Expert analyses of case reports say insufficient evidence to make causal connection

The biggest subject on the front burner in the herb community since last November is the question of the potential hepatotoxicity of kava (Piper methysticum Forst, Piperaceae). European and American regulatory officials have begun investigating kava because of case reports of liver toxicity associated with the use of the popular South Pacific herb. News stories first began surfacing in November, increasing in December and culminating at the end of the year 2001 with a front-page article in USA Today.¹ In mid-January, after The New York Times² and Washington Post³ published articles, things started picking up again, culminating in an Associated Press article that was carried in presumably hundreds of papers across the country.⁴ In March the U.S. Food and Drug Administration (FDA) issued a Consumer Advisory that was reported on by the Associated Press,⁵ CBS News, NBC Nightly News, and other major media.

Isolated case reports of hepatotoxicity associated with ingestion of kava first started becoming known in 1998.⁶ Subsequently, news of possible European regulatory action began to surface as increased reports became available in the fall of 2001. The American Botanical Council (ABC) and other groups have been collecting an extensive volume of data on this subject and have been in contact with German regulatory officials, with leading scientists at universities and phytomedicine manufacturers here and in Europe, and with industry trade organizations in Germany and the United States. What follows is an attempt to put some of the kava situation into a chronological perspective. This article will not, however, attempt to evaluate the safety issue itself; that task is being addressed in other publications, some of which are summarized below.

Kava (often referred to as kava kava in Polynesia or awa in Hawaii) is the traditional ritual tranquilizing beverage from the rootstock of a plant in the pepper family, believed to be native to the Pacific Island of Vanuatu. It has long enjoyed a reputation as the most revered medicinal and ritual herb in Polynesia, where it has been long used for ceremonial purposes, as a symbolic welcome for VIPs, as a beverage for culminating the marriage ceremony, and as a medicine for relaxation and also used to treat urinary tract infections.⁷ There have been at least 13 clinical studies conducted on 619 participants demonstrating kava’s positive effects for conditions including anxiety, mental function, reaction time, sleep quality, and peri-menopausal symptoms,⁸ with two published meta-analyses of three of the trials that met the highest criteria for evaluation, that kava is safe and more effective than placebo in treating symptoms of anxiety.^9,10 Because of its historical use and modern scientific data to demonstrate safety and efficacy, the German Commission E, a special expert committee of the German Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, or BfArM), analogous to the U.S. Food and Drug Administration, has approved kava preparations as nonprescription drugs for the treatment of anxiety disorders, stress and restlessness.¹¹

Kava in Europe

At least 28 cases of liver toxicity associated with the use of kava have been reported in Switzerland (4) and Germany (24), prompting regulatory actions by authorities.¹² The hepatic adverse event reports (AERs) in these cases include cholestatic hepatitis (inflamed liver with obstruction of bile flow), icterus (jaundice), increased liver enzymes (a sign of liver dysfunction), liver cell impairment, severe hepatitis with confluent necrosis, and irreversible liver damage (required transplant in four cases).

The Swiss government issued an advisory on kava in 2000,¹³ based on the initial four case reports, and later banned a trademarked, acetone-based kava extract called Laitan^®. This was a highly concentrated product, standardized to 70 percent kavalactone content (also sometimes referred to as kavapyrones), manufactured by a leading German phytomedicine company, W. Schwabe (Karlsruhe, Germany). While only the acetone extracts were removed from the Swiss market at this time, the government gave manufacturers of ethanolic extracts three years to market these products and monitor their safety to see if any liver problems were reported. In addition, all manufacturers of ethanolic extracts in Switzerland have been asked to substantiate safety by conducting new toxicological and clinical trials, which may pose a significant financial challenge to these (mostly small) companies.

It should come as no surprise that the initial reports in Switzerland pertain to the acetone kava extract; since it dominates the market there with an approximately 80–85 percent share. Ethanolic kava extracts comprise the balance. Ironically, the acetone extract has been the subject of the most scientific scrutiny; with at least seven controlled clinical trials on this product indicating its safety and efficacy compared to placebo.^8-10 Interestingly, the decision by the Swiss authorities was made despite the fact that in three cases other potentially hepatotoxic drugs (e.g., diclofenac) had been taken by those affected and concurrent viral infections, and alcohol use or abuse could not be ruled out in these cases.

