Clamp-like Index: a Novel, Highly Sensitive Insulin Sensitivity Index to Calculate Hyperinsulinemic Clamp Glucose Infusion Rates from Oral Glucose Tolerance Test in Nondiabetic Humans

Abstract

Objective. Insulin resistance (IR), the underlying pathophysiological mechanism of the metabolic syndrome, could predict not only type-2 diabetes (T2DM) development, but also cardiovascular disease. Thus, precise IR measurement in individuals at risk for metabolic diseases would support clinical risk stratification. However, the “gold-standard” to measure IR, the hyperinsulinemic clamp-test, is too labor-intensive to be performed in large clinical studies/settings.

Research Design and Methods. Using plasma glucose and C-peptide concentrations from oral glucose tolerance tests (oGTTs), we developed the novel “clamp-like index” (CLIX) for insulin sensitivity calculation, and compared CLIX to clamp glucose infusion rates (100-120min) (GIR_100-120min). We evaluated CLIX in eighty-nine nondiabetic subjects (f/m=58/31, aged:45±1years, BMI=27.5±0.8kg·m^-2) who underwent frequently-sampled three-hour-(75g)-oGTTs and two-hour hyperinsulinemic-(40mU·min^-1·m^-2)-isoglycemic clamp-tests.

Results. CLIX, calculated as serum creatinine(×0.85 if male)/(mean AUC_glucose×mean AUC_C-peptide)×6600, highly correlated (r=0.670, p<10^-12) with and was comparable to clamp GIRs_100-120min.

Results. In subgroup analyses, GIRs_100-120min were lower (p<0.005) in T2DM-offspring (OFF) (6.2±0.7 mg·min^-1·kg^-1) than in gender-, age- and BMI-matched subjects without T2DM family history (NOFF) (8.6±0.5 mg·min^-1·kg^-1), which was also reflected by CLIX (OFF:6.4±0.6 vs. NOFF:9.0±0.5, p<0.002). When compared to normal-weight subjects (GIR:8.8±0.4 mg·min^-1·kg^-1; CLIX:9.0±0.5), both GIRs_100-120min and CLIX of obese (5.2±0.9 mg·min^-1·kg^-1; 5.7±0.9) and morbid obese (2.4±0.4 mg·min^-1·kg^-1; 3.3±0.5) humans were lower (each p<0.02).

Conclusion: CLIX, a novel index obtained from plasma oGTT glucose and C-peptide levels and serum creatinine without inclusion of anthropometrical measures to calculate insulin sensitivity in nondiabetic humans, highly correlates with clamp GIRs and reveals even slight insulin sensitivity alterations over a broad BMI-range as sensitively as the hyperinsulinemic clamp-test.