Pubertal changes in serum leptin levels in adolescents with type 1 diabetes mellitus: A controlled longitudinal study

The mechanism of the pubertal delay seen in some adolescents with type 1 diabetes mellitus is not entirely clear. Since leptin has been implicated as a neuroendocrine modulator of puberty, we measured serum leptin levels longitudinally in 24 post-'honeymoon' patients with diabetes mellitus (M/F = 15/9) with a mean (+/-SD) age of 10.5 +/- 0.9 years and 26 controls (M/F = 15/11) with a mean age of 10.0 +/- 1.1 years. Physical examinations; serum leptin, IGF-I, IGFBP-3 and IGFBP-1 levels; and bone age X-rays were performed annually for up to 48 months. Glycosylated hemoglobin (HbA(1c)) was measured 2-4 times a year in patients with diabetes mellitus. Serum leptin levels strongly correlated with the body mass index z-scores (BMI-Z) in both controls (r = 0.666, p <0.00001) and diabetic patients (r = 0.577, p <0.00001). Girls had increased serum leptin levels for a given BMI compared to boys (p <0.005). There were no significant differences in serum leptin levels of patients with diabetes mellitus compared to controls, nor were differences seen when the groups were stratified by age, Tanner stage, or gender. There were also no significant correlations between serum leptin levels and degree of metabolic control (i.e. HbA(1c)) or insulin dose standardized for body weight. Although there was no significant diabetes-related or metabolic control-related delay in bone age z-score or pubertal development, there was a significant negative correlation between HbA(1c) and growth velocity z-score, indicating that children with poor diabetes control had modest but significant slowing of growth. It is concluded that neither pubertal development nor serum leptin levels are significantly altered in adolescents with diabetes mellitus managed with standard therapy. The potential role of leptin in initiation of pubertal development is not easily demonstrable in observational studies.