Plasma leptin levels and body fat distribution

The relation between body fat distribution and plasma leptin levels in the human was investigated in 51 obese and 41 non-obese subjects. Plasma levels of leptin showed a positive correlation with body mass index and subcutaneous fat area at the umbilicus level. However, a significant correlation between its plasma levels and visceral fat area was found in neither non-obese nor obese subjects. These results suggest that plasma leptin levels might be attributed mainly to the extent of subcutaneous adiposity in human obesity.     

Circadian rhythm of plasma leptin levels in upper and lower body obese women: influence of body fat distribution and weight loss

Plasma leptin concentrations were measured every 20 min for 24 h in eight normal weight women and in eight upper body and eight lower body obese women matched for body mass index. The circadian rhythm of leptin, which could mathematically be described by a cosine, was characterized by an acrophase just after midnight in all subjects. The amplitude of a cosine fit as well as the average 24-h leptin concentration were increased by 280% and 420%, respectively, in obese compared to normal weight women. All characteristics of leptin concentration profiles were similar in upper body and lower body obese women, except for a significantly higher amplitude in the lower body obese group. Visceral and sc body fat depots were measured using magnetic resonance imaging in all three groups. Average 24-h leptin concentrations were strongly correlated with sc fat (r = 0.84), whereas visceral fat was not an independent predictor of the plasma leptin level. A loss of 50% of the overweight was associated with a 55% decrease in the average 24-h leptin concentrations in obese women (95% confidence interval, 12.3, 26.6), whereas the characteristics of the circadian rhythm of leptin remained unchanged. Finally, it was observed that a fasting plasma leptin concentration is not an acceptable indicator of the average leptin concentration over 24 h.     

Gender differences in body fat content are present well before puberty

To determine whether gender differences in body fat could be detected in prepubertal children using dual energy X-ray absorptiometry (DEXA), body composition was measured in 20 healthy boys aged 3-8 y matched for age, height and weight with 20 healthy girls. Although boys and girls did not differ in age, height, weight, body mass index (BMI) or bone mineral content, the boys had a lower percentage of body fat (13.5 +/- 5.1 vs 20.4 +/- 6.1%, P < 0.01), a lower fat mass (3.2 +/- 2.0 vs 4.9 +/- 3.1 kg, P < 0.01), and a higher bone-free lean tissue mass (18.6 +/- 4.3 vs 17.0 +/- 3.5 kg, P < 0.01) than the girls. Girls had approximately 50% more body fat than the boys. This is the first DEXA study to show that boys aged 3-8 y have less body fat than girls of similar age, height and weight. Thus, this technology demonstrates that significant gender differences in body composition are evident, well before the onset of puberty.     

Correlation between insulin receptor binding in isolated fat cells and insulin sensitivity in obese human subjects

This study examined the relationship between receptor binding of insulin in a metabolically significant target tissue in vitro and sensitivity to insulin in vivo in obese human subjects. Specific insulin binding was measured at 24 degrees C in isolated enlarged fat cells obtained from 16 patients, by observing the effect of increasing concentrations of unlabeled insulin on the binding of [125I]insulin. Scratchard plots of the binding data were curvilinear with an upward concavity, similarity shaped, and essentially parallel. Kinetic studies on the dissociation of [125I]insulin from fat cells indicated that these curvilinear Scratchard plots could be explained by the presence of site:site interactions of the negative cooperative type. Differences in binding between individual patients were predominantly due to differences in the numbers of receptor sites whether expressed in relation to cell number, cell volume, or cell surface area. These findings were not accounted for by differences in [125I]insulin degradation. Acute exposure of adipose tissue to insulin in vitro had no significant effect on [125I]insulin binding to isolated cells. The number of receptor sites was directly correlated with insulin sensitivity in vivo, measured as the rate constant (Kitt) for the fall in blood glucose after intravenous insulin, and was inversely correlated with the level of fasting plasma insulin. These findings corroborate those from other studies using human mononuclear leukocytes and various tissues from the obese mouse, which indicate that decreased insulin binding is a characteristic feature of insulin resistance in obesity.     

