Effects of Weight Loss on Mechanisms of Hyperglycemia in Obese Non-Insulin-Dependent Diabetes Mellitus

Abstract

To quantitate the effects of weight loss on the mechanisms responsible for hyperglycemia in non-insulin-dependent diabetes mellitus (NIDDM), eight obese subjects with NIDDM were studied before and after weight reduction with posttreatment assessment after 3 wks of isocaloric (weight maintenance) refeeding. After weight loss of 16.8 ± 2.7 kg (mean ± SE), the fasting plasma glucose level decreased 55% from 277 ± 21 to 123 ± 8 mg/dl. The individual fasting glucose levels were highly correlated with the elevated basal rates of hepatic glucose output (HGO) (r = 0.91, P < .001), which fell from 138 ± 11 to 87 ± 5 mg · m2 · min1 after weight loss. The change in fasting plasma glucose also correlated significantly with the change in the basal rates of HGO (r = 0.74, P < .05). This was associated with reduced fasting serum levels of glucagon (from 229 ± 15 to 141 ± 12 pg/ml), reduced free fatty acids (from 791 ± 87 to 379 ± 35 μeq/L), and unchanged basal insulin levels (17 ± 4 to 15 ± 2 fill/ml). Peripheral glucose disposal, assessed by the euglycemic glucose-clamp technique, at insulin infusion rates of 120 and 1200 mU · m2 · min1 increased between 135 and 165%, from 128 ± 17 to 288 ± 24 mg · m2 · min1 during the 120-mU · m2 · min1 studies and from 159 ± 19 to 318 ± 24 mg · m2 · min1 during the 1200-mU · m2 · min1 clamp studies, despite comparable steady-state serum insulin levels at each infusion rate before and after weight loss. HGO during the 120-mU · m2 · min1 clamp studies increased from 85% to complete (100%) suppression after treatment. Adipocyte size was reduced 44% (851 ± 91 to 475 ± 48 pi), whereas surface area decreased by 32% (4.30 × 104 to 2.92 × 104 μm2cell). Insulin binding to isolated adipocytes was unchanged, whereas basal in vitro rates of 3-O-methylglucose transport increased from 0.21 ± 0.13 to 0.53 ± 0.24 pmol/(2 × 109 μm2) × (10 s−1) and maximal glucose transport rates increased from 0.64 ± 0.29 to 1.18 ± 0.48 pmol/(2 × 105 cells) × (10 s−1) and 0.42 ± 0.20 to 1.04 ± 0.30 pmol/(2 × 109 μm2) × (10 s−1). Finally, absolute serum insulin levels during oral glucose-tolerance tests and meal-tolerance tests were unchanged by weight reduction, whereas plasma glucose levels were markedly reduced. We conclude that in obese NIDDM, weight loss results in improved glucose homeostasis with 1) reducedbasal HGO predominently responsible for the lowering of fasting glucose levels; 2) improved postprandial glucose excursions with marked amelioration of peripheral insulin resistance due to improved postreceptor insulin action, which is at least partly due to enhanced glucose transport system activity; and 3) unchanged absolute insulin levels in the face of markedly reduced glycemia, indicative of enhanced p-cell sensitivity to insulinogenic stimuli.