Béatrice Aussilhou

Reappraisal of the risks and benefits of major liver resection in patients with initially unresectable colorectal liver metastases.

Objectives:To determine short- and long-term outcomes of major hepatectomy in patients with downstaged colorectal liver metastases considered initially unresectable (IU). Background:Improvements in both surgical technique and efficacy of chemotherapy have increased the rate of resection for IU colorectal liver metastases. The outcome of these patients needs to be reassessed. Patients and Methods:From 2000 to 2011, 257 patients underwent major hepatectomy for colorectal liver metastases. Of these, 87 (34%) IU patients required portal vein occlusion after chemotherapy downstaging. Patients requiring less than 12 cycles and 12 or more cycles of chemotherapy before resection were defined as fast responders and slow responders, respectively. Results:Compared with fast responders, slow responders had increased mortality (0% vs 19%, P = 0.003) and major morbidity rates (20% vs 55%, P < 0.001) despite almost identical tumor characteristics and similar procedures. In multivariate analysis, the only factor associated with increased major morbidity was the existence of a number of chemotherapy cycles of 12 or more (hazard ratio [HR]: 5.32, confidence interval [CI]: 1.69–16.7, P = 0.004). One-, 3-, and 5-year disease-free survival rates for the entire population were 48%, 17.5%, and 13%, respectively. Multivariate analysis found that slow responders (HR: 2.89, CI: 1.67–5.04, P < 0.001) and patients without adjuvant chemotherapy (HR: 2.38, CI: 1.33–4.35, P = 0.004) had a significantly decreased disease-free survival. All slow responders postoperatively recurred within 3 years. Conclusions:Liver resection in slow responders, that is, IU patients requiring 12 or more chemotherapy cycles and portal vein occlusion to achieve resectability, is associated with poor short- and long-term outcomes. These patients would probably benefit from more conservative strategies.

Extended liver resection for polycystic liver disease can challenge liver transplantation.

Abbreviations:PLD, polycystic liver disease; PKD, polycystic kidney disease; LT, liver transplantation; LR, liver resection; CT, computed tomography; MRI, magnetic resonance imaging; US, ultrasonography; KT, kidney transplantation; ICU, intensive care unit; POD, postoperative days; INR, prothrombin index; GI, gastrointestinal; HV, hepatic vein; and IVC inferior vena cava. Objective:In an attempt to refine the indications for liver resection (LR) for highly symptomatic polycystic liver disease (PLD), we focused on the characteristics, technical difficulties, postoperative outcome, and long-term follow-up of PLD patients who underwent either LR or liver transplantation (LT). Methods:Since 1995, among 72 patients with massive hepatomegaly, 45 patients underwent LR associated with contralateral cyst fenestration whereas 27 underwent LT associated with simultaneous kidney transplantations in 23. The LR group was characterized by absence of end-stage renal insufficiency, absence of ascites, and better nutrition status. In the LR group, the volumetry of the spared noncystic parenchyma was preoperatively assessed whereas pathological analysis focused on fibrosis and vascular congestion. Results:After LR, the mortality was nil and overall morbidity was 71%, including biliary leak in 20% and ascites in 42%. Persistent and massive ascites was observed in 8 patients who have undergone extensive resection and had significantly more frequently fibrosis on the analysis of their resected surgical specimens (P = 0.002). A volume of the remnant noncystic parenchyma less than 30% and the presence of vascular changes on the specimen were associated with higher risk of complications. The 5-year survival was 95% and among the 43 survivors, after a mean follow-up of 41 months, 36 (83%) patients stated that they were satisfied, with complete relief of symptoms in 30 (70%). After LT, the postoperative mortality was 15% and the overall morbidity was 85%, including 12 patients who required reoperation. Severe complications were more frequent in the presence of denutrition and preoperative ascites. The 5-year survival was 85% and after a mean follow-up of 36 months all survivors had complete relief of symptoms due to hepatomegaly. Conclusions:In case of massive hepatomegaly from PLD without end-stage renal failure, LR should be considered first when preserved remnant liver volume represents at least 30% of the total volume liver in the absence of vascular changes or fibrosis.

Impact of total pancreatectomy: short- and long-term assessment.

