Effects of nursing a large litter and ovarian response to gonadotropins at weaning on subsequent fertility in first parity sows

Abstract

Post-weaning fertility failures occur more often in parity 1 (P1) sows due to high metabolic demands for lactation and their inability to meet energy requirements for maintenance, growth, and reproduction. We hypothesized that body condition loss occurs more frequently in P1 sows nursing a large litter, resulting in impairment of ovarian follicle development during lactation and post-weaning, which can negatively impact estrus and subsequent fertility. At 24 h post-farrowing, P1 sows (n = 123) were assigned to treatment (TRT) based on sow weight and the number of functional teats to receive a high number (HN, 15 to 16) or low number (LN, 12) of nursing piglets. At weaning, sows in each TRT were assigned to receive PG600 or None (Control). During lactation, sow body measures were obtained and ovarian follicles were assessed in mid-lactation and post-weaning. Lactation data were analyzed for the effects of TRT, and fertility data after weaning were assessed for TRT x PG600, but there were no interactions (P > 0.10). During lactation, 22.2 % of HN sows lost ≥ 4 piglets due to death or removal, and so these sows were excluded from further analysis. The HN sows were lighter (−6.2 kg), had less backfat (−1.0 mm), had lower body condition score (−0.4), and lost more nursing piglets (−1.2) than LN sows (P < 0.05). However, HN sows weaned more pigs (14.0) than LN sows (11.0). There was no effect of TRT on wean to estrus interval (4.2 d), but the interval was 0.5 days shorter for PG600 (P = 0.004) than control. There were no effects of TRT or PG600 on estrus within seven days after weaning (87.3 %), but PG600 induced smaller (P = 0.002) follicles at estrus (6.7 mm) than control (7.3 mm). In the subsequent parity, there were no effects of TRT or PG600 on farrowing rate (93.9%) and total born (13.2). Overall, HN sows lost more piglets and body condition but still weaned more pigs without any detrimental effects on subsequent reproductive performance.

A proportion of primiparous sows nursing a large litter lose more body weight and a greater number of piglets during lactation compared to those nursing fewer piglets. However, the prolific modern sow demonstrated excellent potential to wean more piglets without detrimental consequences to subsequent fertility.

Introduction

Primiparous sows make up the second largest population in a parity distribution in a swine breeding herd and are also the most susceptible to reproductive failures (Koketsu et al., 2017; Tani et al., 2018). Although the majority of sows are expected to return to estrus within seven days after weaning (Bates et al., 2000; Tummaruk et al., 2010), it is common to note increased wean to estrus intervals and lower fertility in parity 1 (P1) sows. It has been estimated that 20.0%–27.5% of P1 sows have delayed estrus (≥7 and ≥6 days, respectively) after weaning (Yatabe et al., 2019). In each case, the delays increase feed costs and space utilization for non-productive days and can prevent farms from reaching weekly production targets.

The origin of the delayed estrus in primiparous sows has been attributed to several risk factors such as season, lower body fat reserves, reduced feed intake, weight loss in lactation, and the impact of lower body condition at weaning (Koketsu et al., 2017). These factors can combine to allow energy expenditure to exceed energy intake which results in a negative energy balance leading to weight loss during lactation. The prolonged interval between weaning and estrus may also occur more frequently in P1 due to the same effects and because these sows need more nutrients as they are still in the process of growing in body size. The result of reduced nutrient intake, high need for milk production and maintenance, and lower body reserves in a growing female could decrease the secretion of gonadotropins and limit the growth and development of ovarian follicles (Koketsu et al., 1996a, 2017). There is sufficient evidence to suggest that the fertility problems in a sub-population of P1 sows may arise from the hypothalamic-pituitary-axis, the ovary, or both. Evidence that the hypothalamus and pituitary may not respond or function normally to release FSH and LH, comes from studies where P1 sows were treated with PG600 at weaning and showed increased follicle development and estrus expression (Vargas et al., 2006; Hidalgo et al., 2014). There is also data to suggest the problem may partly reside at the level of the ovary, as not all P1 sows grow follicles in response to exogenous gonadotropins (Nissen et al., 1995; Youngquist and Threlfall, 2006; Knox, 2015). Additional studies also suggest that prolonged weaning to estrus intervals can result in lower farrowing rates and litter size (Hoshino and Koketsu, 2008; Tummaruk et al., 2010). The reduced fertility may be due to the shorter duration of estrus and a shorter interval between estrus and ovulation (Kemp and Soede, 1996), increasing the risk of suboptimal time for insemination.

