Aims Paediatric obesity is a growing concern for the health service. There is currently no consensus for routine screening of metabolic profiles and medical treatment in obese paediatric patients. We aimed to determine medium-term outcomes of Metformin treatment on BMI, glucose and insulin levels in obese paediatric patients.

Methods In a retrospective review, data were collected from obese paediatric patients treated with Metformin for insulin resistance between October-09 and October-14. Demographic data was collected from these patients including presence of acanthosis nigricans (AN) and family history of T2DM. Changes in BMI SDS, glucose and insulin were analysed. Paired sample T-tests were used to compare pre and post treatment results (treatment washout period of 1 month).

Results 70 patients were treated with metformin (50=female) (35=British White, 18=Pakistani) at a mean age of 12.7 (6.1–17.2) years. Mean BMI 35.2 (24.2–48.5 kg/m²) and BMI SDS 3.4 (2.2–4.7). All the patients had insulin resistance at start of treatment. 21 (30%) of these had both family history of T2DM and AN. An additional 22 patients had AN only (total with AN=43, 49%). Those with AN had significantly higher basal insulin levels (p < 0.05) than those without. Metformin was associated with reduced BMI z-score at 6–12 months (-0.1 SDS, p < 0.05) and 18–24 months (-0.2 SDS, p < 0.05). Reduction in fasting and postprandial glucose levels were (-0.1 mmol/L, p = 0.17) and (-0.5 mmol/L, p = 0.17) respectively, with a significant reduction in postprandial glucose levels in those patients with impaired glucose tolerance (n = 13) (-1.9 mmol/L, p < 0.05). Metformin was associated with a reduction in fasting insulin (-3.0 mU/L,p = 0.44), and significantly reduced 2 h insulin (-118.0 mU/L, p < 0.05) after treatment for 12–18 months.

Conclusions Acanthosis nigricans is a good clinical indicator for the presence of insulin resistance. Metformin treatment is significantly associated with reduction in BMI z-score from 6 months and reduced postprandial insulin levels after treatment. It should be considered as a treatment modality in normoglycaemic obese paediatric patients for weight stabilisation and improvement of insulin resistance, which may have longer term implications on metabolic health.