The relationship between plasma glucose and insulin responses to oral glucose, LDL oxidation, and soluble intercellular adhesion molecule-1 in healthy volunteers.

This study was initiated to describe the relationships between plasma glucose and insulin responses to oral glucose and the concentrations of partially oxidized low density lipoprotein (poxLDL) and soluble intercellular adhesion molecule-1 (sICAM-1) in 23 healthy, non-diabetic volunteers. Results demonstrated that plasma glucose (r=0.65, P<0.002) and insulin (r=0.58, P<0.007) responses to a 75-g oral glucose challenge were highly correlated to poxLDL concentrations. Plasma glucose (r=0.63, P<0.002) and insulin (r=0.68, P<0.001) concentrations also significantly correlated with sICAM-1 concentrations. Furthermore, concentrations of poxLDL and sICAM-1 were significantly related (r=0.55, P<0.001). These relationships remained statistically significant when adjusted for differences in age, gender, body mass index, and lipoprotein concentrations. These results provide further evidence that circulating LDL particles are more highly oxidized in insulin resistant states, and demonstrate the presence of an in vivo relationship between insulin resistance, LDL oxidized state, and sICAM-1 concentrations. These results help explain why soluble forms of adhesion molecules are increased in clinical conditions characterized by insulin resistance, and support the possibility that LDL oxidizability is increased in insulin resistant subjects, and that the increase in sICAM-1 results from stimulation of cellular adhesion molecules by more highly oxidized LDL.

Relationship between several surrogate estimates of insulin resistance and quantification of insulin-mediated glucose disposal in 490 healthy nondiabetic volunteers.

OBJECTIVE: The goal of this study was to define the relationship between a quantitative measure of the ability of physiological hyperinsulinemia to stimulate glucose disposal and several surrogate measures of insulin resistance. RESEARCH DESIGN AND METHODS: Insulin-mediated glucose disposal was quantified in 490 healthy nondiabetic volunteers by determining the steady-state plasma glucose (SSPG) concentration in response to a continuous infusion of somatostatin, insulin, and glucose. Because the steady-state plasma insulin concentration was similar in all subjects during the infusion (approximately 60 microU/ml), the SSPG concentration provided a direct estimate of insulin-mediated glucose disposal. Relationships between this specific measure of insulin resistance and several surrogate estimates of insulin resistance based on plasma glucose and insulin concentrations were then defined. RESULTS: The surrogate measure of insulin resistance most closely related to the direct determination of insulin action was the total integrated insulin response to a 75-g oral glucose challenge with correlation coefficients (r) varying from 0.67 to 0.79. Fasting plasma insulin concentration was significantly correlated (r = 0.61, P<0.001) to the specific estimate of insulin action. Two other surrogate estimates of insulin action, the ratio of fasting glucose-to-fasting insulin concentration and the homeostasis model assessment for insulin resistance, were no more closely related to SSPG than the fasting plasma insulin concentration. CONCLUSIONS: The total integrated insulin response to oral glucose is the best surrogate measure of insulin resistance, accounting for approximately two-thirds of the variability in insulin-mediated glucose disposal. Fasting insulin concentration accounted for approximately one-third of the variability in insulin-mediated glucose disposal, and the use of fasting plasma glucose and insulin concentrations to calculate more sophisticated estimates of insulin resistance appears to offer little advantage over the fasting plasma insulin concentration. Given the large number of nondiabetic individuals in this study, the results should have general application in population-based studies, providing evidence for both the utility and limitation of the use of these surrogate measures.

Comparison of fasting plasma leptin concentrations in healthy subjects with high and low plasma insulin.