BACKGROUND The aim was to assess the outcome of a total pancreatectomy (TP). METHODS From 1993 to 2010, 56 patients underwent an elective TP for intraductal papillary mucinous neoplasia (n = 42), endocrine tumours (n = 6), adenocarcinoma (n = 5), metastases (n = 2) and chronic pancreatitis (n = 1). Morbidity and survival were analysed. Long-term survivors were assessed prospectively using quality-of-life (QoL) questionnaires. RESULTS Five patients developed gastric venous congestion intra-operatively. Post-operative morbidity and mortality rates were 45% and 3.6%, respectively. An anastomotic ulcer occurred in seven patients, but none after proton pump inhibitor therapy. There were five inappropriate TPs according to definitive pathological examination. Overall 3- and 5-year survival rates were 62% and 55% respectively; five deaths were related to TP (two postoperative deaths, one hypoglycaemia, one ketoacidosis and one anastomotic ulcer). Prospective evaluation of 25 patients found that 14 had been readmitted for diabetes and that all had hypoglycaemia within the past month. The glycated haemoglobin (HbA1c) was 7.8% (6.3-10.3). Fifteen patients experienced weight loss. The QLQ-C30 questionnaire showed a decrease in QoL predominantly because of fatigue and diarrhoea, and the QLQ-PAN26 showed an impact on bowel habit, flatulence and eating-related items. DISCUSSION Morbidity and mortality rates of TP are acceptable, although diabetes- and TP-related mortality still occurs. Endocrine and exocrine insufficiency impacts on the long-term quality of life.

Parietal Peritoneum as an Autologous Substitute for Venous Reconstruction in Hepatopancreatobiliary Surgery.

OBJECTIVE To evaluate the parietal peritoneum (PP) as an autologous substitute for venous reconstruction during hepatopancreatobiliary (HPB) surgery. BACKGROUND Venous resection during liver or pancreatic resection may require a rapidly available substitute especially when the need for venous resection is unforeseen. METHODS The PP was used as an autologous substitute during complex liver and pancreatic resections. Postoperative anticoagulation was standard and venous patency was assessed by routine computed tomographic scans. RESULTS Thirty patients underwent vascular resection during pancreatic (n = 18) or liver (n = 12) resection, mainly for malignant tumors (n = 29). Venous resection was an emergency procedure in 4 patients due to prolonged vascular occlusion. The PP, with a mean length of 22 mm (15-70), was quickly harvested and used as a lateral (n = 28) or a tubular (n = 2) substitute for reconstruction of the mesentericoportal vein (n = 24), vena cava (n = 3), or hepatic vein (n = 3). Severe morbidity included Clavien grade-III complications in 4 (13%) patients but there was no PP-related or hemorrhagic complications. Histological vascular invasion was present in 18 (62%) patients, and all had an R0 resection (100%). After a mean follow-up of 14 (7-33) months, all venous reconstructions were patent except for 1 tubular graft (97%). CONCLUSIONS A PP can be safely used as a lateral patch for venous reconstruction during HPB surgery; this could help reduce reluctance to perform vascular resection when oncologically required. Clinical trials identification: NCT02121886.

Hand-assisted laparoscopic total pancreatectomy: a report of two cases.

BACKGROUND Because of improvement in surgical technique and endocrine and exocrine insufficiency management, total pancreatectomy is being frequently performed, especially for benign or low-potential malignant diseases. The laparoscopic approach is rarely performed. SUBJECTS AND METHODS Our aim is to report two cases operated by the assisted laparoscopic approach and to describe a standardized surgical technique. RESULTS Two patients underwent laparoscopic total pancreatectomy with assisted minilaparotomy or the hand-assisted technique for degenerated intraductal papillary mucinous neoplasms (IPMNs) and neuroendocrine tumors with operative times of 270 and 360 minutes, estimated blood loss of 200 and 300 mL, and a hospital stay of 12 and 18 days, respectively. One patient was re-operated on postoperative Day 10 for bleeding from the hepaticojejunostomy probably related to an inadequate dose of antiproton inhibitors, necessitating refection of the anastomosis with an uneventful course. Pathological examination revealed degenerated IPMNs (T3N1R0) and well-differentiated neuroendocrine tumors (T2NOR0) with complete resection. After 6-10 months of follow-up, the diabetes is well controlled with insignificant episodes of hypoglycemia in 1 patient without any evidence of tumor relapse. CONCLUSIONS In selected patients laparoscopic total pancreatectomy appears safe and had many advantages over the open and other laparoscopic pancreatic resection approaches, including first laparoscopic abdominal exploration and no pancreatic anastomosis. Oncological rules can be respected, but further larger studies are needed before drawing conclusions.

