Technical note: development of an indirect calorimetry system to determine heat production in individual lactating sows

Abstract

The ability to determine total heat production (THP) in individual sows and litters can be logistically difficult and often requires the use of multiple animals to generate data on a per room basis. Furthermore, these systems may be costly to construct, precluding their use by many researchers. Therefore, the objective was to develop a low-cost indirect calorimetry system to determine THP in individual lactating sows and litters. Six indirect calorimeters were constructed to house 1 sow and litter in a crate throughout farrowing and a 21-d lactation period. Farrowing crates were placed within a high-density polyethylene pan filled with water and then a polyvinyl chloride frame was constructed around the crate. The frame provided a structure to hold the inlet and outlet air pipes, feed and water inlets, air circulation fans, and a polyethylene plastic sheet that was secured at the bottom of the frame and submerged under water to maintain an air tight seal. Chamber accuracies for O₂ and CO₂ were evaluated by ethanol combustion. One week pre-farrowing, 6 pregnant multiparous sows (parity 2.9 ± 0.9; 218.3 ± 38.6 kg BW) were housed individually in each farrowing crate and the calorimeters were maintained at thermoneutral conditions (20.9 ± 2.6°C and 43.7 ± 18.6% relative humidity) throughout lactation. On lactation day 4, 8, 14, and 18, indirect calorimetry was performed on all sows and their litters, as well as 2 piglets from a sentinel litter to determine THP and the respiratory quotient (RQ). Sentinel piglet data were used to estimate THP and RQ for the sows independent of the litter. Sow + litter THP (kcal/h) increased (P = 0.01; 16.6%) on day 8 compared to day 4 and was greater (27.3%) on day 14 and day 18 compared to day 4 and day 8. Sow THP was greater (P = 0.01) on day 8 (401.19 ± 17.15 kcal/h) and day 14 (430.79 ± 12.42 kcal/h) compared to day 4 (346.16 ± 16.62 kcal/h), and was greater on day 14 compared to day 8, and on day 18 (386.16 ± 20.02 kcal/h) compared to day 14. No sow + litter RQ differences (P = 0.21; 1.02 ± 0.04) were detected by day of lactation. However, sow RQ was reduced (P = 0.01) on day 14 (0.98 ± 0.02) compared to day 4 (1.03 ± 0.03), day 8 (1.02 ± 0.02), and day 18 (1.04 ± 0.03). In summary, this cost-effective system (total cost: $1,892 USD) can allow researchers to accurately evaluate THP in individual lactating sows and their litters.

INTRODUCTION

Indirect calorimetry is used to measure heat production in living organisms and depends on the measurement of energy exchange taking place within the animal’s living tissues (Nienaber et al., 2009). It is a valuable tool for researchers to measure energy balance (Labussière et al., 2015) and understand the thermodynamics of homeothermic animals to their environments (Nienaber et al., 2009; Stinn and Xin, 2014). However, many systems are costly and determining total heat production (THP) in lactating sows and litters often requires the use of multiple animals in a room to generate data for the group rather than the individual (Bond et al., 1959; Brown-Brandl et al., 2014; Stinn and Xin, 2014), precluding determination of animal to animal variability. In addition, methodological difficulties may be present that reduce the ability of researchers to determine THP in individual lactating sows independent of the litter (Theil et al., 2003; Jakobsen et al., 2005). Therefore, the development of a cost-effective system to determine metabolic heat production at the individual lactating sow level may be an important research tool.

Improved genetic potential has increased swine productivity resulting in greater THP in modern sows (Stinn and Xin, 2014). Consequently, modern pigs have likely become more heat stress sensitive, especially when heat stress is combined with physiological states such as lactation (Black et al., 1993; Cabezon et al., 2017a). This is because lactation increases THP in sows primarily due to milk production and greater feed intake placing them at increased risk for heat stress (Quiniou and Noblet, 1999). Therefore, the need to measure THP in lactating sows can be important when evaluating feed additives to reduce the negative effects of heat stress or to determine thermodynamics for building design and environmental management systems (i.e., ventilation). Methods to evaluate the environmental impact on THP in lactating sows are often limited to developing estimation models (Cabezon et al., 2017b) or measuring at the room or group level rather than in individual animals (Bond et al., 1959; Brown-Brandl et al., 2014; Stinn and Xin, 2014) due to animal husbandry practices and indirect calorimeter design. Therefore, the study objective was to develop and test a cost-effective indirect calorimetry system capable of measuring THP in individual lactating sows and their litters and separating out the sow from the litter THP. We hypothesized that this cost-effective system would accurately measure THP in individual lactating sows and their litters as compared with previous literature.

