A reproducible extended ex-vivo normothermic machine liver perfusion protocol utilising improved nutrition and targeted vascular flows

Abstract

Background: Normothermic machine perfusion of donor livers has become standard practice in the field of transplantation, allowing the assessment of organs and safe extension of preservation times. Alongside its clinical uses, there has been expanding interest in extended normothermic machine perfusion (eNMP) of livers as a potential vehicle for medical research. Reproducible extended normothermic machine perfusion has remained elusive due to its increased complexity and monitoring requirements. We set out to develop a reproducible protocol for the extended normothermic machine perfusion of whole human livers.

Methods: Human livers declined for transplantation were perfused using a blood-based perfusate at 36 °C using the Liver Assist device (XVIVO, Sweden), with continuous veno-venous haemofiltration in-parallel. We developed the protocol in a stepwise fashion.

Results: Perfusion techniques utilised included: targeted physiological vascular flows, phosphate replacement (to prevent hypophosphataemia), N-acetylcysteine (to prevent methaemoglobin accumulation), and the utilisation of sodium lactate as both a nutritional source and real-time measure of hepatocyte function. All five human livers perfused with the developed protocol showed preserved function with a median perfusion time of 168 h (range 120-184 h), with preserved viability throughout.

Conclusions: Livers can be reproducibly perfused in excess of 120 (range 121-184) hours with evidence of preserved hepatocyte and cholangiocyte function.

Fig. 1. Schematic of the final liver assist set-up.

The Liver Assist perfusion device consists of two separate circuits, with a centrifugal pump for both portal vein (PV) and hepatic artery (HA) circuits. The hepatic veins drain into a reservoir (where the liver sits) which is then drains back into the HA and PV circuits. Our continuous veno-venous haemofiltration (CVVH) unit was added in parallel to the portal vein circuit, with afferent blood taken off prior to the PV oxygenator and efferent blood entering back into the PV oxygenator. Heparin was infused into the CVVH circuit. Drugs were infused at two separate points in the circuit: the HA circuit and post-CVVH PV circuit. Figure created in BioRender.

Fig. 2. Figure showing median vascular flow and resistance in both artery and vein in the two cohorts: Livers A–D (n = 4) displayed in red, and Livers 1–5 (n = 5) displayed in blue.

The source data for Fig. 2 can be found in Supplementary Data document.

Fig. 3. Figure showing median methaemoglobin (%) and range in no NAC cohort (Livers A-C, red, n = 3) and NAC cohort (Livers D & 1-5, blue, n = 6).

The source data for Fig. 3 can be found in Supplementary Data document.

Fig. 4. Graphs showing lactate challenges performed at 24 h intervals in Livers 1-5.

The source data for Fig. 4 can be found in Supplementary Data document.

Fig. 5. Figure showing median and range of electrolytes in Livers A & B (no CVVH cohort, red, n = 2) and Livers C-D & Livers 1-5 (CVVH cohort, blue, n = 5).

The source data for Fig. 5 can be found in Supplementary Data document.

Fig. 6. Figure displaying bile pH, bile glucose, delta glucose (perfusate glucose – bile glucose), and cumulative bile production during Livers 1-5.

Each Liver is represented by a different colour (Liver 1, red; Liver 2, green; Liver 3, yellow; Liver 4, blue; Liver 5, black). The source data for Fig. 6 can be found in Supplementary Data document.

Fig. 7. H&E Staining for Livers 1-5.

From top to bottom, figures show H&E staining for Livers 1-5 before perfusion (a) on the left side, and at the end of perfusion (b) on the right side. A 50 μm scale bar is included in the bottom right of each figure.