A descriptive survey of lesions from cull sows harvested at two Midwestern U.S. facilities

Preventive Veterinary Medicine 82 (2007) 198–212 www.elsevier.com/locate/prevetmed

A descriptive survey of lesions from cull sows harvested at two Midwestern U.S. facilities§ M. Knauer a,1, K.J. Stalder a,*, L. Karriker b, T.J. Baas a, C. Johnson a, T. Serenius a, L. Layman b, J.D. McKean b a

Department of Animal Science, 109 Kildee Hall, Iowa State University, Ames, IA 50011-3150, USA b Veterinary Diagnostic and Production Animal Medicine Department, 1710 Veterinary Medicine, Iowa State University, Ames, IA 50011-3150, USA Received 1 November 2006; received in revised form 30 April 2007; accepted 21 May 2007

Abstract Physical and reproductive conditions of cull sows (3158) from two U.S. Midwestern harvest plants were assessed. Body condition, feet, shoulders, teeth, lungs, and reproductive tracts were visually evaluated for gross lesions on harvested sows. PROC FREQ (SAS, Cary, NC) was used to calculate the frequency of each binary trait event. Pearson chi-square tests were used to test the alternative hypothesis that a linear association existed between binary traits and body condition score (BCS). The most common foot lesions observed were rear (n = 2064, 67.5%) and front (n = 1024, 32.9%) heel lesions. Cracked hooves were found on the front feet of 703 (22.6%) and rear feet of 552 (18.1%) sows. Rear digital overgrowth was observed in 644 (21.1%) sows. The most common reproductive gross lesion observed among harvested cull sows was acyclic ovaries (n = 277, 9.0%). Presence of acyclic ovaries increased ( p < 0.01) as BCS decreased. Cystic ovaries were found in 192 (6.3%) sows, which increased ( p < 0.01) as BCS increased. Pneumonia was observed in 298 (9.7%) sows, and increased in frequency as BCS decreased ( p < 0.01). The most frequently observed shoulder lesion among harvested cull sows was shoulder abrasions (n = 394, 12.5%). The presence of shoulder abrasions increased ( p < 0.01) as BCS decreased. The prevalence of reproductive lesions detected in the present study was less than the reported percentage of sows culled for reproductive failure from previous studies based on record keeping summaries. # 2007 Elsevier B.V. All rights reserved. Keywords: Cull sows; Gross lesions; Feet/leg lesions; Body condition score

§ This journal paper of the Iowa Agric. and Home Econ. Exp. Stn., Ames, IA, Project No. 3456, was supported by Hatch Act and State of Iowa funds. Additional funding and support was provided by the National Pork Board, Project No. 04127. * Corresponding author. Tel.: +1 515 294 4683; fax: +1 515 294 5698. E-mail address: [email protected] (K.J. Stalder). 1 Current address: Department of Animal Science, North Carolina State University, 233-D Polk Hall, Raleigh, NC 27695, USA.

0167-5877/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.prevetmed.2007.05.017

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

199

1. Introduction Poor sow longevity in commercial pork production systems can lead to economic inefficiency and animal welfare concerns (Stalder et al., 2004). Swine herds located within the U.S. and that participate in the PigCHAMP record keeping system were rank by their annual breeding female culling rate. The annual breeding female culling rates from the lowest tenth percentile have exceeded 60% in recent years (PigCHAMP, 2002, 2003, 2004). Dagorn and Aumaitre (1979) reported reproductive failure as the most common reason sows are culled from farms. Traditional culling studies are based on retrospective farm data as they are easy and economical to obtain. Producers typically report one reason for culling each sow without reporting of co-morbid conditions. Additionally, these reasons are typically based on external signs or indications and do not incorporate evidence of internal lesions or results from diagnostic testing. Sow harvest plants offer researchers an opportunity to confirm farm data and to characterize and enumerate factors which may cause culling. However, studies investigating the presence of gross lesions in harvest plants are few. Post-mortem examinations of female pig reproductive organs are a potential source of information concerning sow reproductive failure. Foot lesions, body condition, disease status, and other problems may also contribute to sow culling decisions. The objective of this study was to characterize the physical condition and gross lesions of cull sows from U.S. Midwestern sow harvest plants. 2. Materials and methods 2.1. Data collection Harvest facilities and hence the cull sow populations represented by the harvest facilities were chosen to represent the range of sows culled from U.S. pork operations. Because the type of sows harvested by individual facilities is driven by varied purchasing criteria based on anticipated output products from those facilities, we chose two different Midwestern plants for data collection. Plant 1 (n = 1321 sows) harvested a ‘‘leaner’’ type of sow and plant 2 (n = 1837 sows) harvested a ‘‘fleshier’’, meatier type of sow. Based on body condition score, ultrasonic backfat and loin muscle area, and weight, we achieved our original goal of evaluating a wide variety of cull sows. At each designated sampling day, 100% of the cull sows harvested were evaluated by the investigators. The cull sows evaluated always represented multiple sources and multiple lots on any given day—harvest facility combination. Twelve trips, six to each harvest plant, were made between 7 April and 29 September 2005. All trial work was previously reviewed and approved by the Iowa State University Animal Care and Use Committee. All harvest occurred under USDA-FSIS inspection. Body condition, feet, shoulders, teeth, lungs, and reproductive tracts were visually evaluated for gross lesions on harvested sows. The following body composition traits were measured and recorded: backfat, longissimus dorsi area, longissimus depth, and body condition score (BCS). A National Swine Improvement Federation certified real-time ultrasound technician measured backfat, longissimus dorsi area, and longissimus dorsi depth from a cross sectional 10th rib image using an Aloka 500V SSD ultrasound machine (Corometrics Medical Systems, Inc., Wallingford, CT). At both plants, ultrasound evaluations were performed ante mortem while sows were restrained in a chute prior to harvest. Body condition score was evaluated using a scale of 1 (thin) to 5 (fat) (Patience and Thacker, 1989). At plant 1, BCS was evaluated as sows were suspended from a gambrel with their backs facing the observer. At plant 2, BCS was evaluated just prior to stunning as the animal was

