Hyperinsulinemia in horses has been shown to be related to risk of pasture associated laminitis and the equine metabolic syndrome. Additionally, some tests of endocrine function in horses, including the dexamethasone suppression test and plasma adrenocorticotropin concentration have seasonal variation. The aim of this study was two-fold, first to determine the prevalence and risk factors for hyperinsulinemia in clinically normal, adult horses and second, to determine if those concentrations vary with season.
For the first part of the study, a convenience sample of 300 horses (138 mares, 143 geldings and 19 stallions; 4 to 30 years of age) was drawn from 18 farms that partially comprised the Ohio State University field service population. Plasma insulin and glucose were measured after a 10-12 hour period of grain and grass withholding. Measures of body condition score, height, weight tape and neck circumference as well as pertinent history and management practices were obtained at that time. A univariable logistic regression and a multivariable logistic regression with a random effect of farm were used to estimate odds ratios.
The prevalence of hyperinsulinemia, defined by a resting insulin concentration >15 μU/L, was 22.3% (67/300). Multivariable analysis revealed that increasing age (odds ratio per year of increasing age 1.1, 95% CI 1.03-1.17) and body condition score (odds ratio 2.43, 95% CI 1.6-3.7) increased the risk for hyperinsulinemia while access to pasture (odds ratio 0.34, 95% CI 0.14-0.86) decreased the risk. Breed, sex, exercise, and measures of neck and girth circumference were not associated with risk of hyperinsulinemia.
For the second part of the study, twenty-nine healthy horses, as determined by physical examination, were evaluated at three-month intervals over a one year period. Plasma insulin and glucose concentrations were measured after a 10-12 hour period of grain and grass withholding. Body condition score, height, weight tape and neck circumference were measured at the same time. Friedman’s test and Wilcoxon’s signed rank test were used to compare the overall and individual differences between the four seasons in the values of insulin and glucose. The type 1 error rate was 5%.
There was an effect of season on the concentration of both insulin and glucose (P<0.001). The concentration of insulin in summer was lower (median 4.6 μU/L, range 1.2 to 22.0) than any of the other three months (P<0.001). There was no significant difference between any of the other seasons (autumn 13.4, range 4.3 to 271.7; winter 13.6, range 5.3 to 401.7; spring 12.5, range 2.0 to 412.4). The concentration of glucose in autumn was less than in any of the other three seasons (P<0.001), and the concentration in summer was less than in spring (p=0.041).
These studies demonstrate that risk factors associated with hyperinsulinemia can be identified and that an apparent seasonal variation exists in the plasma concentrations of both insulin and glucose in normal horses.