Evaluation of predicted ration nutritional values by NRC (2001) and INRA (2018) feed evaluation systems, and implications for the prediction of milk response.

Net energy and protein systems (hereafter called feed evaluation systems) offer the possibility to formulate rations by matching feed values (e.g., net energy and metabolizable protein) with animal requirements. The accuracy and precision of this approach relies heavily on the quantification of various animal digestive and metabolic responses to dietary changes. Therefore, the aims of the current study were, first, to evaluate the predicted responses to dietary changes of total-tract digestibility (including organic matter, crude protein, and neutral detergent fiber) and nitrogen (N) flows at the duodenum (including microbial N and undigested feed N together with endogenous N) against measurements from published studies by 2 different feed evaluation systems. These feed evaluation systems were the recently updated Institut National de la Recherche Agronomique (INRA, 2018) and the older, yet widely used, National Research Council (NRC, 2001) system. The second objective was to estimate the accuracy and precision of predicting milk yield responses based on values of net energy (NEL) and metabolizable protein (MP) supply predicted by the 2 feed evaluation systems. For this, published studies, with experimentally induced changes in either NEL or MP content, were used to calibrate the relationship of NEL and MP supply, with milk component yields. Based on the slope, root mean square prediction error, and concordance correlation coefficient (CCC), the results obtained show that total nonammonia nitrogen flow at the duodenum was predicted with similar accuracy and precision, but considerably better prediction was achieved when the INRA model was used to predict organic matter and neutral detergent fiber digestibility responses. The average NEL and MP content predicted by both models was similar, but NEL and MP content of individual diets differed substantially between both models as indicated by determination coefficients of 0.45 (NEL content) and 0.50 (MP content). Despite these differences, this work shows that when response equations are calibrated with NEL and MP values either from the INRA model or from the NRC model, the accuracy and precision (slope, root mean square prediction error, and CCC) of the predicted milk component yields responses is similar between the models. The lowest accuracy and precision were observed for milk fat yield response, with CCC values in the range of 0.37 to 0.40, compared with milk lactose and protein yields responses for which CCC values were in the range of 0.75 to 0.81.

Review: Use and misuse of meta-analysis in Animal Science.

In animal sciences, the number of published meta-analyses is increasing at a rate of 15% per year. This current review focuses on the good practices and the potential pitfalls in the conduct of meta-analyses in animal sciences, nutrition in particular. Once the study objectives have been defined, several key phases must be considered when doing a meta-analysis. First, as a principle of traceability, criteria used to select or discard publications should be clearly stated in a way that one could reproduce the final selection of data. Then, the coding phase, aiming to isolate specific experimental factors for an accurate graphical and statistical interpretation of the database, is discussed. Following this step, the study of the levels of independence of factors and of the degree of data balance of the meta-design represents an essential phase to ensure the validity of statistical processing. The consideration of the study effect as fixed or random must next be considered. It appears based on several examples that this choice does not generally have any influence on the conclusions of a meta-analysis when the number of experiments is sufficient.

Net hepatic release of glucose from precursor supply in ruminants: a meta-analysis.

For their glucose supply, ruminants are highly dependent on the endogenous synthesis in the liver, but despite the numerous studies that evaluated hepatic glucose production, very few simultaneously measured hepatic glucose production and uptake of all precursors. As a result, the variability of precursor conversion into glucose in the liver is not known. The present study aimed at investigating by meta-analysis the relationships between hepatic glucose net release and uptake of precursors. We used the FLuxes of nutrients across Organs and tissues in Ruminant Animals database, which gathers international results on net nutrient fluxes at splanchnic level measured in catheterized animals. Response equations were developed for intakes up to 41 g DM intake/kg BW per day of diets varying from 0 to 100 g of concentrate/100 g DM in the absence of additives. The net hepatic uptake of propionate, α-amino-N and l-lactate was linearly and better related to their net portal appearance (NPA) than to their afferent hepatic flux. Blood flow data were corrected for lack of deacetylation of the para-aminohippuric acid, and this correction was shown to impact the response equations. To develop response equations between the availability of precursors (portal appearance and hepatic uptake) and net glucose hepatic release, missing data on precursor fluxes were predicted from dietary characteristics using previously developed response equations. Net hepatic release of glucose was curvilinearly related to hepatic supply and uptake of the sum of precursors, suggesting a lower conversion rate of precursors at high precursor supply. Factors of variation were explored for the linear portion of this relationship, which applied to NPA of precursors ranging from 0.99 to 9.60 mmol C/kg BW per h. Hepatic release of glucose was shown to be reduced by the portal absorption of glucose from diets containing bypass starch and to be increased by an increased uptake of ß-hydroxybutyrate indicative of higher body tissue mobilization. These relationships were affected by the physiological status of the animals. In conclusion, we established equations that quantify the net release of glucose by the liver from the net availability of precursors. They provide a quantitative overview of factors regulating hepatic glucose synthesis in ruminants. These equations can be linked with the predictions of portal absorption of nutrients from intake and dietary characteristics, and provide indications of glucose synthesis from dietary characteristics.

