Patterns of Insulin Concentration During the OGTT Predict the Risk of Type 2 Diabetes in Japanese Americans

• Published in: Diabetes care Vol. 36; no. 5; pp. 1229 - 1235
• Main Authors: HAYASHI, Tomoshige, BOYKO, Edward J, SATO, Kyoko Kogawa, MCNEELY, Marguerite J, LEONETTI, Donna L, KAHN, Steven E, FUJIMOTO, Wilfred Y
• Format: Journal Article
• Language: English
• Published: Alexandria, VA American Diabetes Association 01.05.2013
• Subjects: Asian Americans; Biological and medical sciences; Dextrose; Diabetes; Diabetes Mellitus, Type 2 - epidemiology; Diabetes Mellitus, Type 2 - metabolism; Diabetes. Impaired glucose tolerance; Endocrine pancreas. Apud cells (diseases); Endocrinopathies; Etiopathogenesis. Screening. Investigations. Target tissue resistance; Female; Glucose; Glucose metabolism; Glucose Tolerance Test - methods; Glucose tolerance tests; Humans; Hyperglycemia; Insulin; Insulin - metabolism; Insulin resistance; Male; Medical sciences; Metabolic diseases; Middle Aged; Miscellaneous; Odds Ratio; Original Research; Public health. Hygiene; Public health. Hygiene-occupational medicine; Risk factors; Type 2 diabetes; Asia; United States; North America; Japan; America; Endocrinopathy; Type 2 diabetes; Human; Pancreatic hormone; American; Nutrition; Prediction; Metabolic diseases; Risk; Glucose tolerance test; Concentration; Insulin; Japanese; Risk factor; Endocrinology
• ISSN: 0149-5992; 1935-5548
• DOI: 10.2337/dc12-0246

To examine whether the patterns of insulin concentration during the oral glucose tolerance test (OGTT) predict type 2 diabetes. We followed 400 nondiabetic Japanese Americans for 10-11 years. Insulin concentrations at 30, 60, and 120 min during a 2-h 75-g OGTT at baseline were used to derive the following possible patterns of insulin: pattern 1 (30-min peak, higher insulin level at 60 than at 120 min), pattern 2 (30-min peak, lower or equal level at 60 vs. 120 min), pattern 3 (60-min peak); pattern 4 (120-min peak, lower level at 30 than at 60 min), and pattern 5 (120-min peak, equal or higher level at 30 vs. 60 min). Insulin sensitivity was estimated by homeostasis model assessment of insulin resistance (HOMA-IR) and Matsuda index. Insulin secretion was estimated by the insulinogenic index (IGI) [Δinsulin/Δglucose (30-0 min)] and disposition index (IGI/HOMA-IR). There were 86 incident cases of type 2 diabetes. The cumulative incidence was 3.2, 9.8, 15.4, 47.8, and 37.5% for patterns 1, 2, 3, 4, and 5, respectively. Compared with pattern 1, patterns 4 and 5, characterized by a lasting late insulin response, were associated with significantly less insulin sensitivity as measured by the Matsuda index and lower early insulin response by the disposition index. The multiple-adjusted odds ratios of type 2 diabetes were 12.55 (95% CI 4.79-32.89) for pattern 4 and 8.34 (2.38-29.27) for pattern 5 compared with patterns 1 and 2. This association was independent of insulin secretion and sensitivity. The patterns of insulin concentration during an OGTT strongly predict the development of type 2 diabetes.