P ercutaneous coronary intervention (PCI) is a common and established therapy for coronary artery disease. Drugeluting stents have been shown to reduce neointimal hyperplasia and restenosis in stent segments. However, restenosis in reference segments adjacent to the proximal and distal border of the stent (so-called stent edge restenosis [SER]) remains a potential limitation of drug-eluting stents.
1Stenting from normal to normal is effective to minimize SER. However, the reference segments are rarely normal. Especially in diffuse disease, it is difficult to determine appropriate plaque-free reference segments for stent landing. Previous intravascular ultrasound studies have disclosed that greater residual plaque burden and stent overexpansion are associated with SER. 2,3 Optical coherence tomography (OCT) is a high-resolution (10-20 μm) intravascular imaging technique that uses near-infrared light to create images. An excellent contrast between lumen and vessel wall in OCT allows accurate lumen measurements. 4 Tissue characterization by OCT enables us to identify 3 types of coronary plaques such as lipid, fibrous, and fibrocalcific. 5 In addition, OCT has a high sensitivity for detection of suboptimal stent findings such as intrastent tissue protrusion, incomplete stent apposition, stent edge dissection, and intrastent thrombus. 4 Recently, OCT is becoming increasingly widespread as a clinical tool to guide PCI. The aim of this study was to determine OCT predictors for angiographic SER after everolimus-eluting stent implantation.Background-Stent edge restenosis (SER) remains a potential limitation of drug-eluting stents. The aim of this study was to determine optical coherence tomography (OCT) predictors for angiographic late SER after everolimus-eluting stent implantation. Methods and Results-We retrospectively analyzed 319 patients who underwent OCT immediately after everolimus-eluting stent implantation and scheduled 9-to 12-month follow-up angiography. The binary angiographic SER rate was 10% (32/319) in the patients, 8.4% (32/382) in lesions, and 4.4% (33/744) in stent edge segments. In the stent edge segments at post stenting, OCT-derived lipidic plaque (61% versus 20%; P<0.001) was more often observed in the SER group, and OCT-measured minimum lumen area (4.13±2.61 versus 5.58±2.46 mm 2 ; P=0.001) was significantly smaller in the SER group compared with the non-SER group. Multivariate analysis identified lipidic plaque (odds ratio: 5.99; 95% confidence interval: 2.89-12.81; P<0.001) and minimum lumen area (odds ratio: 0.64; 95% confidence interval: 0.42-0.96; P=0.029) as independent predictors of binary SER. Receiver-operating characteristic analysis demonstrated that lipid arc of 185° (sensitivity: 71%; specificity: 72%; area under the curve: 0.761) and minimum lumen area of 4.10 mm 2 (sensitivity: 67%; specificity: 77%; area under the curve: 0.787) were optimal cutoff values for predicting ischemiadriven SER. Conclusions-The present OCT study demonstrated that lipidic plaque and minimum lumen area in t...