Primary biliary cholangitis metachronously complicated with combined hepatocellular carcinoma-cholangiocellular carcinoma and hepatocellular carcinoma.

Primary biliary cholangitis (PBC) is a progressive cholestatic liver disease characterized by the presence of highly specific antimitochondrial antibodies, portal inflammation and lymphocyte-dominated destruction of the intrahepatic bile ducts, which leads to cirrhosis. While its pathogenesis remains unclear, PBC that shows histological progression to fibrosis carries a high risk of carcinogenesis; the same is true of viral liver diseases. In patients with PBC, the development of hepatocellular carcinoma (HCC) is rare; the development of combined hepatocellular carcinoma and cholangiocellular carcinoma (cHCC-CCC) is extraordinary. Herein, we report a rare case of PBC metachronously complicated by cHCC-CCC and HCC, which, to the best of our knowledge, has never been reported. We present a case report of a 74-year-old Japanese woman who was diagnosed as PBC in her 40's by using blood tests and was admitted to our department for further management of an asymptomatic liver mass. She had a tumor of 15 mm in size in segment 8 of the liver and underwent a partial resection of the liver. Subsequent pathological findings resulted in the diagnosis of cHCC-CCC, arising from stage 3 PBC. One year after the initial hepatectomy, a second tumor of 10 mm in diameter was found in segment 5 of the liver; a partial resection of the liver was performed. Subsequent pathological findings led to HCC diagnosis. The component of HCC in the initial tumor displayed a trabecular growth pattern while the second HCC showed a pseudoglandular growth pattern, suggesting that metachronous tumors that arise from PBC are multicentric.

Effect of prolonged administration of pegylated interferon/ribavirin therapy in genotypes 2a and 2b: propensity score-matched analysis.

BACKGROUND AND AIM: Chronic hepatitis C genotype 2 patients show high susceptibility to pegylated interferon plus ribavirin therapy (PEG/RBV). However, the differences in response to therapy between genotypes 2a and 2b, and the efficacy of prolonged therapy for refractory patients have not been evaluated. We investigated the differences in response to PEG/RBV between each genotype and examined the efficacy of prolonged therapy. METHODS: A total of 343 chronic hepatitis patients infected with hepatitis C virus (HCV) genotype 2 (2a: n = 195; 2b: n = 148) were enrolled in this study. All patients received PEG/RBV for 24 (24 week group, n = 242) or more weeks (prolonged group, n = 101). We analyzed the differences in virological response between genotypes 2a and 2b. Clinical and virological factors of patients in the 24-week group and the prolonged treatment group were matched using propensity score analysis, and the efficacy of prolonged therapy established by comparing time of serum HCV disappearance for each genotype. RESULTS: Virological response tended to be higher for genotype 2a compared with genotype 2b; however, there was no significant difference in sustained virological response rates between genotypes (2a: 78.3%; 2b: 70.2%; P = 0.19). After propensity score matching, the adjusted P-value for sustained virological response rate was significantly different for genotype 2b patients with undetectable HCV-RNA between weeks 5 and 8, and for genotype 2a patients with detectable HCV-RNA at week 8. CONCLUSION: Prolonged therapy with PEG/RBV may be effective when serum HCV-RNA is detectable at week 4 and week 8 for genotype 2b and 2a patients, respectively.

Follow up of the 987 blood donors found with hepatitis C virus infection over 9-18 years.

AIM: To follow up blood donors found with hepatitis C virus (HCV) infection, to improve the outcome by antiviral treatments. METHODS: Between 1991 and 2001, 3377 of the 1 925 860 donors (0.18%) were found to have HCV infection at the Hiroshima Red Cross Blood Center in Japan. Of them, 987 were able to be followed regularly over 9-18 years until 2009, and received antiviral treatments as required. RESULTS: At the start, chronic hepatitis was diagnosed in 541 (54.8%), cirrhosis in five (0.5%) and hepatocellular carcinoma (HCC) in one (0.1%), whereas the remaining 439 (44.5%) had persistently normal aminotransferase levels (PNAL). Hospital visits were terminated voluntarily in 24.3% within the first year, 46.8% by 10 years and 50.9% by 17 years. Liver disease improved in 178 (18.0%), remained stable in 606 (61.4%) and aggravated in 170 (17.2%). Of the 541 donors with chronic hepatitis, HCC developed in 28 (5.2%) and cirrhosis in 11 (2.0%), whereas HCV infection was cleared in 107 (19.8%) by antiviral treatments. In addition, HCV infection resolved in 54 of the 439 donors (12.3%) with PNAL after they had developed chronic hepatitis and received treatments. In donors with chronic hepatitis, the cumulative incidence of HCC was 4.1% at 10 years. By multivariate analysis, age and diagnosis of chronic hepatitis at the entry were found to be independent risk factors for the development of HCC. CONCLUSION: Individuals with undiagnosed HCV infection need to be identified and receive medical care. They have to be motivated to merit from this health-care program.

