Estimation of the tryptophan requirement in piglets by meta-analysis

There is no consensus concerning the Trp requirement for piglets expressed relative to Lys on a standardized ileal digestible basis (SID Trp : Lys). A meta-analysis was performed to estimate the SID Trp : Lys ratio that maximizes performance of weaned piglets between 7 and 25 kg of BW. A database comprising 130 experiments on the Trp requirement in piglets was established. Thenutritional values of the diets were calculated from the composition of feed ingredients. Among all experiments, 37 experiments were selected to be used in the meta-analysis because they were designed to express the Trp requirement relative to Lys (e.g. Lys was the second-limiting amino acid in the diet) while testing at least three levels of Trp. The linear-plateau (LP), curvilinear-plateau (CLP) and asymptotic (ASY) models were tested to estimate the SID Trp : Lys requirement using average daily gain (ADG), average daily feed intake (ADFI) and gain-to-feed ratio (G : F) as response criteria. A multiplicative trial effect was included in the modelson the plateau value, assuming that the experimental conditions affected only this parameter and not the requirement or the shape of the response to Trp. Model choice appeared to have an important impact on the estimated requirement. Using ADG and ADFI as response criteria, the SID Trp : Lys requirement was estimated at 17% with the LP model, at 22% with the CLP modeland at 26% with the ASY model. Requirement estimates were slightly lower when G : F was used as response criterion. The Trp requirement was not affected by the composition of the diet (corn v. a mixture of cereals). The CLP model appeared to be the best-adapted model to describe the response curve of a population. This model predicted that increasing the SID Trp : Lys ratiofrom 17% to 22% resulted in an increase in ADG by 8%.