Long-term survivors after resection for primary liver cancer. Clinical analysis of 19 patients surviving more than ten years

Portal Vein Embolization: Rationale, Techniques, and Outcomes to Maximize Remnant Liver Hypertrophy with a Focus on Contemporary Strategies

Hepatectomy remains the gold standard for curative therapy for patients with limited primary or metastatic hepatic tumors as it offers the best survival rates. In recent years, the indication for partial hepatectomy has evolved away from what will be removed from the patient to the volume and function of the future liver remnant (FLR), i.e., what will remain. With this regard, liver regeneration strategies have become paramount in transforming patients who previously had poor prognoses into ones who, after major hepatic resection with negative margins, have had their risk of post-hepatectomy liver failure minimized. Preoperative portal vein embolization (PVE) via the purposeful occlusion of select portal vein branches to promote contralateral hepatic lobar hypertrophy has become the accepted standard for liver regeneration. Advances in embolic materials, selection of treatment approaches, and PVE with hepatic venous deprivation or concurrent transcatheter arterial embolization/radioembolization are all active areas of research. To date, the optimal combination of embolic material to maximize FLR growth is not yet known. Knowledge of hepatic segmentation and portal venous anatomy is essential before performing PVE. In addition, the indications for PVE, the methods for assessing hepatic lobar hypertrophy, and the possible complications of PVE need to be fully understood before undertaking the procedure. The goal of this article is to discuss the rationale, indications, techniques, and outcomes of PVE before major hepatectomy.

Safety and outcomes of pre-operative portal vein embolization using N-butyl cyanoacrylate (Glue) in hepatobiliary malignancies: A single center retrospective analysis

Aims and Objectives:

To evaluate the outcome of preoperative portal vein embolization (PVE) using N-butyl cyanoacrylate (NBCA) for change in future liver remnant (FLR) volume, biochemical changes, and procedure-related complications. The factors affecting FLR hypertrophy and the rate of resection was also evaluated for this cohort.

Materials and Methods:

From 2012 to 2017, PVE utilizing NBCA mixed with lipiodol (1:4) was performed using percutaneous approach in 28 patients with hepatobiliary malignancies with low FLR. All patients underwent volumetric computed tomography (CT) assessment before and at 3–5 weeks after PVE and total liver volume (TLV), FLR volume, and FLR/TLV ratio, changes in portal vein diameter and factors affecting FLR were evaluated. Complications and the resectability rate were recorded and analyzed.

Result:

PVE was successful in all 28 patients. The mean FLR increased by 52% ± 32% after PVE (P < 0.0001). The FLR/TLV ratio was increased by 14.2% ± 2.8% (P < 0.001). Two major complications were encountered without any impact on surgery. There was no significant change seen in liver function test and complete blood counts after PVE. Eighteen patients (64.28%) underwent hepatic resection without any liver failure, and only three patients developed major complication after surgery. Remaining ten patients did not undergo surgery because of extrahepatic metastasis detected either on follow-up imaging or staging laparotomy. Patients with diabetes showed a lower rate of hypertrophy (P < 0.05).

Conclusion:

Preoperative PVE with NBCA is safe and effective for increasing FLR volume in patients of all age group and even in patients with an underlying liver parenchymal disease with hepatobiliary malignancy. Lesser hypertrophy was noted in patients with diabetes. A reasonable resectability was achieved despite having a high rejection in gall bladder cancer subgroup due to rapid disease progression.

Liver operations in Sweden in 1987-99

BACKGROUND AND AIMS

Liver surgery is developing with new techniques and treatment modalities. The aim of this study is to describe liver surgery over a long period of time in a country with a public health care system.

PATIENTS AND METHOD

A register study including adult patients admitted for liver resection in Sweden (population 8.8 million) selected from the Inpatient Register 1987-99. Additional data were collected from the Swedish Cancer Register and the Cause of Death Register. Analyses of the patients, indications, mortality and causes of death are presented.

RESULTS

2,405 operations were performed (21 per million per year). The most frequent indication was colorectal metastases (27%). The 5-year survival after an operation for primary liver cancer and colorectal liver metastases was 27% and 26%, respectively.

CONCLUSIONS

Few patients were admitted for liver operations compared to expected figures. The survival rates are in conformity with those previously published. With an increasing awareness of the relatively favourable prognosis and the introduction of new methods, the volume of liver operations will probably increase in Sweden.

