Risk factors and nomogram model construction for postoperative recurrence in patients with non-valvular atrial fibrillation underwent one-stop surgery guided by intracardiac echocardiography

Li Jingchang; Qiao Chong; Zhou Qian; Liu Lin; Chai Zhiyong; Han Keli

doi:10.3969/j.issn.1674-4055.2025.08.16

Abstract：Objective To analyze the risk factors for postoperative recurrence in patients with non-valvular atrial fibrillation(NVAF)underwent one-stop surgery guided by intracardiac echocardiography(ICE),and to establish a nomogram model.Methods NVAF patients(n=170)underwent one-stop surgery guided by ICE were selected and their basic data was collected by using information questionnaire in Nanyang Central Hospital from Jan.2023 to Feb.2024,and all patients were followed up for 12 months after the surgery.There were 14 cases missed during follow-up period,and finally 156 patients were included the study.According to results of surface electrocardiogram or ambulatory electrocardiogram for determining NVAF recurrence status,the patients were divided into recurrence group(n=62)and non-recurrence group(n=94).All data was used to establish a model,which was internally validated by the Bootstrap method.The factors influencing NVAF recurrence were analyzed by using Cox regression model with univariate analysis first followed by multivariate analysis.A nomogram model was established based on significant predictors identified in multivariate analysis.The calibration curve and ROC curve were used to evaluate the calibration and discrimination of the model,respectively.Results The results of multivariate Cox regression analysis showed that disease duration＞3 y,complicated by hypertension,increased preoperative platelet derived growth factor(PDGF),indoxyl sulfate(IS),scores of managed ventricular pacing(MVP)and brain natriuretic peptide(BNP),and decreased preoperative left atrial appendage velocity(LAAV)were independent risk factors for NVAF recurrence(P＜0.05).From the calibration curve,the prediction model showed a good degree of fit with the validation set.Based on Hosmer-Lemeshow goodness-of-fit test,the calibration of the nomogram model in training set(χ2=0.723,P=0.698)was considered acceptable and of high quality.The AUC of training set composed of all data was 0.834(95%CI:0.826～0.849).Additionally,Bootstrap resampling with 1000 iterations was used for internal validation,and the calculated C-index was 0.807.The C-index was approximately equivalent to the AUC of ROC curve.These results indicate that the nomogram model had good discriminative ability and favorable predictive performance.Conclusion The recurrence risk is higher in NVAF patients with above risk factors after one-stop surgery guided by ICE,and the nomogram model established based on these risk factors has a good predictive performance.