Development and Validation of Nomograms for predicting Coronary Artery Calcification and Severe Coronary Artery Calcification: a retrospective cross-sectional study

IntroductionThere is a significant lack of effective pharmaceutical interventions for treating coronary artery calcification (CAC). Severe CAC (sCAC) poses a formidable challenge to interventional surgery and exhibits robust associations with adverse cardiovascular outcomes. Therefore, it is imperative to develop tools capable of early-stage detection and risk assessment for both CAC and sCAC. This study aims to develop and validate nomograms for the accurate prediction of CAC and sCAC. MethodsThis retrospective cross-sectional study was conducted in Taizhou, Jiangsu Province, China. CAC assessment was performed using non-gated thoracic CT scans. Demographic data and clinical information were collected from patients who were then randomly divided into a training set (70%) or a validation set (30%). Least absolute shrinkage and selection operator (LASSO) regression as well as multiple logistic regression analyses were utilized to identify predictive factors for both CAC and sCAC development. Nomograms were developed to predict the occurrence of CAC or sCAC events. The models performance was evaluated through discrimination analysis, calibration analysis, as well as assessment of their clinical utility. ResultsThis study included 666 patients with an average age of 75 years, of whom 56% were male. 391 patients had CAC, with sCAC in 134 cases. Through LASSO and multiple logistic regression analysis, age increase, hypertension, carotid artery calcification, CHD, and CHADS2 score were identified for the CAC risk predictive nomogram with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.845(95%CI 0.809-0.881) in the training set and 0.810(95%CI 0.751-0.870) in the validation set. Serum calcium level, carotid artery calcification, and CHD were identified for the sCAC risk predictive nomogram with an AUC of 0.863(95%CI 0.825-0.901) in the training set and 0.817(95%CI 0.744-0.890) in the validation set. Calibration plots indicated that two models exhibited good calibration ability. According to the decision curve analysis (DCA) results, both models have demonstrated a positive net benefit within a wide range of risks. ConclusionsThe present study has successfully developed and validated two nomograms to accurately predict CAC and sCAC, both of which have demonstrated robust predictive capabilities.