Department of Production Animal Medicine
Faculty of Veterinary Medicine
University of Helsinki
Finland
Parturition and subsequent uterine health and
fertility in sows
Stefan Björkman
ACADEMIC DISSERTATION
To be presented with the permission of the Faculty of Veterinary Medicine of the University
of Helsinki, for public examination in the Auditorium XII, University main building, on 15
September 2017, at 12 noon.
Helsinki 2017
Supervisors by
Professor Olli Peltoniemi
Dr. Claudio Oliviero
Department of Production Animal Medicine
University of Helsinki
Finland
Professor Nicoline Soede
Department of Animal Sciences
Wageningen University & Research
The Netherlands
Reviewed by
Professor Geert Opsomer
Department of Obstertcs, Reproduction and Herd Health
Ghent University
Belgium
Dr. Ylva Sjunnesson
Department of Clinical Sciences
Swedish University of Agricultural Sciences.
Sweden
Opponent
Professor Olav Reksen
Department of Production Animal Clinical Science
Norwegian University of Life Sciences
Norway
ISBN 978-951-51-3665-7 (paperback)
ISBN 978-951-51-3666-4 (PDF)
Helsinki University Printing House
Helsinki 2017
To Venla, Valtteri and Frida
4ACKNOWLEDGEMENTS
I would first like to thank my thesis supervisors Olli Peltoniemi and Claudio Oliviero of the
Department of Production Animal Medicine at the University of Helsinki as well as Nicoline
Soede of the Department of Animal Sciences at the Wageningen University and Research
for their unfailing support and assistance.
Prof. Peltoniemi was an excellent supervisor and consistently allowed this dissertation to be
my own work, approved my own research ideas and steered me in the right direction
whenever he thought I needed it. Furthermore, he also was an excellent life-coach and
friend during the last years. His door was always open and he was always available
whenever I ran into troubles or had questions.
Prof. Soede was also more than just an excellent supervisor. She welcomed me with great
hospitality every time I visited her institution. Although going through a difficult time herself,
she always found time and energy to enormously increase the scientific quality of my
research through interesting and productive discussions. She has impressed me the most
and I wish to be one day as outstanding in research as she is.
I am also very grateful to Dr. Oliviero for his supervision who also always greatly supported
me with advice and solutions when facing practical issues during my trials.
I would also like to thank my two pre-examiners, Prof. Geert Opsomer of the Department
of Obstertcs, Reproduction and Herd Health at the Ghent University, and Dr. Ylva
Sjunnesson of the Department of Clinical Sciences, Division of Reproduction at the Swedish
University of Agricultural Scienes for taking the time to read through this thesis.
A very special gratitude goes out to all my workmates of the pig research group at the
University of Helsinki for help and discussion whenever it was needed as well as to all my
co-authors of my scientific papers, Jinhyeon Yun, Mikael Niku, Mari Heinonen, Anna Valros,
Päivi Rajala-Schultz and Johannes Kauffold, for their valuable input and comments. It was
fantastic to have the opportunity to work with them!
I also would like acknowledge my deepest gratitude to the farmers, funding agencies and
sponsors. Without them, none of the studies would have been possible. The farmers Jari
and Veera Ollikkala and Elina and Mikko Yrjövuori were always welcoming me on their pig
farms for conducting my research. The Finnish Ministry of Agriculture (2017/312/2011),
Vetcare Oy, Atria Oy, Figen Oy and Hankkija Oy provided me with the major funding for
the studies. The Finish Veterinary Foundation, The Finnish Foundation of Veterinary
Research and the Doctoral Programme in Clinical Veterinary Medicine of the University of
Helsinki provided me with additional research and travel grants. Also I would like to thank
Agfa HealthCare Finland Oy for providing me with a PACS system.
Finally, and most importantly, I must express my very profound gratitude to my parents,
parents-in-low and the rest of my family for their support throughout my years of study and
through the process of researching and writing this thesis. Especially, I would like to deeply
thank Jenni, my beloved wife, and Venla, Valtteri and Frida, my beloved kids. They have
been my eternal cheerleaders and their support, encouragement, quiet patience, smiles
and unwavering love were undeniably the bedrock upon which this thesis has been built.
5LIST OF ORIGINAL PUBLICATIONS
This thesis is based on the following publications:
I Oliviero, C., Kothe, S., Heinonen, M., Valros, A., & Peltoniemi, O. (2013). Prolonged
duration of farrowing is associated with subsequent decreased fertility in sows.
Theriogenology, 79 (7), 1095-1099.
II Björkman, S., Oliviero, C., Rajala-Schultz, P. J., Soede, N. M., & Peltoniemi, O. A. T.
(2017). The effect of litter size, parity and farrowing duration on placenta expulsion
and retention in sows. Theriogenology, 92, 36-44.
III Björkman, S., Oliviero, C., Kauffold, J., Soede, N. M., & Peltoniemi, O. A. T. (2017).
Prolonged parturition and impaired placental expulsion increase the risk of
postpartum metritis and delay uterine involution in sows. Theriogenology, in print.
IV Yun, J., Björkman, S., Oliviero, C., Soede, N. M., & Peltoniemi, O. A. T. (2017). The
effect of farrowing duration and parity on plasma oxytocin concentrations and
follicular development in post-weaning sows. Theriogenology, in print.
V Björkman, S., Yun, J., Niku, M., Oliviero, C., Soede, N. M., & Peltoniemi, O. A. T. (2017).
Serial transvaginal ultrasound-guided biopsy of the porcine corpus luteum in vivo.
Reproduction, Fertility and Development, 29(5), 931-939.
The publications are referred to in the text by their roman numerals.
6CONTENTS
1. Abstract…………………………………………………………………………………………….. 9
2. Introduction and literature review……………………………………..…………………….. 13
2.1. Introduction ……………………………………………………...………….………………..14
2.2 Parturition……………………………………………………………………………………… 18
2.3. Puerperium…………………………………………………………….……….................… 22
2.4. Return of cyclical activity………………………………………………………….…….... 26
2.5. Puerperal diseases………………………………………………………………………….. 30
2.6. Corpus luteum function and biopsy of luteal tissue ………………………………..… 34
3. Aims and hypotheses of the study……………………………………………………………. 37
4. Materials and methods………………………………………………………………………… 41
4.1. Animals, housing, and management………………………………………………….... 42
4.2. Sample size and study design……………………………………………………………... 43
4.3. Farrowing observations…………………………………………………………………….. 45
4.4. Sonographic evaluation of the uterus…………………………………………………… 46
4.5. Sonographic evaluation of the ovaries……………………………………….………… 47
4.6. Sonographic evaluation of backfat……………………………………………..……… 48
4.7. Blood collection and oxytocin assay………………………………………………….…. 49
4.8. Transvaginal ultrasound-guided biopsy of the corpus luteum…………………….... 50
4.9. Statistical analysis……………………………………………………………………………. 53
75. Results……………………………………………………………………………………………… 57
5.1. Effect of prolonged farrowing on subsequent pregnancy rate…………………… 58
5.2. Effect of prolonged farrowing on placenta expulsion & retained placentae…… 59
5.3. Effect of prolonged farrowing on postpartum uterine size & intrauterine fluid…….63
5.4. Effect of prolonged farrowing on follicular development after weaning and
oxytocin release at subsequent estrus……………………………………………………...... 67
5.5. Transvaginal ultrasound-guided biopsy of the corpus luteum……………………… 71
6. Discussion…………………………………………………………………………………………. 75
6.1. Parturition and subsequent uterine health and fertility……………………………….. 76
6.2. Transvaginal ultrasound-guided biopsy of the corpus luteum……………………… 81
7. Conclusion, practical implications and future research…………………………………. 83
7.1. Conclusions…………………………………………………………………………………... 84
7.2. Practical implications……………………………………………………………………….. 85
7.3. Future research………………………………………………………………………………. 86
8. References……………………………………………………………………………………….. 89
9. Original articles…………………………………………………………………………………... 97
8
9Abstract
10
There has been a steady increase in litter size, and consequently farrowing duration, in
modern pig production during recent decades. At the same time, annual removal rates
due to reproductive failure have increased and an association with litter size
demonstrated. Underlying physiological mechanisms for the associations between these
phenomena are not well established. It is plausible that the increase in farrowing duration
is causally linked with the increase in litter size because it takes longer to farrow a larger
litter. Therefore, it is possible that prolonged farrowing negatively affects subsequent
reproductive performance of sows.
Postpartum dysgalactia syndrome (PDS), the major puerperal disease complex of sows,
can cause subfertility, but it is not clear that this is linked with prolonged farrowing.
However, PDS and prolonged farrowing represent similar risk factors regarding increased
backfat, constipation, low dietary fiber, and restricted movement in farrowing crates. In
some studies there was evidence of a connection between prolonged farrowing and
increased incidence of general clinical symptoms such as fever and anorexia. Therefore, it
is feasible that prolonged farrowing increases the risk for PDS, which in turn increases the
risk of subsequent subfertility.
In order to understand better these tentative relationships it makes sense to look at the
different symptoms individually, it being most likely that prolonged parturition in particular
affects the incidence of metritis. Puerperal metritis is an acute inflammation and infection
of the uterus that occurs during the first week after parturition and is characterized by an
enlarged uterus that results from fluid accumulation. In other species, including the cow, a
long and difficult parturition increases the risk of puerperal metritis, which in turn delays
uterine involution and subsequently reduces fertility by disturbing postpartum ovarian
follicular growth and ovulation. Furthermore, long and difficult parturitions can cause
retained placenta, which is of major consequence in the etiology of postpartum metritis.
We hypothesized that prolonged farrowing decreases subsequent fertility, that is,
pregnancy rate. Furthermore, we hypothesized that prolonged farrowing causes retention
of placentae and metritis, delays uterine involution and perturbs follicular growth after
weaning. We also hypothesized that sows that undergo prolonged farrowing release less
oxytocin at subsequent estrus in response to boar stimulation. It is also known that
prolonged parturition is linked with reduced oxytocin concentrations. We therefore
speculated that this is the case also at subsequent breeding. At that time, oxytocin is
needed for sperm transport to the site of fertilization and formation of the sperm reservoirs
and thus, impairment of transport may have a negative effect on the establishment of
pregnancy.
Furthermore, corpus luteum function is important for establishment and maintenance of
pregnancy. The primary function of the corpus luteum is secretion of progesterone, the
pregnancy hormone, and this function depends on other hormones active before
embryonic attachment and maternal recognition. Whether corpus luteum function is
compromised in sows with prolonged farrowing was not investigated in this study, but
validation and testing of a transvaginal ultrasound-guided biopsy method for luteal tissue
was done. This represents an easy and reliable tool for future investigations of corpus
luteum function.
In the first part of the study we determined the farrowing duration (time between expulsion
of first and last piglet) of sows and explored whether there is a negative effect on
subsequent post-weaning pregnancy rate (n = 148). In the following part, we explored the
relationship between farrowing duration and placenta expulsion (n = 142), and
postpartum uterine size and intrauterine fluid (n = 99). For that, placenta expulsion was
11
observed until 24 h after birth of the last piglet and ultrasonography was used during the
first week postpartum to examine the uteri of the sows. Uterine size and intrauterine fluid
were used as indicators for initial uterine involution and puerperal metritis. Furthermore, we
determined the farrowing duration of sows (n = 30) and monitored the subsequent
follicular development using transrectal ultrasound twice a day between the third day
after weaning and ovulation. At estrus, blood samples were collected in the presence of a
boar in order to determine the endogenous oxytocin concentrations and release patterns.
The results show that sows with a prolonged parturition (> 300 min) were 3.4 (odds; P =
0.027) more likely not to be pregnant. Farrowing duration was highly associated with
retained placentae. Sows that experienced total (no expulsion of placental parts; 3%) and
partially retained placentae (no expulsion of placental parts after birth of the last piglet;
3%) had longer farrowing durations (1009 ± 275 and 734 ± 136 min) than sows without
retained placentae (369 ± 202 min; P = 0.021 and P = 0.004). Otherwise, farrowing duration
conformed to a quadratic relationship with the number of expelled placental parts (P =
0.001), placental expulsion duration (time between expulsion of first and last placental
part; P = 0.002) and time between expulsion of last piglet and last placental part (P =
0.024). On the other hand, use of oxytocin increased number of expelled placental parts
(LSmeans ± SD, 3.8 ± 0.2 vs. 2.9 ± 0.3, P = 0.035), decreased the placental expulsion
duration (172 ± 44 vs. 328 ± 26 min, P = 0.011) and time between expulsion of last piglet
and last placental part (148 ± 48 vs. 300 ± 24 min, P = 0.025). Furthermore, prolonged
farrowing (β ± SE, Wald χ2, Odds; 2.0 ± 0.5, 13.1, 7.6; P = 0.001), obstetrical intervention (1.5
± 0.7, 4.4, 4.3; P = 0.036) and two or more stillborn piglets (1.4 ± 0.7, 3.8, 3.9; P = 0.052)
increased the risk of having enlarged uterine size whereas oxytocin administration (- 1.5 ±
0.7, 4.7, 0.2, P = 0.040) decreased the risk. Two or more stillborn piglets (2.6 ± 0.9, 8.7, 13.7; P
= 0.003), obstetrical intervention (1.8 ± 0.8, 5.0, 6.0; P = 0.025), prolonged farrowing (1.7 ±
0.8, 4.3, 5.7; P = 0.039) and impaired placenta expulsion (3.3 ± 1.7, 4.0, 26.9; P = 0.044) were
associated with intrauterine fluid. After weaning, OT concentrations were higher in sows
with prolonged farrowing than in sows with shortened farrowing (LSmeans ± SD, 28.0 ± 7.7
vs. 20.6 ± 7.7, P < 0.01). OT concentrations correlated with diameters of the follicles
measured 5 d after weaning and when the follicles reached their maximum size after
weaning (r = 0.61, P < 0.01 and r = 0.57, P < 0.01, respectively).
The results show that sows with prolonged farrowing can show subsequent subfertility,
which may be linked with compromised uterine health. Sows with prolonged farrowing are
at higher risk of impaired placental expulsion, puerperal metritis and impaired uterine
involution. Using exogenous oxytocin after parturition may be useful in sows with
prolonged farrowing in order to lower the risk for retained placentae and subsequent
metritis. Obstetrical intervention should be careful and done under antiseptic conditions.
Proper parturition management should be used in order to prevent piglets being stillborn.
Against expectations, sows with prolonged farrowing duration had higher oxytocin
concentrations and larger follicles at subsequent estrus. This merits further investigation.
In addition, we adopted and tested a transvaginal ultrasound-guided biopsy method for
luteal tissue, which can be used to study corpus luteum function. No effects were
observed on the cyclicity and morphology of the reproductive organs of the sows: not on
corpus luteum function, pregnancy rate, gestation length or subsequent litter size. A
biopsy was obtained for 50% of all attempts. Sows from which at least one biopsy was
obtained were older (parity 5.0 ± 2.8 vs. 2.2 ± 0.4; mean ± SD), heavier (290 ± 26 vs. 244 ± 27
kg) and had more backfat (11.4 ± 2.7 vs. 6.4 ± 2.5 mm; P < 0.05 for all) compared with sows
from which no samples could be obtained.
12
13
Introduction and literature review
14
1. Introduction
There has been a steady increase in litter size and consequently farrowing duration in
modern pig production over recent decades (Rutherford et al., 2013; Edwards and Baxter,
2015; Tab 1). Both trends are probably causally linked, many authors having reported that
it takes a longer time to farrow larger litters than it does smaller litters (Fahmy and Friend,
1981; Van Dijk et al., 2005; van Rens and van der Lende, 2004). Concurrently, the annual
removal rate of gilts and sows has increased and an association with litter size has been
established (Andersson et al., 2016; Fig. 1); with the most common reason for removal
being conception failure (Lucia et al., 2000). Nevertheless, it has not been determined
whether farrowing duration is a causal factor in this association and would have a
negative impact on subsequent reproductive health and fertility.
Table 1. Data from Danish Pig Research Center (PRC) annual reports (Rutherford et al., 2013)
presenting the increase in litter size between 1996 and 2011.
