Procedural Predictors of Angiographic Restenosis After Bifurcation Coronary Stenting (from the Choice of Optimal Strategy for Bifurcation Lesions With Normal Side Branch and Optimal Stenting Strategy for True Bifurcation Lesions Studies)

Most subordinate techniques accompanying bifurcation stenting have not been supported by relevant clinical trials. The aim of this study was to investigate the impact of technical specifications involved in bifurcation stenting on angiographic outcomes. We analyzed patients enrolled in a cohort consisting of the patients in 2 randomized studies: one comparing routine final kissing inflation (FKI) versus leave-alone strategy after the 1-stent technique for bifurcations without side branch (SB) stenosis (<50%) and the other comparing crush versus the 1-stent technique for lesions with SB stenosis (≥50%). The effects of subordinate techniques and devices on 8-month angiographic restenosis were studied using multivariate models. Of 514 patients whose 8-month angiograms were available, 58 (11.3%) were found to have restenosis involving, in total, 35 main branches (MBs) and 27 SBs. Using multivariate models, we identified FKI as the only independent predictor of MB restenosis in the technically modifiable variables. The effect of FKI was significantly different across subgroups defined by bifurcation lesion type and stenting technique (test for homogeneity, p = 0.003): FKI was unrelated to MB restenosis in the 1-stent for diseased SB (odds ratio [OR] 0.41, 95% confidence interval [CI] 0.10 to 1.72; p = 0.22) and the 2-stent groups (OR 0.14, 95% CI 0.01 to 1.36; p = 0.09) but predictive of MB restenosis in the 1-stent for normal SB group (OR 4.90, 95% CI 1.58 to 15.16; p = 0.006).