Sorting Out Sow Mortality
Over the past decade, many farms and production systems have experienced problems with high sow death loss. Thorough investigation of sow mortality is an important area where veterinary expertise can add value to client herds. Unfortunately, veterinary intervention may not be sought until mortality rates exceed 10%. Partial budget analysis of the economic benefits of lowering sow mortality may provide substantial evidence to support the cost of veterinary services. The expense associated with sow mortality can be categorized as replacement costs, opportunity costs when a female dies during gestation, and the negative impact that mortality can have on worker morale. Decreased morale is difficult to value, but replacement and opportunity costs can account for $400-$500/sow death. One financial analysis predicted that lowering sow mortality by 4% would save $18 per inventoried female per year. Another study showed a 4:1 return on investment for an intervention strategy aimed at decreasing sow mortality.
Sow mortality rate has been defined as the number of breeding females that died or were euthanized on a farm throughout the year divided by the average female population. Sow death rate has been calculated similarly to mortality rate but does not account for females that were euthanized on the farm. Essential components of a sow mortality investigation should include a thorough herd record evaluation, multiple necropsy examinations, and an epidemiologic evaluation of risk factors present on the farm. A systematic approach to evaluating sow mortality has been described previously in detail.
The most common causes of sow mortality are discussed below.
Heart failure: Several criteria have to be taken into account to make a diagnosis of heart failure. First, other causes of death must be excluded and evidence of heart failure such as transudate in body cavities, pericardial effusion, pulmonary edema, passive congestion of lungs, liver, kidneys, and spleen, must be observed or altered cardiac chamber measurements must be observed. In a prospective study, heart failure accounted for 31.4% of all sow deaths. Approximately 2/3 of these females died during the peripartum period, defined as the period from 3 days prior to 3 days after farrowing. Ambient temperatures greater than 32.0oC (89.6oF) have also been associated with an increased risk of heart failure in sows. One recent conflicting study suggested, however, that heart failure was not a common cause of sow mortality because microscopic lesions in the ventricular wall or interventricular septum were rare.
Torsions and other accidents of abdominal viscera: The diagnosis of torsions and other accidents of abdominal viscera is based on characteristic gross findings, including gastric dilation and torsion, splenic torsion, hepatic torsion, intestinal volvulus, intussusception, herniation, and intestinal rupture. No correlation was established between death from torsions or other abdominal accidents and housing type or feeding management. Deaths due to gastric dilation occurred most frequently within a few hours after feeding. One study concluded that feeding three times per day resulted in a lower risk of sow mortality than feeding twice per day. A genetic predisposition for gastric torsion was suspected when all sows in a herd that died from this condition were traced back to a single Landrace boar. Once the boar was removed, no further cases of death from gastric torsion were reported.
Cystitis-pyelonephritis: Urinary tract infections in sows are usually the result of ascending bacterial infections. Isolates obtained from cases of cystitis-pyelonephritis included Escherichia coli, Actinobaculum (Eubacterium) suis, Proteus spp, Klebsiella spp. and Enterococcus faecalis. One study showed that the mean aqueous humor area concentration from sows that died from cystitis-pyelonephritis was 52.3 mmol/L compared to 9.9 mmol/L observed in sows that died of other causes. In this study, the types of waterer did not correlate with the sow deaths attributed to cystitis-pyelonephritis, although adequate water supply is thought to decrease incidence of urinary tract infections.
Clostridium novyi infection: C. novyi infection may cause sudden death in pigs. Gross lesions reported in these cases include a large, bronze-colored, emphysematous liver, generalized edema, subcutaneous emphysema, foul-smelling sanguinous effusion in body cavities and pericardium, and congested submandibular and superficial inguinal lymph nodes. The diagnosis can be difficult in sows that have been dead for an extended period of time prior to necropsy because clostridial bacteria proliferate rapidly after death. Fluorescent antibody (FA) tests on liver smears were reported to have sensitivity superior to that of culture; however, the frequency of false positive FA test results increased as duration between death and necropsy increased. C. novyi types A and B have been isolated from sows with the lesions described above. Death occurs as the result of toxemia.