15
Figure 1. Predicted probability (with 95% CI) of a sow producing ≥ 4 litters in her lifetime versus total
number of piglets born in her first parity litter (Andersson et al., 2016).
In other species, e.g. cow (Parkinson, 2009), mare (LeBlanc, 2008b; Paccamonti and
Pycock, 2009) and bitch (Fellows et al., 2012), prolonged and difficult parturition is known
to increase the risk of subsequent subfertility due to compromised uterine function
postpartum. Uterine function can be compromised because of inflammation (Gilbert et
al., 1998), infection (Savc et al., 2016b) or incomplete involution (Savc et al., 2016a; Mee
et al., 2009). The most severe impairment of uterine function is when postpartum metritis
occurs. Postpartum metritis is an acute infection of the uterus occurring during the first
week after parturition. Many predisposing factors have been identified and summarized
as factors at parturition which contribute to 1) prolonged dilatation of the cervix –
prolonged duration of parturition and delayed uterine involution; 2) bacterial invasion of
the uterus – dystocia and obstetrical intervention; 3) intra-uterine bacterial growth –
retention of fetuses or placentae; and 4) delayed uterine clearance – decreased uterine
contractions (Sheldon et al., 2006; Troedsson & Liu, 1990; Linde-Fosberg & Eneroth, 2005;
LeBland, 2008).
The majority of postpartum uterine infections begin with bacterial contamination of the
uterine lumen. In the case of a difficult parturition, trauma to the uterus and bacterial
contamination are increased and a uterine contamination can become a uterine
infection. The presence of pathogenic bacteria in the uterus causes further inflammation
and lesions of the endometrium and delays uterine involution. Subsequently, as in cattle,
uterine bacterial infection, bacterial products and/or the associated inflammation
suppress postpartum pituitary LH secretion and therefore perturb postpartum ovarian
follicular growth and ovulation (Peter et al., 1988, 1989; Sheldon et al., 2002; Opsomer et
al., 2000).
16
Savc (2016b) used vaginal mucus scoring and ultrasonographic assessment of uterine
content at first pre-breeding examination after parturition as predictors of future
reproductive performance in dairy cattle. Conception rate to first service, breeding
window, and positive pregnancy status were significantly worse for cows with a uterine
score of 3 to 4 (≥ 0.2 cm of mixed echogenicity fluid with hyperechoic particles in the
lumen without infolding of the endometrium) and a positive vaginal mucus score (pus)
compared with cows with a uterine score of 1 to 2 (small volume ≤ 0.2 cm or moderate
volume ≥ 0.2 cm ≤ 0.5 cm of anechoic fluid in the lumen with infolding endometrium) and
a negative vaginal mucus score (no or clear mucus, or mucus with specks of purulent
material) (Fig. 2).
Figure 2. Kaplan–Meier survival curve for all 493 cows which shows days after earliest service date
(ESD) until the cow got pregnant of all animals with the different uterine scores (UTSs). UTS scores of
two and higher were defined as uteri with abnormal content and these animals had a decreased
likelihood of getting in calf. Pregnancy rate hazard for cows with UTS0, 1, 2, 3, and 4 were 1, 0.84,
0.53, 0.78, 0.28, respectively (P < 0.001; Šavc et al., 2016b).
Thus, the general aim of this thesis was to explore whether prolonged parturition decreases
subsequent pregnancy rates and whether similar associations between parturition and
puerperal disorders such as placental retention, metritis, and delayed uterine involution, as
describes above for the cow and other species, exist also in sows. We speculated that
modern hyperprolific sows with increased farrowing duration have higher risks of
experiencing puerperal disorders as well as impaired resumption of ovarian cyclical
activity and fertility after subsequent weaning.
17
In the first parts of this literature review, normal parturition, normal puerperium and normal
return of ovarian cyclical activity after weaning will be described. An outline is presented
in Figure 3. In the last part of this literature review, the focus will be about diseases of the
puerperium and possible associations with previous parturition and subsequent ovarian
cyclical activity
Figure 3. Overview of reproductive events between two subsequent pregnancies and their normal
durations
18
2. Parturition
Sufficient corpus luteum function is important for maintenance of pregnancy. The primary
function of the corpus luteum is to release at sufficient concentrations progesterone.
Progesterone is released in pulses which is triggered by pulsatile LH release from the
pituitary gland (Ellendorf et al., 1979). Progesterone concentrations become independent
of LH and decrease about one to two days before parturition, following activation of the
pituitary adrenal axis of the fetuses and subsequent increase in cortisol levels in the fetal
plasma (Randall, 1983). These events are mainly mediated by intrauterine production of
prostaglandin F2α (Whittle et al., 2000). The increased cortisol levels in the fetal plasma
enhance the expression of prostaglandin synthetase within the fetal trophoblast cells of
the placenta. As a result, the concentration of prostaglandin E2 increases, which
stimulates the placental conversion of pregnenolone to estrogen in the placenta. The
change in the steroid balance (decrease of progesterone and increase in estrogen)
enhances the expression of prostaglandin synthetase within the endometrium, leading to
increased prostaglandin F2α within the uterine wall (Fig. 4). An overview of the hormonal
changes around parturition is presented in Figure 5.
Figure 4. Endocrine events associated with onset of parturition and increased uterine contractility
(Taverne and Noakes, 2009).
19
Figure 5. Trends in hormone concentrations in the peripheral circulation of the sow around the time
of parturition (Taverne and Noakes, 2009). Farrowing at day 0
Another result of the change in steroid balance is production of oxytocin and oxytocin
receptors. Oxytocin, prostaglandins and estrogens have local positive effects on
myometrial contractility and dilatation of the cervix, and general effects on nest-building
behavior. Within 24 h before the birth of the first piglet, sows isolate themselves from the
rest of the group and have an innate need to build a nest where the litter is to be born
(Jensen,1986; Jensen & Redbo, 1987). This nest-building behavior is characterized by
collecting branches, leaves or grass and by rooting and pawing activity (Jensen & Redbo,
1987). The behavior is expressed at highest activity 3–8 h before the birth of the first piglet
and gradually ceases prior to the birth of the first piglet (Hartsock & Barczewski, 1997;
Westin, 2014).
The first phase of farrowing starts about 12 h before expulsion of the first piglet and is
marked by cervical dilatation and increased myometrical activity. The second phase of
farrowing involves abdominal straining of the sow in lateral recumbency and expulsion of
the piglets. This stage can last for 4 to 6 h, with piglets born at 20 min intervals on average.
During the third stage of farrowing the placentae are expelled. This expulsion may already
start before the end of the second stage and takes from 20 min up to 12 h in sows, but is
usually completed within 4 h (Jones, 1966). A study in gilts reported that placenta
expulsion may take from 15 min up to 8 h, but is, on average, completed within 2.5 h (van
Rens and van der Lende, 2004).
20
Studies on factors affecting placental expulsion are scarce. To date, only the study in gilts
investigated factors affecting onset, defined as time of expulsion of the first placental part
relative to the last piglet, and duration of placenta expulsion duration, defined as time
duration between expulsion of first and last placental part (van Rens and van der Lende,
2004). A placental part was defined as an individual placenta or two or more intergrown
placentae. The number of expelled placental parts was smaller than the number of
piglets, with a curvilinear relationship between relative placental parts and the number of
piglets (Fig. 6). Farrowing duration was the only factor affecting the onset and duration of
placenta expulsion.  An increase in farrowing duration was significantly related to an
increase in duration of placental expulsion (Fig. 7a) and to a decrease in the onset of
placenta expulsion (Fig. 7b). On average, the onset of placenta expulsion occurred
simultaneously with the expulsion of the last piglet, but could occur as early as 3 h before
and as late as 2 h after the last piglet. In 40% of the gilts placenta expulsion started before,
in 11% of the gilts simultaneously and in 49% after the birth of the last piglet. Litter size,
gestation length, percentage of males in the litter, average and total litter weight,
bodyweight and genotype of the sow had no effect on placental expulsion. Moreover,
reports on incidence rates and causes of retained placentas or remnants of the placentas
are also scarce. It is considered rare and a retained fetus as the only cause (Jones, 1966).
One possible explanation for the association between farrowing and placental expulsion
duration is oxytocin. Oxytocin is one of the most important modulators for uterine
contractions. Therefore, sufficient concentrations during parturition are needed for rapid
and successful expulsion of piglets and placentae. The baseline concentration of oxytocin
and the frequency of its pulsatile release start to increase about 6 h before parturition,
peak 2-4 h after start of parturition and decrease thereafter (Castrén et al., 1993; Oliviero
et al., 2008). Furthermore, there are pulsatile releases between births of the piglets (Oliviero
et al., 2008). The farrowing duration was shown to correlate negatively with peripheral
oxytocin levels. Sows with long farrowings had lower average basal and peak levels as
well as post-expulsion pulse concentration of oxytocin during farrowing compared with
sows with short farrowings (Castrén et al., 1993; Oliviero et al., 2008).
In domesticated species other than the pig, dystocia and subsequent reduction in uterine
contractions are known to cause retained placentae. Retained placentae is of major
importance in the etiology of postpartum metritis, which is considered to be one of the
most significant causes of subsequent infertility (LeBlanc, 2008b; Parksinson, 2009; Fellows et
al., 2012). In the pig, studies on consequences of retained placentae are lacking.
Therefore, the second aim of the study was to investigate whether an increase in
farrowing duration is related to an increase in duration of placental expulsion also in sows
as observed in gilts (van Rens and van der Lende, 2004). Furthermore, we speculated that
in sows with prolonged farrowing placental expulsion can be impaired and therefore
retained placentae occurs.
21
Figure 6. Relation of number of placental parts found per gilt (expressed as percentage of total
number of piglets born (TNB) in relation to TNB; van Rens and van der Lende, 2004).
Figure 7. Relation of placenta expulsion duration (a) and time interval between expulsion of last
piglet and first placental part (b) to farrowing duration (X-axes log scale; Y-axis in (a) log scale; van
Rens and van der Lende, 2004).
22
3. Puerperium
The puerperium is that period after the completion of parturition, often including the third
stage of parturition, when the genital tract is returning to its normal non-pregnant state
(Noakes, 2009). It is important that the puerperium proceeds normally, since it is practice
under most conventional production systems to wean piglets and breed sows fairly soon
after puerperium is completed (Fig. 4). Therefore, any extension of the puerperium may
have negative effects on subsequent fertility as it has, e.g., in cattle (Šavc et al., 2016b).
The most important events of the puerperium are elimination of bacterial contamination,
uterine involution and restoration of the endometrium, and resumption of ovarian function
(Noakes, 2009; Fig. 4). Initial uterine involution is rapid during the first week postpartum
when the uterine weight is reduced by 65% (Palmer et al., 1965). After that, uterine
involution is continuous and progresses uniformly, and is completed during the third week
postpartum (Palmer et al., 1965; Kudlac and Groch, 1979; Busch, 2007; Fig. 8 and 9).
Compared with the uterus, involution of the cervix is very rapid and completed within one
week (Busch, 2007; Kudlac and Groch, 1979). As a result, resorption of uterine content
needs to occur. During the initial period, uterine weight loss is mainly due to atrophy and
resorption of intraluminal fluid, whereas after the initial period, uterine weight loss is mainly
due to changes in the myometrium, notably a reduction in cell numbers, cell size and
amounts of connective tissue (reorganization) (Busch, 2007).
Figure 8. Uterine weight loss after parturition (Kudlác and Groch, 1979)
23
Figure 9. Uterine weight at various intervals after farrowing and weaning (Palmer et al., 1965)
At the end of uterine involution, during the third week postpartum, many authors reported
an increase in the uterine weight of about 5% (Palmer et al., 1965; Kudlac and Groch,
1979; Busch, 2007). The same observation was made regarding the length of the uterine
horns (Busch, 2007). Busch (2007) interpreted the increase in length and weight of the
uterus during week three postpartum as proliferation due to increased estrogen
production by growing follicles and as a sign of completed uterine involution. Follicular
development during early lactation is suppressed and characterized by a large
population of small follicles (<  3 mm) and a small population of medium-sized follicles (3-4
mm) (Britt, 1985). During the second week postpartum the number of small follicles
increases and during the third week the overall number of follicles and the number of
medium-sized follicles increases (Busch, 2007; Fig. 10). During this time also the weight of
the ovaries increases concomitantly with the increase in uterine weight and length (Busch,
2007), which is due to an increase in the number of follicles and an increase of the
diameter of the follicular pool (Lucy et al., 2001; Busch et al., 2007).
24
Figure 10. Number of small (< 3mm; A) and large (>3mm; B) follicles at the ovary during lactation
(Busch, 2007).
This shift in the follicular pool occurs not only alongside completion of uterine involution but
also alongside restoration of LH-pulsatility. After parturition, the ovarian axis is inhibited.
Weaning sows immediately after parturition, or at less than 3 weeks postpartum, would
result in a higher incidence of anestrus and ovulation failure (Soede et al., 2011). Sows that
fail to ovulate either show follicular regression with a short/intermittent estrus or develop
ovarian cysts. If sows are weaned at more than 3 weeks, WEI seems to be normal and
about 98% of sows show estrus, and 98% of which ovulate (Knox and Rodriguez Zas, 2001).
25
Anestrus and cystic ovaries in sows weaned at less than 3 weeks can be due to the lack of
a preovulatory. LH-surge. In cyclic gilts, the occurrence of cystic follicles has been related
to a lack of the preovulatory LH-surge (Reviewed by Ryan and Reaside, 1991). The lack of
the LH-surge is believed to be due to high levels of corticosteroids (Ryan and Raeside,
1991) or an impaired responsiveness to a positive feedback of estradiol-17β produced by
the follicles at the hypothalamic-pituitary level (Soede et al., 2011). The impaired
responsiveness may be related to depleted LH stores after farrowing that are progressively
restored during lactation (Bevers et al., 1981). On the other hand, mean concentrations of
circulating LH are high during the first two to three days after parturition and only
decrease thereafter (Tokach et al., 1992; De Rensis et al., 1993; Fig. 11). These
concentrations and the number of LH pulses remain low during day four to fourteen and
gradually increase subsequently (Shaw and Foxcroft, 1985; De Rensis et al., 1993; Fig 11). In
addition to this restoration of LH-pulsatility, a pre-ovulatory LH surge will not occur before
weaning in most of the sows because of suckling and piglet proximity (De Rensis et al.,
1993).
Figure 11. Schematic pattern of LH secretion during lactation (Quesnel & Prunier, 1995).
If and how uterine involution affects resumption of ovarian function and restoration of LH
concentration and pulsatility is not known. The restoration of LH-pulsatility and shift in the
follicular pool occur around the same time as when uterine involution ends; during the
third week postpartum. Therefore, uterine involution could also play a role. On the other
hand, no data are available on the interaction between uterine involution and ovarian
activity in pigs (Quesnel & Prunier, 1995).
Furthermore, studies on factors affecting uterine involution are lacking. We speculated
that uterine clearance after prolonged parturition may be impaired: it was shown that
prolonged parturition is associated with reduced oxytocin concentration (Castren et al.,
1993; Oliviero et al., 2008), a hormone considered to be important in uterine contractions
and therefore clearance. Thus, it is plausible that sows with prolonged parturition may
have more uterine content to absorb because of reduced uterine clearance postpartum
and, therefore, this may have a negative effect on the duration of uterine involution.
26
4. Return of cyclical activity
After weaning, pulsatile LH release shifts from a lesser frequency of a greater amplitude
pattern to a greater frequency of a lesser amplitude pattern (Fig. 12). This shift is important
for selection of antral follicles and subsequent follicular development and ovulation (Kemp
et al., 1998; Knox et al., 2005; Soede et al., 2011). Once LH has stimulated follicular
development, these follicles start to produce 17β-oestradiol, which triggers the LH surge
(Fig. 13), causing ovulation of the follicles. Ovulation takes place at 30 ± 3h (mean ± SD)
after the peak of the LH surge (Soede et al., 1994).
Few factors affecting the LH surge have been described. For instance, Turner and Tilbrook
(2005) concluded that prolonged increase in cortisol levels (sustained stress) can result in a
delayed LH surge or even prohibit it through reduction in GnRH output. Furthermore,
lactation weight loss and feed intake during lactation have a major effect on follicular
growth and LH-release after weaning as reviewed by Soede and Kemp (2015). Sows in
bad metabolic state due to high milk production or low feed intake have a high degree
of inhibition of LH secretion before weaning and therefore decreased LH concentration
and pulsatility after weaning (reviewed by Quesnel and Prunier, 1995).
