Skip to content
Winnowleaf home

Supplements for Cirrhosis: BCAAs, Zinc, Probiotics and What to Avoid

For cirrhosis, no supplement replaces medical care, and several can harm the liver. BCAAs (branched-chain amino acids) reduced hepatic encephalopathy, the confusion caused by a failing liver, in a Cochrane review of 18 trials (grade B, low certainty), and probiotics helped more people recover from it (grade B). Eat often, add a night-time snack and don't cut protein. Avoid green tea extract, turmeric pills, kava and comfrey, and ask your hepatology team first.

Published How we grade evidenceEditorial policy

Cirrhosis is advanced scarring of the liver. As the scar tissue builds up, pressure rises in the portal vein, the vein that carries blood from the gut to the liver (portal hypertension). That can lead to fluid in the belly, swollen veins that bleed, and hepatic encephalopathy: confusion, poor memory, personality changes or sleep problems when the liver can't clear toxins from the blood (NIDDK). Supplements are add-ons at best. Medicines and your hepatology team come first, and some popular supplements can harm a scarred liver. Here is what the trials show.

The short version

  • Eat often, have a snack at night and don't cut protein. A 2025 US guideline advises small frequent meals, a night-time snack and no protein restriction, even with hepatic encephalopathy. Grade B (Singal 2025; Chen 2019).
  • BCAAs (branched-chain amino acids) reduced hepatic encephalopathy in a Cochrane review of 18 trials, on low-certainty evidence, with no clear effect on deaths. Grade B (Aamann 2026).
  • Probiotics helped more people recover from hepatic encephalopathy than placebo or no treatment, though most trials had a high risk of bias. Grade B (Dalal 2017).
  • Zinc, taken on top of lactulose, improved a single thinking test in mild hepatic encephalopathy. Grade C (Shen 2019).
  • Milk thistle did not change the course of liver disease in the better-run trials. Grade B for no clear benefit (Rambaldi 2005).
  • Avoid green tea extract, turmeric or curcumin pills, kava, high-dose niacin, comfrey and high-dose vitamin A, which have caused liver injury; the sources are in the avoid list below (Navarro 2014; Halegoua-DeMarzio 2023; Yu 2017; Chen 2010).

Can supplements help cirrhosis?

Only as add-ons to medical care. Doctors treat the cause of the liver damage, treat portal hypertension with medicines that lower the pressure in the portal vein, and treat hepatic encephalopathy with medicines that help lower toxins in the brain (NIDDK). Lactulose and rifaximin are medicines used for hepatic encephalopathy, and the 2025 American College of Gastroenterology (ACG) guideline says BCAAs add to how well they work (Singal 2025). NIDDK advises talking with your doctor before you take any dietary supplement, including herbal products, and before over-the-counter pain relievers or sleep aids. Every supplement decision below belongs with your hepatology team.

What should you eat with cirrhosis?

Grade B for a night-time snack. The ACG guideline advises stable outpatients to eat small, frequent meals, have a night-time snack between 7 PM and 10 PM, and drink 2 or more cups of coffee a day. It says protein should not be restricted, even in decompensated cirrhosis (when complications have started) or hepatic encephalopathy, and that vegetable protein seems better tolerated than animal protein in hepatic encephalopathy (Singal 2025). In a pooled analysis of 8 trials with 341 patients, a late evening snack improved albumin, ammonia and clotting time, and fewer people developed fluid in the belly or hepatic encephalopathy, though the Child-Pugh score (an overall severity score) did not change (Chen 2019).

Grade C for coffee. In 9 observational studies (432,133 people), each extra 2 cups a day was tied to a 44% lower risk of cirrhosis (Kennedy 2016). That is about getting cirrhosis, not about treating it.

Do BCAAs help hepatic encephalopathy?

Grade B, on low-certainty evidence. A 2026 Cochrane review of 18 trials with 934 adults found BCAAs reduced hepatic encephalopathy (risk ratio 0.79) but may have little to no effect on deaths. Ten trials gave BCAAs by mouth and 8 by drip, and the pooled result covers both. Nausea and diarrhoea occurred in 12% of people on BCAAs against 3% on control diets, though the review was very uncertain about that difference (Aamann 2026).