Repeatability of traits for characterizing feed intake patterns in dairy goats: a basis for phenotyping in the precision farming context.

In ruminants, feeding behaviour variables are parameters involved in feed efficiency that show variation among individuals. This study aimed to evaluate during the first two production cycles in ruminants the repeatability of feed intake pattern, which is an important aspect of feeding behaviour. Thirty-five dairy goats from Alpine or Saanen breeds were housed in individual pens at four periods (end of first gestation, middle of first and second lactations and middle of second gestation which is also the end of first lactation) and fed a total mixed ration (TMR) ad libitum. Individual cumulative dry matter intake (DMI) was automatically measured every 2 min during the last 4 days of each period. Feed intake pattern was characterized by several measures related to the quantity of feed eaten or to the rate of intake during the 15 h following the afternoon feed delivery. Two main methods were used: modelling cumulative DMI evolution by an exponential model or by a segmentation-clustering method. The goat ability to sort against dietary fibre was also evaluated. There was a very good repeatability of the aggregate measures between days within a period for a given goat estimated by the day effect within breed and goat, tested on the residual variance (P > 0.95). The correlations between periods were the highest between the second and either the third or fourth periods. With increasing age, goats sorted more against the fibrous part of the TMR and increased their initial rate of intake. Alpine goats ate more slowly than Saanen goats but ate during a longer duration. Principal component analysis (PCA) was performed on all the aggregate measures of feed intake patterns. The factor score plots generated by the PCA highlighted the opposition between the different measures of feed intake patterns and the sorting behaviour. The projection of the animals on the scoring plots showed a breed effect and that there was a continuum for the feed intake pattern of goats. In conclusion, this study showed that the feed intake pattern was highly repeatable for an animal in a given period and between periods. This means that phenotyping goats in a younger age might be of interest, either to select them on feeding behaviour and choose preferentially the slow eaters or to adapt the quantity offered and restrict feed delivery to the fast eaters in order to increase feed efficiency and welfare by limiting the occurrence of acidosis, for example.

A new Tier 3 method to calculate methane emission inventory for ruminants.

Livestock is the main source of methane (CH4) emissions. It is important to accurately determine emissions from ruminants that meet standardized international guidelines for national greenhouse gas inventories. A new method to improve the accuracy of CH4 emissions that complies with IPCC rules for a Tier 3 method is described and evaluated. This method, developed by INRA (French Institute for Agricultural Research), was applied to the French inventory of CH4 emissions by ruminants and compared with the IPCC Tier 2 method. For enteric CH4, depending on the animal category, the INRA CH4 emission estimates lay between 88% and 114% of IPCC's. The INRA/IPCC ratio for enteric emission was close to unity and did not differ between methods (P = 0.43) for adult cows (i.e. most cattle). In France, feedlot manure is stored in aerobic conditions, and so the INRA/IPCC fit for manure emission was poorer (P < 0.05). The INRA/IPCC fit for enteric CH4 was very close between methods to that for total CH4 (P = 0.39), enteric CH4 representing 93% of total emissions. The main improvement is the use of a robust equation (from numerous data and diets), based on digestible organic matter intake (DOMI) corrected for the digestive interactions, to predict CH4 consistently from enteric and manure sources. It was developed for the French livestock inventory but is customizable for other countries. This new improved CH4 estimation method, based on equations from a large literature database, complies with IPCC rules for a Tier 3 method.

Review: Converting nutritional knowledge into feeding practices: a case study comparing different protein feeding systems for dairy cows.

