The MELD Score Is Superior to the Maddrey Discriminant Function Score to Predict Short-Term Mortality in Alcohol-Associated Hepatitis: A Global Study.

INTRODUCTION: Several scoring systems predict mortality in alcohol-associated hepatitis (AH), including the Maddrey discriminant function (mDF) and model for end-stage liver disease (MELD) score developed in the United States, Glasgow alcoholic hepatitis score in the United Kingdom, and age, bilirubin, international normalized ratio, and creatinine score in Spain. To date, no global studies have examined the utility of these scores, nor has the MELD-sodium been evaluated for outcome prediction in AH. In this study, we assessed the accuracy of different scores to predict short-term mortality in AH and investigated additional factors to improve mortality prediction. METHODS: Patients admitted to hospital with a definite or probable AH were recruited by 85 tertiary centers in 11 countries and across 3 continents. Baseline demographic and laboratory variables were obtained. The primary outcome was all-cause mortality at 28 and 90 days. RESULTS: In total, 3,101 patients were eligible for inclusion. After exclusions (n = 520), 2,581 patients were enrolled (74.4% male, median age 48 years, interquartile range 40.9-55.0 years). The median MELD score was 23.5 (interquartile range 20.5-27.8). Mortality at 28 and 90 days was 20% and 30.9%, respectively. The area under the receiver operating characteristic curve for 28-day mortality ranged from 0.776 for MELD-sodium to 0.701 for mDF, and for 90-day mortality, it ranged from 0.773 for MELD to 0.709 for mDF. The area under the receiver operating characteristic curve for mDF to predict death was significantly lower than all other scores. Age added to MELD obtained only a small improvement of AUC. DISCUSSION: These results suggest that the mDF score should no longer be used to assess AH's prognosis. The MELD score has the best performance in predicting short-term mortality.

Cost-effectiveness and budget impact of interferon-free direct-acting antiviral-based regimens for hepatitis C treatment: the French case.

We evaluated the cost-effectiveness and the budget impact of new DAA-based regimen use in France. A Markov model simulated chronic hepatitis C (CHC) treatment interventions with IFN-based and IFN-free regimens at stage of fibrosis ≥F3, ≥F2 or regardless of fibrosis stage, and treatment either with the least or the most expensive combination. It estimated quality-adjusted life years (QALYs) and incremental cost-effectiveness ratios (ICERs). It also assessed the budget impact over 5 years of treating all CHC-screened patients, regardless of fibrosis, assuming ≤20 000 patients treated/year and priority to ≥F3. Sensitivity analyses were also conducted. For genotypes (G) 1-4, the initiation of IFN-free regardless of fibrosis was a cost-effective strategy compared to prior standard of care (SOC) initiated at stage F2: €40 400-88 300/QALY gained in G1; similar results were obtained for patients infected with G4. Considering G2-3, the most cost-effective strategy was IFN-based regimens regardless of fibrosis compared to prior SOC initiated at stage F2: €21 300 and €19 400/QALY gained, respectively; the strategy with IFN-free regimens being more effective but not cost-effective at current costs. The budget impact of treating all CHC-screened patients over 5 years would range between 3.5 and 7.2 billion €, depending on whether one considers the least or the most expensive combination of new DAAs and whether one treats G2-3 with IFN-based or IFN-free new DAAs. In France, treatment initiation with new DDAs regardless of fibrosis stage is cost-effective, but would add 3.5-7.2 billion € to an already overburdened medical care system.

Good outcome of liver transplantation in patients with pre-existing renal cell carcinoma.

The presence of a pre-existing or recent extra-hepatic solid tumor was considered for a long time as an absolute contraindication to liver transplantation, by fear of futility with an unacceptable increase in non-liver-related mortality. However, cancer-related mortality in solid malignancies is heterogeneous, and experts suggest that case-by-case multidisciplinary decisions should be made. Here, we report the cases of 3 patients with favorable oncological and liver outcome in patients with renal cell carcinoma detected during pre-transplant evaluation that nonetheless underwent liver transplantation.

