The objective of the present study was to determine whether there are differences in prececal amino acid digestibility between commonly used slow- and fast-growing broiler strains when the regression approach is applied. ISA J-275 and Ross 308 were selected as common representatives of slow- and fast-growing broiler strains, respectively. The experimental diets with soybean cake at levels of 0, 100, and 200 g/kg were offered for ad libitum consumption between 22 and 29 d post-hatch. Titanium dioxide was used as an indigestible marker. Each treatment was tested with six pens comprising 10 birds each. Digesta samples were collected on a pen basis from the distal two-thirds of the intestine section between Meckel's diverticulum and 2 cm anterior to the ileocecal-colonic junction. The prececal amino acid digestibility of soybean cake was calculated by linear regression simultaneously for both strains. There was no significant interaction between broiler strain and inclusion level of soybean cake with respect to the prececal CP and amino acid digestibility of complete diets; there was a significant strain effect for 5 out of the 16 measured amino acids. The prececal CP and amino acid digestibility of soybean cake did not differ significantly between strains and was numerically almost identical. The results of the present study provide evidence of the transferability between broiler strains of prececal amino acid digestibility data, determined using the regression approach, thus improving the accuracy of diet formulation without drawbacks.

1. Herein, it was investigated whether different particle size distributions of feed ingredients achieved by grinding through a 2- or 3-mm grid would have an effect on precaecal (pc) amino acid (AA) digestibility. Maize and soybean meal were used as the test ingredients. 2. Maize and soybean meal was ground with grid sizes of 2 or 3 mm. Nine diets were prepared. The basal diet contained 500 g/kg of maize starch. The other experimental diets contained maize or soybean meal samples at concentrations of 250 and 500, and 150 and 300 g/kg, respectively, instead of maize starch. Each diet was tested using 6 replicate groups of 10 birds each. The regression approach was applied to calculate the pc AA digestibility of the test ingredients. 3. The reduction of the grid size from 3 to 2 mm reduced the average particle size of both maize and soybean meal, mainly by reducing the proportion of coarse particles. Reducing the grid size significantly (P < 0.050) increased the pc digestibility of all AA in the soybean meal. In maize, reducing the grid size decreased the pc digestibility of all AA numerically, but not significantly (P > 0.050). The mean numerical differences in pc AA digestibility between the grid sizes were 0.045 and 0.055 in maize and soybean meal, respectively. 4. Future studies investigating the pc AA digestibility should specify the particle size distribution and should investigate the test ingredients ground similarly for practical applications.

In this study, we determined whether deficient dietary amino acid (AA) concentrations influence the precaecal (pc) AA digestibility when determined using the regression approach. We mixed two basal diets. Basal diet 1 was deficient in essential AAs, whereas adequate AA concentrations were ensured in basal diet 2 by adding free AAs. Rapeseed cake and full‐fat soya beans as test ingredients were included in the basal diets at levels of 100 and 200, and 150 and 300 g/kg, respectively, at the expense of maize starch. Each diet was tested with six replicates of 10 broiler chickens each. The feed intake of the chickens that were fed diets based on basal diet 2 was similar, whereas the feed intake of the chickens that were fed diets based on basal diet 1 differed considerably. The numerical differences in pc AA digestibility determined with basal diet 1 or 2 ranged from 2.6 percentage points to 20.8 percentage points in rapeseed cake and from 0.5 percentage points to 15.2 percentage points in soya beans. Across all measured AAs, the average differences were 10.1 percentage points and 5.4 percentage points in rapeseed cake and soya beans, respectively. The differences in the estimated pc AA digestibility between the basal diets were probably caused by different basal endogenous AA losses in the digesta between treatments as a consequence of different feed intake. Adequate AA concentrations and test ingredient levels that are specifically adjusted to avoid a negative effect on feed intake are recommended for future studies.

A regression approach was applied to determine the influence of feed provisioning prior to digesta sampling on precaecal (pc) amino acid (AA) digestibility in broiler chickens. Soybean meal was used as an example test ingredient. Five feed-provisioning protocols were investigated, four with restricted provision and one with ad libitum provision. When provision was restricted, feed was provided for 30 min after a withdrawal period of 12 h. Digesta were sampled 1, 2, 4 and 6 h after feeding commenced. A diet containing 300 g maize starch/kg was prepared. Half or all the maize starch was replaced with soybean meal in two other diets. Average pc digestibility of all determined AA in the soybean meal was 86% for the 4 and 6-h protocols and 66% and 60% for the 2 and 1-h protocols, respectively. Average pc AA digestibility of soybean meal was 76% for ad libitum feed provision. Feed provisioning also influenced the determined variance. Variance in digestibility ranked in magnitude 1 h > ad libitum > 2 h > 6 h > 4 h for all AA. Owing to the considerable influence of feed-provisioning protocols found in this study, comparisons of pc AA digestibility between studies applying different protocols prior to digesta sampling must be treated with caution. Digestibility experiments aimed at providing estimates for practical feed formulation should use feed-provisioning procedures similar to those used in practice.

No abstract