Sex hormones, body fat distribution, resting metabolic rate and glucose-induced thermogenesis in premenopausal obese women

The topographic specificity of upper body obesity is known to be at the origin of a series of metabolic complications. In contrast to this negative effect, women with abdominal obesity usually can lose more body weight than women with gluteal-femoral obesity. In order to find some contributive explanations for this effect, we studied resting metabolic rate (RMR) and glucose-induced thermogenesis (GIT) in both types of obesity. Since upper body obesity is characterized by androgen excess, a relationship between body fat distribution, sex hormones, RMR and indices of thermogenesis was studied. Of 39 obese women who were recruited (mean age: 32.4 +/- 9.3 years), 30 were compared for analysis. Upper body obesity (waist-to-hip ratio (WHR): 0.84 +/- 0.02; body mass index (BMI) 36.2: 36.2 +/- 6.0) is not characterized by differences in RMR, whereas glucose-induced thermogenesis is significantly higher in this subgroup (P < 0.008), expressed as percentage increase above RMR (18.3 +/- 8.5 vs. 11.9 +/- 3.6%) or as percentage of metabolisable energy intake (8.2 +/- 3.3 vs. 5.8 +/- 2.3%). Correlation coefficient data show that GIT determinants are closely related to WHR (r = 0.43; P < 0.01) and not to BMI. Resting metabolic rate, both in absolute terms and corrected for fat-free mass (FFM), is not related to indices of androgenicity, but is negatively related to serum oestradiol levels; this negative relationship with oestradiol disappears when RMR is corrected for both fat mass (FM) and FFM. GIT parameters are not related to free testosterone or oestradiol, regardless of the phase of the menstrual cycle.(ABSTRACT TRUNCATED AT 250 WORDS)     

Inverse correlation between serum testosterone and leptin in men

Besides its role in the regulation of energy balance, leptin seems to be involved in linking energy stores to the reproductive system. A gender-dependent difference exists in plasma leptin concentration and leptin messenger ribonucleic acid expression in rodents and humans. This difference does not seem to be explained simply by differences in the amount of body fat between genders. To elucidate the relationship of endogenous testosterone and leptin, we studied the serum leptin concentrations in 269 elderly nondiabetic men. In addition, to assess whether exogenously administered testosterone could influence leptin production, we followed the serum levels of leptin in 10 healthy men during a 12-month treatment with 200 mg testosterone enanthate, i.m., weekly for contraceptive purposes. We found that the serum leptin concentration correlated inversely (r = -0.39; P < 0.001) with that of testosterone in elderly men. This inverse correlation was still present when body mass index and plasma insulin were included in the analysis. The administration of testosterone to young men suppressed serum leptin from the pretreatment level of 3.4 +/- 1.4 to 1.9 +/- 0.6 micrograms/L during the therapy. After cessation of testosterone injections, serum leptin concentration returned back to the pretreatment level. It is concluded that testosterone has a suppressive effect on leptin production, as reflected by circulating levels of this hormone.     

Serum leptin concentration and insulin sensitivity in men with abdominal obesity

Papillary immature metaplasia (PIM) of the cervix (immature condyloma) is a variant of low-grade squamous intraepithelial lesions (LSIL). It is frequently associated with human papillomavirus (HPV) types 6 and 11. The purpose of this study was to characterize the cytologic changes associated with this lesion. We analyzed 10 cases of PIM from our files and reviewed the Papanicolaou smears taken proximate to the time of the biopsy. Four cases had either reactive epithelial changes (2 cases) or cytologic findings typical of low-grade SIL, with koilocytosis (2 cases). Six cases displayed a spectrum of metaplastic cells with varying maturation that ranged from atypical reactive cells to atypical immature metaplastic cells. Binucleation was common. Some cells exhibited features characteristic of SIL, although the degree of nuclear atypia generally was less than that associated with high-grade SIL. Papanicolaou smears from all cases were interpreted as atypical (ASCUS) metaplasia or low-grade SIL. Follow-up biopsy in one case revealed a PIM in association with a high-grade SIL, the latter undiagnosed by smear alone. PIM is a distinct histologic entity that can present with a spectrum of cytologic findings. Its recognition histologically can resolve some cytologic/histologic discrepancies. Confusion with an immature HSIL or atypical immature metaplasia can occur in some instances and the diagnosis of PIM by cytology alone is not recommended, unless the diagnosis is qualified.     

