Safety and Benefits of Kava-Kava

Submission to the Committee for the Safety of Medicines 12th March 2002


On November 8th 2001 the German medicines regulator announced that it was initiating an official benefit/risk assessment of products containing dried rhizome or extracts of kava-kava (Piper methysticum rhizome "hereinafter referred to as "kava") after receiving evidence of hepatic adverse events linked to its consumption. The UK Committee for the Safety of Medicines (CSM) and the Medicines Control Agency (MCA), having considered the German initiative, also announced that they would review the efficacy and safety of kava products available to the British public.

On December 13th 2001 the Board of the British Herbal Medicine Association (BHMA) advised its members voluntarily to withdraw supplies of kava from the public and commenced its own review of information on the efficacy and safety of kava. On January 21st we submitted an initial response, taking into account information from other sources, such as the coordinated and detailed review of each of the adverse reports in German. We have also seen a later review of the adverse reports, presented at a briefing meeting at the MCA on February 12th, and the following revised submission takes this into account.

Public exposure to kava

Kava has been used therapeutically in the UK since the end of the 19th century, initially for urinary complaints. It was official in the British Pharmacopoeia 1914 ("Kavae Rhizoma" p.199), and featured in the 1983 edition of the British Herbal Pharmacopoeia published by the BHMA (Appendix 1). The MCA and CSM have confirmed that they know of no adverse effects from consumption of kava in the UK. Kava is currently available to the British public in a number of formulations, principally:

As an ingredient of three herbal medicinal products with full product licenses; as kava extract products, some imported from Germany; in the form of tinctures, mainly in combination formulas, by herbal practitioners to their patients.

Industry estimates are that 2001 consumption in the UK totaled around a million doses, with signs of a rapid increase in the months before suspension of sales. The Health Food Manufacturers Association (a major partner with the BHMA in the Herbal Registration Forum) has surveyed its members and has compiled a list of kava products available in the UK - this is attached Appendix 2.

The three UK licensed products conform with the General Sale List maximum (single) dose of 625 mg of dried rhizome and hence probably a daily dose of less than 2 g of dried rhizome.

Dosages in Germany have been higher at times. However, in recent years, as determined from successive editions of the German [1] Rote Liste (the relevant entries in the 2001 edition are attached), it appears that recommended dosages have been reduced. This may have been to bring them into line with recommendations in the earlier German Commission E monograph which states that Daily dosage: drug and preparations equivalent to 60-120 mg of kavalactones)[2].

Two prominent examples clearly illustrate the point:
Schwabe Laitan 100
100 mg per tablet of acetone extract providing 70 mg of kavalactones.
1998 dosage: Three times daily, one tablet.
2001 dosage: Once daily, one tablet.

Krewel Meuselbach Antares 120
96% ethanol extract providing 120 mg of kavalactones per tablet.
1998 dosage: 1-2 tablets daily.
2001 dosage: Twice daily half a tablet, or once daily one tablet.

It is likely that the higher doses were recommended at the time most of the adverse reactions were recorded. At these earlier levels a standard dose of 300 mg per day of a concentrated kava extract (containing 210 mg of kavalactones) is equivalent to about 4 g of dried rhizome. This assumes 5.25% of kavalactones in dried rhizome. A standard German reference text, Hagers Handbuch 1994[3], says frequently over 5%. However, the kavalactone content varies widely; in 121 cultivars from 51 Pacific islands tested in 1996, Lebot et al. [4] found 4-18% (average 11%) of kavalactones. On the other hand, our calculations suggest extraction rates of kavalactones from the dried rhizome in the main proprietary examples may be lower than expected - so for the purpose of this review we have assumed the 4 g rhizome equivalent.

Dosage levels in terms of dried rhizome could therefore be summarized as follows.

Recommended by or for Maximum daily dose (approx.) Earlier German preparations, including those likely to be involved in the cases concerned 4 g of dried rhizome or 210 mg of kavalactones. Current German preparations (with possible exceptions) amount to 2 g of dried rhizome or 120 mg of kavalactones. German Commission E list a limit of 120 mg of kavalactones; UK General Sale List has a limit (MD 625 mg of dried rhizome)2 g of dried rhizome.

