Periprocedural quantitative coronary angiography after Palmaz-Schatz stentimplantation predicts the restenosis rate at six months - Results of a meta-analysis of the Belgian Netherlands Stent Study (BENESTENT) I, BENESTENT II pilot, BENESTENT II and MUSIC trials

OBJECTIVES We aimed to identify periprocedural quantitative coronary angiog raphic (QCA) variables that have predictive value on long term angiographic results and to construct multivariate models using these variables for pos tprocedural prognosis. BACKGROUND Coronary stent implantation has reduced the restenosis rate sign ificantly as compared with balloon angioplasty in short de novo lesions in coronary arteries >3 mm in size. Although the postprocedural minimal lumina l diameter (MLD) is known to have significant bearing on long-term angiogra phic results, no practically useful model exists for prediction of angiogra phic outcome based on the periprocedural QCA variables. METHODS The QCA data from patients who underwent Palmaz-Schatz stent implan tation for short (<15 mm) de novo lesions in coronary arteries >3 mm and co mpleted six months of angiographic follow-up in the four prospective clinic al trials (BENESTENT I, BENESTENT II pilot, BENESTENT II and MUSIC) were po oled. Multiple models were constructed using multivariate analysis. The Hos mer-Lemeshow goodness-of-fit test was used to identify the model of best fi t, and this model was used to construct a reference chart for prediction of angiographic outcome on the basis of periprocedural QCA variables. RESULTS Univariate analysis performed using QCA variables revealed that ves sel size, MLD before and after the procedure, reference area before and aft er the procedure, minimal luminal cross-sectional area before and after the procedure, diameter stenosis after the procedure, area of plaque after the procedure and area stenosis after the procedure were significant predictor s of angiographic outcome. Using multivariate analysis, the Hosmer-Lemeshow goodness-of-fit test showed that the model containing percent diameter ste nosis after the procedure and vessel size best fit the data. A reference ch art was then developed to calculate the expected restenosis rate. CONCLUSIONS Restenosis rate after stent implantation for short lesions can be predicted using the variables percent diameter stenosis after the proced ure and vessel size. This meta-analysis indicates that the concept of "the bigger the better" holds true for coronary stent implantation. Applicabilit y of the model beyond short lesions should be tested. (J Am Coll Cardiol 19 99;34:1067-74) (C) 1999 by the American College of Cardiology.