Processing weights of chickens determined by dual-energy X-ray absorptiometry: 1. Weight changes due to fasting, bleeding, and chilling

The increasing breast weight of modern broilers highlights the convenience of studying the dynamics of processing weights across the grow-out period. For models predicting these traits to work with non-fasted live birds under longitudinal studies, it is necessary to determine the changes in the weights of birds due to fasting and bleeding and carcasses due to chilling. This study aimed to develop a model to predict the fasting weight loss of broilers, determine the weight changes due to bleeding and chilling, and assess the percentage of gizzard fat in the abdominal fat. Models were fitted to fasting BW loss data from literature. A bleeding BW loss value obtained in a pilot study was compared to literature. This one and a single-source percent gizzard in the abdominal fat value were used as reference values. Birds from two experiments were selected at different ages to determine the following variables: non-fasted and fasted BW, bled weight, unchilled and chilled carcass weights, and gizzard fat and abdominal fat pad weights. Predicted and reference values were compared to the observed data based on their 95% confidence intervals, and the fitted model and reference values were validated if no statistical differences were inferred. The fasting BW loss model was refitted to the data from experiments, and the estimates of the model fitted to both datasets were compared. The fasting BW loss followed a three-parameter exponential model with reducing values as the bird aged, and the predicted and observed values showed no statistical difference (P > 0.05). The reference bleeding BW loss (2.76%) and gizzard fat percentage (18%) values showed no difference (P > 0.05) to data from broiler experiments, and a chilling carcass weight gain of 2.79% was determined. The scale and relative growth rate estimates of the fasting BW loss model showed no difference (P > 0.05) between datasets, and the asymptote showed a possible but negligible difference. The fasting BW loss model, the bleeding BW loss coefficient, and the gizzard fat percentage were validated. The model and coefficients obtained herein will not only be applied to predict processing weights using Dual-Energy X-ray Absorptiometry but also may be used in other modeling approaches.