Background Retransplantation candidates are disadvantaged owing to lack of good-quality liver grafts. Strategies that can facilitate transplantation of suboptimal grafts into retransplant candidates require investigation. The aim was to determine whether late liver retransplantation can be performed safely with suboptimal grafts, following normothermic machine perfusion. Methods A prospectively enrolled group of patients who required liver retransplantation received a suboptimal graft preserved via normothermic machine perfusion. This group was compared with both historical and contemporaneous cohorts of patient who received grafts preserved by cold storage. The primary outcome was 6-month graft and patient survival. Results The normothermic machine perfusion group comprised 26 patients. The historical (cold storage 1) and contemporaneous (cold storage 2) groups comprised 31 and 25 patients respectively. The 6-month graft survival rate did not differ between groups (cold storage 1, 27 of 31, cold storage 2, 22 of 25; normothermic machine perfusion, 22 of 26; P = 0.934). This was despite the normothermic machine perfusion group having significantly more steatotic grafts (8 of 31, 7 of 25, and 14 of 26 respectively; P = 0.006) and grafts previously declined by at least one other transplant centre (5 of 31, 9 of 25, and 21 of 26; P < 0.001). Conclusion In liver retransplantation, normothermic machine perfusion can safely expand graft options without compromising short-term outcomes.

Introduction: ERAS improves short-term outcomes in patients undergoing pancreaticoduodenectomy (PD), but it is not clear if elderly patients benefit from ERAS. The aim of this study was to evaluate the effect of ERAS on outcomes in elderly patients after PD. Methods: An ERAS pathway was introduced after PD in our unit in December 2012. Short-term outcomes were compared pre- (2009-2012) and post-ERAS (2013-2019) in elderly (≥75 years) and non-elderly (<75 years) patients. Data regarding demographics and postoperative outcomes were collected from a prospectively maintained database. Multivariable analysis was performed to evaluate factors associated with postoperative length of stay (LOS). Results: 830 consecutive patients underwent PD (ERAS 577, pre-ERAS 253), including 170 elderly patients and 660 non-elderly patients. Both elderly (p<0.001) and non-elderly (p=0.039) ERAS patients had significantly higher comorbidity than pre-ERAS. ERAS was associated with significantly fewer medical complications (12.4% v 22.4%; p=0.002) in non-elderly patients, but there was no difference in elderly patients (23.6% v 14.0%; p=0.203). There was no difference in surgical complications after ERAS in either elderly (42.5% v 39.5%; p=0.858) or non-elderly (41.8% v 38.6%; p=0.445) patients. On multivariable analysis, ERAS was independently associated with reduced LOS in elderly (p=0.034) and non-elderly (p<0.001) patients. Factors independently associated with increased LOS were comorbidity (p<0.001), pancreatic fistula risk (p<0.001), additional procedures (p=0.003) and postoperative complications (p<0.001). Conclusion: An ERAS pathway after pancreaticoduodenectomy is associated with a significantly shorter hospital stay in both elderly and non-elderly patients. This finding is particularly notable given that patients operated during the ERAS era had higher comorbidity.

Introduction: Risk factors for clinically relevant postoperative pancreatic fistula (CR-POPF) after pancreaticoduodenectomy (PD) are well established. However, an ability to differentiate between biochemical leak (BL) and CR-POPF in the early postoperative period may be beneficial. The aim of this study was to identify early postoperative variables associated with CR-POPF. Methods: All consecutive patients diagnosed with BL and CR-POPF after PD were identified from a prospective database (2009-2019). Demographics, intraoperative parameters, postoperative drain fluid amylase (DFA) and C-reactive protein (CRP) on the first five postoperative days were retrospectively collected. Preoperative, intraoperative and early postoperative (POD1-5) factors were compared between groups. Cut-off values for DFA and CRP were calculated using ROC analysis. Results: 210/830 patients (25.3%) developed a POPF (104 BL, 106 CR-POPF). DFA1 (p=0.006), DFA3 (p=0.006), DFA5 (p<0.001), CRP2 (p=0.014), CRP3 (p<0.001), CRP4 (p<0.001) and CRP5 (p<0.001) were all significantly higher in patients with CR-POPF compared to BL. There was no significant difference in body mass index (27 v 26; p=0.150), Birmingham fistula score (p=0.05), pancreatic duct width (p=0.216) or soft gland consistency (p=1.00) between groups. CRP4 (AUC 0.677, 95%CI 0.588-0.765; p<0.001) and DFA5 (AUC 0.712, 95%CI 0.628-0.795; p<0.001) outperformed other variables in differentiating between BL and CR-POPF. Conclusion: In patients with a confirmed pancreatic fistula after pancreaticoduodenectomy, it may be possible to differentiate between a biochemical leak and a clinically relevant fistula using early postoperative variables, particularly drain fluid amylase and C-reactive protein values.