Tumor Recurrence After Partial Hepatectomy for Liver Metastases in Rats: Prevention by In Vivo Injection of Irradiated Cancer Cells Expressing GMCSF and IL-12

BACKGROUND Adjuvant treatment could be helpful in prevent recurrence after partial hepatectomy for liver metastases. The purpose of this study was to assess the benefit of in vivo injection of irradiated autologous cancer cells expressing cytokines after partial hepatectomy in rats. METHODS Fifty-four BDIX rats were injected with 3 x 10(6) DHD-K12 cancer cells into the portal vein to induce multiple hepatic metastases. A 70% hepatectomy was carried out 3 days after the injection. The rats were then randomized into three groups of 18 rats each. Rats were given three injections at days 8, 15, and 21 of 5 x 10(6) irradiated DHD-K12 cancer cells expressing either interleukin 12 (IL-12) (group 1), granulocyte monocyte colony stimulating factor (GM-CSF) (group 2), or only saline solution (control group). At day 30, animals of each group were divided into two subgroups: 10 rats of each group were killed for pathological examination and cytofluorimetric analysis and 8 rats of each group were maintained for survival follow-up. RESULTS At day 30, mean number of tumors on liver surface was lower in rats treated by irradiated cancer cells expressing IL-12 than those in GM-CSF group or in the control group. Furthermore, peritoneal carcinomatosis was significantly more frequent in the control group: 3/9 (33%) than in pooled IL-12 and GM-CSF groups: 1/19 (5%) (P < 0.05). In the survival study, we observed a significant increased survival in treated rats compared with the control group (P = 0.0008). CONCLUSION Our results suggest that vaccination with autologous irradiated cancer cells expressing either IL-12 or GM-CSF induced a systemic immune antitumoral response that may be useful as an adjuvant therapy after surgical resection for liver metastases.

Portomesenteric shunt for liver transplantation and complete portal vein thrombosis.

Portal Vein Thrombosis W e read the article by Rodrı́guezCastro et al. (1) published in the December 2012 issue of Transplantation entitled ‘‘Management of Nonneoplastic Portal Vein Thrombosis in the Setting of Liver Transplantation: A Systemic Review’’ with great interest. We totally agree that the subject is very interesting and so are the authors’ results, but we would like to describe in this letter our results in a surgical technique that can overcome this problem but which is underestimated and not well known for liver transplant (LT) surgeons. LT in the setting of cirrhosis and nonneoplastic portomesenteric thrombosis (NPMT) is controversial and even contraindicated in many transplant centers, especially in patients with diffuse thrombosis. NPMT is frequent (8%), complicates surgery, significantly increases postoperative mortality and morbidity, and is a major risk of long-term portal vein thrombosis (PVT) (1Y3). Although NPMT has been considered a contraindication to LT, many alternative surgical techniques have been developed to sustain portal flow to the liver graft. These alternative surgical techniques depend mainly on the grade of NPMT and probably the experience of the surgical team. Four types of NPMT have been described including partial PVT (grade I), complete PVT (grade II), partial superior mesenteric vein (SMV) thrombosis(gradeIII),andcompleteSMV thrombosis (type IV) (2). In cases of LT and NPMT, graft portal flow can be established by anatomical (portal vein thrombectomy, portomesenteric shunts, or combined small bowel and LT) or nonanatomical (renoportal shunts or portocaval hemitransposition) routes (4, 5). Nonanatomical shunts mainly depend on the coexistence of abnormally developed venous shunts between the splanchnic and caval circulations. These shunts are inconstant even in patients with severe portal hypertension, which can explain the high morbidity and mortality rate of LT using nonanatomical shunts (6). Although anatomical (especially portomesenteric) shunts have the advantage of directly perfusing the liver graft by the splanchnic blood independent of thepresence of abnormal venous shunts, these shunts are technically difficult and are usually contraindicated in grade IV thrombosis.

Laparoscopic fenestration of pancreatic serous cystadenoma: Minimally invasive approach for symptomatic benign disease.

Serous cystadenoma (SC) is a benign pancreatic cystic tumor. Surgical resection is recommended for symptomatic forms, but laparoscopic fenestration of large symptomatic macrocystic SC was not yet described in the literature. In this study, 3 female patients underwent laparoscopic fenestration for macrocystic SC (12-14 cm). Diagnosis was established via magnetic resonance imaging and endoscopic ultrasound, with intra-cystic dosage of tumors markers (ACE and CA19-9) in 2 patients. All patients were symptomatic and operated on 15-60 mo after diagnosis. Radiological evaluation showed constant cyst growth. Patients were informed about this new surgical modality that can avoid pancreatic resection. The mean operative time was 103 min (70-150 min) with one conversion. The post-operative course was marked by a grade A pancreatic fistula in one patient and was uneventful in the other two. The hospital stay was 3, 10, and 18 d, respectively. The diagnosis of macrocystic SC was histologically-confirmed in all cases. At the last follow-up (13-26 mo), all patients were symptom-free, and radiological evaluation showed complete disappearance of the cyst. Laparoscopic fenestration, as opposed to resection, should be considered for large symptomatic macrocystic SC, thereby avoiding pancreatic resection morbidity and mortality.