Regulatory consistency is apparent, therefore, in that maximum daily dosages should be around 2 g of dried rhizome or 120 mg of kavalactones, but this was exceeded for a short period in the German market.

However, both these maximum dosages would generally be lower than in traditional use of kava as a recreational drink in various South Pacific islands and among aboriginal populations in Australia. The Australian Therapeutic Goods Administration has recently set a comparatively high upper daily dose of 250 mg of kavalactones.


A number of published pharmacological and clinical research studies demonstrate the efficacy of kava or kava extract. A copy of the monograph in Principles and Practice of Phytotherapy (Mills and Bone, Churchill Livingstone 2000), which summarizes these, is appended (Appendix 4). Some of the most critical clinical studies are summarized below.

In a randomized, placebo-controlled, double blind study, 58 patients with anxiety not caused by psychotic disorders received standardized kava extract (300 mg per day, containing 210 mg kava lactones) or placebo over a 4-week period. For patients receiving the kava extract, there was a significant reduction of anxiety as measured by the Hamilton Anxiety Scale (total score, p<0.02). The difference in anxiety between kava and placebo began in the first week and increased during the course of treatment. No adverse effects were reported for the kava extract.[5]

A standardized extract of kava was compared to benzodiazepine drugs in a randomized, controlled, double blind study. 176 outpatients were divided into three approximately equal groups. One group received kava extract equivalent to 210 mg of kavalactones per day, the second group received 15 mg of oxazepam per day and the third group received 9 mg of bromazepam. Statistical analysis showed that kava treatment was equivalent to the benzodiazepine drugs on Hamilton Anxiety Score ratings. Side effects were higher in the conventional drug groups.[6]

Although positive outcomes from the use of kava were demonstrated in the above controlled trials, none of the trials lasted for more than six weeks, the inclusion criteria were insufficiently defined and patient numbers were relatively small. These issues were addressed in a subsequent study.

In a randomized, placebo-controlled, double blind, multicentre study, 100 patients presenting with nervous anxiety, tension and restlessness of non-psychotic origin (DSM-III-R) were assessed over a period of six months. Patients were randomized to receive either 300 mg per day of a concentrated kava extract containing 210 mg of kavalactones or placebo. Comparison of the pre- and post-therapy Hamilton Anxiety Scores revealed a significant (p=0.0015) superiority of the kava treatment as against placebo. The difference between the two treatment groups was apparent even at 8 weeks (p=0.055). In addition, the accompanying depressive component was positively influenced by kava. There were no significant changes in biochemical parameters during the study period and the overall tolerability of kava was rated as excellent.[7]

In a meta-analysis published in 2000 of seven controlled clinical studies in which the Hamilton Anxiety Scale was used [8], kava was found to be superior to placebo and to lead to a significant reduction in anxiety.

In a recently published study, kava extract reduced anxiety symptoms among patients who had previously been treated with benzodiazepines for an average of 20 months.[9]

The German BAH/BPI report cites three unpublished double-blind studies but the BHMA has not had a chance to evaluate the original papers.[10] ,[11] ,[12]

Kava has a long history of relatively safe use in the Pacific Islands except for cases of serious overdosage (noted below). A recent epidemiological study found a substantially lower incidence of cancer among kava users, which contrasts strongly with the use of alcohol and tobacco.[13]


Early reports were published of adverse effects from substantial overdosage of kava. Heavy users of kava-kava beverages in Australia and Pacific islands were reported to develop a typical scaly rash due to pellagroid dermatopathy, attributed to interference with cholesterol and reversible on cessation of kava-kava ingestion.[14]

A detailed study in an Australian aboriginal community, involving clinical, physiological and biochemical assessments, showed that chronic heavy consumption of kava-kava (up to 440 g of dried rhizome per week) led to various other detrimental effects: reduced plasma levels of albumin, protein, urea and bilirubin, increased HDL cholesterol, haematuria, changes in haematological parameters, and breathlessness.[15]

