心房颤动患者射频消融术后复发风险列线图预测模型的建立与评估
Establishment and evaluation of a nomogram prediction model for recurrence risk of atrial fibrillation patients after radiofrequency ablation.
作者信息
Zhou Xiao-Juan, Zhang Li-Xiang, Xu Jian, Zhu Hong-Jun, Chen Xia, Wang Xue-Qi, Zhao Mei
机构信息
School of Nursing, Anhui Medical University Hefei, Anhui, China.
Department of Cardiology, The First Affiliated Hospital of University of Science and Technology of China Hefei, Anhui, China.
出版信息
Am J Transl Res. 2021 Sep 15;13(9):10641-10648. eCollection 2021.
OBJECTIVE
To explore the risk factors for recurrence of atrial fibrillation (AF) in patients after radiofrequency ablation and construction of a targeted nomogram prediction model.
METHODS
A prospective cohort study design was used to select 312 patients who were separated into two groups; a recurrence group (n = 79) and a non-recurrence group (n = 233) with or without AF, who underwent radiofrequency ablation for the first time between January 2017 and December 2017, with a completed a 12-month follow-up after surgery. The recurrence of AF within 12 months after follow-up was recorded. The nomogram prediction model was established. The original data were resampled using the Bootstrap method. The recurrence risk after resampling was predicted using a nomogram model. The calibration curve and ROC curve of the nomogram model were established. The predicted calibration degree and discrimination degree of the nomogram model were evaluated with the Hosmer-Lemeshow deviation test and area under the curve.
RESULTS
The 12-month follow-up showed that a total of 79 patients (25.32%) had recurrence of AF. The type of AF, sex, gender, disease course, left atrial anteroposterior diameter, left atrial volume, and cardiac function classification were independent risk factors for the recurrence of AF (P < 0.05). After the nomogram prediction model passed the Bootstrap self-sampling 1000 times, Hosmer-Lemeshow deviation test: χ = 8.070, P = 0.427; the area under ROC curve was 0.852 (95% CI: 0.806-0.898), the sensitivity was 78.48%, and the specificity was 81.12%, suggesting that the nomogram model has better predictive calibration and discrimination.
CONCLUSION
The recurrence rate in patients with AF after radiofrequency ablation is high. The nomogram model based on the risk factors of AF recurrence has high prediction accuracy and can be used to predict the recurrence risk of AF in patients after radiofrequency ablation.
目的
探讨心房颤动（AF）患者射频消融术后复发的危险因素，并构建靶向列线图预测模型。
方法
采用前瞻性队列研究设计，选择312例患者分为两组，即复发组（n = 79）和未复发组（n = 233），这些患者在2017年1月至2017年12月期间首次接受射频消融，并在术后完成了12个月的随访。记录随访后12个月内AF的复发情况。建立列线图预测模型。使用Bootstrap方法对原始数据进行重采样。使用列线图模型预测重采样后的复发风险。建立列线图模型的校准曲线和ROC曲线。采用Hosmer-Lemeshow偏差检验和曲线下面积评估列线图模型的预测校准度和区分度。
结果
12个月随访显示，共有79例患者（25.32%）发生AF复发。AF类型、性别、病程、左心房前后径、左心房容积和心功能分级是AF复发的独立危险因素（P < 0.05）。列线图预测模型经Bootstrap自抽样1000次后，Hosmer-Lemeshow偏差检验：χ = 8.070，P = 0.427；ROC曲线下面积为0.852（95%CI：0.806 - 0.898），灵敏度为78.48%，特异度为81.12%，表明列线图模型具有较好的预测校准度和区分度。
结论
AF患者射频消融术后复发率较高。基于AF复发危险因素的列线图模型具有较高的预测准确性，可用于预测AF患者射频消融术后的复发风险。
Zotero 插件