Impact of ribavirin dose reduction on the efficacy of pegylated interferon plus ribavirin combination therapy for elderly patients infected with genotype 1b and high viral loads.

AIM: To examine the impact of ribavirin dose reduction on the efficacy of pegylated interferon (PEG IFN) plus ribavirin combination therapy for elderly patients infected with genotype 1b and high viral loads. METHODS: A total of 72 patients, over 65 years old, were recruited for this study. Patients were divided into groups receiving either 600-800 mg of ribavirin according to bodyweight (Group 1, n = 36) or 400 mg of ribavirin (Group 2, n = 36) plus 1.5 µg/kg (range: 1.3-2.0 µg/kg) of PEG IFN-α-2b for 48 weeks. RESULTS: Total ribavirin doses were administrated at 9.80 ± 2.39 mg/kg per day (3.29 ± 0.80 g/kg) for Group 1 and 5.87 ± 1.82 mg/kg per day (1.97 ± 0.61 g/kg) for Group 2 (P < 0.001). According to the total clearance (CL/F) of ribavirin, 34 of 36 patients in Group 1 received over-doses of ribavirin. In contrast, numbers of those receiving equivalent doses of ribavirin were two of 36 patients in Group 1 and 36 of 36 patients in Group 2, respectively (P < 0.001). End-of-treatment response (ETR) rates were observed in 23 of 36 patients (63.9%) in the standard ribavirin dose protocol and in 23 of 36 patients (63.9%) in the reduction ribavirin dose protocol (NS). Sustained virological response (SVR) rates were observed in 11 of 36 patients (30.6%) in the standard ribavirin dose protocol, and in 13 of 36 patients (36.1%) in the reduced ribavirin dose protocol (NS). CONCLUSION: Reduction of ribavirin doses for elderly patients did not affect the outcome for the 48-week combination therapy.

IL-28B predicts response to chronic hepatitis C therapy--fine-mapping and replication study in Asian populations.

Type I interferon (IFN) is used for the treatment of chronic hepatitis C virus (HCV) infection. Despite advances in antiviral therapy, a large proportion of patients remain infected following current therapies. Through a genome-wide scan, we found two variants (rs8099917 and rs12979860) in the IL-28B locus that affect the outcome of PEG-IFN and ribavirin combination therapy, consistent with recent studies (P = 6.52×10(-8); odds ratio 2.46 and P = 8.63×10(-8), odds ratio 2.40, respectively). Significant associations were also observed in the case of IFN monotherapy for HCV genotypes 1b and 2a. With rs8099917, HCV genotype 1b patients had a significantly lower frequency of the favourable genotype (86.6 %) compared with healthy controls (91.7 %), and HCV genotype 2a patients had an intermediate frequency (89.9 %). Similar results were found for rs12979860. Fine-mapping analysis revealed that rs8099917 had the strongest association with treatment outcome and 14 others, including four novel single nucleotide polymorphisms, had comparable associations. Haplotype analysis revealed that none of the haplotypes showed stronger association than any single marker. Early non-responders who could not achieve 2 log viral decline during the first 12 weeks of treatment had higher odds ratios for these two variants. The favourable allele of rs8099917 is also associated with initial viral decline at 2 and 4 weeks following the start of therapy. Multivariate analysis of PEG-IFN and ribavirin-treated patients showed that rs8099917 genotype, viral load, fibrosis and age were significant predictors of response to therapy. Common variation at the IL-28B locus is predictive of various IFN-based therapies for HCV independent of regimen or HCV genotype.

Predicting the response to 48-week combination therapy with peginterferon alpha-2b plus ribavirin from the estimated HCV RNA load index after negative serum change in genotype 1b hepatitis C patients.

