Doi:10.1016/j.livsci.2007.04.007

Number of services and the reservice intervals in relation to suboptimal reproductive performance in female pigs School of Agriculture, Meiji University, Tama-ku, Kawasaki 214-8571, Japan Received 21 February 2007; received in revised form 6 April 2007; accepted 6 April 2007 This study investigated the associations of the number of services, reservice intervals (RI), parity, and weaning-to-first-mating intervals (WMI) with farrowing rate and pigs born alive (PBA) on commercial farms. The selected 117 farms included 115,442service records and 93,867 farrowing records in 54,502 female pigs (females). A service was defined as one or more mating eventswithin a 10-day time period of estrus. The number of services was categorized into three groups: non-return to service, firstreservice, and second or later reservice. The RI was divided into 8 seven-day interval groups from 11 to 66 days and a group of 67–150 days. The effects of the number of services, RI, WMI, and parity on farrowing rate and PBA were analyzed by using themixed-effects models. The frequency distributions of non-return to service, first reservice, and second or later reservice groupswere 88.6, 9.7, and 1.7%, respectively. Farrowing rate decreased by approximately 18% with each service (P b 0.05). Reservicedfemales exhibited a different reproductive performance as compared to non-return to service females depending on parity andWMI. In the non-return to service group, sows with WMI 7–12 days had the lowest farrowing rate (P b 0.05). Meanwhile, in thefirst and second or later reservice groups, no differences between the WMI groups were found in farrowing rate. At parity 1 and 2,the first reservice group had the most PBA (P b 0.05). However, at parity 0, 3–5, and ≥6, no differences between the number ofservice groups were found in PBA. In the WMI 0–6 and 7–12 day groups, the first reservice group had the most PBA (P b 0.05).
The mean of RI was 44.5 ± 0.28 days in 13,183 reservice records. RI decreased from 47.4 to 39.2 days as parity increased from 0 to≥6 (Pb0.05). The frequency distributions of the RI 18–24, 39–45, and 67–150 days were 39.3, 12.4, and 18.8%, respectively.
Gilts had lower proportions of the RI 18–24 days and higher proportions of RI 67–150 days than sows at parity 3–5 (P b 0.05). Nodifferences between RI groups were found in farrowing rate and PBA. Increasing farrowing rate in non-return to service femalesand minimizing RI in females at low parity improved herd productivity.
2007 Elsevier B.V. All rights reserved.
Keywords: Fertility; Mating; Non-productive female days; Repeat breeders and non-productive female days, and decreased herdproductivity on commercial farms ).
A failure to detect a female pig (female) returning to The RI accounted for non-productive female days by estrus after service accumulated reservice intervals (RI) approximately 30% Furthermore,reserviced females had a lower farrowing rate than ⁎ Corresponding author. Tel.: +81 44 934 7826; fax: +81 44 934 7902.
non-return to service females (non-returned females), but had more pigs born alive (PBA) than non-returned 1871-1413/$ - see front matter 2007 Elsevier B.V. All rights reserved.
doi: Please cite this article as: Takai, Y., Koketsu, Y. Number of services and the reservice intervals in relation to suboptimal reproductiveperformance in female pigs on commercial farms. Livest. Sci. (2007), doi: Y. Takai, Y. Koketsu / Livestock Science xx (2007) xxx–xxx Weaned sows were housed in pens or stalls in a breeding No research has been undertaken to compare farrowing area. Both natural mating and AI were generally prac- rate and PBA between non-returned, first reserviced, ticed. The recommendation was to service females after and second or later reserviced females.