This study was initiated to evaluate the role of hyperinsulinemia in the regulation of fasting plasma leptin. We measured plasma leptin and insulin concentrations in 404 healthy nondiabetic subjects. For analytical purposes, the population was divided into quartiles on the basis of the lowest (quartile 1) and highest (quartile 4) plasma insulin response to oral glucose, and fasting plasma leptin values in these 2 dichotomous groups were compared. The total plasma integrated insulin response was 4-fold greater in quartile 4, associated with significantly higher (P < .001) fasting plasma leptin (12.60+/-0.85 v8.53+/-0.56 ng/mL). Fasting plasma leptin concentrations remained significantly higher in the hyperinsulinemic quartile when comparisons were made after subdividing the population on the basis of gender, body mass index (BMI), or waist to hip ratio (WHR). These results demonstrate that fasting plasma leptin concentrations are significantly higher in hyperinsulinemic individuals, and this difference is independent of either overall or central obesity.

Plasma insulin concentration is more tightly linked to plasma leptin concentration than is the body mass index.

This study tested the hypothesis that the integrated plasma insulin response to oral glucose is a more sensitive predictor of the fasting plasma leptin concentration than the body mass index (BMI) or waist to hip ratio (WHR). For this purpose, we determined the fasting plasma leptin concentration and plasma insulin response to a 75-g oral glucose challenge in 76 healthy female subjects, with a BMI of 19.1 to 36.6 kg/m2 and a WHR of 0.57 to 1.1. The results demonstrated that fasting plasma leptin concentrations were significantly correlated with both the BMI (r = .64, P < .001) and the plasma insulin response to glucose (r = .61, P < .001), but not with the WHR (r = .27). Since the BMI and the insulin response were also significantly related (r = .34, P = .003), multivariate analysis was performed to determine if the BMI and insulin response were independent determinants of the fasting leptin concentration. This analysis indicated that both the BMI and insulin response were significantly related to plasma leptin (P < .001). To pursue this issue further, the population was divided into tertiles on the basis of the (1) plasma leptin concentration, (2) BMI, and (3) integrated insulin response. The two variables that were most closely linked to each other were the leptin concentration and insulin response. In contrast, the BMI was relatively easily disassociated from the other two variables. These results indicate that while both the plasma insulin response to glucose and the BMI are significantly associated with the fasting plasma leptin concentration, the plasma insulin response appears more closely associated with the plasma leptin concentration.

Alzheimer disease and vascular dementia: relationships with fasting glucose and insulin levels.

Cerebrovascular disease and Alzheimer disease are the leading causes of dementia in elderly subjects. In spite of it, relatively little is known about the pathogenesis and risk factors for dementia. We evaluated fasting plasma glucose and insulin, albumin, lipids, Lp(a) and uric acid levels in nondiabetic patients of both sexes affected by vascular dementia (VD) and senile dementia of the Alzheimer type (SDAT) as well as in a control group of age-matched nondemented subjects. Following a covariance analysis by gender, body mass index, albumin levels and prevalence of arterial hypertension, total and LDL cholesterol as well as HDL cholesterol levels were not significantly different among the three groups. Fasting glucose (p < 0.001 and p < 0.005, respectively) and insulin levels (p < 0.05 for both differences) were higher in patients with VD and SDAT than in control subjects. Our data show that nondiabetic patients with VD or SDAT have higher fasting glucose and insulin levels than healthy control subjects. These metabolic characteristics were not influenced by differences in gender, adiposity, nutritional status, lipids or presence of arterial hypertension.

Roles of insulin resistance and obesity in regulation of plasma insulin concentrations.

Plasma glucose, insulin, and C-peptide concentrations were determined in response to graded infusions of glucose, and insulin secretion rates were calculated over each sampling period. Measurements were also made of insulin clearance, resistance to insulin-mediated glucose, uptake, and the plasma glucose, insulin, and C-peptide concentrations at hourly intervals from 8:00 AM to 4:00 PM in response to breakfast and lunch. Plasma glucose, insulin, and C-peptide concentrations were significantly (P < 0.01) higher in obese women in response to the graded intravenous glucose infusion, associated with a 40% (P < 0.005) greater insulin secretory response. Degree of insulin resistance correlated positively (P < 0.05) with the increase in insulin secretion rate in both nonobese (r = 0.52) and obese (r = 0.58) groups and inversely (P < 0.05) with the decrease in insulin clearance in obese (r = -0.46) and nonobese (r = -0.39) individuals. Weight loss was associated with significantly lower plasma glucose, insulin, and C-peptide concentrations in response to graded glucose infusions and in day-long insulin concentrations. Neither insulin resistance nor the insulin secretory response changed after weight loss, whereas there was a significant increase in the rate of insulin clearance during the glucose infusion. It is concluded that 1) obesity is associated with a shift to the left in the glucose-stimulated insulin secretory dose-response curve as well as a decrease in insulin clearance and 2) changes in insulin secretion and insulin clearance in obese women are more a function of insulin resistance than obesity.

