| 赖碧银,李叶阔,徐宣寿,周盼妍,张恒.基于超声心动图左心房参数列线图识别心房颤动[J].中国医学影像技术,2024,40(12):1831~1836 |
| 基于超声心动图左心房参数列线图识别心房颤动 |
| Nomogram based on left atrial echocardiography parameters for identifying atrial fibrillation |
| 投稿时间：2024-07-26 修订日期：2024-10-21 |
| DOI：10.13929/j.issn.1003-3289.2024.12.006 |
| 中文关键词: 心房颤动 心房功能,左 超声心动描记术,四维 列线图 |
| 英文关键词:atrial fibrillation atrial function, left echocardiography, four-dimensional nomograms |
| 基金项目: |
|
| 摘要点击次数: 388 |
| 全文下载次数: 87 |
| 中文摘要: |
| 目的 观察基于超声心动图左心房(LA)参数列线图识别心房颤动(AF)的价值。方法 回顾性纳入66例成年AF患者及65名非AF者,随机将其分为训练集(n=79)与测试集(n=52),行LA四维容积超声心动图检查,并以四维左心房自动定量(4D LAQ)技术测量LA容积及应变参数。比较训练集AF组(n=38)与非AF组(n=41)超声心动图参数,构建用于识别AF的logistic回归模型,采用受试者工作特征(ROC)曲线及其曲线下面积(AUC)评估模型识别AF的效能;基于logistic回归模型绘制列线图,以校准曲线评估校准度,采用决策曲线分析(DCA)评估临床收益。结果 训练集中,相比非AF组,AF组LA内径(LAD)、LA最小容积(LAVmin)、LA最大容积(LAVmax)、LA最大容积指数(LAVImax)均增加,而左心室射血分数(LVEF)、LA排空容量(LAEV)、LA排空分数(LAEF)及LA储存期纵向应变(LASr)、LA储存期周向应变(LASr_c)应变大小均降低(P均＜0.05)。LASr[OR(95%CI)=0.780(0.636,0.956)]及LAEF[OR(95%CI)=0.850(0.757,0.954)]均与AF独立相关,以之构建的识别AF的logistic回归模型在训练集的AUC为0.950,在测试集为0.955;其列线图亦具有较好校准度,并可带来一定临床净收益。结论 基于超声心动图LASr及LAEF所获列线图能有效识别AF。 |
| 英文摘要: |
| Objective To observe the value of nomogram based on echocardiographic functional parameters of left atrium (LA) for identifying atrial fibrillation (AF). Methods Sixty-six adult patients with AF and 65 individuals without AF were retrospectively enrolled and randomly divided into training set (n=79) and test set (n=52). Four-dimensional LA volumetric echocardiography was performed, and LA volume and strain parameters were measured using four-dimensional left atrial automatic quantification (4D LAQ) technology. Echocardiographic parameters were compared between AF group (n=38) and non-AF group (n=41) in training set, and a logistic regression model for identifying AF was constructed. The performance of the model was evaluated with receiver operating characteristic (ROC) curve and the area under the curve (AUC). Then a nomogram was developed based on the logistic regression model, and the calibration was assessed using a calibration curve, as well as the clinical net benefit was evaluated through decision curve analysis (DCA). Results In training set, compared to those in non-AF group, LA diameter (LAD), the minimum LA volume (LAVmin), the maximum LA volume (LAVmax) and the maximum LA volume index (LAVImax) increased, while the left ventricular ejection fraction (LVEF), LA emptying volume (LAEV), LA emptying fraction (LAEF), and the magnitude of LA reservoir longitudinal strain (LASr) and LA reservoir circumferential strain (LASr_c) decreased in AF group(all P＜0.05). LASr (OR[95%CI]=0.780[0.636, 0.956]) and LAEF (OR[95%CI]=0.850[0.757, 0.954]) were both independently associated with AF. The logistic regression model constructed for identifying AF achieved an AUC of 0.950 in training set and 0.955 in test set. The nomogram demonstrated good calibration and provided a certain level of net clinical benefit. Conclusion Nomogram constructed on basis of LASr and LAEF was effective for identifying AF. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
