Variation of Insulin Sensitivity Estimates in Horses

In the horse, resting insulin concentration (INS), the glucose-to-insulin ratio (G:I), and the reciprocal of the square root of insulin (RISQI = 1/√INS) are commonly used to estimate insulin sensitivity, whereas the modified insulin-to-glucose ratio (MIRG = [800 – 0.30 × (INS -50)2]/(GLU – 30) is used to estimate pancreatic beta-cell responsiveness. Because no estimates of their within-horse variability and repeatability have been reported, the objective of this study was to evaluate the within-horse variation of these estimates. Resting blood samples were obtained from six healthy equids (three geldings, two mares; mean ± SD body weight, 525.0 ± 43.36 kg; mean age, 9.8 ± 8.2 years; and one pony gelding: 293 kg; 12 years) on three consecutive days in week 1 and again in week 2. Samples were collected at 12:00 noon, approximately 6 hours postprandially. Serum insulin and plasma glucose (GLU) concentrations were analyzed and used to calculate G:I, RISQI, and MIRG, as well as the insulin to glucose ratio (I:G). The coefficient of variation was used to determine within-horse variation, and repeatability was determined using the repeatability coefficient (RC; measurements from a single horse should differ less than the RC for 95% of the pairs). The mean coefficients of variation (CVs) for resting GLU, INS, G:I, I:G, MIRG, and RISQI were 5.5%, 33.7%, 36.0%, 31.6%, 22.3%, and 18.6%, respectively. All variables had values that differed more than the RC in at least one horse. These data suggest that care should be taken when interpreting insulin sensitivity estimates from a single blood sample.