Leptin measurement in urine in children and its relationship to other growth peptides in serum and urine.

OBJECTIVE: Leptin has been implicated in the interaction between nutrition, energy balance and sexual maturation in humans. A non-invasive method of measuring leptin would greatly facilitate longitudinal studies of changes in leptin in normal children. The aim of this study was to evaluate the use of urinary leptin as a surrogate for serum leptin measurements. DESIGN: We have modified and validated a serum immunoradiometric assay (IRMA) kit for the measurement of leptin in urine, and subsequently investigated the relationship between urinary leptin and other growth-related proteins. METHODS: Cross-sectional study: urinary leptin, measured in the first morning urine voided and expressed as ng excreted overnight, and serum concentrations of leptin, IGF-I, IGF-II, IGFBP-3 and IGFBP-1 were determined in a cohort of 188 healthy schoolchildren aged 5-19 years (88 males, 100 females). Height, weight and pubertal status were assessed in all children. Longitudinal study: urinary levels of leptin, IGF-I and GH were measured daily in two adults (one male, one female) over a period of 6 weeks. RESULTS: The detection limit of this modified assay was 0.59 ng/L. The intra- and interassay coefficients of variation range was 4-8% and 4-9%, respectively. The recovery of recombinant leptin added to urine was 98-108%, and the assay had a recovery rate for serial dilution in the range of 106-112%. Urinary leptin correlated significantly with serum leptin (r = +0.65, P < 0.01). Urinary leptin showed similar changes through puberty to those of serum leptin, with levels rising in females throughout puberty, whereas in males levels peaked at G2/G3 then decreased. BMI SDS was the main determinant of urinary leptin, as it was for serum leptin. In the cross-sectional study urinary leptin correlated significantly with serum IGF-I (r = +0.41, P = 0.001), IGF-II (r = +0.19, P = 0.05), IGFBP-3 (r = +0.29, P = 0.001) and IGFBP-1 (r = -0.25, P = 0.001). In the adult study, leptin was also detected in urine with similar night-to-night variability to that found for IGF-I and GH. CONCLUSION: Urinary leptin is a valid marker of serum leptin concentrations, and therefore this non-invasive assay would be a useful tool for longitudinal assessment of changes in leptin in children.

Serum leptin and IGF-I during growth hormone treatment in chronic renal failure.

Serum leptin decreases during growth hormone (hGH) treatment and pre-treatment values have been suggested as a predictor of the response to hGH in GH deficiency (GHD) but not in non-GHD syndromes. To investigate whether this holds true in children with chronic renal failure (CRF), we evaluated changes in serum leptin, insulin-like growth factor-I (IGF-I) and height before(b) and during the 1st year (3 months, 6 months, 9 months, 12 months) of hGH treatment (1 IU/kg per week) in 11 children (median age(b) 10.1 years, mean height(b) -2.9 SDS) with CRF. Serum leptin and IGF-I were compared with values from healthy children. Each patient also served as his/her own control, with values during treatment compared with those before treatment. Growth improved in all patients during treatment (mean change(12 m) +7.2 cm, change in height SDS(12 m) +0.5, P=0.001). Weight decreased (median decrease(12 m) 0.3 SDS, P=0.02) but body mass index (BMI) and serum leptin did not change during treatment. Serum IGF-I levels were low before (mean -1.1 SDS) but increased during hGH treatment, the increment being greatest at 10 days (mean increment +1.9 SDS, P<0.0001). Serum leptin(b) did not correlate with change in serum IGF-I(10d), height(12 m) or weight(12 m). Serum IGF-I SDS(b) correlated with height SDS at 12 months ( r=0.80, P=0.006) of hGH treatment. Serum leptin(b) correlated with BMI ( r(s)=0.75, P=0.01). Levels adjusted for BMI did not differ from values in healthy children and did not change during treatment. Despite an IGF-I and growth response during hGH treatment, serum leptin did not change and pre-treatment values did not predict the growth response in these children with CRF.