AimTo identify risk factors predictive of or associated with gastrointestinal side effects (GISE) of liraglutide in patients with type 2 diabetes (T2DM).MethodsT2DM patients treated with liraglutide for the first time were obtained and grouped into patients without GSEA and patients with GSEA. Baseline variables, including age, sex, body mass index (BMI), glycemia profiles, alanine aminotransferase, serum creatinine, thyroid hormones, oral hypoglycemic drugs and history of gastrointestinal diseases, were tested for possible associations with GSEA outcome. Significant variables were entered into univariate and multivariate logistic regression (forward LR) analyses. Receiver operating characteristic (ROC) curves to determine clinically useful cutoff values.ResultsA total of 254 patients (95 female) were included in this study. 74 cases (29.13%) reported GSEA and 11 cases (4.33%) discontinued treatment. The results of univariate analyses showed that sex, age, thyroid stimulating hormone (TSH), free triiodothyronine, α-glucosidase inhibitor (AGI), and concomitant gastrointestinal diseases were associated with GSEA occurrence (all p <0.05). In the final regression model, AGI use (adjusted OR=4.01, 95%CI: 1.90-8.45, p<0.001), gastrointestinal diseases (adjusted OR=3.29, 95%CI: 1.51-7.18, p=0.003), TSH (adjusted OR=1.79, 95%CI: 1.28-2.50, p=0.001) and male sex (adjusted OR=0.19, 95%CI: 0.10-0.37, p<0.001) were independently associated with GSEA. Furthermore, ROC curve analysis confirmed that TSH values of 1.33 and 2.30 in females and males, respectively, were useful thresholds for predicting GSEA.ConclusionThis study suggests that the combination of AGI, concomitant gastrointestinal diseases, female sex and higher TSH levels are independent risk factors of GSEA of liraglutide treatment in patients with T2DM. Further research is warranted to elucidate these interactions.