Introduction

Early detection and adequate phenotyping are highly desirable for diagnosis and treatment of COPD^1,2. Therefore, MRI of inhaled fluorinated gas provides a cost-effective and non-invasive tool for direct ventilation measurement in the human lung³. Examining the gas diffusion behavior allows for quantification of pulmonary alveolar surface-to-volume ratio (S/V) which quantifies lung microstructure for assessing e.g. emphysematous changes in COPD^4,5.

Methods

In this study, 20 healthy volunteers and 22 patients with COPD (GOLD stage I – III) were examined on a 1.5 T scanner. The study participants inhaled a mixture of 79% C₃F₈ and 21% O₂ for nine breath-holds with a fixation of the inhaled gas volume to 1/8 inspiratory vital capacity. During the first eight breath-holds (duration 6 seconds each), ventilation imaging was performed using a 3D gradient echo pulse sequence with golden-angle stack-of-stars k-space encoding. During the ninth breath-hold, four images were acquired for S/V mapping in a single breath-hold (duration 14 seconds). Three images were diffusion weighted using additional 3D trapezoidal gradients with a fixed b-value (b = 12.59 s/cm²), fixed gradient pulse length (δ = 1 ms), and varying diffusion times (Δ = 6 ms, 4 ms, 2 ms) and gradient strengths, respectively (|G| = [18.7, 23.3, 34.6] 10^-5 T/cm). The first image was acquired without diffusion weighting as a reference.
All patients underwent spirometry and ¹⁹F MR imaging on the same day and a second ¹⁹F MR scan after maximum 7 days for repeatability measurement. Hyperpolarized ¹²⁹Xe diffusion-weighted MR imaging was performed in 11 COPD patients prior to the first ¹⁹F imaging session using a stack-of-spirals sequence as previously described ⁶. Detailed information on the imaging parameters is provided in table 1.
¹⁹F ventilation images were reconstructed using the parallel imaging and compressed sensing algorithm of the BART toolbox⁷. In order to counter the low SNR, total variation and directional total variation to the eighth ventilation image were regularized when reconstructing the diffusion weighted images.^8,9
To obtain S/V maps, the diffusion weighted images (S) were divided by the unweighted image (S₀) followed by voxel-wise fitting to the first-order approximation of the model function for apparent diffusion in a porous medium $$ \frac{S}{S_0} = \exp{\left(-b\cdot D_{app}\right)} = \exp{\left( -b\cdot D_0 \left( 1 - \frac{\alpha}{3}\frac{S}{V}\sqrt{D_0\Delta}\right)\right)}, $$ where α is a prefactor compensating for the deviation from the narrow-pulse approximation^10,11. Since the free diffusion D₀ depends on the ratio of C₃F₈ to O₂ in the specific voxel, its value was also determined within the fit procedure¹².

Results

Fluorinated gas imaging procedures and maneuvers were well tolerated and no adverse events were reported. Results are stated as medians over the whole lung with first and third quartile, unless otherwise stated.
S/V maps from exemplary patients can be found in figure 1, S/V values in different stages of COPD are shown in figure 2.
The median S/V was 164 cm^-1 (160 cm^-1 – 184 cm^-1) in healthy volunteers, 157 cm^-1 (152 cm^-1 – 164 cm^-1) in COPD patients with GOLD stage I, 132 cm^-1 (105 cm^-1 – 151 cm^-1) in COPD patients with GOLD stage II, and 81 cm^-1 (64 cm^-1 – 116 cm^-1) in COPD patients with GOLD stage III. Taken together, patients with COPD had a median S/V of 126 cm^-1 (87 cm^-1 – 144 cm^-1).
Significant differences could be found between healthy volunteers and patients with GOLD II (P = .007), between patients with GOLD I and GOLD III (P = .025), and between healthy volunteers and patients with GOLD III (P < .0001). A significant difference in S/V was measured between healthy volunteers and COPD patients (P < .0001).
Bland-Altman plots comparing measurements on the first and the second scan day are shown in figure 3. No significant differences for S/V values between first and second scan could be found in any of the patient groups. The Bland-Altman plots show a mean bias of 0.8 cm^-1 and the 95% limits of agreement of ± 45 cm^-1 for healthy volunteers as well as a mean bias of 9 cm^-1 and 95% limits of agreement of ± 37 cm^-1 for COPD patients. The median coefficient of variation between first and second measurement for healthy volunteers was 4.3% and for COPD patients 6.7%.
For 11 COPD patients, a strong linear correlation between median S/V values measured with ¹⁹F and ¹²⁹Xe could be found (r = 0.85, P = .001). Bland-Altman plot analysis shows a non-significant bias of -11 cm^-1 (P = .145) and the 95% limits of agreement of ± 47 cm^-1. Correlation and Bland-Altman plots are shown in figure 4.
A moderate linear correlation between ¹⁹F S/V values and the predicted forced expiratory volume in one second (FEV₁) could be found in all examined subjects (r = 0.68, P < .0001). A correlation plot is shown in figure 5.

