Department of Cardiology, The First Hospital of Jilin University, Changchun, Jilin Province, China.
Medicine (Baltimore). 2023 Feb 10;102(6):e32910. doi: 10.1097/MD.0000000000032910.
Dilated cardiomyopathy (DCM) is one of the most common causes of heart failure. Therefore, screening and early diagnosis of potential DCM patients is beneficial. Electrocardiogram (ECG) can be an inexpensive and easily available screening tool. We aimed to construct a simple screening model for DCM based on electrocardiogram. In this retrospective observational study, we consecutively enrolled 117 DCM patients between July 1, 2016 and July 1, 2021 as the DCM group, while 117 patients hospitalized in the same period with normal echocardiography and ECG were selected as the non-DCM group. Patients were randomly assigned to the training and validation sets at 8:2. ECG parameters of left ventricular related leads were exacted. Logistic regression was performed to evaluate screening indicators of ECG parameters and a nomogram was conducted. The screening ability of the model was evaluated using receiver operating characteristic analysis. Furthermore, the nomogram was assessed using calibration curve and decision curve analysis. Screening indicators included in the nomogram were the amplitude of S wave in V1 and V3 leads, the amplitude of R wave in aVF and V6 leads, and PR interval. The nomogram performed satisfactory discrimination in the training (area under the receiver operating characteristic curve = 0.904) and validation (area under the receiver operating characteristic curve = 0.878) sets and good calibration (Hosmer-Lemeshow P = .066). Decision curve analysis demonstrated the model can generate a net benefit of 0.33 when the threshold probability was 0.543. The nomogram based on electrocardiogram is a simple and practical screening tool for potential DCM patients.
扩张型心肌病（DCM）是心力衰竭最常见的原因之一。因此，对潜在 DCM 患者进行筛查和早期诊断是有益的。心电图（ECG）可以作为一种廉价且易于获取的筛查工具。我们旨在构建一种基于心电图的简单 DCM 筛查模型。在这项回顾性观察研究中，我们连续纳入了 2016 年 7 月 1 日至 2021 年 7 月 1 日期间的 117 例 DCM 患者作为 DCM 组，同时选择同期心电图和超声心动图正常的 117 例患者作为非 DCM 组。患者被随机分为训练集和验证集，比例为 8:2。提取左心室相关导联的心电图参数。使用逻辑回归评估心电图参数的筛查指标，并建立列线图。使用接受者操作特征曲线分析评估模型的筛查能力。此外，还通过校准曲线和决策曲线分析评估了列线图的性能。纳入列线图的筛查指标包括 V1 和 V3 导联 S 波振幅、aVF 和 V6 导联 R 波振幅以及 PR 间期。该列线图在训练集（接受者操作特征曲线下面积=0.904）和验证集（接受者操作特征曲线下面积=0.878）中具有良好的区分能力，且校准良好（Hosmer-Lemeshow P=0.066）。决策曲线分析表明，当阈值概率为 0.543 时，该模型可产生 0.33 的净收益。基于心电图的列线图是一种简单实用的潜在 DCM 患者筛查工具。