Litter sizes have been increasing over years due to improvements in genetic selection technology and animal management (Knol et al., 2016; Zak et al., 2017). The economic importance of litter size born to the increased number of weaned pigs places great emphasis on large litters and piglet survival to weaning. The increase in litter size requires more piglets nursing (>15 piglets) a single sow, which could predispose some hyper-prolific primiparous sows to negative energy balance and subsequent reproductive failure. Further, evidence indicates that stimulating teat development by nursing a larger litter in the first parity aids in the development of those teats in the second parity (Fraser et al., 2010; Farmer et al., 2012). However, nursing a larger litter increases metabolic demands due to the necessity of mobilizing body reserves to support milk production and body maintenance, often leading to body weight and body condition loss. Within the female, the high metabolic demand may reduce nutrients available for normal reproductive hormone production and release leading to compromised development of larger follicles (Quesnel and Prunier, 1995; Quesnel et al., 2007) and problems in subsequent reproductive performance (Kemp and Soede, 2012).

We hypothesized that body condition losses occur more often in P1 sows nursing a large number of piglets and lead to reduced ovarian follicle development, delays in estrus after weaning, and reduced fertility at the subsequent parity. The origin of the problem may be attributed to inadequate hormone release from the hypothalamic-pituitary axis in response to ovarian estrogen feedback or the inability of the ovarian follicles to respond normally to gonadotropins. To address each possibility sows nursing a large or a small litter would either receive exogenous gonadotropins or none at weaning.

Materials and Methods

The use of animals for this experiment was approved by the Institutional Animal Care and Use Committee of the University of Illinois at Urbana-Champaign (#19027).

The experiment was performed in five replicates from September to November 2020 at a 6,500 sow, breed-to-wean farm located in western Illinois. The farm housed females in individual stalls (1.24 m²) from breeding until 113 ± 1 d of gestation. Thereafter, females were relocated to rooms with individual farrowing stalls until piglets were weaned after a 22-day lactation. The feed composition used in breeding and gestation was the same and was a corn-soybean meal, DDGS diet that met or exceeded the recommendations of the 2012 NRC requirements (NRC, 2012).

The females used in this study arrived at the farm at 3 to 4 weeks of age and were housed in group pens of 50 gilts for 8 weeks in an isolation barn next to the sow farm. At 12 weeks of age, the gilts (PIC North America, Hendersonville, TN) were moved to the gilt development unit (GDU), connected to the sow farm. The farm protocol for puberty stimulation and estrous detection in the GDU began at ~160 d and continued until 260 d of age in pens of 50 to 70 gilts. Gilts received once-daily physical boar exposure in their pen using two mature Meishan cross boars for 1 min per boar per gilt. Gilts that had previously expressed one or two normal estrous events, and were ≥210 d of age and ≥136 kg, were moved to the sow farm to enter a breeding group.

Initially, a total of 126 parity 1 (P1) sows were assigned to treatment (TRT) within 24 h of farrowing their first litter. Sows were assigned to receive an HN, 15 to 16 or an LN, 12 of nursing piglets, based on their number of functional teats (producing milk, elongated, pointed with no visual defects, and can be suckled by the piglet) and post-farrowing weight to ensure a minimum of 15 functional teats in HN sows and to balance initial lactation weight between HN and LN sows. Before farrowing, females were fed 1.8 kg per day and after farrowing females were fed ad libitum. Three females were excluded from the study during the experiment due to early weaning unrelated to the TRT, and 14 females were excluded from the HN group due to piglet death or removal of an excessive number (≥4 piglets) of nursing piglets and required their removal from the HN TRT category during the 21.9 ± 0.2 d lactation period. Thus, a total of 109 females were used for analysis. At weaning, sows were assigned by TRT and sow weight at weaning to receive PG600 by i.m. injection (400 I.U. of eCG and 200 I.U. hCG, Merck Animal Health, Rahway, NJ) or no injection (control).

Sow measures during lactation

During lactation, individual sow weight was measured using a floor scale (Salter Brecknell Floor Scales, 2000 x 1 lb, Brooklyn, NY) within 24 h of farrowing (day −21) and on the day before weaning (day −1) to calculate weight changes. On days −21, −10, and −1, classification for body condition score (BCS) was performed using visual appraisal based on a scale of 1 (extremely thin) to 5 (over-conditioned), the use of the Caliper to measure angularity of sows (Knauer and Baitinger, 2015), and with B-mode ultrasound to measure backfat thickness (BF) and loin muscle depth (LMD). Ultrasound was performed using an Aloka Prosound (Hitachi Aloka Medical, Ltd., Wallingford, CT) with a 5.0 MHz linear array transducer and video images digitally recorded. Backfat was measured at the P2 site on both sides of the sow (6 cm from the midline after the last rib) and then averaged. Loin muscle depth was measured behind the scapula between the third and fourth rib (6 cm from the midline after the last rib from the right side).

