S1 Table : Studies on Survival Benefits from SVR
First Author and Reference / Study Type / Country / Number of Patients / Key Finding
Simmons et al 2015 (11) / Meta-analysis (31 studies) / Global / 33,360 / The adjusted hazard ratio of mortality for patients achieving SVR vs non-SVR was 0.50 (95% CI, .37–.67) in the general population, 0.26 (95% CI, .18–.74) in the cirrhotic group, and 0.21 (.10–.45) in the coinfected group. The pooled 5-year mortality rates were significantly lower for patients achieving SVR compared with non-SVR in all 3 populations.
Nyberg et al 2015(15) / Retrospective cohort study at Kaiser Permanente / USA / 24,968 / The mortality rate among HCV treated patients was 5.4% (127/2348) in patients who achieved an SVR, while for those without SVR the mortality rate was 18.9% (540/2855)
Lu et al 2015(16) / Observational study with 4 large US health systems (Chronic Hepatitis Cohort Study) / USA / 9,143 / Treatment reduced mortality (aHR = 0.40, 95% CI 0.28-0.56 for treated patients who achieved SVR; aHR = 0.69, 95% CI 0.56-0.85 for treated patients without SVR. Patients with SVR had a lower risk of HCC (aHR = 0.54, 95% CI 0.34-0.84) compared to treated patients without SVR, while treated patients without SVR had a higher risk of HCC than untreated patients (aHR = 1.97, 95% CI 1.48-2.63).
Tada et al 2016(19) / Retrospective analysis / Japan / 2,743 / The eradication of HCV reduced all-cause mortality (hazard ratio (HR), 0.265; 95% confidence interval (CI), 0.058-0.380) including non-liver-related mortality (HR, 0.439; 95% CI, 0.231-0.834), and the incidence of HCC (HR, 0.275; 95% CI, 0.156-0.448).
Jeźequel et al 2015(17) / Retrospective analysis / France / 1,381 / Median survival at 5, 10 and 15 years after the first liver biopsy was 97.4%, 93.1% and 87% in F0-F1 patients and 93.2%, 83.4% and 65.4% in F2-F4 patients. Survival was higher in treated patient with SVR than in untreated patient or treated without SVR regardless fibrosis stage. 15% rapid progression of fibrosis, and at 15 years, survival was 92% for those with an SVR versus 82% for treatment failures and 87% for those not treated.
Kobayashi et al 2014(20) / Retrospective analysis / Japan / 1,125 / The propensity score adjusted Cox proportional analysis showed that when compared with non- SVR and no treatment, achieving a SVR significantly reduced total mortality risk (HR = 0.077, CI = 0.011 to 0.550; P = 0.011) and HCC risk (HR = 0.118, CI = 0.029 to 0.476; P = 0.003).
Rutter et al 2015(22) / Retrospective analysis / Austria / 714 / Five- and 10-year mortality rates were 1.8% (10/551) and 2.7% (15/551) in the SVR group and 8.6% (14/163) and 19.1% (31/163) in the non- SVR patients (P < 0.001). In 29 patients, decompensation of liver disease [SVR: 9 (1.6%) vs. non- SVR: 20 (12.3%); P < 0.001] occurred and in 29 patients, HCC developed during follow-up [ SVR: 10 (1.8%) vs. non- SVR: 19 (11.7%); P < 0.001]. Non- SVR was an independent predictor for developing (i) HCC [HR: 2.36 (95% CI: 1.07-5.23; P = 0.034], (ii) liver-related complications [HR: 2.62; (95% CI: 1.18-5.81; P = 0.018] and (iii) mortality (HR: 3.46; 95% CI: 1.91-6.29; P < 0.001). For patients with early stages of fibrosis (F0-F2), a survival benefit of SVR patients could not be demonstrated.
Kutala et al 2015(21) / Retrospective analysis / France / 427 / Hazard rate for risk of death or liver transplantation in SVR versus non-SVR patients was 0.35 and 0.51, respectively (P = 0.019 and 0.038, respectively)
S2 Table : Studies on Liver-Related Morbidity
First Author / Study Type / Country / Number of Patients / Key Finding
McCombs et al 2015(27) / Retrospective / US / 187,860 / For patients who received treatment before FIB4>1.00, morbidity and death reduced by 41% and 36%, respectively