Hepatic venous pressure gradient in the assessment of portal hypertension before liver resection in patients with cirrhosis.

BACKGROUND: Preoperative measurement of hepatic venous pressure gradient (HVPG) is not performed routinely before hepatectomy in patients with cirrhosis, although it has been suggested to be useful. This study investigated whether preoperative HVPG values and indirect criteria of portal hypertension (PHT) predict the postoperative course in these patients. METHODS: Between January 2007 and December 2009, consecutive patients with resectable hepatocellular carcinoma (HCC) in a cirrhotic liver were included in this prospective study. PHT was assessed by transjugular HVPG measurement and by classical indirect criteria (oesophageal varices, splenomegaly and thrombocytopenia). The main endpoints were postoperative liver dysfunction and 90-day mortality. RESULTS: Forty patients were enrolled. A raised HVPG was associated with postoperative liver dysfunction (median 11 and 7 mmHg in those with and without dysfunction respectively; P = 0·017) and 90-day mortality (12 and 8 mmHg in those who died and survivors respectively; P = 0·026). Oesophageal varices, splenomegaly and thrombocytopenia were not associated with any of the endpoints. In multivariable analysis, body mass index, remnant liver volume ratio and preoperative HVPG were the only independent predictors of postoperative liver dysfunction. CONCLUSION: An increased HVPG was associated with postoperative liver dysfunction and mortality after liver resection in patients with HCC and liver cirrhosis, whereas indirect criteria of PHT were not. This study suggests that preoperative HVPG measurement should be measured routinely in these patients.

Dunnigan lipodystrophy syndrome: French National Diagnosis and Care Protocol (PNDS; Protocole National de Diagnostic et de Soins).

Dunnigan syndrome, or Familial Partial Lipodystrophy type 2 (FPLD2; ORPHA 2348), is a rare autosomal dominant disorder due to pathogenic variants of the LMNA gene. The objective of the French National Diagnosis and Care Protocol (PNDS; Protocole National de Diagnostic et de Soins), is to provide health professionals with a guide to optimal management and care of patients with FPLD2, based on a critical literature review and multidisciplinary expert consensus. The PNDS, written by members of the French National Reference Center for Rare Diseases of Insulin Secretion and Insulin Sensitivity (PRISIS), is available on the French Health Authority website (in French). Dunnigan syndrome is characterized by a partial atrophy of the subcutaneous adipose tissue and by an insulin resistance syndrome, associated with a risk of metabolic, cardiovascular and muscular complications. Its prevalence, assessed at 1/100.000 in Europe, is probably considerably underestimated. Thorough clinical examination is key to diagnosis. Biochemical testing frequently shows hyperinsulinemia, abnormal glucose tolerance and hypertriglyceridemia. Elevated hepatic transaminases (hepatic steatosis) and creatine phosphokinase, and hyperandrogenism in women, are common. Molecular analysis of the LMNA gene confirms diagnosis and allows for family investigations. Regular screening and multidisciplinary monitoring of the associated complications are necessary. Diabetes frequently develops from puberty onwards. Hypertriglyceridemia may lead to acute pancreatitis. Early atherosclerosis and cardiomyopathy should be monitored. In women, polycystic ovary syndrome is common. Overall, the management of patients with Dunnigan syndrome requires the collaboration of several health care providers. The attending physician, in conjunction with the national care network, will ensure that the patient receives optimal care through regular follow-up and screening. The various elements of this PNDS are described to provide such a support.

Long-term metabolic outcome and quality of life after laparoscopic adjustable gastric banding in obese patients with type 2 diabetes mellitus or impaired fasting glucose.

