The Association Between Antidiabetic Agents and Clinical Outcomes of COVID-19 Patients With Diabetes: A Bayesian Network Meta-Analysis

Abstract

Aims: This study aimed to assess the impact of different antidiabetic agents on individuals with diabetes and COVID-19. Methods: We searched PubMed, Web of Science, Embase, and Cochrane Library databases from inception to October 31, 2021 and included seven antidiabetic agents. The data were pooled via traditional pairwise meta-analysis and Bayesian network meta-analysis. Results: The pairwise meta-analysis included 35 studies. Metformin (odds ratio (OR), 0.74; P=0.001), dipeptidyl peptidase-4 inhibitors (DPP4i) (OR, 0.88; P=0.04), sodium-glucose cotransporter-2 inhibitors (SGLT2i) (OR, 0.82; P=0.001), and glucagon-like peptide-1 receptor agonists (GLP1RA) (OR, 0.91; P=0.02) treatment were associated with lower COVID-19 mortality in individuals with diabetes compared to respective non-users. However, insulin treatment resulted in higher mortality (OR, 1.8; P=0.001). Mortality did not significantly differ in sulfonylurea (OR, 0.97; P=0.56) and thiazolidinediones (TZDs) (OR, 1.00; P=0.96) users. Furthermore, due to limited data, we analyzed five antidiabetic agents (metformin, DPP4i, sulfonylurea, insulin, and SGLT2i) and found no association between them and severe disease risk (all P>0.05). The Bayesian network meta-analysis included 18 studies. GLP1RA and SGLT2i had the highest first and second rank probability (67.3% and 62.5%, respectively). Insulin showed the maximum probability of ranking seventh (97.0%). Metformin had the third and fourth highest rank probability of 44.8% and 38.9%, respectively. Meanwhile, DPP4i had the fifth-highest rank probability of 42.4%, followed by sulfonylurea at 45.1%. Conclusion: Metformin, DPP4i, SGLT2i, and GLP1RA treatments were highly possible to reduced COVID-19 mortality risk in individuals with diabetes, while insulin might be related to increased mortality risk. Sulfonylurea and TZDs treatments were not associated with mortality. None of the antidiabetic agents studied were associated with the risk of severe disease. Additionally, GLP1RA probably had the most significant protective effect against death, followed by SGLT2i and metformin. Systematic Review Registration: PROSPERO (CRD42021288200)