Endometritis: The diagnosis of endometritis is based on characteristic gross lesions in the uterus, for instance edematous mucosa and presence of exudate, and on the isolation of the etiologic agent. The most common isolates from cases of endometritis have included E. coli, group C streptococci, and Arcanobacterium (Actinomyces) pyogenes. Severe PRRS outbreaks have been reported to cause death losses ranging from 5-10% over the course of the outbreak. This increased mortality may partly result from increased bacterial endometritis and septicemia associated with retained fetuses.
Pneumonia: The diagnosis of pneumonia is based on distinctive changes in the lungs, including abnormal color and texture of the pulmonary parenchyma and frequently presence of exudate, as well as on the isolation of the specific etiologic agent. Bacterial isolates from sows that died of pneumonia have included Actinobacillus pleuropneumoniae, Pasteurella multocida, Arcanobacterium (Actinomyces) pyogenes, Streptoccus spp., and Klebsiella spp.
Gastric ulcers: The diagnosis of gastric ulcers is based on the characteristic gross lesions consisting of mucosal defects in the stomach. The etiology of the gastric ulcers has been an area of recent controversy. Small particle size of cereal grains, pelleting, and grain type used in the food, stressful conditions, and heritability have been implicated as predisposing factors for the gastric ulcer development. Speculation has been made regarding Helicobacter spp. infection as a cause of gastric ulcers in pigs, but this has not yet been clearly established.
Downer sows: The pathologic changes observed in downer sows include vertebral abscesses, vertebral fractures, arthritis, osteochondrosis, osteomalacia/osteoporosis, and degenerative joint disease. One study evaluating sow mortality reported that musculoskeletal problems were the cause of death or reason for euthanasia in 38.2% of all sows. Histologically, necrosis of muscle fibers and an associated inflammatory response have raised questions of whether myodegeneration is a primary process, i.e., a direct result of injury, or if it occurs secondary to a debilitation process. Regardless of the cause, the end result is an animal that is unable to rise, ultimately leading to natural death or euthanasia.
Enteropathy:: Causes of sow death attributable to pathologic conditions of the intestine include swine dysentery, idiopathic hemorrhagic enteropathy, proliferative hemorrhagic enteropathy, salmonellosis, and cecal perforation. One study reported that all cases of proliferative hemorrhagic enteropathy were observed in parity zero females.
Uterine prolapse: The diagnosis of uterine prolapse is based on characteristic gross findings. This condition is likely underreported since it is often already diagnosed on the farm with no need for further examination by a veterinarian.
Septicemia: The diagnosis of septicemia is based on bacterial culture of spleen, liver, kidney, or other affected organs. Isolates from cases of septicemia have included Erysipelothrox rhusiopathiae, Klebsiella spp., Streptococcus spp., and Salmonella spp.
Effects of parity: One study reported that gilts had the lowest risk of mortality, with the risk of death increasing with each subsequent increase in parity. Authors of another study, however, reported a substantial variation in parity effects on risk of sow mortality. The initial parity had high mortality risk, but older sows were also at high risk. These authors concluded that sow mortality rates in older parities may be dependent on sow culling practices and size of the available gilt pool. Reportedly, younger sows tend to be at high risk of death from endometritis, pneumonia, and downer cow syndrome, whereas older sows tend to be at high risk of death from torsion or other accidents of abdominal organs, cystitis-pyelonephritis, uterine prolapse, and C. novyi infection.
Effects of season: In one study, 1/3 of all sow deaths occurred during July and August. Another study also concluded that the sow mortality in the United States (US) was significantly higher during the summer months of July through September. A Canadian study retrospectively evaluated sow deaths during the seven consecutive days in 1994 when the average temperature was the highest. Three of these seven days made up only 0.8% of the year, but accounted for 11% of the annual death loss.
