Rationale, design and methods of the HEALTHY study nutrition intervention component.

The HEALTHY study was a randomized, controlled, multicenter and middle school-based, multifaceted intervention designed to reduce risk factors for the development of type 2 diabetes. The study randomized 42 middle schools to intervention or control, and followed students from the sixth to the eighth grades. Here we describe the design of the HEALTHY nutrition intervention component that was developed to modify the total school food environment, defined to include the following: federal breakfast, lunch, after school snack and supper programs; a la carte venues, including snack bars and school stores; vending machines; fundraisers; and classroom parties and celebrations. Study staff implemented the intervention using core and toolbox strategies to achieve and maintain the following five intervention goals: (1) lower the average fat content of foods, (2) increase the availability and variety of fruits and vegetables, (3) limit the portion sizes and energy content of dessert and snack foods, (4) eliminate whole and 2% milk and all added sugar beverages, with the exception of low fat or nonfat flavored milk, and limit 100% fruit juice to breakfast in small portions and (5) increase the availability of higher fiber grain-based foods and legumes. Other nutrition intervention component elements were taste tests, cafeteria enhancements, cafeteria line messages and other messages about healthy eating, cafeteria learning laboratory (CLL) activities, twice-yearly training of food service staff, weekly meetings with food service managers, incentives for food service departments, and twice yearly local meetings and three national summits with district food service directors. Strengths of the intervention design were the integration of nutrition with the other HEALTHY intervention components (physical education, behavior change and communications), and the collaboration and rapport between the nutrition intervention study staff members and food service personnel at both school and district levels.

Challenges in the discovery and development of new agents for the treatment of obesity.

New pharmacologic agents are being sought to help manage the epidemic of obesity and its consequences. Understanding the challenges of the history of obesity drugs is wise before investing in new obesity agents. Expectations of patients, physicians, and drug company executives are not consistent with characteristics of current agents or most potential new ones. Owing to the complex biology underlying body weight regulation, combinations of agents may be necessary to improve weight loss efficacy.

Relationship between glucose tolerance and glucose-stimulated insulin response in 65-year-olds.

BACKGROUND: Decreased insulin secretion may contribute to the deterioration of glucose tolerance associated with aging. METHODS: We studied the insulin response to a 3-hour hyperglycemic clamp (10 mM) of 19 young (24 +/- 1 y) subjects with normal glucose tolerance and 60 older (65 +/- 1 y) subjects with various levels of glucose tolerance. RESULTS: The noninsulin dependent diabetic (NIDDM) group had a diminished first phase immunoreactive (IR)-insulin response compared to young and nondiabetic older groups (p < .05). The older groups had a lower rate of change in IR insulin concentration during the third hour of hyperglycemia compared to the young group (p < .05). This was not, however, a universal finding, because a decreased third hour response was not seen in a subgroup of older subjects whose glucose tolerance was similar to that of the young group. Another subgroup of older subjects with a decrease in glucose tolerance mild enough to be considered normal by the National Diabetes Group Criteria tended to have both an increase in the early insulin response and a decrease in the third hour response. More severe decreases in glucose tolerance were associated with blunting of the early response. CONCLUSION: Aberrations in early and late phase glucose-stimulated insulin responses appear to be present in older subjects with even mildly decreased glucose tolerance. Some individuals, however, show no evidence of deterioration of glucose tolerance or insulin response to glucose with aging, at least up to age 70 years.

Insulin resistance in aging is related to abdominal obesity.

Studies have shown that insulin resistance increases with age, independent of changes in total adiposity. However, there is growing evidence that the development of insulin resistance may be more closely related to abdominal adiposity. To evaluate the independent effects of aging and regional and total adiposity on insulin resistance, we performed hyperinsulinemic euglycemic clamps on 17 young (21-33 yr) and 67 older (60-72 yr) men and women. We assessed FFM and total and regional adiposity by hydrodensitometry and anthropometry. Insulin-stimulated GDRs at a plasma insulin concentration of approximately 450 pM averaged 45.6 +/- 3.3 mumol.kg FFM-1 x min-1 (mean +/- SE) in the young subjects, 45.6 +/- 10.0 mumol.kg FFM-1 x min-1 in 24 older subjects who were insulin sensitive, and 23.9 +/- 11.7 mumol.kg FFM-1 x min-1 in 43 older subjects who were insulin resistant. Few significant differences were apparent in skin-fold and circumference measurements between young and insulin-sensitive older subjects, but measurements at most central body sites were significantly larger in the insulin-resistant older subjects. Waist girth accounted for > 40% of the variance in insulin action, whereas age explained only 10-20% of the total variance and < 2% of the variance when the effects of waist circumference were statistically controlled. These results suggest that insulin resistance is more closely associated with abdominal adiposity than with age.

