In vitro starch digestion and predicted glycemix index of cereal grains commonly utilized in pig nutrition

Starch is the largest constituent in diets for pigs and its structure is considered to have a great impact on the rate of starch digestion. Despite several works showing starch is digested at varying rates in vitro, limited information exist on the characterization of single feedstuffs commonly utilized in pig nutrition through a predicted glycemic index (pGI) approach. Our objectives were: (1) to assess the variability of in vitro starch digestibility of cereal grains as a mean of screening samples by involving a pGI approach, (2) to investigate the role of heat processing and amylose level on pGI and (3) to investigate relationships among pGI, chemical variables and Englyst starch fractions. For this purpose, 137 samples of cereal grains [normal and low-amylose and heat processed (NA, LA and HP, respectively)] were evaluated using an in vitro method based on the Englyst-assay for nutritional classification of starch and a non-linear model was used to fit starch digestibility data. Starch digestion potential greatly differed among cereal grains (P<0.05). Consequently, pGI values differed (P<0.05), ranging from 15.9 for sorghum to 106.9 for HP rice. In general, higher pGI scores were recorded for HP and LA cereals than respective NA counterparts (P<0.05). Differences (P<0.05) were measured both in the rate of starch digestion (ranging from 0.017/min for NA maize to 0.182/min for HP rice) and in the starch digestible fractions. In particular, rapidly digestible starch (RDS) ranged from 119 g/kg dry matter (DM) for sorghum to 704 g/kg DM for HP rice, whereas resistant starch (RS) ranged from 32 g/kg DM for HP barley to 275 g/kg DM for sorghum. The loading plots of principal component (PC) analysis revealed significant relationships between pGI and variables of interest, showing that both chemical compositions and starch fractions could affect the pGI. In particular, the PC analysis conducted on all samples revealed that pGI was positively related with glucose and RDS (P<0.05) and negatively related to crude lipid, slowly digestible starch and RS (P<0.05) contents. The pGI approach could be a rapid, laboratory-based functional evaluation of starch value of cereal feeds entering in pig diet