Can changes in plasma insulin concentration explain the variability in leptin response to weight loss in obese women with normal glucose tolerance?

The aim of this study was to test the hypothesis that the fall in circulating insulin concentration associated with moderate weight loss determines the associated decrease in plasma leptin concentration. For this purpose, 12 healthy, nondiabetic, obese women were studied before and after an average weight loss of 9.5 kg (11.2% of initial body weight). Plasma leptin concentrations fell from a mean (+/-SE) value of 35 +/- 3 to 17 +/- 2 ng/mL (P < 0.001) in association with the loss of weight. However, there was no correlation between the decline in leptin concentration and the associated fall in weight, body mass index, fat mass, or percent body fat. Furthermore, no correlation was seen among changes in fasting plasma glucose or insulin concentrations, the 8-h integrated plasma glucose response to breakfast and lunch, or the estimate of insulin-mediated glucose disposal. The only measured variable that correlated with the fall in plasma leptin concentration (r = 0.78; P < 0.005) was the decline in the 8-h integrated plasma insulin response after weight loss (from 304 +/- 44 to 232 +/- 36 microU/8 h x mL; P < 0.001). Finally, multivariate regression analysis, using various estimates of degree of obesity, insulin resistance, integrated glucose response, and integrated insulin response as dependent variables, indicated that only the insulin response was independently related to the decrease in leptin concentration (P = 0.035). The fall in integrated insulin response accounted for 66% of the variance in leptin concentrations after weight loss, and this was true no matter what the estimate of change in degree of obesity. In addition to offering an explanation for the variance in postweight loss leptin concentrations, these data provide further evidence of the importance of ambient insulin concentrations in the regulation of plasma leptin concentrations.

Differences in insulin resistance do not predict weight loss in response to hypocaloric diets in healthy obese women.

The current study was initiated to determine whether insulin resistance and/or hyperinsulinemia affected the ability of obese individuals to lose weight in response to hypocaloric diets. Thirty-one obese, nondiabetic women, with values for body mass index ranging from 28.0-35.0 kg/m2, volunteered for this program. Resistance to insulin-mediated glucose disposal was assessed by determining their steady state plasma insulin and glucose concentration during the last 30 min of a 180-min infusion of somatostatin, insulin, and glucose. The total integrated insulin response to breakfast and lunch was also determined. After the baseline measurements, volunteers were placed on a hypocaloric diet calculated to lead to a minimum weekly loss of 1% of ideal body weight. Individuals who met the criteria after 30 days of dieting were defined as weight loss successes (n = 20) and continued on the diet for another 30 days. Individuals not meeting the criteria were designated as weight loss failures (n = 12) and were discharged from the study. There was a mean (+/-SEM) weight loss at 60 days of 9.2 +/- 0.4 kg in the 20 individuals defined as weight loss successes, but there was no correlation between weight loss and either steady state plasma glucose or the total integrated insulin response (r < 0.1; P > 0.83). Furthermore, using the same criteria to define insulin sensitivity and insulin resistance as those for therapeutic successes, the therapeutic failures comprised six insulin-sensitive and five insulin-resistant subjects. In summary, insulin-mediated glucose disposal varied widely in nondiabetic, obese women, and there was no relationship between baseline insulin resistance or total integrated insulin response and weight loss. It is concluded that the ability to lose weight on a calorie-restricted diet over a short time period does not vary in obese, healthy women as a function of insulin resistance or hyperinsulinemia.