The effect of exercise on food intake in men and women.

To study the effect of acute exercise on caloric intake in normal-weight young people, food intake was monitored in 10 men and 10 women during consecutive 5-d periods, one with and one without exercise. Food intake during the exercise period was compared with that during the control period. Caloric intake during the control period was 2467 +/- 165 kcal/d (means +/- SEM) for men and 1831 +/- 103 kcal/d for women. During the exercise period the men increased their caloric intake to 2658 +/- 188 kcal/d and the women's caloric intake remained unchanged, 1830 +/- 91 kcal/d. Caloric intake was not affected by sequence of treatment or duration of protocol. Men responded to 5 d of acute exercise with increased caloric intake (208 +/- 64 kcal/d), which was insufficient to compensate for the caloric cost of exercise (596 kcal/d above resting metabolic rate). Women did not change their caloric intake despite expending 382 kcals/d during exercise. Consequently, both normal-weight men and women were in negative caloric balance during the exercise period.

Effects of treadmill exercise to exhaustion on the insulin response to hyperglycemia in untrained men.

The effects of a single bout of exercise to exhaustion on pancreatic insulin secretion were determined in seven untrained men by use of a 3-h hyperglycemic clamp with plasma glucose maintained at 180 mg/100 ml. Clamps were performed either 12 h after an intermittent treadmill run at approximately 77% maximum O2 consumption or without prior exercise. Arterialized blood samples for glucose, insulin, and C-peptide determination were obtained from a heated hand vein. The peak insulin response during the early phase (0-10 min) of the postexercise clamp was higher (81 +/- 8 vs. 59 +/- 9 microU/ml; P less than 0.05) than in the nonexercise clamp. Incremental areas under the insulin (376 +/- 33 vs. 245 +/- 51 microU.ml-1.min) and C-peptide (17 +/- 2 vs. 12 +/- 1 ng.ml-1.min) curves were also greater (P less than 0.05) during the early phase of the postexercise clamp. No differences were observed in either insulin concentrations or whole body glucose disposal during the late phase (15-180 min). Area under the C-peptide curve was greater during the late phase of the postexercise clamp (650 +/- 53 vs. 536 +/- 76 ng.ml-1.min, P less than 0.05). The exercise bout induced muscle soreness and caused an elevation in plasma creatine kinase activity (142 +/- 32 vs. 305 +/- 31 IU/l; P less than 0.05) before the postexercise clamp. We conclude that in untrained men a bout of running to exhaustion increased pancreatic beta-cell insulin secretion during the early phase of the hyperglycemic clamp. Increased insulin secretion during the late phase of the clamp appeared to be compensated by increased insulin clearance.

Insulin secretory capacity in endurance-trained and untrained young men.

Insulin secretion in response to glucose stimulation is reduced in endurance-trained humans. In this study, a modified hyperglycemic clamp, with a superimposed arginine infusion and fat meal, was performed on eight endurance-trained and nine untrained men to determine whether insulin secretory capacity is reduced by exercise training. Raising the plasma glucose concentration to approximately 450 mg/dl resulted in a plasma insulin response in the trained men that was approximately 64% lower than that of the untrained (peak values: 54 +/- 8 vs. 149 +/- 35 microU/ml; P less than 0.001). When a primed continuous infusion of arginine was superimposed on the hyperglycemia, the plasma insulin response was also markedly lower (66%) in the trained subjects, reaching peak values of 333 +/- 68 and 974 +/- 188 microU/ml for trained and untrained subjects, respectively (P less than 0.005). When insulin secretion was further stimulated during the arginine-infused hyperglycemia by the ingestion of a high-fat meal, peak insulin concentrations averaged 989 +/- 205 microU/ml in the trained compared with 2,232 +/- 455 microU/ml in the untrained subjects (P less than 0.01). The response of gastric inhibitory polypeptide (GIP) to the fat meal was delayed and blunted, suggesting that some enteric factor(s) other than GIP mediated the insulinotropic effect of the fat meal. The reduced plasma insulin response in trained people to the stimuli investigated suggests that regular exercise produces either several adaptations within the beta-cell or a single alteration of the beta-cell that results in an attenuation of the insulin secretory response to glucose, arginine, and fat ingestion.

