Assessment of right ventricular function with 320-slice volume cardiac CT: comparison with cardiac magnetic resonance imaging

Abstract

To evaluate the accuracy and feasibility of right ventricular function parameters measurement using 320-slice volume cardiac CT. Retrospective analysis of 50 consecutive patients (23 men, 27 women) with suspected pulmonary diseases was performed in electrocardiogram (ECG)-gated cardiac CT and cardiac magnetic resonance (CMR). Parameters including right ventricular end-diastolic volume (RVEDV), right ventricular end- systolic volume (RVESV), right ventricular stroke volume (RVSV), right ventricular cardiac output (RVCO), and right ventricular ejection fraction (RVEF) were semi-automatically and separately calculated from both CT and CMR data. Significant difference between measurements was measured by paired t test and two-variable linear regression analysis with Pearson's correlation coefficient. Bland-Altman analysis was performed in each pair of parameters. There was little variability between the measurements by the two observers (kappa = 0.895-0.980, P < 0.05). There was good correlation between all parameters obtained by CT and CMR (P < 0.001): RVEDV (108.5 ± 21.9 ml, 113.5 ± 24.8 ml, r = 0.944), RVESV (69.8 ± 33.4 ml, 73.2 ± 35.4 ml, r = 0.972), RVSV (39.0 ± 13.2 ml, 40.2 ± 13.3 ml, r = 0.977), RVCO (2.6 ± 0.7 l, 2.6 ± 0.7 l. r = 0.958), RVEF (38.8 ± 19.1 %, 39.1 ± 19.3 %, r = 0.990), and there was no significant difference between CT and CMR measurements in RVEF (n = 50, t = -0.677, P > 0.05). 320-slice volume cardiac CT is an accurate non-invasive technique to evaluate RV function.

The right ventricle in patients with chronic heart failure and atrial fibrillation

“…There was also good correlation between all parameters obtained by CT and CMR (p < 0.001): RV end-diastolic volume (RVEDV 108.5 ± 21.9 mL, 113.5 ± 24.8 mL, r = 0.944), RV end-systolic volume (RVESV 69.8 ± 33.4 mL, 73.2 ± 35.4 mL, r = =0.972), RV systolic volume (RVSV 39.0 ± 13.2 mL, 40.2 ± 13.3 mL, r = 0.977), RV cardiac output (RVCO 2.6 ± 0.7l, 2.6 ± 0.7l, r = 0.958), and there was no significant difference between CT and CMR measurements in RVEF (38.8 ± 19.1%, 39.1 ± ± 19.3%, r = 0.990, n = 50, t = -0.677, p > 0.05). The results of the study suggest, that 320-slice volume cardiac CT is an accurate non-invasive technique to evaluate RV function [58]. Other paper shows that cardiac CT fully quantifies LV size and function.…”

Cardiac CT for Measurement of Right Ventricular Volume and Function in Comparison with Cardiac MRI: A Meta-Analysis

“…The number of CT slices was ≥ 64 (15,(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29) in 13 studies and < 64 in 7 studies. In studies with ≥ 64 CT slices (6, 11-14, 16, 17), EDV and ESV showed a lower weighted bias and narrower LOA than in studies with < 64 CT slices.…”

“…Table 1 and Supplementary Table 1 summarize the study characteristics and details of image acquisition and RV analysis method in the included studies. The radiation dose from cardiac CT was 2.7-20 mSv in the ten included studies (11,15,16,(20)(21)(22)(24)(25)(26)(27).…”

“…The Simpson's method was used in 14 studies for RV volume measurement on CT (12-20, 22, 24, 26, 28, 29), while 3D-based methods were used in 5 studies (6,21,23,25,27), and Koch et al (11) used both methods in the same study population. Studies using Simpson's method showed a lower bias and narrower LOA for EDV and ESV than kjronline.org those using a 3D-based method.…”

“…With the recent improvements in the temporal and spatial resolution of computed tomography (CT) scanners, cardiac CT can be used to assess RV volume and function (10). Previous studies that compared CT measurements of RV volume and function with CMRI as the reference standard showed variable results regarding the agreement between cardiac CT and CMRI (6,(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29). These variable results may be attributable to the differences in RV segmentation methods in cardiac CT and MRI; for example, two-dimensional (2D)-vs. threedimensional (3D)-based methods, manual vs. semiautomatic vs. automatic segmentation, and threshold-based segmentation vs. simplified contouring (30)(31)(32)(33).…”