J Buckels

Portal vein thrombosis in adults undergoing liver transplantation: risk factors, screening, management, and outcome

BACKGROUNDnPortal vein thrombosis (PVT) has been seen as an obstacle to liver transplantation (LTx). Recent data suggest that favorable results may be achieved in this group of patients but only limited information from small size series is available. The present study was conducted in an effort to review the surgical options in patients with PVT and to assess the impact of PVT on LTx outcome. Risk factors for PVT and the value of screening tools are also analyzed.nnnMETHODSnAdult LTx performed from 1987 through 1996 were reviewed. PVT was retrospectively graded according to the operative findings: grade 1: <50% PVT +/- minimal obstruction of the superior mesenteric vein (SMV); grade 2: grade 1 but >50% PVT; grade 3: complete PV and proximal SMV thrombosis; grade 4: complete PV and entire SMV thrombosis.nnnRESULTSnOf 779 LTx, 63 had operatively confirmed PVT (8.1%): 24 had grade 1, 23 grade 2, 6 grade 3, and 10 grade 4 PVT. Being male, treatment for portal hypertension, Child-Pugh class C, and alcoholic liver disease were associated with PVT. Sensitivity of ultrasound (US) in detecting PVT increased with PVT grade and was 100% in grades 3-4. In patients with US-diagnosed PVT, an angiogram was performed and ruled out a false positive US diagnosis in 13%. In contrast with US, angiograms differentiated grade 1 from grade 2, and grade 3 from grade 4 PVT. Grade 1 and 2 PVT were managed by low dissection and/or a thrombectomy; in grade 3 the distal SMV was directly used as an inflow vessel, usually through an interposition donor iliac vein; in grade 4 a splanchnic tributary was used or a thrombectomy was attempted. Transfusion requirements in PVT patients (10 U) were higher than in non-PVT patients (5 U) (P<0.01). In-hospital mortality for PVT patients was 30% versus 12.4% in controls (P<0.01). Patients with PVT had more postoperative complications, renal failure, primary nonfunction, and PV rethrombosis. The overall actuarial 5-year patient survival rate in PVT patients (65.6%) was lower than in controls (76.3%; P=0.04). Patients with grade 1 PVT, however, had a 5-year survival rate (86%) identical to that of controls, whereas patients with grades 2, 3, and 4 PVT had reduced survival rates. The 5-year patient survival rate improved from the 1st to the 2nd era in non-PVT patients (from 72% to 83%; P<0.01), in grade 1 PVT (from 53% to 100%; P<0.01), and in grades 2 to 4 PVT (from 38% to 62%; P=0.11).nnnCONCLUSIONSnThe value of US diagnosis in patients with PVT depends on the PVT grade, and false negative diagnoses occur only in incomplete forms of PVT (grades 1-2). The degree of PVT dictates the surgical strategy to be used, thrombectomy/low dissection in grade 1-2, mesoportal jump graft in grade 3, and a splanchnic tributary in grade 4. Taken altogether, PVT patients undergo more difficult surgery, have more postoperative complications, have higher in-hospital mortality rates, and have reduced 5-year survival rates. Analysis by PVT grade, however, reveals that grade 1 PVT patients do as well as controls; only grades 2 to 4 PVT patients have poorer outcomes. With increased experience, results of LTx in PVT patients have improved and, even in severe forms of PVT, a 5-year survival rate >60% can now be achieved.

Selective decontamination of the digestive tract reduces gram-negative pulmonary colonization but not systemic endotoxemia in patients undergoing elective liver transplantation.

