Acute liver failure is marked by the rapid deterioration of liver function in a previously well patient over period of days to weeks. Though relatively rare, it is associated with high morbidity and mortality. This makes it a challenging disease to study clinically, necessitating reliance on preclinical models as means to explore pathophysiology and novel therapies. Preclinical models of acute liver failure are artificial by nature, and generally fall into one of three categories: surgical, pharmacologic or immunogenic. This article reviews preclinical models of acute liver failure and considers their relevance in modeling clinical disease.

Normothermic machine perfusion (NMP) has emerged as a valuable tool in the preservation of liver allografts before transplantation. Randomized trials have shown that replacing static cold storage (SCS) with NMP reduces allograft injury and improves graft utilization. The University of Alberta's liver transplant program was one of the early adopters of NMP in North America. Herein, we describe our seven-year experience applying NMP to extend preservation time in liver transplantation using a 'back-to-base' approach. From 2015 to 2021, 79 livers were transplanted following NMP compared to 386 after SCS only. NMP livers were preserved for a median time of 847min compared to 288.5min in the SCS cohort (p<0.0001). Despite this, we observed significantly improved 30-day graft survival (p=0.030), though no differences in long term patient survival, major complications or biliary or vascular complications. We also found that while SCS time was strongly associated with increased graft failure at one year in the SCS cohort (p=0.006), there was no such association amongst NMP livers (p=0.171). Our experience suggests that NMP can safely extend the total preservation time of liver allografts without increasing complications.

The liver is susceptible to ischemia-reperfusion injury (IRI) during hepatic surgery, when the vessels are compressed to control bleeding, or liver transplantation, when there is an obligate period of ischemia. The hallmark of IRI comprises mitochondrial dysfunction, which generates reactive oxygen species, and cell death through necrosis or apoptosis. Cyclosporine (CsA), which is a well-known immunosuppressive agent that inhibits calcineurin, has the additional effect of inhibiting the mitochondrial permeability transition pore (mPTP), thereby, preventing mitochondrial swelling and injury. NIM-811, which is the nonimmunosuppressive analog of CsA, has a similar effect on the mPTP. In this study, we tested the effect of both agents on mitigating warm hepatic IRI in a murine model. Before ischemic insult, the mice were administered with intraperitoneal normal saline (control); CsA at 2.5, 10, or 25 mg/kg; or NIM-811 at 10 mg/kg. Thereafter, the mice were subjected to partial warm hepatic ischemia by selective pedicle clamping for 60 min, followed by 6 h of recovery after reperfusion. Serum alanine transaminase (ALT) was measured, and the liver tissue was examined histologically for the presence of apoptosis and the levels of inflammatory cytokines. Compared with the control mice, the mice treated with 10 and 25 mg/kg of CsA and NIM-811 had significantly lower ALT levels (P < 0.001, 0.007, and 0.031, respectively). Moreover, the liver tissue showed reduced histological injury scores after treatment with CsA at 2.5, 10, and 25 mg/kg and NIM-811 (P = 0.041, <0.001, 0.003, and 0.043, respectively) and significant decrease in apoptosis after treatment with CsA at all doses (P = 0.012, 0.007, and <0.001, respectively). Levels of the pro-inflammatory cytokines, particularly interleukin (IL)-1β, IL-2, IL-4, IL-10, and keratinocyte chemoattractant/human growth-regulated oncogene significantly decreased in the mice treated with the highest dose of CsA (25 mg/kg) than those in the control mice. Premedication with CsA or NIM-811 mitigated hepatic IRI in mice, as evidenced by the decreased ALT and reduced injury on histology. These results have potential implications on mitigating IRI during liver transplantation and resection.

Background.: Limited information is available regarding outcomes of islet cell isolation (ICI) and transplantation (ITx) using medical assistance in dying (MAiD) donors. We aimed to assess the feasibility and outcomes of ICI and ITx in MAiD donors. Methods.: ICI and ITx from MAiD were compared with donation after circulatory death (DCD) type III between 2016 and 2023. Differences of isolated islet equivalents (IEQs), numeric viability and other quantitative in vitro metabolic measures were assessed. Results.: Overall, 81 ICIs were available of whom 34 (42%) and 47 (58%) from MAiD and DCD-III, respectively. There were no differences of pancreas and digested tissue weight and islets viability among the 2 groups; however, cold ischemic time was longer in MAiD (11.5 versus 9.1 h; P = 0.021). The IEQ (P < 0.001) and percent trapped (P < 0.001) were higher in the DCD-III; however, MAiD islets demonstrated a higher purity (P = 0.020). Overall, 15 ITx were performed of whom 3 (8.8%) and 12 (25.5%) from MAiD and DCD-III, respectively (P = 0.056). Patients had a median fasting C-peptide of 0.51 ng/mL (interquartile range, 0.30–0.76 nmol/L), with no differences between groups (MAiD = 0.52 versus DCD-III = 0.51; P = 0.718). The median HbA1c was 6.2% (interquartile range, 5.7%–7%) (MAiD = 6.3% versus DCD-III = 6.1%; P = 0.815) and BETA2 scores (MAiD = 7.4 versus DCD-III = 12.8; P = 0.229) did not differ. Conclusions.: ICI from MAiD donor pancreas may be successfully transplanted with comparable outcomes to DCD-III and may be used for research. These results justify additional efforts to consider MAiD as another valuable source of grafts for ITx. Further multicenter studies and larger clinical experience are needed to validate our findings.