In an older year-long trial of 174 people with advanced cirrhosis, oral BCAAs lowered a combined measure of death and worsening compared with a milk protein (odds ratio 0.43), but not significantly compared with a sugar drink. Hospital admissions were lower, but many people did not keep taking them (Marchesini 2003). For muscle, BCAAs disappointed: in 150 people followed for 12 months, they did no better than the same amount of whey protein for muscle strength, muscle mass or frailty, and people disliked the taste more often (Hey 2024). See our BCAA page.

Does zinc help hepatic encephalopathy?

Grade C. In a pooled analysis of 4 trials (247 people with cirrhosis and mild hepatic encephalopathy), zinc added to lactulose for 3 to 6 months improved the number connection test, a timed join-the-dots test, in 3 trials (227 people). It made no clear difference to a symbol-matching test or to blood ammonia in 2 trials (137 people) (Shen 2019). In a later trial, 69 people with minimal hepatic encephalopathy took 45 mg of elemental zinc or a placebo for 12 weeks; thinking tests and quality of life improved on zinc (Janyajirawong 2021). The ACG guideline names zinc deficiency as one form that malnutrition takes in liver disease (Singal 2025).

Upper limit for adults: 40 mg of zinc a day, because more can cause copper deficiency; take more only with blood tests or your clinician's advice (NIH ODS; check your stack). See our zinc page.

Do probiotics help cirrhosis?

Grade B for hepatic encephalopathy, with caveats. A Cochrane review of 21 trials (1,420 people) found that, against placebo or no treatment, fewer people on probiotics failed to recover (10 trials, 574 people, risk ratio 0.67) and fewer developed overt hepatic encephalopathy (10 trials, 585 people). Ammonia was lower, but deaths did not change. Against lactulose, there was no clear difference. Nineteen of the 21 trials had a high risk of bias, and the products varied (Dalal 2017).

Grade C for portal pressure. Pooling 2 small trials (66 people), probiotics lowered the hepatic venous pressure gradient, the standard measure of portal pressure (Chang 2024). Taking antibiotics too? See can I take probiotics with antibiotics? and our probiotics page.

What about L-carnitine and LOLA?

Grade C for acetyl-L-carnitine. A Cochrane review of 5 trials (398 people), all run by one team in Italy and all at high risk of bias, found it lowered blood ammonia compared with placebo, but made no clear difference to quality of life or fatigue. The reviewers rated all the evidence very low quality (Martí-Carvajal 2019). Our cirrhosis protocol page shows L-carnitine as grade B for ammonia; we grade it C because this review rated every outcome very low quality. See our carnitine page.

Grade C for LOLA (L-ornithine L-aspartate), sold as a medicine in some countries. A Cochrane review found fewer cases of hepatic encephalopathy across 22 trials (1,375 people) and fewer deaths across all trials, but neither benefit held up in the trials at low risk of bias, and all the evidence was very low quality (Goh 2018).

Does vitamin D help cirrhosis?

Grade C. It corrects low levels, but trials show little beyond that. In 164 people with cirrhosis, a year of vitamin D raised blood levels (33.7 against 23.1 ng/mL on placebo) but did not improve bone density at the spine or hip (Grover 2021). In 46 trials of people with chronic liver disease of any kind (4,084 people), vitamin D did not change survival at 6 or 12 months (4 trials each), though liver enzymes fell slightly (Martinekova 2025).

Upper limit for adults: 4,000 IU of vitamin D a day, because more can raise blood calcium too high; take more only with blood tests or your clinician's advice (NIH ODS; check your stack). See our vitamin D page.

Can anything lower portal pressure?

Medicines do, and that choice belongs to your hepatology team (NIDDK). Beyond the small probiotic trials above, one curious result: in 22 people with cirrhosis, a liquid meal with dark chocolate blunted the usual rise in portal pressure after eating (up 11.5% against 26.4% with white chocolate) (De Gottardi 2012). Grade D, from a single meal in one small trial, so it says nothing about long-term use.

Do milk thistle or SAMe help?

