The prevention of restenosis after coronary angioplasty has been marked over recent years by the failure of trials of drug treatment based on inhibition of arterial smooth muscle cell proliferation. This failure could be due to an insufficient concentration of the orally or intravenously administered drug in the lesion to be treated. Another reason for this failure of drug treatment could be the nonexclusive role of intimal hyperplasia in the pathophysiology of restenosis, which also appears to be related to a education of the overall calibre of the artery at the site of dilatation. The pathogenesis of this phenomenon, called "remodelling", remains obecure and is only partly prevented by insertion of stents, which is currently the only treatment able to decrease the number of new revascularization procedures for restenosis. This benefit is related to optimization of the initial result (stents avoiding early "recoil", secondary to the elastic recoil forces of the arterial wall), and possibly to prevention of late remodelling of the vessel at the site of dilatation, either tonic (vasomotor) or trophic. On the other hand, the benefit related to the absence of "remodelling" of the stended lesions is partly limited by intimal hyperplasia, exacerbated by the presence of the stent. The future therapeutic strategy could combine insertion of stents and prevention of smooth muscle cell proliferation by new treatment strategies acting at the molecular level. Encouraging preliminary results have already been obtained in animals with chimeric toxins, antisense strategies and especially gene therapy using defective adenoviral vectors for replication.