A Review of Reproduction in the Pig

By David Chennells MA Vet MB CertPM MRCVS - The performance of the breeding herd is fundamental to the financial success of any pig farm. If pigs are not born they cannot be sold. Recent surveys by MLC/BPEX indicate that the industry is producing only an average of 18.5 pigs per sow per year (p/s/y) which is woeful compared to many of our European competitors who claim in excess of 22. Even making allowances for our high incidence of PMWS this still points to a serious under performance of our national breeding herd of between 2.5 to 3.5 p/s/y. This is very expensive; merely calculating the number of empty days in an average herd results in an extra cost of £71 per sow per year (J.D Mackinnon). Likewise, increasing the output by 2.5 pigs per year can reduce the cost of producing a piglet by £3.92; which would be worth £32,928 to a 400 sow unit.
calendar icon 11 July 2005
clock icon 14 minute read

There are three main areas where production is assessed. The number of litters produced per sow per year (or conversely the number of non productive "empty" days), the number of piglets produced per litter, and the viability of those piglets. A possible fourth is the lifetime productivity of any female within the herd, as replacements are expensive, and within the UK herd there is a worrying waste of gilts and young sows, far too many of whom do not make their most productive parities (3, 4 and 5)

The main problem for both the producer and the advisor is that reproductive failure is multifactorial involving all aspects of sow and boar (or AI) management throughout the lifetime of the sow. J.D Mackinnon suggested 36 interacting headings in a list that was not exhaustive. This is complicated all the more by (bad?) procedures becoming habit forming and entrenched, old and/or under invested buildings, a reducing and ageing often under-trained workforce; and finally a significant over emphasis of the involvement of infectious disease in reproductive failure. There is also a plethora of information and advice available which is often conflicting, as well as a widespread belief by many throughout the industry that mass medication of sows is the ultimate answer to everything. Or is, to paraphrase Churchill, "mass medication of sows is the refuge of a true scoundrel", a truer picture.

Investigation of reproduction problems in herds also encompasses absolutely my 3 primary rules of pig production.

  • Don't mess about with what is working.
  • No two pig farms are the same.
  • Most "diseases" are management related.
Whole books are written on this subject but I will attempt to highlight some pertinent areas.

Investigation of Records

This is the fundamental place to start, but remember that the records are only as good as the recorder, and the recorder only records what he sees. The vast majority of events go unrecorded in so much as only "outcomes" tend to be recorded rather than true details. For instance, it is common find units where first returns are low, 12%, but farrowing rates are poor, 76%, ie another 12% of served "non returned" sows do not farrow often with little recorded information. These contribute significantly to empty days. It is also very important to look back over many years of records to compare the herd historically. This will indicate consistencies or sudden changes and any trends of seasonal infertility. If a herd has suffered a low farrowing rate and numbers born for years, and which has always been worse from summer servings then this will indicate two things. Summer infertility problem and that your chances of making a significant change to improve things are low. It also would suggest that a sudden introduction of a "new" disease is unlikely!

Producers can often be stimulated into immediate concern by one "bad week", a visit to the bank manager or passing comments by company representatives, when the real trends have been evident for years.

Remember that any investigation of infertility will usually only involve 10 to 15% of the herd as in most cases 75% will conceive and farrow without a problem and it is unreasonable to consistently expect a farrowing rate of more than 90%.

Feeding the Sow

The concept of lifetime care of nutrition of the sow is fundamental, from the time of gilt selection until culling after at least eight parities. There are no stages at which attention to detail should lapse. However, this detail should be simple and straight forward as too many complexities or changes will result in more mistakes or misunderstandings by stockmen. To quote, "complex is clever but simple is safer" (Whittemore 1998).