Exogenous androgens influence body composition and regional body fat distribution in obese postmenopausal women--a clinical research center study

Abdominal fat distribution is influenced by androgen levels in both men and women. The purpose of this study was to assess the effects on fat distribution of administering nandrolone decanoate (ND; an anabolic steroid with weak androgenic activity) or spironolactone (SP; an antiandrogen) in obese postmenopausal women. The design was a randomized, placebo-controlled, 9-month trial with simultaneous calorie restriction for weight loss. Women in all three groups lost comparable amounts of weight, but the ND-treated women gained lean mass relative to the other two groups (P < 0.0005) and lost more body fat than women in the SP group (P < 0.01). The resting metabolic rate also increased slightly in the ND group. ND treatment produced a gain in visceral fat, as determined by computed tomography scan, and a relatively greater loss of sc abdominal fat. SP-treated women lost significantly less sc fat than the other two groups. Serum cholesterol decreased in the placebo group, but increased slightly in the other two groups (significant for SP vs. placebo, P < 0.05). High density lipoprotein cholesterol decreased significantly in the ND-treated women. There were no significant changes in fasting glucose or insulin sensitivity. We conclude that administration of exogenous androgens modulates body composition in obese postmenopausal women and independently affects visceral and sc abdominal fat.     

Relationships between dietary fat, body fat, and serum lipid profile in prepubertal children

OBJECTIVE: The purpose of the study was to test the hypothesis that dietary fat components were associated with the serum lipid profile independent of ethnicity, body fat, and fat distribution in prepubertal children. RESEARCH METHODS AND PROCEDURES: Sixty-six children (45 African American and 21 Caucasian), aged from 4 to 10 years, were recruited into the study. Dietary total fat, saturated fat, monounsaturated fat, and polyunsaturated fat were estimated by averaging two 24-hour diet recalls. Fasting serum triacylglycerol, total cholesterol, and high-density lipoprotein cholesterol were analyzed, and low-density lipoprotein cholesterol (LDL-C) was calculated by the method of Friedewald. Body composition and fat distribution were measured by dual energy X-ray absorptiometry and computed tomography. RESULTS: Children in both ethnic groups tended to overreport their dietary intake relative to total energy expenditure by 18%. African American children consumed more energy from total fat (35.3% vs. 31.5%, p<0.05), saturated fat (13.7% vs 12.2%, p<0.05), protein (16.4% vs. 13.2%, p=0.02), and less from carbohydrate (48% vs. 57.1%, p<0.01) than Caucasian children. There was no significant correlation between dietary fat and either serum lipids or body fat indices after adjusting for nonfat energy intake and total lean tissue mass. Total body fat (r=0.32), subcutaneous abdominal adipose tissue (r=0.39), and intra-abdominal adipose tissue (r=0.42) were positively related to serum triacylglycerol; these associations remained significant in a multiple linear regression model in which body fat indices were adjusted for ethnicity, total lean tissue, dietary total fat, and nonfat intake. DISCUSSION: Our results do not support a link between dietary fat and serum lipids; instead, our data suggest that body fat may play a more important role than dietary fat in the course of cardiovascular disease development in prepubertal children.     

Cardiac autonomic nerve function and insulin sensitivity in obese subjects

OBJECTIVE: To investigate whether obesity influences cardiac autonomic nerve function. DESIGN: Comparing two groups of subjects with different degrees of obesity to normal weight controls. SUBJECTS: 19 healthy controls (mean age 33 y, BMI 21.7 +/- 0.2 kg/m2) and 17 obese non-diabetic subjects (mean age 39 y, BMI 33.7 +/- 1.8 kg/m2). MEASUREMENTS: Insulin sensitivity was calculated by an oral glucose tolerance test. Autonomic nerve function was evaluated by analysing the variation of the heart frequency at rest (coefficient variation of R-R intervals, REST 1), during deep respiration, at a Valsalva maneuver (longest/shortest R-R interval during inspiration hold) and by the Ewing test (ratio between the 30th and 15th R-R interval after reaching up-right position). RESULTS: The obese showed a lower insulin sensitivity than healthy controls (3.09 vs 4.60 mg x l2/mmol x mU x min, P < 0.001). Their variation in heart frequency was reduced (REST 1: 1.95 vs 2.9, P < 0.01, Valsalva: 1.30 vs 1.52 and Ewing test: 1.03 vs 1.14, P < 0.05). However, patients with moderate (BMI 31.7 kg/m2) or severe obesity (39.0 kg/m2) with identical insulin sensitivity had no significant difference in autonomic nerve function. Except for the Ewing test all measured parameters for the evaluation of cardiac autonomic nerve function correlated with the degree of diminished insulin sensitivity (REST 1: r = 0.475, P < 0.001). CONCLUSION: Moderate obesity with significantly decreased insulin sensitivity is associated with impaired cardiac autonomic nerve function.     