AIM: We estimated viral dynamics after serum hepatitis C virus (HCV) RNA became negative and assessed the relation between the estimated viral load at the end of treatment (EVE) index and the response to the combination therapy with peginterferon alpha-2b plus ribavirin. METHODS: Patients with chronic HCV, genotype 1b, and a high viral load were treated with this combination therapy for 48 weeks, and serum HCV RNA was measured frequently during the treatment period. In the patients showing an end-of-treatment response (ETR), the viral load profile from the start of treatment until serum HCV RNA became negative was expressed by an approximate curve. Then the EVE index was calculated by using the expression obtained from the curve, and differences between the sustained virologic response (SVR) and relapse groups were investigated. RESULTS: The SVR rate increased as the EVE index became lower, and the EVE index was significantly lower in the SVR group than in the relapse group. The SVR rate was higher for those in whom the EVE index was below the cut-off point. CONCLUSION: Prediction of SVR and relapse from the EVE index is more useful than prediction from viral dynamics at the time when HCV RNA becomes negative or when HCV RNA shows a decrease of 2-log or more.

Dose comparison study of pegylated interferon-alpha-2b plus ribavirin in naïve Japanese patients with hepatitis C virus genotype 2: a randomized clinical trial.

BACKGROUND AND AIM: To compare the efficacy and safety of pegylated interferon (PEG-I) at 1 and 1.5 microg/kg, and in combination with ribavirin (RBV) for 24 weeks in naïve Japanese patients infected with hepatitis C virus genotype 2. METHODS: The present study was an open-label, randomized trial of 55 patients receiving PEG-I (1 or 1.5 microg/kg body weight [BW], subcutaneously, once a week) and RBV for 24 weeks. The patients were followed up for 24 weeks without treatment. RESULTS: The intention-to-treat analyses showed that the proportion of patients with a sustained virological response (SVR) in the 1-microg/kg PEG-I-RBV group (38.5%, 10/26) was lower than that of the 1.5-microg/kg PEG-I-RBV group (74.1%, 20/27; P = 0.013). The PEG-I dose was reduced in two of the 26 patients of the 1-microg/kg PEG-I-RBV group (one because of thrombocytopenia at 2 weeks, and one because of generalized fatigue at 20 weeks), and four of the 27 patients of the 1.5-microg/kg PEG-I-RBV group (one because of neutropenia at 20 weeks, and three because of generalized fatigue at 1, 5, and 8 weeks). The multivariate analysis identified age (< 60 years) and dose of PEG-I (1.5 microg/kg) as significant determinants of SVR. CONCLUSION: The dose of PEG-I to be used at the start of therapy should be 1.5-microg/kg BW in naïve Japanese patients infected with hepatitis C virus genotype 2.

Surgically resected hepatocellular carcinomas in patients with non-alcoholic steatohepatitis.

BACKGROUND/AIMS: Surgically resected hepatocellular carcinomas (HCC) in patients with non-alcoholic steatohepatitis (NASH) have rarely been described and the clinicopathological characteristics of HCC and non-cancerous liver tissue are still obscure. METHODOLOGY: From 1997 to 2004, 242 patients with HCC underwent hepatic resection at the Hiroshima Red Cross Hospital and Atomic Bomb Survivors' Hospital. Among this group, the diagnosis of NASH was made in 3 patients. RESULTS: All 3 patients with HCC had cirrhosis. The tumor cells contained Mallory bodies and fat. The non-cancerous areas showed nodular regeneration with fatty changes, ballooning degeneration, and mild inflammatory infiltrates, as well as perivenular and perisinusoidal fibrosis. CONCLUSIONS: Patients with NASH and cirrhosis may progress to HCC, and careful follow-up based on tumor markers and imaging modalities, is essential to detect resectable HCC in patients with NASH and cirrhosis.

Prospective study of short-term peginterferon-alpha-2a monotherapy in patients who had a virological response at 2 weeks after initiation of interferon therapy.

BACKGROUND AND AIMS: Long-term interferon (IFN) therapy is effective in eliminating hepatitis C virus (HCV). However, it carries the risk of adverse effects and reduced quality of life. To assess whether short-term IFN therapy effectively eliminates HCV, we performed a prospective pilot study of pegylated (peg)IFN-alpha-2a therapy for 8 or 24 weeks. METHODS: After excluding patients with high titers of genotype-1, 55 HCV patients received pegIFN-alpha-2a. Patients who became negative for HCV-RNA at week 2 were allocated to either an 8-week (n = 19) or 24-week (n = 15) course of IFN. We evaluated the efficacy of and tolerance to IFN therapy. RESULTS: The sustained virological response rate was excellent in the two groups (8 weeks, 89.5% [17/19]; 24 weeks, 100% [15/15], respectively,). IFN dose reduction was required in one patient of the 8-week group, but in six patients of the 24-week group (P = 0.028). Treatment was completed by all patients of the 8-week group, but discontinued in five patients of the 24-week group (P = 0.011). CONCLUSIONS: The 8-week IFN therapy is more tolerable than the 24-week therapy and had similar outcomes. Excluding the patients with high titers of genotype-1, we recommend switching to an 8-week course of pegIFN-alpha monotherapy once patients show an ultra rapid virological response at week 2 from the start of IFN therapy.