Effects of management: A study evaluating the production data obtained in 1997 from 604 US farms concluded that a higher risk of sow mortality was associated with a larger herd size. A case-control study reported that multiplier herds were more likely to have higher sow mortality than commercial herds. Lactation length of 28 days or more resulted in a higher risk of mortality, but shorter lactation lengths have also been shown to result in higher mortality. High herd prevalence of metritis, urinary tract infections, or lameness were also associated with high levels of sow mortality rate.
Effects of stage in reproductive cycle: It has been reported that 42% of all sow deaths occurred in the peripartum period, 35% occurred during gestation, 16.5% during lactation, and 6% post-weaning. In another study, nearly half of all mortalities occurred within the first three weeks after farrowing, and a sow had 24% more chance of dying if she had one or more stillbirths.
Conclusion: Sow mortality can be a large expense for the modern swine producer. Systematic diagnostic intervention, including proper record analysis, multiple necropsies, and evaluation of the risk factors, are necessary to determine the specific causes of this problem and to formulate intervention strategies. Fortunately, veterinarians are in a unique position to provide valuable diagnostic services to farms experiencing problems with sow mortality. Often a small decrease in mortality can lead to a big increase in farm profitability.
-by Jason Kelly, Class of 2003
-edited by Ingeborg Langohr, ADDL Graduate student
Abiven N, Seegers H, Beaudeau F et al: 1998. Risk factors for high sow mortality in French swine herds. Prev Vet Med 33: 109-119.
Chagnon M, D'Allaire S, Drolet R: 1991. A prospective study of sow mortality in breeding herds. Can J Vet Res 55:180-184.
D'Allaire S, Drolet R, Brodeur D: 1993. Sow mortality associated with high ambient temperatures. Can Vet J 37: 237-239.
D'Allaire S, Drolet R, Brodeur D: 1991. The causes of sow mortality: a retrospective study. Can Vet J. 32:241-243.
Deen J, Xue J: 1999. Sow mortality in the US: an industry-wide perspective. ProcAllen D. Leman Swine Conf. pp 91-94.
Drolet R, D'Allaire, Chagnon M: 1992. Some observations on cardiac failure in sows. Can Vet J 33: 325-329
Duran CO: 2001. Sow mortality. Comp Cont Ed Pract Vet. 23:S76-S84.
Duran Co, Walton JR: 1997. Clostridium novyi sudden death in sows: toxaemia or post mortem invader? Pig J 39:37-53.
Duran CO: 1994. Causes of sow mortality. Pig J 32: 107-112.
Friendship RM: 2000. Gastric ulceration in swine: overview and infectious etiology control. Proc Allen D Leman Swine Conf. pp. 92-95.
Geiger J, Pretzer IC,: 1999. Assessing sow mortality. Proc Allen D Leman Swine Conf pp. 84-87.
Glock RD, Geiger JO, Henry SC: 1999. Microscopic lesions associated with sow mortality. Proc Allen D. Leman Swine Conf. pp. 88-90.
Halibur P, Bush E: 1997. Update on abortion storms and sow mortality. J Swine Health Proc. 5(2): 73.
Hancock JD: 2000. The role of feed processing in development of gastric ulcers. Proc Allen D. Leman Swine Conf. pp 98-99.
Koketsu Y: 2000. Retrospective analysis of trends and production factors associated with sow mortality on swine-breeding farms in USA. Prev Vet Med 46:249-256.
Pretzer SD, Irwin CK, Geiger JO, Henry SC: 2000. A standardized protocol to investigate sow mortality. J Swine Health Prod 8(1): 35-37.
Sanz M, Roberts J, Almond G et al: 2002. What we see with sow mortality. Proc Allen D. Leman Swine Conf. pp.181-184.
Schultz R, Dau D, Cast W et al: 2001. A sow mortality study - the real reason sows die. Proc AASV. pp 387-395.
Smith W: 1984. Sow mortality-limitedsurvey. Proc IPVS. p. 368.