Alcoholic liver disease (ALD) encompasses a spectrum from simple hepatic steatosis through alcoholic hepatitis (AH) to cirrhosis and hepatocellular carcinoma. Severe AH carries 28-day mortality exceeding 30% without treatment, and corticosteroid therapy remains the only pharmacological intervention with demonstrated mortality benefit in this subset, despite its limitations. The accurate identification of patients likely to respond – and the early recognition of non-response – are as important as the treatment decision itself.
The Maddrey discriminant function (DF) is calculated as 4.6 × (prothrombin time – control prothrombin time in seconds) + serum bilirubin in mg/dL. A DF score of 32 or greater identifies patients with severe AH who have a 28-day mortality of approximately 35–45% without treatment and who may benefit from corticosteroid therapy. Prednisolone 40 mg orally daily for 28 days is the standard regimen; oral prednisolone is preferred over intravenous methylprednisolone because it avoids the renal metabolism step required to convert hydrocortisone to its active form, though both are used clinically. Corticosteroids suppress the systemic inflammatory response driving AH by inhibiting NF-kappaB (NF-kappaB: nuclear factor kappa-B)-mediated transcription of pro-inflammatory cytokines including TNF (tumor necrosis factor) and IL-8 (interleukin-8), reducing hepatocyte apoptosis and necroinflammation. The absolute contraindications to steroid therapy in severe AH include active gastrointestinal hemorrhage, uncontrolled sepsis or infection, acute kidney injury (unless hepatorenal syndrome is the primary driver), and active tuberculosis or fungal infection.1
The Lille model addresses a critical limitation of starting all eligible patients on a full 28-day course: early non-responders gain no survival benefit from completing therapy and face cumulative immunosuppression risk. The Lille score is calculated at day 7 of prednisolone therapy and incorporates age, renal function, albumin, prothrombin time, bilirubin at day 0, and bilirubin at day 7. A Lille score of 0.45 or above at day 7 defines non-response, predicting 6-month mortality of approximately 75% and indicating that prednisolone should be discontinued to avoid ongoing infection risk without therapeutic benefit. Lille scores below 0.16 define complete response, with 6-month mortality under 15%, and corticosteroids should be continued for the full 28 days. Scores between 0.16 and 0.45 represent partial response – a zone where clinical judgment governs the continuation decision.3
Pentoxifylline, a non-selective phosphodiesterase inhibitor previously used for severe AH on the basis of its proposed anti-TNF and anti-fibrotic properties, is no longer recommended following the STOPAH (Steroids or Pentoxifylline for Alcoholic Hepatitis) trial. STOPAH was a large randomized controlled trial demonstrating that prednisolone significantly reduced 28-day mortality (hazard ratio 0.72) while pentoxifylline had no significant effect on 28-day, 90-day, or 1-year mortality compared to placebo.2 Pentoxifylline also failed to reduce the incidence of hepatorenal syndrome, previously its primary claimed benefit. Current EASL (European Association for the Study of the Liver) and AASLD (American Association for the Study of Liver Diseases) guidelines do not recommend pentoxifylline for AH management, and its use should be abandoned in clinical practice.4 The Model for End-Stage Liver Disease (MELD) score is used alongside DF for risk stratification and transplant listing; a MELD score above 20 with AH portends high short-term mortality and in carefully selected patients who fail medical therapy, early liver transplantation has been performed with acceptable outcomes at experienced centers.
Corticosteroids in AH substantially increase the risk of bacterial and fungal infections, which are already elevated in this population due to cirrhosis-associated immune dysregulation. Before initiating prednisolone, mandatory screening includes blood and urine cultures, chest imaging for pneumonia, and diagnostic paracentesis if ascites is present to exclude spontaneous bacterial peritonitis. Active infection that is not fully treated is an absolute contraindication. Prophylactic antifungals (fluconazole) are used in some centers, though not universally mandated. Repeat infection screening during the 28-day course is warranted if the patient deteriorates clinically.
Step 1 — Screen and stage: Calculate Maddrey DF. If DF ≥32, assess for corticosteroid contraindications (infection, active hemorrhage, renal failure, tuberculosis). Also calculate MELD for risk stratification.
Step 2 — Initiate prednisolone: If no contraindications, start prednisolone 40 mg orally daily. Ensure alcohol abstinence, nutritional support (enteral nutrition preferred if oral inadequate), and thiamine supplementation (prevent Wernicke encephalopathy).
