Department of Cardiology, Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio.
J Am Soc Echocardiogr. 2013 Nov;26(11):1267-73. doi: 10.1016/j.echo.2013.07.017. Epub 2013 Aug 28.
Although not recommended in isolation, visual estimation of echocardiographic ejection fraction (EF) is widely applied to confirm quantitative EF. However, interobserver variability for EF estimation has been reported to be as high as 14%. The aim of this study was to determine whether self-directed education could improve the accuracy and interobserver variability of visual estimation of EF and whether a multireader estimate improves measurement precision.
Thirty-one participants provided single-point EF estimates for 30 echocardiograms with a spectrum of EFs, image quality, and clinical contexts in patients undergoing cardiac magnetic resonance (CMR) within 48 hours. Participants received their own case-by-case variance from CMR EF, and the 10 cases with the largest reader variability were discussed along with corresponding CMR images. Self-directed learning was undertaken by side-by-side review of echocardiographic and CMR images. Two months later, 20 new cases were shown to the same 31 participants, using the same methodology.
The baseline interobserver variability of ±0.120 improved to ±0.097 after the intervention. EF misclassification (defined as ±0.05 of CMR EF) was reduced from 56% to 47% (P < .001), and the intervention also resulted in a decrease in the absolute difference between CMR and echocardiography for all cases and all readers (from 0.07 ± 0.01 to 0.06 ± 0.01, P = .0001). This improvement was most prominent for the readers with lower baseline accuracy. A combined physician-sonographer EF estimate improved the precision of EF determination by 25% compared with individual reads.
In readers with varying levels of experience, a simple, mostly self-directed intervention modestly decreased interobserver variability and improved the accuracy of EF measurements. Combined physician-sonographer EF reporting improved the precision of EF estimates.
尽管不单独推荐，但超声心动图射血分数（EF）的目测估计被广泛应用于确认定量 EF。然而，EF 目测估计的观察者间变异性报道高达 14%。本研究旨在确定自我指导教育是否可以提高 EF 目测的准确性和观察者间变异性，以及多读者估计是否可以提高测量精度。
31 名参与者在心脏磁共振（CMR）检查后 48 小时内，对 30 个 EF、图像质量和临床背景各不相同的超声心动图进行单点 EF 估计。参与者根据 CMR EF 获得了自己的个案方差，并且对 10 个观察者间变异性最大的病例进行了讨论，并结合相应的 CMR 图像进行了讨论。自我指导学习是通过并排查看超声心动图和 CMR 图像进行的。两个月后，使用相同的方法向相同的 31 名参与者展示了 20 个新病例。
干预前观察者间变异性为±0.120，干预后改善为±0.097。EF 分类错误（定义为 CMR EF 的±0.05）从 56%减少到 47%（P<.001），干预还导致所有病例和所有读者的 CMR 和超声心动图之间的绝对差异减小（从 0.07±0.01 到 0.06±0.01，P=.0001）。对于基线准确性较低的读者，这种改善更为明显。与单独阅读相比，医生-超声科医师联合 EF 估计可提高 EF 测定的精度 25%。
在经验水平不同的读者中，简单的、主要是自我指导的干预可以适度降低观察者间变异性，提高 EF 测量的准确性。联合医生-超声科医师 EF 报告提高了 EF 估计的精度。
