Purpose

The ability to monitor and investigate isolated organs outside the body is becoming increasingly relevant. This is due to several factors, including a need for novel transplantation applications as well as the desire to perform ever more detailed investigations into metabolism and function. Here we present the use of an in-house developed MRI compatible perfusion system, capable of investigating hemodynamic and metabolic function in ex-vivo kidney porcine models through the use of hyperpolarized and conventional MRI.

Materials and methods

One kidney (125±16 g) and approximately 1.2 L heparinized whole blood was retrieved from four fully anesthetized female pigs (40 kg body weight), followed by the termination of the animal. The kidney was flushed with cold Ringer-acetate (Fresenius Kabi, Bad Homburg, GE) and cooled down to 5 °C. The renal artery and ureter was cannulated, and the kidney was connected to the perfusion system, see Figure 1 and 2. The perfusion system is comprised of a BioMedicus Medtronic Bio Console 540 centrifugal pump (Medtronic, Minneapolis, MN, US) to maintain physiological flow at approximately 170 mL/min. The perfusate is heated to 37 °C and oxygenated using a Medos Hilite 1000 neonatal oxygenator (Xenios, Heilbronn, GE) and a water heater/pump, see Figure 2. Temperature, flow and pressure sensors mounted on the perfusion lines allow for continuous monitoring. Glucose (Fresenius Kabi, Bad Homburg, GE), amino acids (Vaminolac, Fresenius Kabi, Bad Homburg, GE) and insulin (Humulin, Eli Lilly Demark A/S, Herlev, DK) was infused continually to keep blood gas parameters in the physiological range¹. Vasodilator (Veraloc, Orion Pharma, Copenhagen, DK) was infused to ease the perfusion, and the produced urine was collected in a separate bag to avoid contamination of the blood supply. Physiological and hemodynamic parameters was monitored throughout the perfusion. The perfused kidney is placed in the bore of a 3.0 T Signa HDx MRI scanner (GE Healthcare) equipped with proton and carbon-13 imaging capability. Intra-renal anatomy was assessed using T₁ weighted FLAIR and T₂ weighted PROPELLER imaging sequences with the following parameters. T₁ FLAIR: TR/TE/TI = 4.4 s/24.2 ms/1.6 s, flip angle = 111°, FOV/matrix = 200×200 mm²/256×256. T₂ PROPELLER: TR/TE = 7.4 s/94.1 ms, flip angle = 142°, FOV/matrix = 200×200 mm²/256×256. Both sequences were used to acquire 15 long axis slices of 4 mm slice thickness in 2 and 4 averages respectively, see Figure 3. Metabolic ¹³C imaging was performed following a 9 mL injection of hyperpolarized [1-¹³C]pyruvate, using a spectral spatial (SPSP) imaging sequence with parameters: TR/TE = 0.5 s/1 ms, FOV/matrix = 120×120 mm²/128×128, one long axis slice with 40 mm slice thickness and a 90° or 8° flip angle on lactate/bicarbonate/alanine or pyruvate, respectively. The effective TR for pyruvate and its metabolites was 1 and 3 seconds respectively, with three pyruvate and one lactate/bicarbonate/alanine acquisition every 3 seconds. ¹H DCE MRI was acquired following an injection of 0.3 mL Dotarem (279.3 mg/mL) using a 3D fast gradient echo sequence with parameters: TR/TE = 1 s/1.7 ms, flip angle = 12°, FOV/matrix = 240×240×240 mm³/256×256. Region of interest (ROI) analysis is performed in Matlab (MathWorks, Natic, MA, US) using a custom segmentation method², where each whole kidney ROI was divided into 10 equidistantly spaced segment layers, see Figure 3.

Results

Ex-vivo renal perfusion with accurate control of physiological parameters was verified with ¹H MRI and [1-¹³C]pyruvate MRI. Our preliminary results from these investigations display differences in intra renal heterogeneity, see Figure 4. The renal cortex shows a predominant lactate production, while alanine production is mostly confined to the renal medullary region.

Discussion and conclusion

Improved understanding of the role of deranged renal metabolism in the donor graft prior to transplantation and during storage has the potential to increase utilization of marginal organs, and thereby combat organ shortage. This follows from the heightened potential of enhanced therapeutic strategies, e.g. in the form of storage and pharmaceutical interventions, with the effect of preventing graft degradation or modulating the graft viability. With further analysis we hope to clarify the observations presented here, and look for potential correlations between the metabolic distribution, renal function and the outcome following transplantation. This study demonstrates the ability to monitor ex-vivo graft metabolism and function in a large animal model, resembling human renal physiology

Acknowledgements

No acknowledgement found.