Profitable pork production can be achieved by not only enhancing the breeding efficiency but also by accurate detection of pregnancy for rebreeding the non-pregnant sows and/or culling the non-performing females. Therefore, a reliable and accurate pregnancy diagnosis method post-service is the need of the hour in swine production. The aim of the current study was to determine if measurements of salivary progesterone, alternative to serum progesterone, could be used for pregnancy diagnosis in sows. Progesterone levels in both saliva and serum samples of day 10 and day 21 post-service and non-pregnant animals were estimated using a competitive enzyme immunoassay. Both serum and salivary progesterone levels among non-pregnant and pregnant (day 10 and 21 post-service) sows were significantly different (p<0.0001). The progesterone concentration in saliva on day 21 post-service was found to be more than 2 ng/mL. Moreover, a significant difference in ratio of saliva:serum progesterone levels was observed between non-pregnant and pregnant (day 10 and 21) animals. The findings of the present study support the hypothesis of pregnancy diagnosis based on salivary progesterone levels in sows as revealed by the course of salivary levels of progesterone in early gestation period. This study forms an essential step in diagnosing pregnancy status based on progesterone levels in saliva instead of serum and could be regarded as a pilot reference study, results of which could benefit similar studies in the future.
Ensuring the pregnancy status of a female farm animal at the earliest after mating has become the top priority for profitable pig production. Early identification of non-pregnant animals for rebreeding or removal from the herd can reduce non-productive days, thus can optimize the profit. Consequently, the development and implementation of highly effective pregnancy diagnostic methods is an important component of an efficient reproductive management program. Traditionally, non-return to oestrus after 18–22 days of service has been regarded as a sign of pregnancy. However, some times the detection of oestrus can be difficult and time consuming and failure to return to oestrus may be due to showing no signs of oestrus, anoestrus or ovarian cysts. Later application of ultrasonic methods like Doppler, A-mode ultrasound, B mode or real time ultrasonography (RTU) have been introduced in swine pregnancy diagnosis (Williams et al., 2008; Stančić et al., 2012) but the major limitation with ultrasonography is variable detection of non-pregnant animals. Though this method is performed sow-side, the apparatus can be expensive and the degree of experience of the operator may greatly affect the results. Pregnancy diagnosis by determination of serum progesterone concentration revealed that a >5 ng/ml level of progesterone in the serum, collected during 17-22 days after mating, was considered an indication of pregnancy (Krüger and Bilkei, 2002; Boma and Bilkei, 2008). However, due to lack of superficial blood vessels, repeated blood sampling is extremely invasive in pigs (Bushong et al., 2000). Though a non-invasive method, collection of urine is very difficult in pigs and it cannot be performed from the animals of a single experimental group at any one time. As collection of blood and urine samples is technically difficult in pigs, saliva offers an alternate sample to estimate hormones like progesterone during early pregnancy without causing stress to the animals. Suthanthirakannan and Rameshkumar (2014) reported higher salivary progesterone levels during pregnancy period in bovine when compared to prepubertal, estrous and lactation stages. Considering the fact that the level of progesterone is maintained throughout gestation and progesterone enters the saliva through passive diffusion from the salivary glands, in the present study, estimation of salivary progesterone levels was carried out to determine the pregnancy status in non-pregnant animals and sows during early post-service period.
Materials and Methods
Animals and Sample Collection
In the present study, twelve Large White Yorkshire (LWY) sows were selected from the instructional livestock farm complex and pig farm of all India coordinated research project (AICRP) on pigs of the College of Veterinary Sciences and Animal Husbandry, Central Agricultural University, Aizawl, India. Blood and saliva samples were collected from sows (6 LWY) on day 10 and day 21 post service and also from non-pregnant sows (6 LWY; on day 5 post-weaning) serving as controls. The collection of saliva samples was carried out using cotton swabs. Both saliva and serum samples from sows of day 10 and day 21 post-service were considered for further studies only after their farrowing. All the animal experiments were carried out strictly as per the guidelines issued by Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Ministry of Environment, Forests & Climate change, Government of India and was approved by Institute Animals Ethics Committee (IAEC) of College of Veterinary Sciences and Animal Husbandry, Central Agricultural University, Aizawl, India.
Progesterone Estimation by Enzyme Immunoassay
Progesterone levels in both saliva and serum samples of non-pregnant, day 10 post-service and day 21 post-service sows were estimated using a commercial competitive enzyme immunoassay kit (Enzo Life Sciences, Farmingdale, NY, USA) according to the manufacturer’s instructions. After performing enzyme immunoassay, a standard curve was produced automatically by the microtitre plate reader software program. This allowed the progesterone concentration of the unknown samples to be read off from the standard curve.
