The use of implantable microchips for body temperature collection in cattle

Thermal regulation via physiological processes has allowed mammals to survive and thrive in a number of diverse climates. The process of regulating body temperature is complicated and tied to many homeostatic systems in the body. That connection to different systems in the body allows for the use of body temperature to identify deviations from baseline. Rectal temperature (RT) has been used in cattle to identify immune system challenges (febrile response), heat stress, ovulation, and calving, but the ability to monitor body temperature in a production setting is labor intensive. Recently, implantable devices have been developed that have the ability to give real time temperature readings via radio frequency data transfer. The objective of these studies is to determine how well the temperature readings of these implantable devices correlate to RT in cattle exposed to thermoneutral and elevated environmental temperatures, with and without a lipopolysaccharide challenge, and in mature cattle during the periparturient phase (parturition and early lactation disease detection) and around the time of estrus. In the first experiment it was hypothesized that in response to lipopolysaccharide (LPS) challenge, temperature patterns from three radio frequency implants (RFI) at three peripheral implantation sites {s.c. at the ear (ET), poll (PT), and umbilical fold (UT)} would be similar to RT patterns in weaned steers. Rectal temperature rose rapidly to 39.9 ± 0.30°C after LPS injection, but ET, PT and UT declined in similar fashion. These data do not support the hypothesis that core and peripheral temperature move in synchrony after LPS challenge. The ET implant site had the most robust differences from RT and showed promise as suitable implant location for future studies. In the second experiment it was hypothesized that the temperature readings from RFIs implanted in the ear of pregnant cows would be positively correlated with baseline RT and negatively correlated with RT during a known health event, and that calving time could be predicted from temperatures measured by using RFIs. Rectal and ET temperatures were positively correlated during both the dry and post-calving periods, and were both higher during periods of high ambient (AMB; > 31 °C) when compared with periods of lower AMB (< 21 °C). A Multiple Local Property Correlation analysis correctly identified all animals experiencing a health event, but had a high false positive rate (flagged 75% of all animals). A negative correlation between RT and ET around parturition or during diagnosed health events (n=12) was not observed therefore this approach has limited value in prediction of parturition or early disease detection when ET is measured at 6 h intervals. The hypothesis of the third experiment was that temperature measured by a RFI, implanted in the ear of cows (n = 32), would be positively correlated to RT during estrus. Rectal temperature and ET were positively correlated when temperatures were measured every 6 h (6H) as well as every 1 h (1H). Rectal temperature increased by roughly 0.6 °C around estrus using 6H, but ET did not increase significantly during that time, but the patterns of RT and ET were similar. The results of this study indicate that ear temperature monitoring via a RFI is not adequate to detect cattle experiencing estrus. The hypothesis of the fourth experiment was that temperatures obtained from RFI in steers would be positively correlated with RT under high AMB conditions (20 °C vs. 34 °C) and negatively correlated with RT during a LPS challenge under high AMB. Pearson correlation coefficients for RFI and RT were 0.3 during heat stress, 0.20 during heat stress with LPS challenge, 0.34 during the thermoneutral period, and -0.42 during the thermoneutral period with LPS challenge. Individual response varied; some animals exhibited negative correlation while others exhibited positive correlation. These data do not support the hypothesis and suggest that individual response be considered when identifying models for use of RFI in temperature monitoring. The results of these studies indicate that the BioThermo microchips correlate well with RT under thermoneutral conditions. The positive correlation was similar to that in both dry and lactating cattle not experiencing disease events and continued in those cows before and after estrus in the third experiment. The positive correlation was again observed in a second cohort of young steers experiencing thermoneutral ambient temperature in the fourth experiment. There is a repeatable positive correlation between the microchip and RT during periods of thermoneutral ambient temperature and when animals are not experiencing a disease event. The results of these studies indicate that the BioThermo microchip system has potential for use in identifying animals experiencing a range of biological changes through the use of peripheral body temperature. In order for the system to be utilized in common commercial applications will require the development of longer read distances, more frequent data collection, and better data collection and analysis techniques. More research is needed to develop more robust models that take into account varying ambient conditions and individual animal variation