Ng et al 2011(12) / Systematic review (containing 67 studies) / Global / 17,025 / SVR reduced liver-related mortality among patients with chronic hepatitis C (3.3- to 25-fold), the incidence of hepatocellular carcinoma (1.7- to 4.2-fold), and hepatic decompensation (2.7- to 17.4-fold).
Singal et al 2010(13) / Meta-Analysis (assessed 26 studies) / Global / 15,621 / Among patients with advanced fibrosis who failed treatment, rates of liver-related mortality (2.73%/year; 95% CI, 1.38–4.080), HCC (3.22%/year, 95% CI, 2.02–4.42), and hepatic decompensation (2.92%/year; 95% CI, 1.61–4.22) were substantial. Patients with SVR are significantly less likely than patients who experienced treatment failure to develop liver-related mortality (relative risk [RR], 0.23; 95% CI, 0.10–0.52), HCC (RR, 0.21; 95% CI, 0.16–0.27), or hepatic decompensation (RR, 0.16; 95% CI, 0.04–0.59).
Baser et al 2015(28) / Retrospective / Turkey / 12,990 / Mortality (2.27% vs. 5.31%; P < 0.001) and HCC rates (0.69% vs. 1.96%; P < 0.001) were found to be lower for treated patients.
Nahon et al 2015(24) / Prospective cohort / France / 1,323 / Achievement of SVR versus no SVR was associated with decreased 5 year cumulative incidence of HCC (3.3% vs. 21.8%, HR=0.21 [0.13- 0.36], P<0.001) and hepatic decompensation (4.2% vs. 24.0%, HR=0.17 [0.10-0.27], P<0.001).
Labarga et al 2015(68) / Retrospective analysis / Spain / 527 / During a mean follow-up of 70.5 months, hepatic Decompensation events or liver-related deaths occurred less frequently in cured patients (4/138; 2.9%) than in treatment failures (28/201; 13.9%) or untreated (25/188; 13.3%) patients (P<0.001). Sustained HCV clearance following pegylated interferon/ribavirin therapy is associated with a reduced incidence of liver complications and death in HIV-HCV-coinfected patients.
Aleman et al 2013(26) / Prospective cohort / Sweden / 351 / Incidences of HCC, any liver complication, liver-related death, and overall death per 100 person-years were significantly lower in SVR time with 1.0, 0.9, 0.7, and 1.9, compared to 2.3, 3.2, 3.0, and 4.1 in non- SVR and 4.0, 4.9, 4.5, and 5.1 in untreated time.
S3 Table : Meta-Analysis Summary on Extrahepatic Manifestations
Extra hepatic manifestation / Studies included in the meta-analysis / Prevalence
Mixed cryoglobulinemia (MC) / 17 / Pooled prevalence estimate of MC (n=14 studies) was 32% (95% CI: 21-43%) while the prevalence in non-HCV group (n= 3 studies) was 3% (95% CI: 0-8%).
Diabetes mellitus (DM) / 16 / Pooled prevalence of DM (n= 16 studies) among HCV patients was 15% (95% CI: 13-18%) compared to 10% (95% CI: 4-15%) in the non-HCV population. The odds of having DM in patients with HCV was 50% higher compared to non-HCV group (OR=1.50; 95% CI: 1.19-1.81).
Chronic kidney disease/end stage renal disease (CKD/ESRD) / 11 / Pooled OR of developing CKD/ESRD in patients with HCV compared to non-HCV group (n=11 studies) was 1.29 (95% CI: 1.13-1.45).
B-cell lymphoma / Not specified / The risk of lymphoma was 64% higher (OR= 1.64; 95% CI: 1.18-2.11) in patients with HCV compared to non-HCV population
Lichen planus / 11 / Pooled prevalence estimate of lichen planus among HCV (n=11 studies) was 2.1% (95% CI: 1.1-3.1%) while the prevalence in non-HCV (n=3 studies) was 1.4% (95% CI: 0-3.4%).
Sjogren's syndrome / 7 / Pooled prevalence of Sjogren's syndrome (n=6 studies) was 1.11% (95% CI: 0.17-0.53%) compared to 0.11% (95% CI: 0.09-0.13%) in non-HCV group (n=1 study).
Porphyria cutanea tarda (PCT) / 7 / Pooled prevalence of PCT among HCV (n=5) was 0.7% (95% CI: 0.2-1.1%) whereas the prevalence in non-HCV group (n= 1 study) was 0.06% (95% CI: 0.05-0.07).

