A Novel Cardiac Computed Tomography‐Based Calcium Scoring System for Predicting the Use of Atherectomy Devices in Severely Calcified Coronary Lesions

Background Cardiac computed tomography (CT) is a well‐established process used to diagnose coronary artery disease; however, its specific advantages in predicting the use of atherectomy devices during percutaneous coronary intervention (PCI) for moderate to severe calcified lesions remain to be determined. This study aimed to develop a risk scoring system for predicting the use of atherectomy devices in PCI on the basis of morphological findings obtained by preoperative cardiac CT. Methods In this retrospective, multicenter, observational study, we screened patients who underwent cardiac CT 6 months before PCI for the target lesion. Consequently, 498 de novo stenotic lesions from 415 patients with moderate to severe calcification were included. Then, 70% of the cohort (derivation cohort, 349 lesions) was randomly selected and used for developing the risk model of predicting the use of an atherectomy device; the remaining 30% (validation cohort, 149 lesions) was used to evaluate the discriminative ability and calibration of the model. Results Multivariate analysis in the derivation cohort identified four factors as predictors of the use of atherectomy devices: circumferential angle ≥ 270°, bifurcation, calcified mass, and longitudinal distance ≥ 9 mm. The discriminative ability and calibration of the risk model that was developed using the four factors was excellent in the validation cohort (c‐statistic = 0.87, Hosmer–Lemeshow goodness‐of‐fit test, p = 0.53). Conclusions The developed model accurately predicted the use of atherectomy devices in patients with moderate to severe coronary calcification. Estimating atherectomy usage by an easily usable objective tool on the basis of the preoperative cardiac CT findings has the potential for widespread implementation in clinical settings.