Grade B for no clear benefit from milk thistle. Across 13 trials (915 people with alcohol-related or hepatitis B or C liver disease), it did not change deaths from any cause, complications or liver biopsies. Liver-related deaths fell when all trials were pooled, but not in the better-run trials (Rambaldi 2005). See our milk thistle page.

Grade C for no clear benefit from SAMe (S-adenosylmethionine). In the largest trial, 123 people with alcohol-related cirrhosis took SAMe by mouth or a placebo for 2 years; death or liver transplant fell from 30% to 16%, but the difference was not statistically significant, and it reached significance only when the 8 people with the most advanced cirrhosis were left out (Mato 1999). In a later 24-week trial in 37 people with alcohol-related liver disease who stopped drinking, SAMe did no better than placebo on any liver test or biopsy score (Medici 2011).

What helps muscle cramps in cirrhosis?

Grade C. A 2022 review of 7 randomized trials and 17 other studies found taurine (an amino acid) and some muscle-relaxing medicines safe and helpful in recent trials, and reported benefits with L-carnitine, BCAAs, zinc and vitamin D; results for vitamin E were contradictory (Kalia 2022). Most of these studies were small, so ask your team which to try. See our taurine page.

Which supplements should you avoid with cirrhosis?

Liver injury from supplements is real. In the US Drug-Induced Liver Injury Network, 130 of 839 cases (15.5%) were caused by herbal and dietary supplements, and the non-bodybuilding ones more often led to death or a liver transplant than medicines did (13% against 3%) (Navarro 2014). In a US survey, 4.7% of adults were taking one of six botanicals linked to liver injury: turmeric, green tea, ashwagandha, garcinia, red yeast rice or black cohosh (Likhitsup 2024). With cirrhosis, avoid:

  • Green tea extract. In 1,021 women with normal liver tests, 5.1% of those on high-dose extract developed moderate or worse liver test changes, with odds about 7 times those on placebo (Yu 2017). See our green tea extract page.
  • Turmeric or curcumin pills. The US registry found 10 cases since 2011; 5 people were hospitalized and 1 died of liver failure, and 3 of the 7 products tested also contained black pepper extract, which raises absorption (Halegoua-DeMarzio 2023). See our curcumin page.
  • Kava. It has been linked to rare cases of liver injury, some serious or fatal (NCCIH). See our kava page.
  • Comfrey and other herbs with pyrrolizidine alkaloids. These plant chemicals have blocked the small veins of the liver in people taking herbal preparations (Chen 2010), and comfrey contains as many as 14 of them (Mei 2010).
  • High-dose niacin. Upper limit for adults: 35 mg of niacin a day from supplements, because more can cause flushing and liver damage; take more only with blood tests or your clinician's advice (NIH ODS).
  • High-dose vitamin A. Upper limit for adults: 3,000 mcg (10,000 IU) of preformed vitamin A a day, because more can cause birth defects and liver damage; take more only with blood tests or your clinician's advice (NIH ODS).

Look up any other product in LiverTox, the US government's liver-injury database, and read our guide to supplements and your liver.

What we'd skip

  • Milk thistle for cirrhosis. In the better-run trials, deaths, complications and liver biopsies did not change (Rambaldi 2005).
  • SAMe. No significant effect on death or transplant in the largest trial in alcohol-related cirrhosis, and no better than placebo in a later trial (Mato 1999; Medici 2011).
  • BCAAs taken only to build muscle. A 12-month trial found BCAAs did no better than the same amount of whey protein for strength, muscle mass or frailty (Hey 2024).
  • Anything on the avoid list above, however it is marketed.

Safety and interactions

  • Tell your hepatology team about every supplement, including herbal products, and ask before over-the-counter pain relievers or sleep aids (NIDDK).
  • Rifaximin is an antibiotic. For timing probiotics around antibiotics, see can I take probiotics with antibiotics?, and ask your pharmacist.
  • BCAAs can cause nausea and diarrhoea (Aamann 2026). No interaction studies with cirrhosis medicines were found for the supplements above.
  • Bleeding risk: some supplements, such as ginkgo, can raise bleeding risk (NCCIH); with cirrhosis, ask your hepatology team before taking anything that affects clotting, including fish oil with a blood thinner (can I take fish oil with Eliquis?).
  • Zinc and vitamin D have adult upper limits (see each section). Taking several supplements? Check your stack against the upper limits.
  • Don't stop a prescribed medicine on your own; ask the prescriber or pharmacist. If you are pregnant, breastfeeding, or under 18, ask a clinician.