Insulin sensitivity, insulin action, and fibrinolysis activity in nondiabetic and diabetic obese subjects

Because inconsistencies occur with regard to the relative contribution of insulin to the hypofibrinolysis characteristic of obesity and diabetes, we explored the relationship between insulin and fibrinolysis, assessing both insulin sensitivity and insulin action. Seventeen markedly obese subjects (body mass index [BMI], 34.0+/-1.6 kg/m2; 12 nondiabetic and five diabetic) were studied using the three-step euglycemic-hyperinsulinemic clamp technique. Since the circadian rhythm of the fibrinolytic system may obscure a true effect of insulin, variations in fibrinolysis parameters observed during the glucose clamp were compared with those occurring spontaneously because of the circadian rhythm. Compared with six normal-weight subjects (BMI, 21.0+/-0.9 kg/m2), all obese subjects exhibited basal hyperinsulinism (fasting plasma insulin, 16.0+/-1.4 v 9.8+/-1.3 microU/microL, P < .001; fasting plasma C-peptide, 1.4+/-0.2 v 0.5+/-0.2 ng/mL, P < .001), hypofibrinolysis (euglobulin lysis time [ELT], 378+/-29 v 222+/-31 minutes, P=.01; tissue plasminogen activator [tPA] antigen, 7.8+/-0.9 v 4.2+/-0.5 ng/mL, P=.04; plasminogen activator inhibitor type 1 [PAI-1] activity, 22.2+/-2.5 v3.9+/-0.6 AU/mL, P=.004), and marked insulin resistance (M value, ie, the maximal glucose disposal rate, 9.1+/-0.6 v 18.6+/-0.8 mg/(kg x min), P < .001). The M value correlated inversely with tPA antigen (r=-.46, P=.05). During insulin infusion, values for fibrinolysis parameters decreased, but were not different compared with variations due to the circadian rhythm. In conclusion, our findings together with previously reported data reinforce the idea that chronic hyperinsulinism is linked to hypofibrinolysis, but insulin does not seem to acutely regulate the fibrinolysis system.     

Glucose effectiveness, peripheral and hepatic insulin sensitivity, in obese and lean prepubertal children

Vascular injuries in lumbar disk surgery, although rare, are serious complications which may be overlooked due to a broad range of clinical manifestations. It is important that surgeons and radiologists be aware of these potentially fatal complications and develop an appropriate symptom-based diagnostic paradigm. We reviewed 8099 consecutive cases of lumbar disk surgery, performed over a 14-year period at a single institution, for postoperative vascular complications. We identified four patients (0.05%) with lumbar disk surgery-related vascular complications: intraoperative lacerations of the abdominal aorta and median sacral artery, an arteriovenous fistula between the left common iliac artery and vein detected 19 days postdiskectomy, and a partially thrombosed aortic aneurysm with an arteriovenous fistula between the aneurysm and the inferior vena cava, diagnosed 11 months after surgery. The majority of cases in the literature of vascular injury in lumbar disk surgery were reported prior to 1965. Diagnostic approaches described in that period do not reflect the great range of diagnostic techniques available today. Angiography remains the gold standard for diagnosis and guidance as to surgical repair. However, a high index of suspicion based on clinical signs and/or the use of sonography or CT is important in the detection of these complications.     

Vitamin A distribution among fat globule core, fat globule membrane, and serum fraction in milk

In cream suspensions and partly skimmed milks obtained by successive centrifugation from whole milk, the concentration of vitamin A per gram of fat was inversely related to the mean size of fat globules. This result indicated that not all the vitamin A of milk was located in the fat globule core, which was further supported by studies of buttermilk and butter oil fractions obtained by churning of cream. These studies indicated that a portion of the vitamin A was associated with the fat globule membrane. The vitamin A content per gram of fat of purified fat globules was similar to vitamin A content per gram of fat of the whole milk from which they were isolated. This result indicated that the amount of vitamin A in the serum fraction was negligible and that vitamin A was localized in the fat globule and fat globule membrane.     