Of all the stages lactation is probably the most important. I believe that it is essential to feed a lactator diet (14MJ/kg and 1% lysine). This should be built up steadily by no more than 0.5kg per feed and depending on the sow and the size of a litter can be virtually ad lib (ie to full appetite) from as early as day five and at least by day ten of lactation. In effect, once full milk production is underway you cannot feed too much during lactation. With large litters sows will inevitably lose weight, and this loss should be minimal. The ovarian follicles for the next pregnancy will be maturing during late lactation at a time when litter demand is greatest, so this period becomes more important. To maintain appetite food should be replenished frequently, bowls kept clean, stale food removed, water should be clean, fresh and readily available, as restriction of water intake will dramatically reduce appetite, and ambient temperatures around 16-18ºC; (too hot will depress intake and too cold increases maintenance demands).

Remember that the metabolic rate of lactating sows is very high so they will soon become too hot in higher ambient temperatures. This can be indicated by sows splashing water onto floors to keep cool. Also, excessive intakes during the dry period can depress lactation appetite, and conversely if they farrow thin they will always be thin.

Weaning to Service

It is essential to maintain intake during this period by feeding a lactator diet again to appetite (at least twice maintenance level). Remember that it is the removal of the litter that initiates involution of the mammary glands and the start of the next oestrus cycle, not the reduction of food intake. There is strong evidence that good nutrition, particularly of high fibre diets, at this stage improves the maturity of the oocytes, the number of oocytes released, and the subsequent early viability of the embryos produced. This ultimately is reflected in both the number of piglets produced and the quality of those piglets. There is a proven relationship between food intake during lactation and subsequent average litter weight at birth (White 1996). It appears that the size of a piglet at birth is determined at this stage, and not their position in the uterus as is often assumed (Ashworth). This effect is believed to be by the influences on the hormones within the sow as well as purely the available nutrients. These "underweight" piglets or "runts" can be identified at day 30 of pregnancy and always remain smaller. These have a far higher loss rate after birth and subsequently grow slower, mature at a lower bodyweight and carry more body fat.

Post Service Nutrition

At service sows naturally reduce their intake dramatically and can lose a surprising amount of weight. Post service sows should revert to a dry sow ration at maintenance level (around 2.2 - 2.5kg) for at least 3 days (White 1996). Thereafter there is some debate as to what is the "safe" level to feed. Most will agree that in essence sows should be fed to condition, but that excessive levels should be avoided and that increase should be gradual.

During the dry period sows should be fed to condition to achieve a condition score of 3.5 (out of 5) at farrowing. It is vital to remember that this is unequivocally the stockman's job, not that of the computers or the "rule book". Ideally this should be achieved during the middle trimester of pregnancy. There is strong evidence that increasing feed levels by both using increased quantities and/or higher density diets during the final trimester will increase litter weight and mammogenisis. However, this must be carefully monitored as over fat sows can have problems at parturition and there are suggestions that overfeeding can be related to problems with periparturient mastitis, metritis and agalactia; although these problems do appear to be very farm specific.

Problems with sow condition at farrowing are almost universally related to poor observation, monitoring, and attention to detail by the stockmen during the dry period. This is far more demanding in the loose housed systems employed in the UK compared with the stall systems allowed in other countries.

The gilt

Good nutritional preparation of the gilt from selection to first farrowing is essential so that they are mature, well grown, and in good body condition at service. This will produce a good conception rate and litter numbers. Maintenance of growth with a high quality (ideally gilt breeder ration) will produce good mammogenisis and a well fleshed animal ready to withstand a period of lactation. Far too many gilts in the UK are weaned in poor condition and do not conceive or have poor second litters as a result. This is an inexcusable waste of breeding potential. Producers should be prepared to skip a heat, or better, breed from more mature gilts in the first place to avoid these problems. Also excessive culling (loss) of second parity sows produces a disrupted herd parity pattern.