Purpose: Hepatic Echinococcus Granulosus disease (HEGD) requiring surgery is uncommon in the United Kingdom. The annual disease incidence in England and Wales is 10 cases. The aim of this study was to assess the surgical management of HEGD at a tertiary liver surgery uni in England. Methods: Retrospective review of an electronically maintained database of all patients' undergone liver surgery for HEGD, between 2006 and 2019, was performed. Details of patient demography, perioperative medical management, investigations, post-operative morbidity, and presence of live scolices on histology were reported. Perioperative medical management was compared with available local and regional guidelines. Results: 30 patients underwent liver surgery for HEGD (17male:13female). Mean age was 44 yrs(17-76). Right lobe disease was predominant (27/30patients). Average diameter of the cysts was 9.5cm at the time of surgery. Surgical procedures performed were deroofing of liver cysts: 16, Non-anatomical liver resection: 5, Pericystectomy:7, major liver resection:2. Bile leaks encountered (≥Grade IIIa):3 patients. Live scolices were present in 8 histological specimens. There were two perioperative medical management protocols available, and compliance to these was only 55%. There were no drug therapy related complications. Two patients (6%) had recurrent disease. Conclusion: The number of patients requiring surgical intervention for HEGD remains low (two cases/year in this series). Standardisation and compliance with perioperative medical management needs ensured with support from local infection diseases teams.

Purpose: Hermes workstation (Hermes Medical Solutions AB, Stockholm, Sweden) is increasingly used to assess the hepatic uptake of Technetium Mebrofenate for preoperative liver function assessment. The workstation also allows assessment of anatomical and functional liver volumes on SPECT-CT scan. The aim of the current study is to compare the agreement of the semiautomated liver segmentation for the assessment of total and segmental liver volumes with post resection liver specimen volumes using Hermes workstation. Methods: A retrospective analysis was carried out using a prospectively maintained database. A total of 144 patients underwent standard right or left hepatectomy (Brisbane 2000 nomenclature) between January 2015 and December 2016 were included. Total, resection and remnant anatomical liver volumes were measured retrospectively using Hermes workstation. Postresection volumes measured by assessment of dry weight of pathological specimen were compared with the resection volumes. The Pearson's correlation coefficient (r) was used to calculate the linear correlation using SPSS. Results: Right hepatectomy was performed in 79% and left hepatectomy in 21% of the included patients. There was no significant inter reader variability (AO,BD) as assessed on 15 initial consecutive scans (+/-15ml). The mean TLV calculated with the Hermes Software was 1688.41±582.86. The mean specimen weight measured during the histopathological analysis was 880.4±469.38. The volume of the resection segments calculated with the Hermes software significantly correlated with the weight of the specimen, with the Pearson correlation coefficient being 0.84 (p<0.001) (Figure 1). Conclusion: Hermes workstation is a reliable, and efficient tool to measure liver anatomical segmental volumes.

Abdominal wall transplant is developed in the context of intestinal and multivisceral transplant, in which it is often impossible to perform a primary wall closure. Despite the fact that abdominal wall closure is not as consequential in liver transplant, there are circumstances in which it might determine the success of the liver graft, especially in situations that compromise the abdominal cavity and facilitate an abdominal compartment syndrome.

Introduction: Stenosis of pancreaticodigestive anastomosis (pancreaticogastrostomy and pancreaticojejunostomy) is a very rare complication that usually develops several years after pancreaticoduodenectomy (PD). Only a few cases have been previously reported. We review the literature and present 2 more cases of pancreaticojejunostomy stenosis Results: We report two patients, 66 and 45 years old respectively, presenting with a clinical picture of recurrent acute pancreatitis at 6 years and 1 year after PD. Diagnosis of anastomotic stricture was established based on symptoms of pancreatitis (12–15 episodes between 5 and 20 months after PD), and CT scan and MRI findings. ERCP was attempted in both patients prior to surgery, being unsuccessful in both cases. Resection of the pancreaticojejunostomy stenosis was performed, including a slice of 1cm of the pancreatic parenchyma in contact with the jejunum, followed by a new well-vascularized two-layer end-to-side PJ with external drainage of the Wirsung duct. Catheter drainage was exteriorized through a jejunal limb using the Witzel technique. Postsurgical course was uneventful in both cases, and after a follow-up period of 3 and 2.5 years respectively, the patients remain asymptomatic but have developed endocrine and exocrine insufficiency. We believe Witzel technique is useful in patients with pancreatico-jejunostomy stenosis, as it contributes to reduce fistula rate and maintains sufficient Wirsung duct patency in order to avoid postoperative pain. Conclusion: Resection of the PJ and construction of a new PJ with external stent drainage of the Wirsung duct is our preferred surgical option in cases of PJ stenosis after a Whipple procedure.