Evaluation of fasting as a technique for use in feeding experiments in which dairy cows are strip-grazed

Abstract

The effects of length of fast (0-72 h) on the productivity, liveweight and body condition of dairy cows in mid-lactation were measured, together with the effects of this treatment on the responses obtained in a subsequent feeding experiment. It was also considered whether fasting reduced the variation and improved the estimates of liveweight change and condition score change between treatments in the feeding experiment. Fasting did not influence the responses to level of feeding in a subsequent feeding experiment, as evidenced by the lack of interaction between length of fast and feeding level. Furthermore, fasting did not improve the precision of estimates of liveweight or condition score change. Although the cows were slow to recover, particularly from the longest fast, the major effect of fasting on the conduct of a subsequent experiment was associated with the possibility of health problems. The responses of milk yield, milk fat content and liveweight to length of fast were curvilinear while those of milk fat yield and body condition score were linear. After a 72 h fast, the reduction in milk yield was approximately 85%, while that of liveweight was 12%. Observers also assigned lower body condition scores to fasted cows; after a 72 h fast, the assigned condition score was 0.56 units lower, on average. In the light of these results, the use of fasting in feeding experiments is discussed.