Epinephrine's effect on metabolic rate is independent of changes in plasma insulin or glucagon.

Epinephrine's effect to increase metabolic rate is accompanied by changes in the plasma concentrations of insulin, glucagon, and metabolic substrates. Because both glucagon and insulin have been reported to affect thermogenesis, these hormones might contribute to or modify the thermogenic response to epinephrine. To determine if the epinephrine-induced increase in metabolic rate is secondary to changes in glucagon or insulin or to changes in the fuels modulated by these hormones, metabolic rate was measured by indirect calorimetry in five normal weight post-absorptive young men on three occasions: study A, an intravenous epinephrine infusion alone; study B, a 4-h "islet clamp" consisting of somatostatin infusion with basal insulin and glucagon replacement; and study C, an intravenous epinephrine infusion combined with the islet clamp. A 1-h base-line period preceded 2 h of epinephrine infusion. During the 4-h islet clamp (study B), metabolic rate and plasma concentrations of epinephrine, insulin, glucagon, and glucose remained unchanged. During the infusion of epinephrine alone (study A), metabolic rate and concentrations of glucagon, free fatty acids, and C-peptide increased as expected. Also as expected, the glycemic response to epinephrine infusion was much larger when insulin and glucagon levels were fixed with the islet clamp (study C). In contrast, the metabolic rate and the free fatty acid concentration responded similarly to epinephrine infusion when insulin and glucagon were fixed (study C) and when they were changing (study A). We conclude that epinephrine increases metabolic rate independently of physiological changes in plasma glucagon or insulin or the circulating fuels they modulate.

Measurement of fat distribution by magnetic resonance imaging.

Because of the important role of intra-abdominal fat in predicting increased risk for diabetes, hypertension, and heart disease, methods to quantify intra-abdominal fat are needed. Computed tomography defines quantity of intra-abdominal fat but is associated with significant radiation risk. We explored using magnetic resonance imaging (MRI) to measure amount and distribution of intra-abdominal fat. Because MRI has no known risk, the same subject can be studied repeatedly. Six subjects with percent body fat ranging from 14% to 44% had MRI scans of the chest, abdomen, and thigh on two separate occasions. Total abdominal fat and subcutaneous abdominal fat correlated with percent total body fat as determined from hydrostatic weighing (r = .99, P less than .001). Intra-abdominal fat correlated with the ratio of widest abdominal to widest hip circumference (r = .85, P less than .05). Reproducibility of the MRI measurements of fat was less than 3% for total body areas, less than 5% for subcutaneous fat areas, and less than 10% for internal fat areas. Reproducibility was better in individuals with higher percent total body fat. We conclude that MRI can reliably measure fat areas with no radiation risk to the patient.

Effects of lack of exercise on insulin secretion and action in trained subjects.

We employed the hyperglycemic clamp technique to investigate the effects of short-term inactivity on insulin secretion in nine (8 men, 1 woman) well-trained subjects. A 3-h hyperglycemic clamp (plasma glucose approximately 180 mg/100 ml) was performed approximately 16 h after a usual training bout and again 14 days after stopping exercise training. There was no significant change in body composition during this short period of inactivity. The mean plasma insulin response to an identical glycemic stimulus was 67% higher after 14 days without exercise (45 +/- 7 after vs. 27 +/- 4 microU/ml before stopping exercise training). Marked increases in the early (0-10 min, 150 +/- 28 vs. 101 +/- 15 microU.ml-1.min) and late (10-180 min, 6,051 +/- 1,257 vs. 3,521 +/- 749 microU.ml-1.min) incremental insulin areas were observed as a result of the physical inactivity. Incremental areas for C-peptide were also elevated significantly in the inactive state for early (12 +/- 2.0 vs. 7 +/- 1 ng.ml-1.min) and late (567 +/- 90 vs. 467 +/- 85 ng.ml-1.min) phases. Urinary excretion of C-peptide increased from 12.1 +/- 1.5 ng/240 min in the exercising state to 21.8 +/- 3.6 ng/240 min in the inactive state. Rates of whole body glucose disposal were not different between exercising and inactive states, indicating a large increase in resistance to the action of insulin. These findings indicate that the decreased insulin secretory response to a glucose stimulus in people who exercise regularly is a relatively short-term effect of exercise.

