External validation of the PROGRESS-CTO perforation risk score: Individual patient data pooled analysis of threeregistries.

Abstract

Coronary artery perforation is one of the most feared and common complications of chronic total occlusion (CTO) percutaneous coronary intervention (PCI). To assess the usefulness of the recently developed PROGRESS-CTO (NCT02061436) perforation risk score in independent cohorts. Individual patient-level data pooled analysis of three registries was performed. Of the 4566 patients who underwent CTO PCI at 25 centers, 196 (4.2%) had coronary artery perforation. Patients with perforations were older (69 ± 10 vs. 65 ± 10, p < 0.001), more likely to be women (19% vs. 13%, p = 0.009), more likely to have a history of prior coronary artery bypass graft (34% vs. 20%, p < 0.001), and unfavorable angiographic characteristics such as blunt stump (62% vs. 48%, p < 0.001), proximal cap ambiguity (52% vs. 34%, p < 0.001), and moderate-severe calcification (60% vs. 49%, p = 0.002). Technical success was lower in patients with perforations (73% vs. 88%, p < 0.001). The area under the receiver operating characteristic curve of the PROGRESS-CTO perforation risk model was 0.76 (95% confidence interval [CI], 0.72-0.79), with good calibration (Hosmer-Lemeshow p = 0.97). We found that the CTO PCI perforation risk increased with higher PROGRESS-CTO perforation scores: 0.3% (score 0), 2.3% (score 1), 3.1% (score 2), 5.5% (score 3), 7.5% (score 4), 14.6% (score 5). Given the good discriminative performance, calibration, and the ease of calculation, the PROGRESS-CTO perforation score may facilitate assessment of the risk of perforation in patients undergoing CTO PCI.