We studied bone mineral density (BMD), bone metabolism, and body composition in 47 children with central precocious puberty (n = 36) or early puberty (n = 11) before, during, and after cessation of GnRH agonist. Bone density and body composition were measured with dual energy x-ray absorptiometry and expressed as SD scores. Bone age and biochemical parameters of bone turnover were assessed. Measurements were performed at baseline, after 6 months, and on a yearly basis thereafter. Mean lumbar spine BMD SD scores for chronological age were significantly higher than zero at baseline and decreased during treatment. Lumbar spine bone mineral apparent density and total body BMD did not differ from normal at baseline and showed no significant changes during treatment. In contrast, BMD SD scores for bone age were significantly lower than zero at baseline and at cessation of therapy. Two years after therapy, bone mineral apparent density and BMD SD scores for bone age and chronological age did not differ from normal. Markers of bone turnover decreased during treatment, mainly in the first 6 months. Patients had increased percentage of fat and lean body mass at baseline. After an initial increase of percentage body fat during treatment, percentage body fat decreased and normalized within 1 yr after cessation of treatment. Our longitudinal analysis suggests that peak bone mass or body composition will not be impaired in patients with precocious or early puberty after GnRH agonist therapy.