Improving milk nitrogen efficiency through a reduction of CP supply without detrimental effect on productivity requires usage of feeding systems estimating both the flows of digestible protein, the exported true proteins and from these predict milk protein yield (MPY). Five feeding systems were compared in their ability to predict MPY v. observed MPY in two studies where either protein supply or protein and energy supply were changed. The five feedings systems were: Cornell Net Carbohydrate and Protein System (v6.5.5), Dutch protein evaluation system (1991 and 2007), Institut National de la Recherche Agronomique in France (INRA), National Research Council and NorFor. The key characteristic of the systems with the best predicted MPY was the inclusion of a variable efficiency of utilisation of protein supply taking into account the supply of both protein and energy. The systems still using a fixed efficiency had the highest slope bias in their prediction of MPY. Therefore, the development of new feeding systems or improvement of existing systems should include a variable efficiency of utilisation of the protein related to both the protein and energy supply. The limitation of the current comparison did not allow determining if additional factors, as used in INRA, were beneficial. This concept should also probably be transferred to essential amino acids.

Review: To be or not to be an identifiable model. Is this a relevant question in animal science modelling?

What is a good (useful) mathematical model in animal science? For models constructed for prediction purposes, the question of model adequacy (usefulness) has been traditionally tackled by statistical analysis applied to observed experimental data relative to model-predicted variables. However, little attention has been paid to analytic tools that exploit the mathematical properties of the model equations. For example, in the context of model calibration, before attempting a numerical estimation of the model parameters, we might want to know if we have any chance of success in estimating a unique best value of the model parameters from available measurements. This question of uniqueness is referred to as structural identifiability; a mathematical property that is defined on the sole basis of the model structure within a hypothetical ideal experiment determined by a setting of model inputs (stimuli) and observable variables (measurements). Structural identifiability analysis applied to dynamic models described by ordinary differential equations (ODEs) is a common practice in control engineering and system identification. This analysis demands mathematical technicalities that are beyond the academic background of animal science, which might explain the lack of pervasiveness of identifiability analysis in animal science modelling. To fill this gap, in this paper we address the analysis of structural identifiability from a practitioner perspective by capitalizing on the use of dedicated software tools. Our objectives are (i) to provide a comprehensive explanation of the structural identifiability notion for the community of animal science modelling, (ii) to assess the relevance of identifiability analysis in animal science modelling and (iii) to motivate the community to use identifiability analysis in the modelling practice (when the identifiability question is relevant). We focus our study on ODE models. By using illustrative examples that include published mathematical models describing lactation in cattle, we show how structural identifiability analysis can contribute to advancing mathematical modelling in animal science towards the production of useful models and, moreover, highly informative experiments via optimal experiment design. Rather than attempting to impose a systematic identifiability analysis to the modelling community during model developments, we wish to open a window towards the discovery of a powerful tool for model construction and experiment design.

Modeling homeorhetic trajectories of milk component yields, body composition and dry-matter intake in dairy cows: Influence of parity, milk production potential and breed.

The control of nutrient partitioning is complex and affected by many factors, among them physiological state and production potential. Therefore, the current model aims to provide for dairy cows a dynamic framework to predict a consistent set of reference performance patterns (milk component yields, body composition change, dry-matter intake) sensitive to physiological status across a range of milk production potentials (within and between breeds). Flows and partition of net energy toward maintenance, growth, gestation, body reserves and milk components are described in the model. The structure of the model is characterized by two sub-models, a regulating sub-model of homeorhetic control which sets dynamic partitioning rules along the lactation, and an operating sub-model that translates this into animal performance. The regulating sub-model describes lactation as the result of three driving forces: (1) use of previously acquired resources through mobilization, (2) acquisition of new resources with a priority of partition towards milk and (3) subsequent use of resources towards body reserves gain. The dynamics of these three driving forces were adjusted separately for fat (milk and body), protein (milk and body) and lactose (milk). Milk yield is predicted from lactose and protein yields with an empirical equation developed from literature data. The model predicts desired dry-matter intake as an outcome of net energy requirements for a given dietary net energy content. The parameters controlling milk component yields and body composition changes were calibrated using two data sets in which the diet was the same for all animals. Weekly data from Holstein dairy cows was used to calibrate the model within-breed across milk production potentials. A second data set was used to evaluate the model and to calibrate it for breed differences (Holstein, Danish Red and Jersey) on the mobilization/reconstitution of body composition and on the yield of individual milk components. These calibrations showed that the model framework was able to adequately simulate milk yield, milk component yields, body composition changes and dry-matter intake throughout lactation for primiparous and multiparous cows differing in their production level.

