Ghrelin does not regulate the GH response to insulin-induced hypoglycaemia in children but could be involved in the regulation of cortisol secretion.

OBJECTIVE: Ghrelin activates the growth hormone secretagogue receptor GHS-R. It strongly stimulates GH secretion and has a role in energy homeostasis. The relationship between plasma ghrelin and cortisol levels during insulin-induced hypoglycaemia in prepubertal and pubertal children has not yet been investigated. The aim of the present study was to establish whether insulin-induced hypoglycaemia stimulates ghrelin secretion and whether changes in ghrelin concentrations are related to changes in GH and cortisol in children. DESIGN AND PATIENTS: We studied a group of 20 children and adolescents (five girls, 15 boys, mean age 10.8 +/- 3.7 years) undergoing insulin tolerance tests (ITTs) for clinical investigation of GH deficiency. MEASUREMENTS: Stimulation tests were performed to investigate the relationship between ghrelin, GH, cortisol and glucose levels according to age and pubertal stage by determining the ghrelin profiles during insulin-induced hypoglycaemia (at 0, 60 and 120 min). RESULTS: Ghrelin was significantly and inversely related to body weight, height, body mass index (BMI) and age of children (P < 0.05). Significant changes in ghrelin levels (P = 0.00013) were found after the insulin bolus, with a decline at 60 min and an increase to baseline values at 120 min. Changes in cortisol levels were negatively correlated with changes in ghrelin at 60 min (r = -0.59, P = 0.004) and at 120 min (r = -0.605, P = 0.003). CONCLUSIONS: This study shows that ghrelin might not regulate the GH response to insulin-induced hypoglycaemia in prepubertal and pubertal children. A role for ghrelin in the regulation of cortisol secretion can be hypothesized concerning the negative correlation between changes in ghrelin and cortisol. Furthermore, the results imply that ghrelin secretion is age dependent and is a function of growth.

Subcutaneous adipose tissue layers as a stable correlate of leptin in response to short term energy restriction in obese girls.

AIMS: We studied the relationship of subcutaneous adipose tissue layers (SAT-layers) measured at 15 specified body sites with leptin before and after a weight loss program for three weeks. SUBJECTS AND METHODS: In 70 obese girls, SAT-layers were measured by means of the optical device, lipometer. Fat mass (FM) was estimated by means of bioelectrical impedance. RESULTS: At the beginning of the study, all estimates of adiposity, insulin, and SAT-layers from the upper body (from 1-neck to 6-lateral chest) were correlated to leptin at a P-value of<0.0001. Percentage FM together with SAT-layer 4-upper back and insulin explained 75% of the variation in leptin (P<0.0001). After three weeks, estimates of adiposity and leptin were reduced (all P<0.0001). Most SAT-layers were reduced, but SAT-layers 8-lower abdomen and 9-lower back were significantly increased. Changes in leptin were best explained by initial leptin, but percentage change (Delta) in insulin, Delta SAT-layer 1-neck, and Delta SAT-layer 3-biceps contributed to the Delta leptin (adj. r(2)=0.47, P<0.0001). In the weight-reduced state, circulating leptin was best explained by three SAT-layers and insulin (adj. r(2)=0.67, P<0.0001). DISCUSSION: The results suggest that Delta changes in leptin are attributable to changes in the endocrine state and subcutaneous fat, and SAT-layers may serve as a stable correlate of leptin in the weight-reduced state.

The influence of weight loss on fibrinolytic and metabolic parameters in obese children and adolescents.

We studied i) whether short-term weight loss alters plasminogen activator inhibitor-1 antigen (PAI-1-Ag) and tissue-type plasminogen activator antigen (tPA-Ag) in obese children, and ii) whether changes in body composition and/or abdominal adiposity are responsible for changes in PAI-1 and tPA-Ag. 20 obese boys (mean age 11.9 yr) and 40 obese girls (mean age 12 yr) were studied before and after three weeks of low-caloric diet and physical activity. Body composition was assessed by means of bioelectrical impedance, and the waist-to-hip ratio (WHR) was measured. Blood samples were determined for insulin, glucose, triglycerides, PAI-1-Ag, tPA-Ag, and the fasting insulin resistance index (FIRI) was calculated. Boys had a greater WHR, higher levels of glucose, and a slightly greater FIRI than girls. Estimates of adiposity, insulin, and triglycerides were correlated with PAI-1 and tPA-Ag. WHR was significantly correlated with fibrinolytic parameters only in girls. Insulin and tPA-Ag contributed to PAI-1 (adj. R2 = 0.36, p <0.0001), whereas percentage fat mass and triglycerides contributed to tPA-Ag (adj. R2 = 0.469, p <0.0001). The weight loss program significantly reduced adiposity, abdominal adiposity, and lowered fibrinolytic and metabolic parameters. Initial levels of PAI-1 and changes in body mass contributed to the fall in PAI-1 (adj. R2 = 0.18, p = 0.0016) and initial levels of tPA-Ag contributed significantly to changes in tPA-Ag (adj. R2 = 0.57, p <0.0001). The results suggest that changes in fibrinolytic parameters are associated with the loss in body mass but can occur independently of a concomitant reduction in fatness. Although initial PAI-1 and tPA-Ag predict the changes of these fibrinolytic parameters, the results do not exclude the possibility that the improvement in metabolic state and changes in unmeasured parameters related to physical activity and low-caloric diet could have influenced our findings.

