Monitoring the reproductive status of dairy cows using cow-side oestrus detection technologies

Abstract

Reproductive failure remains one of the main reasons for culling in dairy herds worldwide. Prolonged inter-calving periods occasioned by failure or delay in conceiving reduces the lifetime productivity of the cows and increases the herd replacement rate. The postpartum open state is a critical stage for the mature dairy cow that determines its future productivity. The ability of the postpartum cow to return to cyclicity, express oestrus, conceive when inseminated, and carry the pregnancy to term within a prescribed period, assures a cow’s productivity and its chances of remaining in the herd. One of the key activities in the management of the postpartum dairy cow is oestrous detection. Oestrus is defined as a recurring period of sexual receptivity and fertility in female mammals. Successful conception requires that oestrus be detected accurately and cows that are identified to be in oestrus inseminated at the right time, in relation to ovulation. The goal of this thesis was to evaluate and compare the degree of agreement, effectiveness and robustness of three cow-side oestrous detection tools (ODTs) in three production and climatic conditions. The three ODTs selected for investigation were CowAlert®, (IceRobotics, UK), a cow activity monitoring platform, Estrotect™ (Rockway Inc, USA), a scratch card-based mount detector and P4 Rapid (Ridgeway Sciences, UK), a lateral flow technique based on milk Progesterone. Three experiments were carried out, one each in a large-scale farm in a temperate region, a large-scale farm in the tropics and small-holder farms in the tropics. The purpose of the different production environments was to understand how the effectiveness and robustness of the ODTs changed with each production environment. The expectation was that at the end of the study recommendations would be made on which of the tested ODT was best suited for that production environment. The first experiment was carried out Scotland’s Rural College’s (SRUC) Dairy Research and Innovation Centre in Dumfries, UK. This served as the large-scale farm in a temperate region. In this experiment, the three ODTs were evaluated for their degree of agreement, effectiveness and robustness. Degree of agreement was evaluated by carrying out a Cochran’s Q test, the effectiveness evaluated through sensitivity analysis while the robustness was evaluated by carrying out the sensitivity analysis on groups of cows categorised according to their body condition and locomotion scores, milk yield(Kg), weight(kg), days in milk and parity. The ODTs were applied concurrently to the cows to allow for accurate comparison. All three ODTs did not agree in their ability to detect oestrus . P4 Rapid had the highest effectiveness as measured by the sensitivity (0.86 (0.79, 0.90)) in detecting oestrous followed by CowAlert® (0.50(0.43, 0.58) then Estrotect™ (0.31 (0.24 , 0.37)). The body condition and locomotion scores, milk yield and weight did not influence the effectiveness of the ODTs. For all three ODTs there was a statistically significant difference in their effectiveness when cows were grouped in their different parities and between P4 Rapid and both CowAlert® and Estrotect™ when the cows were grouped in their days in milk. A closer examination at the influence of parity on the effectiveness showed that P4 Rapid was statistically significantly different to Estrotect™ in all parities recorded but only statistically significantly different to CowAlert® in cows from parity four and higher. For the days in milk, there was a statistically significant difference in the effectiveness between P4 Rapid and both CowAlert® and Estrotect™ in cows up to 60 days in milk. For this production system P4 Rapid was deemed the most effective and robust while Estrotect™ was found to be the least effective. The second experiment followed the same experimental design as the first one and was carried out in College of Agriculture and Veterinary Sciences (CAVS), University of Nairobi (UON) dairy farm in Kenya. This served as the large-scale farm in a tropical environment. In this experiment, P4 Rapid did not agree with either CowAlert® or Estrotect™ in its effectiveness in detecting oestrus. However, CowAlert® and Estrotect™ agreed in their effectiveness in detecting oestrus. P4 Rapid had the highest effectiveness as measured by the sensitivity (0.98(0.94,1.0)). In contrast to the first experiment Estrotect™ had the next highest sensitivity (0.26(0.19, 0.34)) while CowAlert® had the least sensitivity at (0.2(0.14,0.28)). However, in this experiment both CowAlert® and Estrotect™ had a sensitivity of less than 30%. In deviation from the first experiment, only the parity had an influence on the effectiveness of the ODTs. For this production system P4 Rapid was deemed the most effective and robust while CowAlert® was found to be the least effective. However, there were certain drawbacks associated with the use of Estrotect™ as it allowed for potentially infective ticks to burrow under the strip, providing a haven for them against acaricides. While detecting oestrus is an important management activity, it is prudent that it is not done at the expense of the health of the cows. The third experiment was set in the tropics with 34 small-holder farms. Based on the information collected in the first two experiments, the experimental design was different from the first two experiments. With the small-holder farms, the first objective was to evaluate the reproductive status monitoring practices in use, with a focus on oestrous detection and pregnancy diagnosis. All the farms sampled used visual observation to detect signs of oestrus prior to insemination and their absence (non-return to oestrus) as an indicator of conception. None of the farms sampled had used an ODT prior to this study. The second objective was to evaluate the performance P4 Rapid and CowAlert® in detecting oestrus both at insemination and as an indicator of non-conception up to 24 days following insemination when compared with visual oestrous detection. P4 Rapid was found to be the most effective in detecting oestrus during insemination and non-oestrus up to 24 days post insemination. In conclusion, P4 Rapid was consistently the most sensitive in detecting an oestrous event in the three production systems tested. This is likely because P4 Rapid is a direct measure of an intrinsic parameter, progesterone. Progesterone concentration has a distinct pattern as the cows moves though the oestrous cycle and has been used successfully to identify ovulation and thus signifying that oestrus occurred. However, both CowAlert® and Estrotect™ occur in response to elevated levels of Oestrogen and not direct measures of an intrinsic parameter. Increased activity can also occur outside oestrus for the case of CowAlert® and for Estrotect a scratched strip is not an indicator of mounting activity. This study proved particularly useful for small-holder farms where there was little documented use of ODTs. A natural progression of this work would be to investigate the economic implication of adoption of P4 Rapid as a oestrous detection tool by large and small-holder farms.