Predicting in vivo starch digestibility coefficients in newly weaned piglets from in vitro assessment of diets using multivariate analysis.

The study was based on correlating a dataset of in vivo mean starch digestibility coefficients obtained in the immediate post-weaning phase of piglets with a range of dietary in vitro variables. The paper presents a model that predicts (R2 0·71) in vivo average starch digestibility coefficients in the 0·5 small-intestinal region of newly weaned piglets fed cereal-based diets using seven in vitro variables describing starch properties that are fundamentally associated with the quality of feed materials, i.e. hydration, structure and amylolytic digestion. The variables were: Rapid Visco Analyser (RVA; measures the viscosity of materials when sheared under defined hydration and temperature regimens); RVA end viscosity; RVA (gelatinisation) peak viscosity; ΔH (gelatinisation enthalpy that provides an estimate of helical order or degree of crystallinity in starch); water solubility index (WSI; that denotes the amount of soluble polysaccharides released from starch granules to the aqueous phase); grain endogenous amylase (concentration of endogenous α-amylase in cereals, assessed by pasting cereal flours in 25 g of AgNO3, an amylase inhibitor v. water using RVA).