Extent of liver resection influences the outcome in patients with cirrhosis and small hepatocellular carcinoma

BACKGROUND

The long-term outcome after resection of hepatocellular carcinoma (HCC) is influenced by parameters related to the tumor and the underlying liver disease. However, the extent of the resection, which can be limited or anatomical (including the tumor and its portal territory), is controversial.

METHODS

Among 64 Child-Pugh A patients with cirrhosis who underwent curative liver resection for small HCC (< or = 4 cm) between 1990 and 1996, 34 patients underwent limited resection with a margin width of at least 1 cm, and 30 patients underwent anatomic resection of at least 1 liver segment with complete removal of the portal area containing the tumor. The 2 groups were comparable in terms of epidemiologic and pathologic parameters. The major end points were: (1) in-hospital mortality and morbidity; (2) overall and disease-free survival; and (3) rate and topography of recurrence.

RESULTS

The 30-day mortality (6% vs 7%) and morbidity (52% vs 47%) rates after limited and anatomic liver resection were not statistically different. The 5- and 8-year overall survival rates after limited versus anatomic resection were, respectively, 35% versus 54% (P <.05) and 6% versus 45% (P <.05). The 5- and 8-year disease-free survival rates were, respectively, 26% versus 45% and 0% versus 21% (P <.05). Local recurrence was more frequently observed after limited resections than after anatomic resections (50% vs 10%, P <.05).

CONCLUSIONS

In patients with cirrhosis and a small HCC, anatomic resection achieves better disease-free survival than limited resection without increasing the postoperative risk. Therefore, anatomical resection should be the treatment of choice and considered as the reference surgical treatment compared with other treatments.

Clinicopathologic features of long-term survivors and disease-free survivors after resection of hepatocellular carcinoma: a study of a prospective cohort

PURPOSE

This study aims to clarify the clinicopathologic features of long-term survivors and disease-free survivors after resection of hepatocellular carcinoma (HCC).

PATIENTS AND METHODS

The clinicopathologic features of 5-year survivors and disease-free survivors were elucidated in a cohort of 230 patients prospectively observed for > 5 years (64 to 192 months) after curative resection of HCC.

RESULTS

The incidence of 5-year overall and disease-free survivors were 37% (85 of 230) and 20% (45 of 230), respectively. Clinicopathologic features associated with 5-year survivors included female sex (P =.024), preoperative serum albumin > or= 40 g/L (P =.033), AST < 50 u/L (P =.001), tumor < 5 cm (P =.001), solitary tumor (P =.035), encapsulated tumor (P =.021), no venous invasion (P =.001), no microsatellite nodule (P =.001), and early pathologic tumor-node-metastasis (pTNM) stage (I or II, P <.001). Features favoring 5-year disease-free survivors were preoperative serum AST < 50 u/L (P =.007), tumor < 5 cm (P =.005), encapsulated tumor (P =.007), no venous invasion (P <.001), no microsatellite nodule (P =.001), and early pTNM stage (I or II, P <.001). By multivariate analysis, pTNM stage was the only significant predictive factor for both overall and disease-free survival.

CONCLUSION

This study shows that long-term disease-free survival > 5 years after resection of HCC can be achieved in patients with favorable tumor characteristics. Early pTNM stage was the most reliable predictor of both long-term overall and disease-free survivors.

Hepatocellular carcinoma

Early detection of hepatocellular carcinoma (HCC) is feasible, particularly in patients known to be at risk from chronic hepatitis and chronic liver disease. The optimal surveillance strategy is unknown. HCC usually presents as an incurable disease even when detected on surveillance. Surgical resection is the treatment of choice, but the coexistence of chronic liver disease and the insidious nature of HCC make it unresectable in most patients. Orthotopic liver transplantation for selected patients or ablative techniques may offer an opportunity to render patients disease-free even if the tumor is unresectable. There are numerous therapies offered to patients with unresectable HCC, including chemotherapy, hormonal therapy, and regional intra-arterial treatments. While potentially palliative, none of these approaches has been demonstrated to prolong survival in these patients. If possible, the treatment of patients with HCC should be done on clinical trials.

Telomerase activity in peripheral blood for diagnosis of hepatoma

BACKGROUND

Telomerase activity may be used as a molecular marker for the detection of circulating hepatoma cells in blood of patients with hepatoma.

