Prediction of energy requirement for growing sheep with the Cornell Net Carbohydrate and Protein System.

This study evaluates the suitability of the Cornell Net Carbohydrate and Protein System for sheep (CNCPS-S) to predict average daily gain (ADG) of lambs. This model was also used to compare the efficiency of use of metabolizable energy (ME) to net energy (NE) for growth (k_g) from the Agricultural Research Council (ARC, 1980), the Australian system (CSIRO, 1990), the National Research Council (NRC, 2000) and a theoretical equation by Tedeschi et al. (2004), which uses a decay equation as a function of the composition of the gain. In addition, the equations used by ARC (1980), NRC (1985) and CSIRO (1990) to predict the energy content of empty body gain (EVG) were compared. Forty-two data points from nine published studies were used to investigate the adequacy of CNCPS-S and of the above equations to estimate ADG. Regardless of the k_g prediction equation used, the CNCPS-S markedly underpredicted ADG, due to an overprediction of ME requirements for maintenance (ME_m). When the factors causing overprediction of ME_m were corrected, the CNCPS-S underpredicted ADG when the NRC (1985) and CSIRO (1990) equations were used to estimate k_g, while good precision and accuracy were achieved when k_g was predicted with the Tedeschi et al. (2004) and ARC (1980) equations. With the ARC (1980) equation, the CNCPS-S model explained 82% of the variation in ADG, with small mean bias (-4 g/day) and root mean squared predicted error (RMSPE) (40 g/day); the simultaneous test of the intercept and slope did not reject (P>0.1) the hypothesis that they were statistically similar from zero and unity, respectively. The model had an accuracy of 0.90 when evaluated with the concordance correlation coefficient test. The comparison of three different equations to predict EVG indicated that the best CNCPS-S prediction of ADG was obtained with the CSIRO (1990) approach. We concluded that a modified CNCPS-S model can be used to accurately predict ADG of growing lambs.