BACKGROUND: The long-term outcome of type 2 diabetes mellitus after laparoscopic adjustable gastric banding (LAGB) is unknown. METHODS: A longitudinal cohort study was undertaken of patients with grade 3 obesity and type 2 diabetes or impaired fasting glucose (IFG) undergoing LAGB. Metabolic outcomes and quality of life (QoL) were assessed before and 5 years after LAGB. RESULTS: At 5 years, data for 22 out of 23 patients with type 2 diabetes and 51 out of 53 with IFG were available. Mean(s.d.) excess weight loss was 41(25) and 41(27) per cent in patients with type 2 diabetes and IFG respectively, and was associated with a significant decrease in haemoglobin (Hb) A1c, fasting and postprandial blood glucose, insulin and triglyceride levels, and in liver steatosis. There were significant increases in insulin sensitivity, beta-cell function, disposition index, high-density lipoprotein-cholesterol and QoL (Nottingham Health Profile). Good metabolic control (HbA1c 7 per cent or less) was obtained in 13 diabetic patients, but complete diabetes remission was maintained in only four. Longer duration of diabetes, and poor preoperative glucose control and beta-cell function at baseline were associated with a less favourable outcome. CONCLUSION: LAGB improved metabolic outcomes and QoL in patients with grade 3 obesity with IFG or type 2 diabetes but rarely led to prolonged remission in long-standing diabetes.

Mycophenolate mofetil monotherapy for severe side effects of calcineurin inhibitors following liver transplantation.

Withdrawal of calcineurin inhibitors (CNI) followed by mycophenolate mofetil (MMF) monotherapy after liver transplantation (LT) remains controversial due to the increased risk of acute rejection and graft loss. The aim of the present study, performed in a large cohort of liver-transplanted patients with severe CNI-induced side effects, was to assess renal function recovery, and safety in terms of liver function, of complete CNI withdrawal and replacement by MMF monotherapy. Fifty-two patients treated with MMF monotherapy for CNI-induced toxicity were analyzed. Mean estimated glomerular filtration rate (eGFR) increased significantly during the period of MMF monotherapy, from 37 +/- 10 to 44.7 +/- 15 mL/min/1.73 m(2) at 6 months (p = 0.001) corresponding to a benefit of +17.4% in renal function. eGFR stabilized or improved in 86.5%, 81% and 79% of cases, and chronic renal dysfunction worsened in 13.5%, 19% and 21% of cases, at 6, 12 and 24 months after CNI withdrawal, respectively. Only two patients experienced acute rejection. MMF monotherapy may be efficient at reversing/stabilizing CRD, and appears relatively safe in terms of liver graft function in long-term liver-transplanted patients. However, clinicians must bear in mind the potential risk of rejection and graft loss, and should be very cautious in the management of such 'difficult-to-treat patients'.

A 2-log drop in viral load at 1 month is the best predictor of sustained response in HCV patients with normal ALT: a kinetic prospective study.

The most reliable predictor of a sustained virological response in patients with persistently normal ALT has not been identified. We analysed 17 patients with genotype 1 chronic HCV who underwent therapy with pegylated interferon alfa 2b and ribavirin for 48 weeks. Two patients discontinued therapy within 28 days because of side effects and the remaining 15 patients were analysed in detail. An analysis of on treatment virological response using area under the receiver operating characteristic analyses showed that a 2 log drop in HCV RNA at day 28 was the best predictor of a sustained virological response and a failure to reduce viral load by 2 logs correctly identified patients with a low (<15%) probability of achieving a sustained virological response. Introduction of this early discontinuation rule in patients with normal ALT would allow nearly half of the patients to discontinue futile therapy at an early stage.

[Alcohol and the liver]. / L'alcool et le foie.