METHODS

Telomerase activity in peripheral blood from hepatocellular carcinoma (HCC) patients was assessed by using a highly sensitive and non-radioisotope telomerase polymerase chain reaction (PCR) ELISA. Initially, tissue telomerase activity was measured in the hepatoma and non-tumour portions by using PCR ELISA within the same specimen, to compare its sensitivity with the conventional telomeric repeat amplification protocol (TRAP) method. Second, telomerase activity was measured in the peripheral blood obtained from patients with HCC, patients with chronic liver disease and in healthy controls.

RESULTS

Of the 17 HCC patients, telomerase activity was found to be positive in 14 (82%) by using TRAP and 15 (88%) by using PCR ELISA, indicating that PCR ELISA is a reliable tool for the measurement of telomerase activity. By using the Telomerase PCR ELISA assay, telomerase activities in the peripheral blood of 20 HCC patients was 1.65 +/- 0.78 units. This was significantly greater than the results obtained for 20 chronic liver disease patients (0.43 +/- 0.36 units) and 20 healthy controls (0.39 +/- 0.14 units; P < 0.0001).When the arbitrary cut-off level was set at 0.7 units (maximum value of healthy controls + 0.1), the positive frequency of telomerase activity was 25% for chronic liver disease and 80% for HCC patients (sensitivity 80%, specificity 75%). Among the HCC patients, high telomerase activity in the peripheral blood was shown at stage III HCC with vascular invasion (2.10 +/- 0.62 units, n = 9). This was significantly higher than patients at stage II of HCC (1.28 +/- 0.72 units, n = 11, without vascular invasion; P = 0.012).

CONCLUSION

These results suggest that peripheral blood telomerase activity, which may reflect haematogenous micrometastasis, is potentially a practical diagnostic/predictive marker of HCC.

Hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is a disease that is extremely difficult to manage and is increasing markedly in incidence. This presents both an opportunity and a challenge. At-risk patients can be identified and early detection of HCC is feasible. New surgical techniques and postoperative therapies, including hepatic intra-arterial radiation, may improve the outlook for some patients with resectable cancer. Unfortunately, the vast majority of patients with HCC will have unresectable cancers. Regional treatments may shrink or necrose tumors, but no clear benefit to such therapies has been demonstrated. Recent evidence suggests combination chemotherapy may help some patients, although this needs validation. Perhaps the best hope is that the further elucidation of the genetic and molecular features of HCC will lend us insight into innovative strategies to manage this difficult cancer.

Staged hepatectomy after emergency transcatheter arterial embolization for ruptured hepatocellular carcinoma

BACKGROUND

Staged hepatectomy after emergency transcatheter arterial embolization (TAE) has been advocated in ruptured hepatocellular carcinoma (HCC). However, there have been no reports of clinical series of this strategy. The purpose of this study was to evaluate the protocol of this therapeutic strategy.

METHODS

Sixteen patients with suspected rupture of HCC were included in the study. After emergency TAE, tumor resectability was assessed, followed by staged hepatectomy or repeated TAE. The patients were reevaluated with regard to rupture of HCCs.

RESULTS

Primary hemostasis was achieved successfully in all patients. Eleven patients were finally judged to have experienced HCC rupture. Seven of them underwent staged hepatectomy; the other four underwent repeated TAE because their tumors were considered unresectable. Survival time tended to be longer, but not to a significant extent, in patients who underwent hepatectomy (range, 139 to 1527 days; median, 375 days) than in those treated by TAE alone (range, 43 to 1317 days; median, 158 days).

CONCLUSIONS

Staged hepatectomy after TAE is a rational treatment for patients with ruptured HCC. Although TAE is highly effective for initial hemostasis, hepatectomy appears to provide better long-term survival.

Hepatic cryosurgery for recurrent hepatocellular carcinoma after hepatectomy: a preliminary report

BACKGROUND AND OBJECTIVES

The treatment of choice for recurrent hepatocellular carcinoma (HCC) is repeated resection. However, only a small percentage of patients are suitable for further hepatic resection. The aim of this study was to evaluate the surgical risk and operative outcome of hepatic cryosurgery in patients with recurrent HCC.

METHODS

A retrospective analysis of patients with recurrent HCC after previous curative hepatectomy treated with cryosurgery. Four patients with recurrent HCC not suitable for further resection were enrolled for cryosurgery, their clinical parameters, the operative details and outcome were studied.