Parity, lactation length, and weaning-to-first-mating intervals (WMI) were recognized as factors affecting farrowing rate and PBA ). However, few studies investigated interactions Of the data received from 119 breeding farms, two between the associations of the number of services and farms had just started the operations, and these two these three factors with farrowing rate and PBA.
farms were excluded from this study. The database of Returning to estrus after service within 18 to 24 days, 117 farms contained 116,018 service records and 94,200 and 42 ± a few days has been defined as a regular RI, and farrowing records in 54,722 females. Missing records of returning in the outside of a regular RI has been defined farrowing events, records of sows with WMI longer than 120 days, and females with RI longer than 150 days in ). The previous researchers have considered the 576 service records and 333 farrowing records of 220 RI of around 42 days as “regular”, because most of the females were considered as extreme, and were excluded.
first returns postmating may be weak and could not be Hence, 115,442 service records and 93,867 farrowing detected. Regular returning to estrus suggested an records in 54,502 females were used in the further study.
occurrence of conception or implantation failure, whileirregular returning to estrus implied successful concep- tion but subsequent early embryonic loss (No studies reported an association Weaning records without previous farrowing records between RI and reproductive performance.
could not be input in the PigCHAMP®. Inputting a service Our objectives were to characterize a reservice oc- record with RI N 115 days in the software was frequently currence and RI on commercial farms; and to investigate elicited a warning of a missing record of farrowing and associations of the number of services, RI, parity, weaning. Therefore, records of RI ≥ 115 days rarely in- lactation length, and WMI with farrowing rate and PBA.
cluded a missing event of farrowing.
Females included gilts and sows. A gilt was defined as a female entered into a herd but not farrowed, and a Data was extracted from an existing database (Meiji sow was a female farrowed at least once. A mating was University, Kawasaki, Japan). The database was con- defined as any single insemination of a female during structed in the following manner. Approximately 140 estrus. A service included one or more matings in a 10- commercial farms using a farm management program PigCHAMP® in Japan were requested from December 2002 to mail their data files to the university, when they days from the last mating during the service to the first purchased the software or renewed their yearly mating during the subsequent service.
maintenance contract. By August 31, 2003, the datafiles were received from 119 breeding farms. Measure- 2.6. Categorization of production factors ments of serviced females for 2002 were extracted fromeach data file and used in this study.
The number of services was categorized into three groups: non-return to service, first reservice, and second or later reservice. The RI was divided into 8 seven-dayintervals and a group of 67–150 days: 11–17, 18–24 Females in this study were mainly F1-crossbreds (regular), 25–31, 32–38, 39–45 (regular), 46–52, 53–59, between Large White and Landrace, produced within 60–66, and 67–150 days. The nine RI groups were defined farm or purchased from breeding companies. Breeding by modifying the grouping criteria presented in previous stocks on these farms were originally imported from the Europe or USA. Lactation and gestation diets were Five parity groups were built: 0, 1, 2, 3–5, and ≥ 6.
formulated by using imported corn and soybean meal.
Four groups of lactation length were formed: 0–13, Please cite this article as: Takai, Y., Koketsu, Y. Number of services and the reservice intervals in relation to suboptimal reproductiveperformance in female pigs on commercial farms. Livest. Sci. (2007), Y. Takai, Y. Koketsu / Livestock Science xx (2007) xxx–xxx Farrowing rate by the number of services and weaning-to-first-mating Reservice intervals by parity and the number of services1 67,033 85.7 ± 0.14ax 6305 65.3 ± 0.60y 1078 50.1 ± 1.52z 113,183 reservice records were used.
a, b, cValues (within a column) followed by different superscript letters Values (within a column) followed by different superscript letters Values are presented as means ± S.E.M.
x, y, zValues (within a row) followed by different superscript lettersdiffer (P b 0.05).
contrasts was applied for binary data (farrowing rate) in the Values are presented as proportions ± estimated S.E.
GLIMMIX procedure. Linear mixed-effects models withTukey–Kramer multiple comparisons were applied for 14–20, 21–27, and ≥ 28 days. Three WMI groups continuous data (PBA and RI) in the MIXED procedure.
were constructed: 0–6, 7–12, and ≥ 13 days ( A number of statistical models were built to investigate the relationships between reproductive formed: January to March, April to June, July to performance and various factors. Model 1 was built to determine associations of the number of services withfarrowing rate and PBA. The independent variables for all females were the number of services and paritygroups. The independent variables for sows were the The observational unit was the service record or the number of services, parity, lactation length, and WMI farrowing record. All statistical analyses were done with groups. Model 2 was constructed to analyze relation- ships of RI with farrowing rate and PBA. The error of farrowing rates and the proportions of each RI independent variables were RI, the number of services, group was obtained by a standard method ( and parity groups. In Model 3, the factor associated with ). Mixed-effects logistic regression analysis with RI was investigated. The independent variables were the Table 2Pigs born alive following service by parity and weaning-to-first-mating interval groups at each service group a, b, c, dValues (within a column) followed by different superscript letters differ (P b0.05).
x, y, zValues (within a row) followed by different superscript letters differ (P b0.05).