However, abuse of alcohol is often widespread in such communities. As well as the obviously damaging effect of high alcohol intake there is some evidence of an interaction between kava and ethanol, in which the latter makes the kava more likely to be toxic in animals.[16]

Studies in Australia have shown kava drinkers to have markedly elevated levels of the liver enzyme gamma-glutamyltransferase (GGT) and this has been construed as further proof of the inherent hepatotoxicity of kava.[17]

However, while a survey of heavy kava drinkers in New Caledonia (8 g of kavalactones per week) did find evidence of raised GGT in about one-third, there were no signs of liver damage;[18] the authors concluded that the probable explanation for elevated GGT is enzymatic induction (as occurs with phenobarbital users)and not hepatotoxicity.

They concluded that cases of hepatotoxicity linked to kava are due to a rare immunoallergic mechanism.

All the effects so far reported result from very high consumption of kava and the record of adverse effects at lower therapeutic levels was very scanty until the reports from Germany and Switzerland.

The BfArM noted that, of the 24 reports it had received, it considered 18 to suggest a probable or possible" link between kava consumption and the adverse effect. The BfArM added a further 4 cases, 2 of which had been ranked as probable" and 2 possible".

Based on a review of the BfArM details and the BAH/BPI response, we earlier rearranged the BfArM spreadsheet to indicate the probabilities of an association, in categories of decreasing likelihood. The MCA's subsequent review, categorizing 6 cases as indicating a "probable" association with kava intake, is essentially in accordance with the BHMA submission although the role of concomitant medication is always open to further contention.

The initial BfArM conclusion was that kava is inherently hepatotoxic. Given the large number of cases where there was concomitant dosage with potentially hepatotoxic medication this does seem at least arguable. It may be that the observed cases are examples of idiosyncratic reactions to the herb. Unfortunately there is little evidence on this from in vivo or in vitro studies. The potential hepatotoxicity or otherwise of kava or its phytochemical constituents has not been studied.

However, there are some data suggesting immune-mediated or hypersensitivity reactions. A letter describing one of the Swiss cases mentioned above provides evidence that the hepatotoxicity was immune-mediated,[19] suggesting as a predisposing factor a deficiency of the drug-metabolizing enzyme CYP2D6 (which occurs in 9% of the population). A delayed skin hypersensitivity reaction to kava has also been reported.[20]

Sourcing and definition issues

The action of the BfArM followed a report published in 2000 by Rudolf Stoller of the "Pharmacovigilance Center" of the IKS.[21] His reports are included in the BfArM dossier. Apparently the Swiss cases all involved the consumption of a high dose acetone extract standardized to 70% kavalactones, the most popular product in Germany and Switzerland. This led to the banning of the acetone product in Switzerland.

In the current reports there is also a higher prevalence of reaction for the acetone extract of kava. It could be due to the popularity of this extract, but could also reflect the particular phytochemical balance of an acetone-based extract. The rationale for acetone extraction is based in part on higher levels of "active" kavalactones; whether there are losses of any other important constituents in this process is uncertain. It is also possible that synthetic kavalactones have effects different from those of kava rhizome for similar reasons. "Kavain" may refer to such synthetic products. In a recent review one product "Kavain Harras plus" cited in case 19, is stated to contain a combination of 30 mg synthetic dl-kavain and 250 mg of an ethanol extract concentrated to 8% kavalactones, delivering an additional 20 mg of kavalactones (apparently the woman concerned was taking almost twice the recommended dose of 6 tablets daily).[22]

If mode of preparation of kava is an issue, so might be the plant part used. In spite of earlier pharmacopoeial monographs specifying use of the whole (peeled or unpeeled) rhizome, in Europe kava preparations are now often apparently manufactured from the rhizome peelings, which represent a concentrated source of kavalactones. In the South Pacific the kava rhizome is often peeled and the locals drink this pale yellow rhizome and export the darker peelings. Kava preparations made from the whole peeled rhizome, as used traditionally, could be less likely to cause hepatotoxicity (given the lack of reports of liver damage from Fiji and Vanuatu).


