We studied the chemical composition, protein quality, in vitro protein digestibility (IVPD), and in vivo apparent ileal protein digestibility (AIPD) and AMEn content, of 27 soybean meal (SBM) samples from Argentina (ARG), Brazil (BRA), and USA. The samples (n = 9 per country) were collected at random from European crushers and feed mills by specialized personnel, and analyzed for main nutrients (CP, amino acids, crude fiber, neutral detergent fiber, sucrose, oligosaccharides, and minerals) and protein quality indicators (urease activity, KOH solubility, protein dispersibility index, and trypsin inhibitor activity). The IVPD of the meals was determined using a two-step test that simulated the digestive processes occurring in the gastrointestinal tract of the chick. In addition, the AIDP and the AMEn content of the SBM were determined in 21 d-old broilers. Broilers were fed a commercial crumble diet based on cereals and SBM from hatch to 17 d of age and then, they received their corresponding experimental diet ad libitum for 4 d. The experimental diets consisted of 53% of a common N-free diet (44.0% dextrose, 3.5% soybean oil, 3.5% vitamins and minerals, and 2.0% lignocellulose) and 47% of the SBM tested, as the unique source of N. All data were analyzed as a completely randomized design using the MIXED procedure of SAS with the origin of the SBM as main effect. Additionally, for the in vivo data, SBM sample within each origin was used as a nested effect where each of the 27 samples was replicated 6 times and the experimental unit was a cage with 8 broilers. On 88% DM bases, SBM from BRA had more CP (46.5, 45.4, and 45.5%; P < 0.05) and less sucrose (5.21, 6.28, and 6.47%; P < 0.001) and stachyose (4.20, 4.66, and 4.78%; P < 0.05) than SBM from USA and ARG. Protein quality indicators values were higher for the USA meals than for the BRA and ARG meals (P < 0.05). At 3 h post-incubation, IVPD values tended to be higher for the USA meals than for the ARG and BRA meals (P = 0.078). At 21 d of age, AIPD of the SBM tended to be higher (87.1, 86.5, and 86.1%, respectively; P = 0.063) for the USA meals than for the ARG and BRA meals. Also, the AMEn of the SBM was 2,323, 2,307, and 2,281 kcal/kg for the USA, ARG, and BRA meals, respectively, differences that were not significant in spite of the higher CP content of the BRA meals.
In summary, the chemical composition, protein quality indicators, and nutrient quality of the SBM varied with the origin of the beans. Feed mill managers and nutritionists should take into account the composition and origin of the beans, to cre-ate a more accurate matrix of nutritional values for commercial SBM.