Reproducibility of fractional ventilation derived by Fourier decomposition after adjusting for tidal volume with and without an MRI compatible spirometer

Abstract

Purpose: To reduce the influence of tidal volume on fractional ventilation (FV) derived by Fourier decomposition (FD). Methods: Twelve volunteers were examined on a 1.5 Tesla scanner. Spoiled gradient echo imaging of coronal and sagittal slices of the lung were performed. The tidal volume variations between different acquisitions were studied by reproducibility and repeatability measurements. To adjust the FV derived by FD for tidal volume differences between the measurements, during all acquisitions, the lung volume changes were measured by a spirometer and used to calculate a global FV parameter. As an alternative, using the FD data, the lung area changes were calculated and used for the adjustment. Results: Reproducibility analysis of unadjusted coronal FV showed a determination coefficient of R2 = 71% and an intraclass correlation coefficient of ICC = 93%. Differences in the measurements could be ascribed to different tidal volumes. Area adjusted values exhibited an increased R2 of 84% and a higher ICC of 97%. For the coronal middle slice/sagittal slices in free breathing, the inter-volunteer coefficient of variation was reduced from 0.23/0.28 (unadjusted) to 0.16/0.20 (spirometer) or 0.12/0.13 (area). Conclusion: The calculation of lung area changes is sufficient to increase the reproducibility of FV in a volunteer cohort avoiding the need for an MRI compatible spirometer. Magn Reson Med 76:1542–1550, 2016. © 2015 International Society for Magnetic Resonance in Medicine.