Objective: To examine the effect of selective antibiotic decontamination of the digestive tract in patients undergoing elective orthotopic liver transplantation. Design: Prospective, randomized, concurrent allocation to either selective decontamination or standard antibiotic prophylaxis. Setting: Operating theater and intensive care unit at a tertiary referral, university teaching hospital. Patients: Fifty‐nine adult patients were recruited into the study and underwent liver transplantation. Interventions: Thirty‐two patients were randomized to standard treatment (control group) and 27 patients were randomized to receive selective decontamination. After early deaths and exclusions, 31 controls and 21 decontamination patients were available for analysis. Measurements and Main Results: Portal and systemic endotoxemia, colonization and infection rates, severity of illness (organ system failures, Acute Physiology and Chronic Health Evaluation II score, Therapeutic Intervention Scoring System score), antibiotic costs, and hospital survival rates were measured. Selective decontamination significantly reduced pulmonary infections and enteric, aerobic, and Gram‐negative bacillary colonization without facilitating the emergence of resistant organisms, but selective decontamination had no effect on endotoxemia or the development of organ system failures. The financial costs of the selective decontamination regimen outweighed the advantages gained from an associated reduction in antibiotic usage. Conclusion: The failure of selective decontamination to enhance survival rates in many studies of the regimen in critically ill patients may, in part, be related to the inability of selective decontamination to abolish endotoxemia. (Crit Care Med 1994; 22:40‐49)

Neuroendocrine Tumours of the Pancreas: Predictors of Survival after Surgical Treatment

Aims: Neuroendocrine tumours of pancreatic and duodenal origin (NETP) are rare and we present a significant experience from a single centre. Methods: Data was collected on 44 patients who underwent surgery between 1988 and 2002. Since 1997, data have been recorded prospectively on a dedicated database. Results: Twenty-four patients had functioning tumours (16 insulinomas, 3 gastrinomas, 2 somatostatinomas, 1 vipoma, 1 glucagonoma and 1 carcinoid tumour). Nine functioning tumours and 13 non-functioning had a malignant phenotype. Twenty pancreaticoduodenectomies, 9 local excisions, 7 distal and 2 total pancreatectomies, 5 bypasses and 1 exploratory laparotomy were performed. Fourteen patients (31.8%) had surgical complications, 1 died peri-operatively (2.3%). The overall actuarial survival for resected cases was 74.4 and 42.5% at 5 and 10 years, respectively. Lymph node invasion and metastases were significant predictors of survival by univariate analysis and only the presence of metastases retained significance on multivariate analysis. Conclusion: Surgical resection is the only curative treatment for NETP. Resection can be safely carried out in a specialist centre and is associated with good long-term survival. The presence of metastases was a significant predictive factor for survival in patients with NEPT in this series.

Haemorrhage following Pancreaticoduodenectomy: Risk Factors and the Importance of Sentinel Bleed

Aim: To document the prevalence and to evaluate the management strategies of haemorrhagic complications following pancreaticoduodenectomy (PD). Methods: All patients who underwent PD from 1/2000 to 10/2005 and experienced at least one episode of haemorrhage during the 30 first days postoperatively were recorded. Etiology of haemorrhage, treatment strategy and mortality rate were recorded and analyzed. Results: A total of 362 patients underwent PD during this period and 32 (8.8%) had haemorrhage postoperatively of whom 15 died (47% mortality rate). Primary intraluminal haemorrhage was recorded in 13 patients, primary intra-abdominal haemorrhage in 5 patients and secondary haemorrhage in 14 patients. Successful management of haemorrhage with angioembilization occurred in 2 patients in the study group. Statistical analysis revealed sepsis and sentinel bleed as risk factors for post-PD haemorrhage and pancreatic leak and sentinel bleed as risk factors for secondary haemorrhage (p < 0.05). Conclusions: Haemorrhage after PD is a life-threatening complication. Sepsis, pancreatic leak, and sentinel bleed are statistical significant factors predicting post-PD haemorrhage. Sentinel bleed is not statistically significant associated with postoperative mortality, but with the onset of secondary haemorrhage. The effectiveness of therapeutic angioembolization was not demonstrated in our study.