Step 3 — Assess Lille score at day 7: Score ≥0.45 = non-response; discontinue prednisolone. Score <0.16 = complete response; continue full 28 days. Score 0.16–0.44 = partial response; clinical judgment governs continuation.
Step 4 — Plan long-term care: Alcohol cessation is the single most important determinant of long-term survival. Arrange addiction medicine referral. Evaluate for liver transplantation listing in refractory severe AH at experienced centers following center-specific protocols.
The nomenclature for fatty liver disease not attributable to alcohol has been revised from nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) to MASLD (metabolic dysfunction-associated steatotic liver disease) and MASH (metabolic dysfunction-associated steatohepatitis), reflecting the centrality of metabolic dysfunction in pathogenesis. MASLD affects an estimated 25% of the global population. The subset with MASH and significant fibrosis (stage F2 or above) carries the highest risk of cirrhosis, liver failure, and hepatocellular carcinoma. Until 2024, no pharmacological agent was approved specifically for MASH; this changed with FDA approval of resmetirom.
Lifestyle modification – caloric restriction achieving 7–10% body weight loss combined with aerobic exercise – remains the foundational and most evidence-based intervention in MASLD (metabolic dysfunction-associated steatotic liver disease)/MASH (metabolic dysfunction-associated steatohepatitis), producing histological improvement in steatohepatitis and fibrosis regression in patients achieving sustained weight loss. The pharmacological treatment landscape targets steatohepatitis, fibrosis, or both through distinct molecular mechanisms. Vitamin E (α-tocopherol) at 800 IU (international units) daily, with borderline evidence for fibrosis improvement. Its mechanism involves antioxidant reduction of reactive oxygen species generated by lipid peroxidation in steatotic hepatocytes. However, vitamin E is not recommended in diabetic MASH patients (due to absence of efficacy data in that population) or in patients with significant cardiovascular risk, and concerns about a modest increase in all-cause mortality with high-dose long-term use limit its appeal.
Pioglitazone, a thiazolidinedione that activates PPAR-gamma (peroxisome proliferator-activated receptor-gamma), improves insulin sensitivity primarily in adipose tissue, reduces free fatty acid flux to the liver, and decreases hepatic de novo lipogenesis. In MASH, pioglitazone at 30–45 mg daily has demonstrated histological improvement in steatosis, inflammation, and fibrosis in both diabetic and non-diabetic patients in randomized trials. The PIVENS (Pioglitazone versus Vitamin E versus placebo for NASH study) trial showed significant improvement in the NAFLD (now MASLD) activity score with pioglitazone vs. placebo, with a higher rate of histological resolution of NASH than placebo.5 Weight gain of 2–5 kg is nearly universal with pioglitazone, and heart failure is an absolute contraindication given its sodium-retaining effects and potential for fluid redistribution. Bladder cancer risk with long-term use (beyond 2 years) is a labeling concern, though causality remains debated. Pioglitazone is a reasonable option for MASH patients with concurrent type 2 diabetes or prediabetes where insulin sensitization is additionally beneficial.
Resmetirom (Rezdiffra), approved by the FDA in March 2024, is the first drug specifically indicated for MASH with moderate-to-severe fibrosis (stages F2–F3) in adults. Resmetirom is a selective THR-beta (THR-beta: thyroid hormone receptor-beta) agonist. THR-beta is the predominant thyroid hormone receptor isoform in the liver; THR-beta agonism increases hepatic fatty acid oxidation, reduces hepatic lipogenesis, lowers LDL (low-density lipoprotein) cholesterol, and promotes mitochondrial biogenesis without the systemic thyrotoxic effects driven by THR-alpha (receptor-alpha) activation in cardiac muscle and bone. In the MAESTRO-NASH (Multiple Ascending doses Efficacy and Safety trial of Resmetirom) phase 3 trial, resmetirom 80 mg or 100 mg daily for 52 weeks achieved MASH resolution (primary endpoint) in 26% and 30% of patients respectively versus 10% in the placebo group, and achieved fibrosis improvement of at least one stage in 24% and 26% versus 14% with placebo. Both co-primary endpoints were met.6 Resmetirom at 80 mg or 100 mg daily is taken once daily; dose selection is guided by tolerability. Principal adverse effects include nausea, diarrhea, and pruritus; the most clinically notable finding is dose-related reduction in LDL-C (LDL-cholesterol) and triglycerides (TG), which are pharmacodynamic effects of THR-β activation and not adverse events per se.
