Effect of furosemide on performance of Thoroughbreds racing in the United States and Canada.

OBJECTIVE: To determine the effect of furosemide on performance of Thoroughbreds racing on dirt surfaces at tracks in the United States and Canada. DESIGN: Cross-sectional study. ANIMALS: All Thoroughbreds (n = 22,589) that finished a race on dirt surfaces at tracks in the United States and Canada between June 28 and July 13, 1997 in jurisdictions that allowed the use of furosemide. PROCEDURE: Race records were analyzed by use of multivariable ANOVA procedures and logistic regression analyses to determine the effect of furosemide on estimated 6-furlong race time, estimated racing speed, race earnings, and finish position. Principal component analysis was used to create orthogonal scores from multiple collinear variables for inclusion in the models. RESULTS: Furosemide was administered to 16,761 (74.2%) horses. Horses that received furosemide raced faster, earned more money, and were more likely to win or finish in the top 3 positions than horses that did not. The magnitude of the effect of furosemide on estimated 6-furlong race time varied with sex, with the greatest effect in males. When comparing horses of the same sex, horses receiving furosemide had an estimated 6-furlong race time that ranged from 0.56 +/- 0.04 seconds (least-squares mean +/- SE) to 1.09 +/- 0.07 seconds less than that for horses not receiving furosemide, a difference equivalent to 3 to 5.5 lengths. CONCLUSIONS AND CLINICAL RELEVANCE: Because of the pervasive use of furosemide and its apparent association with superior performance in Thoroughbred racehorses, further consideration of the use of furosemide and investigation of its effects in horses is warranted.