The purpose of the present study was to identify gene polymorphisms that confer susceptibility to recurrent restenosis after bare-metal stenting of coronary arteries, and thereby to assess the genetic risk for this condition. The study population comprised 527 unrelated Japanese individuals, including 28 subjects who developed in-stent restenosis two or more times and 499 subjects without restenosis. The genotypes for 142 polymorphisms of 121 candidate genes were determined with a method that combines the polymerase chain reaction and sequence-specific oligonucleotide probes with suspension array technology. Eleven polymorphisms were related (P<0.05) to the prevalence of recurrent in-stent restenosis as determined by the Chi-square test. Multivariable logistic regression analysis with adjustment for age, sex, body mass index, and the prevalence of smoking, hypertension, diabetes mellitus, and hypercholesterolemia revealed that the -55Cright curved arrow T polymorphism of the uncoupling protein 3 gene (UCP3) was significantly (P=0.0006 in a recessive model) associated with the prevalence of recurrent in-stent restenosis, with the T allele representing a risk factor for this condition. A stepwise forward selection procedure showed that the UCP3 genotype significantly (P=0.0014, recessive model) affected the prevalence of recurrent in-stent restenosis. Determination of the genotype for UCP3 may thus contribute to assessment of the genetic risk for recurrent in-stent restenosis.