RESULTS

No intraoperative or postoperative complications were noted. The duration of operation ranged from 3-5.2 hr and the operative blood loss from 173-1,300 ml. All patients are still alive with survival after cryosurgery ranging from 12-23 mo (25-63 mo after the hepatic resection). Three patients showed evidence of recurrent disease and one patient was disease free.

CONCLUSIONS

Hepatic cryosurgery is a safe therapy for patients with unresectable recurrent HCC.

Utility of lentil lectin affinity of alpha-fetoprotein in the diagnosis of hepatocellular carcinoma

AIMS/METHODS

Frozen sera obtained from 70 patients (35 with hepatocellular carcinoma and 35 with benign chronic liver disease) with serum alpha-fetoprotein > 20 ng/ml were studied to evaluate the diagnostic indices of lentil lectin affinity of alpha-fetoprotein in detecting hepatocellular carcinoma.

RESULTS

The proportion of alpha-fetoprotein-L3 was significantly higher in patients with hepatocellular carcinoma than in those with benign chronic liver disease (41.0 +/- 33.6% vs. 16.4 +/- 15.3%, p < 0.001). This difference led to a sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 57, 89, 83, 67 and 73%, respectively, in detecting hepatocellular carcinoma using the proportion of alpha-fetoprotein-L3 > 35% as a parameter. Within a 1-year period, 1500 high-risk persons were collaborating, leading to 22 cases with serum total alpha-fetoprotein > 20 ng/ml. These 22 cases included six pregnant women. The parameter, alpha-fetoprotein-L3 > 35% was used along with sonography to detect hepatocellular carcinoma for the remaining 16 cases. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 75, 83, 60, 91 and 81%, respectively, by the proportion of alpha-fetoprotein-L3 > 35%; and 100, 92, 80, 100 and 94%, respectively, by sonography.

CONCLUSIONS

Lentil lectin affinity of alpha-fetoprotein provides a moderately high sensitivity and a high specificity in the detection of hepatocellular carcinoma for persons with high alpha-fetoprotein levels. It may be a useful adjuvant tool of sonography and total alpha-fetoprotein level in a mass survey of hepatocellular carcinoma for a high-risk population.

Screening for hepatocellular carcinoma in chronic carriers of hepatitis B virus: incidence and prevalence of hepatocellular carcinoma in a North American urban population

OBJECTIVE

To prospectively determine the prevalence and annual incidence of hepatocellular carcinoma in hepatitis B carriers in a heterogeneous urban North American population and to assess the diagnostic accuracy of tests used for screening for this cancer.

DESIGN

Prospective cohort study of 1,069 chronic carriers of hepatitis B virus using screening with alpha-fetoprotein alone or in combination with ultrasonography every 6 months.

RESULTS

The mean age of the cohort was 39 +/- 12 years (+/- SD), 65% were men, 71% were Asians. At the first screening visit, serum alpha-fetoprotein was > or = 20 micrograms/L in 4%. In those subjects who were also screened by ultrasonography during the first visit, 9% were found to have focal lesions. Only 3 subjects were found to have hepatocellular carcinoma at the first screening, giving a prevalence of 281/100,000 chronic carriers of hepatitis B virus. The cohort was followed for 2,340 person-years (mean, 26 months follow-up, with a range from 6 to 60 months). During this period, 11 more subjects, 10 men and 1 woman, were diagnosed to have hepatocellular carcinoma (annual incidence, 470/100,000). In men only, the annual incidence was 657/100,000. During the study, 5 subjects died from hepatocellular carcinoma (annual mortality rate, 214/100,000). Sensitivity and specificity of serum alpha-fetoprotein > 20 micrograms/L were 64.3% and 91.4%, respectively. For ultrasonography, sensitivity was 78.8% and specificity 93.8%.

CONCLUSIONS

These data suggest that the incidence and prevalence of hepatocellular carcinoma in hepatitis B carriers in our area, an urban North American setting, are as high as in countries where hepatitis B is endemic. Current screening tests have significant false-positive and false-negative rates raising questions about the cost-benefit of screening for hepatocellular carcinoma in our study population.