200

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

in a standing position. Individuals evaluating BCS were consistent across plants and visits, and had extensive livestock evaluation experience. Front and rear foot lesions were evaluated and recorded by a trained technician. The presence of claw and hoof wall cracks (cracked hooves) included side-wall lesions, cracks in the white line, and toes (Gjein and Larssen, 1995a). The presence of pad (heel) lesions as described by Gjein and Larssen (1995a) was assessed. Feet were examined for the presence or absence of abscesses on any surface of the foot. Both hooves and dew claws were evaluated for abnormal overgrowth conditions (digital overgrowth) as defined by upward or inward curvature of the toes or excessive hoof growth (2 cm longer than normal toe). Missing dew claws were recorded where observed. The two people that evaluated feet had extensive experience in evaluating feet and leg soundness of pigs. Shoulder lesions were evaluated and assigned into one of the following categories ‘‘none,’’ ‘‘abscess,’’ ‘‘abrasion,’’ or ‘‘open’’ lesions. Lesion score was classified as ‘‘none’’ if the skin appeared normal over the point of the shoulder, an ‘‘abscess’’ lesion score was assigned upon the presence of an open or closed abscess, an ‘‘abrasion’’ lesion score was designated if nodules of fibrous tissue were evident at the point of the shoulder (Ritter et al., 1999), and an ‘‘open’’ lesion score was used for open, draining sores/healing sores apparent at the point of the shoulder (Ritter et al., 1999). Teeth were evaluated by a trained technician. Top and bottom teeth were counted and scored for severity of wear. The following three definitions were used to categorize teeth wear: minimum—sharp points present on molars and incisors; moderate—points on molars and incisors worn but grooves between points still evident; severe—no points or grooves present on molars and incisors. Reproductive tracts were removed from the carcass by harvest plant personnel and immediately visually inspected by the research veterinarian. From the macroscopic appearance of the ovaries, sows were classified as normal (corpora lutea, corpora hemorrhagica, or mature follicles were present), cystic (multiple follicular cysts, >1.5 cm in diameter without corpora lutea or corpora hemorrhagica), or acyclic (no corpora lutea, corpora hemorrhagica, and small follicles covering less <50% of ovary). Ovaries representing each macroscopic appearance category were submitted to the Iowa State Veterinary Diagnostic Laboratory for histopathology characterization of status and confirmation of criteria. Pregnancy status was determined and fetal tissues (if present) were classified as normal, decomposed, or mummified. The thoracic and abdominal cavities and organs were visually evaluated for lesions by the research veterinarian. The presence or absence of peritonitis, pleural adhesions, and pneumonia was recorded. If pneumonia was visually diagnosed, an estimate of the percentage of total lung involvement was made. Lesions other than those previously mentioned were noted when visual evidence was present. At plant 1, sows were weighed and ultrasonically evaluated the night prior to harvest. The following morning sows were evaluated at five individual stations throughout the facility. Sows were rendered unconscious using electrical stunning methods which followed USDA approved plant protocol. Next, sows were suspended on a gambrel by their rear legs from where front and rear foot lesions were evaluated and ear tag identification numbers recorded. Then, BCS and presence/absence of shoulder lesions were evaluated. Additionally, ear tag identification number was recorded and on every fifth gambrel a colored piece of vinyl flagging tape (Gempler’s, Madison, WI), was attached. The colored tape was used to track and validate sow identities throughout the remaining evaluations at different locations throughout the harvest facility. Following carcass evisceration, the research veterinarian evaluated all thoracic and abdominal

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

201

organs. After masseter muscle removal, the number of teeth was counted and teeth wear was evaluated. At plant 2, sows were evaluated at 4 individual stations throughout the facility. Immediately prior to stunning, each sow was ultrasonically evaluated following the methodology previously outlined, and the BCS and shoulder lesions of each sow were evaluated. Sows were rendered unconscious using electrical stunning methods which followed USDA approved plant protocol. Next, sows suspended on a gambrel were evaluated for front and rear foot lesions and ear tag identification number recorded. Upon excision of the head from the carcass, plant personnel attached matching sequential identification numbers to both the head and carcass. Following carcass evisceration, the research veterinarian evaluated all thoracic and abdominal organs. After masseter muscle removal, teeth number and wear were evaluated. All data were recorded and reported at the sow level. In this study, attributing ‘‘left,’’ ‘‘right’’ or ‘‘bilateral’’ characteristics to lesions in paired organs did not provide additional information as to cause or source and was not reliably obtained given the carcass fabrication process. Additionally, for the present study we did not find value in determining side or bilateral characteristics. 2.2. Statistical analysis Statistical analyses were carried out using SAS (2003). Mixed model methodology (PROC MIXED of SAS) was used to evaluate and least square means were derived (LSMEANS option in SAS) for the dependent continuous traits in this study. The model for continuous traits included harvest facility as a fixed effect. The frequency of each binary trait event was calculated (PROC FREQ in SAS) and the Pearson chi-square tests were used to test the null hypothesis of no association between two binary traits. The phi coefficient is derived from the Pearson chi-square statistic and was utilized to measure the association between two binary traits. The phi coefficient has a range between negative one and positive one for 2  2 frequency tables and is similar to the correlation coefficient in its interpretation. Generalized linear mixed models (PROC GLIMMIX of SAS) were used to analyze response variable from a non-normal distribution and obtain probabilities of binary traits by BCS. Like linear mixed models, generalized mixed linear models assume normal (Gaussian) random effects and conditional on the normally distributed random effects, data can have any distribution in the exponential family. In the analysis of traits by BCS, BCS 4 and 5 were combined because only 28 sows (0.9%) evaluated were classified as a BCS of 5. The PROC MIXED procedure computing pair-wise differences was used to compare the number of teeth in each BCS category. This model included BCS as a fixed effect and day as a random effect. 3. Results 3.1. Feet evaluation The frequency and percentage of lesions measured on cull sows at two Midwestern harvest facilities are presented in Table 1. The most common foot lesions observed among harvested cull sows were rear (n = 2064, 67.5%) and front (n = 1024, 32.9%) heel lesions. The frequency of lesions measured on cull sows by BCS is presented in Table 2. The frequency of rear and front heel lesions from sows with a BCS of was lower ( p < 0.05) than BCS categories 2, 3, and 4. Associations between lesions on cull sows and their corresponding p-values are shown in Table 3.