MATERIALS AND METHODS

Indirect Calorimeter Design

Calorimeter frame.

Six identical positive pressure indirect calorimeters were constructed and tested within an environmentally controlled room at the USDA-ARS Livestock Behavior Research Unit Food Animal Behavior Laboratory in West Lafayette, IN. Each calorimeter was designed to house 1 individual sow and her litter within a crate (2.44 m × 1.52 m × 1.50 m [L × W × H]) throughout farrowing and a 21-d lactation period. Each farrowing crate was placed within a 3.05 m × 2.13 m × 0.15 m (L × W × H) custom pan (Shape Master Plastic Forming, Ogden, IL) constructed out of 0.635 cm thick high-density polyethylene (HDPE; Figure 1). A frame (2.74 m × 1.83 m × 1.52 m) constructed from 3.81 cm diameter polyvinyl chloride (PVC) pipes was then erected around each farrowing crate leaving 15.24 cm of space between the bottom of the frame and the farrowing crate and the edge of the HDPE pan, respectively (Figures 1 and 2). Centered on the top of each end of the frame, a 0.56 cm thick acrylic sheet (0.91 m × 0.38 m; Shape Products; model #220-3036) was attached to the PVC pipe using zip ties (Figure 2). Holes were cut into each acrylic piece using a Dremel drill (Dremel; model #4000-2/30) to accommodate air inlet and outlet pipes (Figures 2 and 3), air circulation fans (Multifan S3, 120 mm, 52 CFM; AC Infinity, City of Industry, CA), a water line, and a PVC pipe with a sealable cap to deliver feed while the calorimeter was sealed (Figures 1 and 2). Rubber gaskets (Danco rubber gasket sheet; model #9D00059849) were secured around each hole to seal the calorimeter when the pipes were secured in place. The feeding pipe was constructed out of a 5.1 cm PVC shower drain body (Sioux Chief; model #825-L20P), a 0.31 m section of 5.1 cm diameter PVC pipe, a 5.1 cm diameter male PVC adapter (NIBCO; model #K03070C), and a 5.1 cm diameter threaded PVC cap (NIBCO; model #F01170D). Two air circulation fans were attached to 22.86 cm × 38.10 cm sections of acrylic and secured to the sides of the PVC frame using zip ties. Finally, a 5.79 m × 4.80 m section of 6 mil polyethylene sheeting (Polar Plastics 6-Mil Clear Rolled Poly All-Purpose Plastic; model #5680075) with grommets in place at the bottom was draped over the entire structure, sealed at the pipe inlets using the rubber gaskets, and secured to the bottom of the PVC frame using brass hooks (3.81 cm; National Hardware; model #N119-727; Figures 1 and 2).

Figure 1.

Image of constructed calorimeters.

Figure 2.

Schematic of indirect calorimeter frame.

Figure 3.

Schematic of calorimeter inlet and outlet air pipes.

Inlet and outlet air pipes.

The air inlet pipe was responsible for directing airflow into the chamber at a constant rate and provided a source to collect air samples entering the chamber (Figure 3). It was constructed with the following material: two 0.15 m long sections of 10.2 cm diameter PVC pipe, two 0.15 m long sections of 7.6 cm diameter PVC pipe, a radon pump fan (Tjernlund PVC4 Radon Mitigation Fan; 90 CFM; 10.2 cm diameter; model #PVC4), a 7.6 cm diameter PVC Tee (NIBCO 7.6 cm sanitary PVC Tee DWV; model #K09870C), a 7.6 cm PVC gate valve (Valterra PVC gate valve; model #6301) to adjust airflow into the chamber, a 10.2 cm diameter male PVC adapter (NIBCO 10.2 cm PVC male adapter; model #K03100C), a 7.6 cm 90-degree PVC elbow (NIBCO 7.6 cm 90-degree PVC elbow DWV; model #K05555C), a 0.64 cm brass hose barb for collecting air samples going into the chamber (Sioux Chief 0.64 cm ID barb × 0.64 cm MIP brass adapter; model #17700135), and a 10.2 cm diameter locknut (10.2 cm rigid conduit locknut; model #61940) to secure the inlet pipe to the acrylic glass and seal the polyethylene plastic sheeting to the rubber gasket (Figures 1 and 3).

