OBJECTIVES: To evaluate utility of the complex percutaneous coronary intervention (PCI) criteria in real-world practice. BACKGROUND: Applicability of procedural complexity criteria for risk stratification has not been adequately evaluated in real-world practice. METHODS: Among 13,087 patients undergoing first PCI in the CREDO-Kyoto registry cohort-2, the study population consisted of 7,871 patients after excluding patients with acute myocardial infarction and those without stent implantation. Complex PCI was defined as PCI, which fulfills at least one of the followings: three vessels treated, > = 3 stents implanted, > = 3 lesions treated, bifurcation with two stents, >60 mm total stent lengths, and target of chronic total occlusion. RESULTS: The cumulative incidences of and adjusted risks for the primary ischemic (myocardial infarction/ischemic stroke), and bleeding (GUSTO moderate/severe) endpoints were significantly higher in patients with complex PCI (N = 2,777 [35%]) than in those with noncomplex PCI (N = 5,094 [65%]) (15.4% vs. 10.9%, log-rank p < .001; odds ratio (OR): 1.53, 95% confidence interval (CI): 1.31-1.79, p < .001, and 11.9% vs. 9.9%, log-rank p = .004; OR: 1.24, 95% CI: 1.05-1.46, p = .01). In the 30-day landmark analysis, the higher risks of patients with complex PCI for ischemic and major bleeding events were only seen within 30 days after PCI (ischemic; within 30 days: HR: 2.19, 95% CI: 1.79-2.69, p < .001; beyond 30 days: HR: 1.11, 95% CI: 0.92-1.34, p = .26, and bleeding; within 30 days: HR: 1.56, 95% CI: 1.13-2.16, p = .007; beyond 30 days: HR: 1.11, 95% CI: 0.94-1.31, p = .22). CONCLUSIONS: Patients with complex PCI as compared with patients with noncomplex PCI had a higher risk for both ischemic and bleeding events mainly within 30 days after PCI.