Also lactation length is affecting the return of cyclical activity after weaning (Soede and
Kemp, 2015). A short lactation length of less than 21 days can cause anestrus, ovulation
failure or follicular cysts, as discussed in the last chapter. On the other hand, Kirkwood et
al. (1984) showed that sows weaned at day 35 of lactation had a higher pre-ovulatory LH
surge than sows weaned at 10 d of lactation.
Figure 12. Pulsatile LH-patterns on the day before weaning (A) and the day of weaning (B) in a
primiparous sow. Weaning took place at day 24 of lactation (Soede et al., 2011).
27
Figure 13. Illustration of hormonal profiles of LH (ng/mL), FSH (ng/mL), estradiol (pg/mL),
progesterone (ng/mL), and GnRH (pg) during proestrus in relation to the timing of estrus and
ovulation in the sow (Knox et al., 2015).
At estrus, increased myometrial activity is needed for sperm transport from the site of
deposition, the cervix and caudal part of the uterine body, towards the sperm reservoir,
located at the utero-tubal junction and the caudal part of the oviduct. Myometrial
contractions start to increase in frequency and amplitude from 15 to 22/h and from 20 to
40 mmHg at the end of the proestrus, reach their maximum during estrus, and decrease
thereafter (Langendijk et al., 2002). Myometrial activity is, for example, in landrace sows,
constant throughout the estrus (Hoang-Vu, 1987), but sow-dependent variation has been
demonstrated and is positively correlated with tissue and plasma concentrations of
certain hormones, such as estradiol and oxytocin, and sensory-related stimulus, including
presence of the boar (Langendijk et al., 2005). The presence of a boar increases oxytocin
release as well as uterine activity, with a causal link between both (Langendijk et al., 2002
and 2003, Madej et al., 2005). Langendijk et al. (2003) assessed the effects of exogenous
oxytocin on uterine activity and compared three different sexual stimuli in their effects on
endogenous oxytocin release and uterine activity. Exogenous oxytocin increased
frequency, amplitude or both of uterine contractions. Overall, frequency of contractions
increased by 5/h and amplitude of contractions by 32 ± 5.8 mmHg after administration of
exogenous oxytocin. Furthermore, boar presence increased the frequency of contractions
by 3/h and the amplitude of contractions by 10 mmHg, especially when the frequency of
spontaneous contractions before boar introduction was below 26/h (Fig. 14).
28
Figure 14. Example trace of an intrauterine pressure recording in one sow. The trace shows 25 min of
spontaneous uterine activity recording, before introduction of a boar at t= 51 min. The points below
the x-axis indicate individual contractions (Langendijk et al., 2003).
This increase in uterine contractility was related to the release of endogenous oxytocin.
Boar introduction in combination with application of manual pressure on the back of the
sow caused a clear, immediate, surge-like release of oxytocin, with mean peak
concentrations of 80 pg/ml (Fig. 15). Oxytocin concentrations rose above basal level
within 1 min and reached maximum concentrations between 1 and 3 min after
introduction of the boar. Oxytocin levels returned to basal levels 20-30 min later. This
observation was confirmed by the study of Madej et al. (2005), who showed that boar
presence and high human stimulation (reverse riding test and manipulation of costae-,
scapulae- and inguinal area) increased oxytocin within 10 min after initiation of stimulation
compared with only application of manual pressure on the back of the sow. Furthermore,
Langendijk et al. (2002) showed that also the intrauterine infusion of estrogens increased
frequency of uterine contractions.
29
Figure 15. Mean (± S.E.) plasma oxytocin relative to the start of an application of three different
boar stimuli (t = 0 min): BPT: back pressure test (manual stimulation of the back and the flanks of the
sow), BPT + spray (back pressure test in combination with boar pheromone spray), BPT + boar (back
pressure test in combination with presence of a boar; Langendijk et al., 2003).
Therefore, speculated that sows with prolonged farrowing have perturbed follicular
growth after subsequent weaning and oxytocin release during boar stimulation at
subsequent estrus. As discussed, oxytocin release and subsequent uterine contractions are
not only needed at parturition but also during breeding at estrus when they are important
for ovulation and sperm transportation towards the sperm reservoirs. Furthermore, oxytocin
release during the weaning-to-estrus interval is important for follicular growth. We
speculated that if prolonged farrowing can be associated with reduced oxytocin
concentrations then also with impaired follicular growth after subsequent weaning and
reduced oxytocin release during boar stimulation at subsequent estrus.
30
5. Puerperal diseases
One of the most important puerperal diseases in sows is postpartum dysgalactia syndrome
(PDS), recently reviewed by Maes et al. (2010). PDS is a multifactorial disease and
characterized by insufficient colostrum and milk production during the first days after
farrowing. Martineau et al. (1992) summarized a list of symptoms for the herd, piglets and
sow. Local symptoms for the sow can include mastitis and/or vaginal discharge due to
metritis and general symptoms can include fever and/or anorexia.
PDS is mediated by different pathways, including endotoxin production by bacteria in the
gut, mammary gland and/or uterus (Martineau et al., 1992). In a study by Bostedt (1998),
24.4% of gilts with feverish puerperal illness exclusively had signs of mastitis, in 29.5% there
was a combination of mastitis and inflammatory affection of the genital tract and in 46.1%
a solitary infection of the reproductive tract. E. coli, Staphylococcus spp. and
Streptococcus spp. were the most predominant bacteria isolated from the genital tract.
The current hypothesis is that interactions between endotoxins produced by Gram-
negative bacteria in the gut and alterations of the immune and endocrine functions play
a central role in the development of PDS (Maes et al., 2010). In dairy cows, it was shown
that those cows suffering significant immunosuppression around parturition were both
predisposed to uterine disease and less able to clear the infections on their own (LeBlanc,
2008a). The impaired immune system also puts the animal at increased risk of developing
retained fetal membranes and suppression of appetite, all with repercussions on
subsequent health and reproductive performance (Dobson et al., 2008).
There are studies on sows showing that prolonged parturition can be linked with a delayed
decline in progesterone (Oliviero et al., 2008). Progesterone is known to impair not only
uterine contractility but also the immune response. Furthermore, a study by Tummaruk et
al. (2013) showed that farrowing duration can affect the incidence of postparturient fever
and reduced appetite. The percentage of sows with fever during the first 24 h postpartum
increased from 40 to 100% when the farrowing duration increased from <2 to >4 h (Fig. 16).
On day 3 postpartum, the percentage of sows with reduced appetite and a farrowing
duration of 4-8 h was higher than for those with a farrowing duration of 2-3 h (Fig. 17). A
study by Bostedt et al. (1998) examined gilts suffering from feverish puerperal illness. Also in
this study, the farrowing duration had a significant effect. 85.9% of gilts with consequent
puerperal illness had farrowing durations of longer than 6 h, whereas 78.8% of control gilts
of the same age finished parturition in <3 h. Other significant factors affecting the
incidence of feverish puerperal illness were frequency of obstetrical intervention and
stillbirth rate. The frequency of obstetrical intervention measured in the group of patients
was 27% and only 9.5% in the group of control sows. In addition to puerperal fever, 75% of
the gilts were anorexic and 66% had abnormal fecal consistency as general symptoms.
31
Figure 16. Percentage of sows with fever (rectal temperature ≥39.0 °C) on the day of parturition in
relation to farrowing duration. Different letters (a-c) differ significantly (Tummaruk et al., 2013).
Figure 17 4. Percentage of reduced appetite (daily food intake relative to provided food) sows at
Days 1, 2, and 3 postpartum by farrowing duration (Tummaruk et al., 2013).
32
Furthermore, there is evidence that prolonged parturition and PDS are associated with
similar risk factors. Risk factors for PDS were reviewed by Maes et al. (2010), and are
constipation, ad libitum feeding, high body condition, restricted movement, high ambient
temperature and heat stress, rapid changes in the housing only a few days before
parturition, farrowing induction and no farrowing supervision. Risk factors for prolonged
farrowing were reviewed by Peltoniemi and Oliviero (2015) and include high body
condition, restricted movement, and constipation.
Subsequently, PDS has been associated with reduced subsequent fertility (Hoy, 2006).
Sows with PDS had a higher rate of anestrus and late onset of estrus, a higher return and
mortality rate, more abortions and smaller litters at subsequent farrowing (Hoy, 2006). An
overview of the current knowledge about associations between prolonged farrowing,
puerperal disease and subsequent fertility is presented in Figure 18.
33
Figure 18. Associations between prolonged farrowing, puerperal disease and subsequent fertility
34
6. Corpus luteum function and biopsy of luteal tissue
The primary function of the corpus luteum is the production and secretion of
progesterone. Therefore, corpus luteum function is important for establishment and
maintenance of the pregnancy. The amount of progesterone is related to total luteal
mass (r = 0.26-0.45; Athorn et al., 2012) and further depends on the blood flow and the
capacity of the luteal tissue to synthesize progesterone (Niswender et al., 2000). This
capacity depends on several hormones, such as LH (Peltoniemi et al., 1995; Tast et al.,
2000; Virolainen et al., 2003), growth factors (Miller et al., 2003), enzymes (Bao and
Garverick, 1998), and pathways (Niswender et al., 2000), as well as management and
environmental factors, such as pre-mating nutrition (Wientjes et al., 2011), post-mating
nutrition (Athorn et al., 2012), season (Langendijk and Peltoniemi et al., 2013), and
lactation weight loss (Hoving et al., 2012).
The maximum size of the corpora lutea are established at day 8-9 of the pregnancy and
maintained autonomous until day 10-12 (Fig. 19). Meduri et al. (1996) showed a decrease
in density of LH receptors in luteal cells, which increased again 6 d after ovulation. This
finding is indicative of an LH-independent autonomous time period during early corpus
luteum development. After that, maintenance of corpus luteum function seems to be
dependent on LH (Fig. 19; Peltoniemi et al., 1995; Tast et al., 2000; Virolainen et al., 2004).
Chronic treatment with a GnRH agonist from days 14 to 21 of pregnancy eliminated LH
secretion and resulted in a decrease in progesterone secretion and loss of pregnancy
(Peltoniemi et al., 1995).
Figure 19. Conceptual diagram illustrating the dynamics in luteal mass (cross sectional area of
corpora lutea, closed symbols) and systemic progesterone (open symbols) during LH independent
and LH dependent stage of the luteal phase in pregnant sows (Langendijk, 2015)
35
The LH release pattern during the luteal phase is similar to the release pattern of a
lactating sow, thus characterized by a lesser frequency of greater amplitude LH pulses.
Furthermore, it seems that also progesterone release by corpora lutea occurs in a pulsatile
pattern related to preceding LH pulses (Fig. 20).
Figure 20. Plasma progesterone profiles from the caudal vena cava (P_VC) and the jugular vein
(P_VJ) and luteinizing hormone (LH) profile during a 12 h bleeding window at 30 min intervals on
day 22 of pregnancy for two gilts (Virolainen et al., 2005).
As discussed earlier, in other species like the cow, difficult calving can subsequently
interrupt LH release and therefore negatively interfere with ovulation and establishment of
pregnancy. Considering that LH seems to play an important role in maintaining luteal
function in sows, it should be possible to study corpus luteum function in sows with
prolonged parturition. Besides blood sampling for LH, luteal tissue collection for gene
expression analysis of LH receptors, for example, may be an interesting approach to study
corpus luteum function. In order to do so, it would be useful to have a practical and
minimally invasive method available that allows repeated sampling of luteal tissue over a
period of time. Until now, luteal tissue could be sampled only once, for example after
euthanasia (Conley and Ford, 1989), or with labor-intensive and invasive methods, such as
ovariectomy under general anesthesia (Forni et al., 2003; Ribeiro et al., 2007). In other
species, for instance cows (Kot et al., 1999) and horses (Beg et al., 2005), an alternative
method, termed transvaginal ultrasound-guided biopsy, is used. This method requires only
local anesthesia (Tsai et al., 2001), if any, and can be performed by one to two persons
within a few minutes (Beg et al., 2005).
Therefore, a secondary aim of this study was to develop and test a similar ultrasound-
guided biopsy method as used in other species such as cows and horses. Furthermore, we
aimed to investigate whether there are adverse effects of this method on subsequent
ovarian function and reproductive performance. We also aimed to investigate if sufficient
amounts and quality of luteal tissue can be obtained for subsequent genetic studies.
36
37
Aims and hypotheses of the study
38
I
The first aim of the study was to investigate the correlation between farrowing duration
and subsequent pregnancy rate in sows breed after a lactation of about four weeks and
a weaning-to-estrus interval of less than seven days (TRIAL-I).
We hypothesized that sow with a prolonged parturition have a lower pregnancy rate
compared with sows with a normal parturition.
II
The second aim of the study was to investigate the correlation between farrowing
duration and placental expulsion duration as well as retention of placentae (TRIAL-II).
We hypothesized that prolonged parturition prolongs placenta expulsion duration
compared with normal parturition. Furthermore, we hypothesized that sows with severely
prolonged parturition experience retained placentae.
III
The third aim of the study was to investigate the correlation between farrowing duration
and the uterine size (indicator for uterine involution) and presence of intrauterine fluid
(indicator for metritis) during the first week postpartum (TRIAL-III).
We hypothesized that prolonged parturition causes an increase in intrauterine fluid and
uterine size as signs of an increased risk for metritis and delayed uterine involution
compared with normal parturition
IV
The fourth aim of the study was to investigate whether sows with prolonged farrowing
have perturbed follicular growth after subsequent weaning and oxytocin release during
boar stimulation at subsequent estrus (TRIAL-IV).
We hypothesized that sows with prolonged parturition have smaller follicles at weaning
and ovulation compared with sows with normal parturition. Furthermore, we hypothesized
that these sows release less oxytocin at estrus during boar stimulation compared with sows
with normal parturition.
39
V
The fifth aim of the study was to adopt and validate a similar ultrasound-guided biopsy
method for pigs as used for other species, such as cows and horses. Furthermore, the aim
was to investigate whether adequate amounts of luteal tissue for genetic studies can be
obtained without adverse effects on ovarian function and reproductive performance of
the sows.
Figure 21. Overview of the aims, hypothesis, and the outline of the studies
40
41
Materials and methods
42
1. Animals, housing, and management
The trials were conducted on two different pig herds (HERD-I and HERD-II) in southern
Finland. HERD-I included a sow pool farm with 280 Yorkshire x Large White crossbred sows
and four satellite farms (Peltoniemi et al., 2009). HERD-II included a fully integrated farm
with about 400 Yorkshire x Large White crossbred sows. TRIAL-I took place in the sow pool
farm and two of the satellite farms (SATELLITE-I and SATELLITE-II) in HERD-I. TRIAL-II and TRIAL-
III took place in SATELLITE-I of HERD-I and TRIAL-IV in the sow pool farm and SATELLITE-I of
HERD-I and HERD-II. TRIAL-V was conducted in the sow pool farm of HERD-I.
In the sow pool farm of HERD-I, sows were housed during gestation in groups of twenty in
pens (20 x 5m) with straw bedding. Sows had ad libitum access to water from a nipple
drinker and were fed a standard pregnancy diet (Emakko Pekoni, Suomen Rehu, Finland)
two times per day via an automatic liquid feeding system. Sows were transported from the
sow pool farm to SATELLITE-I or SATELLITE-II about three weeks before the expected date of
farrowing where they were group-housed in the waiting unit for about two weeks. About
one week before the expected date of farrowing, sows were moved from the waiting unit
to the farrowing and lactation unit. This unit was cleaned and disinfected before the sows
were moved in and consisted of farrowing pens with adjustable farrowing crates on semi-
slatted floors. Sows in SATELLITE-I had no bedding material whereas sows in SATELLITE-II had
sawdust bedding. Feed consisted of a commercial diet for lactation (Emakkotäysrehu,
RehuxOy, Finland) and was served three times a day via an automatic dry feeding system.
After about four weeks of lactation, the sows were returned to the sow pool farm.