Prediction of the amino acid digestibility of legume seeds in growing pigs: a meta-analysis approach.

On pig farms, a high proportion of the cost of production comes from feed costs. However, the use of alternative ingredients such as legume seeds may help to reduce this cost. In fact, legume seeds are an important source of essential amino acids (EAA) and can therefore be an alternative to oilseed meals. However, the accurate use of these legume seeds requires a precise knowledge of the standardized ileal digestibility (SID) of EAA, which may vary depending on its botanical variety. A meta-analysis was performed on a database compiling data from 41 studies published between 1981 and 2013 and 178 dietary treatments. Models of prediction of the SID of EAA as well as the dietary concentration of digestible standardized EAA (dEAA) were obtained, based on the chemical composition of ingredients reported in the publications. The effect of the type of legume seeds (faba bean, lupin, pea and soya bean), surgical procedures (T-cannula, re-entrant cannulas, post valve T-cannulas and ileo-rectal anastomosis), and BW of pigs (BW⩽25 kg BW>25 kg) were also tested in each model. Results showed that dietary CP and crude fibre (CF) were, respectively, the best predictors of each EAA SID for faba bean, lupin and pea (R 2=0.42 to 0.89) and soya bean (R 2=0.32 to 0.77). For the dEAA content, the best prediction models included dietary CP and ADF for faba bean, lupin and pea and soya bean, respectively, with R 2 ranging from 0.66 to 0.98. Models developed in this study allow predicting the digestibility of EAA in these alternatives feedstuffs.

A method to estimate cow potential and subsequent responses to energy and protein supply according to stage of lactation.

Milk responses to dietary change are influenced by the relative production level, that is, the distance between observed production and potential production. The closer the animal is to its potential, the smaller the expected response is to extra nutrients. Therefore, the aim of this work was to provide a method to quantify cow potential, to estimate subsequent responses to changes in nutrient supply. The observed efficiencies in net energy for lactation (NEL) and metabolizable protein (MP) are proposed as a basis to estimate the relative production level of the animal. The rationale for using NEL and MP efficiency (ratios of milk energy yield/NEL above maintenance supply and milk protein yield/MP above maintenance supply) builds on the uniformity of the observed relationships between size of the milk responses and extra NEL supply and MP supply, when centered on a given efficiency. From there, a pivot nutritional situation where MP and NEL efficiency are 0.67 and 1.00, respectively, was defined, from which milk responses could be derived across animals varying in production potential. An implicit assumption of using response equations centered on reference efficiency pivots is that the size of the response to a fixed change in nutrient supply, relative to the pivot, is identical for animals with different production capacities. The proposed approach was evaluated with 2 independent data sets, where different dietary treatments were applied during the whole lactation. In these data sets, MP and NEL above maintenance supply were calculated weekly using the recently updated INRA Systali feed units system. Differences in NEL and MP supply above maintenance between the extreme dietary treatments were large, on average 667 g of MP/d and 13 MJ of NEL/d (3.11 Mcal/d) in the first data set, and 513 g of MP/d and 29 MJ of NEL/d (6.93 Mcal/d) for the second data set. Milk energy yield and milk component yields were predicted with root mean square prediction errors between 7.6 and 13.5% and concordance correlation coefficients between 0.784 and 0.934, respectively. Assessed by the Akaike's information criterion, significant differences existed in the accuracy of prediction for milk energy yield and milk component yields between stages of lactation. However, the effects of stage of lactation were not consistent between data sets and, for most of the predicted variables, relatively small. We concluded that the pivot concept can be used to predict milk energy yield and milk component yields responses to dietary change with a good accuracy for diets that are substantially different and across all stages of lactation.

Milk yield and milk composition responses to change in predicted net energy and metabolizable protein: a meta-analysis.

