250 Associations Between Sow and Litter Parameters and Piglet Birth Weight and the Within-Litter Coefficient of Variation of Piglet Birth Weight

Abstract

Abstract The objective was to evaluate associations between sow, litter, and piglet parameters and both piglet birth weight (BW), and within-litter coefficient of variation of piglet birth weight (CV). Data (1,005 sows/litters, 12,857 piglets) were from 5 studies conducted on the same commercial farrowing facility using the same management. Independent variables included sow parity, use of farrowing induction, number of piglets born alive, litter weight, BW (for analysis of CV), and piglet sex (for analysis of BW) and birth order (for analysis of BW; on a sub-sample of 289 litters/3,643 piglets). Data were analyzed using PROC REG of SAS. Models accounted for linear, quadratic, and cubic effects of each independent variable. The BW and CV averaged 1.49 ± 0.271 kg and 19.5 ± 6.69%, respectively. All independent variables were related (P ≤ 0.05) to BW (Table 1), with litter size having the strongest relationship (R2 = 0.11), and all other variables having weak relationships (R2 ≤ 0.01). All independent variables were related (P ≤ 0.05) to CV, with the exception of farrowing induction. Litter size explained the greatest amount of variation (R2 = 0.17), followed by litter weight and average piglet birth weight (R2 = 0.10 and 0.07, respectively). Other variables each explained a relatively small proportion (R2 ≤ 0.05) of variation in CV. Multivariate models including all significant (P ≤ 0.05) variables explained 13 and 22% of the variation in BW and CV, respectively. In conclusion, the primary variable associated with both BW and CV was litter size. Increases in litter size were associated with reductions in BW and increases in CV. Other variables showed weak relationships. However, even when all of these variables were accounted for, there was a large proportion of variation in both BW and CV that was not explained.