Source:Younossi et al 2015(14)

S4 Table : Studies on Extrahepatic Manifestations (Covering Multiple EHMs)
First Author / Study Type / Country / Number of Patients / Key Finding
El-Serag et al 2002(69) / Retrospective case control study / US / 34,204 / A significantly greater proportion of HCV-infected patients had porphyria cutanea tarda (PCT), vitiligo, lichen planus, and cryoglobulinemia. There was a greater prevalence of membranoproliferative GN among patients with HCV but not membranous glomerulonephritis (GN).
Cacoubet al 1999 (70) / Prospective single cohort / France / 1,614 / Overall prevalence of at least 1 clinical EHM was 74%. Five manifestations had a prevalence >10%: arthralgia (23%), paresthesia (17%), myalgia (15%), pruritus (15%), and sicca syndrome (11%). Four biologic abnormalities had a prevalence >5%: cryoglobulins (40%), antinuclear antibodies (10%), low thyroxine level (10%), and anti-smooth muscle antibodies (7%). Only vasculitis, arterial hypertension, purpura, lichen planus, arthralgia, and low thyroxine level were associated with cryoglobulin positivity.
Mazzaro et al 2012(32) / Retrospective / Italy / 440 / EHM were found in 33% of patients: mixed cryoglobulinemia in 29%, type II IgM k (87.6%) and type III (12.4%), purpura of the lower extremities (9%), weakness (9%) and arthralgias (8%). Peripheral neuropathy was found in 11 %, sicca syndrome in 3%, elevated levels of rheumatoid factor in 27% of cases, autoimmune thyroiditis in 1.4% of cases, nephropaty cryoglobulinemic 1.4% and non Hodgkin's lymphoma (NHL) in 4.3% of cases.
Zarebska-Michaluk et al 2010(33) / Retrospective / Poland / 340 / 210 patients with CHC (61.7%) presented at least 1 EHM, including mixed cryoglobulinemia (37.1%), thrombocytopenia (27.6%), thyroid autoimmunity (16.2%), dermatological disorders (4.1%) and type 2 diabetes (4.1%).
Cheng et al 2014(31) / Retrospective / China / 297 / Sixty-two percent (184/297) of the patients had at least one EHM, including fatigue (29.4%), type 2 diabetes mellitus (28.2%), renal involvement (12.5%), lymphadenopathy (9.6%), fever (9.4%), thyroid dysfunction (8.1%), and arthralgia (7.4%).
Coldea et al 2015(29) / Retrospective / Romania / 162 / 16 patients with endocrine manifestations (30.18%), 22 patients with crioglobulinemia (41.53%); 5 patients with skin manifestations (9.43%); 8 patients with hematological manifestations (15.09%), 2 patients other manifestations (3.77%). One third of patients present EHMs.
Yilmaz et al 2015(30) / Retrospective / Turkey / 62 / At least 1 EHM was identified in 37.1% of the patients. The most frequent clinical manifestations were type II diabetes mellitus (8.1%) and arthralgia or arthritis (8.0%).
S5 Table : Studies on Effect of SVR on Extrahepatic Manifestations
First Author / Study Type / Country / Number of Patients / Key Finding
Kawamura et al 2007(53) / Retrospective / Japan / 3,209 / In the non-interferon group, the cumulative rates of malignant lymphoma development were 0.6% at the 5th year, 2.3% at the 10th year, and 2.6% at the 15th year. The cumulative rates of malignant lymphoma development in interferon-treated patients with sustained virologic response were 0% at the 5th year, 0% at the 10th year, and 0% at the 15th year. The cumulative rates of malignant lymphoma development with persistent infection were 0.4% at the 5th year, 1.5% at the 10th year, and 2.6% at the 15th year. The malignant lymphoma development rate was higher in patients with persistent infection than in patients with sustained virologic response (P=.0159). The hazard ratio of lymphomagenesis in 1048 patients with sustained virologic response was significantly lower than in patients with persistent infection (hazard ratio: 0.13; P=.049).
Arase et al 2009(54) / Retrospective / Japan / 2,842 / SVR causes a two-thirds reduction in the risk of type 2 diabetes development in HCV-positive patients treated with INF.
