The impact of lesion complexity on long-term vascular response to cobalt–chromium everolimus-eluting stent (CoCrEES) remains unclear. We sought to evaluate them using optical coherence tomography (OCT). A total of 47 patients with 58 lesions treated only with CoCrEES and no target-vessel events within 5 years after implantation were prospectively enrolled and underwent 5-year follow-up OCT. Quantitative parameters and qualitative characteristics of the neointima were evaluated using multilevel logistic or linear regression models with random effects at three levels: lesion, cross-section (CS), and strut. According to the lesion complexity, the lesions were classified into the two groups: the complex lesion (CL) and non-CL group. The CL was defined as having at least 1 high-risk feature such as acute coronary syndrome lesion, lesion length > 20 mm, severe calcification requiring rotational atherectomy, and chronic total occlusion at the index procedure. A total of 11,034 struts (CL, n = 6240; non-CL, n = 4794) and 1202 (CL, n = 683; non-CL, n = 519) CSs were analyzed. The percentage of uncovered and malapposed struts did not differ significantly between the CL and non-CL groups (0.90 vs. 0.54%, P = 0.78; 0.56 vs. 0.10%, P = 0.16, respectively). The incidence of neoatherosclerosis was comparable between both groups in the CS- and lesion-level analysis (3.5 vs. 4.6%, P = 0.91; 32.0 vs. 24.2%, P = 0.52, respectively). At 5 years, CoCrEES shows an excellent vascular healing and similar frequency of neoatheroslerosis in patients without target-vessel events, regardless of the lesion complexity.