Background: Guidelines recommend that Percutaneous coronary intervention (PCI) should be performed in hospitals with onsite cardiac surgery (CS). However, emerging data suggest that there is no significant difference in clinical outcomes following primary, or elective PCI between the two groups. We performed a meta-analysis and meta-regression to assess the safety and efficacy of performing PCI in centers with, and without on-site CS. Methods: We conducted electronic database searches in PubMed, CENTRAL, EMBASE. The Cochrane Register, Google Scholar databases, and the scientific session abstracts were searched for eligible studies. Risk ratios and 95% confidence intervals were computed with the Mantel-Haenszel method. Fixed-effect models were used; if heterogeneity (I 2 )>25 was identified, effects were obtained with random models. Meta-regression analyses were performed to determine whether the effects of PCI without on-site CS were modulated by pre-specified study-level factors Results: Twenty-seven studies were included with total n=8,558,618 patients. No significant difference was observed for all-cause mortality (RR 1.02, 95% CI 0.86-1.21, p=0.82, I 2 =97.2%), cardiovascular mortality rates (RR 1.18, 95% CI 0.93-1.50, p=0.17, I 2 2.98%), myocardial infarction rates (RR 0.89, 95% CI 0.62-1.29, p=0.55, I2= 88.5%), repeat revascularization (RR 0.87, 95% CI 0.43-1.76, p=0.69, I 2 =98.8%), stroke (RR 1.28, 95% CI 0.56-2.91, p=0.55, I 2 98.8%), shock (RR 0.76, 95% CI 0.43-1.35, p=0.35, I 2 = 93.7%), mechanical circulatory support (RR 0.83, 95% CI 0.46-1.50, p=0.53, I 2 99.8%), bleeding (RR 0.88, 95% CI 0.43-1.81, p=0.73, I 2 =99.6%), and emergency CABG (RR 0.97, 95% CI 0.64-146, p= 0.87, I 2 =84.1%). In a meta-regression analysis, the effect of PCI without on-site CS, baseline clinical features did not affect the long-term all-cause mortality outcome. Conclusion: There was no significant difference in complications rates, and clinical outcomes for PCI performed at centers without on-site CS compared to centers with on-site CS.