Liver transplantation for hepatocellular carcinoma: results with preoperative chemoembolization

At the University of California, San Francisco, 17 patients who met the following criteria-hepatic tumor unresectable because of location or inadequate liver reserve, no metastases, HBsAg negative, no tumor larger than 5 cm in diameter, and no more than three tumors--were enrolled prospectively in a protocol employing preoperative chemoembolization to assess whether orthotopic liver transplantation (OLT) could cure a majority of highly selected patients with hepatocellular carcinoma (HCC). Thirteen patients had biopsy-proven HCC, 2 had the fibrolamellar variant, and 2 had radiological findings of HCC but no biopsy confirmation. Fourteen had underlying liver disease. All arteriographically apparent lesions were chemoembolized using a mixture including Gelfoam powder, doxorubicin, mitomycin-c, and cisplatin. Eight patients with poor hepatic reserve were chemoembolized when a donor organ became available, whereas 9 patients were chemoembolized and then placed on the waiting list. The only complication of chemoembolization was a gangrenous gallbladder in 1 patient. Thirteen patients underwent liver transplantation (2 patients without prior histological confirmation of carcinoma had no identifiable tumor at OLT); 3 patients developed metastases between the time of enrollment and donor organ availability and subsequently died; and 1 patient underwent a trisegmentectomy. Ten of the 11 patients with biopsy-proven HCC who underwent transplantation remain free of recurrent cancer at a median of 40 months; 1 patient died at 6 months of lymphoproliferative disease with no cancer found at autopsy. Although the role of chemoembolization is uncertain, these data show that the majority of carefully selected patients with HCC may achieve long-term survival with OLT.

Intraoperative microwave tissue coagulation as treatment for patients with nonresectable hepatocellular carcinoma

BACKGROUND

The microwave tissue coagulator (2450 MHz) has been used clinically in the treatment of hepatocellular carcinoma (HCC) to transection of the liver parenchyma and has proven an excellent method for hemostasis. There are, however, few reports on the application of this coagulator to the induction of tumor necrosis.

METHODS

Microwave tissue coagulation (MTC) was applied at laparotomy in eight patients with nonresectable multiple HCCs. All patients were treated with a combination of resection or intrahepato-arterial chemotherapy and MTC. A total of 222 bouts of MTC were applied to 21 tumors, the largest of which was 65 mm in largest dimension. The monopolar needle electrode was inserted directly into the tumor and the procedure was repeated at approximately 5 mm intervals.

RESULTS

Levels of alpha-fetoprotein in serum were found to have decreased in all patients one month after surgery with MTC. Contrast-enhanced computerized tomography (CT) showed the complete absence of blood flow in all tumors subjected to MTC. Needle biopsy one month after MTC confirmed tumor necrosis in all cases. All patients are alive at the time of this report, with the longest survival period being 24 months. In three of eight patients, new tumors were confirmed by angiographic CT at sites separate from the treated tumors. MTC resulted in fewer adverse effects on liver function and less extensive inflammatory reactions than liver resection.

CONCLUSION

Intraoperative MTC appears to be an effective method for inducing local tumor necrosis, and may be of use in combination with palliative surgery for multiple HCC when radical liver resection is not feasible.

Results of major hepatectomy for large primary liver cancer in patients with cirrhosis

Major hepatectomy is associated with a high operative risk in patients with cirrhosis. Between June 1983 and December 1991, 33 cirrhotic patients, 31 with good liver function, underwent major liver resection for a large hepatic primary cancer. The hospital mortality rate was 3 per cent; the only death resulted from liver failure after extended right hepatectomy. None of the patients had variceal bleeding during the postoperative period. Survival rates at 1, 2 and 3 years were 66, 43 and 37 per cent respectively. Recurrence was the most common reason for late death. These results suggest that the operative risk of major hepatectomy in cirrhotic patients with a large tumour and good liver function is comparable to that of minor liver resection. Late survival is also similar to that of patients with a small tumour. Cirrhotic patients with hepatic carcinoma and good liver function are suitable for major hepatectomy.

Predictable factors for estimating prognosis of patients after resection of hepatocellular carcinoma

BACKGROUND

Although many studies have been concerned with a clarification of the relation between various clinicopathologic factors and the prognosis of operated hepatocellular carcinoma (HCC), few studies have focused on the prognostic predictability of mitotic index and anti-hepatitis C virus (anti-HCV).

METHODS

One hundred forty cases of HCC with hepatic resection were observed from 1 to 11 years, and the relationship among various clinicopathologic factors, including the mitotic index and anti-HCV, and prognosis was evaluated.