Timing of Service

This is critical to success. It depends on observation, observation, observation! It is essential to induce sows to a strong standing oestrus and to observe this oestrus from its onset. Most sows will start to show oestrus around the fourth day after weaning and it will generally last 36 - 72 hours, but some will delay onset by a further 3 days or more. There is a direct negative correlation between the weaning to oestrus interval (WOI) and subsequent conception rates and litter size (ie the longer WOI the poorer the results). So it is essential to both stimulate the sows well in this period and to maintain their comfort, by feeding (see above), boar contact, thermal and environmental comfort lighting (16 hours of good light), and suitable group size to avoid bullying. The later that oestrus begins the shorter the oestral period lasts and usually in all cases ovulation occurs two thirds of the way through oestrus. Good stimulation of the sow results in a shorter ovulation period, usually this takes 4-8 hours. Unfertilized ova deteriorate very rapidly, within hours, and there is a marked fall off of successful fertilization after 8 hours from ovulation.

The pertinent points for successful service are (after Soede):

  • Oestrus occurs on average 5 days post weaning.
  • On average ovulation occurs 24 - 48 hours after the onset of oestrus
  • Ovulation occurs 2/3rd of the way through oestrus.
  • Fertilization takes place in the oviduct.
  • The shorter the duration of ovulation the better the results.
  • Sperm require at least 2 hours for capacitation and up to 8 hours to reach the oviduct.
  • Sufficient sperm will survive for more than 24 hours and some for several days in the uterus.
  • Optimum time of insemination is 8 hours before ovulation.
  • Employ two inseminations at a 24 hour interval.
  • There is no advantage in inseminations at less than a 24 hour interval.
  • The best results are likely from two inseminations at 12 and 36 hours after the onset of oestrus.
  • It is essential to start detecting oestrus from day 3 after weaning.
  • It is essential to detect oestrus twice daily.
  • Stress factors in the sow can delay and diminish all these processes.
  • Oestrus onset and duration varies between farms.
  • Oestrus duration varies between sows.
  • There is no such thing as the ideal programme.
When using AI, remember also particularly.
  • It is essential to have boar contact to detect oestrus
  • Intermittent boar contact is far better to detect and stimulate oestrus
  • It is essential to have boar contact during insemination
  • Nose to nose or body contact provides the greatest stimulus.
  • Contractions within the uterus are required to achieve sperm transport to the oviduct.
  • Sight, sound, smell and contact with a boar produce the best stimulation for this.
It can be seen from the above that on problem farms it is essential to know exactly the time of onset and duration of oestrus to analyse results. You can now see why I stated earlier that most events go unrecorded. Vital to good performance on natural service farms is the management of boars in prime physical condition and being lean and fit. Boar usage should be recorded (ideally two and a maximum of three serves per week) and to avoid over use of the "easy" boar. For farms using AI good collection, processing and transport are vital. Likewise, correct storage, stock control and hygiene on the farm of use are essential, and commonly are found to be deficient (Donadeau). Perfectly inseminated dead sperm serve no purpose!

All of the detail procedures above apply to finding and serving 21 day returns. It is totally inadequate to rely on chaser boars for this, as live-in boars do not stimulate sows to the same degree as "fresh" boars. I think that this is a major failing of yarded systems. Likewise, all sows should be actively "pregnancy diagnosed" around 21-35 days (depending on what equipment is available). It has been shown that sows which are empty at 21 days may not show oestrus unless stimulated by a boar, and will remain anoestrus until 28 days.

Maintenance of Pregnancy

During the period of 12-20days of pregnancy implantation (or attachment) of the embryos usually occur, although it can take until the 28th day. It is in this period that most embryos are lost. If all are lost, or insufficient survive, then the sow will return to oestrus at an "irregular" interval. It is during this first 28 days of pregnancy, and particularly between the 7th - 21st days that all stresses, physical and physiological should be avoided such as mixing, moving, dramatic nutritional changes, sunburn etc. This is why the EU will continue to allow the use of sow stalls for four weeks post service.

Disease (The alleged real cause of infertility!)

