Surprisingly enough, insulin has recently been suspected of promoting cardiovascular complications, provided it becomes effective in excess concentrations at the vascular walls. This applies to both endogenous hyperinsulinemia and insulin substitution performed exogenously with high insulin doses. Several large population studies in non-diabetics - carried out in Helsinki, Busselton and Paris - have proved recently that serum insulin concentration is an independent risk predictor for the occurrence of coronary heart disease. According to in vitro trials, insulin stimulates the proliferation of the smooth muscle cells in vascular media and the lipid synthesis, as well as lipid incorporation into the vascular wall. Very recent large-scale population studies in randomly selected type II diabetics (the Schwabing Study and one performed at Oxford) have revealed a close association between an endogenous insulin requirement, on the one hand, and the risk of macroangiopathy or coronary heart disease on the other.