Pagourelias Efstathios D, Mirea Oana, Duchenne Jürgen, Van Cleemput Johan, Delforge Michel, Bogaert Jan, Kuznetsova Tatyana, Voigt Jens-Uwe
From the Departments of Cardiovascular Diseases (E.D.P., O.M., J.D., J.V.C., J.-U.V.), Hematology (M.D.), Radiology (J.B.), Cardiovascular Sciences (T.K.), University Hospital Leuven, Catholic University Leuven, Belgium.
Circ Cardiovasc Imaging. 2017 Mar;10(3):e005588. doi: 10.1161/CIRCIMAGING.116.005588.
A plethora of echo parameters has been suggested for distinguishing cardiac amyloidosis (CA) from other causes of myocardial thickening with, however, scarce data on their head-to-head comparison. This study aimed at comparing the diagnostic accuracy of various deformation and conventional echo parameters in differentiating CA from other hypertrophic substrates, especially in the gray zone of mild hypertrophy (maximum wall thickness ≤16 mm) or normal ejection fraction (EF).
We included 100 subjects, of which 40 were patients with newly diagnosed, biopsy-proven CA (65.5±10.8 years, 65% male, 62.5% amyloidosis light chain [AL] type), 40 patients with hypertrophic cardiomyopathy matched for demographics and maximum wall thickness (60.1±14.8 years, 85% male), and 20 hypertensives with prominent myocardial remodeling. Quantifiable conventional morphological and functional parameters along with multidimensional strain and strain-derived ratios indices, previously suggested to diagnose CA, were analyzed. EF global longitudinal strain ratio showed the best performance to discriminate CA (area under the curve, 0.95; 95% confidence intervals, 0.89-0.98; <0.00005). Traditional echo indices showed overall low sensitivities and high specificities (among them myocardial contraction fraction ratio had the highest area under the curve, 0.80; 95% confidence intervals, 0.7-0.87; <0.0001). In the challenging subgroups (maximum wall thickness ≤16 mm and EF>55%), EF global longitudinal strain ratio remained the best predicting parameter of CA diagnosis (multiple logistic regression models <0.00005 and =0.0002, respectively) independent of the CA type.
Our study demonstrated that in patients with thickened hearts, EF global longitudinal strain ratio has the best accuracy in detecting CA, even among the most "challenging" patient subgroups as those with mild hypertrophy and normal EF.
已有大量超声心动图参数被推荐用于鉴别心脏淀粉样变性（CA）与其他导致心肌增厚的病因，然而，关于这些参数直接比较的数据却很少。本研究旨在比较各种变形参数和传统超声心动图参数在鉴别CA与其他肥厚性心肌病变方面的诊断准确性，尤其是在轻度肥厚（最大壁厚≤16mm）或正常射血分数（EF）的灰色区域。
我们纳入了100名受试者，其中40例为新诊断的、经活检证实的CA患者（65.5±10.8岁，65%为男性，62.5%为轻链淀粉样变性[AL]型），40例肥厚型心肌病患者，其人口统计学特征和最大壁厚相匹配（60.1±14.8岁，85%为男性），以及20例有明显心肌重构的高血压患者。分析了可量化的传统形态学和功能参数以及先前推荐用于诊断CA的多维应变和应变衍生比率指标。EF全局纵向应变率在鉴别CA方面表现最佳（曲线下面积，0.95；95%置信区间，0.89 - 0.98；<0.00005）。传统超声心动图指标总体敏感性较低而特异性较高（其中心肌收缩分数比率曲线下面积最大，0.80；95%置信区间，0.7 - 0.87；<0.0001）。在具有挑战性的亚组（最大壁厚≤16mm且EF>55%）中，EF全局纵向应变率仍然是CA诊断的最佳预测参数（多因素逻辑回归模型分别为<0.00005和=0.0002），与CA类型无关。
我们的研究表明，在心脏增厚的患者中，EF全局纵向应变率在检测CA方面具有最佳准确性，即使在最“具挑战性”的患者亚组中，如轻度肥厚和EF正常的患者。
