Acute Effects of Acarbose and Canagliflozin, Alone and in Combination, on Postprandial GLP-1 and Glycaemic Responses in Post-Bariatric Hypoglycaemia: A Randomised Crossover Study

AIM: To evaluate the acute effects of acarbose and canagliflozin, alone and in combination, on postprandial total GLP-1 and glycaemic responses during a mixed meal test in adults with post-bariatric hypoglycaemia (PBH). MATERIALS AND METHODS: This single-centre, randomised, four-period crossover physiological study included 12 adults with PBH after Roux-en-Y gastric bypass. Participants underwent meal tests after placebo, acarbose 100 mg, canagliflozin 300 mg or combination treatment. Co-primary outcomes were total GLP-1 area under the concentration-time curve (AUC) from 0 to 240 min and peak total GLP-1 concentration. Secondary outcomes included glucose, insulin, C-peptide, GIP and glucagon. RESULTS: Eleven participants completed all four visits and all 12 were analysed. During the placebo mixed meal test, 3/12 participants reached glucose < 3.0 mmol/L, while 10/12 reached < 3.5 mmol/L and 10/12 reported postprandial symptoms suggestive of hypoglycaemia. All active treatments reduced peak glucose and glucose excursion versus placebo, without changing glucose nadir or glucose AUC 0-240 min. Peak insulin, insulin AUC 0-240 min, peak C-peptide and C-peptide AUC 0-240 min were lower with both acarbose-containing regimens than with placebo. Total GLP-1 AUC 0-240 min did not differ between treatments, while peak total GLP-1 was higher with canagliflozin than with the other treatments. Peak GIP was lower with acarbose-containing regimens than with placebo, and GIP AUC 0-240 min was lower with combination treatment than with placebo. Peak glucagon was higher with combination treatment than with placebo. CONCLUSIONS: Acarbose-containing regimens attenuated postprandial beta-cell responses, but adding canagliflozin did not provide clear additive benefit over acarbose alone on the prespecified GLP-1 outcomes or glucose nadir. Changes in glycaemic and beta-cell responses without a parallel reduction in total GLP-1 suggest that mechanisms beyond GLP-1 secretion contribute to the acute postprandial phenotype in PBH.