Rapid analysis of right ventricular volumes and systolic function using Cardiovascular Magnetic Resonance Imaging

Introduction

The function of the right ventricle (RV) is an important prognostic factor in heart disease. Assessment of RV function using the gold standard disk-area (or Simpson) technique is time consuming and semi-quantitative evaluation (e.g. TAPSE) shows unsatisfactory correlation with quantitative analysis. The assessment of left ventricular function with a biplane area-length method has been validated for invasive ventriculography and for CMR. This method has never been used for the RV because of its particular geometry.

Purpose

This study investigates the use of an area-length technique for the rapid assessment of the volumes and function of the RV using cardiovascular magnetic resonance imaging (CMR).

Methods

12 healthy volunteers and 25 patients with RV dilation and/or dysfunction (9 with pulmonary hypertension (PHT), 9 with dilated RV without PHT, 7 with ischemic heart disease) underwent cine CMR on a 1.5 T scanner. RV end-diastolic volume (RVEDV), end-systolic volume (RVESV) and ejection fraction (RVEF) were calculated using a stack of short-axis cines as the standard of reference (Simpson technique). A 4-chamber view and a perpendicular RV 2-chamber view were acquired to calculate RV volumes and RVEF using a simple area-length method (A*A/L) without correction coefficient in both end-diastolic and end-systolic phases with A the RV area, and L the apex to base length. This was tested for both a single and a biplane long-axis view. Finally, TAPSE was measured on the 4-chamber view. Area-length methods and TAPSE were compared to the standard of reference.

Results

Analysis of the area-length method using long-axis orientations was faster (~2 min) than the Simpson method (~20 min). The estimation of RVEF using biplane area-length method correlated well with short-axis calculation (r²_2LA = 0.72, p < 0.01). However, this method slightly overestimated RVEF (difference 5.5%, p = 0.045). The correlation between RVEF and single long-axis analysis (4-chamber) or TAPSE was weaker (r²_1LA = 0.49, p < 0.01 and r²_TAPSE = 0.37, p < 0.01, respectively). There was a good correlation between the two orientations in RVEDV measurements (r²_2LA = 0.78, p < 0.01) and in ESV measurements (r²_2LA = 0.82, p < 0.01) using the biplane method. There was no significant volume differences between the two methods (EDV: mean_2LA = 196.5 ± 92.4 mL, mean_SIMPSON = 208.4 ± 78.7 mL, p = 0.55; ESV: mean_2LA = 104.2 ± 70 mL, mean_SIMPSON = 119.0 ± 58.4 mL, p = 0.33).

Conclusion

Measurements of RV volumes and RVEF using biplane area-length method correlate well with the gold standard of contiguous short-axis measurements, despite the particular geometry of the RV. However, this method leads to a 5% overestimation of RVEF which should be taken in consideration in the RVEF calculation. As analysis time is significantly shorter, this method can be used as a rapid, quantitative screening tool.