Abstract
BACKGROUND: Diabetes is an independent predictor of poor outcomes in patients with COVID-19. We compared the effects of the preadmission use of antidiabetic medications on the in-hospital mortality of patients with COVID-19 having type 2 diabetes. METHODS: A systematic search of PubMed, EMBASE, Scopus and Web of Science databases was performed to include studies (except case reports and review articles) published until November 30, 2021. We excluded papers regarding in-hospital use of antidiabetic medications. We used a random-effects meta-analysis to calculate the pooled OR (95% CI) and performed a sensitivity analysis to confirm the robustness of the meta-analyses. MAIN FINDINGS: We included 61 studies (3,061,584 individuals), which were rated as having low risk of bias. The OR (95% CI) indicated some medications protective against COVID-related death, including metformin [0.54 (0.47-0.62), I(2) 86%], glucagon-like peptide-1 receptor agonist (GLP-1RA) [0.51 (0.37-0.69), I(2) 85%], and sodium-glucose transporter-2 inhibitor (SGLT-2i) [0.60 (0.40-0.88), I(2) 91%]. Dipeptidyl peptidase-4 inhibitor (DPP-4i) [1.23 (1.07-1.42), I(2) 82%] and insulin [1.70 (1.33-2.19), I(2) 97%] users were more likely to die during hospitalization. Sulfonylurea, thiazolidinedione, and alpha-glucosidase inhibitor were mortality neutral [0.92 (95% CI 0.83-1.01, I(2) 44%), 0.90 (95% CI 0.71-1.14, I(2) 46%), and 0.61 (95% CI 0.26-1.45, I(2) 77%), respectively]. The sensitivity analysis indicated that our findings were robust. CONCLUSIONS: Metformin, GLP-1RA, and SGLT-2i were associated with lower mortality rate in patients with COVID-19 having type 2 diabetes. DPP-4i and insulin were linked to increased mortality. Sulfonylurea, thiazolidinedione, and alpha-glucosidase inhibitors were mortality neutral. These findings can have a large impact on the clinicians' decisions amid the COVID-19 pandemic.