A first model of the fate of dietary calcium and phosphorus in broiler chickens

Abstract

To reduce P excretion and increase the sustainability of poultry farms, one needs to understand the mechanisms surrounding P metabolism and its close link with Ca metabolism to precisely predict the fate of dietary P and Ca and related requirements for birds. This study describes and evaluates a model developed to estimate the fate of Ca and P consumed by broilers. The Ca and P model relies on three modules: (1) digestion of Ca and P; (2) dynamics of Ca and P in soft tissue and feathers; and (3) dynamics of body ash. Exogenous phytase affects the availability of Ca and P; thus, to predict the absorption of those minerals, the model also accounts for the effect of phytase on Ca and P digestibility. We used a database to estimate the consequences of dietary Ca, P, and phytase over feed intake response. This study followed a four-step process: (1) Ca and P model development and its coupling with a growth broiler model; (2) model behavior assessment; (3) sensitivity analysis to identify the most influential parameters; and (4) external evaluation based on three databases. The proportion of P in body protein and the Ca to P ratio in bone are the most sensitive parameters of P deposition in soft tissue and bone, representing 91 and 99% of the total variation. The external evaluation results indicated that body water and protein had an overall mean square prediction error (rMSPE) of 7.22 and 12.3%, respectively. The prediction of body ash, Ca, and P had an rMSPE of 7.74, 11.0, and 6.56%, respectively, mostly errors of disturbances (72.5, 51.6, and 90.7%, respectively). The rMSPE for P balance was 13.3, 18.4, and 22.8%, respectively, for P retention, excretion, and retention coefficient, with respective errors due to disturbances of 69.1, 99.9, and 51.3%. We demonstrated a mechanistic model approach to predict the dietary effects of Ca and P on broiler chicken responses with low error, including detailed simulations to show the confidence level expected from the model outputs. Overall, this model predicts broilers’ response to dietary Ca and P. The model could aid calculations to minimize P excretion and reduce the impact of broiler production on the environment. A model inversion is ongoing that will enable the calculation of Ca and P dietary quantities for a specific objective. This will simplify the use of the model and the feed formulation process.