Cholangiocarcinoma Complicating Primary Sclerosing Cholangitis: A 24-Year Experience

Aim: To report the prevalence and outcome of cholangiocarcinoma arising in primary sclerosing cholangitis for a British tertiary referral centre. Methods: All patients diagnosed with primary sclerosing cholangitis and concurrent cholangiocarcinoma were identified from a prospectively maintained departmental database, and the mode of presentation, management and outcome were determined. Results: Of 370 patients with primary sclerosing cholangitis, 48 patients (13%) were diagnosed with a cholangiocarcinoma within a median time of 0.51 months (range: 0–73.12) from presentation to the unit. Mode of presentation included: inoperable tumours (n = 14); incidental findings in transplant hepatectomy specimens (n = 13); primary sclerosing cholangitis follow-up (n = 9); transplant work-up (n = 5); transplant waiting list (n = 5); suspected tumour confirmed at transplant (n = 1), and incidental finding at cholecystectomy (n = 1). The diagnosis was confirmed by: radiology-guided biopsy (n = 27); MRI (n = 3); CT (n = 2); laparoscopy or laparotomy (n = 2), and frozen section at transplant (n = 1). Management consisted of: transplantation (n = 14, including 1 abandoned); hepatic resection (n = 8), and palliation through stenting (n = 26). The overall median survival of the cohort was 4.9 months (range: 0.09–104.5). Median survival ranged from 2.6 months (range: 0.09–35.3) for palliation to 7.6 months (range: 0.6–99.6) for transplantation and 52.8 months (range: 3.7–104.5) for resection. There was no difference in survival between the transplant and resection groups (p = 0.14). Conclusions: Cholangiocarcinoma is a common finding in primary sclerosing cholangitis and regular screening of this cohort of patients at referring centres is advocated to detect early tumours, as surgical treatment at an early stage offers significantly better outcomes for this cohort of patients.

Reduced acute rejection after liver transplantation with Neoral-based triple immunosuppression

Vol 349 • March 8, 1997 701 risk, whereas factors preventing ovulation, such as pregnancy and use of oral contraceptives, are protective. The rate of increase for the incidence of ovarian cancer diminishes after menopause. We speculate that initiating mutations accumulate during the repeated cycles of cell divisions that epithelial ovarian cells undergo with ovulation, thereby forming the basis for malignant transformation upon loss of chromosome 17. Our data show that this event increases with age. On the assumption that epithelial tumours have a common cellular origin and that specific mutational events contribute to tumour histology and grade, our hypothesis explains the non-random distribution with age of ovarian tumours of different histology.

Treatment of Hydatid Disease of the Liver

Background: Hydatid disease of the liver though endemic in many countries, is rare in the UK. We evaluated a 16-year experience of treating hydatidosis using a management protocol combining surgery with anti-scolicidals. Patients and Methods: There were 30 patients. 14 (47%) males, median age 41 (range 25–72) years, of whom 21 (70%) were symptomatic. Diagnosis was by serological tests and imaging. All had disease confined to the liver and received peri-operative anti-scolicidal drug therapy. Results: The initial 4 (13%) patients received praziquantel combined with albendazole for 2 weeks and the following 26 (87%) patients received two cycles of albendazole 400 mg twice daily for 28 days, with a 14-day break in between. However, 2 (7%) patients could not tolerate albendazole, one due to GI side effects and the other developed deranged liver functions. These 2 patients subsequently received praziquantel for 2 weeks. All patients underwent surgery. Subtotal cystectomy was carried out on 29 (96%) patients and 1 patient required a segmentectomy. Cystobiliary communications were identified in 15 (50%) of patients which were oversewn using fine absorbable sutures. Of these, 7 had the bile ducts decompressed using a T tube, with only 1 developing a post-operative bile leak. In comparison, 8 were not drained of which 6 leaked (p = 0.03). The median post-operative hospital stay was 8 days (range 5–24). Patients who developed post-operative bile leaks, however, needed prolonged abdominal drainage for a median of 21 days (range 18–24). Two (7%) patients developed histologically proven recurrent disease. The median follow-up was 56 months (range 3–87). Conclusion: Surgery combined with anti-scolicidal therapy proved effective. Cystobiliary communications are common and, when identified, should result in the biliary system being drained, to avoid post-operative bile leaks.

