Master of Science (MS)


Animal Science (Animal, Dairy, and Poultry Sciences)

Document Type



Three experiments were conducted to further study the use of intravenous insulin injection to measure insulin sensitivity in horses. In the first experiment, glucose responses to multiple doses of recombinant human insulin were compared in insulin sensitive and insensitive mares. Regression lines describing the responses in insensitive mares had less (P = 0.003) steep slopes and greater (P = 0.006) effective doses at 50% response (ED50) compared to sensitive mares. Curvilinear regression models provided a good fit (R2 = 0.95) for the prediction of ED50 from the individual responses to 50 mU/kg BW insulin. The second experiment estimated the repeatability of the glucose responses to a fixed dose of insulin in 41 horses (mares and geldings) injected twice, between 7 and 14 days apart. Overall agreement between the responses was significant (P < 0.001) but not high enough to be predictive (R2 value = 0.384). Intra-horse coefficients of variation (CV) ranged from 0 to 68.7%, and averaged 23%. Mares and geldings were similar in their repeatabilities. Horses in the bottom half of the sensitivity rankings had a greater (P = 0.03) average intra-horse CV (28.7 ± 4.6%) than horses in the top half (16.7 ± 2.2%). In the third experiment (single switch-back design), glucose responses to a fixed dose of insulin were compared after an overnight (about 12 hours) versus a 24-hour period of feed deprivation in geldings previously determined to be insulin sensitive (n = 4) versus insensitive (n = 4). The longer period of feed deprivation decreased (P = 0.06) the percentage decrease in blood glucose concentrations, indicating a lesser sensitivity relative to the 12-hour period of deprivation. Insulin sensitivity also affected (P = 0.0003) the glucose response, with sensitive geldings exhibiting the greater response. It was concluded that the response to a single dose of insulin is often a good predictor of a horse's insulin sensitivity, but that the lack of repeatability in some horses could lead to errors in diagnosis. The decrease in glucose response seen after 24 hours of feed deprivation is likely due to the elevation of fatty acid concentrations in the blood.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Thompson, Donald L., Jr.