TCT-1059 Prediction of Patient Radiation Dose During Chronic Total Occlusion Percutaneous Coronary Intervention: A Machine Learning Augmented Analysis
Recommended Citation
Mutlu D, Strepkos D, Carvalho P, Alexandrou M, Ser OS, Kladou E, Williford N, Alaswad K, Davies R, Azzalini L, Goktekin O, Jaffer F, Frizzell J, Uluganyan M, Choi J, Young L, Raj L, Ybarra L, Rangan B, Mastrodemos O, Sandoval Y, Burke M, Brilakis E. TCT-1059 Prediction of Patient Radiation Dose During Chronic Total Occlusion Percutaneous Coronary Intervention: A Machine Learning Augmented Analysis. J Am Coll Cardiol 2025; 86(17):B452.
Document Type
Conference Proceeding
Publication Date
10-28-2025
Publication Title
J Am Coll Cardiol
Abstract
Background: There is limited information regarding the patient radiation dose during chronic total occlusion (CTO) percutaneous coronary artery intervention (PCI). Methods: We evaluated the association of clinical and angiographic characteristics with patient air kerma (AK) radiation dose in 10,806 patients underwent 10,816 CTO PCIs at 51 centers between 2012 and 2025 from an international registry (PROGRESS-CTO). Feature selection was performed with the Random Forest classification algorithm. A generalized linear mixed model (GLMM) with a gamma distribution and random intercept per study center was utilized for the prediction model. Results: The median AK radiation dose was 2.0 (1.1-3.4) Gray. 11% of patients received >5 Gy of AK radiation (high dose). Patients in the high dose group were more likely to be men, higher body mass index (BMI), diabetes, and more complex lesions (J-CTO score: 2.8±1.2 vs. 2.4±1.2, p<0.001). Technical and procedural success were lower and major adverse cardiovascular events were higher in the high dose group. On multivariable GLMM analysis, age, male gender, BMI, diabetes, heart failure, prior CABG, serum creatinine, contrast volume, retrograde crossing strategy, complex lesion characteristics were independently associated with higher radiation dose, whereas interventional collateral presence and intravascular imaging use were associated with lower radiation dose. Substantial center variation was observed, and AK radiation dose has plateaued in recent years. [Formula presented] Conclusion: Several patient and angiographic characteristics were independently associated with AK radiation dose during CTO PCI. Categories: CORONARY: Complex and Higher Risk Procedures for Indicated Patients (CHIP)
Volume
86
Issue
17
First Page
B452
