Introduction

Sow mortality is an important economic and animal welfare concern for pork producers. Mortality is an economic drain on sow farms (Duran, 2001). In addition high sow mortality affects employee morale (Deen & Xue, 1999) and is an important welfare indicator (Broom, 1996). Currently, the suggested target for herds of more than 200 sows is to keep annual rates below 5%, and some researchers advocate increasing this target to 8 to 10% for herds of more than 1,200 sows (Duran, 2001). A

Sow mortality is affected by factors related to health, management, and environment. The periparturient period, defined as the last 10% of the gestation period and the first few weeks post-partum, is a time of high risk for sows with over 50% of mortality occurring during this time (Duran, 2001; Anil et al., 2005). Parity, farrowing induction, stillbirths, and season have all been identified as periparturient factors influencing sow mortality in previous studies (Deen & Xue, 1999; Koketsu, 2000; Deen, 2003; Chagnon et al., 1991; D’Allaire eta al., 1996). While not all the factors associated with sow mortality can be controlled, understanding them will assist producers in minimizing death loss. The present retrospective study analyzes the association of induction and other periparturient risk factors with sow mortality.

Materials and methods

This study involved 312,000 parity records from 16 U.S. commercial farms retrieved from the PigCHAMP database (PigCHAMP Inc. Ames, Iowa) for sows serviced between January 2001 and December 2004.

The production outcome evaluated was sow death (including euthanasia). Data on parity (1-2, 3-5, > 5), day of farrowing (weekday or weekend), pigs born alive per litter (continuous), stillbirths (0, > 0), mummies (0, > 0), season of farrowing (summer, other), induced (yes, no), and number of services (1, > 1) were retrieved from the PigCHAMP databases of the respective farms.

A multivariate logistic (Proc Logistic) regression models, with farm as a random effect variable (Glimmix macro), were fitted to evaluate the association of the selected variables with sow mortality. All statistical analyses were performed using the statistical software package, SAS v. 8.2 2001 (SAS Institute, Cary, N.C.). The reported reasons for sow removals were also collected from the PigCHAMP database, and the proportion of removals was calculated. Death and culling rates were obtained from the PigCHAMP performance monitor reports.

Results and discussion

Over the four-year period, the sow death rate on the farms in this study ranged from 4.8% to 19.1% with a mean of 8.9%. Farm records indicated that 24.9% of sow deaths were due to problems related to farrowing. Other studies have also found this period to be critical to sows as more than 50% of mortality occurs in this time period (Duran, 2001; Deen & Xue, 1999; Anil et al., 2005; Chagnon et al., 1991).

When we applied the multivariate model, we found that parity is positively associated with sow death. This is consistent with previous studies. However, individual farm culling practices have great influence on the type of older sow removals (Deen & Xue, 1999).

Sows having litters with stillborn pigs were more likely to die than those with no stillborn pigs (P < 0.0001). Litter stillborns are related to both the viability of the pigs born and the attention given to sows during parturition (Holyoake et al., 1995). Thus the relationship of stillborns and mortality is reflective of both sow condition and management of the farrowing barn.

Only a few studies have addressed sow injury or mortality associated with farrowing induction. Studies examining different induction protocols have shown an increased need for assistance and interrupted piglet delivery with certain common induction drugs (Yang et al., 1996; Kirkwood & Aherne, 1998; Cassar et al, 2005; Kirkwood, 1999). These effects are related to more painful, stressful delivery, and greater chance for sow injury (Kirkwood & Aherne, 1998; Cassar et al., 2005), leading to further complications and sow death. In our study, sows that were induced were 11% more likely to die than those that farrowed naturally (P = 0.002).

The presence of mummies in a litter, farrowing on a weekend or weekday, number of services required for conception, and total-born-alive litter size were not associated with sow mortality, when controlling for other factors.

In conclusion, this study found measurable sow attributes, such as parity, stillbirths and induction, are significantly associated with sow mortality. Attention should be given to the periparturient sow. Inducing farrowing can increase sow mortality risk and should not be used without justification and care. This study is a retrospective observational study using computerized records from commercial farms. The results could be affected by housing, nutrition, environment, and genotype which were not measured. Results could be further biased because the data was both producer recorded and volunteered. Discrepancies between the type of removal (death, euthanasia, or culling) and reported reasons for removal were noted and are a potential source of errors.

References

1. Duran, C.O. 2001. Sow Mortality. Compendium on continuing education for the practicing veterinarian. 23(8): S76-S84.
2. Deen, J. and J. Xue. 1999. Sow mortality in the US: An industry-wide perspective. Proceedings Allen D. Leman Swine Conference. pp.91–94.
3. Broom, D.M. 1996. A review of animal welfare measurement in pigs. Pig News and Information. 17:109N-114N.
4. Anil, S.S., L. Anil, and J. Deen. 2005. Evaluation of patterns of removal and associations among culling because of lameness and sow productivity traits in swine breeding herds. JAVMA. 226(6):956–961.
5. Koketsu, Y. 2000. Retrospective analysis of trends and production factors associated with sow mortality on swine-breeding farms in USA. Preventive Veterinary Medicine. 46:249–256.
6. Deen, J. 2003. Periparturient mortality. Allen D. Leman Swine Conference. P203–204.
7. Chagnon, M., S. D’Allaire, and R. Drolet. 1991. A prospective study of sow mortality in breeding herds. Can J Vet Res. 55:180–184.
8. D’Allaire, S., R. Drolet, and D. Brodeur. 1996. Sow mortality associated with high ambient temperatures. Canadian Veterinary Journal. 37:237–239.
9. Holyoake, P.K., G.D. Dial, T. Trigg, and V.L. King. 1995. Reducing Pig Mortality Through Supervision During the Perinatal Period. J. Anim. Sci. 73:3543–3551.
10. Yang, Ping-Cheng, Wen-Der Fang, San-Yuan Huang, Wen-Bin Chung, and Walter H. Hsu. 1996. Farrowing Induction with a combination of Prostaglandin F2alpha and a peripherally acting 2- Adrenergic agonist AGN 190851 and a combination of prostaglandin F2alpha and Oxytocin. Theriogenology. 46(7–8):1289–1293.
11. Kirkwood, R.N. and F.X. Aherne. 1998. Increasing the predictability of cloprostenol-induced farrowing in sows. Swine health and Production 6(2):57–59.
12. Cassar, G., R.N. Kirkwood, R. Friendship, Z. Poljak. 2005. Sow and litter performance following farrowing induction with prostaglandin: Effect of adjunct treatments with dexamethasone or oxytocin. Journal of Swine Health and Production. 13(2): 81–85.
13. Kirkwood, R.N. 1999. Pharmacological intervention in swine reproduction. Swine Health and Production. 7(1):29–35.

October 2007