Using a meta-analysis of literature data, this study aimed to quantify the dry matter (DM) intake response to changes in diet composition, and milk responses (yield, milk component yields and milk composition) to changes in dietary net energy for lactation (NEL) and metabolizable protein (MP) in dairy cows. From all studies included in the database, 282 experiments (825 treatments) with experimentally induced changes in either NEL or MP content were kept for this analysis. These treatments covered a wide range of diet characteristics and therefore a large part of the plausible NEL and MP contents and supplies that can be expected in practical situations. The average MP and NEL contents were, respectively (mean±SD), 97±12 g/kg DM and 6.71±0.42 MJ/kg DM. On a daily supply basis, there were high between-experiment correlations for MP and NEL above maintenance. Therefore, supplies of MP and NEL above maintenance were, respectively, centred on MP supply for which MP efficiency into milk protein is 0.67, and NEL above maintenance supply for which the ratio of NEL milk/NEL above maintenance is 1.00 (centred variables were called MP67 and NEL100). The majority of the selected studies used groups of multiparous Holstein-Friesian cows in mid lactation, milked twice a day. Using a mixed model, between- and within-experiment variation was split to estimate DM intake and milk responses. The use of NEL100 and MP67 supplies substantially improved the accuracy of the prediction of milk yield and milk component yields responses with, on average, a 27% lower root mean square error (RMSE) relative to using dietary NEL and MP contents as predictors. For milk composition (g/kg), the average RMSE was only 3% lower on a supply basis compared with a concentration basis. Effects of NEL and MP supplies on milk yield and milk component yields responses were additive. Increasing NEL supply increases energy partitioning towards body reserve, whereas increasing MP supply increases the partition of energy towards milk. On a nitrogen basis, the marginal efficiency decreases with increasing MP supply from 0.34 at MP67=-400 g/day to 0.07 at MP67=300 g/day. This difference in MP67 supply, assuming reference energy level of NEL100=0, equates to a global nitrogen efficiency decrease from 0.82 to 0.58. The equations accurately describe DM intake response to change in dietary contents and milk responses to change in dietary supply and content of NEL and MP across a wide range of dietary compositions.

Meta-analysis of the amino acid digestibility of oilseed meal in growing pigs.

Oilseed meal is an important source of essential amino acids (EAA) for livestock production. It is the second most important ingredient in pig feed after grains. Optimal use of these ingredients requires precise knowledge of amino acid standardized ileal digestibility (SID), which may vary depending on several factors including botanical variety or processing treatments. A meta-analysis was performed in order to derive models for predicting the SID of soybean, cotton and rapeseed meal EAA, based on chemical composition data such as CP, total concentration of each EAA and fibre (crude fibre, ADF and NDF) content. A database of 47 references (224 experimental treatments) was built. A model incorporating processing method of the meals (e.g. cold pressed, expeller pressed, solvent extracted), experimental surgical procedure (T-cannula, re-entrant cannula, post valve T-cannula and ileo-rectal anastomosis) and pig growth stage (BW⩽ or ⩾25 kg) was tested. Results indicated that neither processing nor BW affected EAA SID. NDF was the best predictor of SID (R 2=0.944, 0.836, 0.779, 0.899 and 0.814, respectively, for Lys, Met, Thr, Trp and Val). The total EAA content was the best predictor of digestible content (g/kg diet) for each EAA (R 2=0.990, 0.985, 0.977, 0.985 and 0.978, respectively, for Lys, Met, Thr, Trp and Val). This study shows that routine chemical analyses may be used to predict EAA digestibility with satisfactory accuracy.

Quantification of the main digestive processes in ruminants: the equations involved in the renewed energy and protein feed evaluation systems.

The evolution of feeding systems for ruminants towards evaluation of diets in terms of multiple responses requires the updating of the calculation of nutrient supply to the animals to make it more accurate on aggregated units (feed unit, or UF, for energy and protein digestible in the intestine, or PDI, for metabolizable protein) and to allow prediction of absorbed nutrients. The present update of the French system is based on the building and interpretation through meta-analysis of large databases on digestion and nutrition of ruminants. Equations involved in the calculation of UF and PDI have been updated, allowing: (1) prediction of the out flow rate of particles and liquid depending on the level of intake and the proportion of concentrate, and the use of this in the calculation of ruminal digestion of protein and starch from in situ data; (2) the system to take into account the effects of the main factors of digestive interactions (level of intake, proportion of concentrate, rumen protein balance) on organic matter digestibility, energy losses in methane and in urine; (3) more accurate calculation of the energy available in the rumen and the efficiency of its use for the microbial protein synthesis. In this renewed model UF and PDI values of feedstuffs vary depending on diet composition, and intake level. Consequently, standard feed table values can be considered as being only indicative. It is thus possible to predict the nutrient supply on a wider range of diets more accurately and in particular to better integrate energy×protein interactions occurring in the gut.