The outlet air pipe was responsible for allowing air to escape the chamber and provided a source to collect air samples exiting the chamber (Figure 3). It was constructed from the following material: two 0.31 m long sections of 5.1 cm diameter PVC pipe, one 5.1 cm diameter PVC elbow fitting (NIBCO 5.1 cm 90-degree PVC vent elbow; model #K05340C), one 5.1 cm × 7.6 cm PVC reducer fitting (Drain Waste Vent Reducer Fitting; model# D102-338), one 0.64 cm brass hose barb for collecting air samples going out of the chamber (Sioux Chief 0.64 cm ID barb × 0.64 cm MIP brass adapter; model #17700135), and one PVC shower drain fitting (Sioux Chief 5.1 cm PVC shower drain body; model #825-L20P) to secure the outlet pipe to the acrylic glass and seal the polyethylene plastic sheeting to the rubber gasket (Figures 1 and 3).

Sentinel piglet calorimeter.

A pair of separate calorimeters were constructed to estimate the litter THP and separate it from the sow + litter THP to estimate sow THP. The chamber was a clear plastic box (57.2 cm × 40.6 cm × 32.4 cm [L × W × H]) with an airtight sealed lid (Sterilite 51.1 L gasket box clear with blue latches; model #002-02-0895). A 7 cm diameter hole was cut into the 40.6 cm × 32.4 cm side of the container, 7.6 cm from the bottom of the container, and a 8.1 cm diameter air inlet fan (Multifan S1, 80 mm, 26 CFM; AC Infinity, City of Industry, CA) was attached and inlet air samples were collected from the room, 1 cm in front of the inlet fan prior to entering the chamber. On the lid, at the opposite side of the chamber, an air outlet pipe was constructed and attached using gaskets as previously described (Figure 2). Finally, a heating pad (Sunbeam heating pad; model #732-500) was placed within the chambers to provide supplementary heat to the piglets during testing. The chambers provided 0.24 m² total floor space for the piglets.

Performance Testing

All performance testing procedures involving animal use were approved by the Purdue University Animal Care and Use Committee (protocol #1708001612), and animal care and use standards were based upon the Guide for the Care and Use of Agricultural Animals in Research and Teaching (Federation of Animal Science Societies, 2010). Following indirect calorimeter construction, confirmation of an airtight seal was performed on sow and litter as well as sentinel piglet chambers by placing a smoke emitter (Smoke Emitter 90 s; model #S103; Regin Products Ltd., Huntingdon, PE) within the chamber, sealing and pressurizing the chamber, and then visually monitoring the outside of the chamber for smoke leaks. In addition, accuracies for O₂ and CO₂ measurements were evaluated by ethanol combustion in which the respiratory quotient (RQ) was confirmed as 0.67 for each chamber as previously described (Nienaber et al., 2009). Based on results from the smoke emitter test and ethanol combustion, it was confirmed that no air leaks occurred and that all indirect calorimeters could accurately measure O₂ and CO₂ use and production, respectively. One week prior to farrowing over 2 mo (n = 3 test sows and litters per month), a total of 6 pregnant multiparous test sows (Landrace × Yorkshire; parity 2.9 ± 0.9; 218.3 ± 38.6 kg BW) that were bred to Duroc sires were housed individually in each farrowing crate. Ambient temperature and relative humidity within the calorimeters were measured in 5-min intervals throughout performance testing with data loggers (Hobo; data logger temperature/RH; Onset, Bourne, MA; accuracy ± 0.2°C). It was determined that the calorimeters were able to maintain sows and litters in thermoneutral conditions (20.9 ± 2.6°C and 43.7 ± 18.6% relative humidity) during testing in accordance with the Guide for the Care and Use of Agricultural Animals in Research and Teaching (Federation of Animal Science Societies, 2010) without the use of supplementary environmental maintenance equipment (i.e., air conditioner, etc.) inside the chambers. Heating pads (Stanfield heat pad; 0.91 m × 1.83 m; Osborne Industries Inc., Osborne, KS) were placed within all crates to assist piglets with thermoregulation following farrowing. In addition to the 3 test sows per repetition, 1 sentinel sow and litter per repetition were housed within an individual farrowing crate in the same room during each month. Piglets were processed (i.e., tails clipped, ear notched, etc.) 48 h post-farrowing, and then litter size was balanced for all sows (n = 11.4 ± 1.4 piglets per sow) through cross-fostering. Sows and their litters were weighed on day 1, 10, and 21 post-farrowing and all sows were provided ad libitum access to water and feed throughout the entire trial. Diets were corn and soybean based and were formulated to meet or exceed nutrient requirements of lactating sows (National Research Council [NRC], 2012; Supplementary Table 1). Individual sow feed intake was measured on each day of calorimetry testing (day 4, 8, 14, and 18 of lactation).

