Plasma lipid metabolic and transfer processes have recently been suggested to play an important role in the development of early restenosis, a major complication of percutaneous transluminal coronary angioplasty (PTCA); in particular, the common variants of genes for cholesteryl ester transfer protein (CETP) and paraoxonase (PONA) have been implicated. We had the opportunity to investigate this question in a large, prospective cohort characterized by quantitative coronary angiography in all subjects. The CETP-TaqIB (intron 1), CETP-MspI (intron 8), and PONA-AlwI (exon 2) polymorphisms were characterized in a cohort of 779 patients of whom 342 ("cases") had developed restenosis (as defined by > 50% loss of lumen compared with immediate postprocedure results) at repeat angiography at 6 months post PTCA. Selected frequencies for CETP B1 and B2 alleles (absence/presence of TaqIB site) were 0.65 and 0.35 (cases) and 0.65 and 0.35 (controls), respectively; frequencies for CETP M1 and M2 alleles (absence/presence of MspI site) were 0.20 and 0.80 (cases), 0.21 and 0.79 (controls), respectively; frequencies for PONA A and B alleles (absence/presence of AlwI site) were 0.73 and 0.27 (cases), 0.72 and 0.28 (controls), respectively. All observed genotype frequencies were in Hardy-Weinberg equilibrium. There was no evidence for gene-gene interaction, or an association between genotype and restenosis or degree of lumen loss (adjusted for covariates). Our data, collected in the largest study of its kind so far, indicate that the common variants for CETP and PONA are not associated with incidence of restenosis after PTCA, and are therefore not useful markers for risk assessment.