Effect of housing system during pregnancy on reproductive parameters of sows

The aim of this study was to evaluate the reproductive performance of sows in individual stalls (Stall) or group-housed into groups on days 3 to 5 (Pens) or 38 to 42 (Pen42) after breeding. The reproductive data was collected from the operating system of the farm establishing the average of the weekly performances of the sows that gave birth, for four years, except for the PenS system, which was evaluated for three years. Statistical analyses were performed using SAS (R), using MIXED (for quantitative traits), NPAR1WAY (for categorical traits) and LOGISTIC (binomial traits) procedures, using sows as a repeated measure. Sows in Stall system had piglets with higher birth weight them sows in the group-housed system (P<0.05), however the piglets born alive, total piglets weaned, average litter weight, and duration of farrowing were lower in this system than in the group-housed (P<0.0001). There was no difference between PenS and Stall systems for gestation period and mummified piglets. The Pen42 system had a higher percentage of mummified piglets and a shorter gestation period, when compared PenS and Stall systems (P<0.0001), and similar results to the PenS system for duration of farrowing, piglets born alive, stillbirths, total piglets weaned, average litter weight and birth weight (P<0.0001).