Impact of obesity on the surgical outcome following repeat hepatic resection in Japanese patients with recurrent hepatocellular carcinoma.

AIM: To evaluate the impact of obesity on the posto-perative outcome after hepatic resection in patients with hepatocellular carcinoma (HCC). METHODS: Data from 328 consecutive patients with primary HCC and 60 patients with recurrent HCC were studied. We compared the surgical outcomes between the non-obese group (body mass index: BMI < 25 kg/m(2)) and the obese group (BMI > or = 25 kg/m(2)). RESULTS: Following curative hepatectomy in patients with primary HCC, the incidence of postoperative complications and the long-term prognosis in the non-obese group (n = 240) were comparable to those in the obese group (n = 88). Among patients with recurrent HCC, the incidence of postoperative complications after repeat hepatectomy was not significantly different between the non-obese group (n = 44) and the obese group (n = 16). However, patients in the obese group showed a significantly poorer long-term prognosis than those in the non-obese group (P < 0.05, five-year survival rate; 51.9% and 92.0%, respectively). CONCLUSION: Obesity alone may not have an adverse effect on the surgical outcomes of patients with primary HCC. However, greater caution seems to be required when planning a repeat hepatectomy for obese patients with recurrent HCC.

Liver disease in hepatitis C virus carriers identified at blood donation and their outcomes with or without interferon treatment: Study on 1019 carriers followed for 5-10 years.

AIM: To portray liver disease and project outcomes in carriers of hepatitis C virus (HCV) in the general population. METHODS: Liver disease was evaluated in 1019 individuals who were found with HCV infection at blood donation, and they were followed for 5-10 years with or without receiving interferon (IFN). RESULTS: At baseline, chronic hepatitis was detected in 529 (51.9%) HCV carriers and more frequently in men than in women (62.6% [299/478]vs 42.5% [230/541], P < 0.01); cirrhosis was diagnosed in five (0.5% [three men included]) and hepatocellular carcinoma (HCC) in one (0.1% [man]). Of the carriers who were followed for 5 years or longer, loss of HCV-RNA from serum was achieved in 61 (31.0%) of the 197 treated with interferon (IFN) and only one of the 211 (0.5%) without IFN (P < 0.0001). HCC developed in 14 carriers including six ofthe 211 (2.8%) without IFN and eight of the 197 (4.1%) with IFN (six non-responders included). Follow ups of the 949 carriers identified age (P < 0.002), male gender (P < 0.01) and cirrhosis at the baseline (P < 0.0001) as factors contributing to the development of HCC. Cumulative incidence rates of HCC during 10 years among carriers found with chronic hepatitis increased in parallel with the age at the baseline. CONCLUSION: Identification of HCV carriers in the general population and treating those indicated with IFN would help decrease the development of HCC and lift its medical, as well as economic, burdens off society.

Serum HBV RNA is a predictor of early emergence of the YMDD mutant in patients treated with lamivudine.

Lamivudine (LAM) is a nucleoside analogue widely used for the treatment of chronic hepatitis B virus (HBV) infection. Emergence of resistant strains with amino acid substitutions in the tyrosine-methionine-aspartate-aspartate (YMDD) motif of reverse transcriptase is a serious problem in patients on LAM therapy. The amount of covalently closed circular DNA in the serum is reported to be higher in patients who develop YMDD mutants than in those without mutants. However, there is no useful serum marker that can predict early emergence of mutants during LAM therapy. Analysis of patients who were treated with entecavir (n=7) and LAM (n=36) showed some patients had high serum levels of HBV RNA. Median serum levels of HBV RNA were significantly higher in patients in whom the YMDD mutant had emerged within 1 year (n=6, 1.688 log copies/ml) than in those in whom the YMDD mutant emerged more than 1 year after treatment (n=12, 0.456 log copies/ml, P=0.0125) or in whom the YMDD mutant never emerged (n=18, 0.688 log copies/ml, P=0.039). Our results suggest that HBV RNA is a valuable predictor of early occurrence of viral mutation during LAM therapy.

