Since 1980 we have performed the identical Greenville gastric bypass (GGB) procedure on 479 morbidly obese patients with an acceptable morbidity and a mortality rate of 1.2%. The weight loss in the series was well maintained over the follow-up period of 10 y. The GGB can control non-insulin-dependent diabetes mellitus (NIDDM) in most patients. The group of 479 patients included 101 (21%) with NIDDM and another 62 (13%) who were glucose impaired. Of these 163 individuals, 141 reverted to normal and only 22 (5%) remained with inadequate control of their carbohydrate metabolism. Those patients who were older or whose diabetes was of longer duration were less likely to revert to normal values. The gastric bypass operation is an effective approach for the treatment of morbid obesity. Along with its control of weight, the operation also controls the hyperglycemia, hyperinsulinemia, and insulin resistance of the majority of patients with either glucose impairment or frank NIDDM.

OBJECTIVE -To determine if weight loss may prevent conversion of impaired glucose tolerance (IGT) to diabetes, because weight loss reduces insulin resistance. The prevalence of 1GT in the U.S. population is estimated at 11.2%, more than twice that of diabetes. Furthermore, because an oral glucose tolerance test is needed for its detection, most of these patients are undiagnosed. Screening for IGT would be meaningful if progression to diabetes could be delayed or prevented. RESEARCH DESIGN A N D METHODS-For an average of 5.8 years (range 2-10 years), 136 individuals with IGT and clinically severe obesity (>45 kg excess body weight) were followed. The experimental group included 109 patients with IGT who underwent bariatric surgery for weight loss. The control group was made up of 27 subjects with IGT who did not have bariatric surgery. The criteria of the World Health Organization was used to detect IGT and diabetes in this population. The main outcome measure of this nonrandomized control trial is the incidence density, or number of events (development of diabetes) divided by the time of exposure to risk. RESULTS-Of the 27 subjects in the control group, 6 developed diabetes during an average of 4.8 ± 2.5 years of postdiagnosis follow-up, yielding a rate of conversion to diabetes of 4.72 cases per 100 person-years. The 109 individuals of the experimental group were followed for an average of 6.2 ±2.5 years postbariatric surgery. Based on the 95% confidence interval of the comparison group, we would expect to find that between 22 and 36 subjects in the experimental group developed diabetes over the follow-up period. Only 1 of the 109 experimental-group patients developed diabetes, resulting in a conversion rate of the experimental group of only 0.15 cases per 100 person-years, which is significantly lower (P < 0.0001) than the control group.C O N C L U S I O N S -Weight loss in patients with clinically severe obesity prevents the progression of IGT to diabetes by >30-fold.From the

The purpose of this study was to determine if 14 wk of exercise training would increase insulin-sensitive glucose transporter protein (GLUT-4) concentration in skeletal muscle of previously sedentary middle-aged men (47.2 +/- 1.3 yr; n = 13). Muscle samples (lateral gastrocnemius) and insulin action [insulin sensitivity index (ISI), minimal model] were obtained in the sedentary condition and 48 h after the final training bout. GLUT-4 protein concentration increased (P < 0.001, 2,629 +/- 331 to 4,140 +/- 391 absorbance units/100 micrograms protein) with exercise training by 1.8-fold. ISI increased by twofold (P < 0.05, 2.1 +/- 0.5 to 3.4 +/- 0.7 SI x 10(5) min/pM) with training. The percentage of GLUT-4 rich type IIa muscle fibers increased by approximately 10% (P < 0.01), which may have contributed to the elevation in transporter protein. GLUT-4 concentration and citrate synthase activity (1.7-fold, P < 0.001) also increased by similar increments. These findings indicate that GLUT-4 protein concentration is elevated in middle-aged individuals with exercise training.

A major defect contributing to impaired insulin action in human obesity is reduced glucose transport activity in skeletal muscle.This study was designed to determine whether the improvement in whole body glucose disposal associated with weight reduction is related to a change in skeletal muscle glucose transport activity and levels of the glucose transporter protein GLUT4. Seven morbidly obese (body mass index = 45.8±2.5, mean ± SE) patients, including four with non-insulin-dependent diabetes mellitus (NIDDM), underwent gastric bypass surgery for treatment of their obesity. In vivo glucose disposal during a euglycemic clamp at an insulin infusion rate of 40 mU/mi per min was reduced to 27% of nonobese controls (P < 0.01) and improved to 78% of normal after weight loss of 43.1±3.1 kg (P < 0.01). Maximal insulin-stimulated glucose transport activity in incubated muscle fibers was reduced by _ 50% in obese patients at the time of gastric bypass surgery but increased twofold (P < 0.01) to 88% of normal in five separate patients after similar weight reduction. Muscle biopsies obtained from vastus lateralis before and after weight loss revealed no significant change in levels of GLUT4 glucose transporter protein. These data demonstrate conclusively that insulin resistance in skeletal muscle of mobidly obese patients with and without NIDDM cannot be causally related to the cellular content of GLUT4 protein. The results further suggest that morbid obesity contributes to whole body insulin resistance through a reversible defect in skeletal muscle glucose transport activity. The mechanism for this improvement may involve enhanced transporter translocation and/or activation. (J. Clin.

To determine the cellular basis for insulin resistance observed in patients with uremia, we investigated insulin action in vivo and in vitro using skeletal muscle obtained from patients with chronic renal failure. Uremic subjects had significantly reduced rates of insulin-stimulated glucose disposal, as determined by a 3-h intravenous glucose tolerance test and using the hyperinsulinemic euglycemic clamp technique. Hepatic glucose production was similar before (control, 76.2 +/- 6.3 vs. uremic, 74.2 +/- 6.9 mg.kg-1.min-1) and during insulin infusion at 40 mU.m-2.min-1 (control, -60.9 +/- 6.6 vs. uremic, -53.9 +/- 6.3 mg.kg-1.min-1). In incubated human skeletal muscle fiber strips, basal 2-deoxy-D-glucose transport was unchanged in uremic subjects compared with controls. However, the increase in insulin-stimulated glucose transport was significantly reduced by 50% in muscles from uremic patients (P = 0.012). In partially purified insulin receptors prepared from skeletal muscle, 125I-labeled insulin binding, beta-subunit receptor autophosphorylation, and tyrosine kinase activity were all unchanged in uremic subjects. The abundance of insulin-sensitive (muscle/fat, GLUT-4) glucose transporter protein measured by Western blot using Mab 1F8 or polyclonal antisera was similar in muscles of control and uremic patients. These findings suggest that the insulin resistance observed in skeletal muscle of uremic patients cannot be attributed to defects in insulin receptor function or depletion of the GLUT-4 glucose transporter protein. An alternative step in insulin-dependent activation of the glucose transport process may be involved.