Establishment and evaluation of a nomogram prediction model for risk of atrial fibrillation recurrence after the cox-maze IV procedure.

Abstract

The Cox-Maze IV (CMIV) procedure is the mainstay in surgical treatment of atrial fibrillation (AF), but the rate of AF recurrence after the CMIV procedure in patients with persistent AF is difficult to accurately evaluate. In this study, we aimed to develop and validate a risk prediction model of AF recurrence within 1 year after undergoing the Cox-Maze IV procedure. We retrospectively enrolled 303 consecutive patients who underwent the Cox-Maze IV procedure for persistent AF concomitant with other cardiac procedures at our institute between 2019 and 2021. A nomogram was developed using multivariate logistic regression analysis, and the concordance statistic (C-statistic) was computed. Differentiation, calibration, clinical suitability, and bootstrapping were performed to verify the model. Among the 303 patients, 71 developed recurrent AF within 1 year of CMIV. Factors predictive of postoperative AF recurrence included age, left ventricular hypertrophy (LVH), early atrial tachyarrhythmias (ATAs), and congenital heart disease surgery (namely, ventricular septal defect repair and atrial septal defect repair). Based on the training dataset, the nomogram had a C-statistic of 0.864 (95% CI 0.811-0.918) for predicting AF recurrence. According to the receiver operating characteristic curve, (ROC curve), the cutoff value of the model was 0.293, and the specificity and sensitivity were 0.841 and 0.789, respectively. This model can predict the risk of AF recurrence after the CMIV procedure. Its discrimination, calibration, and clinical applicability are strong, and its clinical application is simple and easy to promote.