Eicosapentaenoic acid could permit maintenance of the original ribavirin dose in chronic hepatitis C virus patients during the first 12 weeks of combination therapy with pegylated interferon-alpha and ribavirin. A prospective randomized controlled trial.

OBJECTIVE: To evaluate the efficacy of eicosapentaenoic acid (EPA) against ribavirin (RBV)-associated hemolytic anemia during the first 12 weeks in chronic hepatitis C virus (HCV) combination therapy. METHODS: This study was a prospective open-label, randomized controlled trial. 100 HCV patients were randomized to either the EPA group (n = 49) or non-EPA group (n = 51) who received combination therapy with or without EPA. We compared the changes in hemoglobin level and RBV plasma concentrations at week 12 in each group with RBV dose reduction rate and performed multivariate analysis to identify independent variables associated with RBV dose reduction. RESULTS: 8 patients (17%) in the EPA group and 20 patients (29%) in the non-EPA group required RBV dose reduction, respectively. The cumulative RBV reduction rate was significantly lower in the EPA group than in the non-EPA group (p = 0.017), while the decrease of hemoglobin and RBV plasma concentrations from baseline was not significantly different. However, in the multivariate analysis, treatment with EPA showed significant variables for the reduction of RBV dose (odds ratio 3.235, p = 0.023). CONCLUSION: EPA could prevent the RBV dose reduction during the first 12 weeks in combination therapy, although further large-scale double-blind randomized controlled trials are required.

Prolonged negative HCV-RNA status led to a good outcome in chronic hepatitis C patients with genotype 1b and super-high viral load.

OBJECTIVE: We examined whether a sustained negative HCV-RNA status for 48 weeks affects the outcome in patients with genotype 1b and super-high viral load, and also investigated whether the outcome is affected by the induction therapy of twice-daily pre-administrated interferon (IFN)-beta. METHODS: 78 eligible patients were divided into four groups. 40 were patients assigned to the short treatment protocol. 13 patients received 3 MU IFN-beta twice daily for 2 weeks followed by IFN-alpha2b+ribavirin for 22 weeks (beta-induction group: group 1). 27 patients received IFN-alpha2b+ribavirin for 24 weeks (standard combination group: group 2). 38 patients were assigned to the maintenance treatment protocol. All of the 13 in the beta-induction group (group 3) and 21 of 25 patients in the standard combination group (group 4) who were negative HCV-RNA PCR at week 24 had IFN monotherapy to maintain a negative HCV-RNA result for 48 weeks. RESULTS: An HCV-RNA-negative status at week 24 was observed in 96% (25/26) of groups 1 and 3 versus in 79% (41/52) of groups 2 and 4 (p<0.01). The sustained virological response (SVR) was 38% (5/13) in group 1 and 11% (3/27) in group 2 (p<0.05). In the maintenance treatment, SVR was observed in 46% (6/13) of group 3 and 32% (8/25) of group 4 (NS). CONCLUSIONS: A sustained negative HCV-RNA status for 48 weeks might be associated with viral elimination in patients with genotype 1 and super-high viral load.

Liver metastasis from thyroid carcinoma 32 years after resection of the primary tumor: report of a case.

Follicular thyroid carcinoma is a differentiated cancer originating from the follicular cells in the thyroid gland. A 73-year-old woman, who had undergone curative resection of thyroid carcinoma 32 years earlier, was referred to our hospital after ultrasonography showed a solid mass in the liver. Laboratory data revealed positive hepatitis B core antibody, but all other values were normal. Computed tomography showed a round tumor, about 1.5 cm in diameter, which was enhanced early and washed out later, in segment 5 of the liver. She underwent laparotomy and partial resection of the liver. Microscopic examination showed follicular cells with minimal atypia growing in a thyroid follicular pattern with colloids, whereby a diagnosis of metastatic liver cancer from thyroid follicular carcinoma was made. This is a rare case of solitary liver metastasis appearing 32 years after eradication of primary follicular carcinoma. Although the reason for the delayed presentation of the metastatic lesion remains unclear, this case shows that patients with differentiated thyroid cancer should be followed up for their entire life.

Limited hepatic resection for hepatocellular carcinoma in the caudate lobe.