Complete spontaneous liver graft disappearance after auxiliary liver transplantation.

A 59-year old female patient was admitted to the intensive care unit with acute liver failure (ALF) related to Aminata phalloides mushroom poisoning; mushrooms had been ingested 8 hours before symptoms developed. Treatment by N-acetyl cysteine (Flumicil) was begun. Four days after ingestion, a second increase in liver enzymes (transaminases level >1,000 UI/L) was observed with a marked decrease in coagulation factors (prothrombin time [PT] 6%; factor V 9%). Although there was no encephalopathy or altered renal function, the patient was scheduled for emergency liver transplantation because according to the literature and in our experience, rapid deterioration can occur with a fatal outcome if curative treatment is not undertaken. Because of the absence of any underlying liver disease and the relative hemodynamic stability of the patient, auxiliary orthotopic liver transplantation (AOLT) was decided on. Surgery and Early Outcome. Frozen section histology of the native liver parenchyma showed hepatocyte necrosis of 70-80% without fibrosis, indicating that native liver regeneration was possible. A native liver right tri-sectionectomy was performed and segments IV to VIII were removed. A whole cadaveric liver graft was transplanted from a braindead donor and vascular anastomoses were performed to privilege the liver graft. The postoperative course was marked by rapid recovery of liver function tests (PT 5 85%; bilirubin 5 15 lmol/l) on postoperative day 5 and the patient was discharged on postoperative day 26. Long-Term Follow-up. Immunosuppression included glucocorticoids (for 3 months), mycophenolate mofetil, and tacrolimus. Six months after AOLT, functional recovery of the native liver was confirmed by computed tomography (CT) scan volumetry (Fig. 1). There were signs of hypertrophy of the native liver, which was confirmed by liver biopsy showing normal liver architecture with a few inflammatory cells without necrosis. Eleven months after AOLT (Fig. 2), significant native liver hypertrophy was observed and was confirmed by another liver biopsy, which showed marked native liver regeneration with no acute or chronic inflammation. Immunosuppression was gradually tapered down according to our established protocol (0.5 mg 3 2 of tacrolimus, twice weekly) at this time. The graft progressively atrophied as the native liver hypertrophied and immunosuppressive treatment was stopped completely 18 months after AOLT. The graft disappeared completely after 2 years (Fig. 1). The patient is now living a normal life without treatment.

Polycystic liver disease: Hepatic venous outflow obstruction lesions of the noncystic parenchyma have major consequences

In patients with polycystic liver disease (PLD), development of cysts induces hepatic venous outflow obstruction (HVOO) and parenchymal modifications, challenging the paradigm of a normal noncystic liver parenchyma. The aims were to reappraise the pathology of the noncystic parenchyma, by focusing on HVOO lesions; and to investigate the association with outflow obstruction at imaging and perioperative course after liver resection. This is a retrospective study conducted in one tertiary center between 1993 and 2014. PLD patients (n = 125) who underwent resection (n = 90) or transplantation (n = 35) were included. HVOO parenchymal lesions were assessed for all patients and a liver congestion score was built. Imaging was analysed for 45 patients with computed tomography scan, and perioperative course was assessed in resected patients. At pathology, 92% of patients had HVOO lesions, with sinusoidal dilatation being the most common feature. HVOO was more severe in patients who underwent transplantation compared to liver resection, as assessed by the congestion score. At imaging, all patients had HVOO with at least two hepatic veins involved. Mosaic enhancement pattern of the parenchyma was associated with the severity of hepatic vein obstruction (P = 0.045) and the compression of the inferior vena cava (P = 0.014). In case of liver resection, intraoperative course was characterized by hemorrhage, related to HVOO at imaging. Ascites (44%) and liver failure (9%) in the postoperative period were associated with blood losses and transfusions. Conclusion: Hepatic venous outflow obstruction, including development of venous collaterality and parenchymal changes, is frequent in PLD and has major consequences on intraoperative bleeding and postoperative ascites and liver failure. Hepatic venous outflow obstruction should be taken into account to choose the most appropriate surgical treatment. (Hepatology 2017).