The risk of developing alcohol-induced liver injury occurs above a risk threshold around 25-30g/day. A dose-dependent relationship between alcohol intake and the risk of developing alcohol-induced liver injury is observed in alcoholic subjects with a daily consumption above this threshold. Cirrhosis is the first cause of death related to alcohol. Cirrhosis occurs more frequently in heavy drinkers disclosing alcoholic hepatitis or steatosis. Excess weight is a risk factor for alcoholic liver disease. The role of alcohol in mouth, oropharynx and oesophageal cancers is clearly established. The evidence for the role of alcohol in breast cancer has become clear. Corticosteroids improved the short-term survival of patients with severe alcoholic hepatitis. The Lille model identifies patients who will not benefit from corticosteroids. The benefit of liver transplantation is restricted to alcoholic patients with severe cirrhosis. Only abstainers are considered as candidates for liver transplantation. The recent development of non invasive methods opens up new perspectives for the screening of cirrhosis in subjects with heavy drinking.

Natural history and therapeutic management of recurrent hepatocellular carcinoma after liver transplantation.

While the natural history and appropriate diagnostic and management practices are relatively well defined for hepatocellular carcinoma (HCC), data are scarce concerning the characteristic features and treatment modalities for recurrent HCC after liver transplantation. The time of recurrence appears to impact survival more significantly than localization, but to date, guidelines for therapeutic management of recurrent HCC have not been established. Data in the literature shows that late and unifocal recurrence has a better prognosis when treated by surgery or radiofrequency. In the event of early recurrence, surgery cannot be recommended due to the lack of evidence and the high risk of advanced disease. Systemic therapy can be discussed in a situation of multifocal recurrence. Proliferative signal inhibitors exhibit both immunosuppressive and antiproliferative properties and liver transplantation teams tend to introduce such treatment despite the lack of extensive data.

Fatal hemophagocytic syndrome related to human herpesvirus-6 reinfection following liver transplantation: a case report.

Human herpesvirus-6 (HHV-6) has been identified as the causal agent of exanthema subitum in early childhood (also called sixth disease or roseola), a mononucleosis-like disease in adults, and as an opportunistic pathogen in transplant recipients. In the latter setting, most infections are caused by reactivation of the latent virus in the recipient and generally have a paucisymptomatic course. Only limited data on HHV-6 infection are available for liver transplant recipients. Herein we have reported a case of fatal hemophagocytic syndrome related to HHV-6 reactivation 2 weeks after liver transplantation (LT). This case suggests that this virus may be a serious and potentially life-threatening pathogen following LT.

[Nuclear receptor PPAR and hepatology: pathophysiological and therapeutical aspects]. / Récepteurs nucléaires PPAR et hépatologie: implications physiopathologiques et thérapeutiques.

In last few years, the topic of nuclear receptor has been developed in the field of hepatology allowing envisaging therapeutic strategies for the most frequent chronic liver diseases. Peroxysome proliferator-activated receptors (PPAR) contribute to wide physiological processes within the liver such as lipid/glucid metabolisms, inflammatory response, cell differenciation and cell cycle. In vitro experiments and animal studies showed that PPARalpha discloses anti-inflammatory property and PPARgamma discloses anti-inflammatory, antifibrogenic and antiproliferative properties in the liver. Main available agonists are fibrates (PPARalpha) used for 20 years in cases of lipid metabolism abnormalities and glitazones (PPARgamma) used since 2000 for type 2 diabetes. In terms of therapy, animal studies and human trials have been conducted in steatopathies. However, clinicians have to be aware of potential specific side effects related to glitazones especially on cardiovascular system.

[Alcohol, steatohepatitis, insulin resistance and hepatitis C]. / Alcool, stéatohépatite métabolique, insulinorésistance et hépatite C.

Patients with chronic hepatitis C have frequently other morbidities, either because they are frequent in the general population (metabolic syndrome) and/or because the route of contamination (chronic alcohol consumption succeeding to drug abuse). These co-morbidities have a harmfull impact on fibrosis progression during the natural history of HCV infection and reduce the efficacy of antiviral treatments. Thus, it is crucial to diagnose early and treat these different diseases which may be combined. They are the metabolic syndrome and/or chronic alcohol consumption resulting in insuline resistance, infection by the human immune deficiency virus or by the hepatitis B virus as well as chronic tobacco use or excessive consumption of cannabis. An optimal is based on a multidisciplinary approach to reduce fibrosis progression and improve the efficiency of antiviral therapies. However, the hepatologist has to come back to a global care, which is mandatory at the individual level as well as for the public health.

