Validation of the coronary artery specific chronic total occlusion percutaneous coronary intervention angiographic difficulty scores in the PROGRESS-CTO registry

Background: The J-CTO investigators recently developed angiographic difficulty scores for each of the three major coronary arteries in patients undergoing first-attempt chronic total occlusion (CTO) percutaneous coronary intervention (PCI) in de novo occlusions. Methods: We examined the performance of the individual J-CTO scores in a large multicenter registry. Results: The CTO lesion location was as follows: right coronary artery (RCA) 3,805 (54%), left anterior descending artery (LAD) 2,303 (33%), and left circumflex (LCX) 935 (13%). Patients in the PROGRESS-CTO registry were younger, more likely to be female, and had higher J-CTO scores compared with the J-CTO registry. Increasing difficulty scores were associated with lower technical success in the PROGRESS-CTO registry (score 0: 94.4 % - score ≥3: 82.6% for the RCA difficulty score; score 0: 96.4% - score ≥3: 86.1 for the LAD difficulty score; and score 0: 95.4% - score ≥3: 81.2% for the LCX difficulty score). The C-statistic of the coronary artery specific J-CTO scores in the PROGRESS-CTO registry were: LAD 0.69 (95% confidence intervals [CI], 0.64-0.73), LCX 0.63 (95% CI, 0.57-0.69), and RCA 0.61 (95-% CI, 0.58-0.64) with good calibration (Hosmer−Lemeshow p-value >0.05 for all). The AUC of the classic J-CTO score for LAD lesions was similar with the LAD J-CTO score (p-for-difference = 0.26), but worse for LCX (p-for-difference = 0.04) and RCA lesions (p-for-difference = 0.04). Conclusion: In the PROGRESS-CTO registry, the coronary artery specific J-CTO scores did not improve prediction of the technical success of CTO-PCI compared with the classic J-CTO score.