Hyperleptinemia is an essential feature of human obesity. Total body fat mass > % body fat > BMI are the best predictors of circulating leptin levels. Although ob gene is differentially expressed in different fat compartments, apart from total body fat, upper or lower body adiposity or visceral fat does not influence basal leptin levels. Similarly, age, basal glucose levels, and ethnicity do not influence circulating leptin levels. Only in insulin-sensitive individuals do basal levels of insulin and leptin correlate positively even after factoring in body fat. Diabetes does not influence leptin secretion in both lean and obese subjects per se. Independent of adiposity, leptin levels are higher in women than in men. This sexual dimorphism is also present in adolescent children. In eating disorders anorexia nervosa and bulimea nervosa, leptin levels are not upregulated but simply reflect BMI and probably body fat. In spite of strong correlation between body fat and leptin levels, there is great heterogeneity in leptin levels at any given index of body fat. About 5% of obese populations can be regarded as "relatively" leptin deficient which could benefit from leptin therapy. Leptin has dual regulation in human physiology. During the periods of weight maintenance, when energy intake and energy output are equal, leptin levels reflect total bodyfat mass. However, in conditions of negative (weight-loss programs) and positive (weight-gain programs) energy balances, the changes in leptin levels function as a sensor of energy imbalance. This latter phenomenon is best illustrated by short-term fasting and overfeeding experiments. Within 24 h of fasting leptin levels decline to approximately 30% of initial basal values. Massive overfeeding over a 12-h period increases leptin levels by approximately 50% of initial basal values. Meal ingestion does not acutely regulate serum leptin levels. A few studies have shown a modest increase in leptin secretion at supraphysiological insulin concentrations 4-6 h following insulin infusion. Under in vitro conditions, insulin stimulates leptin production only after four days in primary cultures of human adipocytes, which is apparently due to its trophic effects and an increased fat-cell size. Similar to other hormones, leptin secretion shows circadian rhythm and oscillatory pattern. The nocturnal rise of leptin secretion is entrained to mealtime probably due to cumulative hyperinsulinemia of the entire day. Like other growth factors and cytokines, leptin binding proteins including soluble leptin receptor are present in human serum. In lean subjects, the majority of leptin circulates in the bound form whereas in obese subjects, the majority of leptin is present in the free form. When free-leptin levels are compared between lean and obese subjects, even more pronounced hyperleptinemia in obesity is observed than that reported by measuring total leptin levels. During short-term fasting, free-leptin levels in lean subjects decrease in much greater proportion than those in obese subjects. In lean subjects with a relatively small energy store and particularly during food deprivation, leptin circulating predominantly in the bound form could be the mechanism to restrict its availability to hypothalamic leptin receptors for inhibiting leptin's effect on food intake and/or energy metabolism. Unlike marked changes in serum leptin, CSF leptin is only modestly increased in obese subjects and the CSF leptin/serum leptin ratio decreases logarithmically with increasing BMI. If CSF leptin levels are any indication of brain interstitial fluid levels, then hypothalami of obese subjects are not exposed to abnormally elevated leptin concentrations. In the presence of normal leptin receptor (functional long form, i.e., OB-Rb) mRNA expression and in the absence of leptin receptor gene mutations, it is logical to assume defective leptin signaling and/or impaired affector system(s) are the likely causes of leptin resistance in