Epicardial Versus Septal Collateral Channels in Retrograde Chronic Total Occlusion Percutaneous Coronary Intervention: Procedural Safety and Long-Term Outcomes

Background: Septal collateral channels (CCs) are generally preferred for retrograde chronic total occlusion (CTO) percutaneous coronary intervention (PCI) because epicardial CC injury may result in hemopericardium and cardiac tamponade. However, comparative evidence on the procedural safety and long-term clinical outcomes of epicardial versus septal CC use remains limited. Methods: We retrospectively analyzed 750 patients who underwent elective retrograde CTO PCI from January 2011 through April 2023. The primary procedural safety outcome was periprocedural pericardial effusion. The primary long-term outcome was major adverse cardiovascular events (MACEs), defined as all-cause death, nonfatal myocardial infarction, stroke, or unplanned revascularization. Multivariable logistic and Cox regression models were used to evaluate these outcomes. Results: 544 patients underwent retrograde CTO PCI through a septal CC and 206 through an epicardial CC. Pericardial effusion (7.3% vs. 1.7%; p < 0.001) and pericardiocentesis (5.3% vs. 1.1%; p = 0.001) were more frequent with epicardial CC use. CC wire crossing success (90.8% vs. 94.5%; p = 0.067), retrograde technical success (87.9% vs. 90.1%; p = 0.40), and overall recanalization success (90.3% vs. 91.0%; p = 0.80) did not differ significantly. Epicardial CC use was associated with higher odds of pericardial effusion (OR 5.60, 95% CI 2.26–14.70, p < 0.001). During a median follow-up of 1039 days, MACE occurred in 15.0% and 16.5% of the epicardial and septal CC groups, respectively (HR 1.10, 95% CI 0.71–1.70, p = 0.659). Conclusions: Epicardial CC use was associated with more periprocedural pericardial effusion than septal CC use. No statistically significant differences in technical success or long-term MACE were detected; these findings do not establish equivalence and are subject to anatomical selection and residual confounding.