No relationship between leptin and cortisol in obese children and adolescents.

Recent findings have shown that leptin downregulates the steroid producing system in the adrenal. We studied the interactions of leptin, insulin and cortisol in obese children and adolescents at different stages of maturation. In 44 boys (age 11+/-3.1 yr, body mass index [BMI] 29+/-5.3 [mean +/- SD]) and 35 girls (age 11.4+/-2.6 yr, BMI 29+/-4.3), blood levels of leptin, insulin, cortisol, and glucose were determined. Fat mass (FM) was calculated by bioelectrical impedance. No significant differences were found between boys and girls with respect to humoral and anthropometric characteristics. When children were divided according to maturation stage (prepubertal, pubertal, and late/postpubertal) insulin was higher in the more mature groups (p<0.01) and leptin was higher in the pubertal group (p=0.03). In the prepubertal and pubertal groups, the expected positive relationship between adiposity and leptin was found although the magnitude of this association decreased with maturity. In none of the groups studied was cortisol significantly correlated to leptin. Insulin (p=0.03) and glucose (p=0.01) were positively associated with cortisol in the prepubertal group after adjustment for adiposity. However, in the pubertal group an inverse correlation was found between insulin and cortisol (p=0.03), and between insulin and glucose after control for adiposity. In the late/ postpubertal group, no significant correlations were found between estimates of adiposity and humoral parameters even after adjustment for gender. Stepwise multiple regression failed to detect a significant influence of cortisol to explain the variation in leptin, and vice versa. BMI contributed to the variation in leptin (adj. R2 =0.275, p<0.0001), and glucose added 5% to the variation in cortisol (p=0.03). The results do not confirm the inverse association between leptin and cortisol found in adults. Although BMI reflects levels of leptin, it is likely that several other factors in conjunction with fatness modulate the relationship with leptin. Whether leptin per se exerts an influence on the hypothalamic-adrenal-adipo axis remains to be investigated in longitudinal studies.

Relationship between plasminogen activator inhibitor-1 antigen, leptin, and fat mass in obese children and adolescents.

Hyperleptinemia may be associated with cardiovascular risk and is linked with parameters of fibrinolytic processes in adults. We studied whether body fatness, leptin, and insulin interact with plasminogen activator inhibitor-1 antigen (PAI-1-Ag) and tissue-type plasminogen activator antigen (tPA-Ag) in obese children and adolescents. Twenty-three boys (mean +/- SD: age, 10.7 +/- 3.3 years; body mass index [BMI], 28.7 +/- 5.4 Kg/m2) and 19 girls (age, 11.9 +/- 2.7 years; BMI, 29.4 +/- 4.8 Kg/m2) were investigated. Body fat mass (FM) in the children was calculated by bioelectrical impedance analysis, and blood samples were obtained for leptin, insulin, C-peptide, PAI-1-Ag, and tPA-Ag. The children were divided into 3 subgroups according to maturation. Maturity was associated with greater adiposity and higher levels of leptin and C-peptide, but insulin and PAI-1-Ag were not different between prepubertal, pubertal, and late/postpubertal children. PAI-1-Ag was associated with leptin and insulin, but not after adjustment for fatness. PAI-1-Ag was independently associated with tPA-Ag (r = .36, P < .02). Multiple regression analysis showed that tPA-Ag failed to reach the level of significance (P = .07), but FM contributed to the variation in PAI-1-Ag (adjusted R2 = .29). The BMI was the main determinant for the variation in leptin (adjusted R2 = .386) and in insulin (adjusted R2 = .60, all P < .001). Neither gender, maturation, chronological age, or leptin contributed significantly to the variation in either PAI-1-Ag or tPA-Ag. Our data suggest that adiposity and other variables contribute to higher levels of PAI-1-Ag. Leptin seems not to be independently linked with fibrinolytic parameters, but an unfavorable metabolic and fibrinolytic risk profile might emanate from the obese pubertal stage.

No evidence for leptin as an independent associate of blood pressure in childhood and juvenile obesity.

