Abortion occurrence, repeatability and factors associated with abortions in female pigs in commercial herds

• We investigated the abortion occurrences using herd-entry gilt cohorts.
• Abortion rate per year was 3.0% and abortion risk per service was 1.2%.
• Factors for abortions were parity 0 and 5 or higher, summer mating and re-servicing.
• Other factors were having stillbirths or prolonged weaning-to-first-mating interval.
• The repeatability of abortions was estimated to be 24.3–24.5%.

Abortion occurrences disturb the projected production volume and increase non-productive days of female pigs. The objectives of the present study were (1) to define abortion occurrences including abortion rate and abortion risk, (2) to examine risk factors that were associated with abortions, (3) to determine repeatability of abortions and (4) to compare reproductive performance between aborting and non-aborting re-serviced female pigs. Datasets included 122,935 lifetime performance records and 630,363 mating performance records of females in 125 herds between 2008 and 2013 in Spain, Portugal and Italy. Annualized abortion rate was defined as the number of abortion records divided by the sum of the reproductive herd life days×365 days×100. Abortion risk per service was defined as the number of abortion records divided by the number of service records×100. Generalized linear models with random herd effect were applied to the data. Annualized abortion rate (±SE) and abortion risk per service were 3.0±0.05 and 1.2±0.01%, respectively. Factors for an increased abortion risk per service were parity 0 or parity 5 or higher, re-servicing, servicing in July and August, delivering stillborn fetuses in a previous litter and having prolonged weaning-to-first-mating interval (WMI; P<0.05). Gilts (parity 0) and sows in parity 5 or higher had 0.3% higher abortion risk than parity 1 sows (P<0.05). Re-serviced females had 0.5% higher abortion risk than first-serviced females (P<0.05). Females serviced in July and August had 0.4–0.5% higher abortion risk than those serviced in March and April (P<0.05). Sows that had previously delivered stillborn fetuses subsequently had 0.1% higher abortion risk than those that had no stillborn fetuses (P<0.05). Sows mated 7 days or more after weaning had 0.2% higher abortion risk than those mated 0–6 days after weaning (P<0.05). Of the 7187 females that had aborted once, 4.1% experienced a second abortion in the same or a later parity. The repeatability of abortions was estimated to be 24.3–24.4%. Females that were reserviced after aborting had 0.6 lower total number of piglets born than re-serviced females that had not aborted (P<0.05). However, there was no difference in farrowing rate or WMI between aborting and non-aborting re-serviced females (P≥0.11). It is recommended that producers closely monitor female pigs at risk of having an abortion in order to reduce non-productive female days.