Comparison of Simpson's Method and Three-Dimensional Reconstruction for Measurement of Right Ventricular Volume in Patients With Complete or Corrected Transposition of the Great Arteries

Abstract

The right ventricular (RV) volume is commonly measured from magnetic resonance images using Simpson's method from the stack of short-axis images acquired for analysis of the left ventricle. We compared the RV volume measured using Simpson's method to the RV volume measured using 3-dimensional reconstruction and the piecewise smooth subdivision surface (PSSS) method. We studied 6 normal subjects and 18 patients whose right ventricles carried a systemic pressure load, 1/2 with dexto-transposition of the great arteries repaired with an atrial baffle and 1/2 with levo-transposition of the great arteries. The right ventricle was reconstructed from manually traced borders from the short- and long-axis views using the PSSS method. Simpson's analysis was performed on short-axis views alone. The RV volumes were smaller when analyzed using Simpson's method than using the PSSS method. The underestimation averaged 12 +/- 19 ml (7 +/- 12% of PSSS volume; p <0.001), without a significant difference between the groups. The ejection fraction was similar using both methods in patients with transposition of the great arteries and was overestimated in normal subjects. Image review revealed that the volume underestimation using Simpson's method was more frequently due to difficulty in interpreting the basal short-axis images than the apical images. In conclusion, to obtain accurate analysis of the short-axis views for RV volume measurement, it would be helpful to incorporate information from additional images, such as the long-axis views, to assist in delineating this chamber's complex anatomy.