202

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

Table 1 Frequency of lesions on 3158 cull sows evaluated at two U.S. Midwestern harvest facilities in 2005 Trait

Frequency, no.

Percentage (%)

1024 703 109 20 4

32.9 22.6 3.5 0.6 0.1

2064 644 552 152 134

67.5 21.1 18.1 5.0 4.4

Ovaries, n = 3062a Acyclic Cystic

277 192

9.0 6.3

Pregnancy, n = 3070a Pregnant Normal Mummified Decomposed

180 157 15 8

5.9 5.1 0.5 0.3

Systemic lesions, n = 3083a Pneumonia 1–10%b Pneumonia >10% c Pleural adhesion Peritonitis

153 145 174 54

5.0 4.7 5.6 1.7

Shoulder lesions, n = 3146a Abrasions Open Abscesses

394 150 12

12.5 4.8 0.4

Front feet, n = 3117 Heel lesions Cracked hooves Digital overgrowth Abscesses Missing dew claws

a

Rear feet, n = 3058a Heel lesions Digital overgrowth Cracked hooves Missing dew claws Abscesses

a b c

n = number of sows with recorded data. Pneumonia with 1–10% lung involvement. Pneumonia with >10% lung involvement.

Rear heel lesions were positively associated ( p < 0.01) with front heel lesions (0.21) and front hoof cracks (0.06), and negatively associated ( p < 0.01) with acyclic ovaries ( 0.06). Front heel lesions were negatively associated ( p < 0.01) with front hoof cracks ( 0.09). Cracked hooves were the next most frequently observed foot lesion. Cracked hooves were found on the front feet of 703 sows (22.6%) and rear feet of 552 sows (18.1%). The occurrence of both front and rear cracked hooves tended to increase as BCS decreased. Cracked hooves on the front feet were positively associated ( p < 0.05) with rear cracked hooves (0.16), rear foot abscesses (0.05), pleural adhesions (0.05), peritonitis (0.04), and open shoulder lesions (0.06). Cracked hooves on the rear feet were positively associated ( p < 0.05) with abscesses on the rear feet (0.08) and missing rear dew claws (0.04). Rear digital overgrowth was observed in 644 sows (21.1%). The occurrence of rear digital overgrowth BCS 4 sows (12.6%) was lower ( p < 0.05) than BCS categories 1, 2, and 3 (26.5, 22.2, and 21.3%, respectively). Rear digital overgrowth was positively associated ( p < 0.05)

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

203

Table 2 Probability of lesions by BCSa on 3158 cull sows evaluated at two U.S. Midwestern harvest facilities in 2005 Disorder

BCS 1

2

3

4

p-valueb

Front feet Heel lesions (%) Cracked hooves (%) Digital overgrowth (%) Abscesses (%) Missing dew claws (%)

26.3c 32.9e 6.3 e 2.3 d 0.7

35.4d 24.9d 5.3 e 0.3 c 0.1

32.6d 20.4c 2.7 d 0.6 c 0.1

33.1d 17.6c 0.5 c 0.3 c 0

.04 .01 .01 .01 .23

Rear feet Heel lesions (%) Digital overgrowth (%) Cracked hooves (%) Missing dew claws (%) Abscesses (%)

55.1c 26.5d 19.1d 3.7c,d 6.1 e

68.5d 22.2d 20.7d 6.5 d 7.8 e

68.6d 21.3d 17.7d 4.7c,d 2.9 d

70.8d 12.6c 12.6c 3.0 c 1.1 c

.01 .01 .01 .04 .01

Ovaries Normal (%) Acyclic (%) Cystic (%)

76.6c 20.6f 2.8 c

85.2d 9.9 e 4.8 c

85.1d 7.6 d 7.3 d

87.7d 4.2 c 8.1 d

.01 .01 .01

6.7 3.9 2.8 d 0

4.8 4 0.1 c 0.7

6.2 5.8 0.3 c 0.1

6.3 6 0.3 c 0

.45 .15 .01 .16

Systemiclesions Pneumonia 1–10%g Pneumonia >10% h Pleural adhesion (%) Peritonitis (%)

6.2d,e 13.2e 11.0e 8.2 e

6.9 e 6.4 d 7.4 d 2.2 d

4.2c,d 2.6 c 4.1 c 0.6 c

2.6 c 3.1 c 3.7 c 0.5 c

.01 .01 .01 .01

Shoulder lesions None (%) Abrasions (%) Open (%) Abscesses (%)

60.8c 21.6e 16.9e 0.7

69.8d 20.4e 8.9 d 0.9

90.4e 8.3 d 1.2 c 0.1

96.1f 3.9 c 0.0 c 0

.01 .01 .01 .11

Teeth wear Minimum (%) Moderate (%) Severe (%)

16.0d 53.5d 30.6c

10.3c 42.6c 47.1d

8.3 c 46.7c 45.2d

16.1d 52.8d 31.1c

.01 .01 .01

Teeth Top teeth, no. Bottom teeth, no.

