043 Effects of standardized ileal digestible valine-to-lysine ratio on growth performance of twenty-five– to forty-five–kilogram pigs under commercial conditions

Two experiments were conducted to estimate the standardized ileal digestible (SID) Val:Lys requirement for growth performance in 25- to 45-kg pigs. In Exp. 1, 1134 gilts (PIC 337), initially 31.2 kg (SD 2.0) BW, were used in a 19-d trial with 27 pigs/pen and 7 pens/treatment. In Exp. 2, 2100 gilts (PIC 327), initially 25.4 ± 1.9 kg BW, were used in a 22-d trial with 25 pigs/pen and 12 pens/treatment. In both experiments, treatments were blocked by initial BW in a randomized complete block design. In Exp. 1, there were 6 treatments with SID Val:Lys at 59.0, 62.5, 65.9, 69.6, 73.0, and 75.5%. For Exp. 2, there were 7 treatments with SID Val:Lys at 57.0, 60.6, 63.9, 67.5, 71.1, 74.4, and 78.0%. Diets were formulated to ensure that Lys was the second limiting AA throughout the experiments. Responses were analyzed separately for each experiment using general linear and nonlinear heteroskedastic mixed models, including initial BW as an explanatory covariate and BW block as a random effect. In Exp. 1, ADG linearly increased with increasing SID Val:Lys (P= 0.009; 680, 717, 717, 712, 744, and 726 ± 17.1 g, respectively), whereas no significant treatment differences were observed for G:F (0.467, 0.467, 0.472, 0.474, 0.481, and 0.472 ± 0.0084, respectively). In Exp. 2, ADG (quadratic, P= 0.002; 621, 662, 717, 708, 708, 726, and 717 ± 16.1 g, respectively) and G:F increased (linear, P< 0.001; 0.415, 0.420, 0.437, 0.429, 0.433, 0.441, and 0.439 ± 0.0046, respectively) with increasing SID Val:Lys. There was no evidence of experiment × treatment interaction. Therefore, data from the 2 experiments were combined for analysis using experiment and BW block within experiment as random effects. Competing models, namely a broken-line linear model, a broken-line quadratic model, and a quadratic polynomial (QP), were compared using Bayesian information criterion. In the combined analysis, the best-fitting model for ADG was a QP (prediction equation: −1.15 + 4.13 × SID Val:Lys − 2.78 × SID Val:Lys2+ 0.012 × initial BW) with optimum ADG estimated at 74.4% (95% confidence interval [CI] 69.5 to > 78.0) SID Val:Lys. The best-fitting model for G:F was also a QP (prediction equation: −0.04 + 1.36 × SID Val:Lys − 0.94 × SID Val:Lys2) with optimum G:F estimated at 72.3% (95% CI 64.0 to > 78.0) SID Val:Lys. In conclusion, 67% SID Val:Lys was able to capture 99% of maximum ADG and G:F in 25- to 45-kg pigs.