This study was carried out to stratify the risk of stent thrombosis by using three predictors: stent size, poststenting residual dissection, and residual filling defect. In the multicenter clinical trial, 1318 patients had successful deployment of Gianturco-Roubin coronary stent for threatened and acute closure. The 714 (54.2%) patients having none of these risk factors were designated a low-risk group; 484 (36.6%) had one factor and were designated an intermediate-risk group; 120 (9.1%) had two or all three factors and were designated a high-risk group. The incidence of stent thrombosis was 5.6%, 9.4%, and 16.7% in the low-, intermediate-, and high-risk groups; the difference among the three groups was highly significant (p < 0.0001). With these three predictors, the risk of stent thrombosis can be stratified. Avoiding the use of small stents (< 3.0 mm) and achieving optimal angiographic results after stenting for acute or threatened closure are useful strategies in reducing stent thrombosis.