Author + information
Traditional CMR sequences are useful tools for assessing cardiac structure and function, However, the poor image quality and motion artifacts caused by arrhythmia may hamper the diagnostic quality of CMR images. We hypothesized that the optimized temporal parallel acquisition technique (TPAT) may improve this situation by allowing free-breathing of subjects and meanwhile provides a similar diagnostic values.
Experiment: 13 patients with arrhythmia and 13 patients with normal heart rhythm were included in this study. CMR scan was performed using a 3.0T system (MAGNETON Verio, Siemens Healthcare, Erlangen, Germany). Cine images were acquired by bSSFP sequence with retrospective ECG-triggering method and the optimized TPAT technique in the same planes. The optimized TPAT sequence allows for free-breathing scans, which is beneficial for those with a poor breath-hold. Parameters of the conventional scan including: voxel size = 1.3×1.3×5.0 mm3, TR/TE = 41.1/1.51ms, FOV = 340×289 mm2, scan time per slice = 12 s, temporal resolution = 41.1ms. The parameters of TPAT scan were: voxel size = 2.9×2.1×8.0mm3, TR/TE = 60.48/1.12ms, FOV = 340× 289 mm2, scan time per slice = 2.6 s, temporal resolution = 60.48ms, TPAT factor = 4.
Image analysis: For quantitative analysis, LV global myocardial longitudinal strain (GLS), LV circumferential strain and radial strain (GCS and GRS) were measured by dedicated software (QStrain version 2.1 Medis, Leiden, the Netherlands), LV volumes and function were measured using Argus software (Siemens Healthcare, Erlangen, Germany). Specifically, GLS was calculated from 4-chamber, 3-chamber and 2-chamber orientations. GCS and GRS were calculated in the short axis orientation. LV volumes and function were measured in the short axis orientation. Besides, apparent contrast-to-noise ratio (CNR) between LV myocardium and blood pool was also calculated. For qualitative analysis, two radiologists scored the image quality (IQ) using a four-point grading scale (1=very good IQ, no artifacts affecting cardiac anatomy. 2=good/average IQ, artifacts slightly interfering with cardiac anatomy. 3=below-average IQ, artifacts moderately affecting cardiac anatomy. 4=poor IQ with artifacts severely affecting cardiac anatomy) . Wilcoxon signed-rank test was used to compare the measured samples between the conventional method and optimized TPAT technique in patient groups with/without arrhythmia.
There was no significant difference between the conventional method and TPAT technique in evaluation of LV volumes and function, it’s also the same for longitudinal, circumferential and radial strains in both patient groups. For the images of patients with arrhythmia obtained by TPAT technique, there were 6 cases of grade 1 (46%), 6 cases of grade 2 (46%) and 1 case of grade 3 (8%), while for those obtained by conventional method, there were 4 cases of grade 2 (31%), 8 cases of grade 3 (62%), and 1 case of grade 4 (7%). The IQ of TPAT technique was higher than that of conventional method in patients with arrhythmia on the whole. The CNR values in patients with arrhythmia obtained by TPAT were significantly higher than those by conventional method (10.0±3.4 vs 8.2±2.4, p = 0.009).
In conclusion, the application of optimized TPAT technique, despite of its limited CNR compared to conventional ECG-triggering method, provides better IQ in most patients with arrhythmia, and it also has the advantages of allowing free-breathing scans.