Reproductive measures for sows during lactation and after weaning

Before assignment to TRT, sow data were obtained for total piglets born, piglets born alive, stillborns, mummies, and number of functional teats (day −21). At mid-lactation (day −10), at weaning (day 0), and after weaning (until estrus or day 7 if no estrus), counts and sizes of follicles were measured on ovaries by transrectal ultrasound using an Aloka Prosound with a 7.5 MHz linear array transducer (Knox and Althouse, 1999) with continuous digital recording. Ovaries were classified by the presence of small (≤3.0 mm), medium (>3.0 and <6.5 mm), large sized (≥ 6.5 mm), or cystic (≥ 12.5 mm) follicles. To analyze the proportion of females with follicles able to ovulate (≥5.0 mm) at estrus and the proportion of females with large follicles at estrus, only females in estrus within 7 d after weaning were included. For these analyses, a minimum of three follicles of the same size class were required for inclusion.

After weaning, sows were checked for estrus once a day from day 2 until 30 using a mature Meishan cross boar. If females did not stand in estrus within 7 d of weaning, females were moved to an opportunity row of the breeding room and no further information was collected. For sows that did express estrus within 7 d of weaning, females received a post cervical insemination containing 1.5 billion sperm in a 35 mL volume every 24 h while standing in estrus, with a maximum of three inseminations for each female. At 30 d after first insemination, transabdominal ultrasound was performed to detect pregnancy. Data obtained for wean to estrous interval (WEI), duration of estrus, percentage of females in estrus within 7 d of weaning, number of inseminations, and pregnancy rate were analyzed. Records of the subsequent performance as a parity two sow were also obtained to analyze farrowing rate, total born, born alive, stillborn, and mummies among treatments.

Piglet performance

At fostering (within 24 h post parturition), only healthy and larger-looking piglets (≥1.0 kg) were chosen by farm staff for use in the study to improve litter survivability and stimulate teat development in P1 sows. The weight of a subpopulation of litters (n = 69) was obtained within 5 h after cross-fostering and the average piglet weight was calculated. All litters were processed according to the farm standard operating procedure (3 to 5 d post farrowing). Total piglets assigned, the number of piglets lost by day, total piglets weaned, and the proportion of piglets weaned were recorded and analyzed. Any piglets removed as mortality or for falling behind on growth compared to the rest of the litter were counted and recorded with the removal date and reason to be included in the analysis. Piglets that appeared to be not gaining weight or growing in size compared to their litter mates were removed by staff according to the farm protocol and removal was recorded and analyzed as a fall-behind piglet.

Statistical analysis

Data were analyzed using procedures in SAS 9.4 (SAS Institute Inc., Cary, NC). Sows were considered the experimental unit. Data were evaluated for normality of residuals and homogeneity of variances to meet the assumptions for ANOVA using the UNIVARIATE procedure. The models used for analysis during lactation included the main effects of TRT (HN or LN) and the models for analysis after weaning contained the main effects of TRT, PG600 or Control and their interaction. The interaction was excluded from the models when not significant. Replicate was included in the models and removed when not significant. Total piglets born before assignment to TRT was included in the models and removed if not significant. Continuous response measures were analyzed using PROC MIXED. Binary response measures and proportional data were analyzed using PROC GLIMMIX using a binary distribution and a logit-link function. Results were considered significant at P ≤ 0.05, and trends with P > 0.05 and P ≤ 0.10. Data are presented as least square means ± standard error. For the analysis of piglet death and removal, all P1 sows were used (n = 123). However, due to the reduction in the number of piglets in HN (≥4 piglets) during the treatment period, some sows (n = 14) were excluded from further analysis. Also, because females were not assigned based on the number of piglets born and were different between HN and LN, the analysis contained the total number of piglets born as a covariate in the models and was deleted if not significant. For females to be included in the follicle measures analysis, sows needed to express estrus within 7 d of weaning.

Results

Initial parameters before treatment assignment

Total born for the sows first litter was not included as a criteria in the assignment for TRT in this study due to the emphasis on use of number of functional teats and weight of sows. As a result, at assignment, primiparous sows assigned to HN had 1.2 more total piglets born (P = 0.05) than LN sows. However, piglets born alive (14.2 ± 0.4), stillborn (0.6 ± 0.1), and mummies (0.3 ± 0.1) were not different between TRT (P > 0.10, Table 1).

Table 1.

Primiparous sow parameters before assignment to a HN, 15 to 16 or a LN, 12 of nursing pigs at 24 h post farrowing (day −21)

	Treatment
	HN	LN
n	49*	60	SEM	P-value
Total pigs born	15.7	14.5	0.4	0.05
Pigs born alive	14.5	13.8	0.4	0.20
Stillborn	0.7	0.5	0.1	0.21
Mummies	0.5	0.2	0.1	0.11

^*Sows (n = 14) were excluded from HN group due to excessive number of piglets that died or were removed due to poor health or growth.

