Mahabadi Amir A, Lehmann Nils, Sonneck Nina C, Kälsch Hagen, Bauer Marcus, Kara Kaffer, Geisel Marie H, Moebus Susanne, Jöckel Karl-Heinz, Erbel Raimund, Möhlenkamp Stefan
Department of Cardiology, West-German Heart Center, University of Duisburg-Essen, Essen, Germany
Institute for Medical Informatics, Biometry, and Epidemiology, University of Duisburg-Essen, Essen, Germany.
Acta Radiol. 2014 Oct;55(8):917-25. doi: 10.1177/0284185113507446. Epub 2013 Oct 10.
Left atrial (LA) size is associated with cardiovascular mortality and morbidity. Once cardiac computed tomography (CT) is performed, information on LA size is readily available without additional contrast media or radiation exposure.
To determine the association of CT-derived LA area and body surface area-adjusted (BSA) LA index with cardiovascular risk factors and describe age- and gender-specific normative values in a general population cohort.
This study included 3945 participants (mean age, 59 ± 8 years; 53% women) from the community-based Heinz Nixdorf Recall Study. LA area in an axial image at the level of the mitral valve was quantified from non-contrast-enhanced electron-beam CT by manual delineations of the boundaries of the LA with exclusion of subjects with prevalent cardiovascular disease. Definition of normative values was performed in subjects without predictors of LA enlargement.
LA quantification was feasible in all subjects. Men had larger LA size (1856 mm(2) vs. 1677 mm(2), P < 0.0001), while after adjustment for BSA, this effect was inverted (910 mm(2)/m(2) vs. 933 mm(2)/m(2) for men and women, P < 0.0001). Determinants of body size were major predictors of LA size (body mass index [BMI]: R(2) = 0.195, BSA: R(2 )= 0.216, both P < 0.0001). Blood pressure was associated with LA size (parameter-estimate [95% confidence interval] = 51.0 (4.9-57.1) mm(2)/10 mmHg for systolic, 31.4 (25.4-37.4) mm(2)/5 mmHg for diastolic blood pressure, 214.6 (186.9-242.3) mm(2) for antihypertensive medication, P < 0.0001 for all). Cholesterol levels, lipid-lowering therapy, and diabetes were associated with LA in univariable analysis, however, correlations were low (r(2 )≤ 0.026). Current smoking was associated with reduced LA size (-115.9 [-149.0 - -82.8] mm(2), P < 0.0001). In multivariable regression, BMI, blood pressure, antihypertensive medication, and smoking remained associated with LA size (P < 0.005).
Non-contrast-enhanced cardiac CT enables LA quantification with body size, hypertension, and smoking status being predictors of LA size.
左心房（LA）大小与心血管疾病的死亡率和发病率相关。一旦进行心脏计算机断层扫描（CT），无需额外的造影剂或辐射暴露即可轻松获得有关LA大小的信息。
确定CT衍生的LA面积和体表面积调整（BSA）后的LA指数与心血管危险因素之间的关联，并描述一般人群队列中特定年龄和性别的正常值。
本研究纳入了来自社区的海因茨·尼克斯多夫召回研究中的3945名参与者（平均年龄59±8岁；53%为女性）。通过手动勾勒LA边界，从非增强电子束CT的二尖瓣水平轴向图像中量化LA面积，排除患有心血管疾病的受试者。在没有LA扩大预测因素的受试者中定义正常值。
所有受试者的LA量化均可行。男性的LA尺寸更大（1856平方毫米对1677平方毫米，P<0.0001），但在调整BSA后，这种差异颠倒（男性和女性分别为910平方毫米/平方米对933平方毫米/平方米，P<0.0001）。体型决定因素是LA大小的主要预测因素（体重指数[BMI]：R² = 0.195，BSA：R² = 0.216，均P<0.0001）。血压与LA大小相关（参数估计[95%置信区间]：收缩压为51.0（4.9 - 57.1）平方毫米/10 mmHg，舒张压为31.4（25.4 - 37.4）平方毫米/5 mmHg，使用抗高血压药物为214.6（186.9 - 242.3）平方毫米，所有P<0.0001）。在单变量分析中，胆固醇水平、降脂治疗和糖尿病与LA相关，然而，相关性较低（r²≤0.026）。当前吸烟与LA尺寸减小相关（-115.9[-149.0 - -82.8]平方毫米，P<0.0001）。在多变量回归中，BMI、血压、抗高血压药物和吸烟仍然与LA大小相关（P<0.005）。
非增强心脏CT能够量化LA大小，体型、高血压和吸烟状况是LA大小的预测因素。
