The ID (insertion/deletion) polymorphism of the ACE (angiotensin-converting enzyme) gene controls plasma ACE levels. Both have been correlated with ISR (in-stent restenosis) in preliminary analyses, but not confirmed in larger studies. In the present study, baseline and 6-month quantitative coronary analysis were performed in 897 patients who had stent implantation and the ID polymorphism genotyped. Plasma ACE levels were measured in 848 patients (95%). Restenosis rates among genotypes were 31.2% DD, 25.5% ID and 28.8% II (not significant). Plasma ACE levels were significantly higher in restenotic patients compared with patients without restenosis (30.7+/-18.6 units/l compared with 22.8+/-12.8 units/l; P=0.0001) and a strong independent predictor of ISR [OR (odds ratio)=3.70; 95% CI (confidence interval), 2.40-5.71; P<0.0001], except in diabetics. In the subgroup of diabetics and patients with AMI (acute myocardial infarction), the DD genotypes actually had a lower risk of ISR than the II genotypes (diabetics, OR=0.16; 95% CI, 0.04-0.69; P=0.014; and patients with AMI, OR=0.21; 95% CI, 0.061-0.749; P=0.016). After exclusion of diabetics and patients with AMI, ISR rates for genotypes in 632 patients were 31.7% DD, 24.3% ID and 17.6% II (P=0.02; DD compared with non-DD OR=1.57; 95% CI, 1.09-2.25). The association between the D allele and ISR observed in selected populations does not hold with a larger sample size. Other than sample size, clinical variables can modulate the association between ID polymorphism and ISR. Plasma ACE level is a risk factor for ISR, independently of the ID genotype.