Effects of exercise and lack of exercise on insulin sensitivity and responsiveness.

Insulin action is enhanced in people who exercise regularly and vigorously. In the present study, the hyperinsulinemic, euglycemic clamp procedure was used to determine whether this enhanced insulin action is due to an increased sensitivity and/or an increased responsiveness to insulin. To avoid the variability that exists between individuals and complicates cross-sectional studies, the same subjects were studied in the trained exercising state and again after 10 days of physical inactivity. When the plasma insulin concentration was maintained at approximately 78 microU.ml-1 (a submaximal level), glucose disposal rate averaged 8.7 +/- 0.5 mg.kg-1.min-1 before and 6.7 +/- 0.6 mg.kg-1.min-1 after 10 days of activity (P less than 0.001). When the plasma insulin concentration was maintained at approximately 2,000 microU.ml-1 (a maximally effective concentration), the rate of glucose disposal was not significantly different before (15.3 +/- 0.5 mg.kg-1.min-1) compared with after (14.5 +/- 0.4 mg.kg-1.min-1) 10 days without exercise. These results provide evidence that the reversal of enhanced insulin action that occurs within a few days when exercise-trained individuals stop exercising is due to a decrease in sensitivity to insulin, not to a decrease in insulin responsiveness.

Insulin action and secretion in endurance-trained and untrained humans.

To evaluate insulin sensitivity and responsiveness, a two-stage hyperinsulinemic euglycemic clamp procedure (insulin infusions of 40 and 400 mU.m-2.min-1) was performed on 11 endurance-trained and 11 untrained volunteers. A 3-h hyperglycemic clamp procedure (plasma glucose approximately 180 mg/dl) was used to study the insulin response to a fixed glycemic stimulus in 15 trained and 12 untrained subjects. During the 40-mU.m-2.min-1 insulin infusion, the glucose disposal rate was 10.2 +/- 0.5 mg.kg fat-free mass (FFM)-1.min-1 in the trained group compared with 8.0 +/- 0.6 mg.kg FFM-1.min-1 in the untrained group (P less than 0.01). In contrast, there was no significant difference in maximally stimulated glucose disposal: 17.7 +/- 0.6 in the trained vs. 16.7 +/- 0.7 mg.kg FFM-1.min-1 in the untrained group. During the hyperglycemic clamp procedure, the incremental area for plasma insulin was lower in the trained subjects for both early (0-10 min: 140 +/- 18 vs. 223 +/- 23 microU.ml-1.min; P less than 0.005) and late (10-180 min: 4,582 +/- 689 vs. 8,895 +/- 1,316 microU.ml-1.min; P less than 0.005) insulin secretory phases. These data demonstrate that 1) the improved insulin action in healthy trained subjects is due to increased sensitivity to insulin, with no change in responsiveness to insulin, and 2) trained subjects have a smaller plasma insulin response to an identical glucose stimulus than untrained individuals.

Physiological increments in epinephrine stimulate metabolic rate in humans.

Markedly elevated plasma epinephrine is known to increase metabolic rate (MR), but such levels of epinephrine are encountered infrequently in normal free-living subjects. We studied whether epinephrine levels common in usual daily activities can affect MR and thus possibly regulate caloric expenditure. To aid definition of a MR threshold, we first measured the hourly and daily variation in MR within individuals by measuring the MR of four individuals by indirect calorimetry for 6 h on six separate occasions without any intervention. We found that hour-to-hour variation (2.0 +/- 0.9%) and the day-to-day variation (2.7 +/- 0.9%) were low, thus allowing confident detection of small increments in metabolic rate during epinephrine infusion. To define a threshold for epinephrine's effect to increase MR, we studied five normal-weight postabsorptive young men on four separate occasions. During the 1st h of each 5-h study period, saline was infused intravenously. Then, during the subsequent 4 h, subjects received intravenous infusion of saline or epinephrine at 0.1, 0.5, and 1.0 microgram/min (randomized). A significant increase in MR (3.6 +/- 1.0% SE) was measured with the lowest epinephrine infusion rate (venous plasma concentration, 94 +/- 32 pg/ml). The increases in MR correlated (r = 0.85, P less than 0.001) with increases in plasma epinephrine. The threshold concentration (upper 95% confidence limit) of epinephrine to affect MR was 90 pg/ml, a concentration frequently occurring in daily life. Thus epinephrine may play an important role in weight maintenance by affecting energy expenditure.

