Left atrial volume quantification using cardiac MRI in atrial fibrillation: comparison of the Simpson’s method with biplane area-length, ellipse, and three-dimensional methods

Abstract

Left atrial volume is an important predictor of future arrhythmias, and it can be assessed by several different methods. Simpson's method is well accepted as a reference standard, although no standardization exists for cardiac magnetic resonance (CMR). We aimed to compare the estimations of left atrial volumes obtained by the Simpson's method with three other methods. Eighty-one consecutive patients referred for CMR imaging between February 2007 and May 2010 were included in the study (47 males; mean age, 59.4±11.5 years; body mass index, 26.3±3.7 kg/m(2)). Left atrial volume measurements were performed using the Simpson's, biplane area-length, ellipse, and three-dimensional methods. Results were correlated using a Bland-Altman plot and linear regression models and compared by two-tailed paired-sample t tests. Reader variability was also calculated. Left atrial volume measurements using the biplane area-length technique showed the best correlation with Simpson's method (r=0.92; P < 0.001). Quantification values using the ellipse and three-dimensional methods were significantly different than values obtained using the Simpson's method (P < 0.05, for both). All methods showed excellent observer reliability (intra-class correlation coefficient >0.99). The biplane area-length method can be used for left atrial volume measurement when the Simpson's method cannot be performed. If these two methods are not feasible, then all methods are highly reproducible and can be used, but should not be used interchangeably for follow-up studies.