Background Short-term and long-term comparative outcomes after percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) for multivessel coronary artery (MVCA) or left main coronary artery (LMCA) disease are highly debated. Goals We performed a meta-analysis to evaluate the difference between PCI and CABG for the treatment of patients with MVCA or LMCA in long-term follow-up. Methods Literatures were searched in PubMed, EMBASE and The Cochrane Library from January 1, 2000 to January 1, 2021, including RCTs and observational studies (OSs). The primary outcome was all-cause mortality at 10 years follow-up, and the secondary outcomes included cardiac mortality, repeated revascularization, myocardial infarction, and stroke. Results A total of 5 RCTs reporting data from 3013 participants and 4 OSs of 5608 participants were included for analysis. There was no significant difference between PCI and CABG in all-cause mortality (Odds Ratio (OR) 1.03 [95% confidence interval (CI) 0.89 to 1.19]), whereas PCI was associated with higher cardiac mortality (OR 0.76 [95% CI 0.65 to 0.90]) and repeated revascularization rate comparing to CABG (OR 1.77 [95% CI 1.08 to 2.89]; I2 = 94.61%). The difference between PCI and CABG in repeated revascularization in either RCTs or OSs, in myocardial infarction in either RCTs or OSs were not significant. In OSs, stroke rate in PCI group was lower than those in CABG, but not in RCTs. There was a significant increase of stroke rate in CABG comparing to PCI (OR 0.65 [95% CI 0.53 to 0.80]; I2 = 0.00%). No significant difference between PCI and CABG in myocardial infarction was not observed (OR 0.92 [95% CI 0.64 to 1.31]; I2 = 57.84%). Conclusion Evidence from our study and prior studies suggested the superiority of CABG over PCI in improving 5- but not 10-year survival among patients with MVCA. In the contrast, there was no significant difference between CABG and PCI for treating patients with LMCA in either 5- or 10-year survival rate. More long-term trials are needed to better define differences of outcome between 2 techniques.