Article Text

Download PDFPDF

159 Comparing image quality and reporting times for scar identification between 2d and 3d sequences in cardiac magnetic resonance imaging
  1. Nikesh Jathanna1,
  2. Kevin Strachan2,
  3. Hazlyna Kamaruddin2,
  4. Bara Erhayiem2,
  5. Shahnaz Jamil-Copley2
  1. 1Nottingham University Hospitals NHS Trust, Trent Cardiac Centre, Hucknall Road, Nottingham, NGM NG5 1PB, United Kingdom
  2. 2Nottingham University Hospitals NHS Trust


Introduction Compared to standard 2D imaging (2D SAX), 3D free-breathing whole heart imaging (3D-WHI) allows for the acquisition of smaller voxel size. The subsequent increased spatial resolution could provide significant benefit in research and cardiology electrophysiology procedures especially when combined with late gadolinium enhancement (LGE). However, the perception that increased series size may result in longer reporting times and the possibility of worsened image quality have prevented this modality from being adopted into routine clinical practice.We aimed to investigate if the reporting time for 3D-WHI was longer compared to standard 2D SAX imaging and if a difference in image quality was present.Methods15 consecutive cases of clinically indicated myocardial viability scans with same-sitting 2D SAX and 3D-WHI were duplicated with one LGE modality removed resulting in 15 pairs of 2D SAX only and 3D-WHI only cases. LGE visual reporting of paired cases was undertaken in 2 sittings, 3 months apart, by a single level 3 trained, Cardiac imaging consultant and analysed using Medis (Medical Imaging System, Leiden, The Netherlands) according to SCMR recommendations. The interpreter had access to all other non-LGE images obtained during the initial acquisition.Interpretation time was recorded from the time of series opening to report completion. Additional image quality assessment was undertaken quantitatively using features previously described by Klinke et al. (0–19, good to bad respectively) and qualitatively on a likert scale 0–2 (uninterpretable, poor/fair, good respectively) taking diagnostic utility and technical quality into account. Interpretation Time and quantitative image quality was compared with the Wilcoxon signed-rank test. Qualitative scores were compared narratively.

Results Of the 15 cases included, aetiologies were 10 ischaemic, 4 non-ischaemic and 1 normal case with LGE present in all but the normal case. Mean number of slices for the 2D SAX and 3D-WHI were 12 and 115 respectively.There was no difference in mean time taken to interpret 2D SAX and 3D-WHI LGE images (197.9 vs. 175.2 seconds, p=0.609). No statistical difference between 2D SAX and 3D-WHI LGE was seen in mean image quality by quantitative analysis (1.5 vs 1.5, p=0.666). Qualitatively, there was no difference in the number of ‘good’ images in either modality (60% vs 60%).

Abstract 159 Table 1

Comparison image quality and reporting times of late gadolinium enhancement by sequence - 2D SAX and 3D Whole Heart Imaging (3D-WHI).

Conclusion Despite the larger number of slices in the 3D-WHI stacks, there was no difference detected in LGE reporting times or in image quality for routine clinical reporting. This suggests the clinical implications of adopting 3D-WHI into routine clinical scanning may not adversely impact on time or resources. This is a small single operator study which did not assess the impact of sequence on reporting times of quantitative scar assessment which needs to be acknowledged.

Conflict of Interest None

  • Cardiac Magnetic Resonance Imaging
  • Fibrosis
  • late gadolinium enhancement

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.