In HERD-II, approximately 400 crossbred Yorkshire × Large White crossbred sows were
housed during gestation in groups of 8 to 10 (8 × 5 m) with peat bedding where they had
ad libitum access to water from a nipple drinker and were fed a standard pregnancy diet
(Emakko Täysreu, Rehux Oy, Finland) two times a day via an automatic liquid feeding
system. Sows were moved from the gestation unit into the farrowing and lactation unit
about one week before expected date of farrowing where they received a commercial
lactation diet (Emakko Pekoni Plus, Suomen Rehu, Finland) as dry feed three times a day.
This unit was cleaned and disinfected before the sows were moved in and consisted of
farrowing pens with adjustable farrowing crates on fully-slatted plastic floors. After about
four weeks of lactation, the sows were moved to the breeding unit.
Farrowings were not induced in any of the sows included in the study. In TRIAL-II and TRIAL-
III, farm workers supervised the farrowings and injected 0.5 – 1.0 ml oxytocin (Vetox Vet 10
IU/ml, Vetcare Oy, Finland or Partoxin Vet 10 IU/ml, Pharmaxim, Finland) during the
parturition if indicated. An indication was given if: 1) the birth canal was open and
unobstructed; 2) a fetus was not present or present and well positioned cranially in the
uterine cervix; 3) the sow gave an expulsion effort (contractions of abdominal
musculature) but no piglet was delivered over a period of at least 30 min. These criteria
were based on the recommendation of Gilbert (1999).
43
2. Sample size and study design
In TRIAL-I, parturitions of 148 randomly assigned multiparous sows were recorded in
SATELLITE-I (n = 65) and SATELLITE-II (n = 83) of HERD-I during eighteen consecutive months
in eleven batches. Sows were excluded from TRIAL-I if they experienced obstetrical
intervention or administration of exogenous oxytocin during parturition or did not come to
heat within 7 d after weaning. After weaning, sows were brought in the sow pool farm to a
corridor in groups of five for heat detection twice a day, starting from second day after
weaning until standing heat. In the corridor, sows had nose-to-nose contact with boars.
After standing heat was detected, the sows were separated from the others by a fence,
further stimulated (five-point stimulation) and artificially inseminated with semen of 10
Hampshire x Duroc crossbred boars (3 x 109 spermatozoa in each extended AI dose).
Artificial insemination was repeated about 16-24 h later. At four and seven weeks after
insemination, a pregnancy examination was performed (transabdominal ultrasound;
Tringa 50S; Esaote).
In TRIAL-II, parturitions of 142 randomly assigned multiparous sows were recorded in
SATELLITE-I of HERD-I during twelve consecutive months in ten batches. Sows were
excluded from TRIAL-II if they experienced obstetrical intervention. On the other hand,
administration of endogenous oxytocin was allowed and recorded. Two to three days
after parturition, an ultrasound examination of the uterus was performed once per sow
and the uterus was inspected for signs of retained piglets and placentae.
In TRIAL-III, parturitions of 99 randomly assigned multiparous sows were recorded in
SATELLITE-I of HERD-I during twelve consecutive months in eight batches. Obstetrical
intervention and oxytocin administration were allowed and therefore not considered to
be exclusion criteria. During the first week after parturition, an ultrasound examination of
the uterus was performed once per sow and uterine size and fluid accumulation was
determined.
In TRIAL-IV, parturitions of 19 randomly assigned multiparous sows were recorded in HERD-II
during three consecutive months in four batches. Furthermore, parturitions of another 11
randomly assigned multiparous sows were recorded in SATELLITE-I of HERD-I during three
consecutive months in three batches. Sows were excluded from TRIAL-IV if they
experienced obstetrical intervention or administration of exogenous oxytocin during
parturition. After weaning, the subsequent follicular development was monitored twice a
day between the third day after weaning and ovulation for all sows in either the sow pool
farm of HERD-I or in HERD-II. Furthermore, sows were brought to a corridor in groups of five
for heat detection twice a day starting from third day after weaning until standing heat
was established. The day after that, blood samples were collected in the presence of a
boar in order to determine the endogenous oxytocin concentrations and release patterns
(Method described in detail in 4.7).
44
TRIAL-V was conducted in the sow pool farm of HERD-1 and consisted of two sub-trials
(TRIAL-Va and TRIAL-Vb). In TRIAL-Va, four randomly assigned cycling multiparous sows
were chosen and their day of ovulation was determined with the aid of transrectal
ultrasonography. After that, a transvaginal ultrasound-guided biopsy was performed nine
and fifteen days after ovulation for three consecutive estrous cycles. The obtained
samples were submitted to the laboratory for histological staining and evaluation. Before
each biopsy and shortly before each ovulation, a transrectal ultrasound examination of
both ovaries was performed to determine the average size of the corpora lutea and
follicles. In TRIAL-Vb, two groups of randomly assigned multiparous sows were chosen, their
day of ovulation determined, and artificially inseminated twice with the semen of the
same boar. One group served as control group and the other group was subjected to two
TUBs. The first was conducted 10 d and the second 13 d after ovulation. The obtained
samples were submitted to the laboratory for RNA extraction and determination of the
quality and quantity of the extracted RNA. Before each second TUB, a transrectal
ultrasound examination of both ovaries was performed to determine the average size of
the corpora lutea. At day 28 after ovulation, a pregnancy examination was performed
with the aid of transabdominal ultrasonography (MyLab OneVet, Esaote). At farrowing,
gestation length and litter size were determined.
An overview of the animals used in the trials is presented in Table 2.
Table 2. Overview of the animals used in the trials
45
3. Farrowing observations
In TRIAL-I, sows were monitored with a digital video recorder (AVC 787; Avtech)
connected to infrared cameras (TS-6030PSC, Actech). The cameras were placed on the
ceiling above each pen about 2 d before expected farrowing. The recordings were
viewed afterwards and the farrowing duration (defined as the time interval between the
expulsion of the first and the last piglet) was determined.
In TRIAL-II, TRIAL-III, and TRIAL-IV, sows were monitored with WLAN IP cameras
(NICECAM420WL, Niceview Corp.) and IP-camera software (Blue Iris v.2.64, Perspective
Software Corp.). The cameras were placed behind and 1.5 m above the sow about 2 d
before expected farrowing (Fig. 22). The videos were viewed afterwards on a stationary
computer with a media player (Windows media player 12.0, Microsoft Corp.), starting from
the expulsion of the first piglet until 24 h after expulsion of the last piglet. In TRIAL-IV, the
farrowing duration was determined. In TRIAL-II and TRIAL-III, the farrowing duration and the
following placenta expulsion traits were determined: number of expelled placental parts
(number of expelled placental parts from first to last expulsion), placenta expulsion
duration (defined as the time interval between the expulsions of the first and the last
placental parts), first placental part expulsion (time of expulsion of the first placental part
relative to the last piglet) and last placental part expulsion (time of expulsion of the last
placental part relative to the last piglet). Placental part was defined as ‘placental mass’
that was observed to be expelled at once. A ‘placental mass’ consists of an individual
placenta or two or more placentae. Furthermore, it was recorded whether the sow had
total retained placentae (no expelled placental parts during the observation period) or
partially retained placentae (expulsion of last placental part observed before the
expulsion of the last piglet) and whether oxytocin was administered during parturition.
Figure 22. Position of the cameras during parturition.
46
4.  Sonographic evaluation of the uterus
The uterus was examined in TRIAL-II and TRIAL-III with a 4.3 MHz convex array ultrasound
probe (MyLab OneVet, SC3421 Vet, Esaote) through the right flank of the standing sow.
In TRIAL-II the examination was performed 2-3 d after parturition once for each sow. The
uterus was inspected for signs of retained piglets and placentae. Retained placentae
were defined as hyperechoic structures within the lumen of the uterus and/or the
endometrium.
In TRIAL-III, the examination was performed between the second and seventh day after
farrowing once in each sow. The uterus was imaged cranially of the bladder and the clips
were saved, and later exported in DICOM format and analyzed on a stationary computer
using IMPAX 6.5.5 picture archiving and communication system (Agfa Healthcare,
Belgium). The uterine size (cross-sectional area in cm2) at five different locations were
measured (Fig. 23) and averaged. Furthermore, it was recorded if fluid accumulation was
present at any location of the uterine horns. For that, as much as possible of the uterus was
pictured with the ultrasound. When uterine fluid accumulation was detected then the
uterine size was determined at cross-sections without fluid accumulation. Postpartum
uterine size was used as an indicator of initial uterine involution and postpartum uterine
fluid accumulation was used as an indicator of puerperal metritis.
Figure 23. Determination of uterine size (cross-sectional area in cm2) at one location of one uterine
horn. The uterus was imaged cranially of the bladder und dorsally of the intestine. A – sectional
area of the uterine horn; B – urinary bladder; C – intestine.
47
5. Sonographic evaluation of the ovaries
The ovaries were examined in TRIAL-IV and TRIAL-V with a 10.0 MHz linear array ultrasound
probe (MyLab OneVet, SV3513, Esaote) through the rectum of the standing sow. The
ultrasound examinations were saved, exported in DICOM format and analyzed on a
stationary computer using IMPAX 6.5.5 picture archiving and communication system (Agfa
Healthcare, Belgium).
In TRIAL-IV, both ovaries of each sow were scanned twice a day, once in the morning and
once in the afternoon, starting from 3 d after weaning until ovulation. The sizes (diameter
in mm) of the five biggest follicles of each ovary were measured and averaged (Fig. 24A).
In TRIAL-V, both ovaries of each sow were scanned once a day starting from the third day
after weaning until ovulation. The day of ovulation was defined as Day 0. Furthermore, in
TRIAL-Va, ovaries were scanned at day 9, 15 and 21 of three consecutive estrus cycles
and the sizes (area in cm2) of the five biggest corpora lutea (Fig. 24B) of each ovary at
day 9 and 15 and the sizes (diameter in mm) of the five biggest follicles at day 21 were
measured and averaged. In TRIAL-Vb, ovaries were scanned at day 10 and 13 and the
size (area in cm2) of the five biggest corpora lutea of each ovary were measured and
averaged.
Figure 24. Determination of the size of one follicle (diameter in mm; A) and one corpus luteum
(area in cm2; B).
48
6. Sonographic evaluation of backfat
Sonographic evaluation of backfat was used in TRIAL-V. Backfat was determined from all
sows at weaning with a 10.0 MHz linear array ultrasound probe (MyLab OneVet, SV3513,
Esaote). The probe was placed above the last rib on the back of the sow, 5-6 cm to the
right and left of the backbone (P2 position). An image of each position was saved,
exported in DICOM format and analyzed on a stationary computer using IMPAX 6.5.5
picture archiving and communication system (Agfa Healthcare, Belgium). The thickness of
the fat layer (mm) was measured in each image at three different places and averaged
(Fig. 25).
Figure 25. Determination of backfat (thickness in mm). The ultrasound probe was placed above the
last rib on the back of the sow, 5-6 cm to the right and left of the backbone. A – fat layer; B –
Backfat layer, C – muscle layer.
49
7. Blood collection and oxytocin assay
In TRIAL-IV, blood samples were collected at estrus before, during, and after boar
presence to determine plasma oxytocin concentrations.
For that, the sows were catheterized non-surgically through the auricular vein,
approximately 2 d after weaning. Each sow was confined in a stall and restrained with a
rope snare around the upper jaw. The rope snare was tied to the stall and an i.v. catheter
(14 G, 57 mm) was inserted into the auricular vein. Through the catheter, 0.5 m of a 1.5 m
long PVC tube (o.d. 1.5 mm × i.d. 1.0 mm; SteriHealth) was passed and the outer end was
sealed with a cap, attached to the ear with adhesive tape, and the remaining tube
placed into a pouch that was attached to the neck with adhesive tape.
After catheterization, the vinyl tubing was flushed daily with a saline (0.9% NaCl) solution
containing heparin (5 IU/ml) until collection of the blood samples. Blood sample collection
occurred the day after detection of standing estrus. On the day of collection, three blood
samples were obtained before boar presence (minutes -15, -10, and -5), eight samples
during boar presence (minutes 0, 1, 2, 3, 4, 6, 8, and 10) and seven samples after boar
presence (minutes 15, 20, 25, 30, 40, 50, and 60). Thus the boar was present for 10 min.
Minute 0 was the moment of boar introduction.
Before collection of each blood sample, the first 2 ml of blood drawn up was discarded.
The following blood was collected with single-use vacuum needles into ice-chilled EDTA
tubes (HERD-II) containing 500k IU/ml of aprotinin. All blood samples were immediately
centrifuged for 10 min at 2000 rpm. The plasma was separated into 1.5 ml microtubes and
stored at -80 °C until analysis of oxytocin concentrations in the lab.
In the lab, blood samples were purified using two different methods. For HERD-I: 500 μl of
plasma was precipitated with 1 ml of acetone (Chromasolve Plus, for HPLC ≥ 99.9%,
650501/Sigma), and extracted in 2 ml of petroleum ether (Purist. p.a., ACS reagent, boiling
point range 40-60 °C, 77399/SIGMA) twice. Extracted plasma samples were evaporated in
a vacuum evaporator (RVC 2-25 CD, Martin Christ) at 40 °C, and stored at -80 °C for later
analysis of oxytocin. For HERD-II: Oasis HLB columns (Oasis HLB 1 cc Vac Cartridge, 30 mg,
Waters Corporation) columns were conditioned with 1 ml of methanol and 1 ml of water,
then 1 ml of pig plasma with 20 μl of phosphoric acid was added to the columns. Columns
were washed with 1 ml of 10% acetonitrile containing 0.1% trifluoroacetic acid (TFA) in
water and eluted with 1 ml of acetonitrile:water (80:20). Eluted samples were evaporated
in a nitrogen stream at 37 °C and stored at -80 °C for later analysis of oxytocin.
After purification, oxytocin concentrations were measured using an enzyme-linked
immunosorbent assay kit (Oxytocin ELISA kit, Enzo Life Sciences). Purified plasma samples
were reconstituted with 250 μl of assay buffer supplied in the ELISA kit, and analyzed in
duplicates according to the assay kit instruction. Sensitivities of the plasma oxytocin assay
were 7.8 pg/ml and 3.9 pg/ml, average mass recovery were 89.6% and 90.1%, intra-assay
CVs were 14.5% and 9.6%, and inter-assay CVs were15.1% and 7.4% for HERD-I and HERD-II,
respectively.
50
8. Transvaginal ultrasound-guided biopsy of the corpus luteum
A novel method for collection biopsies of the porcine corpus luteum was tested and
validated in TRIAL-V. During each biopsy, a maximum of two attempts was made. If the
first attempt was successful, the biopsy was over. If the first attempt was not successful, a
second attempt was made.
A semiautomatic Tru-Cut biopsy needle (25 cm long, 16 gauge; Ultimate PMO 16-25;
Zamar, Croatia) was used. The biopsy needle consisted of an inner needle with a 1 cm
specimen notch and an echogenic trocar tip, and an outer cutting cannula. Before the
procedure, animals were confined in a crate, feces were removed from the rectum and
the vulva was washed three times with a povidone–iodine solution (7.5% Betadine; Leiras,
Helsinki). The biopsy needle was then inserted into a needle guide (DBSE12X Biopsy kit;
Esaote) that was 18 cm in length and placed on to a 6.8-MHz microconvex array probe
(Endocavity probe, SE3123; Esaote) that was 30 cm in length. The probe was connected
to an ultrasound device (MyLab One VET; Esaote), placed in the vagina and held in place
with one hand. The other hand was placed in the rectum to localize the ovary. In the early
pregnant sow, or during the luteal phase in cycling sows, the ovary is located laterally and
ventrally of the uterine body, approximately 5–6 cm cranially of the pelvic inlet
(Constantinescu, 2007). To facilitate localization of the ovary, the suspensory ligament of
the ovary, which suspends the ovary and extends from the abdominal roof as the cranial
boundary of the broad ligament (König and Liebich 2004), was palpated and followed
ventrally until the ovary was reached. Furthermore, to alleviate grabbing of the ovary, the
ligament was pulled on to bring the ovary more caudally and dorsally of the uterine body.