We studied whether leptin is an independent associate of blood pressure in obese children and adolescence. 102 obese children (48 girls, age: 11.6 +/- 2.22 yr; body mass index [BMI]: 27.45 +/- 4.4; blood pressure: 122.5 +/- 11.1/64.7 +/- 10.6 mm Hg and 54 boys, age: 11.5 +/- 2.4 yr; BMI: 27.6 +/- 4.4; blood pressure: 122.5 +/- 13.2/60.9 +/- 8.1 mm Hg [mean +/- SD]) were investigated. Serum leptin and insulin were measured by RIA; glucose was determined enzymatically. Fat mass (FM) was calculated by bioelectrical impedance. Leptin was higher in girls than in boys (p=0.018) but no significant gender differences were found with respect to indices of adiposity and systolic blood pressure (SBP). Children were divided into three groups, according to pubertal stage (Group 1: prepubertal, 32 boys/13 girls; Group 2: pubertal, 17 boys/25 girls; Group 3: late/postpubertal, 5 boys/10 girls). SBP and DBP correlated with body weight in the whole group (r=0.49, p<0.0001, and r=0.27, p=0.004). In Group 1, BMI showed the highest correlation to SBP; in Group 3 no indices of adiposity were related to SBP. In no case was leptin significantly associated with SBP after adjustment for adiposity. In Group 2, glucose was significantly associated with SBP after adjustment for body weight. In Group 3, however, no correlations were found between SBP, DBP and metabolic characteristics, perhaps due to small sample size. Stepwise multiple regression revealed that body weight and glucose contributed to the variation in SBP in the whole group (R2=0.31, p<0.0001). Insulin accounted for almost 8% of the variation in DBP (R2=0.08, p=0.0034). Body weight contributed significantly to SBP in boys (R2=0.39, p<0.0001) and girls (R2=0.24, p< 0.001). The results imply that body weight contributes independently to the variation in blood pressure. Glucose and insulin contribute to mean blood pressure to some extent, but our data do not support the assumption that leptin per se serves as an independent predictor of blood pressure in obese children and adolescents.

Determinants of haemostatic risk factors for coronary heart disease in obese children and adolescents.

OBJECTIVE: To investigate the contribution of serum lipids, parameters of glucose metabolism, body composition and cardiovascular fitness to the variance of several haemostatic risk factors for coronary heart disease (CHD) in obese children and adolescents. SUBJECTS AND MEASUREMENTS: Forty-two healthy, obese children and adolescents (20 male, 22 female, age 12.6 +/- 3.2y; body mass index (BMI), 30.4 +/- 5.3 kg/m2), were screened for haemostatic and metabolic risk factors for CHD. Thirty-five of the participants (18 male, age 13.5 +/- 2.9y; BMI, 29.9 +/-4.5kg/m2; 17 female, age 12.8+/-2.1 y, BMI, 31.1 +/- 5.3 kg/m2) were assessed for cardiovascular fitness by means of incremental cycle ergometer exercise. RESULTS: After adjustment for age, fat mass correlated significantly with plasminogen activator inhibitor-1 antigen (PAI-1-Ag) in boys and girls and factor VIIc only in girls. Children with lower power output (< or = 2.77W/kg) showed significantly higher values for factor VIIc, fibrinogen and tissue-type plasminogen activator antigen (tPA-Ag). Neither body composition nor cardiovascular fitness contributed independently to the variance of the determined haemostatic risk factors, except PAI-1-Ag, which has been shown to be determined by fat mass. In multiple linear regression analysis, triglycerides and PAI-1-Ag explained significant independent proportions of the variance of tPA-Ag. Factor VIIc was explained by C-peptide, insulin and fibrinogen. Von Willebrand factor antigen (vWF-Ag) was significantly related to glucose and insulin. CONCLUSION: The results suggest that in obese children and adolescents the haemostatic risk factors factor VIIc, vWF-Ag and tPA-Ag are mainly determinated by plasma insulin and triglyceride concentrations, but are primarily independent of body composition and cardiovascular fitness.

Correlation between cholesterol, soluble P-selectin, and D-dimer in obese children and adolescents.

Thirty-eight obese children and adolescents were investigated for a possible relation between cholesterol and markers of platelet activation, endothelial cell dysfunction, and activation of the coagulation system. Soluble P-selectin, von Willebrand factor antigen (vWf-Ag), D-dimer, and prothrombin fragment 1 + 2 (F1 + 2) were determined by enzyme-linked immunosorbent assays, and factor VIII coagulant activity (VIIIc) was measured by means of one-stage clotting assay. Cholesterol correlated significantly with log P-selectin (r = 0.43, P = 0.003) and log D-dimer (r = 0.33, P = 0.02). Cholesterol did not correlate with vWf-Ag, factor VIIIc, and F1 + 2. Log P-selectin correlated significantly with log D-dimer (r = 0.42, P = 0.003), which remained significant after adjustment for cholesterol (P = 0.02). Log D-dimer correlated significantly with F1 + 2 (r = 0.38, P = 0.01). Our study demonstrates that, in obese children and adolescents, cholesterol is significantly associated with P-selectin and D-dimer, and suggests an unfavorable intercorrelation between metabolic and hemostatic risk factors for coronary heart disease in childhood obesity.