Low density lipoprotein cholesterol, lipoprotein(a), and apo(a) isoforms in the elderly: relationship to fasting insulin. Associazione Medica Sabin.

BACKGROUND AND AIM: Insulin resistance/hyperinsulinemia are often associated with aging and could play an important role in the development of glucose intolerance and dyslipidemia in the elderly. We investigated the relationship between plasma fasting insulin with total cholesterol (TC) and low density lipoprotein LDL cholesterol (LDL-C), triglycerides (TG), lipoprotein(a) [Lp(a)] levels apolipoprotein (a) [apo (a)] isoforms in 100 free-living "healthy" octo-nonagenarians. METHODS AND RESULTS: Fasting insulin was positively correlated with TG, whereas a negative relation was found with TC and LDL-C (r = -0.29 and r = -0.28 respectively; p < 0.01), LDL-C/apo B, HDL-C and apo A-I levels. Fasting insulin was also inversely correlated with Lp(a) levels (r = -0.22; p < 0.03), whereas the latter were significantly related with TC and LDL-C (r = 0.30 and r = 0.31; p < 0.005), TG (r = 0.21; p < 0.05) and apo B (r = 0.26; p < 0.02). There was a negative relation between Lp(a) levels and apo(a) isoforms: the greater the apo(a) molecular weight, the lower the Lp(a) level (p < 0.0001). Fasting insulin increased with apo(a) size, though the difference in insulin levels among apo(a) isoforms was not significant (p = 0.4). Multiple regression analysis showed that fasting insulin was the best predictor of LDL-C (R2 = 0.14; p = 0.002) irrespective of age, gender, BMI, waist circumference and TG, while apo(a) isoform size, BMI and waist circumference were related with Lp(a) irrespective of TC and LDL-C, TG and apo B (R2 = 0.35 to 0.37; p < 0.0001). CONCLUSIONS: These results suggest that fasting insulin levels significantly influence LDL-C metabolism in old age. Lp(a) levels seem to be very strongly related to genetic background, although an indirect relation with insulin through adiposity and/or other associated lipid abnormalities cannot be ruled out.

Further evidence for a central role of adipose tissue in the antihyperglycemic effect of metformin.

OBJECTIVE: To evaluate further the relative roles played by liver and adipose tissue in the therapeutic response to metformin in patients with type 2 diabetes. RESEARCH DESIGN AND METHODS: A total of 11 patients with diet-treated type 2 diabetes were given metformin for approximately 3 months. Measurements were made before and after treatment of 1) fasting and postprandial plasma glucose, insulin, and free fatty acid (FFA) concentrations; 2) glucose appearance (Ra) and disappearance (Rd) rates measured overnight with 3-[3H]glucose; and 3) plasma FFA concentrations during a 195-min infusion period at relatively low insulin (approximately 12-24 microU/ml) concentrations. RESULTS: Mean +/- SEM fasting plasma glucose concentration was significantly lower (175 +/- 11 vs. 224 +/- 15 mg/dl; P < 0.01) after treatment with metformin. Mean +/- SEM insulin concentrations measured from 8:00 A.M. to 5:00 P.M. did not change with treatment. However, both glucose and FFA concentrations were significantly lower (P < 0.01) when measured over the same time period, and the decreases in plasma FFA and glucose concentration were highly correlated (r = 0.81; P = 0.03). Overnight glucose turnover studies indicated that neither Ra (hepatic glucose production [HGP]) nor Rd changed significantly with treatment in association with metformin treatment. Since plasma glucose concentration was much lower after metformin treatment, the overnight glucose metabolic clearance rate (MCR) was significantly lower (P < 0.01). Finally, the ability of insulin to inhibit isoproterenol-stimulated increases in plasma FFA concentration was enhanced in metformin-treated patients (P < 0.05). CONCLUSIONS: Metformin treatment was associated with significantly lower fasting plasma glucose concentrations and lower day-long plasma glucose and FFA concentrations. Although overnight HGP was unchanged after treatment with metformin, the overnight glucose MCR was significantly increased, and the antilipolytic activity of insulin was also enhanced. Given these findings, it is suggested that at least part of the antihyperglycemic effect of metformin is due to a decrease in release of FFA from adipose tissue, leading to lower circulating FFA concentrations and an increase in glucose uptake.