Thereafter, the ovary was held between the index and middle fingers by the site of
attachment of the proper ligament, a short and strong ligament between the ovary hilus
and the cranial end of the uterine horn (Constantinescu, 2007), with the ovary on the
palm side of the hand. Subsequently, the ovary was relocated into the pelvic cavity
towards the caudal part of the cervix (Fig. 26). At the same time, the probe was inserted
deeper into the vagina towards the ovary, until the ovary became visible on the
ultrasound screen. When visible, the ovary was pushed against the face of the probe and
the biopsy needle was introduced approximately 5 mm through the vaginal wall into the
ovary. The echogenic tip of the needle allowed visual confirmation on the ultrasound
screen of the correct position of the biopsy needle in the ovary. After the biopsy needle
was in place, the trigger was pushed and the inner needle advanced 1 cm into the
ovarian tissue, exposing the specimen notch, followed by the outer cannula, which cut
and trapped ovarian tissue inside the notch (Fig. 27). After the sample had been
obtained, the biopsy needle and ultrasound probe were withdrawn from the vagina.
51
Figure 26. Illustration of the positioning of the transvaginal ultrasound-guided biopsy device. One
hand is placed in the rectum (a) and the suspensory ligament of the ovary is palpated. After pulling
on the ligament, the ovary is located and the proper ligament of the ovary held between the index
and middle fingers, with the ovary on the palm side of the hand. The ovary (b) is pulled along the
uterine cervix (c) into the peritoneal part of the pelvic cavity towards the cranial part of the vagina
(d). With the other hand, the probe (e) is inserted into the vagina adjacent to the caudal part of
the uterine cervix until the ovary becomes visible on the ultrasound screen.
In TRIAL-Va, biopsies were stained for histological evaluation of their composition. After
biopsy collection, they were drop-fixed in 10% formalin solution. After 24 h, the biopsies
were embedded in paraffin and cut with a microtome into 4 mm sections before being
mounted on glass slides. The slides were deparaffinized with xylene and ethanol, and
stained with hematoxylin–eosin.
In TRIAL-Vb, RNA was extracted from the obtained biopsies and their quality and quantity
were evaluated. After biopsy collection, they were stored in RNAlater (Sigma-Aldrich Life
Sciences) until RNA extraction (GenElute; Sigma Aldrich Life Sciences) according to
manufacturer’s instructions. For assessment of RNA quality, the RNA integrity number (RIN)
was determined (Agilent 2100 Bioanalyzer, RNA Nano LabChip; Agilent Technologies,) and
for determination of RNA concentration, a Qubit fluorometer (Qubit RNA HS assay kit;
ThermoFisher Scientific) was used. Furthermore, the concentrations were multiplied by the
volume of the RNA sample to calculate the total amount of RNA extracted per sample.
52
Figure 27. The transvaginal ultrasound-guided biopsy. (a) The ovary is pushed against the vaginal
wall and the face of the probe. The biopsy needle is pushed approximately 5 mm through the
vaginal wall into one corpus luteum (CL). (b) The echogenic tip (black arrow) of the needle allows
visual confirmation on the ultrasound screen of the correct positioning of the biopsy needle in the
ovary. (c) After the biopsy needle is in place, the trigger is pushed and the inner needle advances
1 cm into the ovarian tissue, exposing the specimen notch (gray arrow). (d) The inner needle is
followed by the outer cannula (white arrow), which cuts and traps ovarian tissue inside the notch.
53
9. Statistical analysis
Statistical processing of all data was done in SPSS Statistics (IBM Corporation) v.18.0 (TRIAL-I
and TRIAL-V), v.23.0 (TRIAL-II), v.24.0 (TRIAL-III), or SAS (SAS Institute Inc.) v.9.4 (TRIAL-IV). All
variables were checked for normal distribution. Logarithmic transformation of data was
applied for farrowing duration and square-root transformation for placenta expulsion
duration and last placenta expulsion to achieve normality. Descriptive data are presented
as mean ± SD. Before a regression model was applied, an initial univariable screening was
performed to identify potential confounders and predictors of the outcome. Based on the
type of variable, either Pearson Chi-square tests, independent t-tests, ANOVAs, Spearman
rank correlation coefficients or Pearson correlation coefficients were used. Details of the
methods and results of the initial univariable screenings can be found in the original
publications. Overview of the models, and outcome and explanatory variables used in
the major statistical analyses is presented in Table 3.
In TRIAL-I, the outcome variable was defined as ‘pregnant’ or ‘not pregnant’. The
explanatory variables were: parity, herd (SATELLITE-I vs. SATELLITE-II), housing (crate vs. pen)
gestation length, farrowing duration, number of total, liveborn and stillborn piglets, backfat
at farrowing, lactation length, number of weaned piglets, weaning-to-estrus interval, boar
semen used at insemination and number of inseminations. Farrowing duration was
transformed into a binary variable based on a 300 min threshold value. Sow and Batch
were included as random factors. A logistic regression model was built and a backward
stepwise elimination with a likelihood-ratio test was performed. When significant predictors
were found, the odds ratio ± SE was calculated.
In TRIAL-II, the outcome variables related to placenta expulsion were: number of expelled
placental parts, placenta expulsion duration, and first and last placental part expulsion.
The following explanatory variables were: parity, gestation length, farrowing duration, and
number of liveborn and stillborn piglets. Each of the outcome variables was tested in a
distinct general linear model. Sow and batch were included as random factors and the
use of oxytocin (yes/no) as a fixed effect. Furthermore, the quadratic term for farrowing
duration was included, except for the ‘first placental part expulsion‘model, because we
hypothesized a positive correlation between farrowing and placenta expulsion duration in
sows with short farrowings, as found in the study of Van Rens and Van der Lende (2004),
and a negative correlation between farrowing and placenta expulsion duration in sows
with prolonged farrowings due to impaired placenta expulsion. Furthermore, because we
hypothesized that there were differences in the correlation between farrowing duration
and placenta expulsion between sows receiving and sows not receiving exogenous
oxytocin during parturition, data were further divided according to the use of oxytocin
and consequently two additional submodels for each outcome were created. Results are
presented as least squares means ± SEM or estimates ± SE. Furthermore, for post hoc
comparisons of means, Tukey's HSD was used. Furthermore, we compared farrowing
duration and other explanatory variables (parity, gestation length, and number of liveborn
and stillborn piglets) between sows with no retained placentas, partially retained
placentas and totally retained placentas. The effects were tested using a multinomial
logistic regression model with ‘no retained placentas’ as the reference category.
54
In TRIAL-III, the outcome variables were uterine size and uterine fluid accumulation. Uterine
fluid accumulation was defined as ‘no fluid’ or ‘fluid’. Uterine size was transformed into a
binary variable and defined as ‘normal’ or ‘enlarged’ based on the postpartum day of
examination and the mean of all sows examined on that specific postpartum day. The
explanatory variables were: parity, gestation length, number of live and stillborn piglets,
application of obstetrical intervention (yes vs. no), farrowing duration, placental expulsion
duration, placentae retention (defined as no expelled placental parts within the
observation period of 24 h after expulsion of the last piglet), relative placental expulsion
duration (placental expulsion duration relative to farrowing duration), and relative number
of expelled placental parts (number of expelled placental parts relative to the number of
total piglets born). These last two variables were used as indicators for impaired placental
expulsion and transformed into binary variables based on observations done on TRIAL-II.
Placental expulsion was defined as ‘normal’ if relative placental expulsion duration and
relative number of expelled placental parts were higher than 10% and defined as
‘impaired’ if they were lower than 10%. Furthermore, farrowing duration was transformed
into a categorical variable and defined as ‘short’ if farrowing was briefer than 400 min
and defined as ‘prolonged’ if it exceeded 400 min. Altogether, two binary logistic
regression models were created, one for each outcome variable. Sow and batch were
included as random factors and use of oxytocin (yes vs. no) and antimicrobials (yes vs. no)
during parturition or between parturition and postpartum day of ultrasound examination
were fixed factors. A backward stepwise elimination with a likelihood-ratio test was
performed. If an explanatory variable was a significant predictor, the odds ratio (± S.E.)
was also calculated.
In TRIAL-IV, the outcome variables were: oxytocin concentration at breeding (divided into
four categories: minutes -15 to -5; minutes 0 to 10, minutes 15 to 60, and minutes -15 to 60)
and follicular development (follicular size at day 3, 4, and 5 after weaning as well as
maximum follicular size during weaning-to-ovulation interval and size at first day of
standing heat). As explanatory variables, farrowing duration and parity were used and
transformed into categorical variables. Farrowing duration was defined as ‘short’ if it was
shorter than 300 min and defined as ‘prolonged’ if it was longer than 300 min. Parity was
defined as ‘young’ of it was three or lower and defined as ‘old’ if it was four or higher.
Linear mixed models for repeated measures were used with a nested design where
farrowing duration and parity were nested within herd (HERD-I vs. HERD-II). For post hoc
comparisons, a Kenward-Rogers approximation was used.
In TRIAL-Va, logistic regression analysis and ANOVA were performed to investigate
differences in backfat, weight and parity between successfully and unsuccessfully
sampled sows. A sow was defined as successfully sampled if at least one biopsy during the
entire sampling period was successful. A biopsy was defined as successful if the first or
second attempt was successful. An attempt was defined as successful if ovarian tissue
was present in the biopsy needle notch.
In TRIAL-Vb, corpus luteum size at day 13 of pregnancy, and subsequent gestation length
and litter size were compared between sows that underwent transvaginal ultrasound-
guided biopsy and sows of the control group using an independent t-test. Furthermore,
the pregnancy rate was compared using a Pearson Chi-square test.
55
Table 3. Overview of the models, and outcome and explanatory variables used in the major
statistical analyses.
1 – Normal farrowing < 300 minutes and prolonged farrowing ≥ 300 minutes
2 – Expulsion of last placental part observed before the expulsion of the last piglet
3 – No expelled placental parts during the observation period
4 – Based on the postpartum day of examination and the mean of all sows examined on that specific
postpartum day. Normal size < mean and enlarged size ≥ mean.
5 – Normal if relative placental expulsion duration and relative number of expelled placental parts were higher
than 10% and impaired if they were lower than 10%.
6 – Young ≤ parity 3 and old > parity 3
56
57
Results
58
1. Effect of prolonged farrowing on subsequent pregnancy
rate
In TRIAL-I, 148 sows were of parity, 4.4 ± 1.9 (mean ± SD), had a gestation length of 116 ±
1.2 d, a backfat at farrowing of 14.3 ± 3.6 mm, and a farrowing duration of 268 ± 142 min.
There were 11.9 ± 2.9 liveborn piglets and 0.8 ± 1.3 stillborn piglets at birth. After a lactation
of 31.6 ± 2.9 d 9.8 ± 1.6 piglets were weaned.  87% (n = 129) of the sows were ‘pregnant’
and 13% (n = 19) of the sows were ‘not pregnant’. In the final logistic regression model,
duration of farrowing was significantly associated with the outcome (P < 0.01; Fig. 28).
Sows with a farrowing duration of ≥ 300 min were 3.4 ± 1.8 (Odds ratio ± SE; P = 0.027) more
likely to be ‘not pregnant’.
Fig. 28: Sows that failed to become pregnant at first insemination after weaning had a previous
farrowing 100 minutes longer than sows that become pregnant.
59
2. Effect of prolonged farrowing on placenta expulsion and
retained placentae
In TRIAL-II, 142 parturitions of 101 sows in 10 batches were recorded: 66 sows had a single
parturition, 29 sows had two consecutive parturitions and 6 sows had three consecutive
parturitions. An overview is provided in Table 4.
Use of exogenous oxytocin towards the end of farrowing (used in 44 out of 142 parturitions)
increased the number of expelled placental parts (3.8 ± 0.2 vs. 2.9 ± 0.3; LSmean ± SEM; P =
0.035) and decreased the placenta expulsion duration (172 ± 44 vs. 328 ± 26 min; P =
0.011) and the time of last placental part expulsion (148 ± 48 vs. 300 ± 24 min; P = 0.025).
If oxytocin was not used, farrowing duration obeyed a quadratic relationship with the
number of expelled placental parts (P = 0.001; Fig. 29), placenta expulsion duration (P =
0.002; Fig. 30) and time of last placental part expulsion (P = 0.024; Fig. 31). Nevertheless,
when the two observations with FAR < 100 min and the five observations with FAR > 800
min were excluded from the model, the quadratic term for farrowing duration was not
significantly correlated with the placenta expulsion duration but only the linear term of
farrowing duration (0.7 ± 0.2; Β ± SE; P = 0.005).
If oxytocin was used, number of expelled placental parts was positively associated with
number of liveborn piglets (0.2 ± 0.1; Β ± SE; P = 0.002) and was affected by parity. 5th
parity sows expelled more placental parts (4.3 ± 0.4; LSmean ± SEM) than 4th (3.2 ± 0.3; P =
0.024) and 3rd parity sows (2.7 ± 0.4; P = 0.008). Furthermore, placenta expulsion duration
was positively associated with number of liveborn piglets (18 ± 8 min; Β ± SE; P = 0.025). First
placental part expulsion was negatively correlated with farrowing duration (0.3 ± 0.1; P =
0.001; Fig. 32) and there was no difference between sows that received oxytocin and
sows that did not.
Table 4: Descriptive statistics for all sows included in TRIAL-II.
60
Figure 29. Relationship between farrowing duration and the number of expelled placental parts in
sows receiving no oxytocin during parturition. Error bars indicate a single SE.
Figure 30. Relationship between the farrowing duration and the placenta expulsion duration in sows
receiving no oxytocin during parturition. Error bars indicate a single SE.
61
Figure 31. Relationship between the farrowing duration and the expulsion of the last placental part
relative to the birth of the last piglet in sows receiving no oxytocin. Error bars indicate a single SE.
Figure 32. Relationship between the farrowing duration and the expulsion of the first placental part
relative to the birth of the last piglet. Error bars indicate a single SE.
62
Four sows experienced total (no expulsion of placental parts) and four sows partial
retained placentae (no expulsion of placental parts after birth of the last piglet). Their
farrowing durations were longer (1009 ± 275 and 734 ± 136 min with P = 0.004 and P =
0.021, respectively) compared with those of sows that had no retained placentas (369 ±
202 min; LSmean ± SD). In 3 sows with total retained placentae, hyperechoic structures
within the lumen of the uterus and/or the endometrium were detected (Fig. 33).
Figure 33. Ultrasound image of one sow with total retained placentae at day 3 after farrowing.
Ultrasound findings are dense structures inside the uterine lumen and endometrium. A – Uterine
horn; B – Uterine lumen and endometrium; C – Enlarged blood vessels.
63
3. Effect of prolonged farrowing on postpartum uterine size
and intrauterine fluid
In TRIAL-III, the uterine size at the 2nd postpartum (p.p.) day was 4.19 ± 1.23 cm2 (mean ±
SD; n/N = 8/99); at the 3rd p.p. day 4.05 ± 1.05 cm2 (24/99); at the 4th p.p. day 3.96 ± 0.89
cm2 (25/99); at the 5th p.p. day 3.75 ± 0.99 cm2 (20/99); at the 6th p.p. day 3.73 ± 0.44 cm2
(13/99); and at the 7th p.p. day 3.53 ± 0.86 cm2 (9/99). 55 sows had a ‘normal’ and 44 sows
had an ‘enlarged’ uterine size (Fig. 34). 15 had accumulation of intrauterine fluid (Fig. 35).
13 out of 15 sows with uterine fluid accumulation had also increased uterine size.
Results for the final binary logistic regression models are presented in Table 5 and 6.
Figure 34. Ultrasound image of one sow with enlarged uterus at day 3 after farrowing. A – Enlarged
uterine horn; B – Enlarged blood vessels.
64
Table 5. Significant predictors of the final binary logistic regression model with ‘normal’ and
‘enlarged’ uterine size as binary outcome variable.
Enlarged uterine size
(in 44/99 sows)
Predictor n/N % B SE Wald χ2 Odds Sig.
