The effects of postprandial glucose and insulin levels on postprandial endothelial function in subjects with normal glucose tolerance

BACKGROUND Previous studies have demonstrated that postprandial hyperglycemia attenuates brachial artery flow-mediated dilation (FMD) in prediabetic patients, in diabetic patients, and even in normal subjects. We have previously reported that postprandial hyperinsulinemia also attenuates FMD. In the present study we evaluated the relationship between different degrees of postprandial attenuation of FMD induced by postprandial hyperglycemia and hyperinsulinemia and differences in ingested carbohydrate content in non-diabetic individuals. METHODS Thirty-seven healthy subjects with no family history of diabetes were divided into 3 groups: a 75-g oral glucose loading group (OG group) (n = 14), a test meal group (TM group) (n = 12; 400 kcal, carbohydrate content 40.7 g), and a control group (n = 11). The FMD was measured at preload (FMD0) and at 60 minutes (FMD60) and 120 (FMD120) minutes after loading. Plasma glucose (PG) and immunoreactive insulin (IRI) levels were determined at preload (PG0, IRI0) and at 30 (PG30, IRI30), 60 (PG60, IRI60), and 120 (PG120, IRI120) minutes after loading. RESULT Percentage decreases from FMD0 to FMD60 were significantly greater in the TM group (-21.19% ± 17.90%; P < 0.001) and the OG group (-17.59% ± 26.64%) than in the control group (6.46% ± 9.17%; P < 0.01), whereas no significant difference was observed between the TM and OG groups. In contrast, the percentage decrease from FMD0 to FMD120 was significantly greater in the OG group (-18.91% ± 16.58%) than in the control group (6.78% ± 11.43%; P < 0.001) or the TM group (5.22% ± 37.22%; P < 0.05), but no significant difference was observed between the control and TM groups. The FMD60 was significantly correlated with HOMA-IR (r = -0.389; P < 0.05). In contrast, FMD120 was significantly correlated with IRI60 (r = -0.462; P < 0.05) and the AUC of IRI (r = -0.468; P < 0.05). Furthermore, the percentage change from FMD0 to FMD120 was significantly correlated with the CV of PG (r = 0.404; P < 0.05), IRI60 (r = 0.401; p < 0.05) and the AUC of IRI (r = 0.427; P < 0.05). No significant correlation was observed between any other FMDs and glucose metabolic variables. CONCLUSION Differences in the attenuation of postprandial FMD induced by different postprandial insulin levels may occur a long time postprandially but not shortly after a meal.