Occurrence of nausea, vomiting and diarrhoea reported as adverse events in clinical trials studying glucagon-like peptide-1 receptor agonists: A systematic analysis of published clinical trials.

AIM: GLP-1 receptor agonists (RAs) may cause nausea, vomiting or diarrhoea. The aim of this study was to assess the risk of adverse events (AEs) with GLP-1 RAs and their relation to dose, background medication and duration of action. RESEARCH DESIGN AND METHODS: The PubMed database was searched and 32 clinical trials with GLP-1 RAs (phase 3) were selected. We performed a systematic analysis and compared the proportion of patients reporting nausea, vomiting or diarrhoea, for different doses and glucose-lowering background medications, and relative to a reference compound within the subclasses of short- (exenatide b.i.d.) and long-acting (liraglutide) GLP-1 RAs, calculating the relative risks ± 95% confidence intervals. RESULTS: The risk of nausea was dose-dependent for long-acting (P = .0063) and across all GLP-1 RAs (P = .0017), and a similar trend was observed for vomiting (P = .23). Diarrhoea was dose-dependent (P = .031). Background treatment with metformin was associated with more nausea (P = .04) and vomiting (P = .0009). Compared to exenatide b.i.d., there was less nausea and diarrhoea with lixisenatide. Compared to liraglutide, there was a similar risk associated with dulaglutide, and less with exenatide q.w. and albiglutide. Long-acting GLP-1 RAs were associated with less nausea and vomiting, but with more diarrhoea than short-acting agents. CONCLUSIONS: GLP-1 RAs are associated with gastrointestinal AEs that are related to dose and background medications (especially metformin) and may vary in a compound-specific manner. Long-acting agents are associated with less nausea and vomiting but with more diarrhoea.