GLP-1 (glucagon-like peptide-1) receptor agonists, particularly semaglutide, have generated substantial MASH trial data. The NASH-specific semaglutide phase 2 trial demonstrated significant histological improvement in MASH resolution (59% vs. 17% placebo) but did not achieve significance for fibrosis improvement at the doses and duration studied.7 The mechanism is multifactorial: weight loss reduces hepatic fat delivery, GLP-1 receptor activation in the liver may directly reduce de novo lipogenesis and inflammatory signaling, and improved glycemic control reduces hepatic glucotoxic damage. Phase 3 MASH trials with semaglutide are ongoing. The broader class benefit for metabolic risk reduction – including cardiovascular outcomes – makes GLP-1 receptor agonists an attractive option for MASH patients with obesity and type 2 diabetes even pending MASH-specific regulatory approval.
Resmetirom is indicated for MASH with fibrosis stages F2–F3 confirmed by biopsy or non-invasive fibrosis assessment. It is not approved for compensated or decompensated cirrhosis (F4). Baseline liver function tests are required; avoid in Child-Pugh B or C cirrhosis. Monitor LFTs at baseline, 3 months, and 6 months. Drug interactions: resmetirom is a substrate of CYP3A4 – avoid strong CYP3A4 inducers (rifampin, phenytoin); dose adjust with strong inhibitors. Resmetirom lowers statin dose requirements due to LDL reduction – reassess statin dosing after initiation. Not recommended in pregnancy (thyroid hormone pathway involvement). The FDA approval is conditional on a requirement to demonstrate cardiovascular outcome safety in a post-marketing trial.
Cholestatic liver diseases are characterized by impaired bile formation or flow, resulting in accumulation of bile acids and bilirubin that cause progressive hepatocyte and biliary epithelial injury. Primary biliary cholangitis (PBC) is an autoimmune disease of small intrahepatic bile ducts that now has three sequential pharmacological lines of evidence-based therapy. Primary sclerosing cholangitis (PSC), in contrast, remains without a proven disease-modifying pharmacological treatment despite extensive trial activity, and management focuses on complication surveillance and transplant evaluation.
Ursodeoxycholic acid (UDCA) is the established first-line treatment for PBC (primary biliary cholangitis), administered at 13–15 mg/kg/day in divided doses. UDCA is a hydrophilic bile acid that displaces toxic hydrophobic bile acids (particularly chenodeoxycholic acid and deoxycholic acid) from the bile acid pool through competitive mechanisms, reducing their cytotoxic effects on cholangiocytes. UDCA also has direct cytoprotective, anti-apoptotic, and immunomodulatory effects on biliary epithelium. Approximately 40% of PBC patients achieve a complete biochemical response to UDCA (normalization of ALP (alkaline phosphatase) and bilirubin, or reduction of ALP to below 1.5 times the upper limit of normal), which predicts excellent long-term prognosis equivalent to the general population. The remaining 60% have inadequate response and require second-line therapy.8
Obeticholic acid (OCA) is an FXR (farnesoid X receptor) agonist and the established second-line agent for PBC in patients with inadequate UDCA response or UDCA intolerance. FXR is a bile acid-activated nuclear receptor that regulates bile acid synthesis, transport, and detoxification. OCA, a potent synthetic bile acid analogue, activates FXR in hepatocytes and enterocytes, suppressing bile acid synthesis via downregulation of CYP7A1 (cholesterol 7-alpha-hydroxylase – the rate-limiting enzyme in the classic bile acid synthesis pathway) and upregulating ileal bile acid binding protein, thereby reducing the enterohepatic burden of bile acids. In the POISE (PBC OCA International Study of Efficacy) phase 3 trial, OCA at 5 mg daily (titrated to 10 mg if tolerated) significantly reduced ALP and total bilirubin compared to placebo in UDCA inadequate responders over 12 months.8 The principal dose-limiting adverse effect of OCA is pruritus, occurring in up to 50–70% of patients and often severe enough to require dose reduction or discontinuation. OCA is contraindicated in decompensated cirrhosis (Child-Pugh B or C) due to a risk of hepatic decompensation reported in post-marketing data. GGT (gamma-glutamyl transferase) elevation is expected as a pharmacodynamic effect and does not indicate hepatotoxicity.
