Jacobs K, Rigdon J, Chan F, Cheng JY, Alley MT, Vasanawala S, Maskatia SA.J Cardiovasc Magn Reson. 2020 May 14;22(1):33. doi: 10.1186/s12968-020-00612-4.PMID: 32404159 Free PMC article.
Abstract
Background: 3D-time resolved flow (4DF) cardiovascular magnetic resonance (CMR) with retrospective analysis of atrioventricular valve regurgitation (AVVR) allows for internal validation by multiple direct and indirect methods. Limited data exist on direct measurement of AVVR by 4DF CMR in pediatric congenital heart disease (CHD). We aimed to validate direct measurement of the AVVR jet as accurate and reliable compared to the volumetric method (clinical standard by 2D CMR) and as a superior method of internal validation than the annular inflow method.
Methods: We identified 44 consecutive patients with diverse CHD referred for evaluation of AVVR by CMR. 1.5 T or 3 T scanners, intravenous contrast, and a combination of parallel imaging and compressed sensing were used. Four methods of measuring AVVR volume (RVol) were used: volumetric method (VOL; the clinical standard) = stroke volume by 2D balanced steady-state free precession – semilunar valve forward flow (SLFF); annular inflow method (AIM) = atrioventricular valve forward flow [AVFF] – semilunar valve net flow (SLNF); and direct measurement (JET). AVFF was measured using static and retrospective valve tracking planes. SLFF, SLNF, AVFF, and JET were measured by 4DF phase contrast. Regurgitant fraction was calculated as [RVol/(RVol+SLNF)]× 100. Statistical methods included Spearman, Wilcoxon rank sum test/Student paired t-test, Bland Altman analysis, and intra-class coefficient (ICC), where appropriate.
Results: Regurgitant fraction by JET strongly correlated with the indirect methods (VOL and AIM) (ρ = 0.73-0.80, p < 0.001) and was similar to VOL with a median difference (interquartile range) of – 1.5% (- 8.3-7.2%; p = 0.624). VOL had weaker correlations with AIM and JET (ρ = 0.69-0.73, p < 0.001). AIM underestimated RF by 3.6-6.9% compared to VOL and JET, p < 0.03. Intra- and inter- observer reliability were excellent for all methods (ICC 0.94-0.99). The mean (±standard deviation) inter-observer difference for VOL was 2.4% (±5.1%), p < 0.05.
Conclusions: In a diverse cohort of pediatric CHD, measurement of AVVR using JET is accurate and reliable to VOL and is a superior method of internal validation compared to AIM. This study supports use of 4DF CMR for measurement of AVVR, obviating need for expert prospective prescription during image acquisition by 2D CMR.
Fig. 1 Graphic and mathematical depiction of the indirect methods calculating atrioventricular valve regurgitation (AVVR). a Ventricular stroke volume (SV) is the sum of AVVR and semilunar forward flow (SLFF); b AVVR is derived from the conservation of mass which assumes that atrioventricular net flow (AVNF) equals semilunar net flow (SLNF) in the absence of an intracardiac shunt
Fig. 2 4DF CMR on Arterys of the direct jet method (a, b) and the indirect annular inflow method (AIMtrack) by valve tracking planes (c, d). Examples of tricuspid regurgitation (row a) and mitral regurgitation (row b) are presented. Rows (c) and (d) represent four-chamber and two-chamber views of AIMtrack
Fig. 3 Bland Altman plots comparing the 4 methods in this study: the indirect methods (VOL and both AIMs) and the direct method (JET). The limits of agreements (dashed lines) are narrower for the JET comparisons (c–e) than for the VOL comparisons (a–c). The mean difference is shown as a solid line. Mitral regurgitation (MR) is denoted by red points, tricuspid regurgitation (TR) is denoted by blue points. VOL = volumetry method, AIMstat = annular inflow method by static plane, AIMtrack = annular inflow method by valve-tracking plane, JET = direct measurement of jet
Fig. 4 Linear regression plots comparing the 4 methods of this study: the indirect methods (VOL and both AIMs) and the direct method (JET). Given non-parametric distribution of data, Spearman’s correlation coefficients are substituted for Pearson’s. Mitral regurgitation (MR) is denoted by red points, tricuspid regurgitation (TR) is denoted by blue points. VOL = volumetry method, AIMstat = annular inflow method by static plane, AIMtrack = annular inflow method by valve-tracking plane, JET = direct measurement of jet
Fig. 5 Bland Altman plots comparing the intra- and inter-observer reliability measurements (a–d and e–h, respectively) using the indirect and direct methods measuring atrioventricular valve regurgitation. Mean difference is represented with a solid line; limits of agreement are represented with dashed lines. Mitral regurgitation (MR) is denoted by red points, tricuspid regurgitation (TR) is denoted by blue points. VOL = volumetry method, AIMstat = annular inflow method by static plane, AIMtrack = annular inflow method by valve-tracking plane, JET = direct measurement of jet
