The study of potentially effective drug therapies and mechanical devices for the prevention of restenosis after percutaneous coronary revascularization has relied heavily on the use of experimental animal models. To date, greater than 50 experimental studies have been reported and have suggested that at least nine different classes of pharmacologic agents inhibit the intimal proliferative response to arterial injury. However, no pharmacologic intervention has yet been shown to reproducibly reduce the incidence of restenosis after coronary balloon angioplasty in humans. To identify the reasons for the apparent nonspecificity of the animal models and to determine which model should most reliably predict the efficacy of individual therapies in humans, the distinguishing characteristics of the experimental models were compared. Particular attention was paid to the size and morphologic structure of the treated artery, the susceptibility of the species to spontaneous and diet-induced arterial disease, the nature of the stimulus to intimal proliferation and several practical and logistic considerations. Finally, the reported efficacies of specific drug therapies in the respective animal models and in humans were compared. This review suggests that significant interspecies and occasionally intraspecies differences do exist among the respective animal models, particularly in the extent and composition of the neointimal thickening.(ABSTRACT TRUNCATED AT 250 WORDS)