BACKGROUND: Discontinuation of metformin therapy, if started beyond menarche in adolescents or young women with hyperinsulinaemia following low birthweight, is rapidly followed by rebound deteriorations in body fat, insulin resistance and blood lipid profile. OBJECTIVE: We hypothesized that early commencement of metformin and its continuation throughout puberty might have more persisting benefits. PATIENTS AND MEASUREMENTS: We followed up on a previously reported randomized study cohort at 12 months and 18 months after treatment discontinuation, including body composition by absorptiometry, fasting insulin, glucose and blood lipids. In that open-labelled, prospective study, 22 low birthweight girls with early normal puberty (Stage 2 breast development at age 8-9 years) were randomized to remain untreated (N = 12] or to receive metformin (850 mg/day; N = 10) for 36 months (between time -36 months to 0 month). RESULTS: The significant improvements previously reported at the end of the 36-month active treatment period in per cent body fat, abdominal fat mass, fasting insulin sensitivity, high density lipoprotein (HDL) cholesterol and triglyceride levels all persisted at follow-up 12 months after treatment discontinuation. Further anthropometry at 18 months off therapy confirmed the persistence of benefits in height, body mass index (BMI) and waist circumference in the previously metformin-exposed girls. CONCLUSION: In low birth weight girls with early normal onset of puberty, metformin treatment for 3 years across puberty resulted in auxological, endocrine and metabolic benefits that persisted for at least 1 year after metformin withdrawal. Further follow-up and longer-term studies are needed to explore the possibility that insulin sensitization therapy during puberty might reprogramme predisposition to metabolic disease.