21.3c,d 21.5c

21.4d 21.8d

21.3c,d 21.8d

21.2c 21.7c,d

Pregnancy Pregnant (%) Normal (%) Mummified (%) Decomposed (%)

a b c d e f g h

BCS = body condition score (possible range 1–5, Patience and Thacker, 1989). p-value = from PROC GLIMMIX testing the difference between traits by BCS. Row means with different subscripts differ ( p < 0.05). Row means with different subscripts differ ( p < 0.05). Row means with different subscripts differ ( p < 0.05). Row means with different subscripts differ ( p < 0.05). % of sows with pneumonia and 1–10% lung involvement. % of sows with pneumonia and >10% lung involvement.

204

Disorder

Peritonitis

Pneumonia Pleural adhesion

Cystic ovaries .02a (.35) b .00 (.87) Acyclic ovaries .12 (.0001) .07 (.0002) Peritonitis .06 (.0005) Pneumonia Pleural adhesion Fc cracked hooves F heel lesions F digital overgrowth Rd cracked hooves R heel lesions R abscesses R digital overgrowth R missing dew claws

.01 .08 .18 .15

(.60) (.0001) (.0001) (.0001)

F cracked F heel hooves lesions .02 .03 .04 .01 .05

(.24) (.13) (.03) (.55) (.004)

.02 .02 .02 .02 .01 .09

(.40) (.25) (.30) (.28) (.43) (.0001)

F digital R cracked overgrowth hooves .01 .04 .04 .02 .00 .03 .03

(.48) (.05) (.02) (.29) (1.0) (.12) (.06)

.03 .01 .01 .00 .01 .16 .02 .01

(.10) (.65) (.59) (.93) (.50) (.0001) (.32) (.58)

R heel lesions .03 .06 .03 .02 .01 .06 .21 .01

(.07) (.0008) (.09) (.37) (.59) (.0009) (.0001) (.76)

.02 (.37)

R abscesses .04 .06 .00 .05 .03 .05 .02 .00

(.05) (.001) (.82) (.003) (.15) (.005) (.37) (.87)

R digital overgrowth .01 .03 .00 .02 .02 .02 .01 .23

(.50) (.06) (.93) (.20) (.35) (.36) (.67) (.0001)

R missing dew claws .01 .01 .02 .03 .01 .01 .03 .04

(.57) (.50) (.39) (.09) (.64) (.43) (.16) (.03)

Shoulder abrasions .02 .02 .00 .06 .02 .00 .01 .01

(.19) (.34) (.88) (.002) (.27) (.94) (.47) (.66)

Open wounds .02 .03 .04 .08 .05 .06 .01 .01

(.30) (.07) (.02) (.0001) (.004) (.001) (.52) (.67)

.08 (.0001)

.00 (.80)

.04 (.02)

.01 (.47)

.00 (.81)

.01 (.51)

.01 (.52) .04 (.05)

.01 (.44) .16 (.0001) .04 (.01)

.01 (.55) .03 (.15) .05 (.003)

.03 (.10) .05 (.005) .03 (.05)

.01 (.45)

.02 (.22)

a Phi coefficient is derived from the Pearson chi-square statistic and was utilized to measure the association between two binary traits. The phi coefficient has a range between negative one and positive one for 2  2 frequency tables. b p-value of Pearson chi-square. c Front. d Rear.

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

Table 3 Phi coefficientsa and corresponding p-values of lesions on 3158 cull sows evaluated at two U.S. Midwestern harvest facilities in 2005

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

205

Table 4 Foot lesion scores on 3158 cull sows evaluated at two U.S. Midwestern harvest facilities in 2005 Score

Front feet (%)

Rear feet (%)

Front and rear feet (%)

0 1 2 3 4 5 6

48.5 43.8 7.3 0.4 0.0 0.0 NAb

19.5 51.3 23.8 4.8 0.6 0.1 NA

13.6a 29.5 33.3 17.0 5.0 1.4 0.2

a b

13.6% of sows had no foot lesions on their front or hind feet. NA = not applicable.

with front digital overgrowth (0.23), missing rear dew claws (0.04), shoulder abrasion lesions (0.05), and open shoulder lesions (0.03), and negatively associated ( p = 0.05) with rear foot abscesses ( 0.04). The frequency of total front, rear, and front/rear combined foot lesions on cull sows is shown in Table 4. More sows had one or more foot lesions on the rear feet (n = 2463, 80.5%) when compared to the front feet (n = 1605, 51.5%). 3.2. Reproductive tract evaluation Grossly normal ovaries were observed in 2592 (84.7%) of sows. The most common reproductive lesion observed among harvested cull sows was acyclic ovaries (n = 277, 9.0%). The occurrence of acyclic ovaries increased ( p < 0.05) as BCS decreased. Acyclic ovaries were positively associated ( p < 0.01) with rear foot abscesses (0.06), pneumonia (0.07), pleural adhesions (0.08), and peritonitis (0.12). Following acyclic ovaries, the next most frequent reproductive lesion among harvested cull sows was cystic ovaries (n = 192, 6.3%). The frequency of cystic ovaries tended to increase as BCS increased. Of the 192 females with cystic ovaries, all but two sows had large (>2.5 cm), multiple, and fluid filled cysts with thickened walls suggestive of luteinized tissue. Two sows had polycystic ovaries with multiple, small (<2.5 cm) thin walled cysts, without visual evidence of a thickened wall. None of the 192 sows had evidence of corpora lutea or corpora hemorrhagica on either ovary and in all cases, cystic ovaries were a bilateral problem affecting both ovaries. Pregnancy was detected in 180 sows (5.9%). Normal pregnancies were determined in 157/180 (87%) of pregnant sows. The occurrence of mummified fetuses from sows having a BCS of 1 (2.8%) was higher ( p < 0.05) when compared to sows from the BCS 2, 3, and 4 categories (0.1, 0.3, and 0.3%, respectively). 3.3. Systemic lesions Pneumonia was found in 298 sows (9.7%). Of the sows with pneumonia, slightly more sows had less than or equal to 10% lung involvement (n = 153, 51% of the total) when compared to sows with greater than 10% lung involvement (n = 145, 49% of the total). The frequency of pneumonia involving greater than 10% inform sows having a BCS of 1 (13.2%) was higher (0.05) than sows from BCS categories 2, 3, and 4 (6.4, 2.6, and 3.1%, respectively). The presence of pneumonia was positively associated ( p < 0.01) with pleural adhesions (0.15), peritonitis (0.06), rear foot abscesses (0.05), shoulder abrasion lesions (0.06), and open shoulder lesions (0.08).

