Determinants of Radiation Exposure and Radiation Overexposure During Percutaneous Coronary Intervention. A Real-World Study of 10,634 Procedures With Development of a Pragmatic Radiation-Alert Risk Score
digne, f.; Darmon, A.; stratiev, v.; Abdellaoui, M.; Dreyfus, J.; taramasso, m.; Nappi, F.; Nejjari, M.
Show abstract
Background: Ionizing radiation is inherent to percutaneous coronary intervention (PCI) and may reach radiation-alert thresholds during complex procedures. Real-world data translating dose determinants into a usable planning-stage alert tool remain scarce. Aims: To determine the prevalence and predictors of radiation overexposure (RO) during PCI and to develop a pragmatic radiation-alert risk score usable before substantial dose accumulates. Methods: We analysed 10,634 consecutive PCI procedures performed between January 2008 and May 2018. Complex PCI was defined as unprotected left main, rotational atherectomy, chronic total occlusion (CTO), or bifurcation/trifurcation PCI. The composite RO endpoint was a dose-area product >500 Gy{middle dot}cm2, cumulative reference-point air kerma (AK) >5 Gy, or fluoroscopy time >60 minutes. A pragmatic score was built from early procedural variables. Results: Overall, 3,457 procedures (32.5%) were complex, and RO occurred in 99 (0.93%). Complex PCI showed higher RO rates than standard PCI (2.1% vs 0.4%; p<0.001). In multivariable analysis, body mass index (BMI), male sex, CTO, rotational atherectomy, bifurcation, and treatment of [≥]2 vessels independently predicted higher AK. A pragmatic score derived from early clinical and procedural variables showed good discrimination (area under the curve 0.772; 95% bootstrap CI 0.725-0.818) and was well calibrated (Brier score 0.0091; observed-to-expected ratio 0.994) for the composite RO endpoint. Conclusions: Radiation-alert thresholds were exceeded in approximately 1% of PCI procedures, but fivefold more often during complex PCI. A pragmatic, internally assessed score based on early procedural variables identified high-risk cases with good discrimination, supporting planning-stage risk stratification to activate dose-sparing strategies before substantial radiation has accumulated.
Matching journals
The top 11 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Longitudinal Changes in the Carotid Arteries of Head and Neck Cancer Patients Following Radiation Therapy: Results from a Prospective Serial Imaging Biomarker Characterization Study 92%
- Artificial Intelligence Uncertainty Quantification in Radiotherapy Applications - A Scoping Review 89%
- Serum miRNA-based signature indicates radiation exposure and dose in humans: a multicenter diagnostic biomarker study 87%
Similar papers in this journal
- Treatment of Slow-flow After Primary Percutaneous Coronary Intervention With Flow-mediated Hyperemia. The Randomized RAIN-FLOW Study 90%
- Patient Screening for Self-Expanding Percutaneous Pulmonary Valves using Virtual Reality 89%
- Artificial intelligence of arterial Doppler waveforms to predict major adverse outcomes among patients evaluated for peripheral artery disease 89%
Similar papers in this journal
Similar papers in this journal
- Ticagrelor vs Clopidogrel: the Impact of Platelet Inhibition on Cerebrovascular Microembolic Events during TAVR 91%
- Direct Oral Anticoagulants Affect Activated Clotting Time During and Bleeding Events After Percutaneous Coronary Intervention 90%
- Risk of Cardiovascular Events after Covid-19: a double-cohort study 88%
Similar papers in this journal
- Evaluation of the second-generation whole-heart motion correction algorithm (SSF2) used to demonstrate the aortic annulus on cardiac CT 91%
- Inconsistency of AI in Intracranial Aneurysm Detection with Varying Dose and Image Reconstruction 90%
- Deep Learning-based Prediction of Early Cerebrovascular Events after Transcatheter Aortic Valve Replacement 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.