Relationship between different subcutaneous adipose tissue layers, fat mass, and leptin in response to short-term energy restriction in obese girls

This study addresses whether the expected relationship of 15 specified subc utaneous adipose tissue layers (SAT layers) from 1-neck to 15-calf and body fat mass (FM) with leptin was influenced by a weight-loss program, In 30 o bese girls (10 prepubertal, 15 pubertal, and 5 late/postpubuertal) SAT laye rs were measured by means of the optical device Lipometer. Fat mass (FM) wa s estimated indirectly by means of bioelectrical impedance. Leptin and insu lin were determined by means of radioimmunoassays. All measurements were pe rformed before (pre) and after (post) 3 weeks of low-caloric diet and physi cal training, At the beginning of the study, there were significant correla tions for all estimates of adiposity and leptin (0.67 to 0.79; P < 0.0001), Five SAT layers from the upper;body and the trunk (0.48 to 0.67; P < 0.01) but none from the abdominal region and lower extremities were correlated w ith leptin. FM. together with SAT layers 4-upper back and 8-lower abdomen ( negative slope) explained 79% of the variation in pre leptin values (P < 0. 0001). The weight-loss program significantly reduced leptin (P < 0.0001), i nsulin (P = 0.04), estimates of adiposity (P < 0.0001), and SAT layers 4-up per back (P = 0.0006), 11-front thigh, 13-rear thigh, and 14-inner thigh (P between <0.03 and <0.01). Although significant, the reductions in the four SAT layers were small. Estimated fat-free mass was significantly increased after three weeks (P < 0.05). Changes in SAT layers from the upper extremi ties and from the trunk were inversely correlated to the decrease in leptin (P between <0.05 and <0.001), Initial leptin was the best correlate of the decrease in leptin (adj. R-2 = 0.815; P < 0.0001). However, when only chan ges in adiposity and insulin were considered in the regression model, chang es in insulin contributed to the fall in leptin (adj. R-2 = 0.23; P = 0.004 ). When changes in SAT layers were added to the model, changes in SAT layer s a-triceps and 10-hip (negative slopes) contributed to the decrease in lep tin (adj. R-2 =: 0.48; P < 0.0001). After weight loss, correlations between estimates of post adiposity and post leptin (0.40, P = 0.01 to 0.57, P = 0 .0005) were lower compared with pre values. SAT layers 4-upper back and 3-b iceps contributed independently to post leptin values (adj. R-2 = 0.50; P < 0.0001). It is suggested that fat mass and SAT layers from the upper body are the main determinants of leptin in obese girls before weight loss. The diet and sports intervention program reduced leptin independent of the redu ction in adiposity. The distribution of subcutaneous fat might be a stable correlate of circulating leptin after a short-term reduction in energy inta ke. (C) 2000 Wiley-Liss, Inc.