【作者】 李娜；

【导师】 张冀东；

【作者基本信息】 河北医科大学 ， 内科学（专业学位）， 2023， 硕士

【摘要】 目的:本研究主要探讨阵发性心房颤动患者导管消融术后复发的危险因素,构建并验证预测术后复发的列线图模型。方法:纳入了2020年6月至2022年6月期间就诊于河北医科大学第二医院心内六科且因症状性阵发性心房颤动首次行射频消融治疗的所有患者共204例,中位随访18.7个月后,术后房颤复发的患者有42例。204例患者以7:3随机分成训练集(143例)与验证集(61例)两组,并对两组间临床基线资料(性别、年龄、病程、既往病史、心脏超声指标、实验室指标)进行比较。在验证集中,应用单因素及多因素COX回归方法分析阵发房颤患者导管消融术后复发的独立危险因素,采用R软件构建预测阵发房颤患者术后复发的列线图模型,并对列线图模型进行内外部验证,根据所有患者列线图分值高低分为高分组及低分组,绘制两组的累计复发概率曲线并用Log-rank检验进行比较。结果:1.训练集与验证集两组间性别、年龄、病程、既往病史、心脏超声指标、实验室指标以及复发情况比较均无统计学意义(P>0.05),两组患者具备同质性。2.在训练集中,经单因素与多因素COX回归分析得出年龄、合并高血压、左房内径、病程是阵发性心房颤动患者导管消融术后复发的独立危险因素(P<0.05)。3.基于年龄、高血压、左房内径、病程建立列线图模型,并评估列线图的有效性。通过1000次Bootstrap自抽样法对列线图进行验证,分别绘制训练集与验证集6个月、1年及2年的校准曲线,校准曲线提示列线图有较好的预测效能。同时计算出训练集一致性指数(C-index)为0.813(95%CI:0.750-0.877),验证集一致性指数(C-index)为0.805(95%CI:0.683-0.928);并绘制两组时间依赖性ROC曲线,其中训练集中预测导管消融术后6个月、1年、2年复发的ROC曲线下面积(AUC)分别为0.848、0.899、0.884,验证集预测导管消融术后6个月、1年、2年复发的R OC曲线下面积(AUC)分别为0.856、0.814、0.841。一致性指数与RO C曲线下面积均提示列线图模型具有良好的区分度及一致性。4.依据所有患者列线图分值分为高分组与低分组,绘制两组的累计复发概率曲线,两组间存在显著差异(Log-rank P<0.0001),高分组术后复发概率明显高于低分组。结论:1.年龄、病程、LAD、高血压是阵发性心房颤动患者导管消融术后复发的独立预测因子。2.以整合年龄、病程、LAD及高血压而构建的预测阵发性心房颤动患者导管消融术后复发的列线图模型具有较好的区分度和一致性。3.列线图得分高的患者随时间延长导管消融术后复发概率显著高于得分低者,提示该列线图模型具有良好的预测效能。更多还原

【Abstract】 Objective: This study mainly explored the risk factors of recurrence after catheter ablation in patients with paroxysmal atrial fibrillation,and constructed and verified a nomogram model for predicting recurrence after catheter ablation.Methods: A total of 204 patients admitted to the sixth Department of Cardiology,the Second Hospital of Hebei Medical University from June 2020 to June 2022 who underwent the first catheter ablation for symptomatic paroxysmal atrial fibrillation were included.After a median follow-up of 18.7months,42 patients had postoperative atrial fibrillation recurrence.The 204 patients were randomly divided into training set(143 cases)and validation set(61 cases)at 7 : 3,and the clinical baseline data(gender,age,course of disease,past medical history,echocardiography index,laboratory index)were compared between the two groups.In the validation set,univariate and multivariate COX regression methods were used to analyze the independent risk factors for recurrence after catheter ablation in patients with paroxysmal atrial fibrillation.R software was used to construct a nomogram model for predicting postoperative recurrence in patients with paroxysmal atrial fibrillation,and the nomogram model was verified internally and externally.According to the nomogram scores of all patients,they were divided into high score group and low score group.The cumulative recurrence probability curves of the two groups were drawn and compared by Log-rank test.Results:1.There was no significant difference in gender,age,course of disease,past medical history,cardiac ultrasound index,laboratory index and recurrence between the training set and the validation set(P>0.05).The two groups of patients have homogeneity.2.In the training set,univariate and multivariate COX regression analysis showed that age,hypertension,left atrial diameter,and course of disease were independent risk factors for recurrence after catheter ablation in patients with paroxysmal atrial fibrillation(P<0.05).3.A nomogram model was established based on age,hypertension,left atrial diameter,and course of disease,and the effectiveness of the nomogram was evaluated.The nomogram was verified by 1000 Bootstrap self-sampling methods,and the calibration curves of the training set and the validation set for6 months,1 year and 2 years were drawn respectively,suggesting that the nomogram has good predictive performance.The nomogram was verified by1000 Bootstrap self-sampling methods,and the calibration curves of the training set and the validation set for 6 months,1 year and 2 years were drawn respectively,suggesting that the nomogram has good predictive performance.The C-index of training set was 0.813(95 % CI 0.750-0.877),and the C-index of validation set was 0.805(95 % CI 0.683-0.928).The area under the time-dependent ROC curve(AUC)of the training set for predicting recurrence at 6 months,1 year and 2 years after catheter ablation was0.848,0.899 and 0.884.The area under the time-dependent ROC curve(AUC)of the validation set for predicting recurrence at 6 months,1 year and 2 years after catheter ablation was 0.856,0.814 and 0.841.The consistency index and the area under the ROC curve indicated that the nomogram model had good discrimination and consistency.4.According to the nomogram scores of all patients,they were divided into high score group and low score group.And the cumulative recurrence probability curve of the two groups shows that here was a significant difference between the two groups(Log-rank P<0.0001).Conclusions:1.Age,course of disease,LAD and hypertension are independent predictors of recurrence after catheter ablation in patients with paroxysmal atrial fibrillation.2.The nomogram model for predicting recurrence after catheter ablation in patients with paroxysmal atrial fibrillation constructed by integrating age,course of disease,LAD and hypertension has good discrimination and consistency.3.Patients with high scores in the nomogram had a significantly higher probability of recurrence after catheter ablation over time than those with low scores,suggesting that the nomogram model has good predictive performance.更多还原