The effects of a high fat diet on leptin mRNA, serum leptin and the response to leptin are not altered in a rat strain susceptible to high fat diet-induced obesity

Osborne-Mendel (OM) and S5B/Pl rats differ in their sensitivity to develop obesity when fed a high fat (HF) diet; OM rats become obese, whereas S5B/Pl rats remain thin. We have investigated the possibilities that either an impaired leptin response or resistance to leptin action underlies the sensitivity to this form of obesity in OM rats. In Experiment 1, OM and S5B/Pl rats fed a nonpurified diet were killed at d 0 or were fed either a HF (56% fat energy) or a low fat (LF, 10% fat energy) diet for 2 or 7 d. The HF diet increased serum leptin significantly by d 2 to levels that were similar in both rat strains. At 7 d, leptin levels were lower than at d 2 but remained higher than levels in the d 0 control groups. The leptin mRNA:18S RNA ratio in epididymal adipose tissue increased to higher levels in HF-fed OM rats than in S5B/Pl rats fed that diet. However, although the LF diet had no effect in S5B/Pl rats, it increased leptin mRNA levels in epididymal adipose tissue of OM rats compared with the controls fed the nonpurified diet. In Experiment 2, OM and S5B/Pl rats were fed HF or LF diets for 5 wk. At that time, their feeding response to a range of leptin doses (0, 1, 5 or 10 microgram) given intracerebroventricularly was tested after overnight food deprivation. There was a similar dose-dependent reduction in energy intake in response to leptin in both OM and S5B/Pl rats. These responses were independent of the diet. The data suggest that the susceptibility of OM rats to HF diet-induced obesity is not related to either a loss of central sensitivity to leptin or a failure to enhance leptin production acutely, although the failure to maintain chronically increased levels of serum leptin could contribute to the obesity.     

Longitudinal changes in maternal serum leptin concentrations, body composition, and resting metabolic rate in pregnancy

OBJECTIVE: We sought to evaluate the longitudinal changes in maternal serum leptin concentrations, body composition, and resting metabolic rate during pregnancy. STUDY DESIGN: Ten women were evaluated before pregnancy, in early pregnancy (12 to 14 weeks), and in late pregnancy (34 to 36 weeks). Leptin concentrations were measured by radioimmunoassay, body composition with hydrodensitometry with adjustment for total body water, and resting metabolic rate by use of indirect calorimetry. RESULTS: Using analysis of variance with repeated measures from pregravid to late pregnancy, a 66% increase (mean +/- SD) was found in leptin concentrations (in nanograms per milliliter) (before pregnancy, 25.4 +/- 19.9; in early pregnancy, 37.5 +/- 26.2; and in late pregnancy, 38.4 +/- 27.3, p = 0.003); a 9% increase in body fat (in kilograms) (before pregnancy, 29.4 +/- 15.7; in early pregnancy, 28.7 +/- 14.0; in late pregnancy, 31.4 +/- 14.6; p = 0.04); a 28% increase in oxygen consumption (in milliliters of oxygen per minute) (before pregnancy, 221.2 +/- 29.5; in early pregnancy, 230.4 +/- 42.9; in late pregnancy, 285.3 +/- 51.9; p < 0.0001); and a 9% increase in oxygen consumption (milliliters of oxygen per kilogram per minute) (before pregnancy, 3.02 +/- 0.43; in early pregnancy, 3.05 +/- 0.30; in late pregnancy, 3.31 +/- 0.37, p = 0.002) with advancing gestation. A significant positive correlation was present between leptin and body fat before pregnancy (r = 0.90, p < 0.0001), in early pregnancy (r = 0.91, p < 0.0001), and in late pregnancy (r = 0.87, p = 0.0005) and between leptin and oxygen consumption before pregnancy (r = 0.80, p = 0.004), in early pregnancy (r = 0.92, p < 0.0001), and in late pregnancy (r = 0.62, p = 0.06). When oxygen consumption was adjusted for maternal and fetal tissue mass, a significant negative correlation was found between leptin and oxygen consumption before pregnancy (r = -0.96, p < 0.0001), in early pregnancy (r = -0.80, p = 0.0034), and in late pregnancy (r = -0.70, p = 0.02). CONCLUSION: We conclude that leptin increases significantly during early pregnancy before any major changes in body fat and resting metabolic rate. These data suggest that pregnancy represents a leptin-resistant state.     