Weight, backfat, loin muscle depth, and body condition of P1 sows during lactation

Response measures for P1 sows assigned to TRT are shown in Table 2. At the start of lactation, sow weight was balanced between HN and LN and was not different (P > 0.10, 179.4 ± 2.4 kg). However, by the end of lactation (day −1), there was a difference (P = 0.03) with HN sows 6.2 ± 2.1 kg lighter than LN sows. The proportion of females that lost weight during lactation was not affected by TRT (P > 0.10, 77.9% ± 6.0%), and for those that lost weight (n = 81), there was no effect of TRT on the kg of sow weight loss (−9.6 ± 1.1 kg, P = 0.30) or percentage of body weight loss (−5.4% ± 0.6 %, P = 0.27). A total of 11 females (11.9% ± 4.8%) lost more than 10 % of their body weight but with no difference between the frequency between the HN (n = 7) and LN (n = 4) treatments. Also, there was no effect (P > 0.10) of TRT on the proportion of females that gained weight (22.2% ± 6.0 %), and for those that gained weight (n = 28), there was no effect of TRT on the amount of weight gained (P > 0.10, 2.1 ± 0.3 kg).

Table 2.

Least square means for body condition measures of primiparous sows assigned to a HN, 15 to 16 or a LN, 12 of nursing pigs 24 h post farrowing (day −21)

	Treatment
	HN	LN
n	49	60	SEM	P-value
Sow weight day −21, kg	177.9	180.8	2.4	0.37
Sow weight day −1, kg	170.0	176.2	2.1	0.03
Proportion of females that lost weight, %	82.5 (39/49)	73.1 (42/60)	6.0	0.24
Sow weight loss during lactation, kg	−10.3	−8.8	1.0	0.30
Sow weight loss during lactation, %	−5.8	−4.9	0.6	0.27
Proportion of females that gained weight, %	17.4 (10/49)	26.9 (18/60)	6.0	0.24
Sow weight gain during lactation, kg	1.7	2.4	0.3	0.09
Backfat day −21, mm	11.1	11.0	0.3	0.52
Backfat day −10, mm	9.9	10.8	0.3	0.06
Backfat day −1, mm	9.8	10.8	0.4	0.05
Proportion of females that lost backfat, %	70.0 (28/40)	56.9 (29/51)	7.0	0.20
Backfat loss by day −10, mm	−1.3	−1.6	0.2	0.24
Backfat loss day −10 to day −1, mm	−1.0	-1.5	0.3	0.39
Backfat loss by day −1, mm	−1.4	−1.5	0.2	0.87
Proportion of females that gained backfat,%	30.0 (12/40)	43.1 (22/51)	7.0	0.20
Backfat gain by day −10, mm	1.1	2.0	0.8	0.57
Backfat gain day −10 to D −1, mm	1.0	1.4	0.2	0.29
Backfat gain by day −1, mm	0.8	1.8	0.3	0.03
Backfat difference (day −1 to −day −21), mm	−0.7	−0.1	0.2	0.08
Loin muscle depth day −21, mm	46.6	48.3	1.2	0.31
Loin muscle depth day −10, mm	48.1	48.8	0.9	0.35
Loin muscle depth day −1, mm	50.8	52.3	0.9	0.22
Proportion of females that lost loin muscle, %	36.7 (11/30)	50.0 (23/46)	8.0	0.25
Loin muscle loss by day −10, mm	−4.3	−6.4	2.2	0.52
Loin muscle loss day −1 to day −1, mm	−2.4	−4.4	1.1	0.19
Loin muscle loss by day −1, mm	−7.0	−10.8	2.2	0.26
Proportion of females that gained loin muscle, %	63.3 (19/30)	50.0 (23/46)	8.0	0.26
Loin muscle gain by day −10, mm	7.5	7.6	1.6	0.96
Loin muscle gain day −1 to day −1, mm	1.4	1.9	0.3	0.31
Loin muscle gain by day −1, mm	10.1	12.8	1.6	0.27
Loin muscle difference (day −1 to day −21), mm	3.4	0.9	1.4	0.20

Lactation length averaged 22.0 ± 0.2 d.

In terms of fat tissue measures, there were no effects of TRT (P > 0.10) on backfat (BF) thickness at the beginning of lactation (11.1 ± 0.3 mm), although there was a trend for greater backfat (+0.9 mm) in LN than HN at mid lactation (day −10, P = 0.06), and a greater (P = 0.05) backfat (+1.0 mm) in LN than HN at the end of lactation (day −1). There was no effect of TRT on the proportion of females that lost BF (P > 0.10, 63.5% ± 7.0%), and from these sows (n = 62), there was also no effect of TRT (P > 0.10) on BF loss during the first half of lactation (−1.4 ± 0.2 mm), the second half of lactation (−1.3 ± 0.3 mm), or throughout the entire lactation period (−1.4 ± 0.2 mm). In addition, there was no effect of TRT on the proportion of females that gained BF (P > 0.10, 36.6 ± 7.0 %), and from these sows (n = 41), there were no effects of TRT (P > 0.10) on BF gain by mid-lactation (1.6 ± 0.8 mm), the second half of lactation (1.2 ± 0.2 mm), but there was an effect of TRT with greater backfat gain (+ 1.0 mm) in LN (P = 0.03) by the end of lactation.