Normal farrowing*  9/44 20
Prolonged farrowing  35/55 65 2.037 0.565 13.012 7.671 0.001
Parity
2*
1/11 9 9.595 0.022
3  10/19 53 2.404 1.274 3.563 11.070 0.059
4  12/35 34 1.867 1.235 2.286 6.466 0.131
5  21/34 61 3.107 1.217 6.520 22.352 0.011
No obstetrical
intervention*
32/82 39
Obstetrical intervention  12/17 70 1.457 0.696 4.381 4.293 0.036
No oxytocin
administration*
26/68 38
Oxytocin administration  18/31 58 - 1.483 0.721 4.721 0.235 0.040
Stillborn piglets
0*
17/45 38 4.870 0.088
1  12/33 36 - 0.165 0.606 0.074 0.848 0.786
≥ 2  15/21 71 1.353 0.697 3.769 3.871 0.052
* - reference category
65
Figure 35. Ultrasound image of one sow with uterine fluid accumulation at day 3 after farrowing. A –
Uterine horn with fluid accumulation.
66
Table 6. Significant predictors of the final binary logistic regression model with ‘no intrauterine fluid’
and ‘uterine fluid’ as binary outcome variable.
Intrauterine fluid
(in 15/99 sows)
Predictor n/N % B SE Wald χ2 Odds Sig.
Normal farrowing* 2/44 5
Prolonged farrowing 13/55 24 1.740 0.843 4.255 5.697 0.039
Stillborn piglets
0* 5/45 11 9.347 0.009
1 3/33 9 - 0.505 0.897 0.317 0.603 0.574
≥ 2 7/21 33 2.616 0.889 8.653 13.674 0.003
No obstetrical intervention* 5/82 6
Obstetrical intervention 10/17 59 1.799 0.802 5.027 6.042 0.025
Normal placental
expulsion* 11/93 12
Impaired placental
expulsion 4/6 67 3.293 1.638 4.040 26.928 0.044
* - reference category
67
4. Effect of prolonged farrowing on follicular development
after weaning and oxytocin release at subsequent estrus
In TRIAL-IV, farrowing duration was 358 ± 235 min. Sows with ‘short’ farrowing duration (n =
14) had 159 ± 29 min and sows with ‘prolonged’ farrowing duration (n= 16) had 533 ± 190
min. ‘Young’ sows (n = 14) had a parity of 2.5 ± 0.8 and ‘old’ sows (n = 16) had a parity of
6.4 ± 2.3.
At subsequent estrus, oxytocin concentrations were higher during boar presence
compared with before and after (P < 0.05, Fig. 36). Furthermore, during the boar presence,
there was an effect of farrowing duration on oxytocin concentrations (P < 0.01; Fig. 36;
Table 7). Sows with ‘short’ farrowing duration had lower oxytocin concentrations
compared with sows with ‘prolonged’ farrowing duration. Post hoc comparison showed
that in ‘young’ sows in particular, oxytocin concentrations during and after boar presence
were higher in sows with ‘prolonged’ farrowing duration compared with sows with ‘short’
farrowing duration (P < 0.001 and P < 0.01, respectively). On the other hand, oxytocin
concentrations of ‘old’ sows were not affected by previous farrowing duration (P > 0.10,
for all).
Figure 36. LS means and SE of plasma OT concentrations of oestrous sows in boar presence are
plotted by SHORT-YOUNG (n = 9), SHORT-OLD (n = 5), LONG-YOUNG (n = 5) and LONG-OLD (n = 11)
groups, according to farrowing duration and parity classes. The red shading indicates 10 min of
boar presence.
68
Table 7. The effects of farrowing duration and parity on the plasma oxytocin concentrations at
subsequent estrus before, during, and after boar exposure as well as during the whole sampling
period.
Young Old P value
Period2
NORMAL
 (n = 9)
PROLONGED
(n = 5)
NORMAL
(n = 5)
PROLONGED
(n = 11)
FD3 Parity4 FD*parity
-15 to -5 min 13.8 ± 1.5 14.8 ± 1.7 13.7 ± 1.7 13.3 ± 1.6 0.79 0.76 0.48
0 to 10 min 18.6 ± 7.6 28.9 ± 7.8 24.2 ± 7.8 27.6 ± 7.7 < 0.01 0.84 0.13
15 to 60 min 13.7 ± 2.5 16.8 ± 2.6 16.6 ± 2.7 16.2 ± 2.6 0.58 0.79 0.07
-15 to 60 min 16.9 ± 2.5b 19.7 ± 2.6a 18.8 ± 2.6ab 18.8 ± 2.5ab 0.44 0.91 0.04
1 – Values represent LS means ± SE of plasma oxytocin concentrations of the sows (total n = 30).
2 – Repeated measures carried out on the values over four periods: 1) -15, -10 and -5 min before
boar presence, 2) 0, 1, 2, 3, 4, 6, 8 and 10 min during 10 min of boar presence, 3) 15, 20, 25, 30, 40,
50 and 60 min after boar presence, and 4) total period.
3 – FD (farrowing duration) was divided into two classes: 1) SHORT: between 103 and 192 min, and 2)
LONG: between 326 and 878 min.
4 – Parity was divided into two classes: 1) YOUNG: between 1 and 3, and 2) OLD: between 4 and 11.
a,b – Different letters indicate that variables were significantly different at P < 0.05.
69
The follicular size at day 4 after weaning tended to be larger in the sows with ‘prolonged’
farrowing duration compared with sows with ‘short’ farrowing duration (P = 0.06, Table 8),
but no other differences were established.
Table 8. The effects of farrowing duration and parity on follicular development and weaning to
oestrus intervals of subsequent oestrous sows.
Young Old P value
SHORT
(n = 9)
LONG
(n = 5)
SHORT
(n = 5)
LONG
(n = 11)
FD1 Parity2 FD*parity
Days3
  WOI 5.0 ± 0.2 5.4 ± 0.2 5.1 ± 0.2 4.8 ± 0.2 0.79 0.32 0.10
  Maximum 4.0 ± 0.4 4.4 ± 0.5 4.7 ± 0.5 4.7 ± 0.4 0.77 0.38 0.61
Diameters, mm4
  Day 3 4.5 ± 0.4 5.0 ± 0.6 4.8 ± 0.5 5.5 ± 1.0 0.56 0.42 0.81
  Day 4 5.5 ± 0.3 5.9 ± 0.5 5.1 ± 0.5 5.9 ± 0.6 0.06 0.83 0.65
  Day 5 5.5 ± 0.3 5.8 ± 0.3 6.1 ± 0.3 6.2 ± 0.2 0.44 0.07 0.65
  Oestrus 6.0 ± 0.4 5.3 ± 0.4 5.7 ± 0.4 6.3 ± 0.3 0.84 0.30 0.10
  Maximum 5.7 ± 0.2 5.6 ± 0.3 6.2 ± 0.3 6.2 ± 0.2 0.80 0.02 0.95
1 – FD (farrowing duration) was divided into two groups: 1) SHORT: between 103 and 192 min, and 2)
LONG: between 326 and 878 min.
2 – Parity was divided into two groups: 1) YOUNG: between 1 and 3, and 2) OLD: between 4 and 11.
3 – Values represent LS means ± SE of the days for 1) weaning-to-oestrus interval, and 2) when the 5
largest follicles reached their maximum size after weaning until ovulation.
4 – Values represent LS means ± SE of the mean diameters of the five largest follicles measured 1) on
day 3, 2) on day 4 and 3) on day 5 after weaning until ovulation, 4) at oestrus in which the sow
showed a standing response to the back test, and 5) when the follicles reached their maximum size
after weaning until ovulation.
70
Furthermore, the follicular size at fifth day after weaning and the maximum follicular size
between weaning and ovulation were correlated with oxytocin concentrations at estrus
before, during and after boar presence (rs = 0.23, P < 0.0001 and rs = 0.38, P < 0.0001,
respectively) (Fig. 37).
Figure 37. Average plasma oxytocin concentrations of the sows during the total sampling period,
i.e. -15 to +60 min from beginning of 10 min of boar presence, and follicle diameters measured at
five days after weaning and when the follicles reached their maximum size after weaning until
ovulation are indicated with Spearman correlation coefficients (rs) (n = 30).
71
5. Transvaginal ultrasound-guided biopsy of corpora lutea
In TRIAL-Va, 18 out of 24 biopsies were successful (75%). In TRIAL-Vb, 20 out of 52 biopsies
were successful (38%). Thus, samples were obtained in 50% of all biopsies. All sows in TRIAL-
Va were successfully sampled whereas only 20 sows (77%) in TRIAL-Vb. Overall, compared
with unsuccessfully sampled sows (n = 6), successfully sampled sows (n = 24) were older
(parity 5.0 ± 2.8 vs. 2.2 ± 0.5), heavier (290 ± 26 vs. 244 ± 27 kg) and had thicker backfat
(11.4 ± 2.7 vs. 6.4 ± 2.5 mm; P < 0.05 for all).
In TRIAL-Va, no abnormal cyclic changes during the sampling period were found (Table 9)
and no major pathological changes after the sampling period (Fig. 38) were found in the
ovary. The only pathological gross findings noted were small focal blood clots and small
multifocal scars.
Figure 38. Macroscopic inspection of the ovaries of one sow after six consecutive biopsies taken
during three consecutive estrus cycles. Pathological changes found were (a) small multifocal scars
(arrows) and (b) small focal blood clots (bolt arrow).
72
Table 9.  Mean sizes and ranges of follicles and/or corpora lutea present in the ovaries before each
transvaginal ultrasound-guided biopsy.
Day of TUB Area corpora lutea (cm2) Diameter follicles (mm)
1. Cycle – Day 9 0.93 (0.90 – 0.99)
1. Cycle – Day 15 0.73 (0.69 – 0.78) 2.6 (2.2 – 3.2)
1. Cycle – Day 21 6.6 (5.9 – 7.2)
2. Cycle – Day 9 0.94 (0.89 – 1.00)
2. Cycle – Day 15 0.73 (0.71 – 0.76) 2.7 (2.4 – 3.1)
2. Cycle – Day 21 6.5 (5.7 – 7.2)
3. Cycle – Day 9 0.95 (0.93 – 0.96)
3. Cycle – day 15 0.72 (0.70 – 0.75) 2.9 (2.6 – 3.4)
3. Cycle – day 21 6.3 (5.9 – 6.8)
In TRIAL-Vb, there were no significant differences (P > 0.05) between the successfully
sampled sows (n = 20) and the control group (n = 26) in terms of corpora lutea size at day
13 (0.89 ± 0.10 vs 0.92 ± 0.11 cm2), pregnancy rate (95% vs 96%), gestation length (115 ± 1
vs 115 ± 1 days) and the total number of piglets born (12.7 ± 2.5 vs 13.3 ± 2.8).
Seventeen samples were subjected to RNA extraction and assessment of RNA quality and
quantity. Three samples (15%) were excluded from analysis because they contained
almost no RNA. Across the remaining samples, the RNA concentration was 520 ± 160 μg
per sample and the RIN was 8.5 ± 1.x (range 6.7 – 10). RNA extracted from 14 samples
(82%) had a RIN >8.
The histological composition of 18 samples was assessed. The shape of the samples was
round and they were yellow–red in color; the samples were ~1 mm in diameter (Fig. 39a).
The histological composition was heterogeneous and contained luteal cells (Fig. 39b),
follicular cells, such as granulosa cells (Fig. 39c), stromal cells (Fig. 39d) and occasionally a
small number of blood cells and cells from the vagina wall.
73
Figure 39. Histological examination of a sample obtained on day 15 of the estrus cycle. (a) The
sample is approximately 1 mm in diameter. (b–d) The histological composition of the sample is
heterogeneous and contains luteal cells (b) and follicular cells, such as granulosa cells (c) and
stromal cells (d).
74
75
Discussion
76
1. Parturition and subsequent uterine health and fertility
This study demonstrated, for the first time, an association between the length of farrowing
and the subsequent pregnancy rate. Furthermore, this study demonstrated, also for the
first time, that retained placentae can occur in sows. In general, the results indicated that
prolonged farrowing has a negative effect on postpartum uterine health. The use of
exogenous at the end of parturition, on the other hand, seems to support placental
expulsion and subsequent uterine health.
On average the last placenta was expelled within 4.5 h after birth of the last piglet and at
latest after 15.5 h, in overall agreement with the general opinion that the placenta is
expelled within 4 h, but may take up to 12 h (Taverne and Noakes, 2009). However, only
about 50% of the sows expelled their placenta within 4 h, which indicates that longer
placenta expulsion durations are not unusual. Overall, the duration of placenta expulsion
depended mainly on the use of exogenous oxytocin. If oxytocin was used, more placental
parts were expelled faster. That indicates that there a direct link between farrowing
duration and placental expulsion since both prolonged farrowing (Oliviero et al., 2008;
Castrén et al., 1993) and prolonged placental expulsion (in this study) were associated
with lack of oxytocin. If oxytocin was not used, farrowing duration and placenta expulsion
duration were significantly quadratically associated. However, when sows with extremely
short and long farrowing duration (< 100 and > 800 min) were excluded from the analysis,
farrowing duration and placenta duration were linearly correlated, as in the study of Rens
and Van der Lende (2004). Thus, it can be confirmed that in sows and gilts the placenta
duration increases with increasing farrowing duration. Furthermore, this study showed that
placenta duration was very short when farrowing duration was longer than 800 min, which
could be a sign of impaired placenta expulsion and therefore placenta retention.
Furthermore, the same phenomenon was recorded for the number of expelled placental
parts, which depended entirely on the use of oxytocin. Only if oxytocin was used did
placental parts depend on the number of liveborn piglets. This is in agreement with the
finding of Van Rens and Van der Lende (2004). In contrast, when oxytocin was not used,
the number of liveborn piglets was not correlated and farrowing duration was
quadratically correlated with the number of expelled placental parts. This is another
indication that sows with prolonged farrowing might have experienced impaired placenta
expulsion due to lower oxytocin concentrations. A third indication is that the expulsion of
the last placental part relative to the last piglet was also influenced by oxytocin. In
contrast, the onset of placenta expulsion seems to be entirely associated with duration of
farrowing. In sows with short farrowing, placental expulsion started after the birth of the last
piglet, whereas in sows with prolonged farrowing, placental expulsion started before the
birth of the last piglet. This observation is in agreement with the report of Van Rens and
Van der Lende (2004). Additionally, we observed delayed expulsion of the first placenta in
older sows that received no oxytocin, which could also be related to lower oxytocin
concentrations in older sows and is supported by the finding of Oliviero et al. (2008), who
showed that older sows housed in farrowing crates had lower oxytocin concentrations.
77
Sows with partially retained placentas (no placenta expulsion after the expulsion of the
last piglet) and sows with totally retained placentas (no placenta expulsion) had
extremely long farrowing durations, of about 11 h and 17 h, respectively. These sows
accounted for 3 - 6% of all sows and none received oxytocin during parturition. This finding
has not been reported earlier for gilts or sows. In other species, causes of retained
placenta are generally considered to be factors that interfere with normal maturation
and/or expulsion of the placenta (Parkinson, 2009; Paccamonti and Pycock, 2009). Birth
before maturation can be caused by, induced parturition or stress, for example. In the
present study, none of the parturitions were induced and stress levels were similar for all
sows because all sows experienced the same farrowing environment and management.
Furthermore, we observed no effect of the gestation length on the placenta expulsion,
which would indicate premature parturition. Thus, it is more likely that the retained
placentas were rather caused by factors interfering with normal placenta expulsion.
Impaired expulsion of the placenta can be due to a physical obstruction, such as retained
piglet or cervical closure, or inadequate uterine contractions (Parkinson, 2009;
Paccamonti and Pycock, 2009). No retained piglets were detected in sows with retained
placentas. On the other hand, cervical involution starts already six hours after start of the
second phase of parturition (Waldmann, 2009). The cervix will be still open but loss of
elasticity, stretchability may complicate expulsion of the placentae. Nevertheless, most
likely inadequate uterine contractility caused the impaired placenta expulsion.
The findings that the farrowing duration, number of stillborn piglets and number of
obstetrical interventions were negatively associated with the occurrence of intrauterine
fluid, and therefore puerperal metritis, is in agreement with the findings of Bostedt et al.