206

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

Following signs of pneumonia, the next most common systemic lesion among harvested cull sows was pleural adhesions (n = 174, 5.6%). The occurrence f pleural adhesions tended to increase as BCS decreased. Pleural adhesions were positively associated ( p < 0.01) with peritonitis (0.18) and open shoulder lesions (0.05). Other than the previously described lesions, any additional gross abnormalities were noted in the present study when observed; however they were rare (less than 1%). 3.4. Teeth evaluation Teeth wear was categorized as minimal (10.5%), moderate (47.0%), or severe (42.5%). No category of teeth wear (minimum, moderate, severe) appeared to have a linear relationship with BCS. No associations were observed between the number of top and bottom teeth present and BCS. Significant differences in the number of teeth in the upper and lower jaw by BCS were observed. However, the differences (less than 1 tooth) were likely attributable to the large number of sows evaluated within each BCS category. Additionally, a biological meaning to these differences is unlikely. 3.5. Shoulder lesions The most observed shoulder lesion, abrasions (n = 394, 12.5%), were more common as BCS decreased. Open lesions were present on 150 (4.8%) of the sows evaluated. The incidence of open shoulder lesions increased as BCS decreased. Presence of open shoulder lesions were positively associated ( p < 0.01) with rear foot abscesses (0.05). 3.6. Body composition evaluation Descriptive statistics of body composition traits of the cull sows evaluated are presented in Table 5. Backfat and longissimus dorsi area ranged from 4.6 to 62.4 mm and 18.0 to 73.9 cm2, respectively. All five body condition scores were observed. Most carcasses were assigned a BCS of 3 (49.2%), 9.7% had a BCS of 1, 28.8% had a BCS of 2, and 12.3% had a BCS of 4 or 5. All correlation coefficients among BCS, backfat, longissimus dorsi area, and longissimus dorsi depth were different ( p < 0.01) from zero. Pearson correlation coefficients between BCS and backfat, longissimus dorsi area, and longissimus dorsi depth were 0.74, 0.51, and 0.47, respectively. Correlations between backfat and longissimus dorsi area and longissimus dorsi depth were 0.30 and 0.26, respectively. Among body composition traits evaluated, the correlation between longissimus dorsi area and longissimus dorsi depth (0.92) was greatest in magnitude.

Table 5 Body composition trait means, standard deviations (S.D.), and ranges of 3158 cull sows evaluated at two U.S. Midwestern harvest facilities in 2005 Trait

Number of sows

Mean

S.D.

Minimum

Maximum

Backfat (mm) Longissimus dorsi area (cm2) Longissimus dorsi depth (cm) BCSa

2083 2074 2074 3149

24.9 47.8 5.5 2.7

9.6 7.7 0.8 0.8

4.6 18.0 2.4 1

62.4 73.9 8.1 5

a

BCS = body condition score (possible range 1–5, Patience and Thacker, 1989).

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

207

4. Discussion 4.1. Internal and external validity In general terms, internal validity represents the degree in which the results from a study, in this case evaluation of cull sows at harvest, are correct for the sample studied, in this case the sample of cull sows (Fletcher and Fletcher, 2005). To address this issue the project coordinators took numerous precautions. The harvest facilities were selected because their configuration allowed sufficient time and physical area for each person conducting a particular evaluation on each carcass to make an accurate assessment of their assigned trait. Additionally, the harvest facilities were chosen because the type of sow harvested (from very thin to very heavy as represented by the different USDA weight classes) was different. Based on the researchers knowledge, the broad spectrum of sows existing in modern pork operations were well represented when the information from the two harvest facilities were combined. Similarly, the study was designed to remove as much bias that could occur in three ways. First, we evaluated all sows at each data collection day that were presented. This removes any chance of bias due to time of harvest at each plant. Secondly, data collection occurred across multiple days at each harvest facility in order to reduce the chance of bias due on a given day due to outside factors. Lastly, over 3000 sow carcasses were evaluated in order to minimize the effect of any single cull sow or small group of cull sows would have on the data. The people examining the carcasses and collecting the data were trained to accurately evaluate the particular trait they were assigned. Furthermore, the researchers tagged every fifth gambrel using a colored piece of vinyl flagging tape (Gempler’s, Madison, WI). The presence or absence of this tape was noted in the data collection for each individual sow as a verification of numerical order in the slaughter process and to ensure that sows were not dropped from the line. This allowed verification and maintenance of the sequence of carcasses presented for evaluation at different locations within the plant during carcass processing. Throughout the project, the evaluation of each trait was done by a maximum of two people. In situations where two people evaluated the same trait, data collection visits were overlapped with both people on a visit simultaneously to allow calibration and agreement of subjective critieria. Minimizing the number of persons evaluating any one trait minimizes the chance for differences to occur because of evaluator. Additionally, means, ranges and variances where appropriate we compared to ensure the evaluation similarity between evaluators in the case where two people evaluated a particular trait. External validity refers to ‘‘generalizability’’ of our results or how well the results of the present study hold to other settings (Fletcher and Fletcher, 2005). Because the harvest facilities that cooperated purchased sows from across the U.S. and Canada, the results of the present study can be useful to operations and clinicians in these areas. Readers from other areas of the world will have to decide if the sample population of the present study is representative of their situation and apply the results accordingly. 4.2. Feet evaluation Rear heel lesions were observed on 67.5% of cull sows. These results are consistent with a Norwegian study (Gjein and Larssen, 1995a) that evaluated sows from 36 herds and reported rear heel lesions on cull sows housed in groups and stalls of 77% and 53%, respectively. This study reported that a large proportion of the group housing systems had partially slatted concrete floors