There were no effects (P > 0.10) of TRT on loin muscle depth (LMD) at the beginning (47.5 ± 1.2 mm), mid (47.9 ± 0.9 mm), and end of lactation (51.6 ± 0.9 mm), and no effect on LMD loss by mid-lactation (−5.4 ± 2.2 mm), the second half of lactation (−3.4 ± 1.1 mm), and by the end of lactation (−8.9 ± 2.2 mm), or on the proportion of sows that lost muscle (43.4% ± 8.0%, n = 39). In addition, there was no effect of TRT (P > 0.10) on LMD gain by mid-lactation (7.6 ± 1.6 mm), in the second half of lactation (1.7 ± 0.3 mm), and during lactation (11.5 ± 1.6 mm), or on the proportion of sows that gained muscle during lactation (56.7 ± 8.0 %, n = 51).

Furthermore, there was no effect of TRT on BCS (P > 0.10, Figure 1) at the beginning of lactation (3.0 ± 0.05) and mid-lactation (2.8 ± 0.05). However, there was an effect of TRT (P < 0.0001) on BCS at the end of lactation (day −1), resulting in a greater BCS in LN (3.0 ± 0.05) than HN (2.6 ± 0.05) sows. There was also no effect of TRT on body condition using the Caliper (P > 0.10, Figure 2) at any of the days for measure throughout lactation.

Figure 1.

Body condition scores (BCS, 1 to 5) of primiparous sows assigned to a HN, 15 to 16 or a LN, 12 of nursing pigs, 24 h post farrowing (day −21), in mid-lactation (day −10), and one day before weaning (day −1). There was an effect of TRT (P < 0.0001) resulting in a greater BCS in LN (3.0 ± 0.05) than HN (2.6 ± 0.05) sows at the end of lactation (day −1).

Figure 2.

Body condition Caliper measurement of primiparous sows assigned to a HN, 15 to 16 or a LN, 12 of nursing pigs, 24 h post farrowing (day −21), in mid-lactation (day −10), and one day before weaning (day −1). There were no effects of TRT on BCS during lactation.

Piglet performance

Piglet loss during lactation based on death or removal due to falling behind in growth compared to other piglets in the litter is shown by TRT for each quartile of the lactation period (Table 3). The total number of piglets nursing assigned by TRT at the beginning of lactation averaged 15.5 piglets in the HN group and 12.0 in the LN group, and based on the subpopulation of litters weighed, the average piglet weight at assignment was not different (Table 4, 1.3 ± 0.03 kg). There was an effect of TRT (P < 0.0001) on the number of piglets lost (removal for death or fall behind piglets), with more losses in HN by mid-lactation (1.0) and by the end of lactation (2.1) compared to LN. The LN females weaned 2.2 less piglets than HN sows (P < 0.0001) but weaned a greater proportion (+5.8%) of the pigs assigned for nursing compared to the HN group (P = 0.002).

Table 3.

Piglet death and removal due to falling behind in growth compared to other pigs in the litter during lactation in primiparous sows assigned to nurse a HN, 15 to 16 or a LN, 12 of pigs.

	Treatment
	HN				LN
	Lactation days relative to weaning (day 0)
	−21 to −17	−16 to −11	−10 to −6	−5 to −1	−21 to −17	−16 to −11	−10 to −6	−5 to−-1
Crushed, %	11.4	4.0	0.0	0.0	5.0	0.0	0.5	0.0
Low viability, %	3.0	2.0	0.0	0.0	1.0	2.0	0.0	0.5
Unknown, %	4.0	0.0	0.5	0.0	1.0	0.0	0.0	0.0
Fall behind pig, %	3.5	14.4	11.9	13.9	1.0	9.4	5.4	5.9

Table 4.

Least square means for piglet losses (death and removal due falling behind in growth compared to other pigs in the litter), and numbers weaned from primiparous sows assigned to a HN, 15 to 16 or a LN, 12 of nursing pigs 24 h post farrowing (day −21)

	Treatment
	HN	LN
n	63	60	SEM	P-value
Total piglets nursing at day −21	15.5	12.0	0.05	.
Average piglet weight at day −21, kg1	1.3	1.3	0.03	0.39
Piglet loss by day −10	1.7	0.7	0.2	<0.0001
Piglet loss by day −10 to day −1	0.4	0.2	0.1	0.26
Total piglet loss	2.1	0.9	0.2	<0.0002
Total piglets weaned	13.3	11.3	0.2	<0.0001
Proportion of pigs weaned, %	86.6	92.4	1.3	0.002

¹Subpopulation of litters weighed (n = 69 sows, HN (n = 36), and LN (n = 33).