Bonkovsky et al 2007(71) / Retrospective / US (HALT-C) / 1,144 / At baseline, patients had poorer scores for all eight SF-36 domains compared to healthy controls. Patients with cirrhosis had lower HRQOL scores than those with bridging fibrosis, as did patients with higher depression scores. SVR patients had significant improvements in seven domains, whereas relapsers had significant worsening in one domain. Sexual scores improved in SVR patients and decreased in relapsers (p=0.03). In multivariate analyses, improvements in HRQOL and sexual scores were significantly associated with SVR but were less striking in patients with lower depression scores.
Hassanein et al 2004(55) / Randomized study / US / 1,121 / Patients receiving peginterferon alfa-2a plus ribavirin reported better HRQOL than those receiving interferon alfa-2b plus ribavirin. These differences were statistically significant for three SF-36 domains and both FSS scores (p<=0.05). Patients receiving peginterferon alfa-2a plus placebo had the least impairment; adding ribavirin significantly decreased five domains of the SF-36 and both FSS scores. SVR was associated with improvement at follow-up on all SF-36 and FSS scores.
Romero-Gómez et al 2008(72) / Retrospective / Spain / 1,059 / The SVR rate was lower in patients with IFG and/or T2DM than in patients with normal glucose concentrations [143/325 (44%) vs. 432/734 (58.8%); P=0.002]. In the follow-up, abnormal glucose concentrations were observed in 74 of 304 (24.3%) non-responders and in 49 of 430 (11.4%) sustained responders (log-rank: 13.8; P=0.00002). Reverse stepwise logistic regression analysis identified the independent variables predictive of IFG or T2DM development as: sustained response (OR: 0.44; 95%CI=0.20-0.97; P=0.004) and fibrosis stage (OR: 1.46; 95%CI=1.06-2.01;P=0.02). Family history of DM, steatosis, gender, HCV viral load, genotype, triglycerides, cholesterol and BMI did not enter the multivariate analysis equation.
McHutchison et al 2001(73) / RCT / US / 912 / Pre-treatment, patients had significant impairment in five of eight SF-36 concepts compared to matched population norms. Sustained responders had a return to normal for four of these five concepts. Quality of life did not improve in non-responders. Improvements in histology, viral load or ALT values predicted improvements in quality of life. Sustained responders also had improvements in work functioning and productivity.
Kang et al 2005(74) / Retrospective / Taiwan / 568 / A total of 371 outpatients were enrolled, including 182 with CH-C and 189 age- and sex-matched subjects without CH-C. CH-C subjects had obviously lower educational status (P<0.01). Mean scores of domains in general health, physical functioning, role-physical, role-emotional, vitality, and mental health of the SF-36 were significantly lower in subjects with CH-C than those without CH-C (P<0.05). In an analysis of 47 CH-C patients who received and completed the whole course of antiviral treatment, mean scores of all domains were significantly lower at wk 12 of treatment compared to baseline. The scores returned to pretreatment values by the end of treatment, but were significantly increased at mo 6 after stopping the treatment. Among the 47 CH-C patients, 21 had sustained responses and 26 had non-sustained responses to antiviral treatment. Compared to pretreatment values, subjects with sustained responses had significantly lower social functioning scores at wk 12 of treatment, and scores for all SF-36 domains returned to pretreatment values, and increased significantly at mo 6 after stopping the treatment. For non-sustained virologic responders, scores of all SF-36 domains significantly decreased at wk 12 of treatment, and did not increase significantly by the end of treatment, or at mo 6 after stopping the treatment when compared to the pretreatment values.
Rasenack et al 2003(75) / Randomized study / Germany / 531 / At weeks 2 and 12, differences favoring peginterferon alpha-2a (40kD) were seen on seven of eight domains and both summary scores of the SF-36 (p < 0.05 to p < 0.01). At weeks 2, 12 and 24, patients receiving peginterferon alpha-2a (40kD) had less disabling fatigue (p < 0.01) than those receiving unmodified interferon alpha-2a.
