Both type 1 and type 2 diabetes are associated with increased risk for cardiovascular disease (CVD) events. This risk seems to be reduced by achievement of euglycemia. However, after the withdrawal of rosiglitazone from the market, the question arose as to whether this risk concerns simply a matter of euglycemia or the distinct role played by each antidiabetic drug with respect to its effect on CVD risk. To address this issue, many studies have been published during the last decade involving old and new antidiabetic agents, which however yielded contradictory results. Briefly, metformin is still considered safe and confers a beneficial effect on CVD risk. Conflicting data exist as concerns sulfonylureas, although the second and third generation representatives are regarded as relatively safe. Pioglitazone use seems to be associated with a reduction in CVD risk, whereas the dipeptidyl-dipeptidase-4 inhibitors (DPP-4i), lixisenatide and exenatide-LAR [from the category of glucagon-like-peptide-1 receptor (GLP-1R) agonists], confer a neutral effect. Two other GLP-1R agonists, liraglutide and semaglutide, as well as the sodium-glucose transporter-2 (SGLT2)-inhibitors, empagliflozin and cangliflozin, have shown an additional effect on CVD risk reduction, although their safety is in doubt. Insulin analogues and newer long-acting compounds are also safe for the cadiovascular system. The aim of this narrative review is to present and critically analyse the current data for each antidiabetic drug category with regard to their effect on CVD risk.