Functional teat assessment during lactation

Assessment of sows for number of functional teats was one of the measures used for assigning the number of piglets nursing in each TRT, and for that reason, the number was different (P < 0.0001, Table 5), with 1.8 more functional teats for HN compared to LN sows. The greater number of functional teats (P < 0.0001) for HN sows was still present at mid-lactation (+1.9) and end of lactation (+1.8). Losses in functional teats by mid-lactation, the second half of lactation, and throughout lactation did not differ between treatments (P > 0.10).

Table 5.

Least square means for functional teats assessment during lactation of primiparous sows assigned to a HN, 15 to 16 or a LN, 12 of nursing pigs 24 h post farrowing (day −21)

	Treatment
	HN	LN
n	49	60	SEM	P-value
Functional teats at day −21	15.8	14.0	0.1	<0.0001
Functional teats at day −10	14.1	12.2	0.2	<0.0001
Functional teats at day −1	13.6	11.8	0.2	<0.0001
Functional teat loss by day −10	−2.0	−2.2	0.2	0.46
Functional teat loss day −10 to day −1	−1.6	−1.5	0.2	0.76
Functional teat loss by day −1	−2.3	−2.5	0.2	0.48

Reproductive sow measures during lactation and after weaning

Ovarian assessment of mid-lactation females (n = 91) revealed no influence of TRT on the average size of the three largest follicles (3.1 ± 0.2 mm, P > 0.10) or on the proportion of females with medium-sized follicles (26.4 ± 6.6 %, P > 0.10), while the remainder of sows had only small sized follicles. From the females that had medium follicles as their largest size at mid-lactation (n = 25), the average size of the three largest follicles was not affected by TRT (4.6 ± 0.2 mm, P > 0.10). Also, the presence of medium follicles in a proportion of the females at mid-lactation did not influence wean to estrus interval (WEI) or expression of estrus within 7 d of weaning.

There were no significant interactions of TRT x PG600 for any reproductive measures (Table 6). Results from analysis for TRT assignment at the start of lactation demonstrated no effect on WEI (4.2 ± 0.2 d, P > 0.10). However, there was an effect of PG600, that resulted in a 0.5 d shorter interval to estrus compared to control (P = 0.004). There were no effects of TRT on expression of estrus within 7 d of weaning (87.3 ± 6.1 %, P > 0.10), but there was a trend for HN Control and LN PG600 (P = 0.09) to have lower expression of estrus within seven days of weaning compared to other treatments. There was no effect of TRT, but a trend for an effect of PG600 (P = 0.09) with a greater proportion of control sows having large follicles on the first day of estrus (+13.9%) compared to PG600. Also, there were no effects of TRT, or PG600 (P > 0.10) on the proportion of females with follicles of ovulatory size (≥5 mm) on the first day of estrus (95.8 ± 3.1 %). Follicle size at estrus was smaller for PG600 (−0.6 mm) than control (P = 0.002). The number of inseminations per sow during estrus (1.9 ± 0.1) did not differ. In the subsequent parity, day 30 pregnancy rate (96.9% ± 3.1%), farrowing rate (93.9% ± 3.3%), total piglets born (13.0 ± 1.0), piglets born alive (12.5 ± 0.8), and stillborn (0.5 ± 0.2) were not affected by TRT or PG600 (P > 0.10).

Table 6.

Least square means for post-weaning reproductive responses and parity 2 (P2) performance of primiparous (P1) sows assigned to a HN, 15 to 16 or a LN, 12 of nursing pigs 24 h post farrowing (day −21) and receiving either PG600 or none (control) at weaning (day 0)

	Treatment					P-value
	HN		LN
	PG600	Control	PG600	Control		TRT	PG600	TRT * PG600
n	24	25	30	30	SEM
WEI, d	4.0	4.4	3.8	4.4	0.2	0.57	0.004	0.55
Estrus within 7 d, %	95.8	80.0	83.3	90.0	6.1	0.60	0.40	0.09
Follicles ≥6.5 mm at estrus, %1	84.2	93.8	65.2	84.0	8.5	0.10	0.09	0.99
Follicles ≥5 mm at estrus, %1	100.0	100.0	83.3	100.0	3.1	0.98	0.97	0.99
Follicle size at estrus, mm2	6.8	7.2	6.5	7.3	0.2	0.63	0.002	0.24
Number of A.I.	1.8	1.9	1.9	1.8	0.1	0.80	0.37	0.33
Pregnancy rate, %	91.3	100.0	96.0	100.0	2.4	0.51	0.97	0.99
Farrowing rate, %	87.0	100.0	88.0	100.0	3.4	0.91	0.97	0.99
P2 Total pigs born	13.2	13.3	13.7	12.3	1.0	0.76	0.35	0.38
P2 Pigs born alive	12.4	12.9	12.9	11.6	0.9	0.55	0.44	0.23
P2 Stillborn	0.4	0.3	0.7	0.6	0.3	0.45	0.63	0.76
P2 Mummies	0.3	0.1	0.1	0.1	0.2	0.39	0.82	0.11