Response of growing ruminants to diet in warm climates: a meta-analysis.

The aim of this work was to establish the response of growing sheep, goats and cattle to different nutritional environments. Data from 590 publications representing 2225 treatments were analysed. The results showed that each 10% increase in NDF was accompanied by 0.11 g/kg live weight (LW) and 0.32 g/kg metabolic live weight (LW0.75) decreases in DMI. Otherwise, the response of DMI to CP (CP%DM) content was curvilinear (P1.4, corresponding to maximum growth (HE+)). No significant difference was observed between LE-- and LE-, and no significant difference was observed between ME+- and HE+. For nitrogen balance, no difference was observed between species for a given level of nitrogen intake.

A meta-analysis of nutrient intake, feed efficiency and performance in cattle grazing on tropical grasslands.

It is essential to quantify the potential of tropical grasslands to allow significant feed efficiency for grazing livestock in controlled conditions such as at pasture. We conducted a quantitative analysis of published studies reporting the experimental results of average daily gains (ADG) and diet characteristics obtained specifically under grazing conditions (17 publications and 41 experiments), which have been less studied compared with controlled conditions in stalls. The database was analyzed to determine the average and range of values obtained for ADG (g/kg BW), dry matter digestibility, intake (DMI) and digestible DMI (DDMI, g/kg BW) and feed conversion efficiencies (FCE), as well as to predict the response of these parameters to the main strategies investigated in the literature - that is, mainly the stocking rate (SR) and the concentrate intake (CI). The ADG reached 1.2 kg BW per day and was directly linked to DDMI (ADG=-1.63+0.42 DDMI -0.0084 DDMI2, n=90, r.m.s.e=0.584, R 2=0.93). The DDMI, which was representative of the nutrient input, was driven mainly by DMI rather than dry matter digestibility, whereas these two parameters did not correlate (r=0.068, P=0.56). The average global FCE (0.11 g ADG/g DDMI) showed a greater association with the metabolic FCE (0.17 g ADG/g DMI) than the digestive FCE (0.62). The CI (g DM/kg BW) increased ADG (ADG=2376+CI 56.1, n=16, r.m.s.e.=441, R 2=0.95). The SR expressed as kg BW/ha decreased the individual ADG by 1.19 g/kg BW per additional ton of BW/ha, whereas the global ADG calculated per ha increased by 0.57 per additional ton BW/ha. When the SR was expressed as kg BW/ton DM and per ha rather than as kg BW/ha, the impact on the individual ADG decreased by 0.18 or 0.86 g per additional ton BW/ha, depending on the initial BW of the cattle. These results provide a better view of the potential performance and feeding of cattle in tropical grasslands. The results provide an improved quantification of the relationships between diet and performance, as well as the overall quantitative impact of SR and supplementation.

Empirical prediction of net splanchnic release of ketogenic nutrients, acetate, butyrate and ß-hydroxybutyrate in ruminants: a meta-analysis.

For energy feeding systems for ruminants to evolve towards a nutrient-based system, dietary energy supply has to be determined in terms of amount and nature of nutrients. The objective of this study was to establish response equations of the net hepatic flux and net splanchnic release of acetate, butyrate and ß-hydroxybutyrate to changes in diet and animal profiles. A meta-analysis was applied on published data compiled from the FLuxes of nutrients across Organs and tissues in Ruminant Animals database, which pools the results from international publications on net splanchnic nutrient fluxes measured in multi-catheterized ruminants. Prediction variables were identified from current knowledge on digestion, hepatic and other tissue metabolism. Subsequently, physiological and other, more integrative, predictors were obtained. Models were established for intakes up to 41 g dry matter per kg BW per day and diets containing up to 70 g concentrate per 100 g dry matter. Models predicted the net hepatic fluxes or net splanchnic release of each nutrient from its net portal appearance and the animal profile. Corrections were applied to account for incomplete hepatic recovery of the blood flow marker, para-aminohippuric acid. Changes in net splanchnic release (mmol/kg BW per hour) could then be predicted by combining the previously published net portal appearance models and the present net hepatic fluxes models. The net splanchnic release of acetate and butyrate were thus predicted from the intake of ruminally fermented organic matter (RfOM) and the nature of RfOM (acetate: residual mean square error (RMSE)=0.18; butyrate: RMSE=0.01). The net splanchnic release of ß-hydroxybutyrate was predicted from RfOM intake and the energy balance of the animals (RMSE=0.035), or from the net portal appearance of butyrate and the energy balance of the animals (RMSE=0.050). Models obtained were independent of ruminant species, and presented low interfering factors on the residuals, least square means or individual slopes. The model equations highlighted the importance of considering the physiological state of animals when predicting splanchnic metabolism. This work showed that it is possible to use simple predictors to accurately predict the amount and nature of ketogenic nutrients released towards peripheral tissues in both sheep and cattle at different physiological status. These results provide deeper insight into biological processes and will contribute to the development of improved tools for dietary formulation.