The BHMA suggests that the benefit/risk balance for the use of kava is more positive than preliminary conclusions reached by the BfArM. Evidence of benefit is accumulating, with new papers on double-blind controlled clinical trials still appearing and growing medical attention on its anxiolytic properties. We also agree with later reviews that have reduced the incidence of "probable" associations between adverse reactions and kava intake. We suggest that a possible enzyme-induction mechanism might be triggered in susceptible individuals, most often when other factors are present. This risk could be substantially reduced with a few usage restrictions (see below).

A number of regulatory options are available in response to adverse reports on a product:

  1. Prohibition
  2. Prescription-Only Medicine (POM) status
  3. Usage restriction (maximum dosage or extract type)
  4. Label warnings (including variations for the 3 licensed products)
  5. No change

Taking account of the many millions of doses of kava consumed in Germany, and the debate about formulation and dosage factors, we would argue that incidence levels of adverse effects associated with kava are insufficient to deny public access to it. As there is almost no medical prescription of kava in the UK both the first two options would have this effect.

On the other hand, we accept that not responding to such adverse reports, especially in the case of unlicensed products, may not be acceptable (although we would also argue that there should be consistent public measures in the case of paracetemol).

As a proportionate response, we therefore propose a combination of measures (3) and (4) above. Given that the majority of products are unlicensed, we also suggest these be on the basis of guidelines and a voluntary code in the first instance, adherence being monitored and kept under review. The regulator can point to the success of the voluntary suspension of kava sales in the months following the adverse notices as evidence that such discipline can operate in the public interest. The BHMA can also point to the value of voluntary codes of conduct in building quality standards in the unlicensed sector.

We propose the following voluntary usage restrictions:

Dosage levels to be within the GSL limit, i.e. the equivalent of 2 g dried rhizome per day containing not more than 120mg kavalactones.
Kava raw material, for use as crude drug or in the preparation of extracts, to consist of whole, peeled or unpeeled, rhizome (not peelings) in accordance with established pharmacopoeial standards.

We would propose warnings similar to the following, to be added to labels for both unlicensed and licensed products.

Kava is a long established remedy with strong traditional endorsement. It shows promise for valuable modern applications in an area with few good therapeutic options. We believe the measures outlined here are proportionate and realistic and would allow the potential of the remedy to be developed in the public interest.


[1] Rote Liste Service GmbH, editor. Psychopharmaka (Section 71). In: Rote Liste 2001. Arzneimittelverzeichnis für Deutschland (einschließlich EU-Zulassungen und bestimmter Medizinprodukte [Red List 2001. Index of medicines for Germany. Aulendorf: Editio Cantor Verlag, 2001.
[2] Piperis methystici rhizoma (Kava-Kava-Wurzelstock). German Commission E Monograph published in: Bundesanzeiger No. 101 of 1 June 1990
[3] Hänsel R, Keller K, Rimpler H and Schneider G, editors. In: Hagers Handbuch der Pharmazeutischen Praxis, 5th ed. Volume 6: Drogen PZ. Berlin-Heidelberg-New York-London: Springer-Verlag, 1994:191-221
[4] Lebot V and Levesque J. Genetic control of kavalactone chemotypes in Piper methysticum cultivars. Phytochemistry 1996, 43, 397-403.
[5] E. Kinzler E, Krömer J, Lehmann E. Wirksamkeit eines Kava-Spezial-Extraktes bei Patienten mit Angst-, Spannungs- und Erregungszuständen nicht-psychotischer Genese. Arzneim.-Forsch./Drug Res. 1991;41:584-588 >
[6] Woelk H, Kapoula O, Lehrl S, Schröter K, Weinholz P. Behandlung von Angst-Patienten. Doppelblindstudie: Kava- Spezialextrakt WS 1490 versus Benzodiazepine. Z. Allgemeinmed. 1993;69:271-277
[7] Volz HP, Kieser M. Kava-kava extract WS 1490 versus placebo in anxiety disorders â€" A randomized placebo- controlled 25-week outpatient trial. Pharmacopsychiat. 1997;30:1-5
[8] Pittler MH, Ernst E. Efficacy of kava extract for treating anxiety: systematic review and meta-analysis. J Clin Psychopharmacol 2000; (1):