208

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

without bedding and stall systems were generally concrete floors without slats. In the present study the frequency of rear and front heel lesions was lowest in sows from the BCS 1 category. One possible explanation for these results is that sow weight generally increases as BCS increases, which may result in increased pressure and damage to the rear heel. Rear heel lesions had a low negative association ( 0.06) with acyclic ovaries. This association is likely confounded by BCS, as rear heel lesions were associated with higher BCS and the presence of acyclic ovaries was higher in sows with a lower BCS. Rear and front heel lesions were positively associated in the present study. This would imply factors affecting heel lesions simultaneously affect rear and front heel lesions. Rear cracked hooves were observed on 18.1% of cull sows in the present study. In contrast, Gjein and Larssen (1995a) reported that sows housed in loose and confined systems both had relatively high prevalence of side wall cracks (80% and 51%, respectively) and cracks in the white line (63% and 46%, respectively) of the feet. In the current study, increased rear and front cracked hooves were associated with decreasing BCS. This could suggest dietary deficiency as an impact on hoof integrity if decreased body condition is a result of reduced feed intake or feed restriction. Studies have shown corn-based diets to be a good source of biotin, but deficiencies sufficient to result in problems with cracked hooves have been demonstrated in studies of diets based on other cereal grains (Brooks et al., 1977; Simmins and Brooks, 1988). These diets, without sufficient supplemental biotin, could make the hoof prone to trauma. This likelihood of trauma could reasonably be expected to be higher on hard surface flooring (Bane et al., 1980). In the current study, flooring and diet exposure of the sows were not known for the entire population. Rear and front cracked hooves were positively associated with each other in the present study. This would imply common factors affecting rear and front cracked hooves simultaneously. Concrete floors, both solid and partially slatted, have been shown to contribute to the presence of cracked hooves (Kornegay et al., 1990). On concrete, sows housed in stalls generally have fewer cracked hooves when compared to sows housed in group housing systems (Gjein and Larssen, 1995a). However, in a comparison of sows housed in concrete stalls or in deep litter bedding, the latter has been shown to produce sows having fewer cracked hooves (Gjein and Larssen, 1995b). The prevalence of rear digital overgrowth (21.1%) across all sows evaluated was relatively high. It is not clear why the prevalence of digital overgrowth was at the prevalence observed and whether it is attributable to genetic, nutritional, environmental, or some other factor(s). A previous report (MAFF, 1981) has reported that pigs housed on plastic slats had a higher prevalence of overgrown claws when compared to pigs housed on concrete slats. The same study suggested that slippery surfaces may force sows to exert more pressure on the back of the foot and reduce claw wear in the front of the foot. An earlier report has indicated a nutritional association with digital overgrowth in sows. Jorgensen and Sorensen (1998) reported sows reared on higher feeding levels as gilts experienced longer dew claws. Rear and front digital overgrowth were positively associated in the present study. This implies factors that affect digital overgrowth simultaneously affect rear and front digital overgrowth. The present study found that 19.5% of cull sows had no rear foot lesions. Similarly, Gjein and Larssen (1995a) reported 20% of cull sows in stalled housing and 4% of cull sows in group housing had no rear foot lesions. In the present study, foot lesions were more prevalent in rear feet (80.5%) when compared to front feet (51.5%). These results are in agreement with Gjein and Larssen (1995a) who reported that lateral rear claws were the most frequent location for lesions. One possible explanation for more rear foot lesions compared to front foot lesions may be that rear claws are known to be less uniform than the front claws (Penny et al., 1963). Foot lesions are