[Adult-onset Still's disease and hepatic angiosarcoma, a fortuitous association or a paraneoplastic syndrome: a case-report]. / Maladie de Still de l'adulte et angiosarcome hépatique, une association fortuite ou un syndrome paranéoplasique: à propos d'un cas.

Adult-onset Still's disease is a systemic disorder without specific histological feature. Diagnosis requires to rule out any other disorder including neoplasia. Nevertheless, patients with paraneoplastic adult-onset Still's disease have been reported. We report a patient with an adult-onset Still's disease who presented with a liver involvement at onset. Two years later, a liver angiosarcoma was diagnosed. This report underlines the difficulty of the diagnosis of the adult-onset Still's disease even in the presence of Yamaguchi et al.'s [J Rheumatol 19 (1992) 424-30] and Fautrel et al.'s [Medicine 81 (2002) 194-200] classification criteria and may suggest a link between the initial clinical picture and the discovery nearly two years later, of a liver angiosarcoma.

Fibrosis progression occurs in a subgroup of heavy drinkers with typical histological features.

BACKGROUND: Studies using consecutive liver biopsies constitute an attractive approach to gaining insight into the pathogenesis of alcoholic liver disease. AIM: To analyse histological factors at baseline, which are predictive of fibrosis progression and recurrence of alcoholic hepatitis. RESULTS: A total of 193 drinkers underwent consecutive biopsies at an interval of 4 years. At baseline, 20 had normal livers, 135 steatosis, five fibrosis and 33 alcoholic hepatitis. The fibrosis score increased from 1.07 +/- 0.07 to 1.7 +/- 0.94 (P < 0.001). In multivariate analysis, only steatosis (P = 0.04), alcoholic hepatitis (P = 0.0004) and stage of fibrosis (P < 0.0001) were independent predictive factors of the fibrosis score at the second biopsy. Cirrhosis developed more frequently in patients with steatosis (11%) and alcoholic hepatitis (39%) than in others (0%, P < 0.0001). Alcoholic hepatitis recurred more frequently in patients with alcoholic hepatitis at baseline: 58% vs. 15%, P < 0.0001. In multivariate analysis, alcoholic hepatitis at the first biopsy was the only predictive factor of its recurrence (P < 0.0001). CONCLUSIONS: In a large cohort of drinkers with consecutive biopsies, steatosis, fibrosis stage and alcoholic hepatitis at baseline were independent predictive factors of fibrosis progression. In terms of mechanisms, we propose a novel concept of multiple hits of alcoholic hepatitis occurring in the same patient.

The negative impact of HBV/HCV coinfection on cirrhosis and its consequences.

