Objectives The aim of the study was to evaluate the outcomes of retrograde versus antegrade approach in chronic total occlusion (CTO) percutaneous coronary intervention (PCI). Background The retrograde approach has increased the success rate of CTO PCI but has been associated with a higher risk for complications. Methods We conducted a meta‐analysis of studies published between 2000 and August 2019 comparing the in‐hospital and long‐term outcomes with retrograde versus antegrade CTO PCI. Results Twelve observational studies (10,240 patients) met our inclusion criteria (retrograde approach 2,789 patients, antegrade approach 7,451 patients). Lesions treated with the retrograde approach had higher J‐CTO score (2.8 vs. 1.9, p < .001). Retrograde CTO PCI was associated with a lower success rate (80.9% vs. 87.4%, p < .001). Both approaches had similar in‐hospital mortality, urgent revascularization, and cerebrovascular events. Retrograde CTO PCI was associated with higher risk of in‐hospital myocardial infarction (MI; odds ratio OR 2.37, 95% confidence intervals CI 1.7, 3.32, p < .001), urgent pericardiocentesis (OR 2.53, 95% CI 1.41–4.51, p = .002), and contrast‐induced nephropathy (OR 2.12, 95% CI 1.47–3.08; p < .001). During a mean follow‐up of 48 ± 31 months retrograde crossing had similar mortality (OR 1.79, 95% CI 0.84–3.81, p = .13), but a higher incidence of MI (OR 2.07, 95% CI 1.1–3.88, p = .02), target vessel revascularization (OR 1.92, 95% CI 1.49–2.46, p < .001), and target lesion revascularization (OR 2.08, 95% CI 1.33–3.28, p = .001). Conclusions Compared with antegrade CTO PCI, retrograde CTO PCI is performed in more complex lesions and is associated with a higher risk for acute and long‐term adverse events.