Khankirawatana Banthit, Khankirawatana Suwanee, Porter Thomas
Section of Cardiology, Department of Internal Medicine, University of Nebraska Medical Center, Omaha, Neb 68198-1165, USA.
Am Heart J. 2004 Feb;147(2):369-74. doi: 10.1016/j.ahj.2003.03.001.
Determination of left atrial (LA) size is important in clinical decision-making. The LA anteroposterior dimension (APD) has been routinely reported as LA size assessment. Early studies indicated that the APD may have limited accuracy in quantification. Conventional 3-dimensional reconstruction (C3DR) of the LA has been validated. However, its process is time-consuming and not applicable for daily practice. To explore an accurate and practical approach, we compared different echocardiographic measurements with C3DR in 141 patients with different LA sizes.
LA size was measured with (1) the cubic equation with APD (Cub); (2) the ellipsoidal formula (Ellp); (3) biplane modified Simpson rule (biplane); and (4) simplified 3-dimensional reconstruction from 3 standard apical views with B spline interpolation (S3VR). All four methods were compared with C3DR. S3VR and biplane methods provided a close agreement to C3DR (y = 0.94x + 3.6, r = 0.95, SEE = 7.6 mL, mean difference = -1.3% for S3VR; y = 0.87x + 2.9, r = 0.91, SEE = 9.0 mL, mean difference = -9.4% for biplane). The Cub and Ellp calculations were less accurate, with significant volume underestimation (P <.001).
LA single dimension is not accurate for LA size measurement. Among four different methods of LA size measurement, biplane and S3VR provide the closest agreement to C3DR. The biplane, which is readily applicable with current echocardiographic equipment, should be routinely applied in clinical practice.
左心房（LA）大小的测定在临床决策中具有重要意义。LA前后径（APD）一直被常规用作LA大小评估指标。早期研究表明，APD在定量方面的准确性可能有限。LA的传统三维重建（C3DR）已得到验证。然而，其过程耗时且不适用于日常实践。为探索一种准确且实用的方法，我们在141例不同LA大小的患者中，将不同的超声心动图测量方法与C3DR进行了比较。
用以下方法测量LA大小：（1）基于APD的立方方程（Cub）；（2）椭圆公式（Ellp）；（3）双平面改良辛普森法则（双平面）；（4）从3个标准心尖视图通过B样条插值进行的简化三维重建（S3VR）。将这四种方法与C3DR进行比较。S3VR和双平面方法与C3DR的一致性较好（S3VR：y = 0.94x + 3.6，r = 0.95，标准误 = 7.6 mL，平均差异 = -1.3%；双平面：y = 0.87x + 2.9，r = 0.91，标准误 = 9.0 mL，平均差异 = -9.4%）。Cub和Ellp计算的准确性较低，存在明显的容积低估（P <.001）。
LA单一维度对于LA大小测量并不准确。在四种不同的LA大小测量方法中，双平面和S3VR与C3DR的一致性最为接近。双平面方法可通过现有的超声心动图设备轻松实现，应在临床实践中常规应用。
