Insulin sensitivity, lipids, and body composition in childhood: is "syndrome X" present?

Syndrome X, or the syndrome of insulin resistance, is a cluster of related metabolic abnormalities of hyperinsulinemia, glucose intolerance, increased very low density lipoprotein (VLDL), decreased high density lipoprotein (HDL), and hypertension in nonobese adults and plays an important role in the genesis of cardiovascular disease. The aim of the present study was to examine the relationships among insulin sensitivity, plasma lipid levels, and body composition in the pediatric age group to determine whether these associations are present in childhood. Twenty healthy Caucasian Tanner stage I (TI) children (age, 10.7 +/- 0.3 yr; body mass index, 18.9 +/- 0.8 kg/m2) and 22 pubertal Tanner stage II-IV (TII-IV) adolescents (age, 14.0 +/- 0.3 yr; body mass index, 20.0 +/- 0.4 kg/m2) were studied. In vivo insulin-mediated glucose disposal (Rd) was evaluated during a 40 mu/m2. min hyperinsulinemic-euglycemic clamp. Body composition was assessed isotopically by the H218O dilution principle. Fasting blood was obtained for cholesterol, triglyceride (TG), VLDL, low density lipoprotein (LDL), and HDL determinations. In both groups, the strongest correlation of Rd was with percent body fat (%BF) (TI: r = -0.82; P < 0.001; TII-IV: r = -0.73; P < 0.001). In addition, in TI, Rd was correlated with TG (r = 0.64; P = 0.001), VLDL (r = 0.64; P = 0.001), and diastolic blood pressure (r = -0.50; P = 0.01). There were no such correlations in TII-IV. In TI, % BF correlated positively with LDL and negatively with TG and VLDL. In TII-IV, % BF correlated positively with cholesterol and LDL. After correcting for %BF, partial correlation analysis revealed no relationship between Rd and lipid levels in either group. This suggests that the relationship of insulin sensitivity to lipid levels was secondary to the effect of body composition on lipid levels. However, regardless of body composition, the basal insulin level was correlated with TG (r = 0.38; P = 0.04) and VLDL (r = 0.40; P = 0.04) in TII-IV subjects. We conclude that 1) the primary correlate of insulin sensitivity is %BF in both prepubertal and pubertal subjects, with no relationship to plasma lipids; 2) in prepubertal children, diastolic blood pressure is negatively correlated with insulin sensitivity and positively with insulin levels, independent of adiposity; and 3) after the onset of puberty, basal insulin levels are positively correlated with VLDL and TG regardless of the degree of adiposity. This observation could be a very early manifestation of the genesis of syndrome X in childhood.