BACKGROUND. Late angiographic narrowing has been observed following coronary implantation of the Wallstent. To identify the angiographic variables that predict restenosis within the stented segment, a retrospective study of data from the European Wallstent core laboratory was performed. METHODS AND RESULTS. Follow-up angiograms (excluding patients with in-hospital occlusions) were analyzed for 214 lesions in 176 patients (78% restudy rate). The incidence of restenosis within the stented segment was 35% by lesion and 35% by patient for criterion 1 (greater than or equal to 0.72 mm loss in minimal luminal diameter) and 24% by lesion and 24% by patient for criterion 2 (diameter stenosis greater than or equal to 50% at follow-up). The association between 16 variables and restenosis was determined by a relative risk ratio assessment. Variables with significant risk ratios for restenosis with criterion 1 were use of multiple stents/lesion (relative risk, 1.56; 95% confidence interval [CI], 1.08-2.25) and oversized (unconstrained stent diameter exceeding reference diameter greater than 0.7 mm) stents (relative risk, 1.64; 95% CI, 1.10-2.45), and for criterion 2, oversizing by more than 0.70 mm (relative risk, 1.93; 95% CI, 1.13-3.31), bypass grafts (relative risk, 1.62; 95% CI, 0.98-2.66), use of multiple stents/lesion (relative risk, 1.61; 95% CI, 0.97-2.67) and residual diameter stenosis more than 20% post stenting (relative risk, 1.51; 95% CI, 0.91-2.50). CONCLUSIONS. It is concluded that several angiographic variables are significantly associated with late angiographic narrowing after stenting in the coronary arteries. We suggest that stent operators avoid excessive oversizing in the selection of stent diameter and the use of multiple stents per lesion to lessen the risk of late restenosis.