Results were expressed as the means ± SEM. A difference with value p<0.05 was considered statistically significant. Data were analyzed by univariate analysis of variance and statistical differences between various treatment group means were determined by Duncan’s multiple range test (DMRT) using the Statistical Product and Service Solutions, Version 17.0.1 software (SPSS Inc., Chicago, IL, USA).
Results and Discussion
The serum progesterone levels in non-pregnant, day 10 post-service and day 21 post-service sows were 924.72±14.51, 2677.48±16.48 and 5374.54±23.38 pg/mL respectively. Whereas the salivary progesterone levels in non-pregnant, day 10 post-service and day 21 post-service sows were 479.1±2.56, 1054.66±7.61 and 2326.81±88.17 pg/mL respectively. Both serum and salivary progesterone levels among non-pregnant, day 10 post-service and day 21 post-service sows were significantly different (p<0.0001) (Fig.1). The serum and saliva concentration of progesterone are high (>5ng/mL and >2ng/mL respectively) in pregnant and low (<1ng/mL) in non-pregnant sows. The serum progesterone levels of sows on day 21 post-breeding were more than 5 ng/mL is in agreement with the earlier findings (Krüger and Bilkei, 2002; Boma and Bilkei, 2008). Interestingly, the corresponding levels of progesterone in saliva on day 21 post service were found to be more than 2 ng/mL.
Fig.1: Progesterone levels in serum and saliva of non-pregnant (NP), day 10 post-service and day 21 post-service sows. Means with different letters significantly different (p<0.05)
Our results corroborate the fact that, like most of the steroids, progesterone enters the saliva through passive diffusion from the salivary glands and the level of progesterone is maintained throughout gestation (Gröschl, 2008). Analysis of difference in ratio of saliva: serum progesterone levels among non-pregnant, day 10 post-service and day 21 post-service sows revealed a significant difference (p<0.01) between non-pregnant (51.85%) and pregnant [day 10 (39.36%) and 21 (42.41%)] animals (Fig. 2).
Fig.2: Saliva to serum ratio (%) of progesterone levels among non-pregnant (NP), day 10 post-service and day 21 post-service sows. Means with different letters significantly different (p<0.05)
In India, pregnancy diagnosis in pigs was carried out by estimating estrone sulphate in sow urine by cuboni test (Lalrintluanga, 2012). However, the accuracy of the test for pregnancy diagnosis from pregnant sow urine samples was found to be only 74% in the early stage of pregnancy. Barium chloride test was also performed for pregnancy diagnosis in pigs, but the results were not encouraging as the accuracy of the test for pregnancy diagnosis was only 64% (Lalrintluanga and Dutta, 2009). As animal welfare is of great importance in animal production systems, collection of saliva offers a non-invasive sampling method that causes minimal discomfort to the animal (Schönreiter and Zanella, 2000). Saliva collection is more straightforward, economical and can be performed by the unskilled personnel. As far as the stability of the sample is concerned, saliva is stable at room temperatures for several weeks (Hofman, 2001). Moreover, repeated sampling of saliva is possible without causing stress to the animal and could be especially useful in pigs where blood sampling is technically difficult (Gutiérrez et al., 2012; Ramirez et al., 2012). Importantly, the estimation of salivary steroid hormones presents an attractive alternative to serum as the steroids are not protein bound in saliva.
In conclusion, the course of salivary levels of progesterone in early gestation period supports the hypothesis of pregnancy diagnosis based on salivary progesterone levels in sows. This gives an opportunity to measure progesterone levels of saliva to determine whether the animal is pregnant or not, as collection of blood samples is technically difficult in pigs. However additional research is warranted in order to statistically assess the findings of this study by including more number of sows. This can be followed by the development of a reliable and accurate early pen-side pregnancy diagnosis test in sows based on salivary progesterone levels which would enable the prompt rebreeding of non-pregnant animals without missing any estrus cycle.
This study was carried out under Central Agricultural University funded Intramural Research Project (CAU-DR/3-2(Vety)/2010/Vol.III/2471). The authors would like to thank the Vice-chancellor, Central Agricultural University, Iroisemba, Imphal, India and the Dean, College of Veterinary Sciences and Animal Husbandry, Central Agricultural University, Selesih, Aizawl, Mizoram, India for providing essential support to conduct the research work.