Abstract
AIMS: COVID-19 has and still is sweeping away the national health systems worldwide. In this review, we sought to determine the evidence base proofs on the antidiabetic treatment capable to reduce the risk of COVID-19-related mortality. METHODS: We have performed a systematic search of published articles using PubMed, and EMBASE from March 2020 to March 31st, 2021. We excluded editorials, commentary, letters to the editor, reviews, and studies that did not have mortality as an outcome. For metformin and insulin only, we performed a meta-analysis using Cochrane RevMan 5.2. RESULTS: Among antidiabetic drugs, metformin was the only drug associated with a reduced risk of mortality. Conversely, insulin appears associated with an increased risk. The other classes of drugs were neutral. CONCLUSIONS: The totality of articles reports retrospective data strongly affected by "channeling bias" so that most of the existing results on each class of drugs are driven by the phenotype of patients likely to receive that specific drug by prescription.