Bezafibrate, a fibric acid derivative that activates PPAR-alpha (PPAR-alpha: peroxisome proliferator-activated receptor-alpha) and to a lesser extent PPAR-gamma and PPAR-delta, has emerged as a third-line option for PBC patients with inadequate response to or intolerance of UDCA plus OCA. The BEZURSO (BEZafibrate in combination with URSO in primary biliary cirrhosis) trial demonstrated that bezafibrate 400 mg daily combined with UDCA produced significant biochemical response (normalization of ALP, AST (aspartate aminotransferase), ALT (alanine aminotransferase), bilirubin, albumin, and prothrombin time) in approximately 30% of patients versus 0% with UDCA plus placebo at 2 years, with significant pruritus improvement as a secondary benefit – making it attractive for the OCA-intolerant patient with severe pruritus.9 The mechanism in PBC involves PPAR-alpha-mediated suppression of bile acid synthesis and upregulation of bile acid transporters. Bezafibrate is not approved in the United States for PBC (it is off-label) but is widely used in Europe. Renal function must be monitored as fibrates can increase creatinine, and myopathy risk increases with statin co-administration.
Primary sclerosing cholangitis (PSC) is characterized by fibro-inflammatory stricturing of both intrahepatic and extrahepatic bile ducts, strongly associated with IBD (inflammatory bowel disease) – particularly ulcerative colitis in approximately 75% of cases. PSC has no pharmacological agent proven to halt or reverse biliary fibrosis or improve transplant-free survival. UDCA at standard (13–15 mg/kg/day) doses has not been shown to improve outcomes in PSC in adequately powered trials, and high-dose UDCA (28–30 mg/kg/day) was associated with increased adverse events in a large randomized trial and is actively contraindicated. Management priorities in PSC are therefore surveillance-oriented: ERCP (endoscopic retrograde cholangiopancreatography) with balloon dilation or short-term stenting for dominant biliary strictures, annual CCA (cholangiocarcinoma) surveillance with MRI (magnetic resonance imaging)/MRCP (magnetic resonance cholangiopancreatography) and CA 19-9, surveillance colonoscopy for IBD-associated colorectal cancer risk (annual if concurrent IBD), and timely liver transplant evaluation when MELD (Model for End-Stage Liver Disease) score approaches the listing threshold. Liver transplantation is the only definitive treatment for end-stage PSC, with 5-year post-transplant survival exceeding 85%.10
Cholestatic pruritus — driven by bile acid accumulation activating peripheral itch receptors and central opioid pathways — is among the most debilitating symptoms in PBC and PSC. First-line: cholestyramine (bile acid sequestrant, 4 g before and after morning meal) or colestipol. Second-line: rifampicin 150–300 mg daily (PXR-mediated upregulation of bile acid detoxification; monitor liver tests monthly for idiosyncratic hepatotoxicity). Third-line: naltrexone 50 mg daily (central opioid antagonism; warn patients about opioid withdrawal syndrome on initiation). Fourth-line: sertraline 75–100 mg daily. Refractory: nasobiliary drainage, MARS (molecular adsorbent recirculating system) in hospital, or liver transplant listing consideration. OCA commonly worsens pruritus; bezafibrate paradoxically improves it.
Hepatic encephalopathy (HE) is a brain dysfunction caused by liver insufficiency and/or portosystemic shunting, manifesting across a spectrum from minimal hepatic encephalopathy (MHE) – detectable only by neuropsychometric testing – through covert HE to overt hepatic encephalopathy (OHE) – clinically evident altered consciousness, cognitive impairment, or asterixis). The pathophysiology centers on gut-derived nitrogenous toxins, particularly ammonia, that accumulate due to impaired hepatic detoxification and cross the blood-brain barrier, triggering astrocyte swelling, neuroinflammation, and altered neurotransmission. Pharmacological management targets ammonia generation in the gut and reversal of identifiable precipitants.