¹Only females that expressed estrus within seven days were included in the analysis

²Follicle size was based on the average of the three largest follicles

Discussion

The results of this study indicate that P1 sows nursing 15 or 16 piglets during lactation are lighter and thinner at weaning and lose more piglets than P1 sows nursing 12 piglets. These findings support the theory that young females are more prone to lose body condition when subjected to higher milk production pressure (Quesnel et al., 2007). However, our study also demonstrated that P1 sows nursing more piglets could wean more piglets, and surprisingly, despite loss of body condition, showed no negative impact on fertility postweaning and in the subsequent parity. The absence of detrimental effects on P1 fertility differs from previous studies that demonstrated the susceptibility of P1 sows to reproductive impairment postweaning, and attributed to the high metabolic demands of nursing, sow growth, and maintenance (Quesnel and Prunier, 1995; Koketsu et al., 1996a, 1996b). Our study also aimed to address the problem of delayed estrus postweaning that may originate from inadequate hormone release from the hypothalamic-pituitary axis or the inability of ovarian follicles to respond to endogenous gonadotropins. We investigated these different sources for failure by administering an exogenous gonadotropin (PG600) at weaning in half of the females in each treatment. There were minimal differences in the fertility responses of P1 females receiving PG600, indicating that the ovaries were not compromised in the ability of their follicles to bind and respond to gonadotropins at weaning and no evidence that estrus responses were altered. The results of the present study were surprising in that nursing the large litter had no detrimental consequences on P1 sows.

The sow metabolic indicators measured during this study confirmed that P1 sows nursing a larger litter were more susceptible to body condition loss. During the lactation period, sows typically mobilize lean and fat mass (McNamara and Pettigrew, 2002; Schenkel et al., 2010) to produce milk and to meet their requirements, and as a result, are more likely to lose body condition. In recent years, the selection for a highly prolific and leaner sow has created a female with a higher capacity for total piglets born and piglets weaned (Kemp et al., 2018), but this has also increased the metabolic demands on the female. These higher demands become especially important for the P1 sow, which, when compared to multiparous females, has lower feed intake capacity and retain requirements for growth as well as maintenance and milk production (Strathe et al., 2017). In the present study, we were not able to measure sow milk production and feed intake. However, a previous study established that milk protein and milk fat production are related to muscle depth mobilization and backfat loss, respectively (Costermans et al., 2020a). This may explain why the HN sows lost more backfat, body weight, and body condition than LN sows, although no negative effect on muscle depth was noted. In terms of feed intake, other studies show that P1 sows nursing 10 to 14 piglets are still more prone to lose weight because they are unable to eat enough feed to meet all their requirements during lactation (Kemp et al., 2018). Considering these pieces of information, lowering the number of piglets that are nursing in P1 sows to prevent body condition loss could cause detrimental impacts in terms of economics and sow longevity. A hyper prolific sow farrowing a larger litter requires a matching number of functional teats. Decreasing the number of piglets nursing in P1 sows could be deleterious for the young females that require full mammary gland stimulation for lifetime teat development and full nursing potential for future parities (Farmer et al., 2012). In addition, the decrease in the number of piglets nursing in P1 sows would require the availability of additional nursing females to occupy farrowing room spaces to accommodate these piglets and would increase the non-productive days at the farm.

With no impact of treatment, 12% of the P1 sows still lost more than 10% of their body weight in the present study. Previous studies demonstrated that primiparous sows that underwent severe weight loss (>10% or >12%) during their first lactation are more likely to display prolonged wean to estrus interval (Tantasuparuk et al., 2001), reduced pregnancy rate (Hoving et al., 2012), and farrowing rate (Thaker and Bilkei, 2005). Studies that focused on the metabolic state of P1 sows related negative energy balance to sow fertility in the subsequent parity, such as effects on oocyte quality and follicle development postweaning (Zak et al., 1997; Costermans et al., 2020b), and embryo development post insemination (Patterson et al., 2011; Hoving et al., 2012). The present study identified no impacts in follicle development, pregnancy, and farrowing for the second litter. Although the modern P1 sow nursing more piglets might show changes in body weight and other metabolic indicators, most P1 sows demonstrated adjustment for high lactation demands in the HN group and remained highly productive.