Values are presented as means ± S.E.M.
Please cite this article as: Takai, Y., Koketsu, Y. Number of services and the reservice intervals in relation to suboptimal reproductiveperformance in female pigs on commercial farms. Livest. Sci. (2007), doi: Y. Takai, Y. Koketsu / Livestock Science xx (2007) xxx–xxx Farrowing rate and pigs born alive following service by nine reserviceinterval groups a The frequency distributions of serviced females by the non-return to service, first reservice, and second or later reservice were 88.6, 9.7, and 1.7%, respectively. The means (±estimated S.E. or S.E.M.) of farrowing rate and PBA in 115,442 service records and 93,867 farrowing records were 81.3 ± 0.11% and 10.1 ± 0.01 pigs, respec- tively. Of 21,575 not-pregnant services in the 115,422 service records, 8392 females were culled for various reasons (e.g. anestrus, reproductive failure, or lameness).
The number of preceding services, parity, lactation length, and WMI groups were associated with farrowing rate and PBA (P b 0.05). Farrowing rate decreased byapproximately 18% with each service (P b 0.05). Sows Values of farrowing rate (%) are presented as proportions ± estimated S.E.
Values of pigs born alive following service are presented as means ± S.E.M.
with lactation length of 0–13 days had ≥15% lower a 13,183 reservice records were used.
farrowing rates and ≥0.4 fewer PBA than those withlactation length ≥ 14 days (P b 0.01). Additionally, the number of services and parity. All possible two-way two-way interaction of the number of services × WMI interactions between independent variables were in- with farrowing rate was found (P b 0.05; ). In the cluded in all the models, but insignificant interactions non-return to service group, sows with WMI 7–12 days (P N 0.05) were removed from the final models.
had the lowest farrowing rate, while sows with WMI 0– All statistical models included farm, mating season, 6 days had the highest farrowing rate (P b 0.05). In the and an interaction of farm × mating season as a random first and second or later reservice groups, no differences effect. The interaction was used to account for a part of between the WMI groups were found in farrowing rate.
individual service within a sow, because one or more Second or later reserviced females had fewer PBA services were done in each female during the study year.
than non-returned and first reserviced females (P b 0.01; The chi-square test was used to compare frequency ). In addition, the 2 two-way interactions of the distributions of the RI among parity groups. The number of services × parity and the number of servi- proportions of service records with each RI group ces × WMI with PBA were also found (P b 0.05). At were compared with the two-sample test for binomial parity 1 and 2, the first reservice group had the most PBA (P b 0.05). In contrast, at parity 0, 3–5, and ≥6, no Table 5Frequency distributions (%) of reservice interval groups by five parity groups a Values are presented as proportions ± estimated S.E.
a 13,183 reservice records were used.
b The parity 3–5 group was used as reference.
⁎ Proportions (within a row) differed from the parity 3–5 group (Pb0.05).
Please cite this article as: Takai, Y., Koketsu, Y. Number of services and the reservice intervals in relation to suboptimal reproductiveperformance in female pigs on commercial farms. Livest. Sci. (2007), Y. Takai, Y. Koketsu / Livestock Science xx (2007) xxx–xxx difference in PBA was found between the number of necessary for producers to consider WMI when they In the WMI 0–6 and 7–12 day groups, the first The result showing 0.3 to 0.5 more PBA in the first reservice group had the most PBA (P b 0.05; In reserviced sows than the non-returned sows at parity 1 the non-return to service group, sows with WMI 7–12 days and 2 might be due to prolonged weaning-to-conception had the fewest PBA (P b 0.05). In the first and second or intervals in reserviced females. A service at second later reservice groups, no differences in PBA were found estrus postweaning (delayed mating) increased report- The mean of RI in 13,183 reservice records was 44.5 ± 0.28 days. Gilts had RI similar to sows at parity 1 and 2, higher concentrations of plasma progesterone 10 to 20 h and had longer RI than those at parity 3–5 and ≥6 after ovulation were found in first estrus postweaning (P b 0.05; The first reservice group had 2.2 days than those in second estrus in low parity sows ( longer RI than the second or later reservice group (P b 0.05). No significant interaction of the number of increased RI by approximately 45 days at parity 1 and 2.
services × parity with RI was found.
