J Attard1,2, D Sneiders2,3, R Laing1,2, Y Boteon1,2, H Mergental2, J Isaac2, D F Mirza2, S Afford1, H Hartog2, D A H Neil1,2,4, M T P R Perera5. 1. Centre for Liver Research, Institute of Immunology and Immunotherapy, University of Birmingham, Birmingham, UK. 2. Liver Unit, Queen Elizabeth Hospital, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK. 3. Erasmus MC Transplant Institute, Department of Surgery, Division of HPB and Transplant Surgery, Erasmus University Medical Center, Rotterdam, the Netherlands. 4. Department of Cellular Pathology, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK. 5. Liver Unit, Queen Elizabeth Hospital, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK. thamara.perera@uhb.nhs.uk.
Abstract
BACKGROUND: Ex vivo normothermic machine liver perfusion (NMLP) involves artificial cannulation of vessels and generation of flow pressures. This could lead to shear stress-induced endothelial damage, predisposing to vascular complications, or improved preservation of donor artery quality. This study aims to assess the spatial donor hepatic artery (HA) endothelial quality downstream of the cannulation site after end-ischaemic NMLP. METHODS: Remnant HA segments from the coeliac trunk up to the gastroduodenal artery branching were obtained after NMLP (n = 15) and after static cold storage (SCS) preservation (n = 15). Specimens were fixed in 10% neutral buffered formalin and sectioned at pre-determined anatomical sites downstream of the coeliac trunk. CD31 immunohistostaining was used to assess endothelial integrity by a 5-point ordinal scale (grade 0: intact endothelial lining, grade 5: complete denudation). Endothelial integrity after SCS was used as a control for the state of the endothelium at commencement of NMP. RESULTS: In the SCS specimens, regardless of the anatomical site, near complete endothelial denudation was present throughout the HA (median scores 4.5-5). After NMLP, significantly less endothelial loss in the distal HA was present compared to SCS grafts (NMLP vs. SCS: median grade 3 vs. 4.5; p = 0.042). In NMLP specimens, near complete endothelial denudation was present at the cannulation site in all cases (median grade: 5), with significantly less loss of the endothelial lining the further from the cannulation site (proximal vs. distal, median grade 5 vs. 3; p = 0.005). CONCLUSION: Loss of endothelial lining throughout the HA after SCS and at the cannulation site after NMLP suggests extensive damage related to surgical handling and preservation injury. Gradual improved endothelial lining along more distal sites of the HA after NMLP indicates potential for re-endothelialisation. The regenerative effect of NMLP on artery quality seems to occur to a greater extent further from the cannulation site. Therefore, arterial cannulation for machine perfusion of liver grafts should ideally be as proximal as possible on the coeliac trunk or aortic patch, while the site of anastomosis should preferentially be attempted distal on the common HA.
BACKGROUND: Ex vivo normothermic machine liver perfusion (NMLP) involves artificial cannulation of vessels and generation of flow pressures. This could lead to shear stress-induced endothelial damage, predisposing to vascular complications, or improved preservation of donor artery quality. This study aims to assess the spatial donor hepatic artery (HA) endothelial quality downstream of the cannulation site after end-ischaemic NMLP. METHODS: Remnant HA segments from the coeliac trunk up to the gastroduodenal artery branching were obtained after NMLP (n = 15) and after static cold storage (SCS) preservation (n = 15). Specimens were fixed in 10% neutral buffered formalin and sectioned at pre-determined anatomical sites downstream of the coeliac trunk. CD31 immunohistostaining was used to assess endothelial integrity by a 5-point ordinal scale (grade 0: intact endothelial lining, grade 5: complete denudation). Endothelial integrity after SCS was used as a control for the state of the endothelium at commencement of NMP. RESULTS: In the SCS specimens, regardless of the anatomical site, near complete endothelial denudation was present throughout the HA (median scores 4.5-5). After NMLP, significantly less endothelial loss in the distal HA was present compared to SCS grafts (NMLP vs. SCS: median grade 3 vs. 4.5; p = 0.042). In NMLP specimens, near complete endothelial denudation was present at the cannulation site in all cases (median grade: 5), with significantly less loss of the endothelial lining the further from the cannulation site (proximal vs. distal, median grade 5 vs. 3; p = 0.005). CONCLUSION: Loss of endothelial lining throughout the HA after SCS and at the cannulation site after NMLP suggests extensive damage related to surgical handling and preservation injury. Gradual improved endothelial lining along more distal sites of the HA after NMLP indicates potential for re-endothelialisation. The regenerative effect of NMLP on artery quality seems to occur to a greater extent further from the cannulation site. Therefore, arterial cannulation for machine perfusion of liver grafts should ideally be as proximal as possible on the coeliac trunk or aortic patch, while the site of anastomosis should preferentially be attempted distal on the common HA.