Introduction

Ex-vivo perfusion systems for pulmonary allografts prior to transplantation are a promising new tool for conditioning of marginal organs. However, reliable parameters for graft function within the system are missing. Washout kinetics of oxygen and fluorinated gases have been introduced to quantify regional lung ventilation and therefore to give local information about lung function. In this proof of concept study the feasibility of using oxygen as well as perfluoropropane washout time in an ex-vivo lung perfusion system was tested. Furthermore, the ability of detecting pathologic changes due to ischemic lung injury was tested for both methods.

Methods

This study was approved by the federal animal welfare committee. Twelve female pigs of German Landrace were anesthetized. Pulmonary and bronchial arteries of the left lung were clamped to induce ischemic injury (76 to 199 min, median 170 min), right lungs remained perfused as controls. After cold flush with Perfadex solution (XVIVO Perfusion AB), lungs were installed in a modified ex-vivo lung perfusion system container (OCS lung system, TransMedics; Figure 1) using full blood and Steen solution (XVIVO Perfusion AB) with a perfusate flow of 1.5 l/min and a respiration rate of 10/min with 500 ml tidal volume. After allowing a minimum wash-in time of 5 minutes for both 100% oxygen or a mixture of perfluoropropane and oxygen the respirator was switched to normal room air. Magnetic resonance imaging was performed on a 1.5 tesla scanner (Avanto, Siemens) using a standard spine coil in combination with body matrices for proton imaging and a dedicated ¹⁹F-coil (Rapid Biomedical) tuned to 59.9 MHz (Helmholtz transmit coil and a 16 channel receive coil) for imaging of perfluoropropane. Oxygen washout dynamics were measured using an Inversion-Recovery HASTE sequence (TI=1100 ms, TE=11 ms, TR=6000ms, flip angle=180°, FOV=45 x 45 cm², matrix size=128 x 128, slice thickness=15 mm) on a central coronal slice. ¹⁹F wash-out dynamics were measured with a 3D spoiled gradient echo sequence (TE=5.1 ms, TR=12 ms, flip angle=50°, FOV=50 x 50 cm², matrix size=64 x 64, slice thickness=40 mm, GRAPPA imaging with acceleration factor of 2 in A/P direction). One central coronal slice corresponding to the oxygen washout image was analyzed. Washout times win were estimated voxelwise by mono-exponential fitting of the temporal signal. At the end of the EVLP run samples from both lungs were collected and weight. After drying in an oven for 3 days samples were weighed again to assess the wet / dry weight ratio as a marker for edema formation. Statistical analysis was performed using the Wilcoxon matched pairs test. Results are given as median [interquartile range].

Results

No ventilation defect was seen on the ¹⁹F images prior to washout (Figure 2a). The washout time for ¹⁹F did not differ between lungs after warm ischemia (67.9 s [54.6 s; 84.6 s]) compared to normal lungs (69.8 s [54.9 s; 79.3 s]) (p=0.97; Figure 2b). The oxygen washout time differed significantly between lungs with and without ischemia (p=0.019). The ischemic lungs showed a prolonged washout time of 44.6 s [43.1 s; 54.7 s] compared to 37.3s [31.2 s; 47.2 s] at an equal breathing rate of 10/min (Figure 2c). Wet / dry ratio of ischemic lungs was significantly higher than in normal control lungs (4.76 [4.55 ; 5.45] vs 4.53 [4.16 ; 4.74], p=.024).

Discussion

Ischemia-reperfusion injury is known to induce an increase in pulmonary vascular permeability and therefore causing pulmonary edema^1,2. The observed wet / dry ratio difference shows the induced edema in ischemic lungs. This edema leads to a thickening of the alveolar walls and as a consequence, to a reduced oxygen diffusion over the basal membrane. Therefore, the observed prolonged washout times for oxygen may reflect this pathophysiologic pattern. As inert fluorinated gases have a low solubility in blood³, fluorine washout primarily reflects the ventilation of the lung, whereas oxygen washout time is a combination of ventilation, perfusion and oxygen diffusion capacity. Therefore, the combination of fluorine with oxygen washout seems to be a promising new tool in pulmonary magnetic resonance imaging, as this might deliver a marker for oxygen diffusion via the membrane.

Conclusion

The combination of oxygen and fluorine washout kinetics might be a promising new marker for graft function and should be evaluated in further studies comparing this parameter with graft function and clinical outcome following transplantation. Furthermore, this combination should be evaluated as a marker for oxygen diffusion in other pulmonary diseases.

Acknowledgements

No acknowledgement found.