In response to the potential for hepatoxicity, in 2000 the BfArM called for labeling of such potential risk on package inserts in kava products.¹⁴ Then, on November 8, 2001, officials at the BfArM sent letters to kava manufacturers proposing withdrawal of drug licenses for kava products, including homeopathic preparations up to a final concentration of D6.¹² Although kava is approved as a safe and effective nonprescription drug by the German Commission E, much of the kava sold in Germany also is prescribed, as many physicians consider it a safe and effective first-line remedy to treat symptoms associated with anxiety, a condition for which it is approved. The action proposed in the November 8 letter from BfArM is pending evaluation of data submitted by the companies. These submissions originally were to be made within four weeks; however, an extension was granted to December 21. During this time, the pharmaceutical giant Merck permanently ceased sales of its kava products Kavadura^® and Kytta-Kava^®, according to a story published in the Financial Times.¹⁵

On December 18, the German Nonprescription Drug Manufacturers Association (Bundesverband der Arzneimittel-Hersteller, or BAH) and the German Pharmaceutical Industry Association (Bundesverband der Pharmazeutischen Industrie, or BPI) filed comments with BfArM,¹⁶ as did several individual parties.^17,18 The BAH/BPI comments noted the following: "… in most of the reported cases, the causality between kava intake and liver reactions is not clear because further medication was used which might have caused liver toxicity. Furthermore, in many cases, detailed information on the patients’ history, co-medication, consumption of alcohol and further particulars are missing, thus not permitting a sound evaluation of these cases."

Some of these comments proposed that kava remain on the market as a prescription drug to be used under the supervision of a physician, where the patient’s medical history and potential susceptibility to liver disease and liver enzyme levels could be monitored to help reduce the potential risk of hepatotoxicity. Discussing the already-existing disclosure of potential hepatic risk in kava package inserts, as required by BfArM in 2000, Barbara Steinhoff, Ph.D., of the BAH, stated that the BAH submission "comes to the conclusion that the data presented [i.e., to the BfArM] on the benefit/risk assessment of Kava- and Kavain-containing medicinal products does not justify the withdrawal of marketing authorisations. The companies have already included risk information in their package leaflets and expert information and consider control of patients applying Kava products by a physician as an appropriate means."¹⁹

Some comments submitted to BfArM evaluated the details of the AERs, showing that many were related to the concomitant use of conventional drugs with potential hepatotoxic effects, and to other issues that raised doubts about the validity of some of the reports.²⁰ It is not certain when the BfArM will complete its review of the industry’s comments. The fate of kava on the German market has been slightly clarified at the time of this writing (mid-April 2002). In late January 2002, a special committee of the BfArM that evaluates drugs for prescription use recommended that BfArM approve kava as a prescription drug.²¹ This approval was consistent with some of the comments from German manufacturers who would rather see kava be retained on the market than to see it banned entirely. It is likely that kava will be approved on a prescription basis starting in July 2002.

Meanwhile, the German situation set off a chain reaction of scrutiny by other governments and a spate of articles in Europe, the U.S., and elsewhere. In early December, members of four major U.S. dietary supplement trade associations — the American Herbal Products Association (AHPA); the Council for Responsible Nutrition (CRN); the National Nutritional Foods Association (NNFA); and the Utah Natural Products Alliance (UNPA) — initiated a meeting with Christine Lewis Taylor, Ph.D., director of the FDA’s Office of Nutritional Products, Labeling and Dietary Supplements (ONPLDS), and other FDA officials to apprise them of the situation in Europe and the actions being taken by the American industry (i.e., retaining a board-certified toxicologist/pharmacologist to review European AERs). On December 18, in a second meeting with the industry groups, the FDA announced that it was sending a letter to physicians requesting that they submit reports of cases of liver toxicity associated with the herb to the FDA’s MedWatch system.²² The FDA letter was not intended as a public warning, but merely as a fact-finding measure. The letter was posted on the FDA website http://www.fda.gov, and generated inquiries from the media during the following days. Since FDA’s request was made, the agency has received 20 reports of possible hepatotoxicity associated with kava, per a CD-ROM released by FDA on March 22, although the validity of these reports have not been fully confirmed by FDA or a third party.²³

On December 19, the dietary supplement industry in the United Kingdom, working with officials at the Medicine Controls Agency, agreed to voluntarily suspend kava sales until such time as the question of liver toxicity is resolved.²⁴ Boots, a leading drug chain in the UK, had previously suspended sales. According to reports from the UK, there are no published cases there of liver toxicity associated with kava.