Leucine metabolism in type II diabetes mellitus.

Severe muscle wasting is a well-recognized characteristic of untreated insulin-deficient diabetes mellitus, a condition in which leucine turnover and oxidation are accelerated. To ascertain whether a similar circumstance exists in type II diabetes when insulin is present but with reduced efficacy, we investigated leucine turnover and oxidation in five obese type II diabetic women by tracer infusion of L-[1-13C,15N]leucine in the postabsorptive state both before and after intensive insulin therapy. With conventional treatment, the type II diabetic women received 61 +/- 33 (SD) U/day of insulin, and their fasting plasma glucose averaged 194 +/- 41 (SD) mg/dl. Leucine carbon flux (QC), nitrogen flux (QN), and oxidation (C) averaged 6.4 +/- 1.2, 15.6 +/- 4.6, and 1.4 +/- 0.3 mmol/h, respectively. These values were not different from the respective values of 6.6 +/- 1.3, 17.0 +/- 8.3, and 1.0 +/- 0.2 mmol/h in matched obese nondiabetic controls, suggesting that leucine metabolism is not altered in insulin-treated type II diabetics. After a week of intensive insulin therapy in which the same diabetic subjects received 94 +/- 36 U/day of insulin, postabsorptive plasma glucose declined to 117 +/- 26 mg/dl. Leucine QC (6.2 +/- 1.0), QN (14.8 +/- 3.7), and C (1.5 +/- 0.5 mmol/h) were unaltered by the increased insulin therapy. Thus, obese type II diabetics had normal leucine kinetics but were hyperglycemic while receiving conventional insulin therapy. Additional intensive insulin therapy in these diabetic subjects improved plasma glucose but did not alter leucine kinetics.(ABSTRACT TRUNCATED AT 250 WORDS)

Regulation of valine metabolism in man: a stable isotope study.

Valine and leucine kinetics were studied in four young healthy men in the postabsorptive state with a 4-h primed infusion of either L-[1-13C,15N] valine or L-[1-13C,15N]-leucine. For 1 wk before each infusion study each subject consumed a diet that provided an adequate amount of energy and 1.6 kg/day of protein. During infusion of tracer, plasma valine or leucine, and expired 13CO2 reached isotopic steady state by 2 h. The valine and leucine carbon fluxes (mean +/- SE) were 80.3 +/- 1.2 and 86.6 +/- 2.0 mumol kg-1h-1, respectively, consistent with the lesser content of valine compared with leucine in body protein. Valine and leucine oxidation rates were 11.8 +/- 0.6 and 15.9 +/- 1.1 mumol kg-1h-1, respectively. From these values and values for valine and leucine nitrogen flux, the rates of valine and leucine transamination were calculated. Valine and leucine deamination were 84.0 +/- 3.5 and 103.0 +/- 6.5 mumol kg-1h-1, and values for reamination were 72.2 +/- 3.3 and 87.1 +/- 7.5 mumol kg-1h-1, respectively. Thus, the patterns of valine and leucine catabolism are similar. However, when the plasma substrate levels are used to estimate transamination rate constants, we estimate that the transamination equilibrium favors leucine transamination over valine by 5-fold.

Usefulness of screening chest roentgenograms in preoperative patients.

We proposed that clinical criteria could define a group of patients very unlikely to have abnormal preoperative chest roentgenograms. Nine hundred five surgical admissions were screened for the presence of clinical factors we thought would make those patients more likely to have abnormal preoperative chest roentgenograms. Of these, 368 had no risk factors. One patient (0.3%) of the 368 had an abnormal x-ray film, which did not affect the surgery. No material abnormalities were found in the remainder of the group without risk factors. Five hundred four patients had identifiable risk factors. Of these, 114 (22%) were found to have serious abnormalities on preoperative chest roentgenogram.