I do not wish to produce an exhaustive list of diseases causing infertility in the pig herd, rather to look at how one may approach a diagnosis.

Diseases which are known to cause reproductive disease.

  • Porcine parvovirus
  • Porcine reproductive and respiratory syndrome virus (PRRS)
  • Other SMEDI viruses - Aujeszky's disease (Not GB) - Swine fever(s) (Not UK) - Porcine enteroviruses - Swine influenza viruses
  • Erysipelas
  • Brucella suis (Not UK)
  • Some Leptospiral diseases (mozdok, icterohaemorrhagiae)
Diseases strongly associated with reproductive disease.
  • Leptospirosis; L. bratislava
  • Chlamydiosis (?UK)
    • Other organisms involved with reproductive disease.
  • Those causing endometrial/vaginal/urinary infections
  • Porcine circovirus type 2 (PCV2)

A Word on PCV2

There is a wealth of evidence to associate this organism with reproductive failure (Done), and it does seem logical to assume that an organism which is associated with devastating disease in the growing pig (PMWS) should also affect foetal tissues. However, the true link between this organism and the pathogenesis of PMWS remains the subject of fierce debate, and so also should its method of involvement in reproductive disease.

The diagnosis of an infectious cause of a fertility problem requires a complete history, reliable records, clinical signs, a range of correct diagnostic samples and tests, sufficient results for adequate interpretation and basic clinical and practical experience (Done et al).

Consequently the collection of fresh and representative samples is essential. The whole of an aborted or stillborn litter is required to increase the chance of a representative sample being collected. Also collect sow serum to relate to any findings. It should be noted that only 25% of porcine samples submitted to the VLA produce a diagnosis (Williamson). This is due to a combination of poor sample selection, the fragility of many organisms, poor supportive history, but probably the majority because of a lack of an infectious cause!

A Word on Serology

Serology is frequently used to "diagnose" fertility problems in sows. Undoubtedly it has its uses in diagnosing the presence or absence of some conditions. However, frequently dramatically inadequate numbers are sampled. For instance, to estimate a 10% prevalence of a disease in a 400 sow herd with 95% (+/- 5%) confidence would require 103 samples (Williamson). Likewise, many organisms causing abortion/embryonic loss, ie parvovirus, need a finite "incubation" period to cross the placenta, so samples taken at the time will be too late for paired serology as the "baseline" was 4 weeks earlier (Bidewell). Only paired samples can give a true indication of serological change relative to disease incidence. Often samples are taken historically and are "pre selected" ie returning sows. Doing this slants the results and does not offer a comparison with the unaffected herd which may have a similar pattern. Many titres ie Leptospiral IgM's decline very rapidly and can give false negative or inconclusively low results. Also different tests for the same organism may have different specificity and sensitivity and are selected for different antigens or different types of antibody so results may not be at all comparable. The only conclusion is 'interpret with care' and at your peril, and the fewer samples and the more slanted the target animals the greater the peril. Essentially, test only to support your clinical diagnosis.

So good luck in solving reproductive problems and remember my 3 rules!


This article is based on papers presented to the "John Walton Day" of the Pig Veterinary Society 13th May 2005 by Nicoline Soede, Wagening University, Holland; Cheryl Ashworth, SAC, Aberdeen; Meritxell Donadeau, PIC UK; Susanna Williamson, Cornelia Bidewell and Stan Done, VLA and John Mackinnon, Pig Health and Production Consultant.


White M.E.C (1996) Reproductive physiology of the pig - theory into practice. In Practice, Vol 18, No3. p108-114 Whittemore C (1998) Feeding requirements for breeding sows: complex is clever, but simple is safer. In Practice, Vol 20, No6. p304-307

Source: UK Pig Veterinary Society - July 2005

A version of this article was originally published in Vet Times, 2005

© 2000 - 2023 - Global Ag Media. All Rights Reserved | No part of this site may be reproduced without permission.