Effect of concentrate level on feeding behavior and rumen and blood parameters in dairy goats: relationships between behavioral and physiological parameters and effect of between-animal variability.

This work aimed first to compare 2 diets differing in their percentage of concentrate [low (LO): 30% concentrate vs. High (HI): 60% concentrate] by measuring simultaneously feeding behavior, rumen parameters, blood and plasma parameters, and milk yield and composition in 8 mid-lactation goats. The second aim was to study the interrelationships between these variables and to analyze the between-animal variability to better understand the between-animal differences in acidosis susceptibility. All of the animals received the 2 diets ad libitum as total mixed ration according to a crossover design of two 4-wk periods. Mean daily DMI was similar between the 2 diets but the variability was higher for the HI than for the LO diet. Goats produced more milk when fed the HI diet compared with the LO diet but with a lower fat:protein ratio (0.81 vs. 0.99). They ate more rapidly the HI than the LO diet but stopped eating sooner after the afternoon feed allowance, and spent less time chewing. The increase in concentrate percentage modified rumen parameters: the pH and acetate:propionate ratio decreased and total VFA, ammonia, and soluble carbohydrate concentrations increased. Hematocrit, plasma NEFA, and blood K and Ca concentrations decreased but glycemia and uremia increased. Other parameters were not modified: milk fat content, blood pH, and bicarbonate and Na concentrations. A large between-animal variability was detected for all the measured parameters, especially for feeding behavior, with important consequences on rumen and blood parameters. This work confirmed the effects of a high percentage of concentrate on feeding behavior, rumen and blood parameters, and milk production, and some known relationships such as the positive link between rumen pH and chewing index. It also pointed out other relationships between parameters seldom measured at the same time, such as rumen redox potential or blood pH and chewing index, or the negative link between blood and rumen pH. When the animals spent a lot of time chewing, they probably produced a lot of saliva that buffered the rumen pH and prevented them from suffering from subacute ruminal acidosis. However, they used part of their blood bicarbonates reserve, which might have induced metabolic acidosis, as rumen and blood pH were inversely related. This could explain why some animals suffer from acidosis and others do not in a herd receiving the same diet, and why some animals seem to suffer more from subacute ruminal acidosis and others from metabolic acidosis.

Nutritional requirements of sheep, goats and cattle in warm climates: a meta-analysis.

The objective of the study was to update energy and protein requirements of growing sheep, goats and cattle in warm areas through a meta-analysis study of 590 publications. Requirements were expressed on metabolic live weight (MLW=LW0.75) and LW1 basis. The maintenance requirements for energy were 542.64 and 631.26 kJ ME/kg LW0.75 for small ruminants and cattle, respectively, and the difference was significant (P<0.01). The corresponding requirement for 1 g gain was 24.3 kJ ME without any significant effect of species. Relative to LW0.75, there was no difference among genotypes intra-species in terms of ME requirement for maintenance and gain. However, small ruminants of warm and tropical climate appeared to have higher ME requirements for maintenance relative to live weight (LW) compared with temperate climate ones and cattle. Maintenance requirements for protein were estimated via two approaches. For these two methods, the data in which retained nitrogen (RN) was used cover the same range of variability of observations. The regression of digestible CP intake (DCPI, g/kg LW0.75) against RN (g/kg LW0.75) indicated that DCP requirements are significantly higher in sheep (3.36 g/kg LW0.75) than in goats (2.38 g/kg LW0.75), with cattle intermediate (2.81 g/kg LW0.75), without any significant difference in the quantity of DCPI/g retained CP (RCP) (40.43). Regressing metabolisable protein (MP) or minimal digestible protein in the intestine (PDImin) against RCP showed that there was no difference between species and genotypes, neither for the intercept (maintenance=3.51 g/kg LW0.75 for sheep and goat v. 4.35 for cattle) nor for the slope (growth=0.60 g MP/g RCP). The regression of DCP against ADG showed that DCP requirements did not differ among species or genotypes. These new feeding standards are derived from a wider range of nutritional conditions compared with existing feeding standards as they are based on a larger database. The standards seem to be more appropriate for ruminants in warm and tropical climates around the world.

