues. Individual stool sampling (about 200 g) was performed directly from the rectum three times a day (always at the same time to reduce the effect of the variability of their composition throughout the day). The individual daily pool of feces was homogenized, an aliquot was used to determine the dry matter content at 103 °C, and another one was dried at 65 °C and used to prepare the individual pool of six test days for each horse. Similarly, for each diet, a sample was created daily to be associated with the feces pool of each animal for the determination of digestibility. The digestibility of the organic substance was evaluated with the internal indicator method, using the insoluble acid ash using the following formula:
where ADC is the apparent digestibility coefficient of organic matter and crude fiber, and Cf and Ca represent the concentration of the AIA with respect to organic matter content in feces and diet, respectively.
2.4. In Vitro Degradability
On the last day of the in vivo tests, a feces sample was taken from each animal, kept in anaerobic conditions at a temperature of 39 °C, immediately transported to the Food analysis laboratory of the Department of Veterinary Medicine and Animal Production, and used for the preparation of the inoculum for the in vitro test, using the GPT [
29,
30]. To this end, according to Macheboeuf et al. [
31], 50 g of feces was mixed with 100 mL of anaerobic buffer at 39 °C, filtered through four layers of gauze, and diluted 1:1 with the buffer, finally obtaining an inoculum for each of the four horses used for each diet. For each diet, about 1.0 g of sample was placed in a 120 mL serum bottle which, after adding 75 mL of medium and 4 mL of reducing agent, was hermetically sealed with a butyl rubber stopper and aluminum, and placed in a thermostat at 39 °C until the internal temperature was balanced. All the steps were carried out under CO
2 insufflation to maintain the anaerobiosis. Thereafter, the bottles were added with 10 mL of inoculum and, after having balanced their internal pressure with the atmospheric one, they were incubated in a thermostat at 39 °C.
For each inoculum, coming from a single animal, 3 replications were carried out to have an average value of all GPT parameters. Furthermore, two bottles were incubated without feed and used as blank. At pre-established times, with intervals of 2–24 h, 20 gas measurements were taken for each bottle using a manual system consisting of a pressure transducer (Cole and Palmer Instrument Co., Vernon Hills, IL, USA) inserting a 21 G × 1″ (0.80 × 25 mm) needle through the vial caps that were attached to the pressure transducer. Then, the transducer was removed, and the needle was inserted into the cap for a few seconds for complete stabilization between internal and external pressures. The gas pressure (psi) measured during the test was transformed into volume (mL of gas). At the end of the gas readings, the bottles were shaken to mix the suspension.
After 120 h of incubation, the bottles were opened, and the pH was determined using a pH meter (ThermoOrion 720 A+, Fort Collins, CO, USA). Subsequently, an aliquot of the liquid present in the bottle was taken to determine the volatile fatty acids (VFA) by gas chromatography (ThermoQuest mod. 8000top, FUSED SILICA capillary column 30 m × 0.25 mm × 0.25 mm film thickness) according to Formato et al. [
32]. After that, the content of each bottle was filtered through pre-weighed porous septum crucibles (Schott-Duran #2), which were placed in an oven at 103 °C and then in a muffle at 550 °C to estimate the residual organic matter; the degraded organic matter (dMO) was calculated by the difference between the incubated one and the residual one, corrected for the blank. The total gas production (corrected for the blank) was related to the incubated organic matter (OMCV, mL/g) and the degraded organic substance (Yield, mL/g).
2.5. Statistical Analyses
For each bottle, the cumulative volumes of gas obtained were related to the incubated organic matter and processed with the one-phase Michaelis–Menten model modified by Groot et al. (1996) [
33]:
where G represents the quantity of gas (mL/d) produced at time t; A is the potential gas production (mL/g); B is the time (h) necessary to produce a quantity of gas equal to A/2; and C is a constant that defines the shape of the curve.
All data relating to in vivo digestibility and GPT parameters were processed by ANOVA using the General Linear Model (GLM) procedure, including the group effect as a fixed effect and the month of sampling as a repeated measure. The differences between means were evaluated with the T-test.
Furthermore, to evaluate the relationships between in vivo and in vitro results, the correlation and possible regressions (CORR and REG procedures, respectively, of SAS, 2000) between the digestibility coefficients and the GPT parameters (dOM, OMCV, A, B, Yield, VFA) were determined.