On day 4, 8, 14, and 18 of lactation, indirect calorimetry was performed on all test sows and their litters as well as on median weight piglets from the sentinel sow’s litter. THP was measured at different days of lactation to compare with previously published data in lactating sows (Jakobsen et al., 2005; Brown-Brandl et al., 2014; Stinn and Xin, 2014), and to characterize the temporal pattern of THP as lactation progressed in individual lactating sows. For test sows and litters, the chamber sides were closed, and the water pan was filled at 1700 h on the day prior to measurements to provide an air tight seal. Test sows and their litters were given 2 h to acclimate to the chambers, and then airflow (m/s) was measured through the outlet pipe using a calibrated vane anemometer (OMEGA HHF 142; accuracy ± 1% of reading; OMEGA Engineering, Inc., Norwalk, CT) held and sealed onto the air sampling port (Figure 3). A manometer (Series 477AV Handheld Digital Manometer; accuracy ± 0.5% of reading; Dwyer Instruments, Inc., Michigan City, IN) was used for all constructed calorimeters to ensure that no pressure changes occurred due to the use of the vane anemometer that would alter the airflow measurement. Immediately following the airflow measurement, input and outlet air samples for individual sows and litters were immediately dried using a 6.67 cm × 28.89 cm tube filled with indicating desiccant (Indicating Drierite; WA Hammond Drierite CO. LTD., Xenia, OH), and collected continuously in 20-liter Mylar bags (Restek Corporation, Bellefonte, PA) from 1900 to 0700 h overnight. The following morning at 0700 h, the filled Mylar bags were removed, airflow rates were monitored, and then an empty set of Mylar bags were attached to the inlet and outlet pipes and a pump (Aqua lifter vacuum pump; model #TM1137; Tom’s Aquatics, Placentia, CA) was used to collect air samples. This air sampling procedure was previously described and validated by Nienaber and Maddy (1985) and was conducted at 0700, 0800, 0900, 1100, 1300, 1500, and 1900 h on each 24-h testing period and combined with the 1900–0700 h data to estimate 24-h THP.

On day 4 and 8, 2 median weight sentinel piglets (n = 1 barrow and 1 gilt) were housed in 1 sentinel piglet calorimeter simultaneously so that they could perform behavioral thermoregulation. However, on day 14 and 18, each sentinel piglet calorimeter only contained 1 median weight sentinel piglet (n = 1 barrow and 1 gilt per calorimeter) due to space constraints. The sentinel piglets were placed into the sentinel piglet calorimeters at 1000 h each day, given 1 h to acclimate, and then airflow was measured through the outlet pipe with a vane anemometer as previously described for the test sow and litter chambers. A manometer was used to ensure no pressure changes occurred due to the use of the vane anemometer that would alter the airflow measurement. Following the airflow measurements, a pump was used to collect dried air samples into an empty set of Mylar bags attached to the inlet and outlet tubes at 60, 90, and 120 min. Following air sample collection, sentinel piglets were immediately returned to the farrowing crate with the rest of the sentinel litter. The sentinel litter data were then used as a correction factor to estimate THP of the individual test sows.

