Safety CVOTs have been performed with GLP-1 receptor agonists and 4 of such trials have been published. All of them recruited diabetic subjects at very high CV risk and called for the addition of the drug of interest (or placebo) to the ongoing diabetes treatment. This needed to be adjusted in the attempt to achieve acceptable and super imposable metabolic control in both arms (equipoise). Analysis of the primary outcome (a composite of CV death, non fatal MI and non fatal stroke) revealed that all GLP-RAs are safe. However, liraglutide and semaglutide were significantly associated with a 13 to 25 % reduction in the number of events. A trend toward reduction barely missing significance was observed for exenatide LAR as well, while no such a trend was observed for lixisenatide. These results suggest that GLP-RA formulations ensuring sustained plasma concentrations of the agent might exert protection against further CV events in high risk diabetic subjects. The “potency” of this protective action might differ among different agents. It has to be kept in mind, however, that in all studies equipoise was attempted but not achieved and that, more or less in the different studies, subjects in the “active” arm had better glucose and blood pressure control and experienced less hypoglycemia and less treatment intensification with insulin and/or sulphonylureas. Thus, it cannot be excluded that the CV protection observed with GLP-1 RAs is at least partially related to the fact that by these agents a better overall control can be achieved with less side effects.