Another objective of our study was to test if delays in P1 sow follicular development and estrus after weaning were attributed to inadequate hormone release from the hypothalamic-pituitary axis or the inability of ovarian follicles to respond to gonadotropins released by the pituitary. If the problem was related to production or release of LH and FSH by the pituitary, then treating P1 sows with PG600 would resolve the dilemma, and females would develop follicles and express estrus. However, if the issue was related to ovarian failure to respond to gonadotropins, then PG600 would have no effect on stimulating follicle development. Our results showed that ~87% of P1 sows were in estrus within 7 d of weaning, with no clear effects of PG600. However, a reduction in the wean to estrus interval was observed for those sows administered PG600, which confirmed previous findings when PG600 was administered on the day of weaning (Kirkwood et al., 1998; Knox et al., 2001) or even when given 2 to 4 d before weaning (De Rensis et al., 2003). Surprisingly, our findings demonstrated smaller follicles on the first day of estrus for those P1 sows treated with PG600 compared to controls, but without any effects on subsequent litter size. In contrast to our findings, Knox et al. (2001) found no effect of PG600 on follicle development in the younger parity groups, which could be explained by significant changes in genetic traits, management approaches, and fertility. Thus, it appears that the exogenous gonadotropins given at weaning in our study stimulated smaller follicles to produce estrogen at high enough levels to advance estrus 0.5 d earlier. The exact reason why the proportion of sows with large follicles was lower and the average size of those smaller is unclear. It is possible that the exogenous application of gonadotropins slightly changed follicle selection and rate of development when compared to the controls, but nevertheless, had no bearing on subsequent measures of fertility.

Besides the information on body condition and fertility of the sows, another valuable aspect of our study was related to piglet survivability in the nursing treatment groups. Overall, 22% of sows assigned to the HN treatment lost or had more than four piglets removed when attributed to either mortality or morbidity, while the other 78% of HN sows still lost significantly more piglets during lactation than the LN females. This outcome is important since female reproductive performance can be defined by the total lifetime production of piglets that survive weaning (Zak et al., 2017). Although the HN group was still capable of weaning a greater number of piglets than the LN group in our study, greater pre-weaning mortality in larger litters is of significant concern because it affects economic and animal well-being aspects of the system (Peltoniemi et al., 2021). Usually, the leading causes of piglet mortality in the first days after farrowing are related to crushing, starvation, and hypothermia, especially in piglets with lower birth weights (Baxter et al., 2008; Devillers et al., 2011). Our study confirms these findings, with the primary reason for piglet death in both treatments occurring in the first five days due to crushing, but which was substantially greater in the HN females nursing a larger litter. From day 6 of lactation until weaning, piglet removal due to low viability was most common in the HN sows, which may have been due to more limited nursing opportunities. One of the limitations of the present study was that the weight of piglets at removal and at weaning was not able to be collected. This data would have been helpful to explain the reasons for removal due to slower growth and would assist with litter performance comparisons between treatments. Collectively, the performance of the HN group held greater advantages over the LN group, because of the increased number of piglets weaned, but the dilemma for this group was that more piglets were lost due to crushing in the first days after farrowing and removal of poor performing piglets in the last days of lactation. There are obvious benefits but some risks with the HN approach. To reduce the chances of piglet loss when allowing more than 15 piglets to nurse a P1 sow, a greater focus on management would be required in both the early stages after farrowing and throughout later lactation to prevent loss of fall behind piglets. Improved management intervention might include drying piglets and minimizing temperature declines during the early post-natal period which have been helpful in avoiding early and later pre-weaning mortality during the cool months of the year (Vande Pol et al., 2021b). Also, cross-fostering by piglet weight categories can be an important tool and can be beneficial especially to light-weight piglets within a large litter (Vande Pol et al., 2021a). Because different farms will have different piglet mortality patterns (Baxter and Edwards, 2018) and as reviewed by Knol et al. (2022), piglet survivability can be affected by genetics, individual farms will need to evaluate the potential for their parity 1 females to support large litters and their strategies for monitoring and supporting piglet survival and growth..

CONCLUSION

This study showed that P1 modern prolific sows can nurse a high number of piglets and still have a high potential to express estrus within seven days, achieve high farrowing rates, and reach target parity two litter sizes. However, a greater proportion of these females lost weight during lactation, so there is still an opportunity to improve management in terms of body condition, although there seems to be little benefit for exogenous gonadotropin application for post-wean follicle development. Even though the HN group clearly showed excellent potential to wean more piglets, the greater loss of piglets due to death and removal, points to a real need for active intervention. Under these circumstances for HN, an approach required may involve increased flexibility in the number of piglets nursing based on functional teats. When establishing the number of piglets allowed to nurse, when greater than 12, and especially when greater than 14, attention would be needed for strategic observation of piglets each day and an intervention plan in place that would allow movement of those slower growing piglets before they become health compromised.

Glossary

Abbreviations

BCS

body condition score

BF

backfat

GDU

gilt development unit

HN

high number of piglets nursing

LMD

loin muscle depth

LN

low number of piglets nursing; P1, parity one

WEI

wean to estrus interval

Conflict of Interest Statement

The authors declare no real or perceived conflicts of interest.