Calorimeter input and output air samples were collected in Mylar bags from each individual sow and litter as well as sentinel piglets and then were analyzed for O₂ with a calibrated paramagnetic O₂ analyzer (Model 600P; California Analytical Instruments, Orange, CA) and for CO₂ with a calibrated infrared CO₂ sensor (Model GMT221; Vaisala Oyj, Helsinki, Finland). THP was calculated with the following formula adapted from Brouwer (1965) by Nienaber et al. (2009), Brown-Brandl et al. (2014), Stinn and Xin (2014), and Chapel et al. (2017) and used to calculate THP in W: THP (W) = [(16.18*(O₂ consumed, mL/s)) + (5.02*(CO₂ produced, mL/s))]. Methane excretion was excluded from the final equation due to its low value in swine energy balance determination and the adjustment for nitrogen excretion was excluded in order to remain consistent with the aforementioned reports (Nienaber et al., 2009; Brown-Brandl et al., 2014; Stinn and Xin, 2014; Chapel et al., 2017). Watts were converted to kilocalorie per hour by multiplying by 0.86 kcal/h/W. To estimate individual test sow THP and RQ, O₂ consumed (VO₂, in mL/s) and CO₂ produced (VCO₂, in mL/s) were calculated with the following equation adapted from Nienaber and colleagues (2009): VO₂/CO₂ (Sow, mL/s) = VO₂/CO₂ (Sow + Litter, mL/s) − [VO₂/CO₂ (sentinel piglet, mL/s·kg) * BW (test litter)]. Test litter BW on d 1 was used to estimate litter THP on d 4 of lactation, test litter BW on d 10 was used to estimate litter THP on d 8 and d 14 of lactation, and test litter BW on d 21 was used to estimate litter THP on d 18 of lactation. These values were then used in the THP equation (Nienaber et al., 2009) to estimate sow THP. The RQ per pig was calculated with the following formula (Nienaber et al., 2009): VCO₂ (mL/s) / VO₂ (mL/s).

Data Analysis

Data were analyzed using the PROC MIXED procedure in SAS 9.4 (Cary, NC). The assumptions of normality of error, homogeneity of variance, and linearity were confirmed post hoc. Farrowing crate was considered the experimental unit for all analyses. Month of testing was included as a random effect for all analyses and lactation day (4, 8, 14, 18) was included as the fixed effect. For repeated analyses with lactation day as the repeated effect, the optimal covariance structure for each response variable was determined by goodness of fit criteria (Littell et al., 1998). Means were separated with a Tukey’s adjustment where appropriate. A statistical significance between comparison was defined when the P ≤0.05, and a tendency was defined as 0.05 < P ≤ 0.10.

RESULTS AND DISCUSSION

Methodological difficulties have contributed to a lack of literature on the bioenergetics of individual lactating sows. This is primarily because the litter must join the sow in the calorimeter to maintain milk production levels and therefore the piglets contribute to the gas exchange and THP and RQ calculations (Theil et al., 2003; Jakobsen et al., 2005). As such, separating the sow’s THP from the litter requires additional steps. Several techniques have been described to separate THP in the lactating sow from the litter such as the RQ method, the piglet factorial method, and the double labeled water method (Theil et al., 2003; Jakobsen et al., 2005). The double labeled water method allows researchers to evaluate THP in a subset of piglets from an individual litter, apply that estimate to the total litter, and then calculate the sow THP by the difference between sow + litter THP and litter THP (Theil et al., 2003; Jakobsen et al., 2005). Unfortunately, this method requires researchers to remove a subset of piglets from the test sow, which may decrease her milk production and subsequently THP since litter size is linearly related to milk production (Auldist et al., 1998). Therefore, to circumvent this, a subset of piglets from sentinel litters of similar genetics, parity, and housing conditions were utilized in the present study to separate sow and litter THP. Although the advantage of this method is that researchers can avoid inadvertently reducing milk production and subsequently THP in test sows, it is important to note that this method assumes that the THP of the sentinel piglets is similar to the test sow’s piglets. Therefore, potential source of error must be acknowledged in selecting the best approach.