Discussion and Conclusion

Although the sequence design was rather simple, this study shows the general feasibility and the good repeatability of in vivo S/V mapping with ¹⁹F diffusion MRI. The provided values seem to be reasonable using the proposed theoretical model and sequence design when compared to measurements from ¹²⁹Xe and spirometry as well as literature values for histological studies and CT imaging^13–15.

Acknowledgements

This work was supported by the German Center for Lung Research (DZL) and the Fritz Behrens Foundation. The authors thank Frank Schröder, Melanie Pfeifer, and Cheng-Kai Huang for experimental support, as well as E. Tobias Krause for in-depth advice on statistical evaluation.

References

1. Laucho-Contreras ME, Cohen-Todd M. Early diagnosis of COPD: myth or a true perspective. European Respiratory Review. 2020;29(158):200131. doi:10.1183/16000617.0131-2020
2. Lange P, Halpin DM, O’Donnell DE, MacNee W. Diagnosis, assessment, and phenotyping of COPD: beyond FEV₁. International journal of chronic obstructive pulmonary disease. 2016;11 Spec Is(Spec Iss):3–12. doi:10.2147/COPD.S85976
3. Couch MJ, Ball IK, Li T, Fox MS, Littlefield SL, Biman B, Albert MS. Pulmonary Ultrashort Echo Time 19F MR Imaging with Inhaled Fluorinated Gas Mixtures in Healthy Volunteers: Feasibility. Radiology. 2013;269(3):903–909. doi:10.1148/radiol.13130609
4. Itoh H, Nishino M, Hatabu H. Architecture of the Lung: Morphology and Function. Journal of Thoracic Imaging. 2004;19(4).
5. Thurlbeck WM. Internal surface area and other measurements in emphysema. Thorax. 1967;22(6):483–496. doi:10.1136/thx.22.6.483
6. Kern AL, Gutberlet M, Moher Alsady T, Welte T, Wacker F, Hohlfeld JM, Vogel-Claussen J. Investigating short-time diffusion of hyperpolarized 129Xe in lung air spaces and tissue: A feasibility study in chronic obstructive pulmonary disease patients. Magnetic Resonance in Medicine. 2020;84(4):2133–2146. doi:https://doi.org/10.1002/mrm.28264
7. Uecker M, Ong F, Tamir JI, Bahri D, Virtue P, Cheng JY, Lustig M. Berkeley advanced reconstruction toolbox. Proc. Intl. Soc. Mag. Reson. Med. 2015;23:2486. doi:10.1002/mrm.25176
8. Obert AJ, Gutberlet M, Kern AL, Kaireit TF, Grimm R, Wacker F, Vogel-Claussen J. 1H-guided reconstruction of 19F gas MRI in COPD patients. Magnetic Resonance in Medicine. 2020. doi:10.1002/mrm.28209
9. Ehrhardt MJ, Betcke MM. Multicontrast MRI Reconstruction with Structure-Guided Total Variation. SIAM Journal on Imaging Sciences. 2016;9(3):1084–1106. doi:10.1137/15M1047325
10. Mitra PP, Sen PN, Schwartz LM. Short-time behavior of the diffusion coefficient as a geometrical probe of porous media. Physical Review B. 1993;47(14):8565–8574. doi:10.1103/PhysRevB.47.8565
11. Miller GW, Carl M, Mata JF, Cates GD, Mugler JP. Simulations of Short-Time Diffusivity in Lung Airspaces and Implications for S/V Measurements Using Hyperpolarized-Gas MRI. IEEE Transactions on Medical Imaging. 2007;26(11):1456–1463. doi:10.1109/TMI.2007.903192
12. Chang Y V., Conradi MS. Relaxation and diffusion of perfluorocarbon gas mixtures with oxygen for lung MRI. Journal of Magnetic Resonance. 2006;181(2):191–198. doi:10.1016/j.jmr.2006.04.003
13. Coxson HO, Rogers RM, Whittall KP, D’Yachkova Y, Paré PD, Sciurba FC, Hogg JC. A Quantification of the Lung Surface Area in Emphysema Using Computed Tomography. American Journal of Respiratory and Critical Care Medicine. 1999;159(3):851–856. doi:10.1164/ajrccm.159.3.9805067
14. Yablonskiy DA, Sukstanskii AL, Woods JC, Gierada DS, Quirk JD, Hogg JC, Cooper JD, Conradi MS. Quantification of lung microstructure with hyperpolarized 3He diffusion MRI. Journal of Applied Physiology. 2009;107(4):1258–1265. doi:10.1152/japplphysiol.00386.2009
15. West JB. Respiratory physiology: the essentials. Lippincott Williams & Wilkins; 2012.