Background: In percutaneous coronary intervention (PCI) for chronic total occlusion (CTO), most experts regard the antegrade approach as the default initial strategy, reserving the retrograde approach for reattempts following antegrade failure. In this study, we aimed to compare the efficacy and safety between the antegrade and retrograde approaches in CTO PCI. Results: Between 2012 and 2013, patients that underwent 321 consecutive attempts by high-volume operators (> 75 total CTO PCI cases during the period) in a tertiary university-affiliated hospital were enrolled. The antegrade approach was used in 152 patients, and retrograde in 169 patients. The duration of occlusion was significantly longer and the J-CTO score higher in the retrograde group. Technical success was achieved in 148 patients of the antegrade group (97.4%), and 163 patients in the retrograde group (96.4%) (p = 0.75). A major procedural complication occurred in 3 patients of the antegrade group (2.0%) and in 6 patients of the retrograde group (3.6%) (p = 0.51). In-hospital major adverse cardiac events (MACE) rates (antegrade 0.7%, n = 152; retrograde 0.6%, n = 169)were comparable. The procedure and fluoroscopy times were significantly longer, with more radiation exposure and contrast medium consumption, in the retrograde group. In the retrograde group, similar success, procedural complication and in-hospitalMACE rates were achieved in the 3 collateral subgroups. Conclusions: In selected cases and with highly experienced operators, retrograde approach in CTO PCI is as effective and safe as antegrade approach at the expense of longer procedure time, more radiation exposure and contrast medium consumption. For retrograde approach, either septal, epicardial or AV groove collaterals can be used with similarly success, complication and in-hospital MACE rates.