Peak flow measurements in patients with severe aortic stenosis: a prospective comparative study between cardiovascular magnetic resonance 2D and 4D flow and transthoracic echocardiography

Abstract

Background Aortic valve stenosis (AS) is the most prevalent valvular disease in the developed countries. Four-dimensional (4D) flow cardiovascular magnetic resonance (CMR) is an emerging imaging technique, which has been suggested to improve the evaluation of AS severity compared to two-dimensional (2D) flow and transthoracic echocardiography (TTE). We investigated the reliability of CMR 2D flow and 4D flow techniques in measuring aortic transvalvular peak systolic flow in patients with severe AS. Methods We prospectively recruited 90 patients referred for aortic valve replacement due to severe AS (73.3 ± 11.3 years, aortic valve area 0.7 ± 0.1 cm2, and 54/36 tricuspid/bicuspid), and 10 non-valvular disease controls. All the patients underwent echocardiography and 2D flow and 4D flow CMR. Peak flow velocity measurements were compared using Wilcoxon signed rank sum test and Bland–Altman analysis. Results 4D flow underestimated peak flow velocity in the AS group when compared with TTE (bias − 1.1 m/s, limits of agreement ± 1.4 m/s) and 2D flow (bias − 1.2 m/s, limits of agreement ± 1.6 m/s). The differences between values obtained by TTE (median 4.3 m/s, range 2.7–6.1 m/s) and 2D flow (median 4.5 m/s, range 2.9–6.5 m/s) compared to 4D flow (median 3.1 m/s, range 1.7–5.1 m/s) were significant (p < 0.001). The difference between 2D flow and TTE were insignificant (bias 0.07 m/s, limits of agreement ± 1.5 m/s). In non-valvular disease controls, peak flow velocity was measured higher by 4D flow than 2D flow (1.4 m/s, 1.1–1.7 m/s and 1.3 m/s, 1.1–1.5 m/s, respectively; bias 0.2 m/s, limits of agreement ± 0.16 m/s). Conclusions CMR 4D flow significantly underestimates systolic peak flow velocity in patients with severe AS. 2D flow, in turn, estimated the AS velocity accurately, with measured peak flow velocities comparable to TTE.

4D FlowVersus2D Phase Contrast MRI in Populations With Bi- and Tricuspid Aortic Valves

“…flow MRI appeared to be better at estimating the higher velocity values at the valvular level in TAV patients. The finding that the peak velocities were significantly higher in BAV as compared to TAV could be attributed to the higher accuracy of the 4D method to detect the decreases in flow at increasing flow velocities as described earlier by Halva et al (8). When measuring the flow velocities upstream of the valve, 4D flow MRI produced systematically higher values indicating its superiority over 2D PC MRI.…”

“…While there is a growing number of reports comparing 4D flow MRI against different clinically used modalities in healthy volunteers, there is still little published information regarding patients with accelerated and complex flow. A preliminary report indicated that 4D flow MRI underestimated peak velocities in cases of severe aortic stenosis compared to TTE and 2D PC MRI (8). However, in patients with a pulmonary valve with lower velocities, the 4D flow method was claimed to be more accurate when compared to 2D PC MRI (9).…”

“…Even though one would expect more elaborate and precise velocity values with MRI technology-based method, and indeed it has been shown to be excellent in measuring velocities in a healthy population, it has been reported to be poor in detecting high-velocity values in patients with severely stenotic valves (8). This has been hypothesized to derive from the intravoxel averaging of the narrow peak jets in cases of severe valvular stenosis.…”

Quantitative 4D flow MRI-derived thoracic aortic normal values of 2D flow MRI parameters in healthy volunteers

“…There are numerous studies regarding the validation of 4 D flow MRI sequences against 2 D flow MRI in vivo showing no significant differences between 2 D flow MRI and 4 D flow MRI, and excellent correlation between both techniques with a correlation coefficient of up to R = 0.98 has been found [13,16,17]. Contrary to those results, other groups reported that 4 D flow MRI significantly underestimates systolic peak flow velocities, while 2 D flow MRI gives accurate results [18,19]. Other groups found significant underestimation of aortic or pulmonary regurgitation and intracardiac flow when using 4 D flow MRI measurements [20][21][22].…”

Kat-ARC accelerated 4D flow CMR: clinical validation for transvalvular flow and peak velocity assessment

“…In addition, this study uniquely compares echocardiography acquired peak velocity through the aortic valve with 4D flow CMR derived peak velocity. A study by Hälvä et al [29] debates the reliability of 4D flow for peak velocity assessment. However, their work used prospectively gated 4D flow sequences, which have the issue of temporal blurring.…”