The most appropriate approach to treating hepatocellular carcinoma (HCC) in the caudate lobe has not yet been determined. A series of 197 patients who had undergone curative hepatic resection for HCC were analyzed. Fifteen patients had HCC in the caudate lobe: three in the Spiegel lobe (SP), three in the caudate process (CP), and nine in the paracaval portion (PC). Patients with HCCs in the SP and CP underwent partial hepatectomy. HCCs in the PC were approached in one of three ways: anterior approach and partial hepatectomy of the PC (Ant+PHx-PC), partial hepatectomy, or left lobectomy. Clinicopathologic variables, including the underlying liver disease, the mean tumor size, and the pathologic characteristics of HCC, did not differ between surgery of the caudate lobe and that of other segments. The overall survival was 88.9% at 3 years and 66.7% at 5 years after resection of HCC in the caudate lobe; the corresponding figures were 86.1% at 3 years and 68.6% at 5 years for the other segments. The recurrence-free survival rate was 51.9% at 3 years and 34.6% at 5 years for the caudate lobe, and it was 52.1% at 3 years and 32.8% at 5 years for the other segments. Clinicopathologic characteristics of HCCs originating in the caudate lobe were not different from those in the other segments. Limited resection of HCC in the caudate lobe confers a similar prognostic value as in other segments.

Significance of thymidine phosphorylase in HCC with chronic liver disease for long-term postoperative recurrence.

BACKGROUND AND OBJECTIVES: Little is known about the significance of thymidine phosphorylase (TP) levels in HCC. The authors investigated the TP activity of HCC and adjacent normal liver tissue in 70 HCC patients, and the prognostic significance of the TP levels. METHODS: We obtained fresh samples (tumors with adjacent normal tissues) from 70 patients with HCC who have undergone curative hepatic resection. The levels of TP activity in unfixed, fresh, frozen HCC tissue, together with adjacent normal liver tissue specimens, were determined by the enzyme-linked immunosorbent assay (ELISA) method. RESULTS: The mean TP activity in HCC was approximately 1.5-fold higher than that in adjacent liver tissues (P = 0.0013). Disease-free survival rates for patients with low and high TP levels in HCC were not significantly different. Although disease-free survival rates for low and high TP levels in adjacent normal liver tissue were also not significantly different, disease-free survival for patients with low and high TP levels in adjacent normal liver tissue, excluding cases of postoperative recurrence within 700 days of hepatectomy, showed a significant difference (P = 0.0494). The group with high TP activity in adjacent liver tissue showed more frequent postoperative recurrence in the long-term than that with low TP activity. CONCLUSIONS: Measurement of TP activity of adjacent normal liver tissue to HCC may allow us to predict metachronous long-term postoperative recurrence.

Hepatic resection for hepatocellular carcinoma existing with liver cirrhosis.

BACKGROUND/AIMS: Hepatocellular carcinoma is usually complicated with liver cirrhosis, which makes its treatment difficult. Also a high rate of recurrence exists after surgical resection. However, how the prognosis after surgical treatment is affected by the severity of coexisting cirrhosis has not been clarified. METHODOLOGY: We compared the postoperative longterm courses of hepatocellular carcinoma patients with cirrhosis according to the liver function. All 112 hepatocellular carcinoma patients in this study underwent curative hepatic resection, and were classified into three groups according to the severity of liver dysfunction. The ICG R15' (indocyanine green retention test) normal: < 10%) was used in this study. Patients whose ICG R15' was less than 20% were classified as group I of 62, patients equal to 20% or between 20% and 30% as group II of 24, and patients equal to and more than 30% as group III of 26. RESULTS: In this series, 76 of 112 patients had recurrence (68%). A second hepatic resection was performed in six cases of group I and one case in group II. Fifty-eight of 76 recurrent cases (76%) were treated with transcatheter arterial chemoembolization. A total of eleven cases had no transcatheter arterial chemoembolization in the three groups: 3 cases in group I, 5 cases in group II, and 3 cases in group III; The three cases of group III had no treatment because of extremely poor liver dysfunction, whilst the 8 patients without transcatheter arterial chemoembolization in groups I and II had hepatocellular carcinoma itself and other diseases. The 1-, 3-, and 5-year survival rates after recurrence were 92%, 48%, and 14%, respectively, in group I; 83%, 37%, 12%, respectively, in group II; and 66%, 30%, 0%, respectively, in group III. The prognosis was significantly worse according to the degree of liver dysfunction (p = 0.0206). CONCLUSIONS: The prognosis of hepatocellular carcinoma with liver cirrhosis is affected not only by hepatocellular carcinoma itself, but also by the severity of the coexisting cirrhosis. Moreover, the cirrhotic liver can decline due to surgery. Surgical resection of this disease should be performed after careful patient selection and using a less invasive technique.