Introduction

MR imaging and spectroscopy of ¹²⁹Xe dissolved in pulmonary red blood cells (RBCs) and parenchymal tissue/plasma (TP) is of interest for investigating gas exchange limitation in interstitial lung diseases.^1,2 Key to the success of dissolved-phase ¹²⁹Xe spectroscopy and imaging is the ability to reliably measure the signal amplitude and chemical shift of ¹²⁹Xe in red blood cells (RBCs) within an MR acquisition. Both the signal amplitude and chemical shift of ¹²⁹Xe dissolved in RBCs have been shown to be modulated by cardiac pulsation.³ This has implications for accurate measurement of RBC oxygenation from the ¹²⁹Xe chemical shift, Dixon-based dissolved ¹²⁹Xe imaging, which assumes a fixed chemical shift difference between the target RBC and TP compartments, and for ¹²⁹Xe MRS techniques which depend on robust quantification of the RBC and TP signals as biomarkers for gas exchange efficiency.⁴ In this study, we demonstrate the use of cardiac-gated pulmonary spectroscopy to mitigate previously observed modulations in the ¹²⁹Xe-RBC signal and chemical shift.

Methods

Hyperpolarized ¹²⁹Xe pulmonary MRS was performed on a healthy male volunteer on a clinical 3 T (Philips, Achieva) scanner with a quadrature flex T-R coil (CMRS). All scans were performed at breath-hold apnoea after inhalation (from functional residual capacity) of 500 mL of hyperpolarized ¹²⁹Xe balanced with 500 mL of medical-grade N₂. Two sequences were implemented with the following parameters: hard pulses centered on the ¹²⁹Xe-RBC resonance, flip angle = 90°, spectral points = 256, bandwidth = 3 kHz. The first sequence was performed with a fixed RF pulse delay of TR = 500 ms. In the other sequence, RF pulses were ECG triggered at end ventricular diastole. The ¹²⁹Xe signal in RBCs and TP was determined by performing zeroth-order phasing on the associated resonances followed by numerical integration between the y-axis zero crossings on either side of the resonant peaks as it was not possible to obtain a good fit due to the non-Lorentzian behavior of the RBC resonance. The ¹²⁹Xe-RBC chemical shift was determined using 0 ppm as a reference for ¹²⁹Xe-TP.

Results and Discussion

The ¹²⁹Xe-RBC and ¹²⁹Xe-TP signals were observed to decay during both MR acquisitions with a time constant characteristic of ¹²⁹Xe-gas in the alveolar reservoir (Fig. 1 (a) to (d)); after each saturation pulse, the dissolved compartments are replenished with ¹²⁹Xe gas from the alveoli with polarization that is undergoing T₁ decay (T₁ = 18.8 s). The mean RBC:TP peak ratios and ¹²⁹Xe-RBC chemical shifts were calculated to be 0.84 ± 0.02 and 0.67 ± 0.04 and 21.5 ± 0.2 ppm and 21.4 ± 0.04 ppm using cardiac-gated and standard pulse sequences, respectively – see Figure 1 and Table 1. The coefficient of variation (CV), defined as the standard deviation divided by the mean value, is a factor of 2.5 higher for the RBC:TP ratio and a factor of 2 higher for the ¹²⁹Xe-RBC shift with the non-gated sequence when compared to the cardiac-gated sequence. The lower CV values observed using cardiac gating can be attributed to sampling the ¹²⁹Xe-RBC signal at a fixed period within the cardiac cycle where there should be a fixed and repeatable influx of oxygen desaturated blood entering the alveolar capillary bed from the right ventricle when compared to the non-gated pulse sequence where the blood flux and RBC oxygenation will be less consistent due to incoherent signal sampling with respect to the cardiac cycle. The larger RBC:TP ratio calculated with the cardiac gated sequence is likely due to more blood volume being sampled downstream of the alveolar-capillary gas exchange surface, as the patient’s heart rate was ~50 beats/minute giving an effective TR = 1200 ms when compared to TR = 500 ms with the non-gated sequence; it has being shown previously⁵ that the RBC:TP ratio increases with TR, plateauing at TR~3000 ms.

Conclusion

¹²⁹Xe cardiac gated pulmonary MR sequences provide more reliable measurements of the RBC:TP ratio and the ¹²⁹Xe-RBC shift when compared to sequences that acquire the ¹²⁹Xe signal at random phases of the cardiac cycle. This has implications for Dixon-based imaging dissolved ¹²⁹Xe MRI and ¹²⁹Xe MRS techniques where accurate values of ¹²⁹Xe-RBC shift and RBC:TP ratio are needed.

Acknowledgements

This work was supported by NIHR grant NIHR-RP-R3-12-027 and MRC grant MR/M008894/1. The views expressed in this work are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health