Bioelectrical impedance analysis as a non-invasive tool to predict body composition in pullet layers

This study aimed to develop and validate bioelectrical impedance analysis (BIA) equations to predict body composition (BC) in pullets from hatch to 16 weeks of age (WOA). A total of 285 birds were reared under ad libitum feeding and received common starter (from 0 to 4 WOA), grower (from 5 to 10 WOA), and developer (from 11 to16 WOA) diets containing 2836, 2739, and 2691 kcal AME/kg; 3.04, 3.00 and 3.07% ether extract; and 19.5, 16.9 and 14.8% crude protein, respectively. Bioelectrical impedance analysis measurements were taken biweekly at nine timepoints (0, 2, 4, 6, 8, 10, 12, 14, and 16 WOA). Birds were euthanized and analyzed for dry matter, protein, fat, ash, and energy content through proximate analysis. Water (% of BW) and protein (% DM) decreased with age (76.5 to 56.0 and 67.4 to 52.4, respectively; P < 0.001), while fat (% DM) and energy (cal/g DM) increased (21.8 to 36.0 and 5,663 to 6,104, respectively; P < 0.001). Ash (% DM) rose rapidly from 0 to 2 WOA (8.11 to 11.8) and declined sharply at 14 WOA (9.18; P < 0.001). Resistance and reactance values declined with age, showing week-to-week fluctuations. Multiple linear regression models were developed to predict BC components, yielding R² values of 0.942, 0.716, 0.704, 0.287, and 0.622 and relative mean prediction errors of 1.97, 4.20, 10.5, 8.99, and 2.69% for water, protein, fat, ash, and energy, respectively. No significant differences were found between measured and predicted values (paired t-test). These findings demonstrate that BIA is a reliable, noninvasive technique for estimating body composition in live pullets and may serve as a practical tool for future research on growth and nutritional management.