Evaluation of left ventricular ejection fraction using through-time radial GRAPPA

• Published in: Journal of cardiovascular magnetic resonance 2014-10, Vol.16 (1), p.79-79, Article 79
• Main Authors: Aandal, Gunhild, Nadig, Vidya, Yeh, Victoria, Rajiah, Prabhakar, Jenkins, Trevor, Sattar, Abdus, Griswold, Mark, Gulani, Vikas, Gilkeson, Robert C, Seiberlich, Nicole
• Format: Article
• Language: English
• Subjects: Breath Holding; Cardiac-Gated Imaging Techniques - methods; Ejection; Electrocardiography; Gating and risering; Health aspects; Heart; Heart Rate; Humans; Image Interpretation, Computer-Assisted - methods; Imaging; Linear Models; Magnetic resonance; Magnetic Resonance Imaging - methods; Observer Variation; Ohio; Patients; Physiological aspects; Predictive Value of Tests; Prospective Studies; Real time; Reproducibility of Results; Stroke Volume; Time Factors; Ventricular Function, Left
• ISSN: 1097-6647; 1532-429X; 1532-429X
• DOI: 10.1186/s12968-014-0079-8

The determination of left ventricular ejection fraction using cardiovascular magnetic resonance (CMR) requires a steady cardiac rhythm for electrocardiogram (ECG) gating and multiple breathholds to minimize respiratory motion artifacts, which often leads to scan times of several minutes. The need for gating and breathholding can be eliminated by employing real-time CMR methods such as through-time radial GRAPPA. The aim of this study is to compare left ventricular cardiac functional parameters obtained using current gold-standard breathhold ECG-gated functional scans with non-gated free-breathing real-time imaging using radial GRAPPA, and to determine whether scan time or the occurrence of artifacts are reduced when using this real-time approach. 63 patients were scanned on a 1.5T CMR scanner using both the standard cardiac functional examination with gating and breathholding and the real-time method. Total scan durations were noted. Through-time radial GRAPPA was employed to reconstruct images from the highly accelerated real-time data. The blood volume in the left ventricle was assessed to determine the end systolic volume (ESV), end diastolic volume (EDV), and ejection fraction (EF) for both methods, and images were rated for the presence of artifacts and quality of specific image features by two cardiac readers. Linear regression analysis, Bland-Altman plots and two-sided t-tests were performed to compare the quantitative parameters. A two-sample t-test was performed to compare the scan durations, and a two-sample test of proportion was used to analyze the presence of artifacts. For the reviewers´ ratings the Wilcoxon test for the equality of the scores' distributions was employed. The differences in EF, EDV, and ESV between the gold-standard and real-time methods were not statistically significant (p-values of 0.77, 0.82, and 0.97, respectively). Additionally, the scan time was significantly shorter for the real-time data collection (p