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

209

known to increase as the difference in size between the medial and lateral claws becomes larger (Kornegay et al., 1990), likely from unequal weight distribution. Wet conditions may also increase the prevalence of rear foot lesions relative to front foot lesions. The rear feet of sows housed in stalls may experience wetter conditions when compared to the front feet because of sow urination (Gjein and Larssen, 1995a). Wet, rough, or slatted floors have all been shown to increase foot lesions (Penny et al., 1965). 4.3. Reproductive tract evaluation Visually normal ovaries were found in 84.7% of sows evaluated in the present study. This percentage is somewhat greater than values reported in previous studies. Dalin et al. (1997) reported normal ovaries lesions (69%) in 115 (34 gilts and 81 sows). However, this may be explained as the ovaries represent only portion of the entire reproductive system. Areas not examined in this study may impact reproductive performance in part or in whole. Similarly, Heinonen et al. (1998) reported 52.3% of 1708 gilts and sows harvested in southern Finland showed normal ovaries. Visually acyclic ovaries were observed in 9.0% of sows evaluated in the present study. Dalin et al. (1997) and Heinonen et al. (1998) found higher a prevalence of acyclic ovaries of 17.0% and 25.1%, respectively. One possible explanation for the difference in prevalence of ovaries that appear acyclic between the current study and previous studies may be in culling strategy. The prevalence of ovaries that appear acyclic among sows culled within 1 day of weaning has been reported to be relatively high (97%) (Einarsson et al., 1982). Frequency of acyclic ovaries was higher in the BCS 1 category (20.6%) when compared to the BCS 4 category (4.2%). Clowes et al. (2003) reported body protein mass loss greater than 9 to 12% rapidly reduced ovarian function. Because protein loss was unknown in the present study, it is not clear if it contributed to acyclic ovaries in a similar manner to that observed in previous studies. However, it could reasonably be assumed that sows in the BCS 1 category likely had at least some body protein loss and hence, this may have contributed to the greater prevalence of acyclic ovaries in the lower BCS sows. Cystic ovaries were observed in 6.3% of sows in the present study. These results are similar to Heinonen et al. (1998) who found 6.2% grossly cystic ovaries. Dalin et al. (1997) found a slightly higher prevalence of cystic ovaries (14%). Possible explanations for the difference found in the Dalin et al. (1997) study are population size and number of herds examined. In that study, 115 cull gilts and sows from a single Swedish sow pool were used compared to 3158 sows from multiple sources used in the current study. In an on-farm study, Castagna et al. (2004), using real-time transcutaneous ultrasound examination in 1990 live Camborough 22 (Pig Improvement Company, Franklin, KY) gilts and sows found 2.4% cystic ovaries. Castagna et al. (2004) reported the rate of return to estrus was greater in sows with cystic ovaries compared to sows with normal ovaries. This study indicates sows with cystic ovaries may be more likely to be culled for reproductive failure. Cystic ovaries can readily be identified using real-time transcutaneous ultrasound due to their immense size and shape (Knox and Althouse, 1999). Real-time transcutaneous ultrasound technology may help pork producers reduce herd non-productive days by identifying and subsequently culling females with ovarian cysts. The present study found a higher prevalence of cystic ovaries was associated with a higher BCS, and these results could be influenced by weaning age or failure to conceive. Sows weaned earlier may lose less body condition from lactation demands when compared to later weaned

210

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

sows and have been shown to have more cystic ovaries. Weaning at less than 14 days of age (Castagna et al., 2004) and at 2 days (Svajgr et al., 1974) has been shown to increase the prevalence of cystic ovaries when compared to older weaning ages. Sows with cystic ovaries have reduced fertility (Castagna et al., 2004) which may lead to the accumulation of body stores prior to culling. 4.4. Systemic lesions Of the systemic lesions recorded, pneumonia was the most commonly observed lesion (9.7%). Peritonitis, pneumonia, and pleural adhesions were all positively associated with each other. It appears common for multiple lesions to be found in coexistence suggesting systemic impacts on health. 4.5. Teeth evaluation Severe teeth wear was found in 42.5% of sows evaluated in the present study. Johnson et al. (2003) reported severe incisor wear and loss of 16% and 31%, respectively, in 32 immature sows and 62% and 34%, respectively, in 82 mature sows. The same study reported 9% of immature sows and 63% of mature sows had severe molar wear. Severe incisor wear and loss and molar wear were defined as a score of 3 or 4 on a 0–4 scale with 4 being the most severe. In the present study, teeth wear and teeth number did not appear to have a linear association with BCS. This may suggest the presence or absence of teeth and teeth wear have minimal effect on culling for poor body condition. However, the age of the sows was unknown in the current study and may influence the results of teeth number related to BCS as pigs full permanent dentition is acquired when they are at least 1.5 years of age (Pond and Mersmann, 2001). 4.6. Shoulder lesions In the present study, 12.5% of sows had shoulder abrasions. A previous study (Ritter et al., 1999) reported a lower prevalence of shoulder abrasions (3.8%) in 1751 carcasses of culled gilts and sows harvested during a 1-week period in July at a Midwest harvest facility. Differences in sow body condition likely contribute to the differences observed between studies. The present study found 9.7% of evaluated sows in the BCS 1 classification whereas Ritter et al. (1999) found 0.8% of examined sows in the BCS 1 class. In agreement with the current study, Ritter et al. (1999) reported sows with poorer body condition had more shoulder lesions. 4.7. Body composition evaluation Positive correlations between BCS and backfat/longissimus dorsi indicate the experienced livestock evaluators and the ultrasonic evaluation similarly evaluated body composition. The relatively high correlation (0.92) between longissimus dorsi area and longissimus dorsi depth suggests either longissimus dorsi area or longissimus dorsi depth measured by experienced realtime ultrasonic scan technicians may be used when assessing body condition. Measuring longissimus dorsi depth compared to longissimus dorsi area when assessing body condition currently takes less time and can be done using less expensive ultrasonic equipment.

