Background Extensive resection for hilar cholangiocarcinoma is the most effective treatment, but high morbidity and poor prognosis remain concerns. Previous data have shown marked differences in ...outcomes between comparable Eastern and Western centers. We compared the outcomes of the management for hilar cholangiocarcinoma at one Japanese and one British institution with comparable experience. Methods Of 298 consecutive patients with hilar cholangiocarcinoma evaluated at Hirosaki University Hospital, Japan and St. James's University Hospital, Leeds, UK, 183 underwent radical resection. Clinicopathologic variables and postoperative outcomes were compared. Results Significant differences were not observed between the Hirosaki and Leeds cohorts in overall outcomes despite several differences in the patient characteristics. Although there was a difference in 90-day mortality (2.5% vs 13.6%, respectively), disease-specific 5-year survival rates were 32.8% and 31.9%, respectively ( P = .767). Multivariate analysis identified trisectionectomy (odds ratio = 2.32; P = .010), combined pancreatoduodenectomy (odds ratio = 7.88; P = .010), and perioperative blood transfusion (odds ratio = 1.88; P = .045) were associated with postoperative major complications, while preoperative biliary drainage associated with postoperative major complications, while preoperative biliary drainage (risk ratio = 2.21; P = .018), perioperative blood transfusion (risk ratio = 1.58; P = .029), lymph node metastasis (risk ratio = 2.00; P = .002), moderate/poorly differentiated tumor (risk ratio = 1.72; P = .029), microvascular invasion (risk ratio = 1.63; P = .046), and R1 resection (risk ratio = 1.90; P = .005) were risk factors for poor survival. Conclusion Disease-specific survival and prognostic factors were similar in both centers. Meticulous operative technique to avoid perioperative blood transfusion may improve long-term survival.
Introduction This meta-analysis aimed to review the percentage increase in future liver remnant (FLR) and perioperative outcomes after portal vein ligation (PVL) and portal vein embolization (PVE) ...before liver resection. Methods An electronic search was performed of the MEDLINE, EMBASE, and PubMed databases using both subject headings (MeSH) and truncated word searches to identify all articles published that related to this topic. Pooled risk ratios were calculated for categorical outcomes and mean differences for secondary continuous outcomes using the fixed-effects and random-effects models for meta-analysis. Results Seven studies involving 218 patients met the inclusion criteria. There was no difference in the increase in FLR between the 2 groups 39% (PVE) versus 27% (PVL; mean difference MD 6.04; 95% CI, −0.23, 12.32; Z = 1.89; P = .06). Similarly, there was no difference in the morbidity (risk ratio RR, 1.08; 95% CI, 0.55, 2.09; Z = 0.21; P = .83) and mortality (RR, 0.87; 95% CI, 0.19, 3.92; Z = 0.18; P = .85) in the 2 groups after liver resection. While awaiting liver resection after PVL and PVE, no difference was noted in the number of patients developing disease progression (RR, 0.93; 95% CI, 0.52, 1.66; Z = 0.24; P = .81). In a subset analysis comparing FLR with PVE and PVL as part of the procedure called an associating liver partition with PVL for staged hepatectomy (ALPPS), there was a significant increase in FLR in favor of ALPPS (MD, −17.09; 95% CI, −32.78, −1.40; Z = 2.14; P = .03). Conclusion PVL and PVE result in comparable percentage increase in FLR with similar morbidity and mortality rates. The ALPPS procedure results in an improved percentage increase in FLR compared with PVE alone.
Background In Western countries, hepatocellular carcinoma (HCC) often presents at a large size, which is seen as a contraindication to transplantation and often resection. Although diagnosis by ...imaging and α-fetoprotein is usually straightforward, nonspecialist units continue to use biopsy to prove the diagnosis before transfer for specialist surgical opinion. We have looked at the impact of this on our practice. Study Design We retrospectively analyzed all large HCCs resected in our unit during the last 12 years. Survival data were calculated according to size and univariate and multivariate analyses were carried out to determine impact of preoperative, operative, and histologic factors affecting outcomes. Results We identified 85 large HCCs (> 3 cm) and classified 42 as giant (> 10 cm). Overall survival at 1, 3, and 5 years was 76%, 54%, and 51%. Size did not influence survival, although more complex surgical techniques were required for giant tumors. Predictors of poorer disease-free survival were positive resection margin (p < 0.001), multiple tumors (p = 0.003), macroscopic vascular invasion (p = 0.015), and preoperative lesion biopsy (p = 0.027). Conclusions Our data shows excellent outcomes after resection for large HCC. This supports the management of such patients in large-volume units that are fully equipped and experienced in the management of these patients. Preoperative biopsy should be avoided, as this unnecessary maneuver appears to have worsened our longterm results.
Background Hepatocellular carcinoma (HCC) most commonly arises in patients with chronic liver disease. Data on outcomes after liver resection in patients with noncirrhotic, nonfibrotic, seronegative, ...referred to as a “normal” liver are limited. We aimed to investigate differences in prognostic factors and outcomes between patients presenting with HCC arising in “normal” liver (NLHCC) and that arising in “diseased” liver (DLHCC). Study Design All patients undergoing resection for HCC between 1994 and 2008 were assessed. Multivariable analysis of clincopathologic data from the NLHCC group was performed by comparing them with data from the group who had surgery for DLHCC during this period. Results During the 15-year study period, 142 patients underwent liver resection for HCC: 81 for NLHCC and 61 for DLHCC. NLHCCs were more often solitary but were larger and required more major resections. There was no significant difference in survival outcomes between patients who had NLHCC or DLHCC, with overall and recurrence-free 5-year survivals of 60% and 51% in NLHCC and 55% and 33% in DLHCC, respectively. In patients with NLHCC, significant factors predicting overall survival were blood transfusion requirement (p = 0.003) and age (p = 0.009), and the only significant factor at predicting recurrence-free survival was presence of multiple tumors (p = 0.025). In contrast, in DLHCC, the only significant prognostic variables were a preoperative tumor biopsy (p = 0.017) or a high neutrophil-to-lymphocyte ratio (p = 0.001), both of which predicted a poorer recurrence-free survival. Conclusions HCC presenting in patients with a normal background liver parenchyma appears to present a different spectrum of the disease. However, excellent outcomes can be achieved after liver resection, although this often requires the use of advanced techniques due to late presentation.