Pedometers as supervision tools for mares in the prepartal period

To assess the practical use of pedometers as supervision tools in the equine prepartal period, two types, IceQube®- and ALT-sensors, were tested. Nine mares were randomly fitted with devices of the respective types on the front legs (ALT: n = 9, IceQube®: n = 7), and with an IceQube®-sensor on a neck collar (n = 7). Measurements of 10 days ante partum for motion activity, lying times and lying bouts with a measuring interval of 15 min were used to analyse changes during the ante partum interval. Technical function, animal acceptance, informative value, and the general use of the assessed data to predict the date of parturition, were observed and investigated, respectively. Therefore, deviations in the animals’ behaviour were statistically determined by estimating least squares means (LSM), and their differences between the days ante partum and the day of parturition to the same hours, respectively (P < 0.05). Mean gestation length of the mares was 337.3 ± 8.6 days. LSM-differences for motion activity showed a highly significant increase two to one hours ante partum. In contrast, relatively constant patterns of motion within the same hours on the other days before were observed. These results were used to develop a real-time method that can detect the upcoming birth. The method is based on the actual moving average of motion activity and the 95th percentile of motion activity of the day before. For most mares, increasing moving average values above the 95th percentile were observed in the last two to one hours before parturition. The correlation for motion activity between the leg-instruments’ data was calculated with r = 0.51 during stable times. Lying times increased significantly ante partum, but this effect was too close to the beginning of parturition, and therefore not useable as a predictive marker. Lying time per day and per hour varied between the animals, and was generally on a very low level. The IceQube®-sensors on the mares’ neck collars recorded a permanent lying bout of these devices, which was increasingly fragmented due to movements of the head and the neck prior to parturition. This observation can be explained as colicky behaviour under pain. We suggest that this parameter could have a high predictive value. The number of lying bouts increased within the last prepartal hours, however, the analysed data of the single mares displayed again a high variability. The correlation for the recorded lying time was r = 0.71 between IceQube®- and ALT-devices at the mares’ front legs. In conclusion, pedometers, regardless of which type tested, possess a high potential as supervision tools in the prepartal period of mares, especially two to one hours before birth. A prerequisite is, however, that the assessed data is analysed thoroughly with a suitable method.