Cacoub et al 2002(76) / Retrospective / France / 431 / Of 1614 patients, 431 met the inclusion criteria (56% male; age 49 years; 60% with significant fibrosis or cirrhosis; 46% with cryoglobulinemia). Seventy-six were untreated; of the treated patients, 83 were sustained responders, 47 relapsers and 225 non-responders. At baseline, fatigue and other extrahepatic manifestations were present in 254 (59%) and 225 (52%) patients. Fatigue was improved in 29 of 83 (35%) responders versus 75 of 348 (22%) patients with detectable hepatitis C virus (HCV)-RNA (P=0.01). The impact of virologic response on fatigue persisted after adjusting for age, gender, fibrosis stage, and depression (odds ratio: 0.34, P<0.001). A cryoglobulin was detectable in two of 34 (6%) responders versus 38 of 115 (33%) patients with detectable HCV-RNA (P<0.001).
Aghemo et al 2012(77) / Retrospective / Italy / 431 / Treatment failure (odds ratio = 2.81, 95% confidence interval = 1.39-5.67, P = 0.004) and a 10% body mass index increase (odds ratio = 6.42, 95% confidence interval = 1.69-24.3, P = 0.006) were significantly associated with the development of de novo IR.
Gragnani et al 2014(78) / Prospective / Italy / 424 / In the majority of SVR patients all MCS symptoms persistently disappeared (36 patients, 57%); in only two (3%) did definite MCS persist. All virological nonresponders were also clinical nonresponders, in spite of a transient improvement in some cases. No evolution to lymphoma was observed.
Kraus et al 2013(79) / Retrospective / Germany / 168 / When tested at least 12 months after termination of antiviral treatment, patients with sustained virologic response (SVR) had improved significantly as compared to their pretreatment performance in three of five TAP subtasks (vigilance, P < 0.001; shared attention: optical task, P < 0.001; working memory, P < 0.001). Patients who failed to eradicate the virus, however, showed no significant long-term changes in neurocognitive performance in all five subtasks assessed (0.194 < P < 0.804). In the post-treatment evaluation, neurocognitive function was significantly better in responders to the antiviral therapy as compared to nonresponders.
Gragnani et al 2015(52) / Retrospective / Italy / 35 / In 19/22 cryoglobulinaemic subjects (86%), the addition of boceprevir induced cryocrit disappearance. Cryocrit behavior was related to virological response, with improvement of symptoms upon undetectable viraemia and reappearance after virological breakthrough. The rate of sustained virological response was lower in cryoglobulinaemic patients than in patients without mixed cryoglobulinaemia (23.8% vs 70% respectively, p=0.01).
Thein et al 2005(80) / Prospective / Australia / 34 / Pretreatment cognitive function, mood status, and HRQOL were similar between the HCV patient groups. Sustained virological response (SVR) rates were similar between HCV-monoinfected (68%) and HIV/HCV-coinfected (73%) groups. SVR was associated with significant improvements in some measures of cognitive function, independent of HRQOL improvement.
Taskoparan et al 2011(81) / Retrospective / Turkey / 29 / In 15 hypertensive patients, after the 8th week of treatment, total cholesterol, LDL, HDL, hemoglobin, white blood cell, platelet and AST/ALT were significantly decreased (p<0.05). Serum triglyceride levels increased significantly (p<0.0001) and HOMA-IR decreased (p=0.07).
Byrnes et al 2012(82) / Retrospective / US / 15 / Significant reductions in basal ganglia Cho/Cr (p=0.03) and basal ganglia MI/Cr (p=0.03) were observed in sustained virological responders (SVRs, n=8), but not non-responders/relapsers (NR/R, n=6), indicative of reduced cerebral infection and/or immune activation in those who cleared virus. SVRs demonstrated significant improvements in verbal learning, memory, and visuospatial memory. A small but significant improvement in neurocognitive function secondary to the practice effect was seen in both HCV controls and HCV subjects during treatment.
Vallisa et al 2005(83) / Retrospective / Italy / 12 / In a study of 12 assessable patients, hematological response (complete and partial, 75%) were highly significantly associated with clearance or decrease in serum HCV viral load following treatment (P = .005)