Appendix 1


  1. Kava Capsule: Ingredients: 250mg Standardised Kava Extract Powder per capsule Strength: 75mg Kavalactones per capsule Dosage: 1 capsule daily
  2. Kava Capsule: Qualitative and quantitative composition for Kava component. Grain alcohol extraction is used to create an extraction. This extract is placed back on the marc, meaning the actual plant material from which it was extracted is used as the carrier. Each capsule contains 500 milligrams of dried extract from kava root (which is not standardized), and there are no additional excipients used. Strength including kavalactone content Each 500 milligram capsule provides 15 different kava lactones, and approximately 4.3% to 5.3% kavalactone per capsule. Recommended daily dosage Three to six capsules daily.
  3. Kava Kava extract 150mg
    Ingredients including details of extraction solvent (qualitative and quantitative composition for Kava component): Kava extraction solvent Ethanol 90deg(v/v), Dicalcium phosphate, Microcrystalline cellulose, silicon dioxide, Stearic acid, magnesium stearate, HPMC glaze Strength including kavalactone content: 150mg kava extract 30% kavalactones Recommended daily dosage:1-3 tablets daily with food.
  4. Valerian Passiflora hops tablets
    Ingredients including details of extraction solvent (qualitative and quantitative composition for Kava component) magnesium oxide, Valerian extract, passion flower, kava kava powder (not extract), Motherwort, wild lettuce, Scullcap, hops, niacin, Thiamin, riboflavin, Microcrystalline cellulose, Dicalcium phosphate, magnesium stearate. Strength including kavalactone content 50mg Kava Kava powder (not extract) Recommended daily dosage 1 tablet daily.
  5. Kava Extract
    Extract of Kava root, ethanol, water (Ethanol content 67% v/v). Extraction solvent: Ethanol/water (86%w/w ethanol according to HAB2000). Quantitative composition of the Kava component: 0.57%w/w (typical analysis by HPLC) Qualitative composition: Product tested for; Appearance, colour, odour, taste, Identity by TLC and UV-VIS spectrometry (HAB1958), Relative density, pH, Dry residue, Ethanol content, Heavy metals, Herbicide/pesticides (Ph.Eur), Aflatoxin, Microbiology (Ph. Eur). Recommended daily dose: 15 to 20 drops (0.6 to 0.8ml) twice daily.
  6. Kira Limited kava kava
    Standardised dried herbal extract containing 120mg (one-a-day dsosage or 45mg (up to on etablet three times daily ) of kavalactones Solvent: ethanol
  7. Kava kava capsules
    200mg per capsule/30% kavalactones Extracted with 50% ethanol/water One capsule twice daily or 1 â€" 2 30 mins before retiring to bed.
  8. Kava Kava capsules
    Standardised extract. 250mg/30% kavalactones 1 â€"3 Capsules per day
  9. Kava kava 100mg kavalactone capsules
    One capsule daily
  10. Pure Kava Root Powder 100mg tablets
    Kavalactone content 4% - 8% On tablet per day
  11. Kava kava root tincture
    1.3 extract 60% ethanol/water solvent 100ml 30 drops 3 x daily
  12. Kava kava extract
    Standardised Kava kava extract tablets 150mg (equiv 1500mg) silicon dioxide/povidone/corn starch/cellulose/sodium Carboxmethyl cellulose/calcium stearate 30's 90's 1 - 3 tablets daily
  13. Valerian 2000 Complex with kava
    2000mg Std. Valerian root/Std kava Root extract equiv. 100mg providing Kavalactones 3.1mg/ Passion Flower/Hops/Chamomile/Magnesium 30s 60s One cap daily