(1998) in gilts, where 85.9% of gilts with consequent puerperal illness had farrowing
durations of longer than 6 h, whereas 78.8% of control gilts of the same age finished
parturition in less than 3 h. Other significant factors affecting the incidence of feverish
puerperal illness were frequency of obstetrical intervention and stillbirth rate. The
frequency of obstetrical intervention measured in the group of patients was 27%, but only
9.5% in the group of control sows. Thus, risk factors for puerperal metritis in gilts and sows
seem to be similar and may be generalized for all female pigs. Furthermore, retention of
placentae has not yet been associated with puerperal metritis in gilts but was evident in
the present study. The reason why retained placentae increase the risk for uterine
infection is not clear but may be due to retained placenta increasing bacterial
contamination and providing a suitable medium for bacterial growth within the uterus, as
mentioned by Maes et al. (1998). He also reported that prolonged farrowing increases the
bacterial contamination within the uterus probably due to either prolonged opening of
the cervix, and therefore prolonged reduction of the physical barrier against
contamination, and/or reduced reduction in uterine contractions (Castrén et al., 1993;
Oliviero et al., 2008), and therefore reduced uterine clearance.
78
Increased uterine contamination may also have been the reason why obstetrical
intervention increased the risk for uterine infection. It is likely that during the obstetrical
interventions additional contamination of the birth canal occurs, especially when the
genital area of the sow is not washed with an antiseptic solution before intervention, as
was the case in the present study. In conclusion, it can be said that prolonged farrowing
most likely increases the odds for puerperal metritis to develop and therefore postpartum
dysgalactia (PDS). It can be supported by evidence that prolonged parturition and PDS
have similar risk factors. Risk factors for PDS were reviewed by Maes et al. (2010) and risk
factors for prolonged farrowing by Peltoniemi and Oliviero (2015). Common risk factors are
constipation, low dietary fiber, high body condition and restricted movement inside
farrowing crates.
Furthermore, the present study showed that farrowing duration could increase the time for
uterine involution because sows with prolonged farrowing have larger uterine size at the
beginning of involution. Considering that farrowing duration has increased substantially
during recent decades, it is possible that uterine involution takes longer in modern
hyperprolific sow lines than it has been reported earlier (Palmer et al., 1965; Kudlac and
Groch, 1979; Busch, 2007). In addition to farrowing duration, all other factors affecting
intrauterine fluid were also found to increase uterine size. The reason may be that an
inflamed uterus is usually enlarged (Kauffold et al., 2005). Thus, puerperal metritis may be
one of the major causes of delayed uterine involution. On the other hand, the
administration of oxytocin supported uterine involution. The reason may be that oxytocin
supports uterine contractions and therefore uterine clearance. Considering that cervical
involution and closure occur rapidly within a week, supported uterine clearance before
that may lead to the uterus needing to resorb less compared with the case for sows that
do not receive oxytocin.
Nevertheless, the present study also showed that postpartum ultrasound may be a
practical and adequate tool to assess uterine involution. Busch (2007) investigated
whether the visual study of the degree of edematization or the formation and length of
the vulva could be used as an indicator for the point in time of uterine involution, but they
did not show them to be appropriate parameters. Nevertheless, whether determination of
the mean sectional area with the aid of ultrasound is an appropriate parameter remains
to be investigated. The present study showed that there are large differences between
sows and follow-up studies are needed on sows with ultrasound until involution is
completed. The present study also shows that at least parity, probably also body size,
should be taken into consideration because parity significantly affected uterine size. On
the other hand, many authors disagree about the importance of the size of the uterus at
the end of involution and whether uterine size is a good indicator (Kauffold et al., 2005).
Nevertheless, ultrasound examination of the uterus at the beginning of the puerperium
may represent a practical tool to anticipate the length of uterine involution and therefore
could be used in reproductive management of the sow.
79
How prolonged parturition may negatively affect subsequent fertility was investigated, but
the explanation for the connection remains unclear and further research is required.
Contrary to the hypothesis, sows with prolonged farrowing did not have impaired follicular
development after weaning and reduced oxytocin release during boar stimulation at
subsequent estrus. In contrast, sows, especially young sows, with prolonged farrowing had
slightly larger follicles and released more oxytocin compared with sows that farrowed
normally. There are two possible explanations. One is that, as with cows, impaired uterine
health, such as a uterine bacterial infection, and/or the associated inflammation,
suppresses postpartum pituitary LH secretion and therefore perturbs postpartum ovarian
follicular growth and ovulation (Peter et al., 1988; Sheldon et al., 2002; Opsomer et al.,
2000). Follicles of sows with suppressed LH secretion will have larger pre-ovulatory follicles
than sows with normal LH secretion because pre-ovulatory follicles may experience
impaired luteinization of the follicular cells. Furthermore, sows with suppressed LH secretion
will either experience ovulation failure or ovulate later than sows with normal LH-secretion.
Ovulation failure will result in regression of follicles or development of follicle cysts. Late
ovulation will cause release of reduced quality oocytes. On the other hand, in this study,
the time between weaning and ovulation did not differ significantly for sows with normal
and prolonged farrowing. Therefore, it is unlikely that there were significant differences in
LH secretion between normal and prolonged-farrowing sows and therefore timing of
ovulation. Another explanation why sows with prolonged farrowing had larger follicles
after weaning could be lactation weight loss, the major factor affecting follicular growth
after weaning (Soede and Kemp, 2015). If weight loss is high, follicular growth is low. Thus,
in this study, sows with prolonged farrowing would have hypothetically lost less weight
compared with those that farrowed normally because they had larger follicles. High
weight loss is mainly caused by either low energy intake (reduced feed intake) or by high
energy output (increased milk production). Since it is possible that postpartum dysgalactia
(PDS) is the link between prolonged farrowing and subsequent subfertility it is thinkable
that sows with PDS loose less weight because one of the major symptoms of PDS is
reduced milk production. On the hand, this explanation is contrary to that of Tummaruk et
al. (2013) and Hoy (2006). Tummaruk et al. (2013) showed that sows with prolonged
farrowing had reduced feed intake for a longer time after farrowing than sows that
farrowed normally. Thus, sows with reduced feed intake should experience higher
lactation weight loss, in line with the findings of Hoy (2006) who reported that sows with
PDS had a higher rate of anestrus after weaning or a higher rate of late onset of estrus
which are typical signs for high lactation weight loss (Soede and Kemp, 2015).
A schematic presentation of the proposed mechanisms playing a role in the aetiology of
puerperal diseases in sows is presented in Figure 40.
80
Figure 40. A schematic presentation of the proposed mechanisms playing a role in the aetiology of
puerperal diseases in sows
81
2. Ultrasound-guided biopsy collection of luteal tissue
The present study shows that it is possible to conduct a transvaginal ultrasound-guided
biopsy (TUB) in pigs. The success rate, defined as the proportion of successful biopsies, of
50% is lower compared with that for cows (92%; Kot et al., 1999) and horses (68%; Beg et
al., 2005). One explanation could be the smaller size of this species, and therefore
narrower vaginal and rectal diameters, which make it harder to guide the ovary and
biopsy needle towards each other. This effect is also seen in the correlation between
success rate and sow factors. The older (after 3rd farrowing) and bigger (weight > 250 kg or
backfat > 6 mm or both) the sow was, the more successful the biopsy. Furthermore, it
seems that the success rate can be improved by increasing the time between two
consecutive biopsies. In the first part of the study, when the success rate was 75%, 10.5 d
was the average time between biopsies, whereas in the second part, when success rate
was 38%, it was only 3 d. One explanation could be the swelling of the rectal wall due to
the manipulation of the ovary. We noticed that with only 3 d between two biopsies the
rectal wall remained swollen, which makes it hard to find, take, and pull the ovary against
the vaginal wall. There seems to be no impact on the reproductive performance, such as
cyclicity and pregnancy status, of the sow.
Our findings suggest that the four sows, included in the first part of the study, cycled
normally during the sampling period of three consecutive cycles because they had
normal sized functional bodies on each sampling day (Soede et al., 2011). Furthermore,
we found no differences in the reproductive performance of the pregnant sows. The
findings of small multifocal blood clots at the ovaries, during post-mortem investigation of
the reproductive tract, and the reddish appearance of the biopsies, indicate that mild
focal bleeding occurred. However, the bleedings appear to have no effect on
reproductive performance. The size of the sample, 1 mm in diameter, is very small
compared with that for cows (11.7 mm long; Kot et al., 1999) and horses (Beg et al., 2005).
One explanation for why the sample is so much smaller could be the smaller size of the
corpus luteum (10 mm; Langendijk and Peltoniemi, 2013) compared with that for cows (20-
30 mm; Quintal-Franco et al., 1999) and horses (20-35mm; Allen 2001). Any minor deviation
of the biopsy needle from the midline of the corpus luteum could potentially lead to a
biopsy of significantly reduced size. That probably explains also the heterogeneous
composition of the biopsy. If the biopsy needle hits the corpus luteum at the periphery
rather than in the center, it is possible also that other cells, such as stroma cells, of the
ovary are sampled. However, if the samples are intended to be used for RNA studies, such
as real time qPCR and other gene expression studies, it is important to realize that the
sample contains heterogeneous tissue. In these circumstances it might be beneficial to
make use of methods that aim at individual cells, i.e. laser microdissection or
immunohistochemistry. Still, it is possible to conduct gene expression studies. A RNA
integrity number (RIN) of higher than 8 shows that RNA of sufficient quantity and quality
can be collected. The RIN ranges from 1 to 10, where 1 indicates that RNA is completely
degraded and 10 that RNA is completely intact (Schroeder et al., 2006). A RIN of higher
than 8 is generally regarded as perfect total RNA and samples can be used for
downstream applications (Fleige and Pfaffl, 2006).
82
83
Conclusions, practical implications and future
research
84
1. Conclusions
1. Prolonged farrowing negatively affects subsequent fertility and sows with prolonged
farrowing may require attention after farrowing.
2. Ultrasound examination of the uterus after farrowing is a practical and relaible tool
to assess whether uterine health is compromised.
3. Prolonged farrowing impairs placenta expulsion and can lead to retained
placentae.
4. Administration of exogenous oxytocin improves placenta expulsion and decreases
uterine size in the first week postpartum and therefore may support uterine
involution.
5. Prolonged farrowing and retained placentae, as well as obstetrical intervention
and more than two stillborn piglets, increase the risk for intrauterine fluid
accumulation and increased uterine size in the first week postpartum.
6. It is possible to collect luteal tissue of sows in vivo for research purposes using a
transvaginal ultrasound-guided biopsy method.
85
2. Practical implications
1. Proper farrowing management and prevention of prolonged farrowing is important
in order to preserve subsequent fertility and should be implemented on-farm.
2. Sows with a farrowing longer than 600 min are at risk of having retained placentae
and developing puerperal metritis and therefore may need administration of
exogenous oxytocin after the birth of the last piglet.
3. Obstetrical intervention of the birth canal during farrowing should only be done
after proper obstetrical examination and in a hygienic and gentle manner.
4. Sows receiving obstetrical intervention and/or having more than two stillborn piglets
may also be at risk of developing puerperal metritis and therefore may require
administration of exogenous oxytocin after the birth of the last piglet.
5. A postpartum ultrasound examination of the uterus, e.g. three days after farrowing,
is indicated for those sows at risk of developing puerperal metritis so as to be able
to diagnose and treat sows with puerperal metritis.
86
3. Future research
1. How prolonged farrowing affects subsequent fertility needs further research.
2. How prolonged farrowing affects uterine involution needs clarification. Thus, uterine
involution should be followed-up until the end and compared between sows with
normal and prolonged farrowing.
3. How incomplete uterine involution at weaning and puerperal metritis affect
subsequent fertility needs further attention. For instance, in other species, puerperal
metritis or incomplete uterine involution can interrupt the subsequent pre-ovulatory
LH- surge. It should be investigated if that is the case also in sows and if it could be a
reason for reduced subsequent fertility.
4. Sows with prolonged farrowing had larger follicles during the weaning-to-ovulation
interval. One of the major factors affecting follicular size and growth is lactation
weight loss. Thus, the relationship between farrowing duration and lactation weight
loss should be investigated.
5. Sows with prolonged farrowing, i.e. a farrowing longer than 300 min, may need
further classification. For instance, those sows may need to be further divided into
those with a farrowing of between 300 and 600 min and those with a farrowing
duration longer than 600 min. Because sows with a farrowing longer than 600 min
are at increased risk of retained placentae, it is possible that those sows are at a
higher risk of becoming clinically diseased, whereas sows with a farrowing of
between 300 and 600 min may be only sub-clinically affected. There may be
differences in the effect on subsequent fertility for clinically and subclinically
diseased sows.
87
88
89
References
90
Allen, W.R. (2001). Luteal Deficiency and embryo mortality in the mare. Reprod. Dom.
Anim. 36, 121-131.
Andersson E, Frössling J, Engblom L, Algers B, Gunnarsson S. Impact of litter size on sow
stayability in swedish commercial piglet producing herds. Acta Vet Scand 2016;58: 31.
Athorn, R.Z., Stott, P., Bouwman, E.G., Chen, T.Y., Kennaway, D.J. and Langendijk, P. (2012).
Feeding level and dietary energy source have no effect on embryo survival in gilts, despite
changes in systematic progesterone levels. Anim. Prod. Sci. 53, 30-37.
Bao, B. and Garverick, H.A. (1998). Expression of steroidogenic enzyme and gonadotropin
receptor genes in bovine follicles during ovarian follicular waves: a review. J. Anim. Sci. 76,
1903-1921.
Beg, M.A., Gastal, E.L., Gastal, M.O., Ji, S., Wiltbank, M.C. and Ginther, O.J. (2005).
Changes in steady-state concentrations of messenger ribonucleic acids in luteal tissue
during prostaglandin F2α induced luteolysis in mares. Anim. Reprod. Sci. 90, 273-285.
Bevers, M. M., Willemse, A. H., Kruip, T. A., & Van de Wiel, D. F. M. (1981). Prolactin levels
and the LH-response to synthetic LH-RH in the lactating sow. Animal Reproduction
Science, 4(2), 155-163.
Bostedt, H., Maier, G., Herfen, K., & Hospes, R. (1998). Clinical examinations of gilts with
puerperal septicemia and toxemia. Tierarztliche Praxis. Ausgabe G, Grosstiere/Nutztiere,
26(6), 332-338.
Britt, J. H., Armstrong, J. D., Cox, N. M., & Esbenshade, K. L. (1984). Control of follicular
development during and after lactation in sows. Journal of reproduction and fertility.
Supplement, 33, 37-54.
Busch, B. (2007). Untersuchungen zur postpartalen Involution und Regeneration des
Reproduktionstraktes beim Schwein (Doctoral dissertation, Universitätsbibliothek Giessen).
Castrén, H., Algers, B., De Passillé, A. M., Rushen, J., & Uvnäs-Moberg, K. (1993). Early milk
ejection, prolonged parturition and periparturient oxytocin release in the pig. Animal
Science, 57(03), 465-471.
Conley, A.J. and Ford, S.P. (1989). Direct luteotrophic effect of oestradiol-17 beta on pig
corpora lutea. J. Reprod. Fertil. 87, 125-131.
De Rensis, F., Cosgrove, J. R., & Foxcroft, G. R. (1993). Luteinizing hormone and prolactin
responses to naloxone vary with stage of lactation in the sow. Biology of reproduction,
48(5), 970-976.
Dobson, H., Walker, S. L., Morris, M. J., Routly, J. E., & Smith, R. F. (2008). Why is it getting
more difficult to successfully artificially inseminate dairy cows?.
Edwards SA, Baxter EM. Piglet mortality: causes and prevention. In: Farmer C, editor. The
gestating and lactating sow: Wageningen Academic Publishers; 2015, p. 253-78.
Ellendorff, F., Taverne, M., Elsaesser, F., Forsling, M., Parvizi, N., Naaktgeboren, C., & Smidt,
D. (1979). Endocrinology of parturition in the pig. Animal Reproduction Science, 2(1), 323-
334.
91
Fahmy MH, Friend DW. Factors influencing, and repeatability of the duration of farrowing
in Yorkshire sows. Canadian J Anim Sci. 1981; 61: 17-22.
Fleige, S. and Pfaffl, M.W. (2006). RNA integrity and the effect on the real-time qRT-PCR
performance. JMAM 27, 126-139.
