Effects of slice orientation on reproducibility of sequential assessment of right ventricular volumes and ejection fraction: short-axis vs transverse SSFP cine cardiovascular magnetic resonance

Luigia D'Errico; Mariana M Lamacie; Laura Jimenez Juan; Djeven Deva; Rachel M Wald; Sebastian Ley; Kate Hanneman; Paaladinesh Thavendiranathan; Bernd J Wintersperger

doi:10.1186/s12968-016-0282-x

Effects of slice orientation on reproducibility of sequential assessment of right ventricular volumes and ejection fraction: short-axis vs transverse SSFP cine cardiovascular magnetic resonance

J Cardiovasc Magn Reson. 2016 Sep 22;18(1):60. doi: 10.1186/s12968-016-0282-x.

Authors

Luigia D'Errico^{1

2}, Mariana M Lamacie¹, Laura Jimenez Juan^{2

3}, Djeven Deva^{2

4}, Rachel M Wald^{1

2

5}, Sebastian Ley^{1

2}, Kate Hanneman^{1

2}, Paaladinesh Thavendiranathan^{1

5}, Bernd J Wintersperger^{6

7

8}

Affiliations

¹ Department of Medical Imaging, Peter Munk Cardiac Centre, University Health Network, Toronto, Canada.
² Department of Medical Imaging, University of Toronto, Toronto, Canada.
³ Department of Medical Imaging, Sunnybrook Health Science Centre, Toronto, Canada.
⁴ Department of Medical Imaging, St. Michaels Hospital, Toronto, Canada.
⁵ Department of Medicine, University of Toronto, Toronto, Canada.
⁶ Department of Medical Imaging, Peter Munk Cardiac Centre, University Health Network, Toronto, Canada. bernd.wintersperger@uhn.ca.
⁷ Department of Medical Imaging, University of Toronto, Toronto, Canada. bernd.wintersperger@uhn.ca.
⁸ Department of Medical Imaging, Toronto General Hospital, 1 PMB-273, 585 University Avenue, M5G 2N2, Toronto, ON, Canada. bernd.wintersperger@uhn.ca.

Abstract

Background: Test-retest reproducibility is of utmost importance in follow-up of right ventricular (RV) volumes and function; optimal slice orientation though is not yet known. We compared test-retest reproducibility and intra-/inter-observer variability of right ventricular (RV) volumes and function assessed with short-axis and transverse cardiovascular magnetic resonance (CMR).

Methods: Eighteen volunteers underwent cine CMR for RV assessment obtaining ventricular coverage in short-axis and transverse slice orientation. Additional 2D phase contrast flow imaging of the main pulmonary artery (MPA) was performed. After complete repositioning repeat acquisitions were performed. Data sets were contoured by two blinded observers. Statistical analysis included Student's t-test, Bland-Altman plots, intra-class correlation coefficient (ICC) and 2-way ANOVA, SEM and minimal detectable difference calculations.

Results: Heart rates (65.0 ± 7.4 vs. 67.6 ± 9.9 bpm; P = 0.1) and MPA flow (89.8 ± 16.6 vs. 87.2 ± 14.9 mL; P = 0.1) did not differ between imaging sessions. EDV and ESV demonstrated an inter-study bias of 0.4 %[-9.5 %,10.3 %] and 2.1 %[-12.3 %,16.4 %] for short-axis and 1.1 %[-7.3 %,9.4 %] and 0.8 %[-16.0 %,17.6 %] for transverse orientation, respectively. There was no significant interaction between imaging orientation and interstudy reproducibility (p = 0.395-0.824), intra-observer variability (p = 0.726-0.862) or inter-observer variability (p = 0.447-0.706) by 2-way ANOVA. Inter-observer agreement by ICC was greater for short axis versus transverse orientation for all parameters (0.769-0.986 vs. 0.625-0.983, respectively). Minimal detectable differences for short axis and transverse orientations were 10.1 mL/11.5 mL for EDV, 8.3 mL/8.4 mL for ESV and 4.1 % vs. 4.7 % for EF, respectively.

Conclusion: Short-axis and transverse orientation both provide reliable and reproducible measures for follow-up of RV volumes and global function. Therefore, additional transverse SSFP cine CMR may not necessarily be required if performed for the sole purpose of quantitative volumetric RV assessment.

Keywords: Cardiac output; Cardiovascular magnetic resonance; Ventricles.