Residual risk according to lesion complexity phenotypes after intravascular imaging-guided percutaneous coronary intervention

Background Intravascular imaging (IVI)-guided percutaneous coronary intervention (PCI) is associated with improved outcomes in complex coronary disease. However, whether IVI guidance uniformly reduces excess risk across distinct lesion complexity phenotypes remains uncertain. Purpose To evaluate residual risk across predefined lesion complexity phenotypes following real-world IVI-guided PCI and to assess the impact of post-procedural stent optimization. Methods We analyzed a prospective single-center registry of consecutive IVI-guided (intravascular ultrasound or optical coherence tomography) PCI procedures performed between January 2015 and March 2025. Eight predefined phenotypes were evaluated: left main, bifurcation, long lesion, severe calcification, in-stent restenosis, chronic total occlusion, ostial lesion, and multiple stents. The reference group comprised non-complex PCI (no phenotype flags). The primary endpoint was 1-year major adverse cardiac events (MACE: all-cause death, myocardial infarction, or target vessel revascularization). Adjusted hazard ratios (HR) were derived using multivariable Cox proportional-hazards models. A secondary analysis evaluated patients achieving an optimal post-procedural minimum stent area (MSA ≥5.5 mm² for non-left main lesions and ≥8.0 mm² for left main lesions). Results Among 3,664 patients, 3339 (91.1%) had ≥1 complex phenotype. At 1 year, MACE occurred in 380 patients (13.6%, 95% CI 12.4-14.9). After multivariable adjustment, in-stent restenosis (HR 2.66, 95% CI 1.46-4.84, p=0.001), chronic total occlusion (HR 2.45, 95% CI 1.31-4.56, p=0.005), multiple stents (HR 2.23, 95% CI 1.29-3.87, p=0.004), long lesion (HR 1.81, 95% CI 1.11-2.95, p=0.017), and bifurcation (HR 1.68, 95% CI 1.03-2.72, p=0.037) remained independently associated with higher MACE compared with non-complex PCI. In contrast, left main, severe calcification, and ostial lesions were not independently associated with excess risk after adjustment. Achievement of optimal post-procedural MSA was associated with lower 1-year MACE; however, high-risk phenotypes remained predictive despite achievement of optimal expansion. Conclusions In real-world IVI-guided PCI, lesion complexity confers heterogeneous residual risk. In-stent restenosis, chronic total occlusion, multiple stents, long lesion, and bifurcation PCI remain independently high-risk despite imaging guidance. These findings suggest that IVI use does not fully attenuate risk across complexity phenotypes and support further refinement of structured, phenotype-specific optimization strategies in complex PCI.