When to see a doctor

  • Vomiting blood, or black or bloody stools: go to a hospital right away, since swollen veins may be bleeding (NIDDK).
  • New confusion, trouble thinking, memory loss, personality changes or sleep changes: these can be signs of hepatic encephalopathy (NIDDK); call your team the same day.
  • A swelling belly, or swelling in the legs, ankles or feet: fluid can build up as portal pressure rises (NIDDK).
  • Before starting any supplement, including the ones graded here.
  • Pregnant, breastfeeding or under 18: ask a clinician before you start any supplement.

FAQ

What supplements are good for cirrhosis? None replaces medical care. BCAAs reduced hepatic encephalopathy in a Cochrane review of 18 trials (grade B, low certainty), probiotics helped more people recover from it (grade B), and zinc on top of lactulose improved a single thinking test (grade C). Ask your hepatology team before starting any.

Should I cut protein if I have hepatic encephalopathy? No. The 2025 ACG guideline says protein should not be restricted, even with hepatic encephalopathy, and that vegetable protein seems better tolerated than animal protein.

Does zinc help hepatic encephalopathy? A little. In 4 trials of people with cirrhosis and mild hepatic encephalopathy, zinc added to lactulose improved a timed join-the-dots test but not blood ammonia (grade C). The adult upper limit is 40 mg a day.

Is coffee good for cirrhosis? The ACG guideline advises 2 or more cups a day for stable outpatients, and people who drink more coffee get cirrhosis less often in observational studies (grade C).

Which supplements are bad for cirrhosis? Avoid green tea extract, turmeric or curcumin pills, kava, comfrey, high-dose niacin and high-dose vitamin A, all linked to liver injury. Check anything else in LiverTox and with your team.

