Reproducibility of lung perfusion and ventilation derived using phase-resolved functional lung (PREFUL) MRI imaging

Abstract

Abstract Background There is a growing demand for non-invasive imaging biomarkers of lung function, allowing more sensitive and timely decision making to achieve better patient outcomes. Recently developed Fourier decomposition MRI methods enable quantitative assessment of lung performance. Objectives To assess intra- and inter-observer agreement of lung perfusion and ventilation derived using the free-breathing, contrast agent-free, phase-resolved functional lung (PREFUL) imaging. Methods Fifty patients with acute ST-segment elevation myocardial infarction underwent CMR imaging using a 3 T MRI scanner (MAGNETOM Skyra). Coronal images were acquired during free breathing, covering the entire lung using a phase-resolved functional lung 2D FLASH sequence. The main scan parameters were: temporal resolution = 2.1 ms, number of measurements = 250. The lung perfusion and ventilation maps were generated (Figure 1). The analysis of the images was performed using prototype software (MR Lung Prototype version 2.0) by two independent observers. Analysis was repeated after four weeks to assess intra-observer agreement. Results PREFUL imaging and analysis were fast with a 35 second scan time and up to one minute post-processing time for a complete quantitative assessment. The lung perfusion and ventilation demonstrated excellent intra-observer (ICC 0.928 (0.870 to 0.959) and 0.994 (0.990 to 0.997) for mean perfusion and ventilation respectively) and inter-observer agreement (ICC 0.966 (0.940 to 0.981) and 0.992 (0.985 to 0.996) for mean perfusion and ventilation respectively). Bland-Altman plots demonstrate intra- and inter-observer reproducibility for lung perfusion and ventilation (Figure 2). Conclusion The lung perfusion and ventilation derived using phase-resolved functional lung (PREFUL) MRI imaging are highly reproducible. The Fourier decomposition method may allow fast quantitative evaluation of lung function. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): Hospital of Lithuanian University of Health Sciences