Lactulose is a synthetic non-absorbable disaccharide hydrolyzed by colonic bacteria to lactic acid, acetic acid, and other short-chain organic acids. The resulting colonic acidification (lowering luminal pH to approximately 5–5.5) converts ammonia (NH3) to ammonium (NH4+), which is ionically charged, membrane-impermeant, and therefore trapped in the colonic lumen for fecal excretion rather than absorbed into the portal circulation.11 Additionally, the osmotic effect of lactulose accelerates intestinal transit, reducing the colonic residence time for bacterial urease activity and ammonia generation. The standard dosing is 15–30 mL orally two to four times daily, titrated to achieve two to three soft stools per day. Over-purging (more than four stools daily) causes dehydration and hypernatremia, which are themselves precipitants of HE and must be avoided. Lactulose enemas (300 mL lactulose in 700 mL water, retained for 30–60 minutes) are used in patients with acute OHE who are unable to take oral medication.13
Rifaximin is a non-absorbable rifamycin-derivative antibiotic with broad-spectrum activity against gram-positive and gram-negative aerobic and anaerobic gut bacteria. Systemic bioavailability after oral administration is less than 0.4%, concentrating its antimicrobial activity entirely within the intestinal lumen. Rifaximin reduces ammonia production by suppressing urease-producing gut bacteria including Enterobacteriaceae, Clostridium species, and other ammonia-generating anaerobes, without substantially altering the overall composition of the gut microbiome at standard doses. In the landmark RFHE3001 (rifaximin HE phase 3 registration trial) randomized controlled trial (RCT) of rifaximin 550 mg twice daily vs. placebo in patients with recurrent OHE on background lactulose therapy, rifaximin reduced the risk of HE breakthrough episode by 58% (hazard ratio 0.42) and reduced HE-related hospitalizations by 50% over 6 months.12 Rifaximin is therefore indicated as add-on therapy to lactulose for secondary prophylaxis of OHE recurrence – not as monotherapy replacing lactulose. The combination lactulose plus rifaximin is superior to either agent alone for secondary prophylaxis. Principal advantages of rifaximin over systemic antibiotics are its non-systemic profile, low resistance-selection potential, and absence of significant drug interactions.
Precipitant identification and correction is the single most impactful intervention in acute OHE and is commonly more effective than any pharmacological agent alone. The major precipitants follow the clinical mnemonic TIPS-FAILURE (each letter representing a precipitant category): TIPS (transjugular intrahepatic portosystemic shunt) insertion or shunt dysfunction, Infection (especially spontaneous bacterial peritonitis – sepsis increases ammonia load and inflammatory cytokine burden), Protein excess, Surgery or procedures, Fluid and electrolyte disturbances (hyponatremia, hypokalemia alkalosis – alkaline pH promotes NH3 formation), Alimentary bleeding (upper GI (gastrointestinal) blood is a large nitrogen load for colonic bacterial metabolism), Inappropriate sedatives and opioids (benzodiazepines are particularly potent precipitants), Loss of bowel regularity (constipation), Uremia (elevated BUN (blood urea nitrogen) increases diffusible ammonia), Renal impairment (reduces ammonia clearance and drug excretion), and Electrolyte correction (overcorrection of hyponatremia). Systematic review of each precipitant category at the time of OHE presentation is mandatory before escalating pharmacotherapy.13
Zinc deficiency is prevalent in cirrhosis due to poor nutritional intake, increased urinary zinc excretion, and competition with copper for intestinal absorption. Zinc is an essential cofactor for urea cycle enzymes including ornithine carbamoyltransferase and argininosuccinate synthetase; zinc depletion impairs hepatic ammonia disposal capacity even in the absence of overt liver failure. Zinc supplementation (220 mg zinc sulfate twice daily or equivalent) has been shown in small randomized trials to reduce blood ammonia and improve MHE parameters, though its impact on clinical OHE episodes is less well established. BCAA (branched-chain amino acids – leucine, isoleucine, valine) supplementation addresses the altered amino acid profile in cirrhosis, where BCAA are catabolized to provide energy in skeletal muscle while aromatic amino acids accumulate and compete with BCAA for transport across the blood-brain barrier, altering neurotransmitter synthesis. Oral BCAA supplementation (particularly enriched enteral formulas) has shown modest benefit in MHE and may improve nutritional status and quality of life in cirrhotic patients; its role in acute OHE management is adjunctive rather than primary.1314
1. Identify and correct precipitants: Obtain blood cultures, urine cultures, diagnostic paracentesis if ascites present. Check electrolytes, renal function, CBC, ammonia. Review medication list for benzodiazepines, opioids, sedatives. Assess for GI bleeding (rectal exam, NG aspirate if indicated).
