ObjectiveTo investigate whether metformin has protective effect on myocardial injury in patients with coronavirus disease 2019 (COVID-19) combined with coronary heart diseases and diabetes.MethodsCOVID-19 patients with coronary heart disease and diabetes who were admitted to Tongji Hospital from January 18 to April 25 in 2020 were enrolled. They were divided into a metformin group and a none-metformin group according to whether the metformin was used. The demographic characteristics, clinical symptoms, laboratory parameters, treatment and clinical outcomes of the two groups were analyzed retrospectively.ResultsThere were 29 patients in the metformin group, 3 patients (12.0%, 3/25) suffered myocardial injury and 1 (3.4%) died of acute respiratory failure complicated by septic shock; 67 patients were in the non-metformin group and 24 (37.5%, 24/64) had myocardial injury but 15 died in hospital among whom 1 died of septic shock complicated by disseminated intravascular coagulation, 1 acute respiratory failure complicated by possible cerebral hemorrhage, 2 acute respiratory failure, 1 fulminant myocarditis, 3 acute myocardial infarction and 7 cardiac arrest. The incidence of myocardial injury (12.0% vs. 37.5%, P=0.019), hospital mortality (3.4% vs. 22.4%, P=0.034) and mortality of cardiovascular events (0.0% vs. 16.4%, P=0.049) in the metformin group were significantly lower than those in the non-metformin group. Multivariate analysis showed that the use of insulins (OR=11.235, P=0.003) was an influencing factor for in-hospital mortality of patients. The use of metformin (OR=0.154, P=0.013) was positively correlated with the myocardial injury.ConclusionWhen patients with coronary heart disease and diabetes are infected with COVID-19, metformin can effectively reduce myocardial damage and has a certain effect on reducing hospital mortality. Combined with clinical considerations, it is worthy of popularization.