BW of the sows + litter increased (P < 0.01) as lactation progressed from day 1 to day 21 (Table 1). However, sow BW remained unchanged from day 1 to day 10 and was reduced by 4.1% on day 21 compared to day 10 (Table 1), which was somewhat surprising considering that sow feed intake increased (P = 0.01) as lactation progressed from day 4 to day 18 of lactation (Table 2). However, these data are similar to previous results by Jakobsen et al. (2005) and Brown-Brandl et al. (2014) in which sow + litter weight increased over time without a subsequent increase in sow weight. The increase in sow + litter BW without a subsequent increase in sow BW can be explained by the increase (P < 0.01) in litter BW from day 1 to day 21 of lactation (Table 1). However, the reduction in sow BW at the end of lactation (Table 1) despite the increase in feed intake (Table 2) is likely due to the greater lactation demands as the piglet’s energy requirements were increased with age and size. Therefore, although sow feed intake increased as lactation progressed, it is likely that the sow’s energy intake was not great enough to keep up with lactation demands (i.e., 65% metabolizable energy directed towards lactation; Bergsma et al., 2009) resulting in body tissue mobilization to support lactation (Rempel et al., 2015), which likely led to a decrease in sow BW (Table 1).

Table 1.

Body weight of sows + litter, sows only, and litters only throughout lactation

	Lactation day				P-value
Parameter	Day 1	Day 10	Day 21	SEM	Day
Sow + litter
Body weight, kg	237.5a	265.8b	282.3c	15.8	<0.01
Sow
Body weight, kg	217.7ab	220.3b	211.3a	15.3	<0.01
Litter
Body weight, kg	19.8a	45.5b	71.0c	1.8	<0.01

^a,b,c P ≤ 0.05.

Table 2.

Feed intake, total heat production (THP, kcal/h), and respiratory quotient of sows and sows + litter throughout lactation

	Lactation day					P-value
Parameter	Day 4	Day 8	Day 14	Day 18	SEM	Day
Sow + litter
THP, kcal/h	430.50a	502.17b	588.17c	599.00c	21.46	0.01
Respiratory quotient	1.06	1.03	1.00	1.03	0.04	0.21
Sow
Feed intake, kg	4.96a	6.59b	6.93bc	7.26c	0.13	0.01
THP, kcal/h	346.16a	401.19b	430.79c	386.16ab	16.55	0.01
Respiratory quotient	1.03a	1.02a	0.98b	1.04a	0.03	0.04

^a,b,c P ≤ 0.05.

As lactation progresses, the THP of lactating sows and their litters increases and this increase is related to greater piglet size over time (as observed in the present study; Table 2) as well as milk production levels by the sow (Jakobsen et al., 2005; Brown-Brandl et al., 2014; Stinn and Xin, 2014). In agreement, sows + litter THP increased (P = 0.01) as days of lactation progressed from day 4 to 18 (Table 2). Specifically, sow + litter THP was greater on day 8 compared to day 4 and was greater on day 14 and day 18 compared to day 4 and day 8 (Table 2). No differences were detected between day 14 and day 18 suggesting that sow + litter THP peaked by week 2 of lactation in the present study, which corroborated previous findings by Brown-Brandl et al. (2014) and Stinn and Xin (2014). Although it is likely that the increase in THP over time was partially related to piglet growth (Brown-Brandl et al., 2014), data from the present study suggests that individual sow THP partially contributed to this increase because sow THP was also greater (P = 0.01) as days progressed from day 4 to 14 (Table 2). Although sow + litter THP values in the present study increased throughout the course of lactation similar to previous studies (Jakobsen et al., 2005; Brown-Brandl et al., 2014; Stinn and Xin, 2014; Supplementary Table 2), THP values were lower when compared to observations by Brown-Brandl et al. (2014) and Stinn and Xin (2014). However, this may be due to differences in sow genetics (i.e., modern hybrids in Brown-Brandl et al. [2014] and Stinn and Xin [2014] vs. conventional genetics in the present study) and measurement techniques (i.e., whole room vs. individual animal measurements) between the studies (Supplementary Table 2). This is because sow + litter THP and sow THP in the present study differed by only 2.3% and 3.1%, respectively, when compared to previously published THP values observed in sows of similar genetics on similar lactation days and measured using a calorimeter (Jakobsen et al., 2005; Supplementary Table 2). To determine this difference, the THP mean from lactation day 10 and day 17 (Jakobsen et al., 2005; Supplementary Table 2) was compared to the THP mean from lactation day 8 and day 18 in the present study (Table 2). Therefore, it may be concluded that THP values as determined by this system are comparable with previously observed values in sows of similar genetics and stages of lactation. Furthermore, the small differences in THP between the Jakobsen et al. (2005) study and the present study may have been due to the fact that THP was calculated using the nitrogen and methane adjustment factors (Jakobsen et al., 2005) while THP in the present study did not include these factors as previously described (Brown-Brandl et al., 2014; Stinn and Xin, 2014; Chapel et al., 2017).