BACKGROUND: Hepatitis B virus (HBV)/hepatitis C virus (HCV) confection has been rarely studied in nonasian series. AIM: To compare the characteristics of HBV/HCV coinfected patients to those of HBV- or HCV-monoinfected patients in the ANRS CO22 HEPATHER cohort study. PATIENTS AND METHODS: Of the 20 936 included patients, 95 had HBV/HCV coinfection (hepatitis B surface antigen, anti-HCV antibody and HCV RNA positive) and were matched with 375 HBV- and 380 HCV-monoinfected patients on age, gender and time since HBV or HCV diagnosis. RESULTS: F3-F4 fibrosis was more frequent in coinfected patients (58%) than in HBV- (32%, P < .0001), but similar in HCV-monoinfected patients (52%, P = .3142). Decompensated cirrhosis was more frequent in coinfected patients (11%) than in HBV- (2%, P = .0002) or HCV- (4%, P = .0275) monoinfected patients. Past excessive alcohol use was more frequent in coinfected patients (26%) than in HBV (12%, P = .0011), but similar in HCV monoinfected patients (32%, P = .2868). Coinfected patients had a higher proportion with arterial hypertension (42%) than HBV- (26%) or HCV-monoinfected patients (25%) (P < .003). Multivariable analysis confirmed the association between F3-F4 fibrosis and HCV infection in HBV-infected patients (OR = 3.84, 95% CI 1.99-7.43) and the association between decompensated cirrhosis and coinfection in HBV infected (OR = 5.58, 95% CI 1.42-22.0) or HCV infected patients (OR = 3.02, 95% CI 1.22-7.44). CONCLUSIONS: HCV coinfection harmfully affects liver fibrosis in HBV patients, while decompensated cirrhosis is increased in coinfected patients compared with HBV- or HCV-monoinfected patients. HCV treatment is as safe and effective in coinfected as monoinfected patients and should be considered following the same rules as HCV monoinfected patients.

Sequential combination of glucocorticosteroids and alfa interferon versus alfa interferon alone for HBeAg-positive chronic hepatitis B.

BACKGROUND: Chronic hepatitis B has serious effects on morbidity and mortality. Alfa interferon has been shown to increase the rates of HBeAg-clearance as well as seroconversion to anti-HBe, but response rates are unsatisfactory. Glucocorticosteroid pretreatment may increase the response to alfa interferon. OBJECTIVES: The objectives were to assess the effects of the sequential combination of glucocorticosteroids and alfa interferon versus alfa interferon alone in hepatitis B 'e' antigen positive chronic hepatitis B on mortality, virological response, biochemical response, liver histology, quality of life, and adverse events. SEARCH STRATEGY: Eligible trials were identified through searches of The Cochrane Hepato-Biliary Controlled Trials Register (May 2005), The Cochrane Central Register of Controlled Trials in The Cochrane Library (Issue 2, 2005), MEDLINE (1950 to May 2005), EMBASE (Excerpta Medica Database) (1980 to May 2005), BIOSIS (1969 to May 2005), and reference lists of relevant articles. Further trials were sought through correspondence with authors of trials and pharmaceutical companies. SELECTION CRITERIA: Randomised clinical trials comparing identical alfa interferon treatment regimens with and without glucocorticosteroid pretreatment for hepatitis B 'e' antigen positive chronic hepatitis. We included trials irrespective blinding, publication status, or language. DATA COLLECTION AND ANALYSIS: Three authors selected the trials independently and one extracted the data, which were then validated. We performed assessments of the outcome measures at the end of treatment and at six months and at maximal follow-up after the end of treatment with alfa interferon. MAIN RESULTS: We included a total of 13 randomised trials with 790 patients. Loss of hepatitis B 'e' antigen (OR 1.41, 95% confidence interval 1.03 to 1.92, P = 0.03) and hepatitis B virus DNA (OR = 1.51, 95% confidence interval 1.12 to 2.05, P = 0.008) were significantly more frequent among patients treated with the sequential combination of glucocorticosteroids and alfa interferon than among patients treated with alfa interferon alone. Glucocorticosteroid pretreatment did not significantly influence seroconversion from hepatitis B 'e' antigen to antibodies to hepatitis B 'e' antigen, loss of hepatitis B surface antigen, normalisation of alanine aminotransferase/aspartate aminotransferase activities, and severity of adverse events. Glucocorticosteroid pretreatment did not significantly affect mortality and adverse events. The effect of glucocorticosteroid pretreatment on liver histology and quality of life could not be assessed due to insufficient data. AUTHORS' CONCLUSIONS: Pretreatment with glucocorticosteroids before treatment with alfa interferon in patients with hepatitis B 'e' antigen positive chronic hepatitis B may be more effective than treatment with alfa interferon alone with regard to loss of hepatitis B 'e' antigen and hepatitis B virus DNA, but evidence for effect on clinical outcomes is lacking.