Application of Portsmouth modification of physiological and operative severity scoring system for enumeration of morbidity and mortality (P-POSSUM) in pancreatic surgery

BackgroundPancreatoduodenectomy (PD) is associated with high incidence of morbidity and mortality. We have applied P-POSSUM in predicting the incidence of outcome after PD to identify those who are at the highest risk of developing complications.MethodA prospective database of 241 consecutive patients who had PD from January 2002 to September 2005 was retrospectively updated and analysed. P-POSSUM score was calculated for each patient and correlated with observed morbidity and mortality.Results30 days mortality was 7.8% and morbidity was 44.8%. Mean physiological score was 16.07 ± 3.30. Mean operative score was 13.67 ± 3.42. Mean operative score rose to 20.28 ± 2.52 for the complex major operation (p < 0.001) with 2 fold increase in morbidity and 3.5 fold increase in mortality. For groups of patients with a physiological score of (less than or equal to) 18, the O:P (observed to Predicted) morbidity ratio was 1.3–1.4 and, with a physiological score of >18, the O:P ratio was nearer to 1. Physiological score and white cell count were significant in a multivariate model.ConclusionP-POSSUM underestimated the mortality rate. While P-POSSUM analysis gave a truer prediction of morbidity, underestimation of morbidity and potential for systematic inaccuracy in prediction of complications at lower risk levels is a significant issue for pancreatic surgery

Factors predicting perioperative cytokine response in patients undergoing liver transplantation.

Objectives: An exaggerated production of proinflammatory cytokines during liver transplantation stimulates the inflammatory process within the graft, and eventually promotes liver failure. This study was conducted to evaluate factors predicting perioperative response of proinflammatory cytokines during liver transplantation. Design: Prospective, consecutive entry study of liver transplant candidates. Setting: University hospital. Patients: Thirty liver transplant recipients. Interventions: Arterial blood samples were obtained perioperatively. Measurements: Interleukin (IL)‐1β, IL‐6, tumor necrosis factor‐α were measured by ELISA. Endotoxin was determined by a chromogenic endotoxin‐specific method. Main Results: The peak concentrations of IL‐1β and IL‐6 in the patients with complications were significantly higher than those in the patients without complications. The peak concentration of IL‐1β was significantly correlated with the level of bilirubin at admission and the intraoperative blood product requirement. The peak concentration of IL‐6 was significantly correlated with the admission bilirubin and the intraoperative blood product requirement. A multivariate regression model revealed that the serum bilirubin and the intraoperative blood product requirement were the independent factors that influenced the peak concentration of IL‐1β or IL‐6. The severely jaundiced patients had a significantly higher plasma concentration of endotoxin at the end of the anhepatic phase. In addition, there was a tendency for these patients to have a higher postoperative peak concentration of endotoxin. Conclusions: Serum level of bilirubin may be a potent preoperative factor influencing perioperative cytokine response in patients undergoing liver transplantation. An enhanced perioperative response of endotoxin seen in severely jaundiced patients suggests the clinical implication of endotoxin removal during the anhepatic phase in liver transplant surgery.

Role of Completion Pancreatectomy as a Damage Control Option for Post-Pancreatic Surgical Complications

Introduction: Management of pancreatic leak and haemorrhage is complex with high mortality rates. In this study, the results of completion pancreatectomy which was performed as a last resort option were analysed. Patients and Methods: 25 patients who had completion pancreatectomy from among 677 patients who had pancreatoduodenectomy or distal pancreatectomy over a period of 18 years were analysed in terms of the indications for completion pancreatectomy, outcome and survival data. Results: Indications for completion pancreatectomy include pancreatic leak in 12 patients (48%), both bleeding and pancreatic leak in 8 (32%), and haemorrhage alone in 5 (20%) patients. 18 (72%) patients also had splenectomy. Median ITU stay was 4 and 8 days for those who survived and died post-completion pancreatectomy, respectively. 36% patients had septicaemia and 32% patients had multiple organ failure. 12 patients survived the operation with a median survival of 52 months. Conclusion: 25 (3.6%) patients required surgical intervention for pancreatic complications. The incidence of splenectomy was 84.6% in those who died after completion pancreatectomy compared to 58.3% of those who survived (Fisher’s exact test two-sided 0.20). Despite significant morbidity and mortality, completion pancreatectomy has a role in the management of post-pancreatic surgical complications.