Although THP increased over the course of lactation, no RQ differences for the sow + litter were detected (P = 0.21; 1.03 ± 0.26) from day 4 to 18 of lactation (Table 2). These data are similar to previously reported values in sows + litters (Brown-Brandl et al., 2014; Stinn and Xin, 2014), and likely indicate that sow + litter energy consumption was sufficient throughout lactation. This is because values of 1.0 designate carbohydrate oxidation (Nienaber et al., 2009), which would be expected in animals at a positive energy balance. Despite the lack of RQ differences for sow + litters, a reduction in RQ was detected for sow’s (P = 0.04) on day 14 (0.98 ± 0.02) compared to day 4 (1.03 ± 0.03), day 8 (1.02 ± 0.02), and day 18 (1.04 ± 0.03), but no differences were observed on day 4, 8, and 18 (Table 2). While this could point towards a more negative energy balance on day 14, this slight reduction from 1.0 likely did not indicate that the sows were in a severe negative energy state because protein and adipose oxidation generally does not occur until the RQ approaches 0.8 and 0.7, respectively (Nienaber et al., 2009).

This indirect calorimetry system can be built and dismantled at relatively low costs ($1,892 USD), which can allow for greater availability to researchers worldwide. In addition, it may allow researchers to overcome methodological difficulties in determining individual lactating sow and litter THP. However, as with any system, there are potential caveats to consider. Although using sentinel piglets to estimate litter CO₂ production and O₂ consumption allows researchers to separate the sow from the litter THP without removing piglets from the test litter, this could be a potential source of error if the sentinel piglet CO₂ production and O₂ consumption is not similar to the test litter. Moreover, determining sentinel piglet CO₂ production and O₂ consumption requires the removal of piglets from the sow for a period of time precluding the piglet’s nursing ability and potentially altering their metabolism. To overcome this obstacle, sentinel piglets were only removed from the sow and tested for a 2-h period and these CO₂ production and O₂ consumption data were assumed to represent the 24-h average. Despite these potential limitations, this system allows researchers to evaluate THP in individual sows and litters at a relatively low cost with a degree of accuracy based on comparisons with previously published data in sows of similar genetics (Jakobsen et al., 2005).

CONCLUSIONS

The use of calorimetry systems to estimate metabolic rate in living organisms is essential to understand livestock bioenergetics. Indirect calorimetry can be used to determine animal energy requirements and assist in formulating diets, or to evaluate heat production to better design environmental management systems in livestock facilities. Herein, we described the development and use of a low-cost indirect calorimetry system to determine THP in individual lactating sows and their litters. It was determined that this system could accurately evaluate THP in sows and their litters both together and separately and that the cost of the sow + litter calorimeter was $1,825 USD and the cost of the sentinel piglet calorimeter was $67 USD. Furthermore, as lactation progressed THP increased while RQ remained static for the sow and litter combined. These data have implications towards providing researchers worldwide with a new cost-effective technology that can allow them to accurately and inexpensively determine the bioenergetics of lactating sows and their litters.