S6 Table : Summary of Cost Effectiveness Studies on HCV Anti-Viral Therapy

Type of CE analysis / Number of studies / Countries
INF-free DAAscost effective versus PI+PEG+RBV / 6 Markov, 2 MCS / 8 / 4 US, 2 Spain, 1 Italy, 1 France
DAAscost effective versus PEG+RBV / 7 Markov, 3 MCS / 10 / 6 US, 1 Spain, 1 Italy, 1 UK, 1 Switzerland
PI+PEG+RBVcost effective versus PEG+RBV / 17 Markov / 17 / 5 US, 3 Italy, 3 UK, 1 Spain, 1 Portugal, 1 Singapore, 1 Australia, 1 Greece, 1 Netherlands

INF is interferon

DAAs is direct-acting antivirals

PEG+RBV is pegylated interferon and ribavirin

MCS is micro simulation

PI is protease inhibitor

Source: (84-117)

REFERENCES

68.Labarga P, Barreiro P, De Mendoza C, Soriano V. Progression to advanced liver fibrosis in HIV/HCV coinfected patients and prioritization of new hepatitis C therapies. Journal of Hepatology. 2015;62:S651-S2.

69.El-Serag HB, Hampel H, Yeh C, Rabeneck L. Extrahepatic manifestations of hepatitis C among United States male veterans. Hepatology. 2002 Dec;36(6):1439-45. PubMed PMID: 12447870. Epub 2002/11/26. eng.

70.Cacoub P, Poynard T, Ghillani P, Charlotte F, Olivi M, Piette JC, et al. Extrahepatic manifestations of chronic hepatitis C. MULTIVIRC Group. Multidepartment Virus C. Arthritis and rheumatism. 1999 Oct;42(10):2204-12. PubMed PMID: 10524695. Epub 1999/10/19. eng.

71.Bonkovsky HL, Snow KK, Malet PF, Back-Madruga C, Fontana RJ, Sterling RK, et al. Health-related quality of life in patients with chronic hepatitis C and advanced fibrosis. J Hepatol. 2007 Mar;46(3):420-31. PubMed PMID: 17196293. Pubmed Central PMCID: PMC1995129. Epub 2007/01/02. eng.

72.Romero-Gomez M, Fernandez-Rodriguez CM, Andrade RJ, Diago M, Alonso S, Planas R, et al. Effect of sustained virological response to treatment on the incidence of abnormal glucose values in chronic hepatitis C. J Hepatol. 2008 May;48(5):721-7. PubMed PMID: 18308416. Epub 2008/03/01. eng.

73.McHutchison JG, Ware JE, Jr., Bayliss MS, Pianko S, Albrecht JK, Cort S, et al. The effects of interferon alpha-2b in combination with ribavirin on health related quality of life and work productivity. J Hepatol. 2001 Jan;34(1):140-7. PubMed PMID: 11211891. Epub 2001/02/24. eng.

74.Kang SC, Hwang SJ, Lee SH, Chang FY, Lee SD. Health-related quality of life and impact of antiviral treatment in Chinese patients with chronic hepatitis C in Taiwan. World J Gastroenterol. 2005 Dec 21;11(47):7494-8. PubMed PMID: 16437722. Pubmed Central PMCID: PMC4725183. Epub 2006/01/27. eng.

75.Rasenack J, Zeuzem S, Feinman SV, Heathcote EJ, Manns M, Yoshida EM, et al. Peginterferon alpha-2a (40kD) [Pegasys] improves HR-QOL outcomes compared with unmodified interferon alpha-2a [Roferon-A]: in patients with chronic hepatitis C. PharmacoEconomics. 2003;21(5):341-9. PubMed PMID: 12627987. Epub 2003/03/12. eng.

76.Cacoub P, Ratziu V, Myers RP, Ghillani P, Piette JC, Moussalli J, et al. Impact of treatment on extra hepatic manifestations in patients with chronic hepatitis C. J Hepatol. 2002 Jun;36(6):812-8. PubMed PMID: 12044533. Epub 2002/06/05. eng.

77.Aghemo A, Prati GM, Rumi MG, Soffredini R, D'Ambrosio R, Orsi E, et al. Sustained virological response prevents the development of insulin resistance in patients with chronic hepatitis C. Hepatology. 2012 Nov;56(5):1681-7. PubMed PMID: 22619107. Epub 2012/05/24. eng.

78.Gragnani L, Fabbrizzi A, Triboli E, Urraro T, Boldrini B, Fognani E, et al. Triple antiviral therapy in hepatitis C virus infection with or without mixed cryoglobulinaemia: a prospective, controlled pilot study. Digestive and liver disease : official journal of the Italian Society of Gastroenterology and the Italian Association for the Study of the Liver. 2014 Sep;46(9):833-7. PubMed PMID: 24953206. Epub 2014/06/24. eng.