Effects of inoculum source, pH, redox potential and headspace di-hydrogen on rumen in vitro fermentation yields.

This in vitro study aimed at understanding how abiotic, that is chemical and electrochemical potentials, and biotic factors combine to impact the outputs of rumen volatile fatty acid (VFA). Using a 48-run design optimized by means of an exchange algorithm, the curvilinear effects of pH, Eh and partial pressure of dihydrogen (H2) on fermentation yields were investigated in 6-h batch cultures of mixed rumen microbes, fed on glucose so as to bypass the enzymatic hydrolysis and conversion steps preceding the glycolytic pathway. The role played by rumen microbiota in the expression of these effects was explored by testing three inocula grown on feeds supplying a microflora adapted to fibre, slowly degradable or readily degradable starch as the dominant dietary polysaccharide. Data were fitted to 2nd-order polynomial models. In fibre-adapted cultures, the yields of major VFA were mainly influenced by pH and H2 partial pressure, in opposite ways. In wheat grain-adapted cultures, the VFA yields underwent the opposite influences of pH, in a curvilinear way for propionate, and Eh since acetate production yield was not significantly modified by any factor. In maize grain-adapted cultures, acetate production yield was not modified by any factor but H2 in a quadratic way when the production yields of higher VFA underwent opposite influences of pH and Eh. In conclusion, the effects of environmental factors were dependent on the nature of the inoculum, a major source of variation, and more particularly on its adaptation to high- or low-fibre diets. These effects were loosely interrelated, the pH being the most active factor before the Eh and H2 partial pressure.

Influence of trough versus pasture feeding on average daily gain and carcass characteristics in ruminants: a meta-analysis.

Quantitative meta-analysis was run on 108 publications featuring 116 experiments and 399 treatments dealing with the effect of trough or pasture feeding environment (FE) on ruminant performances. The objective was to compare the effect of trough or pasture FE on ADG, diet OM digestibility (OMD), various carcass characteristics, and the interaction between FE and complementation modalities. Live weight was adjusted to compare results between species. Results showed that trough-fed animals had higher ADG (+17.89%; P < 0.001), hot carcass yield (HYield; + 2.47%; P < 0.001) and carcass fat content (+ 24.87%; P < 0.001) than pasture-fed animals but lower carcass muscle and bone percentages (-1.60% [P = 0.010] and -7.63% [P = 0.003], respectively). Feeding environment had no effect on diet OMD (P = 0.818), but the number of observations was low. After considering the addition or not of concentrate in the diet (addiCO), FE effect persisted on ADG (P = 0.024) and carcass fat content (P = 0.027) but not on HYield (P = 0.078) or muscle and bone percentages (P = 0.119 and P = 0.581, respectively). After considering the nature of the concentrate (natCO), FE effect persisted on ADG (P < 0.001) and HYield (P = 0.004). Considering the percentage of concentrate in the diet (PCO) erased FE effect on ADG (P = 0.891) and HYield (P = 0.128). In contrast, considering the quantity of concentrate(QCO) erased FE effect on ADG (P = 0.084) but not on HYield (P = 0.006) or on carcass fat and muscle contents (P = 0.040 and P = 0.040, respectively) although the FE effect on carcass bone content persisted (P = 0.550). Animal species and physiological stage had no effect on any of the variables studied (P > 0.05) but experiment did (P ≤ 0.001). The increase in ADG was positively correlated to HYield in cattle (P = 0.002) and small ruminants (P = 0.003) and positively linked to carcass fat content (P = 0.007) but not carcass muscle content, which actually decreased (P = 0.001). Overall, this meta-analysis confirmed previous reports of FE effects and revealed how the differences generally reported result from a confounding effect of FE and nature of the diet. Indeed, in most of the studies used, trough-fed animals were supplemented with concentrate whereas pasture-fed animals were not. This research also highlighted the fact that pasture-fed animals have the potential to achieve the same performances as trough-fed animals when fed a similar diet.