Epidemiological, population and familial studies have revealed the multifactorial aspect of diabetes mellitus. Several mutations implicated in the pathogenesis of diabetes have been described over the last decade. These mutations are localised within genes associated with glucose metabolism, providing a molecular basis for the heterogeneity in the clinical presentation of diabetes mellitus. However, chronic hyperglycaemia associated with other vascular risk factors can only partly explain the incidence of micro and macrovascular complications. Familial studies have revealed the presence of a familial susceptibility for some vascular complications as nephropathy and coronary heart disease. In addition, these two vascular complications of diabetes mellitus are frequently associated in the same individual. This familial susceptibility could not be exclusively explained by environmental factors. Consequently the quest for susceptibility genes of vascular complications appears as a logical approach. The study of genes associated with an increased cardiovascular risk like the renin angiotensin system, the hemostasis cascade or the lipoproteins, may constitute the first step in this new research avenue. Moreover, glycation and oxidation pathways seem to play a role in the development of vascular complications. For example, the paraoxonase genes are good candidates for an increased vascular risk. This enzyme is entirely bound to HDL-cholesterol and could explain its anti-oxidant capacity. The natural substrate of this enzyme is unknown but there is some evidence suggesting that it may participate in the oxidated phospholipids degradation. Functional studies of paraoxonase with other exogenous substrates have revealed different phenotypes associated with different catalytic activities. In addition, varying enzymatic activities seem to be associated with different polymorphisms of the paraoxonase gene recently described (at position 192 and 55 of the paraoxonase gene), and these two polymorphisms have been recently studied in relation with coronary heart disease in non insulin dependent diabetic patients. The two polymorphisms were associated with coronary heart disease. But these initial results still await confirmation in different populations. Such studies will likely open the way to novel approach of vascular complications in diabetes mellitus.