In patients with type 2 diabetes, Roux-en-Y gastric bypass (RYGB), which excludes a portion of the stomach and the proximal intestine from the alimentary circuit, improves glucose metabolism more rapidly and more extensively than is expected from weight loss. The mechanisms of this unique effect of gastrointestinal exclusion appear to be complex and have not yet been clarified. A recent study unveil that intestinal uptake of ingested glucose is diminished by RYGB and restricted to the common limb, where food meets bile and other digestive fluids, resulting in an overall decrease of post prandial blood glucose excursion. the hypothesize that reducing the length of the common limb, which is rarely measured and highly variable in clinical practice, may significantly affect the metabolic outcome of gastrointestinal surgical procedures. The aim of the present study is to compare the impact of two variants of Roux-en-Y gastric bypass with a short common limb, the long alimentary limb or the long biliary limb Roux-en-Y gastric bypass, on type 2 diabetes remission in severely obese patients.