Extreme but asymptomatic hypergastrinemia with gastroparesis in a young woman with insulin dependent diabetes mellitus.

Overt diabetic gastroparesis is a rare long-term complication of diabetes, probably resulting from autonomic neuropathy of vagus nerve. It is now clear that neural damage plays a pivotal role in the pathogenesis of the disease. Some studies showed high basal gastrin levels in patients with diabetic gastroparesis, but the clinical meaning of this observation is still unclear. We report the case of a young woman with Insulin Dependent Diabetes Mellitus (IDDM) who was referred to evaluate nausea and vomiting associated to ketoacidosis. Our hypothesis of autonomic neuropathy with gastroparesis was confirmed. We observed a progressive increase in fasting gastrin concentration (20-fold normal values) in the absence of any clinical and laboratory signs of Zollinger-Ellison (ZE) syndrome. The increasing vomiting induced a severe state of cachexia, which required total parenteral nutrition for a long period. All therapeutic approaches were unsuccessful, and the patient rapidly died, suggesting a possible link between the severity of the clinical picture and the gastrin plasma levels.

Hematocrit and hemoglobin are independently related to insulin resistance and compensatory hyperinsulinemia in healthy, non-obese men and women.

In this study, we evaluated the relationship between resistance to insulin-mediated glucose disposal and hematocrit (Hct) and hemoglobin (Hgb) concentrations in 150 normal, healthy volunteers: 100 men and 50 women. Insulin resistance was defined as the steady-state plasma glucose (SSPG) concentration at the end of a 180-minute infusion of somatostatin, insulin, and glucose. Since the steady-state plasma insulin (SSPI) concentrations are similar in all individuals, the SSPG concentrations provide a direct measure of insulin resistance: the higher the SSPG, the more insulin-resistant the subject. The results indicated that SSPG was significantly (P < .001) related to Hct and Hgb in both men and women, with correlation coefficients (r) ranging from 0.38 to 0.43. A series of other variables were also related to Hct and Hgb, including blood pressure, plasma glucose and insulin responses to oral glucose, and plasma triglyceride and high-density lipoprotein (HDL) concentrations. When multiple regression analysis was used to evaluate these relationships, the only variables that were consistently found to be associated with Hct and Hgb were insulin resistance and plasma insulin response to oral glucose. Thus, these results suggest that Hct and Hgb concentrations be added to the cluster of variables related to insulin resistance and compensatory hyperinsulinemia.

Relationship between insulin resistance and partially oxidized LDL particles in healthy, nondiabetic volunteers.

This study was performed in 36 healthy volunteers to define the relationship between plasma concentrations of partially oxidized low density lipoprotein (poxLDL), plasma glucose and insulin responses to oral glucose, and steady-state plasma glucose (SSPG) concentrations after a 180-minute infusion of somatostatin, insulin, and glucose. The concentration of poxLDL was estimated by determining the amount of conjugated dienes formed during in vitro LDL oxidation in the presence or absence of alanine. Under these conditions, the greater the in vitro antioxidant effect of alanine, the lower the amount of poxLDL that was present in plasma. The results demonstrated that plasma poxLDL concentration was significantly correlated with plasma glucose (r=.53, P<.001) and insulin (r=.43, P<.01) responses, SSPG concentrations (r=.53, P<.001), and plasma triglyceride (r=.42, P<.01) and HDL cholesterol (r=-.50, P<.002) concentrations. Furthermore, these relationships persisted when the data were corrected for differences in age, sex, body mass index, and the ratio of waist to hip girth. Of note, there was no correlation between poxLDL and LDL cholesterol concentration. When SSPG was entered along with age, sex, body mass index, and waist-to-hip ratio in a multiple regression model, SSPG alone was a significant prediction of poxLDL (r-=.37, P<.02). The addition of plasma glucose and insulin responses and triglyceride and HDL cholesterol concentrations increased the r2 to only .47. These results show that the amount of poxLDL in plasma is significantly correlated with insulin resistance (ie, SSPG) and its metabolic consequences.

