Kava and the liver: bans, court rulings, the WHO assessment and warnings

2024-01-25 · 8 min read

By the kava.fyi editorial team · Last updated: 25 January 2024

Almost every piece written about kava ends with the same caveat: it may harm the liver. Where did this come from, what is known after twenty years of research and court cases, and what does it mean for someone who drinks kava? Below are the facts from regulatory documents, court rulings and peer-reviewed analyses, without reassurance beyond what the sources actually say.

Where the accusation came from: Germany and Switzerland around 2000

In Germany, kava had been approved as a medicine for anxiety, in the form of ethanolic or acetonic rhizome extracts. In the late 1990s and early 2000s, reports of liver damage began to come in from people taking such preparations. According to the timeline kept by the German medicines agency BfArM, an investigation began in November 2001, a decision was issued in June 2002 (industry reports describe it as the first withdrawal of approvals), and after disputes with manufacturers the agency withdrew the approvals again in December 2007. During the same period, restrictions or bans were introduced by, among others, Switzerland, France, the United Kingdom and Canada, and on 25 March 2002 the American FDA issued a consumer warning about a possible link between kava supplements and severe liver damage.

German courts: 2014 and 2015

Manufacturers challenged the decision. On 20 May 2014 the Administrative Court in Cologne (case 7 K 2197/12 and parallel cases) overturned the December 2007 withdrawal of approvals. From the ruling: of the 48 German reports cited by the agency, only 26 were sufficiently documented, and the material showed that kava could possibly cause harm, but not a certain causal link. The court also compared the risk with the approved alternatives, benzodiazepines, which show a very high potential for dependence even at therapeutic doses, and found that milder measures were available: a prescription requirement, dose limits, monitoring of liver function tests.

On 25 February 2015 the Higher Administrative Court of North Rhine-Westphalia upheld this ruling in eight proceedings covering 15 medicines in total. It found that the benefit-risk ratio is not unfavourable if certain changes are made to the approvals: a prescription requirement, limits on the daily dose and duration of use, regular liver checks, and avoidance of alcohol and certain medications. The court allowed an appeal to the Federal Administrative Court.

The story does not end there. According to BfArM, in August 2015 the agency amended the approvals instead of withdrawing them (including a requirement to monitor liver function tests); in January 2019 the Cologne court found these conditions largely lawful; and following an opinion from the European Medicines Agency's HMPC committee that the benefit-risk ratio was unfavourable, the agency again withdrew all approvals in December 2019. According to current BfArM information, kava-based medicines, including homeopathic preparations up to a D4 potency, are not authorised for sale in Germany. In a peer-reviewed 2015 commentary, Kuchta, Schmidt and Nahrstedt described the whole affair as an example of poorly defined raw-material identity and a lack of quality control, rather than proven toxicity.

The 2007 WHO assessment

The most extensive independent analysis was prepared by a committee convened by the World Health Organization under the chairmanship of pharmacoepidemiologist David Coulter. The report "Assessment of the risk of hepatotoxicity with kava products" (2007) identified 93 case reports, with a possible small number of duplicates. Seven patients died and 14 underwent a liver transplant. Eight cases were assessed as probably related to kava, 53 as possibly related, and most of the rest could not be assessed for lack of data. In five cases symptoms returned on rechallenge.

The committee's conclusions cut both ways. On one hand: clinical trials of kava did not reveal hepatotoxicity, most experimental studies showed no tendency to damage the liver, and most clinical reviews cast doubt on a causal link. On the other: the causality assessments give significant grounds for concern about a link between kava products and liver damage, and a higher rate of events with acetonic and ethanolic extracts than with synthetic products suggests a non-random effect, which may be caused by constituents other than kavalactones. Risk factors listed included organic extracts, excessive dosing, heavy alcohol consumption, pre-existing liver disease, cytochrome P450 enzyme polymorphisms, and combination with other medicines, particularly anxiolytics, antipsychotics and anticoagulants.

