A novel simple disposition index (SPINA-DI) from fasting insulin and glucose concentration as a robust measure of carbohydrate homeostasis

Abstract

Aims: The widely used dynamic disposition index, derived from oral glucose tolerance testing, is an integrative measure of the homeostatic performance of the insulin-glucose feedback control. Its collection is, however, time consuming and expensive. We, therefore, pursued the question if such a measure can be calculated at baseline/fasting conditions using plasma concentrations of insulin and glucose.

Methods: A new fasting-based disposition index (structure parameter inference approach-disposition index [SPINA-DI]) was calculated as the product of the reconstructed insulin receptor gain (SPINA-GR) times the secretory capacity of pancreatic beta cells (SPINA-GBeta). The novel index was evaluated in computer simulations and in three independent, multiethnic cohorts. The objectives were distribution in various populations, diagnostic performance, reliability and correlation to established physiological biomarkers of carbohydrate metabolism.

Results: Mathematical and in-silico analysis demonstrated SPINA-DI to mirror the hyperbolic relationship between insulin sensitivity and beta-cell function and to represent an optimum of the homeostatic control. It significantly correlates to the oral glucose tolerance test based disposition index and other important physiological parameters. Furthermore, it revealed higher discriminatory power for the diagnosis of (pre)diabetes and superior retest reliability than other static and dynamic function tests of glucose homeostasis.

Conclusions: SPINA-DI is a novel simple reliable and inexpensive marker of insulin-glucose homeostasis suitable for screening purposes and a wider clinical application.

Keywords: disposition index; dynamical compensation; insulin‐glucose homeostasis.

FIGURE 1

(A) SPINA‐DI remains constant as long as changes in insulin sensitivity (as measured by SPINA‐GR) are balanced by adequate variations in beta cell function (SPINA‐GBeta, circles 1, 2, and 3). (B) If dynamic compensation fails, either due to insufficient adaptation to insulin resistance (circle 4) or primary beta cell failure (circle 5), the disposition index breaks down and is markedly reduced. (C) Computer simulations with systematic variations of insulin receptor gain (GR) and secretory capacity of beta cells (GBeta) and resulting blood glucose concentration. (D) SPINA‐GBeta plotted against SPINA‐GR in the definition dataset. Results of persons with normal glucose homeostasis are near the hyperbola of normal SPINA‐DI (green dots). The ability of dynamical beta cell compensation is partly lost in persons with prediabetes (orange dots) or diabetes (red dots), resulting in lower SPINA‐DI. SPINA‐DI, structure parameter inference approach‐disposition index; SPINA‐GBeta, secretory capacity of pancreatic beta cells; SPINA‐GR, reconstructed insulin receptor gain.

FIGURE 2

Flow chart of the three cohorts. The numbers of eligible, excluded, enrolled, and analyzed persons are shown. BMI, body mass index; FAST, study on fast insulin response in OGTT; HbA1c, glycated hemoglobin; NHANES, National Health and Nutrition Examination Survey; OGTT, oral glucose tolerance test.

FIGURE 3

SPINA‐DI has high discriminatory power for the distinction between diabetes and normal glucose homeostasis. Receiver operating characteristic (ROC) curves and corresponding areas under the curve (AUC) are shown for several markers for the function of the insulin‐glucose feedback control in the validation dataset. HOMA‐Beta, homeostatic model assessment of β‐cell function; HOMA‐IR, homeostatic model assessment of insulin resistance; IGI, insulinogenic index; ISI, insulin sensitivity index; QUICKI, quantitative insulin sensitivity check index; SPINA‐DI, structure parameter inference approach‐disposition index; SPINA‐GBeta, secretory capacity of pancreatic beta cells; SPINA‐GR, reconstructed insulin receptor gain.

FIGURE 4

SPINA‐GR (A), SPINA‐GBeta (B), and SPINA‐DI (C) correlate with the area under the glucose curve in oral glucose tolerance testing and with the glucose concentration after 2 h (D). SPINA‐DI also correlates to the disposition index according to Matsuda and DeFronzo (E) and to the area under the insulin curve in oral glucose tolerance test (OGTT) (F). Areas under the curve are baseline corrected. AUC, area under the curve; SPINA‐DI, structure parameter inference approach‐disposition index; SPINA‐GBeta, secretory capacity of pancreatic beta cells; SPINA‐GR, reconstructed insulin receptor gain.