Increased 0.5 pigs in the first reserviced females did not The frequency distributions of regular RI 18–24 days, make up for 45 days of RI which were equivalent to 2.25 regular RI 39–45 days, and irregular RI groups were 39.3, 12.4, and 48.3%, respectively (). No No increase of PBA in reserviced females at parity differences between RI groups were found in farrowing ≥3 can be explained by the previous report showing no rate and PBA (P = 0.20 and P = 0.43, respectively).
difference in embryo survival and progesterone con- Gilts and sows at parity 1 had lower proportions of centrations between the first and the second estrus at regular RI 18–24 day group and higher proportions of regular RI 39–45 day group than sows at parity 3–5 terone concentrations were also associated with higher (P b 0.05; ). In addition, gilts had higher proportions of irregular RI 53–59 day, 60–66 day, and This study showed that regular or irregular RI was 67–150 day groups than sows at parity 3–5 (P b 0.05).
not related to farrowing rate and PBA in reservicedfemales. Therefore, detected females returning to estrus should be immediately serviced to decrease non-pro-ductive female days.
This study estimated that commercial farms had A higher proportion of irregular RI 53–150 days approximately 10% first reserviced females, and had and longer RI in low parity females than high parity approximately 2% second or later reserviced females.
sows may be, at least in part, attributed to shorter estrus These findings were consistent with the studies showing duration in reserviced gilts than in reserviced sows approximately 12% reserviced females found in 55 poor management in pregnancy checking, and reser- viced females at low parity accumulated non-produc- Farrowing rate decreased by approximately 18% tive female days more than those at high parity. with each service. This finding might be partially ex- plained by differences in preovulatory serum LH diagnosed accurately in most gestating females by profiles and ovulation timing between reserviced day 24 after mating using real-time ultrasound. Some hormonal treatments or strict culling policy should be ). In the heifers, repeat breeders had also lower considered for reserviced females at low parity to magnitude of the preovulatory LH peaks than non- a farrowing rate in reserviced females, a matured boarshould be used with AI, because the presence of a boar increased high pulse frequency of LH, and promoteduterine activity Reservice decreased farrowing rate in both gilts and sows, but no influence on PBA was found. Increasing The effects of WMI 7–12 days on suboptimal re- farrowing rate in non-returned females and minimizing productive performances disappeared when sows were RI in reserviced females at low parity improved herd reserviced in this study. Therefore, in practice, it is not Please cite this article as: Takai, Y., Koketsu, Y. Number of services and the reservice intervals in relation to suboptimal reproductiveperformance in female pigs on commercial farms. Livest. Sci. (2007), doi: Y. Takai, Y. Koketsu / Livestock Science xx (2007) xxx–xxx Koketsu, Y., Dial, G.D., King, V.K., 1997a. Influence of various factors on farrowing rate on farms using early weaning. J. Anim.
Sci. 75, 2580–2587.
The authors gratefully thank cooperative swine Koketsu, Y., Dial, G.D., King, V.K., 1997b. Returns to service after producers for providing their valuable data, and the mating and removal of sows for reproductive reasons from PigCHAMP staff (Setagun, Gunma, Japan) for their commercial swine farms. Theriogenology 47, 1347–1363.
technical assistance. This research was supported by Koketsu, Y., 2005. Six component intervals of nonproductive days by breeding-female pigs on commercial farms. J. Anim. Sci. 83,1406–1412.