The report also cites an assessment (after Loew and Gaus, 2002) that kava has at least a 1,500-year history of relatively safe use, and that liver effects were never recorded in the ethnopharmacological data, and states that the water-based extracts drunk in the Pacific are chemically different from the raw material used in European tablets. Its recommendations: use only the root or rhizome of Piper methysticum for medicinal purposes, not the aerial parts; agree on appropriate cultivars; develop and study water-based products instead of acetonic and ethanolic extracts; do not use kava in people with liver disease or heavy alcohol consumption; warn about interactions.

A re-analysis of 26 cases

Hepatologist Rolf Teschke and colleagues assessed 26 cases that the authorities had judged to be linked to kava, using the CIOMS causality scale, in 2008. In 16 cases the link was not assessable, excluded, or unrelated to kava for lack of a temporal relationship or other reasons; in a further two, low scores gave a rating of "excluded" or "unlikely". That left eight cases with varying degrees of probability. Only one of these people had followed the recommendations on dose (up to 120 mg of kavapyrones daily) and duration (up to three months), and suffered toxic liver injury with a probable link to kava. The authors' conclusion: when used as recommended, hepatotoxicity is rare, whereas overdosing, prolonged use and combination with medications can increase the risk.

Noble cultivars, tudei and plant parts

One hypothesis holds that the problem was not kava as such but the raw material: the wrong cultivars and the wrong parts of the plant. Vanuatu regulated this through Kava Act No. 7 of 2002. It defines kava as the Piper methysticum plant or a traditional beverage obtained by a cold-water extract of its underground organs, and sorts cultivars into four schedules: noble, medicinal, two-day (tudei) and wichmannii. Only noble cultivars from plants grown for at least five years before harvest are permitted for export, and every package must state the cultivar, the island of origin and the plant parts used. The noble cultivar schedule includes, among others, Melomelo from Ambae and Borogoru from Maewo. The act does not even allow two-day cultivars to be sold on the local market.

As for plant parts: Lechtenberg and co-authors examined, in 2008, samples of products withdrawn from the German market as well as their own root and stem extracts for the alkaloid pipermethystine, suspected of toxicity. It was absent from all the root and stem samples (limit of detection 45 ppm), while the leaves contained 0.2 percent of it. The authors concluded that in the German cases involving ethanolic extracts, this compound could not have been the culprit. The American LiverTox database, on the other hand, lists pipermethystine and flavokavain B as cytotoxic in liver cell studies, and notes that kavalactones inhibit the CYP2E1 enzyme and probably other cytochrome P450 isoenzymes, which creates a risk of drug interactions. The mechanism remains unresolved.

What is known today and what the warnings say

LiverTox, a database run by the American National Institutes of Health, assigns kava category A: a well-known cause of clinically apparent liver injury. The pattern described is fatigue and nausea usually appearing 2 to 24 weeks after starting use, with hepatocellular-type injury. NCCIH sums up that various kava products have been linked to rare cases of liver damage, including serious and fatal ones, and lists among the possible causes undesirable cultivars, the wrong plant parts, alcohol, contamination, genetic differences and long-term use of high doses. In a 16-week clinical trial from 2020 (171 people, aqueous root extract, 120 mg of kavalactones twice daily), liver test abnormalities were more common in the kava group, although no case met the criteria for drug-induced liver injury. In a 2020 memorandum, the FDA found no basis for treating kava as a food ingredient generally recognised as safe, and noted that Australia has a limit of 250 mg of kavalactones daily from water-based extracts.

A consistent list of warnings emerges from these documents:

liver disease, current or past: the WHO recommends not using kava in such cases;
alcohol: listed as a risk factor by the WHO and NCCIH, and the German courts named avoiding alcohol as a condition for approval;
medications: particularly sedatives, anxiolytics, antipsychotics and anticoagulants, because of cytochrome P450 interactions;
pregnancy and breastfeeding: NCCIH points to particular risk;
dose and duration: Teschke's analysis links most of the assessed cases to overdosing, prolonged use or combination with medications.

An honest summary is this: a traditional water-based drink made from the root of a noble cultivar has a long history of use behind it and a different chemical profile from the extracts that triggered the crisis, but none of the documents cited states that the risk is zero. We describe what traditional preparation looks like in the article on preparing kava. This article is for information only and does not replace medical advice; people with liver disease or taking medication should consult a doctor before trying kava.