Meta-analysis: evaluation of adjuvant therapy after curative liver resection for hepatocellular carcinoma.

AIM: To evaluate adjuvant modalities after curative resection for hepatocellular carcinoma using a meta-analysis of randomized and non-randomized controlled trials. METHODS: In a first step, a meta-analysis of randomized controlled trials was carried out. Sensitivity analyses after inclusion of non-randomized controlled trials were performed. Four therapeutic modalities were evaluated: pre-operative transarterial chemotherapy, post-operative transarterial chemotherapy, systemic chemotherapy and a combination of systemic and transarterial chemotherapy. RESULTS: Only post-operative transarterial chemotherapy improved survival significantly at 2 years [difference, 22.8%; confidence interval (CI), 8.6-36.9%; P = 0.002] and 3 years (difference, 27.6%; CI, 8.2-47.1%; P = 0.005), and decreased the probability of no recurrence at 1 year (difference, 28.8%; CI, 16.7-40.8%; P < 0.001), 2 years (difference, 27.6%; CI, 8.2-47.1%; P = 0.005) and 3 years (difference, 28%; CI, 8.2-47.9%; P = 0.006). In a sensitivity analysis after inclusion of non-randomized controlled trials, post-operative transarterial chemotherapy still improved survival at 1 year (difference, 9.6%; CI, 0.8-18.3%; P = 0.03), 2 years (difference, 13.5%; CI, 0.9-26%, P = 0.04) and 3 years (difference, 18%; CI, 7-28.9%; P < 0.001), and decreased the probability of no recurrence at 1 year (difference, 20.3%; CI, 7.7-33%; P = 0.002), 2 years (difference, 35%; CI, 21.4-46.3%; P < 0.001) and 3 years (difference, 34.5%; CI, 18.7-50.3%; P < 0.001). CONCLUSION: Post-operative transarterial chemotherapy improved survival and decreased the cumulative probability of no recurrence. New randomized controlled trials evaluating this modality are required.

Review article: Overview of medical treatments in unresectable hepatocellular carcinoma--an impossible meta-analysis?

BACKGROUND: Controversies surrounding medical treatment in patients with unresectable hepatocellular carcinoma continue to persist. AIM: To perform a meta-analysis of therapeutic modalities which had been evaluated in two or more randomized trials. METHODS: Fifty-two randomized trials were studied; only 30 were included. This overview identified seven therapeutic modalities which had been evaluated in two or more trials: adriamycin, 5-fluorouracil, interferon, percutaneous ethanol injection, transarterial chemotherapy, the combination of lipiodol with transarterial chemotherapy, and tamoxifen. RESULTS: Comparisons of survival between control groups showed substantial heterogeneity. There was no survival benefit at 1 year with adriamycin (mean difference 4%), 5-fluorouracil (mean difference -3%), percutaneous ethanol injection (mean difference 6%) or transarterial chemotherapy (mean difference -2%). For interferon, the survival benefit was significant with the Der Simonian & Laird method (mean difference 9%, 95% CI = 1-18%, P = 0.04) but not with the Peto et al. method (2.4 mean odds ratio, 95% CI = 0.9-6.8). The meta-analysis of tamoxifen showed a borderline survival benefit (mean difference 25%, 95% CI = 0-49%, P = 0.05). However, in sensitivity analyses, the survival benefit of tamoxifen was no longer significant. CONCLUSIONS: No treatment has clearly proven efficacy in survival. 5-Fluorouracil, adriamycin and transarterial chemotherapy were not associated with survival benefit at 1 year. The number of randomized controlled trials was insufficient to enable a conclusion to be reached for interferon and percutaneous ethanol injection. Controversy persists concerning tamoxifen efficacy. Interferon and tamoxifen require new randomized controlled trials on a larger population of patients.