Pig production farms worldwide strive to optimise their operations and thereby maximise their profits. Along with the overall increase in pork production during the last 20 years, there has been an increase in larger farms with 1000 sows or more and a decrease in the actual number of pig farms. This trend has been particularly evident in eastern and western Europe, North and South America and Australia (Cutler and Holyoake, 2007; Augère-Granier, 2020).
Large production units require high-level coordination of farrowing management to ensure good animal welfare and profitability. Farrowing is a stressful event in the production life of a sow, and it has a substantial impact on piglet production (Mainau and Manteca, 2011; Rootwelt et al., 2012; Peltoniemi and Oliviero, 2015). Piglet production aims towards low stillbirth rate and low pre-weaning mortality. Despite this, piglet mortality estimates in cross-sectional studies range from 10 to 20% (Friendship et al., 1986; Glastonbury, 1976; Koketsu et al., 2006; Tubbs et al., 1993). Main causes of piglet mortality are stillbirth, crushing, starvation and disease which typically occur during the first days of life (Pedersen et al., 2011; Kielland et al. 2018). Reduction in piglet mortality requires coordinated genetic, nutritional, management and stockperson interventions (Edwards and Baxter, 2015). Proper farrowing supervision allows manual assistance for sows exhibiting dystocia, and treatment with oxytocin to help overcome cases of primary and secondary uterine inertia and may also reduce the number of stillbirths (Vanderhaeghe et al., 2013). Additionally, neonatal care, such as providing dry and clean straw, drying and warming, clearing airway and ensuring colostrum intake will help improve chances for survival (Holyoake et al., 1995; White et al., 1996; Kirkden et al., 2013a; Rosvold et al., 2017).
Manual supervision of farrowing is time consuming, and as the gestation length in the sow can range from 105 to 125 days, parturition is difficult to predict in practice (Sasaki and Koketsu, 2007). Today farmers rely on behavioural change prior to farrowing, such as nest-building activity and mammary development, for prediction of the farrowing initiation timepoint. Detection of nest-building activity requires a close monitoring of the sows, which is not always feasible for the farmers. The induction of farrowing by administering the natural hormone prostaglandin F2 alpha (PGF 2 alpha), or a synthetic analogue such as cloprostenol prior to the expected date of farrowing, is recommended in some countries to facilitate farrowing supervision (Sprecher et al., 1974; Herpin et al., 1996; Lawlor and Lynch, 2005). In addition, oxytocin is commonly used as an obstetric intervention to reduce farrowing duration in sows (Kirkden et al., 2013a). However, human intervention by means of exogenous hormones during parturition includes important risks, such as decreased piglet viability, increased risk of dystocia, and reduced sow welfare (Kirkden et al., 2013b; Peltoniemi and Oliviero, 2015).
The maintenance of trained staff for the supervision of farrowing on constant monitoring to determine the onset of parturition is expensive (Kirkden et al., 2013b). Therefore, detection of sows with impending farrowing is crucial for the efficient allocation of human resources in the farrowing unit. Production units should focus on innovative solutions to deal with welfare challenges around farrowing. Some methods have been suggested for this purpose, including the automatic monitoring of activity level during farrowing (Cornou and Lundbye-Christensen, 2012; Erez and Hartsock, 1990; Oliviero et al., 2008) or behavioural patterns (Aparna et al., 2014; Holmqvist, 2012; Khoramshahi et al., 2013). These systems seem to be reliable, but suggestions state that they could be improved if combined with other variables that could indicate imminent parturition (Oliviero et al., 2008). Accordingly, rectal thermography (Damgaard et al., 2009; Hendrix et al., 1978; King et al., 1972; Williams et al., 2013) and telemetry (Bressers et al., 1994; Elmore et al., 1979) data indicate that the sow’s body temperature increase just before farrowing, suggesting a possible application for estimation of time of parturition. However, these thermography techniques are to a certain extent invasive and/or require constant physical monitoring, which could negatively affect the course of farrowing.
The sows’ temperature rise before farrowing could be due to a combination of hormonal changes responsible for the initiation of parturition and the stress of the birth process. Kelley and Curtis (1978) describe an increase in energy-expenditure rate (EER, i.e the sum of the body’s heat production and work on environment) during the prepartal period due to elevated activity levels related to restlessness and nest-building. An increase in prostaglandin (i.e. PGE2 and PGF2α) secretion is an essential step in the induction of parturition (Challis et al., 2000). Indeed, in the pig, a substantial increase in blood concentrations of PGF2α takes place approximately 24 hours before birth, reaching maximum values during parturition (Gilbert et al., 2002; Gooneratne et al., 1983; Guthrie et al., 1987; Watts et al., 1988; Whitely et al., 1990). Experimental evidence indicates that prostaglandins, including PGF2α, can increase body temperature in prepubertal pigs (Feldberg and Saxena, 1975; Parrott and LLoyd, 1995). Parturition is a natural stressful event in many placental mammals. Studies in women indicate that parturition is a source of discomfort that can generate high levels of physical pain and hardship (Garthus-Niegel et al., 2014; Simkin, 2011). A similar scenario is present in domestic animals, including sows (Mainau and Manteca, 2011), and it is known that physical stressors can induce prostaglandin-mediated (i.e. PGE2) hyperthermia in pigs (Parrott and LLoyd, 1995).
Infrared thermography is a non-contact and non-invasive technique used to obtain body surface temperature (Ring and Ammer, 2012). This thermography technique is already applied in veterinary medicine as an ancillary tool for the estimation of disease occurrence (i.e. lameness, mastitis) (Turner, 2001; Kunc and Knizkova, 2012; Schaefer and Cook, 2013; Stelletta et al., 2012). However, the predictive usefulness of infrared thermography for female reproductive events (e.g. pregnancy, ovulation) is less clear (Durrant et al., 2006; Luño et al., 2013; Scolari, 2010; Stelletta et al., 2012; Sykes et al., 2012; Talukder et al., 2014, Weng et al., 2020).
The aim of this study was to determine surface and core temperature of sows in the period before, at, and after farrowing, using infrared and rectal thermography; and investigate the possibility of using infrared thermography to estimate impending parturition in sows, and thereby improve farrowing supervision.