2. Lactulose immediately: 20–30 mL every 1–2 hours until first bowel movement, then titrate to 2–3 soft stools/day. For unable-to-swallow patients, use retention enema.
3. Airway protection: Grade III–IV HE with depressed consciousness requires ICU (intensive care unit) evaluation and often endotracheal intubation for aspiration prevention before lactulose therapy can be administered safely.
4. Rifaximin on discharge: Add rifaximin 550 mg twice daily to lactulose for secondary prophylaxis after first episode of OHE requiring hospitalization. Continue indefinitely unless transplantation occurs or liver function improves significantly.
5. TIPS assessment: If TIPS-related HE, consider TIPS reduction or occlusion in consultation with IR. Discuss transplant listing if MELD appropriate.
European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Management of alcohol-related liver disease. J Hepatol. 2018;69(1):154–181.
doi:10.1016/j.jhep.2018.03.018Thursz MR, Richardson P, Allison M, et al. Prednisolone or pentoxifylline for alcoholic hepatitis. N Engl J Med. 2015;372(17):1619–1628.
doi:10.1056/NEJMoa1412278Louvet A, Naveau S, Abdelnour M, et al. The Lille model: a new tool for therapeutic strategy in patients with severe alcoholic hepatitis treated with steroids. Hepatology. 2007;45(6):1348–1354.
doi:10.1002/hep.21607Thursz MR, Richardson P, Allison M, et al. Prednisolone or pentoxifylline for alcoholic hepatitis. N Engl J Med. 2015;372(17):1619–1628.
doi:10.1056/NEJMoa1412278Sanyal AJ, Chalasani N, Kowdley KV, et al. Pioglitazone, vitamin E, or placebo for nonalcoholic steatohepatitis. N Engl J Med. 2010;362(18):1675–1685.
doi:10.1056/NEJMoa0907929Harrison SA, Bedossa P, Guy CD, et al. A phase 3, randomized, controlled trial of resmetirom in NASH with liver fibrosis. N Engl J Med. 2024;390(6):497–509.
doi:10.1056/NEJMoa2309000Newsome PN, Buchholtz K, Cusi K, et al. A placebo-controlled trial of subcutaneous semaglutide in nonalcoholic steatohepatitis. N Engl J Med. 2021;384(12):1113–1124.
doi:10.1056/NEJMoa2028395Nevens F, Andreone P, Mazzella G, et al. A placebo-controlled trial of obeticholic acid in primary biliary cholangitis. N Engl J Med. 2016;375(7):631–643.
doi:10.1056/NEJMoa1509840Corpechot C, Chazouillères O, Rousseau A, et al. A placebo-controlled trial of bezafibrate in primary biliary cholangitis. N Engl J Med. 2018;378(23):2171–2181.
doi:10.1056/NEJMoa1714519European Association for the Study of the Liver. EASL Clinical Practice Guidelines on sclerosing cholangitis. J Hepatol. 2022;77(3):761–806.
doi:10.1016/j.jhep.2022.05.011Sharma P, Sharma BC, Puri V, Sarin SK. An open-label randomized controlled trial of lactulose and probiotics for the treatment of minimal hepatic encephalopathy. Eur J Gastroenterol Hepatol. 2008;20(6):506–511.
doi:10.1097/MEG.0b013e3282f3e6f5Bass NM, Mullen KD, Sanyal A, et al. Rifaximin treatment in hepatic encephalopathy. N Engl J Med. 2010;362(12):1071–1081.
doi:10.1056/NEJMoa0907893Vilstrup H, Amodio P, Bajaj J, et al. Hepatic encephalopathy in chronic liver disease: 2014 Practice Guideline by the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver. Hepatology. 2014;60(2):715–735.
doi:10.1002/hep.27210Les I, Doval E, García-Martínez R, et al. Effects of branched-chain amino acids supplementation in patients with cirrhosis and a previous episode of hepatic encephalopathy: a randomized study. Am J Gastroenterol. 2011;106(6):1081–1088.
doi:10.1038/ajg.2011.9