Relationships between fasting plasma insulin, anthropometrics, and metabolic parameters in a very old healthy population. Associazione Medica Sabin.

Several studies have shown that insulin resistance and hyperinsulinemia are associated with many metabolic disorders predisposing to coronary heart disease (CHD). This syndrome has been termed syndrome X. However, it is not completely known whether these relationships are still present in the elderly, or whether other factors such as age, gender, and body fat distribution modulate them. Therefore, we investigated the relationship between fasting plasma insulin, total and regional adiposity, fasting plasma glucose and lipids, plasma plasminogen activator inhibitor-1 (PAI-1), fibrinogen, and coagulation factor VII in a sample of 100 healthy free-living octogenarians-nonagenarians (52 men and 48 women) who were disability-free according to the Katz index. By univariate analysis, fasting insulin correlated positively with all anthropometric measures except the waist to hip ratio (WHR) in women. There was a positive correlation between fasting insulin and fasting glucose (r=.40, P < .01), plasma triglycerides ([TGs] r=.21, P < .05), and PAI-1 levels (r=.33, P < .01), whereas a negative relation was found with high-density lipoprotein cholesterol (HDL-C) and apolipoprotein, A-I (apo A-I) levels (r=-.22 and =-.24, respectively, P < .05). These relationships were weaker and less significant in women. In pooled data, stepwise multiple regression analysis showed an independent relationship of both the body mass index (BMI) and fasting insulin level with TGs (R2=.14), while gender and fasting insulin were the best predictors of HDL-C variance (R2=.17). Furthermore, fasting insulin was the only variable independently related to PAI-1 (R2=.12). Our findings support the existence of a metabolic syndrome even in very old age by showing that high insulin levels are related to various metabolic and hemostatic disorders.

Plasma leptin concentrations do not appear to decrease insulin-mediated glucose disposal or glucose-stimulated insulin secretion in women with normal glucose tolerance.

The aim of this study was to test the hypothesis that plasma leptin concentrations contributed to the pathophysiology of NIDDM by decreasing both insulin-mediated glucose disposal and glucose-stimulated insulin secretion. The study was performed in 60 women with normal oral glucose tolerance. Differences in insulin-mediated glucose disposal were determined by comparing the steady-state plasma glucose (SSPG) concentrations attained at the end of a 180-min constant infusion of somatostatin, glucose, and insulin, while comparisons of glucose-stimulated insulin secretion were based on the incremental increase in insulin concentration 30 min after an oral glucose challenge (deltaIns) as compared with the fasting value. The results showed that the higher the fasting plasma leptin concentration, the greater the degree of insulin resistance (r = 0.47, P < 0.01). Furthermore, multiple regression analysis indicated that the relationship between leptin and SSPG was independent of age and degree of obesity as estimated by BMI. However, since the total integrated plasma insulin response was highly correlated with both SSPG (r = 0.80, P < 0.001) and leptin (r = 0.55, P < 0.01), multiple regression analysis was repeated, adding total insulin response to the model. When this was done, the significant relationship between leptin and SSPG disappeared, whereas both BMI (P < 0.03) and insulin response (P < 0.001) were correlated with SSPG. A significant relationship between leptin and deltaIns was seen, but it was a positive one (r = 0.31, P < 0.02), not a negative one as would be expected if circulating levels of leptin inhibited glucose-stimulated insulin secretion. Furthermore, multiple regression analysis could only confirm an independent relationship between deltaIns and SSPG, but not between deltaIns and leptin. The results of these studies do not support the view that circulating leptin has a primary effect on either insulin action or secretion in normal female volunteers. It seems more likely that chronic hyperinsulinemia in insulin-resistant individuals acts to increase adipose tissue leptin synthesis and secretion, leading to higher ambient leptin concentrations.