RESULTS

Age at the time of operation, positive results for hepatitis B surface antigen or anti-HCV, accompanying cirrhosis, and the degree of tumor necrosis due to transarterial embolization did not influence the prognosis significantly. Patients with hepatitis C virus-related cases had a better prognosis than patients with hepatitis B-related cases. Patients with a single and small carcinoma smaller than 2 cm had a significantly better prognosis than those who had larger and/or multiple tumors. A better prognosis also was observed in the carcinomas with no histologic invasion into portal vein branches, low Edmondson grades, and low mitotic activities when compared with the counterpart of each group. Among these factors, the mitotic index was correlated best with prognosis in the current study.

CONCLUSIONS

The examination of mitotic index was quite simple, and the index was a helpful factor in predicting prognosis.

Solitary minute hepatocellular carcinoma. A study of 14 patients

Fourteen patients with clinical Stage I hepatocellular carcinoma (T1N0M0) were studied. All patients were asymptomatic, and their conditions were detected by alpha-fetoprotein (AFP) serosurvey and/or ultrasonography (US) either in the natural population in the early years of the study or in the high-risk population in the later years of the study. Cirrhosis was present in all patients. Radical resection was performed in all patients. There were no operative deaths or hospital deaths in this series. The 5-year survival rate after resection was 100%. There were seven long-term survivors in this series (14.2 years (alive), 11.3 years (alive), 8.8 years (alive), 8.8 years, 7.9 years, 7.6 years (alive), and 7.2 years after resection). The authors discuss aspects concerning early diagnosis, treatment, and prognosis of hepatocellular carcinoma (HCC).

Flow cytometric DNA analysis of hepatocellular carcinoma

The prognostic value of nuclear DNA content was studied retrospectively using flow cytometry in 203 cases of resected hepatocellular carcinoma. The occurrence of DNA aneuploidy, which was detected in 50% of patients, correlated significantly with tumor size and the presence of vascular invasion or intrahepatic metastasis. Overall, patients with DNA aneuploid tumors had a significantly worse prognosis than those with DNA diploid tumors (P less than 0.001) and, also in subdivided groups by tumor size (P less than 0.01). Among DNA aneuploid patients, the survival times were significantly shorter for patients with a low DNA index (less than 1.5) than for those with a high DNA index (greater than or equal to 1.5) (P less than 0.05). In a Cox multivariate analysis, nuclear DNA content provided significant prognostic value (P = 0.008), as did vascular invasion (P = 0.001) and intrahepatic metastasis (P = 0.005). These results indicated that nuclear DNA content has an important prognostic value in hepatocellular carcinoma.

Long-term survivors after hepatectomy for hepatocellular carcinoma

Sixteen long-term (more than 5 years) survivors after hepatic resection performed for hepatocellular carcinoma (HCC) from 1970 to 1988, were reviewed. The mean age of the patients was 51 years. There were 11 males and 5 females. HBs antigen was positive in 9 patients. Liver cirrhosis was associated with 11 patients but its severity was designated as Child's A in all patients except one. The mean tumour diameter was 2.8 cm and was relatively small. At the first operation, limited procedures (i.e. partial hepatectomy and subsegmentectomy) were employed in 87.5% of patients. A large percentage of tumours were located in S5 and S6 segments. A recurrence of HCC occurred in 9 patients after the first resection. A second resection was carried out in 7 patients, in 2 of which a third resection was done. Transcatheter arterial embolization (TAE) was performed on 4 patients. These results show that, in addition to detection of small tumours and early resection, repeated operation or TAE for treatment of recurrent HCC was important in achieving long-term survival after HCC resection.

Extended hepatectomy for hepatocellular carcinoma

The results of extended hepatectomy in 25 patients with hepatocellular carcinoma performed over a 16 year period have been reviewed, analysed and compared with those of 144 patients who underwent lesser liver resection. Five left and 20 right extended hepatectomies were performed for tumours ranging from 3 to 20 cm in diameter. Seventeen (68 per cent) of the patients had non-cirrhotic livers. The major postoperative complications were: haemorrhage in five cases, major bile duct injury in three, subphrenic abscess in two, liver failure in one and wound dehiscence in one. The 30-day (operative) mortality rate was 12 per cent and the median survival duration, including operative mortality, was 9.7 (range 0.2-32.1) months. The survival rate was 46 per cent at 1 year, 33 per cent at 2 years and 22 per cent at 3 years. The morbidity, mortality and survival data of extended hepatectomy were comparable with the results of lesser hepatic resections for hepatocellular carcinoma. We conclude that extended hepatectomy is a worthwhile operation for large hepatocellular carcinomas and a viable alternative to liver transplantation.