Dehydroepiandrosterone decreases serum tumor necrosis factor-alpha and restores insulin sensitivity: independent effect from secondary weight reduction in genetically obese Zucker fatty rats

Dehydroepiandrosterone (DHEA) and its sulfate ester are the most abundant circulating adrenal steroids in humans. Administration of DHEA has been reported to have beneficial effects on obesity, hyperlipidemia, diabetes, and atherosclerosis in obese rodents, although its effects on insulin resistance have not been fully elucidated. In this study, the effects of DHEA treatment on insulin sensitivity were investigated in genetically obese Zucker rats, an animal model of insulin resistance, using the euglycemic clamp technique. After 0.4% DHEA was administered for 10 days to female obese Zucker rats aged 16 weeks, body weight and plasma insulin decreased and glucose disposal rate (GDR), which was normally reduced in obese rats, rose significantly compared with age- and sex-matched control obese rats. On the other hand, although the pair-fed obese rats also showed levels of weight reduction similar to those of DHEA-treated rats, the increase in GDR of DHEA-treated rats was significantly greater than in pair-fed rats, suggesting a direct ameliorating effect of DHEA on insulin sensitivity of obese rats. Serum concentration of tumor necrosis factor (TNF)-alpha, one of cytokines causing insulin resistance, was also reduced significantly in DHEA-treated, but not in pair-fed obese rats. In conclusion, our results suggest that DHEA treatment reduces body weight and serum TNF-alpha independently, and that both may ameliorate insulin resistance in obese Zucker fatty rats.     

Zinc supplementation aggravates body fat accumulation in genetically obese mice and dietary-obese mice

Human colostrum, the first product of lactation, has antioxidant properties and inhibits selected enzyme and bactericidal activities of human neutrophils. We examined the subsequent product of lactation, mature human milk, with respect to its antioxidant activities, its effects on neutrophil enzyme activities (myeloperoxidase, beta-glucuronidase, and lysozyme), and its effects on neutrophil bactericidal and phagocytic activities. Mature human milk displayed antioxidant characteristics similar to those of human colostrum, reducing cytochrome c and consuming H2O2. Mature milk also displayed colostrum-like characteristics in depressing neutrophil myeloperoxidase and beta-glucuronidase activities, but not in altering lysozyme activity. Neutrophil bactericidal activity against Staphylococcus aureus was depressed by both mature milk and colostrum, without dramatic effects on phagocytic activity. These data show that mature milk shares characteristics with human colostrum that may result in anti-inflammatory effects, but the magnitude of these effects is generally smaller.     

Evidence that plasma leptin and insulin levels are associated with body adiposity via different mechanisms

OBJECTIVE: Like insulin, the adipocyte hormone, leptin, circulates at levels proportionate to body adiposity. Because insulin may regulate leptin secretion, we sought to determine if plasma leptin levels are coupled to body adiposity via changes in circulating insulin levels or insulin sensitivity and whether leptin secretion from adipocytes is impaired in subjects with NIDDM. RESEARCH DESIGN AND METHODS: We used multiple linear regression to analyze relationships between BMI (a measure of body adiposity) and fasting plasma levels of leptin and insulin in 98 nondiabetic human subjects (68 men/30 women) and 38 subjects with NIDDM (27 men/11 women). The insulin sensitivity index (Si) was also determined in a subset of nondiabetic subjects (n = 38). RESULTS: Fasting plasma leptin concentrations were correlated to both BMI (r = 0.66, P = 0.0001) and fasting plasma insulin levels (r = 0.65, P = 0.0001) in nondiabetic men and women (r = 0.58, P = 0.0009 for BMI; r = 0.47, P = 0.01 for insulin). While the plasma leptin level was also inversely related to Si (r = -0.35; P = 0.03), this association was dependent on BMI, whereas the association between insulin and Si was not. Conversely, the relationship between plasma leptin and BMI was independent of Si, whereas that between insulin and BMI was dependent on Si. The relationship between plasma leptin levels and BMI did not differ significantly among NIDDM subjects from that observed in nondiabetic subjects. CONCLUSIONS: We conclude that 1) body adiposity, sex, and the fasting insulin level are independently associated with plasma leptin level; 2) because NIDDM does not influence leptin levels, obesity associated with NIDDM is unlikely to result from impaired leptin secretion; and 3) insulin sensitivity contributes to the association between body adiposity and plasma levels of insulin, but not leptin. The mechanisms underlying the association between body adiposity and circulating levels of these two hormones, therefore, appear to be different.