Development and validation of a risk prediction model for new-onset ventricular arrhythmias after coronary artery bypass grafting: A retrospective observational study

Background To examine the risk factors for in-hospital ventricular arrhythmias (VA) in coronary artery disease (CAD) patients after coronary artery bypass grafting (CABG) and develop a nomogram to predict the risk of VA occurrence. Methods This retrospective study involved the analysis of the clinical data (using the MIMIC-IV database) of 5,267 patients who underwent CABG. The risk factors for in-hospital VA were identified using the least absolute shrinkage and selection parameter (LASSO) and multivariate logistic regression analyses. Based on the outcomes, a risk prediction model was then developed. Results The nomogram was constructed using eight predictive indicators: congestive heart failure (CHF), atrial fibrillation (AFib), base excess, systolic blood pressure (SBP), white blood cell (WBC) count, use of milrinone or dobutamine, continuous renal replacement therapy (CRRT), and logistic organ dysfunction system(LODS) score. According to the internal validation results, the model demonstrated a good predictive ability, with area under the curve (AUC) values of 0.777 and 0.743 in the training and validation sets, respectively. Furthermore, the calibration curve revealed that the model’s predicted values were in good agreement with the actual observed values. Moreover, the clinical decision curve analysis (DCA) showed that the model had a significant clinical net benefit at diagnostic thresholds of 0.1–0.95 and 0.1–0.8 in the training and validation sets, respectively. Conclusion Herein, we developed a risk prediction model for VA occurrence. The model demonstrated good discrimination, calibration, and clinical applicability, ensuring early VA prediction, which could facilitate risk stratification, enhance patient monitoring and management post-CABG, and reduce VA incidence in high-risk patients.