Langendijk, P., Bouwman, E.G., Schams, D., Soede, N.M., Kemp, B., 2003. Effects of different sexual stimuli on oxytocin release,uterine activity and receptive behavior in estrous sows. Ther- Clark, L.K., Leman, A.D., Morris, R., 1988. Factors influencing litter size in swine: parity-one females. J. Am. Vet. Med. Assoc. 192, Le Cozler, Y., Dagorn, J., Dourmad, J.Y., Johansen, S., Aumaître, A., 1997. Effect of weaning-to-conception interval and lactation length Clowes, E.J., Aherne, F.X., Foxcroft, G.R., 1994. Effect of delayed on subsequent litter size in sows. Livest. Prod. Sci. 51, 1–11.
breeding on the endocrinology and fecundity of sows. J. Anim. Sci.
Miller, G.M., Breen, S.M., Roth, S.L., Willenburg, K.L., Rodriguez- Zas, S., Knox, R.V., 2003. Characterization of image and labor Dial, G.D., Marsh, W.E., Polson, D.D., Vaillancourt, J.-P., 1992.
requirements for positive pregnancy diagnosis in swine using two Reproductive failure: differential diagnosis, In: Leman, A.D., methods of real-time ultrasound. Swine Health Prod. 11, 233–239.
Straw, B.E., Mengeling, W.L., D'Allaire, S., Taylr, D.J. (Eds.), PigCHAMP, 1996. PigCHAMP Reports Manual. University of Diseases of Swine, 7th edition. Iowa State University Press, Rosner, B., 1989. Fundamentals of Biostatistics, 3rd ed. PWS-Kent Dos Santos, J.M.G., Wentz, I., Bortolozzo, F.P., Barioni Jr, W., 2004.
Early-weaned sows: altrenogest therapy, estrus, ovulation, and Statistical Analysis System (SAS), Version 9.1. SAS Institute, Cary, reproductive performance. Anim. Reprod. Sci. 84, 407–413.
Dyck, G.W., 1974. The effects of stage of pregnancy, mating at the first Steverink, D.W.B., Soede, N.M., Groenland, G.J.R., van Schie, F.W., and second estrus after weaning and level of feeding on fetal Noordhuizen, J.P.T.M., Kemp, B., 1999. Duration of estrus in survival in sows. Can. J. Anim. Sci. 54, 277–285.
relation to reproduction results in pigs on commercial farms.
Elbers, A.R.W., van Rossem, H., Schukken, Y.H., Martin, S.W., van Exsel, Ad C.A., Friendship, R.M., Tielen, M.J.M., 1994. Return to Tubbs, R.C., 1990. Factors that influence the weaning-to-estrus oestrus after first insemination in sow herds [incidence, season- interval in sows. Compend. Contin. Educ. Pract. Vet. 12, 105–115.
ality, and association with reproductivity and some blood Tummaruk, P., Lundeheim, N., Einarsson, S., Dalin, A.-M., 2001.
Repeat breeding and subsequent reproductive performance in Gustafsson, H., Larsson, K., Kindahl, H., Madej, A., 1986. Sequential Swedish Landrace and Swedish Yorkshire sows. Anim. Reprod.
endocrine changes and behaviour during oestrus and metoestrus in repeat breeder and virgin heifers. Anim. Reprod. Sci. 10, 261–273.
Van de Wiel, D.F.M., Booman, P., 1993. Post-weaning anoestrus in Jindal, R., Cosgrove, J.R., Aherne, F.X., Foxcroft, G.R., 1996. Effect primiparous sows: LH patterns and effects of gonadotropin of nutrition on embryonal mortality in gilts: association with injection and boar exposure. Vet. Q. 15, 162–166.
progesterone. J. Anim. Sci. 74, 620–624.
King, V.L., Koketsu, Y., Reeves, D., Xue, J., Dial, G.D., 1998.
Management factors associated with swine breeding-herd produc-tivity in the United States. Prev. Vet. Med. 35, 255–264.
Please cite this article as: Takai, Y., Koketsu, Y. Number of services and the reservice intervals in relation to suboptimal reproductiveperformance in female pigs on commercial farms. Livest. Sci. (2007),

Source: http://www.isc.meiji.ac.jp/~animals/paper/takai.reservice.pdf