Introduction

The functional information of the pulmonary blood-gas exchange could be obtained by the chemical shift saturation recovery (CSSR) method and related xenon uptake model¹. Meanwhile, hyperpolarized xenon diffusion weighted MR imaging (DWI) could offer the information of pulmonary microstructure². However, the blood-gas exchange function and the pulmonary microstructure are generally affected by the lung inflation levels³. Therefore, simultaneously acquiring CSSR data and DWI data in a single breath is necessary to obtain the accurate pulmonary physiological parameters. In this study, we developed a new method, named as Single Breath CSSR-DWI, to quantify the respiratory membrane and the pulmonary microstructure via hyperpolarized ¹²⁹Xe in a single breath.

Methods

The pulse sequence used in this study was shown in Fig. 1 (a total time of 10 s). For the CSSR, three 1.6 ms Gaussian saturation RF pulses with the flip angle of 90° (90° phased, centered at +208 ppm relative to the resonance frequency of xenon in alveoli) separated by gradient spoilers were used to destroy the dissolved xenon. After a series of variable delay that controlled how much xenon gas can enter the lung parenchyma and the blood from the alveolar airspaces, a 1.2 ms Gaussian excitation RF pulse with the flip angle of 90° (0° phased, centered at +208 ppm) was applied for data collecting (TE=0.7 ms, Bandwidth=10 kHz, 512 data points). The process was repeated 21 times in the same breath hold with delay times ranging from 5 ms to 700 ms (a total time of 5.3 s). For the weighted images of DWI, b = 12 s/cm2 (bipolar diffusion gradients, the diffusion time Δ is 3.6 ms, the ramp time τ is 0.3 ms, no gaps between the bipolar gradients, 5 slices) and the compressed sensing (CS) method⁴ was used to accelerate the acquisition (a total time of 4.2 s). After acquiring the DWI data, spectra were collected to calibrate the influence of the RF pulse centered at dissolved xenon resonance frequency on gas xenon (a total time of 0.5 s). All the experiments were performed on a 1.5 T whole-body MRI Scanner (Avanto, Siemens Medical Solutions) using a homebuilt transmit-receive vest RF coil. Enriched xenon (86% ¹²⁹Xe) was polarized using a homebuilt xenon polarizer. One liter of xenon mixture (30% xenon + 70% N2) was inhaled from functional residual capacity before each measurement. Informed consent was obtained from each of 4 healthy young subjects (aged between 23 and 27, HY) and 3 asymptomatic aged non-smoking subjects (aged 63-85 years, HA). The protocol was approved by the institutional review board (IRB). Five parameters were obtained by fitting CSSR data to MOXE: SVRd, d, ξ/d, Hct, t_X. A new parameter SVRd/g was defined from ADC and SVRd to characterize the ‘%-predicted dissolved SVR’. SVRd/g = SVRd/(4/(2(Δ+δ)ADC)^0.5), where δ is the duration time of the diffusion gradients and ADC the mean apparent diffusion coefficient of the whole lung.

Results

For the HY subjects, SVRd/g = 1.24 ± 0.05, ADC = 0.040 ± 0.002 cm²/s. For the HA subjects, SVRd/g = 0.77 ± 0.01, ADC = 0.052 ± 0.005 cm²/s. Figure 2 showed the CSSR curves and the ADC maps of the third slice of HY4 and HA1. Compared to HY subjects, SVRd/g of HA subjects significantly decreased (p<0.05) while ADC significantly increased (p<0.05).

Conclusion

Human pulmonary functional and structure information were successfully obtained in a single breath via hyperpolarized ¹²⁹Xe. The parameters obtained from the MOXE model agreed well with the previous study⁵, and the changes of ADC were in agreement with the results reported by Kaushik⁶. Compared to the healthy young subjects, the ‘%-predicted dissolved SVR’ of the asymptomatic aged subjects decreases while the size of the pulmonary microstructure increases.

Acknowledgements

We acknowledge the support by the National Natural Science Foundation of China (81227902, 81625011, 81601491) and National Program for Support of Eminent Professionals (National Program for Support of Top-notch Young Professionals).