An investigation of the success of production-based sow removal and replacement in the context of herd performance.

Production-based removal and replacement has been used as a method to improve sow herd performance. Limited data are available as to the reliability of this approach. The purpose of this investigation was to use a retrospective case-control study to assess the success of replacement events when herd productivity was greater or less than the mean for removal events attributed to problems with fertility, fecundity, or old age. For each of 3 herds, 1,000 consecutive sows removed between parities 1 and 6 for reasons of fertility, fecundity, or old age were matched to sows with similar histories that were retained in the herd (controls) and to gilts that were first bred into the herd around the time of the case removal events. Controls and gilts were followed until their next parity or removal event, and the outcome was measured as a standardized calculation of born alive per mated female per year. Herd performance at the time of the case removal events was categorized according to greater or less than the mean for fertility or fecundity on monthly farrowing rates and average piglets born alive per litter. Success of removal/replacement events were evaluated according to removal reason and contemporary herd performance. A model was developed to estimate production and financial implications of changes to productivity-based culling, using a Monte Carlo simulation with a 1,000-iteration run. Born alive per mated female per year from gilts was greater (P = 0.0001) than from controls in 1 of 3 herds when herd fertility was greater than the mean, 1 of 3 herds when herd fertility was less than the mean (P = 0.0065), 3 of 3 herds when herd fecundity was greater than the mean (P < 0.030), and 2 of 3 herds when herd fecundity was less than the mean (P < 0.020). The financial model sensitivity analysis indicated greater likelihood of economic advantage for a scenario without production-based removals in parities 1 to 6.