Sources

  1. Singal AK, et al. ACG Clinical Guideline: Malnutrition and Nutritional Recommendations in Liver Disease. Am J Gastroenterol. 2025. PubMed 40314389 · doi:10.14309/ajg.0000000000003379
  2. Chen CJ, et al. Significant effects of late evening snack on liver functions in patients with liver cirrhosis: A meta-analysis of randomized controlled trials. J Gastroenterol Hepatol. 2019. PubMed 30883904 · doi:10.1111/jgh.14665
  3. Aamann L, et al. Branched-chain amino acids for people with cirrhosis and hepatic encephalopathy. Cochrane Database Syst Rev. 2026. PubMed 41542879 · doi:10.1002/14651858.CD001939.pub5
  4. Dalal R, et al. Probiotics for people with hepatic encephalopathy. Cochrane Database Syst Rev. 2017. PubMed 28230908 · doi:10.1002/14651858.CD008716.pub3
  5. Shen YC, et al. Zinc supplementation in patients with cirrhosis and hepatic encephalopathy: a systematic review and meta-analysis. Nutr J. 2019. PubMed 31279342 · doi:10.1186/s12937-019-0461-3
  6. Rambaldi A, et al. Milk thistle for alcoholic and/or hepatitis B or C liver diseases--a systematic cochrane hepato-biliary group review with meta-analyses of randomized clinical trials. Am J Gastroenterol. 2005. PubMed 16279916 · doi:10.1111/j.1572-0241.2005.00262.x
  7. Navarro VJ, et al. Liver injury from herbals and dietary supplements in the U.S. Drug-Induced Liver Injury Network. Hepatology. 2014. PubMed 25043597 · doi:10.1002/hep.27317
  8. Halegoua-DeMarzio D, et al. Liver Injury Associated with Turmeric-A Growing Problem: Ten Cases from the Drug-Induced Liver Injury Network [DILIN]. Am J Med. 2023. PubMed 36252717 · doi:10.1016/j.amjmed.2022.09.026
  9. Kennedy OJ, et al. Systematic review with meta-analysis: coffee consumption and the risk of cirrhosis. Aliment Pharmacol Ther. 2016. PubMed 26806124 · doi:10.1111/apt.13523
  10. Marchesini G, et al. Nutritional supplementation with branched-chain amino acids in advanced cirrhosis: a double-blind, randomized trial. Gastroenterology. 2003. PubMed 12806613 · doi:10.1016/s0016-5085(03)00323-8
  11. Hey P, et al. Branched-chain amino acid supplementation does not improve measures of sarcopenia in cirrhosis: results of a randomised controlled trial. Aliment Pharmacol Ther. 2024. PubMed 38404263 · doi:10.1111/apt.17917
  12. Janyajirawong R, et al. Efficacy of Zinc Supplement in Minimal hepatic Encephalopathy: A prospective, Randomized Controlled Study (Zinc-MHE Trial). Asian Pac J Cancer Prev. 2021. PubMed 34582657 · doi:10.31557/APJCP.2021.22.9.2879
  13. Chang G, et al. Effect of Probiotics on Portal Hypertension (PH) with Cirrhosis: A Systematic Review and Meta-Analysis. Clin Res Hepatol Gastroenterol. 2024. PubMed 38701917 · doi:10.1016/j.clinre.2024.102361
  14. Martí-Carvajal AJ, et al. Acetyl-L-carnitine for patients with hepatic encephalopathy. Cochrane Database Syst Rev. 2019. PubMed 30610762 · doi:10.1002/14651858.CD011451.pub2
  15. Goh ET, et al. L-ornithine L-aspartate for prevention and treatment of hepatic encephalopathy in people with cirrhosis. Cochrane Database Syst Rev. 2018. PubMed 29762873 · doi:10.1002/14651858.CD012410.pub2
  16. Grover I, et al. Effect of Vitamin D Supplementation on Vitamin D Level and Bone Mineral Density in Patients With Cirrhosis: A Randomized Clinical Trial. Am J Gastroenterol. 2021. PubMed 33927126 · doi:10.14309/ajg.0000000000001272
  17. Martinekova P, et al. Role of Vitamin D Supplementation in Chronic Liver Disease: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Nutr Rev. 2025. PubMed 40644459 · doi:10.1093/nutrit/nuaf117
  18. De Gottardi A, et al. Postprandial effects of dark chocolate on portal hypertension in patients with cirrhosis: results of a phase 2, double-blind, randomized controlled trial. Am J Clin Nutr. 2012. PubMed 22811444 · doi:10.3945/ajcn.112.040469
  19. Mato JM, et al. S-adenosylmethionine in alcoholic liver cirrhosis: a randomized, placebo-controlled, double-blind, multicenter clinical trial. J Hepatol. 1999. PubMed 10406187 · doi:10.1016/s0168-8278(99)80263-3
  20. Medici V, et al. S-adenosyl-L-methionine treatment for alcoholic liver disease: a double-blinded, randomized, placebo-controlled trial. Alcohol Clin Exp Res. 2011. PubMed 22044287 · doi:10.1111/j.1530-0277.2011.01547.x
  21. Kalia S, et al. Treatment of Muscle Cramps in Patients With Cirrhosis of Liver: A Systematic Review. J Clin Exp Hepatol. 2022. PubMed 35677500 · doi:10.1016/j.jceh.2021.10.147
  22. Likhitsup A, et al. Estimated Exposure to 6 Potentially Hepatotoxic Botanicals in US Adults. JAMA Netw Open. 2024. PubMed 39102266 · doi:10.1001/jamanetworkopen.2024.25822
  23. Yu Z, et al. Effect of Green Tea Supplements on Liver Enzyme Elevation: Results from a Randomized Intervention Study in the United States. Cancer Prev Res (Phila). 2017. PubMed 28765194 · doi:10.1158/1940-6207.CAPR-17-0160
  24. Chen Z, et al. Hepatic veno-occlusive disease associated with toxicity of pyrrolizidine alkaloids in herbal preparations. Neth J Med. 2010. PubMed 20558855
  25. Mei N, et al. Metabolism, genotoxicity, and carcinogenicity of comfrey. J Toxicol Environ Health B Crit Rev. 2010. PubMed 21170807 · doi:10.1080/10937404.2010.509013