ESPE Abstracts

Background: Metformin is a well-known biguanide approved for treatment of Type2 Diabetes. Metformin is not considered as an anti-obesity drug despite its common off-label use. Currently, there are no data regarding the profile of metformin related gastrointestinal side effects (MRGSE) in obese adolescents.

Aim: To identify the frequency and time course of MRGSE in obese adolescents and assess the presence of any association of side effects with change in BMI standard deviation (∆BMI-SD).

Design-Setting: 6-month prospective, observational study in a tertiary-care center

Patients and Methods: 200 consecutive treatment-naïve adolescents with exogenous obesity aged 10-17yr were enrolled. Metformin was given (maximum dose: 2g/d) to 100 obese patients with glucose intolerance and/or insulin resistance(IR) in addition to standard dietary and life style related recommendations. Standard clinical and biomedical evaluations were performed. MRGSE profile; i.e. nausea, vomiting, diarrhea, abdominal pain and the severity of these side effects was recorded and evaluated with a validated likert scale questionnaire on a monthly basis. Associations between metformin dose, severity of side effects and ∆BMI-SD were investigated.

Results: 173 patients (103females) with the median age of 13.5 yr completed the study. The median BMI-SD at presentation was 2.3 (range:1.8-3.8). Compared to patients in the untreated group, patients in the metformin treated group (MTG)(n=98) had similar BMI-SD (P=0.09) and higher HOMA-IR (P=0.02); as expected. ∆BMI-SD was higher in the MTG than in the untreated group at the sixth month when compared to baseline BMI-SD, but not significantly (P=0.074). In the MTG; severe, moderate and mild MRGSE were noted in 13(13.7%), 14(14.7%), 9(9.5%) patients; respectively at the first month of follow-up. The majority of MRGSE diminished significantly at the fourth month (P=0.007). At the sixth month,13(14%) patients had MRGSE with varying degrees of severity. The presence of MRGSE was unrelated to gender, BMI and HOMA-IR (P> 0.05;for all). In the MTG, the decrease in BMI-SD was greater at the first and second months of follow-up in the patients with MRGSE than those without (P=0.036 and P=0.048; respectively). These significant differences in ΔBMI-SD owing to MRGSE did not persist on further follow-up.

Conclusions: The frequency of MRGSE was variable in our cohort with the majority subsiding at the fourth month. The initial mild decrease in BMI-SD during metformin treatment owing to MRGSE disappeared on follow up, thereby confirming the absence of a direct association of metformin use with weight loss in parallel to the decline in IR.