Adiponectin influences insulin sensitivity and lipid metabolism, but it is not clear whether these effects are correlated with fat mass or distribution. We studied the relationship between plasma adiponectin and leptin levels, insulin sensitivity, and serum lipids by a cross-sectional study (n = 242 subjects) and by an intervention study (95 of 242) to evaluate the effect of weight loss (WL). Considering all subjects both together and subdivided into nonobese (n = 107) and obese (n = 135) groups, plasma adiponectin, but not plasma leptin, was significantly (P < 0.01) correlated with insulin sensitivity [homeostasis model assessment of insulin-resistance index (HOMAIR), insulin sensitivity index (ISI) at oral glucose tolerance test, and clamp in 115 of 242 individuals], high-density lipoprotein cholesterol, and triglycerides. These relationships were still significant (P < 0.01) after adjusting for age, gender, body mass index (BMI), and ISI. After WL (-16.8 +/- 0.8%), plasma adiponectin increased, and plasma leptin decreased (P < 0.0001 for both). Their changes (Delta) were significantly correlated with Delta-BMI (P < 0.05 for both). Delta-Adiponectin, but not Delta-leptin, significantly (P < 0.001) correlated with Delta-high-density lipoprotein cholesterol and Delta-triglycerides; these correlations were independent of age, gender, Delta-BMI, and Delta-ISI (P < 0.005). In conclusion, both cross-sectional and intervention studies indicate that plasma adiponectin level correlates with serum lipids independently of fat mass. The intervention study also suggests that adiponectin increase after WL is correlated with serum lipid improvement independently of insulin sensitivity changes.