Impact of ventricular contrast medium attenuation on the accuracy of left and right ventricular function analysis at cardiac multi detector-row CT compared with cardiac MRI

Abstract

Rationale and objectives: The aim of this study was to investigate the impact of ventricular contrast medium attenuation on the accuracy of left ventricular (LV) and right ventricular (RV) function analysis on coronary computed tomographic angiographic (CCTA) imaging compared to cardiac magnetic resonance imaging (CMR).

Materials and methods: Thirty patients (mean age, 61.9 ± 11.2 years; 14 men) underwent CCTA imaging and CMR. For both the right and left ventricles, end-diastolic volume (EDV), end-systolic volume (ESV), and stroke volume (SV) were computed using multiphase image reconstruction of CCTA data. The accuracy of CCTA imaging was determined by subtracting CCTA measurements from CMR measurements. The accuracy of CCTA imaging was then correlated with the level of LV and RV contrast medium attenuation using regression analysis.

Results: In the right ventricle, there was strong correlation between the accuracy of CCTA functional assessment of EDV (R(2) = 0.78, P < .001), ESV (R(2) = 0.36, P < .001), and SV (R(2) = 0.75, P < .001) and the level of RV contrast medium attenuation. In studies with lower RV enhancement (<176 Hounsfield units; n = 15), the mean CCTA deviations of EDV, ESV, and SV from CMR measurements were 43.6 ± 17.4, 11.2 ± 9.64, and 35.1 ± 11.5 mL, respectively. In studies with higher RV attenuation (>176 Hounsfield units; n = 15), these values were 13.6 ± 10, 8.0 ± 5.28, and 13 ± 4.96 mL, respectively. In the left ventricle, there was weak correlation between functional CCTA accuracy and LV attenuation (mean, 358.31 ± 68.71 Hounsfield units), and there was excellent correlation with CMR for LV EDV (R(2) = 0.86, P < .001), ESV (R(2) = 0.85, P < .001), and SV (R(2) = 0.51, P < .001).

Conclusions: If computed tomographic evaluation of RV function is desired, attention should be paid to the contrast injection protocol, because the accuracy of RV function analysis depends on the level of contrast medium attenuation. The high contrast medium attenuation that is typically achieved in the left ventricle routinely enables highly accurate measurements compared to CMR.