Do the antibiotics aminoglycosides, vancomycin, and metronidazole help people with cirrhosis and hepatic encephalopathy, a brain disorder associated with chronic liver disease?

Key messages

• Antibiotics previously thought to help people with cirrhosis manage the risk of hepatic encephalopathy do not appear to benefit the condition.

• These antibiotics possibly cause more side effects than other treatments for this condition.

• Research is needed to establish whether combining these antibiotics with other treatments may be beneficial.

What are cirrhosis and hepatic encephalopathy?

Cirrhosis is a long-term condition in which the liver becomes badly scarred. The liver normally helps clean the blood, process nutrients, and fight infections. When it is repeatedly damaged over time, the scar tissue which develops makes it harder for the liver to work properly.

People with cirrhosis may develop hepatic encephalopathy, a condition that results in poor brain functioning. The reason why people develop hepatic encephalopathy is complex, but the build-up in the blood of toxins from the gut, particularly of a compound called ammonia, plays a key role. Ammonia is produced in the gut mainly as a result of the activities of gut bacteria.

How do antibiotics work in hepatic encephalopathy?

Antibiotics are medicines used to treat infections caused by bacteria. They work by killing bacteria or stopping them from growing. Certain antibiotics may reduce the levels of ammonia in the blood by interfering with ammonia production in the gut. These antibiotics have been used to treat hepatic encephalopathy for many years, but the evidence that they help is unclear. The antibiotics considered in this review are:

• aminoglycosides (neomycin, paromomycin, and ribostamycin);

• vancomycin;

• metronidazole.

What did we want to find out?

We wanted to know if these antibiotics help to prevent or treat hepatic encephalopathy in people with cirrhosis. We analysed their effects on the number of people who died, whose symptoms of hepatic encephalopathy improved or failed to improve, and the number who experienced serious or non-serious complications.

What did we do?

We searched for studies that compared aminoglycosides, vancomycin, or metronidazole to:

• placebo – an inactive or 'dummy' medicine;

• non-absorbable disaccharides – sugars that mainly act in the gut, and are commonly used to treat some digestive and liver conditions;

• other antibiotics;

• other active agents that may help the condition.

We compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as study methods and size.

What did we find?

We found 24 studies which involved 1405 participants. Of these, 23 studies evaluated the treatment of hepatic encephalopathy, and one study explored the prevention of hepatic encephalopathy in people at risk.

We found evidence for the following comparisons:

• aminoglycosides versus placebo, non-absorbable disaccharides, other antibiotics, and other active agents – 19 studies in total;

• vancomycin versus non-absorbable disaccharides – 2 studies;

• metronidazole versus other active agents – 3 studies.

Main results

Death

• Aminoglycosides may slightly increase the risk of death compared to other active agents.

• We do not know if aminoglycosides make any difference to the risk of death compared to placebo, non-absorbable disaccharides, and other antibiotics.

• We also do not know if there is any difference in the risk of death between:

  • vancomycin and non-absorbable disaccharides;

  • metronidazole and other active agents.

Hepatic encephalopathy: number of people in whom hepatic encephalopathy failed to improve

• Aminoglycosides may make little to no difference to the number of people whose condition improves compared to non-absorbable disaccharides and other active agents.

• Metronidazole may make little to no difference to the number of people whose condition improves compared to other active agents.

• We do not know if there is any difference in the number of people whose condition improves when comparing:

  • aminoglycosides to placebo or other antibiotics;

  • vancomycin to non-absorbable disaccharides.

Serious complications

• Aminoglycosides may slightly increase the risk of serious complications compared to other active agents.

• We do not know if there is any difference in serious complications when comparing:

  • aminoglycosides to placebo or other antibiotics;

  • vancomycin to non-absorbable disaccharides.

Non-serious complications

• Aminoglycosides may slightly increase the risk of non-serious complications compared to placebo or other antibiotics.

• We do not know if there is any difference in the risk of non-serious complications when comparing:

  • aminoglycosides to non-absorbable disaccharides or other active agents;

  • metronidazole to other active agents;

  • vancomycin to non-absorbable disaccharides.

Only one study, using metronidazole, reported on health-related quality of life, so we were unable to draw any conclusions for this outcome.

What are the limitations of the evidence?

We had little to no confidence in the results of the review because the participants and researchers in some studies were aware of which treatment the participants received, which may have influenced the results, while some studies did not report on the outcomes we were interested in. The studies were generally small, and differed in how they were conducted. There were also too few studies to be certain about the outcomes.

How current is this evidence?

The evidence is current to 15 April 2025.