Fibrinolytic and coagulation factors in very old subjects: association with lipoprotein profile and anthropometric variables.

BACKGROUND: We evaluated plasminogen activator inhibitor-1 (PAI-1), factor VII activity (FVII), and fibrinogen in a sample of octo-nonagenarians. Furthermore, we investigated the relationship of these fibrinolytic and coagulation parameters with lipoprotein profile and anthropometric variables in the absence or presence of disability. METHODS: We enrolled a population of 162 octo-nonagenarians, divided in two groups on the basis of presence or absence of disability in the activity of daily living (ADL). All the anthropometric determinations were carried out according to standardized methods. Blood samples for hemostatic and lipid determinations were collected after overnight fasting and resting. RESULTS: PAI-1 activity and fibrinogen levels were significantly higher in disabled (DIS) compared to free-living (FL) adults, whereas FVII did not show differences in the two groups. PAI-1 activity and FVII positively correlated to anthropometric parameters (body mass index, subscapular and tricipital skinfold thickness) in both DIS and FL. No correlations were found between fibrinogen and other variables in FL, whereas a negative relation with high density lipoprotein-cholesterol levels emerged in DIS. FVII was positively related with total cholesterol low density lipoprotein-cholesterol, and apolipoprotein B in both FL and DIS. CONCLUSIONS: In a sample of octo-nonagenarians, PAI-1 activity and FVII show a significant correlation with several anthropometric and lipoprotein parameters, suggesting that these variables are strongly associated with body composition and lipid metabolism independent from age and disability. DIS presented higher PAI-1 and fibrinogen levels; this observation may take in account the high prevalence of vascular diseases and also occult inflammation, which are known to affect these parameters.

Results of a placebo-controlled study of the metabolic effects of the addition of metformin to sulfonylurea-treated patients. Evidence for a central role of adipose tissue.

OBJECTIVE: To define the metabolic effects of metformin in the treatment of NIDDM and to evaluate potential mechanisms for its ability to improve glycemic control. RESEARCH DESIGN AND METHODS: Sulfonylurea-treated patients, with inadequate glycemic control, were treated with metformin in either a placebo-controlled or open fashion. Measurements were made of 1) fasting and postprandial plasma glucose, insulin, and free fatty acid (FFA) concentrations; 2) glucose appearance and disappearance rates measured overnight with 3-[3H]glucose; and 3) plasma FFA concentrations during a 45-min infusion period at relatively low (approximately 60 pmol/l) insulin concentrations. RESULTS: Mean +/- SE hourly plasma glucose, insulin, and FFA concentrations were similar before and after treatment in the placebo group. In contrast, mean hourly plasma glucose concentrations were significantly lower (P < 0.005) after metformin treatment in both the placebo-controlled and open-label groups (-3.9 +/- 1.0 and -4.4 +/- 0.8 mmol/l, respectively). Similarly, day-long hourly FFA levels were lower (P < 0.005) following metformin in the placebo-controlled and open-label groups (-87 +/- 35 and -136 +/- 31 mumol/l, respectively). Plasma insulin concentrations did not change with treatment in any group. Overnight glucose turnover studies indicated that neither the rate of glucose appearance (hepatic glucose production) or glucose disappearance changed significantly with treatment in the placebo or metformin groups. Because plasma glucose concentration was much lower after metformin treatment, overnight glucose metabolic clearance rate was significantly (P < 0.001) lower in this group. Finally, plasma FFA concentrations in response to a low-dosage insulin infusion (5 mU.m-2.min-1) were significantly lower after metformin as compared with the placebo-treated group (P < 0.001). CONCLUSIONS: Metformin treatment was associated with significantly lower day-long plasma glucose and FFA concentrations. Although overnight hepatic glucose production was unchanged following treatment with metformin, the overnight glucose metabolic clearance rate significantly increased. Given these findings, it is suggested that at least part of the antihyperglycemic effect of metformin is due to an increase in glucose uptake, secondary to a decrease in release of FFA from adipose tissue, and lower circulating FFA concentrations.

