Is metformin better taken before or with meals? An acute evaluation of the impact of timing of metformin administration on intestinal glucose absorption, gut hormone secretion and the blood glucose response to an intraduodenal glucose infusion in type 2 diabetes

Following enrolment, each subject will be studied on 4 occasions, separated by at least 7 days, in a double-blind, randomized, crossover design with treatment allocation assigned by the Royal Adelaide Hospital Pharmacy. During the study period, patients will only need to cease their morning dose of metformin on the respective study days to allow for evaluation of the impact of timing of metformin administration on the study outcomes. On the evening preceding the study day (~1900h), participants will be given a standardised evening meal (McCain’s frozen beef lasagne (McCain Foods Proprietary Ltd, Victoria, Australia); 2472kJ) to consume with water. Following this meal, participants will be asked to fast from solids and liquids (water will be allowed until 10 pm) until the following morning, when they will attend the CRF of the AHMS building at 0800h. On each study day, a silicone rubber catheter (Dentsleeve International Ltd., Mui Scientific, Ontario, Canada) will be inserted through an anaesthetised nostril into the stomach, and allowed to pass into the small intestine by peristalsis. The catheter will be positioned with the intraduodenal infusion port located 12 cm distal to the pylorus. The correct positioning of the catheter will be monitored continuously by measurement of the transmucosal potential difference in the stomach (~ -40 mV) and the duodenum (~ 0 mV). For this purpose, an intravenous cannula will be placed subcutaneously in the left forearm and filled with sterile saline as a reference electrode. An intravenous cannula will be placed into a vein on the dorsum of the hand, which will be kept warm with a heat pad to allow sampling of “arterialised” blood. After correct positioning of the catheter, 1000 mg metformin dissolved in 30 mL water, or 0.9% saline as a control, will be administered intraduodenally over 2 min at t = -60, -30 or 0 min, in a double-blind, randomised fashion, i.e. one of the following 4 treatments: (i) 1000 mg metformin at t = -60 min + 0.9% saline at t = -30 and 0 min, (ii) 1000 mg metformin at t = -30 min + 0.9% saline at t = -60 and 0 min, (iii) 1000 mg metformin at t = 0 min + 0.9% saline at t = -60 and -30 min, or (iv) 0.9% saline at t = -60, -30 and 0 min. Commencing at t = 0 min, a solution containing 45 g glucose and 5g 3-OMG (as a marker of intestinal glucose absorption) dissolved in water to a total volume of 180 mL, will be infused into the duodenum over 60 minutes (i.e. 3 kcal/min). At t = 60 min, the catheter will be removed. The participant will be monitored for another 60 min (t = 60 - 120 min) before a light meal is served. “Arterialised” venous blood (~10 mL) will be sampled at t= -60, -30, 0, 30, 60, 90, 120 min. Blood glucose levels will be measured immediately at the bedside with a glucometer (Medisense Precision QID, Abbott Laboratories, Bedford, MA, USA). Plasma and serum samples will be separated from the remainder of each sample and stored at -80 degree Celsius for subsequent measurements. At the same intervals used for blood sampling, appetite and gastrointestinal sensations (including hunger, desire to eat, fullness, nausea, bloating, headache and abdominal pain) will be assessed using 100 mm visual analogue scales. After a final blood sample is collected, subjects will be served with a light lunch, and once the blood concentrations of subjects have stabilised above 5 mmol/L, subjects will be free to leave the laboratory. The total amount of blood drawn during the screening and 4 study visits will be ~300 mL.