On January 9, in an action that appears to be precipitous and premature, the French government banned kava products.²⁵ The Belgian government reportedly has required kava warnings to be posted in drugstores, and the Dutch government has reportedly worked out an arrangement with industry to require more detailed label warnings on kava products. On February 4, the Irish Medicines Board, while acknowledging that there were no reports of liver AERs associated with kava in Ireland, issued a voluntary recall of kava products based on the reports in Germany and Switzerland http://www.imb.ie. On February 26, the Australian Therapeutic Goods Administration posted a Consumer Advisory, Practitioner Alert, and Kava Fact Sheet on its website.^26-28

Review and Evaluation of Kava Adverse Event Reports

A review of the cases reported in Germany and Switzerland, based on the information available so far, is instructive. Of 34 cases reported (including duplicate reports), 18 appear to be associated with the concomitant use of prescription medications, some of which are known or suspected to be hepatotoxic. In some instances, people consumed kava at higher doses than recommended by the Commission E (60–120 mg kavalactones per day)¹¹; one report shows that the subject exceeded this dose by 400 percent. Experts point to other potentially confounding co-factors that raise questions about kava’s culpability. Whether some of these AERs were caused by the simultaneous use of conventional drugs, the interaction of kava and the conventional drugs, or simply the kava products used still remains to be determined. However, what has become increasingly clear is that many of the AERs are clouded by numerous confounding variables. A detailed assessment by a board-certified toxicologist/pharmacologist on behalf of AHPA was released mid-February (described below).

Several analyses have turned a critical eye on these reports.^17,18,29,30 One analysis from Professors Johanne Schulz and Claus-Peter Siegers, of the Medical University of Lübeck in Germany, concludes that although the association of kava and hepatotoxicity was reported as early as 1990, this issue did not attract much scientific or medical attention until 1999. Despite the fact that numerous spontaneous adverse event reports suggest the potential of kava hepatotoxicity, a definite causal relationship has been confirmed in one case only. The increase in the number of spontaneous reports may be due to increased awareness of medical and regulatory officials, chronological association and aggravated pre-existing liver disease. They also point out that duplicate reports and erroneous association of kava in cases of pre-existing liver damage has occurred. The most suitable therapeutic alternatives are the class of conventional drugs known as benzodiazepines (e.g., Valium^®) and antidepressants, which in some cases are known to exert drug toxicity, including hepatotoxicity. These researchers acknowledge that kava preparations can produce adverse reactions (as noted in the Commission E monograph¹¹), and may be associated with rare serious liver toxicity. Schulz and Siegers suggest that the current data support idiosyncratic reactions as the possible mechanism rather than dose-dependent direct toxicity. They conclude that despite the wave of AERs, kava still appears to be a reasonable alternative for long-term treatment of anxiety, although suitable precautions should be taken to minimize patient risks.¹⁸

An analysis by American toxicologist/pharmacologist Prof. Donald Waller, of the College of Pharmacy at the University of Illinois at Chicago, of the approximately 30 hepatic AERs from Germany and 5 submitted to the FDA between May 1998 and September 2001 concluded that there is "no clear evidence that the liver damage reported in the U.S. and Europe was caused by the consumption of kava," and that those cases in which there is a possible association between the use of a kava extract and liver dysfunction "appear to have been hypersensitivity or idiosyncratic base responses."³⁰ However, the report’s author acknowledged that he did not have adequate medical information on all the case reports to adequately assess them. Two American case reports suggest relative safety of kava. In one case in which four prescription drugs, plus relatively high levels of kava were used (300 pills, equivalent to 45,000 mg kava per day!), there was no liver damage observed; in another case a 13-year-old girl consumed 8–10 500 mg tablets in a suicide attempt, but recovered the following morning. "From a toxicologist perspective, these two cases provide some evidence that kava itself is not a direct hepatotoxin even in extremely high concentrations."³⁰