Insulin resistance does not change the ratio of proinsulin to insulin in normal volunteers.

Plasma glucose, insulin, and proinsulin concentrations were measured before and after an oral glucose challenge in 57 nondiabetic individuals. In addition, insulin-mediated glucose disposal was estimated by determining the steady state plasma glucose (SSPG) concentration after a 180-min iv infusion of somatostatin, insulin, and glucose. The plasma glucose concentration after oral glucose administration was used to divide the population into those with normal (n = 36) or impaired glucose tolerance (IGT; n = 21), and the 36 normal glucose-tolerant individuals were further subdivided into an insulin-sensitive (SSPG, < 9.0 mmol/L; n = 15) and an insulin-resistant (SSPG, > 10 mmol/L; n = 21) group. Fasting and postglucose load insulin concentrations were similar in the normal glucose-tolerant insulin-resistant and IGT groups, but were significantly higher (P < 0.02- < 0.001) than those in normal glucose-tolerant insulin-sensitive individuals. Fasting proinsulin concentrations were also higher (P < 0.002) in the normal glucose-tolerant insulin-resistant (15.1 +/- 1.5 pmol/L) and IGT (15.8 +/- 1.8 pmol/L) groups compared to those in normal glucose-tolerant insulin-sensitive volunteers (9.3 +/- 1.2 pmol/ L). However, the ratio of fasting proinsulin to insulin was identical in all three groups (0.12). When the three groups were combined, significant relationships (P < 0.001) existed between SSPG (degree of insulin resistance) and both fasting proinsulin (r = 0.59) and insulin (r = 0.66) concentrations, but not with the ratio of proinsulin to insulin (r = 0.03). These results demonstrate that fasting proinsulin and insulin concentrations are increased in insulin-resistant, nondiabetic subjects, and the more insulin resistant, the greater the increase. In contrast, the ratio of proinsulin to insulin did not vary as a function of insulin resistance. Thus, neither insulin resistance nor the need to secrete more insulin to maintain glucose tolerance necessarily leads to abnormal insulin processing by the beta-cell.

Adherence of mononuclear cells to endothelium in vitro is increased in patients with NIDDM.

OBJECTIVE: To compare the binding to cultured endothelial cells of mononuclear cells isolated from healthy volunteers and patients with NIDDM. RESEARCH DESIGN AND METHODS: Mononuclear cells were isolated from healthy volunteers (n = 11) and patients with NIDDM (n = 14) and incubated with ECV 304 cells, a human umbilical endothelial cell-derived transformed cell line. Following a period of incubation, the adherence of mononuclear cells to endothelial cells was determined. RESULTS: Adherence of mononuclear cells from patients with NIDDM was significantly greater (P < 0.05) than that of cells isolated from the healthy volunteers, and this difference persisted when adjusted for age, sex, and degree of obesity. Mononuclear cell binding to ECV 304 cells correlated significantly with fasting plasma glucose (r = 0.52, P < 0.01), insulin (r = 0.51, P < 0.01), triglyceride (r = 0.54, P < 0.01), and VLDL (r = 0.54, P < 0.01) and HDL cholesterol (r = -0.45, P < 0.05) levels, but not with either total or LDL cholesterol levels or blood pressure. CONCLUSIONS: Since the adherence of mononuclear cells to the endothelium represents the earliest step in atherogenesis, the observation that mononuclear cells from patients with NIDDM bind more avidly to cultured endothelial cells may help explain why accelerated atherosclerosis occurs in patients with NIDDM. The metabolic abnormality, or abnormalities, present in patients with NIDDM that is responsible for the enhanced adhesiveness of mononuclear cells requires further examination.