Fellows EJ, Hazzard TM, Kutzler MA. Gene expression in pre-term, pre-labor and parturient
canince placenta. Reprod Dom Anim 2012;47:182-5.
Forni, M., Zannoni, A., Tamanini, C. and Bacci, M.L. (2003). Opposite regulation of clusterin
and LH receptor in the swine corpus luteum during luteolysis. Reprod. Nutr. Dev. 43, 517-
525.
Gilbert, R. O., Shin, S. T., Guard, C. L., & Erb, H. N. (1998). Incidence of endometritis and
effects on reproductive performance of dairy cows. Theriogenology, 49(1), 251.
Hartsock, T. G., & Barczewski, R. A. (1997). Prepartum behavior in swine: effects of pen size.
Journal of animal science, 75(11), 2899-2904.
Hoang-Vu, C. (1987). Elektrische Aktivita ࡇt des Myometriums beim Schwein wa ࡇhrend des
Zyklus und ihre Beeinflussung durch Infusion von O ࡇstrogenen in das Uteruslumen am Tag
der Rausche. Thesis, Hohenheim University (Germany).
Hoving, L.L., Soede, N.M., Feitsma, H. and Kemp, B. (2012). Lactation weight loss in
primiparous sows: Consequences for embryo survival and progesterone and relations with
metabolic profiles. Reprod. Dom. Anim. 47, 1009-1016.
Hoy, S. The impact of puerperal diseases in sows on their fertility and health up to next
farrowing. Animal Science, 2006, 82.05: 701-704.Kot, K., Anderson, L.E., Tsai, S.J., Witbank,
M.C. and Ginther, O.J. (1999). Transvaginal ultrasound-guided biopsy of the corpus luteum
in cattle. Theriogenology. 52, 987-993.
Jensen, P. (1986). Observations on the maternal behaviour of free-ranging domestic pigs.
Applied Animal Behaviour Science, 16(2), 131-142.
Jensen, P., & Redbo, I. (1987). Behaviour during nest leaving in free-ranging domestic pigs.
Applied Animal Behaviour Science, 18(3-4), 355-362.
Jones JET. Observations on parturition in the sow: Part II: the parturient and post-parturient
phases. Br Vet J 1966; 122: 471-8.
Kemp, B., Soede, N. M., & Hazeleger, W. (1998). Control of ovulation.
Kirkwood, R. N., Lapwood, K. R., Smith, W. C., & Anderson, I. L. (1984). Plasma
concentrations of LH, prolactin, oestradiol-17β and progesterone in sows weaned after
lactation for 10 or 35 days. Journal of reproduction and fertility, 70(1), 95-102.
Knox, R. V., & Zas, S. L. (2001). Factors influencing estrus and ovulation in weaned sows as
determined by transrectal ultrasound. Journal of animal science, 79(12), 2957-2963.
Knox, R. V. (2005). Recruitment and selection of ovarian follicles for determination of
ovulation rate in the pig. Domestic animal endocrinology, 29(2), 385-397.
Knox, R. V. (2015). Recent advancements in the hormonal stimulation of ovulation in swine.
92
Kudlac, V. E., & Groch, L. (1979). Morphologische Veränderungen an Ovarien und Uterus
bei Sauen im Puerperium. Reproduction in Domestic Animals, 14(2), 64-72.
Langendijk, P., Bouwman, E. G., Soede, N. M., Taverne, M. A. M., & Kemp, B. (2002).
Myometrial activity around estrus in sows: spontaneous activity and effects of estrogens,
cloprostenol, seminal plasma and clenbuterol. Theriogenology, 57(5), 1563-1577.
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. Theriogenology, 59(3), 849-861.
Langendijk P, Soede NM, Kemp B. Uterine activity, sperm transport, and the role of boar
stimuli around insemination in sows. Theriogenology. 2005; 63: 500-13
Langendijk, P. and Peltoniemi, O. (2013). How does nutrition influence luteal function and
early embryo survival. In ‘Control of pig reproduction IX’. (Eds H. Rodrigues-Martinez, N.M.
Soede and W.L. Flowers) pp. 145-158. (Context Products: Packington, UK.)
Langendijk, P. (2015). Early gestation feeding and management for optimal reproductive
performance. In The gestating and lactating sow (pp. 1145-1153). Wageningen Academic
Publishers.
LeBlanc, S. J. (2008a). Postpartum uterine disease and dairy herd reproductive
performance: a review. The Veterinary Journal, 176(1), 102-114.
LeBlanc, M. M. (2008b). Common peripartum problems in the mare. Journal of Equine
Veterinary Science, 28(11), 709-715.
Linde-Forsberg, C., & Eneroth, A. (2005). Abnormalities in pregnancy, parturition, and the
periparturient period. Textbook of veterinary internal medicine, 7, 1890-1901.
Lucia, Thomaz, Gary D. Dial, and William E. Marsh. "Lifetime reproductive performance in
female pigs having distinct reasons for removal." Livestock Production Science 63.3 (2000):
213-222.
Lucy, M. C., Liu, J., Boyd, C. K., & Bracken, C. J. (2000). Ovarian follicular growth in sows.
Reproduction (Cambridge, England) Supplement, 58, 31-45.
Madej, A., Lang, A., Brandt, Y., Kindahl, H., Madsen, M. T., & Einarsson, S. (2005). Factors
regulating ovarian function in pigs. Domestic animal endocrinology, 29(2), 347-361.
Maes, D., Papadopoulos, G., Cools, A., & Janssens, G. P. J. (2010). Postpartum dysgalactia
in sows: pathophysiology and risk factors. Tierärztliche Praxis. Ausgabe K,
Kleintiere/Heimtiere, 38(1), S15.
Martineau, G. P., Smith, B. B., & Doizé, B. (1992). Pathogenesis, prevention, and treatment
of lactational insufficiency in sows. Veterinary clinics of North America: Food Animal
Practice, 8(3), 661-684.
Meduri, G., Hai, M. V., Jolivet, A., Takemori, S., Kominami, S., Driancourt, M. A., & Milgrom,
E. (1996). Comparison of cellular distribution of LH receptors and steroidogenic enzymes in
the porcine ovary. Journal of endocrinology, 148(3), 435-446.
93
Mee, J. F., Buckley, F., Ryan, D., & Dillon, P. (2009). Pre-breeding Ovaro–Uterine
Ultrasonography and its Relationship with First Service Pregnancy Rate in Seasonal-Calving
Dairy Herds. Reproduction in domestic animals, 44(2), 331-337.
Miller, E.A., Ge, Z., Hedgpeth, V. and Gadsby, J.E. (2003). Steroidogenic responses of pig
corpora lutea to insulin-like growth factor I (IGF-I) throughout the oestrus cycle.
Reproduction. 125, 241-249.
Niswender, G.D., Juengel, J.L., Silva, P.J., Rollyson, M.K. and McIntush, E.W. (2000).
Mechanisms controlling the function and life span of the corpus luteum. Physiol Rev. 80, 1-
29.
Noakes, D. (2009). The puerperium. Veterinary reproduction and obstetrics, ninth edition.
Philadelphia, WB Saunders, 194-205.
Oliviero C, Heinonen M, Valros A, Hälli O, Peltoniemi OAT, 2008. Effect of the environment
on the physiology of the sow during late pregnancy, farrowing and early lactation. Animal
Reproduction Science; 105: 365-377.
Opsomer, G., Gröhn, Y. T., Hertl, J., Coryn, M., Deluyker, H., & de Kruif, A. (2000). Risk factors
for post partum ovarian dysfunction in high producing dairy cows in Belgium: a field study.
Theriogenology, 53(4), 841-857.
Paccamonti D, Pycock J. Infertility and subfertility in the mare. In: Noakes DE, Parkinson TJ,
England GCW, editors. Veterinary Reproduction and obstetrics: Saunders-Elsevier; 2009, p.
583-631.
Palmer, W. M., Teague, H. S., & Venzke, W. G. (1965). Macroscopic observations on the
reproductive tract of the sow during lactation and early postweaning. Journal of animal
science, 24(2), 541-545.
Parkinson T. Infertility and subfertility in the cow: structural and functional abnormalities,
management deficiencies and non-specific infections. In: Noakes DE, Parkinson TJ,
England GCW, editors. Veterinary Reproduction and obstetrics: Saunders-Elsevier; 2009, p.
393-475.
Peltoniemi, O.A.T., Easton, B.G., Love, R.J., Klupiec, C. and Evans, G. (1995). Effect of
chronic treatment with a GnRH agonist (Goserelin) on LH secretion and early pregnancy in
gilts. Anim. Reprod. Sci. 40, 121-133.
Peltoniemi, O. A. T., & Oliviero, C. (2015). Housing, management and environment during
farrowing and early lactation. In The gestating and lactating sow (pp. 77-84). Wageningen
Academic Publishers.
Peter, A. T., & Bosu, W. T. K. (1988). Relationship of uterine infections and folliculogenesis in
dairy cows during early puerperium. Theriogenology, 30(6), 1045-1051.
Peter, A. T., Bosu, W. T., & DeDecker, R. J. (1989). Suppression of preovulatory luteinizing
hormone surges in heifers after intrauterine infusions of Escherichia coli endotoxin.
American journal of veterinary research, 50(3), 368-373.
Randall, G. C. B. (1983). Changes in the concentrations of corticosteroids in the blood of
fetal pigs and their dams during late gestation and labor. Biology of reproduction, 29(5),
1077-1084.
94
Quesnel, H., & Prunier, A. (1995). Endocrine bases of lactational anoestrus in the sow.
Reproduction Nutrition Development, 35(4), 395-414.
Quintal-Franco, J.A., Kojima, F.N., Melvin, E.J., Lindsey, B.R., Zanella, E., Fike, K.E., Wehrman,
M.E., Clopton, D.T. and Kinder, J.E. (1999). Corpus luteum development and function in
cattle with episodic release of luteinizing hormone pulses inhibited in the follicular and
early luteal phases of the estrous cycle. Biol. Reprod. 61, 921-926.
Ribeiro, L.A., Bacci, M.L., Seren, E., Tamanini, C. and Forni, M. (2007). Characterization and
Differential Expression of Vascular Endothelial Growth Factor Isoforms and Receptors in
swine corpus luteum throughout estrous cycle. Mol. Reprod. Dev. 74, 163-171.
Rutherford KMD, Baxter EM, D’Eath RB, Turner SP, Arnott G, Roehe R, Ask B, Sandoe P,
Moustsen VA, Thorup F, Edwards SA, Berg P, Lawrence AB. The welfare implications of
large litter size in the domestic pig I: biological factors. Anim Welf 2013;22:199-218.
Ryan, P. L., & Raeside, J. I. (1991). CYSTIC OVARIAN DEGENERATION IN PIGS-A REVIEW. 1.
Irish Veterinary Journal, 44(1-3), 22-25.
Šavc, M., Kenny, D. A., & Beltman, M. E. (2016a). The effect of parturition induction
treatment on interval to calving, calving ease, postpartum uterine health, and resumption
of ovarian cyclicity in beef heifers. Theriogenology, 85(8), 1415-1420.
Šavc, M., Duane, M., O'Grady, L. E., Somers, J. R., & Beltman, M. E. (2016b). Uterine disease
and its effect on subsequent reproductive performance of dairy cattle: a comparison of
two cow-side diagnostic methods. Theriogenology, 86(8), 1983-1988.
Schroeder, A., Mueller, O., Stocker, S., Salowsky, R., Leiber, M., Gassmann, M., Lightfood, S.,
Menzel, W., Granzow, M. and Ragg, T. (2006). The RIN: an RNA integrity number for
assigning integrity values to RNA measurements. BMC Mol. Biol. 7:3, 1-14.
Shaw, H. J., & Foxcroft, G. R. (1985). Relationships between LH, FSH and prolactin secretion
and reproductive activity in the weaned sow. Journal of reproduction and fertility, 75(1),
17-28.
Sheldon, I. M., Noakes, D. E., Rycroft, A. N., Pfeiffer, D. U., & Dobson, H. (2002). Influence of
uterine bacterial contamination after parturition on ovarian dominant follicle selection
and follicle growth and function in cattle. Reproduction, 123(6), 837-845.
Sheldon, I. M., Lewis, G. S., LeBlanc, S., & Gilbert, R. O. (2006). Defining postpartum uterine
disease in cattle. Theriogenology, 65(8), 1516-1530.
Soede, N. M., Helmond, F. A., & Kemp, B. (1994). Periovulatory profiles of oestradiol, LH and
progesterone in relation to oestrus and embryo mortality in multiparous sows using
transrectal ultrasonography to detect ovulation. Journal of reproduction and fertility,
101(3), 633-641.
Soede, N.M., Langendijk, P. and Kemp, B. (2011). Reproductive cycles in pigs. Anim.
Reprod. Sci. 124 , 251-258.
Soede, N. M., & Kemp, B. (2015). Best practices in the lactating and weaned sow to
optimize reproductive physiology and performance. In The gestating and lactating sow
(pp. 99-106). Wageningen Academic Publishers.
95
Tast, A., Love, R.J., Clarke, I.J. and Evans, G. (2000). Effects of active and passive
gonadotrophin-releasing hormone immunization on recognition and establishment of
pregnancy in pigs. Reprod. Fertil. Dev. 12, 277-282.
Taverne, M., & Noakes, D. E. (2009). Parturition and the care of parturient animals,
including the newborn. Veterinary Reproduction and Obstetrics. 9th edition. Edinburgh,
UK: Saunders (Elsevier), 154-193.
Tokach, M. D., Pettigrew, J. E., Dial, G. D., Wheaton, J. E., Crooker, B. A., & Koketsu, Y.
(1992). Influence of glucose infusions on luteinizing hormone secretion in the energy-
restricted, primiparous, lactating sow. Journal of animal science, 70(7), 2202-2206.
Troedsson, M. H., & Liu, I. K. (1990). Uterine clearance of non-antigenic markers (51Cr) in
response to a bacterial challenge in mares potentially susceptible and resistant to chronic
uterine infections. Journal of reproduction and fertility. Supplement, 44, 283-288.
Tsai, S.J., Kot, K., Ginther, O.J. and Witbank, M.C. (2001). Temporal gene expression in
bovine corpora lutea after treatment with PGF2α based on serial biopsies in vivo.
Reproduction. 121, 905-913.
Tummaruk, P., & Sang-Gassanee, K. (2013). Effect of farrowing duration, parity number and
the type of anti-inflammatory drug on postparturient disorders in sows: a clinical study.
Tropical animal health and production, 45(4), 1071-1077.
Turner, A. I., & Tilbrook, A. J. (2005). Stress, cortisol and reproduction in female pigs. Society
of Reproduction and Fertility supplement, 62, 191-203.
Van Dijk AJ, Van Rens BT, Van der Lende T, Taverne MA. Factors affecting duration of the
expulsive stage of parturition and piglet birth intervals in sows with uncomplicated,
spontaneous farrowings. Theriogenology 2005;64:1573-90.
Van Rens BTTM, Van der Lende T. Parturition in gilts: duration of farrowing, birth intervals
and placenta expuslion in relation to maternal, piglet and placental traits. Theriogenology
2004;62:331-52
Virolainen, J.V., Love, R.J., Tast, A. and Peltoniemi, O.A. (2003). Effects of a gonadotrophin-
releasing hormone antagonist on luteinising hormone secretion and early pregnancy in
gilts. Reprod. Fertil. Dev. 15, 451-459.
Westin, R. (2014). Strategic use of straw at farrowing (Vol. 2014, No. 69).
Whittle, W. L., Holloway, A. C., Lye, S. J., Gibb, W., & Challis, J. R. G. (2000). Prostaglandin
Production at the Onset of Ovine Parturition Is Regulated by Both Estrogen-Independent
and Estrogen-Dependent Pathways 1. Endocrinology, 141(10), 3783-3791.
Wientjes, J.G.M., Soede, N.M., van den Brand, H. and Kemp, B. (2011). Nutritionally
induced relationships between insulin levels during the weaning-to-ovulation interval and
reproductive characteristics in multiparous sows: II. Luteal development, progesterone
and conceptus development and uniformity. Reprod. Dom. Anim. 47, 62-68.
96
97
Original articles