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

211

5. Conclusion From the observations of this study, BCS was associated with several abnormal conditions of sows. Multiple lesions were associated with BCS. Whether the lesions caused BCS to change, BCS caused the lesions, or the lesions and BCS changed simultaneously is unknown. Most of these abnormal conditions are difficult to reliably observe by production personnel in the farm setting. Observation and attention to BCS may serve as an indicator of other lesions that have the potential to lessen the productivity of the sow. Causative relationships were not established by this study. However, the strong associations observed suggest that sows that do not respond to increased feeding with improved BCS on farm could reasonably be expected to have additional lesions that may have poor prognosis for high performance. The prevalence of reproductive lesions detected in the current study was substantially lower than the reported percentage of sows culled for reproductive failure in previous farm based studies. As previously noted, this may be explained as the ovaries represent only portion of the entire reproductive system and process and factors not evaluated in this study may impact reproductive performance in part or in whole. Additional characterizations may be able to relate on-farm management practices to one or more lesions that had a high occurrence in the present study. References Bane, D.P., Meade, R.J., Hilley, H.D., Leman, A.D., 1980. Influence of d-biotin and housing on hoof lesions. Proc. Int. Pig Vet. Soc. 334. Brooks, P.H., Smith, D.A., Irwin, V.C.R., 1977. Biotin-supplementation of diets; the incidence of foot lesions, and the reproductive performance of sows. Vet. Rec. 101, 46–50. Castagna, C.D., Peixoto, C.H., Bortolozzo, F.P., Wentz, I., Neto, G.B., Ruschel, F., 2004. Ovarian cysts and their consequences on the reproductive performance of swine herds. Anim. Reprod. Sci. 81, 115–123. Clowes, E.J., Aherne, F.X., Foxcroft, G.R., Baracos, V.E., 2003. Selective protein loss in lactating sows is associated with reduced litter growth and ovarian function. J. Anim. Sci. 81, 753–764. Dagorn, J., Aumaitre, A., 1979. Sow culling: reasons for and effect on productivity. Livest. Prod. Sci. 6, 167–177. Dalin, A.-M., Gidlund, K., Eliasson-Selling, L., 1997. Post-mortem examination of genital organs from sows with reproductive disturbances in a sow-pool. Acta Vet. Scand. 38, 253–262. Einarsson, S., Lundeheim, N., Martinsson, K., Persson, N., Persson, I., 1982. Post mortem examination of the genital organs of culling sows from a large herd with relation to fertility data. Proc. Int. Pig Vet. Soc., Mexico 211. Fletcher, R.W., Fletcher, S.W., 2005. Clinical Epidemiology, The Essentials, fourth ed. Lippincott Williams and Wilkins, Baltimore, MD 21201. Gjein, H., Larssen, R.B., 1995a. Housing of pregnant sows in loose and confined systems—a field study. 2. Claw lesions: morphology, prevalence, location, and relation to age. Acta Vet. Scand. 36, 433–442. Gjein, H., Larssen, R.B., 1995b. Housing of pregnant sows in loose and confined systems—a field study. 3. The impact of housing factors on claw lesions. Acta Vet. Scand. 36, 443–450. Heinonen, M., Leppa¨vuori, A., Pyo¨ra¨la¨, S., 1998. Evaluation of reproductive failure of female pigs based on slaughterhouse material and herd record survey. Livest. Prod. Sci. 52, 235–244. Johnson, E.W., Curtis, S.E., Ellis, M., 2003. Dental disease in sows: early findings. In: Proceedings of A. D. Leman Swine Conference. pp. 30–32. Jorgensen, B., Sorensen, M.T., 1998. Different rearing intensities of gilts: II Effects on subsequent leg weakness and longevity. Livest. Prod. Sci. 54, 167–171. Knox, R.V., Althouse, G.C., 1999. Visualizing the reproductive tract of the female pig using real-time ultrasonography. Swine Health Prod. 7, 207–215. Kornegay, E.T., Bryant, K.L., Notter, D.R., 1990. Claw lesion development in gilts and sows housed in confinement as influenced by claw size and claw location. Appl. Agr. Res. 5, 327–334. Maff., 1981. Injuries caused by flooring: a survey in pig health scheme herds. Proc. Pig Vet. Soc. 8, 119–125.

212

M. Knauer et al. / Preventive Veterinary Medicine 82 (2007) 198–212

Patience, J.F., Thacker, P.A., 1989. Swine Nutrition Guide. Prairie Swine Centre, University of Saskatchewan, Saskatoon, pp. 149–171. Penny, R.H.C., Osborne, A.D., Wright, A.I., 1963. The causes and incidence of lameness in store and adult pigs. Vet. Rec. 75, 1225–1240. Penny, R.H.C., Osborne, A.D., Wright, A.I., Stephens, T.K., 1965. Foot-rot in pigs; observations on the clinical disease. Vet. Rec. 77, 1101–1108. PigCHAMP, 2002. Breeding Herd Summary U.S.A., available: http://www.pigchamp.com/2002Datashare.htm. accessed April 13, 2006. PigCHAMP, 2003. Breeding Herd Summary U.S.A., available: http://www.pigchamp.com/2003AnnualDatashare.htm., April 13, 2006. PigCHAMP, 2004. Breeding Herd Summary U.S.A., available: http://www.pigchamp.com/2004YearendDatashare.htm., accessed April 13, 2006. Pond, W.G., Mersmann, H.J., 2001. Biology of the Domestic Pig. Cornell University Press, Ithaca, New York, USA, p. 391. Ritter, L.A., Xue, J.L., Dial, G.D., Morrison, R.B., Marsh, W.E., 1999. Prevalence of lesions and body condition scores among female swine at slaughter. J. Am. Vet. Med. Assoc. 214, 525–528. SAS, 2003. User’s Guide: Statistics. SAS Institute Inc., Cary, NC. Simmins, P.H., Brooks, P.H., 1988. Supplementary biotin for sows: effect on claw integrity. Vet. Rec. 122, 431–435. Stalder, K.J., Knauer, M., Baas, T.J., Rothschild, M.F., Mabry, J.W., 2004. Sow longevity. Pig News Info. 25, 53N–74N. Svajgr, A.J., Hays, V.W., Cromwell, G.L., Dutt, R.H., 1974. Effect of lactation duration on reproductive performance of sows. J. Anim. Sci. 38, 100–105.