| In present study, using Hy-Line Brown rooster as experimental animal, the correlation between simulated and in vivo digestion in gizzard or gastrointestinal tract, and the correlation between bionic digestion method and empting-force feeding method for determing metabolizable energy of feed were investigated. The sensitivity and confidence limit were compared between bionic digestion method and empting-force feeding method. Based on these results, the bionic digestion method was evaluated for its application to accurately determine the metabolizable energy of feeds for rooster. Three experiments were conducted in this study.Experiment1was to compare the energy digestibility of feed in gizzard and gastrointestinal tract of Hy-Line Brown rooster determined using in vivo and in vitro method. Two samples completely randomized design was adopted. Two treatments were consisted of in vivo and in vitro energy digestibility. In in vivo method, each determination contained5replicates of3roosters. In in vitro method, each determination contained5replicates of1digestion tube. The in vivo and in vitro energy digestibilities of corn, soybean meal, wheat bran and wheat were determined. The result showed that in in vivo method, the apparent energy digestibilities of corn, soybean meal, wheat bran and wheat in gizzard were-7.49±9.08%,16.77±17.66%,9.12±11.37%,3.23±5.26%,-3.06±8.21%from the rooster force-fed25g feed, and-24.53±11.37%,-3.17±12.59%,-3.80±6.94%from the rooster force-fed50g feed, respectively. The apparent energy digestibilities of corn, soybean meal, wheat bran and wheat in gastrointestinal tract were74.58±4.06%,49.96±5.55%,38.88±3.25%,75.39±1.05%from the rooster force-fed25g feed, and75.87±3.02%,46.60±6.90%,38.09±10.52%,76.00±1.14%from the rooster force-fed50g feed, respectively. In bionic digestion method, the apparent energy digestibilities of corn, soybean meal, wheat bran and wheat were3.36±0.50%,45.34±0.73%,20.59±0.34%,15.74±0.81%in gizzard, and83.50±0.44%,65.99±0.26%,48.65±0.60%,79.88±0.41%in gastrointestinal tract, respectively. The coefficient of correlation between in vitro apparent energy digestibility in gizzard and in vivo apparent energy digestibility in gizzard from the rooster force-fed25g feed, and between in vitro apparent energy digestibility in gizzard and in vivo apparent energy digestibility in gizzard from the rooster force-fed50g feed were0.55(P<0.05) and-0.66(P<0.05), respectively. The coefficient of correlation between in vitro apparent energy digestibility in gastrointestinal tract and in vivo apparent energy digestibility in gastrointestinal tract from the rooster force-fed25g feed, and between in vitro apparent energy digestibility in gastrointestinal tract and in vivo apparent energy digestibility in gastrointestinal tract from the rooster force-fed50g feed were0.93(P<0.01) and0.90(P<0.01), respectively. There was high correlation between in vitro and in vivo digestion of gastrointestinal tract in rooster.Experiment2was to test the correlation between bionic digestion method and empting-force feeding method for determing metabolizable energy of feed ingredients, then to evaluate the feasibility and accuracy of development of EHGE from bionic digestion method to predict the AME of feed. There were14calibration diets for development of predict model and8validation diets for testing the model. The AME and EHGE of diets were determined using empting-force feeding method and bionic digestion method, respectively. In AME determination,96Hy-Line Brown roosters were randomly divided into4groups. Each group contained6replicates of4roosters per replicate. The diets were randomly selected and assayed in one of three simulated digestion system to determine the EHGE. Each determination contained5replicates of1digestion tube per replicate. The result showed that the mean of AME and EHGE of14calibration diets were13.15±0.69MJ/kg and13.82±0.80MJ/kg DM, respectively. The mean of AME and EHGE of8test diets were13.13±0.24MJ/kg DM and13.83±0.17MJ/kg DM, respectively. The regression equation to predict AME by EHGE was AME=0.829×EHGE+1.688(R2=0.95, RSD=0.21MJ/kg DM, P<0.01). In14calibration diets, the deviation between AME and AMEp ranged from-0.26to0.19MJ/kg DM. In8validation diets, the deviation between AME and AMEp ranged from-0.36to0.33MJ/kg DM. The AMEp of validation diets were all in the95%confidence interval of regression equation developed from calibration diets. Based on these results, the regression equation was suitable to estimate the AME of validation diets.Experiment3was to compare the sensitivity and the confidence limit between in empting-force feeding method (GB/T26437-2010) and in bionic digestion method to further test the feasibility of bionic digestion method. In trial1, a2x5completely randomized design with2levels of basal diets and5levels of diluents (2%、4%、6%、8%、10%of peanut hull in diets) was adopted to determine the response of determined AME using empting-force feeding method to the calculated AME, and the response of detemined EHGE using bionic digestion method to the calculated EHGE. In trial2, according to the variances of AME or EHGE of10diets in trial1and blank samples, the confidence limit was determined for empting-force feeding method or bionic digestion method to differentiate the AME or EHGE values between two samples. The result showed that the mean of AME and EHGE of10dites were13.10±0.63MJ/kg DM, and13.77±0.78MJ/kg DM, respectively. The mean of AME of4blank samples was-1.38±0.28MJ/kg DM, and the mean of EHGE of3blank sample was-0.44±0.04MJ/kg DM. The sensitivity expressed with dAME/dAMEc in empting-force feeding method was0.987, and the sensitivity expressed with dEHGE/dEHGEc in bionic digestion method was1.115. Calculated from the variance of repeated determination, the confidence limit of empting-force feeding method and bionic digestion method were0.27MJ/kg DM and0.08MJ/kg DM, respectively.In conclusion, the bionic digestion method can accurately predict the AME of diets in Hy-Line Brown rooster. Comparing with empting-force feeding method, bionic digestion method has higher sensitivity and lower confidence limit of detection. |
PDF Full Text Request