The report concludes:

… kava when taken in appropriate doses for reasonable periods of time has no scientifically established potential for causing liver damage. However as with any pharmacologically active agent, there is always the possibility of drug interactions, preexisting disease conditions and idiosyncratic or hypersensitivity reactions, which can exacerbate the toxicity of such an agent. Increased surveillance or reports of adverse effects and judicious use of kava-derived products under the conditions recommended by the natural products industry would be a most prudent approach to confirm its safety and minimize any risk of liver damage. The medical community and the general public should be made aware that concomitant intake of prescription drugs associated with liver damage, excessive alcohol consumption and preexisting liver disease or hepatitis with compromised liver function are conditions which may preclude any kava consumption.³⁰

Another analysis of the hepatotoxicity issue was published in the Deutsche Apotheker Zeitung (German Pharmacists Journal) in February by Mathias Schmidt and Adolf Narstadt.¹⁷ They noted that the four Swiss cases were "exemplary" insofar as their documentation was concerned, but described the German cases as "far from complying with the current requirements of the relevant European guidelines" for AER reporting, concluding that "the allocation to Kava [of hepatotoxicity] is not logical in the majority of cases and appears arbitrary. Furthermore, in many cases, available information is not taken into account, e.g., with regard to other causes." An "extreme" example they cite is the one fatality where "it was evidently known … that the cause of liver failure was due to many years of alcohol abuse and that Kava was not involved in the development of liver symptoms — the biopsy showed that cirrhotic processes in the liver had started long before kava was taken!"

This analysis noted that there were no cases reported in France, Belgium, Austria, Great Britain and Canada. The total number of documented cases was 36 (including duplicate listings and 4 cases reported to the FDA). Table 1 shows how Schmidt and Narstadt evaluated these cases.

The analysis also compared risk of kava use to the potential risk of using conventional pharmaceutical anxiolytic drugs (e.g., benzodiazepines, which are not generally considered hepatotoxic). The paper estimates that approximately 250 million daily doses of ethanolic kava extract products have been provided in the previous decade, with only two cases in which a causal relationship appears probable, with one resulting from overdosing at levels exceeding the recommended dose (although this case was included in the following statistical analysis). Based on these two cases, the authors computed an incidence rate of 0.008 AERs to kava in 1 million daily doses. This was compared to known reports for benzodiazepines: 0.90 AERs for 1 million daily doses of bromazepam, 1.23 for oxazepam, and 2.12 cases for diazepam. The authors thus concluded, "A prohibition of Kava and a consequent increase in change-over of patients to benzodiazepines would therefore increase the risk of adverse events and not reduce them."¹⁷

The Schmidt and Narstadt paper also noted that at least eight toxicological studies on kava extracts, as well as the isolated kavalactones, showed no evidence of hepatotoxic effects, including an in vitro study on human liver cells; nor is there any evidence of liver dysfunction associated with long-term kava use in the ethnopharmacological literature.

In a paper published in the same issue of the German Pharmacists Journal, Prof. Dieter Loew provides a general review of kava chemistry, pharmacology, clinical studies and toxicology.²⁹ He notes that kava has had at least a 1,500 year history of relatively safe use, with liver side effects never having arisen in the ethnopharmacological data. While acknowledging that this is not proof for the non-existence of kava toxicity, "the widespread ritual use … cannot be ignored in the assessment of risks of side effects." He concludes, "In the controlled clinical studies and post-marketing surveillance studies, subjective and objective organ-related adverse drug reactions were rarely or very rarely documented. The hepatotoxic reactions reported recently are single cases, whereby the underlying diseases and co-medication [other simultaneously used drugs] suggest that a causal relationship is doubtful." He continues that "an immune-allergic mechanism may be possible … In view of the current scientific knowledge, the hasty condemnation of Kava extracts appears neither logical nor justifiable. Professional judgement and correct risk/benefit analyses are lacking."

AERs in the U.S.

