Comparison of the regression analysis technique and the substitution method for the determination of true phosphorus digestibility and faecal endogenous phosphorus losses associated with feed ingredients for growing pigs

Two digestibility experiments were carried out to comparatively measure true phosphorus (P) digestibility and the faecal endogenous P loss associated with soybean meal (SBM) and wheat middling meal (WM) for growing pigs by the simple linear regression analysis technique and the substitution method. Eight barrows, with an average initial body weight 21 kg, were randomly divided into two groups of four pigs with each group being fed four test diets according to a 4 × 4 Latin square design. Eight maize starch-based diets containing four levels (on dry matter basis) of P (0.20, 0.24, 0.36 and 0.42%) and P (0.181, 0.26, 0.36, and 0.43%) were formulated from SBM and WM as a sole source of P, respectively. Chromic oxide (0.30%) was used as a digestibility marker. Each experimental period comprised 8 d with a 4-d adaptation and a 4-d collection of faecal samples. Levels of the assay ingredient substitutions had no effects (P > 0.05) on true P digestibility and the endogenous P loss associated with SBM and WM measured by the substitution method. There were no differences (P > 0.05) between the regression technique and the substitution method in measuring the true P digestibility (SBM: 49.4 ± 3.51 vs. 50.6 ± 2.4; WM: 63.7 ± 5.0 vs. 63.2 ± 5.4%) as well as the faecal endogenous P loss (SBM: 0.62 ± 0.10 vs. 0.71 ± 0.11; WM: 0.65 ± 0.16 vs. 0.70 ± 0.17 g/kg dry matter intake) for the growing pig. Thus, true P digestibility and the faecal endogenous P loss in feed ingredients for pigs may be measured by using the regression analysis technique and the substitution method.