Quantitative assessment of systolic and diastolic left ventricular twist using Fourier Analysis of STimulated echoes (FAST) and CSPAMM

Abstract

Summary The FAST method for measuring left ventricular twist has been expanded to semi-automatically measure torsion, peak systolic twist rate, peak diastolic untwisting rate, time to peak twist, and duration of untwisting using FAST+CSPAMM. Background Alterations in left ventricular (LV) twist are important for many pathologies including aortic stenosis[1], diastolic dysfunction[2] and aging[3]. LV twist is a measure of the rotation of the apex relative to the base of the heart. Employing a previously validated method, Fourier Analysis of STimulated echoes (FAST)[4], which measures object rotation directly in Fourier space, we performed quantitative analysis of LV twist, torsion, twisting rates, time to peak twist, and duration of untwisting in thirteen (n=13) healthy volunteers using SPAMM and CSPAMM[5] tagged images. FAST analysis typically requires 2-3 minutes of user interaction time. Methods A spoiled gradient echo pulse sequence was modified to support CSPAMM and used to acquire short-axis images in healthy volunteers (n=13) at the LV base and apex with the following parameters: 360-300x300280mm FOV, 5-6mm slice thickness, 192x144 acquisition matrix, 501 Hz/pixel receiver bandwidth, TE/ TR=3.5-3.7/4.7-6.5ms, 10mm tag spacing, 7-8 ky-lines per segment, 3/4 partial Fourier imaging, 14-16 cardiac phases, 15° FA and two averages for SPAMM imaging, and non-linearly ramped imaging flip angles (initial angle 22°) for CSPAMM[6]. Estimates of LV rotation at basal and apical slice levels for all frames were obtained from the rotation of the stimulated echo and stimulated anti-echo about the FID in Fourier Space using the FAST method[4]. Bland-Altman analysis, linear regression, paired t-test, and Wilcoxon signed-rank test were used to compare FAST+CSPAMM to a previously validated FAST+SPAMM technique. Only data obtained during the first 500ms was used for comparison of SPAMM and CSPAMM due to tag fading. P<0.01 was considered statistically significant. Results The comparison of FAST+CSPAMM to FAST+SPAMM for twist, torsion, twisting rates, time of peak twist and duration of untwisting are summarized in Table 1. Table 2 provides peak rotation and time to peak rotation for systole and diastole for the apical and basal slices, which are important metrics in aortic stenosis[1] and aging[3]. No significant differences were found for twist, torsion, or twist rate measurements made with FAST+CSPAMM to FAST+SPAMM using the paired ttest or the Wilcoxon signed-rank test. Bland-Altman analysis and linear regression demonstrate excellent agreement between techniques (Table 1). These values match well with literature[7]. Conclusions The FAST+CSPAMM is a semi-automated method that provides a quick and quantitative assessment of systolic and diastolic left ventricular twist, torsion, and twisting rates. When FAST is used in conjunction with CSPAMM duration of untwisting can be measured, which may provide further insight into left ventricular diastolic dysfunction.