Regarding kava in North America, the information is somewhat confusing. In November 2001, both the ABC and AHPA checked the FDA database of AERs related to herbals and dietary supplements and found no reports of adverse liver effects associated with kava in the U.S. There were, however, 35 AERs suggesting other types of adverse effects purportedly associated with kava use (as of November 26, 2001). A closer inspection of those AERs reveals that 29 of them are based on the infamous "fX" case in which a young man doled out ersatz "kava" doses at a New Year’s Eve rave in Los Angeles in 1996. The fraudulent products contained 1,4-butanediol and contained no kava or any other herbs or natural dietary supplement ingredients.³¹ This case received widespread publicity, the perpetrator’s parents’ business (where he stealthily manufactured the illicit product) was closed by health authorities, and he was tried and convicted. Thus, these AERs have nothing to do with kava and should not even be reported as kava AERs in the FDA database. AHPA has contacted FDA and asked the agency to remove these erroneous and highly misleading reports;³² Christine Taylor, of the FDA’s ONPLDS, has advised AHPA that the FDA press office has been directed to inform the press that these are not kava-related cases.

The New York Times article on January 16 stated that the FDA had 60 kava-related adverse event reports since 1998.² The ABC contacted the FDA and was told that the agency has only 26 AERs related to kava-containing products, with only 5 related to liver.³³ Fortunately, the story on January 22 in the Washington Post correctly reported the 26 cases.³ A group of 26 AERs reflecting reports to the FDA from March 1998 to August 2001 implicating products in which kava is an ingredient (some contain multiple ingredients, raising questions regarding the effect kava may have had on the reported adverse event) were obtained through the Freedom of Information Act process by AHPA President Michael McGuffin. Of these 26 reports, only 5 reported some type of liver disorder.³⁴

How, then, can one resolve the apparent discrepancy in the FDA’s reports on kava, from the 35 on the FDA’s website in November (including the "fX" reports), to the 26 obtained by AHPA in January, to the 60 reported by The New York Times in January? The answer lies in the fact that the reports on the FDA website had not been updated since October 1998. And, to complicate matters further, the Times article included the 29 "fX" cases, even though the reporter had been told previously by this author that the "fX" cases were not associated with kava and should not be listed on the FDA database under "kava." As previously noted, on March 25, 2002, the FDA released to ABC and to various dietary supplement trade associations its revised list of 20 possible hepatic adverse events associated with kava. ²³ The validity of these cases had not yet been confirmed by FDA or any third party.

Kava Safety and Precautions

The kava-liver safety issue comes as somewhat of a surprise to most industry members, health professionals, and consumers familiar with the history and clinical use of kava. There is little evidence in the scientific literature to suggest liver toxicity; and the herb has enjoyed safe use for centuries in the South Pacific. The two meta-analyses on clinical trials on the acetone-based kava extract from Germany do not mention liver toxicity as one of the adverse effects noted in clinical trials, concluding that kava appears to be a safe and effective remedy for anxiety.^9,10 In October, a Duke University clinical trial to assess the adverse effects profile of an American kava extract (KavaPure^®, PureWorld, South Hackensack, New Jersey) found only insignificant elevations of liver enzymes in 3 of the 38 subjects in the trial (one already with elevated levels at the beginning), concluding that kava is relatively safe.³⁵

No one seems to be certain what’s going on. Contamination has been ruled out as the products implicated came from various sources. No definitive mechanism explains the alleged toxicity although several are speculated. Government officials in Vanuatu blame the European extracts, saying that the traditionally used kava beverages do not produce liver problems.³⁶ This assertion has not been proven scientifically. Some have suggested that the acetone solvent used to make the leading European kava extract may be at fault, but analysis of the case reports shows that five of the six most serious cases are related to ethanolic (alcohol) extracts. This writer cautions all concerned not to suggest that differences in products have any toxicological significance until all the evidence has been properly evaluated.

Because of the confusion and the potential seriousness of the issue, on December 20, 2001, ABC issued a press release explaining the situation at the time and suggesting that the public adhere to the following cautions:

• Kava should not be used by anyone who has any liver problems, or by anyone who is taking any drug products with known adverse effects on the liver, or anyone who is a regular consumer of alcohol.

• Since the reports so far are associated with chronic use, kava should not be taken on a daily basis for more than four weeks (without the advice of a qualified professional).

• In addition, consumers should discontinue use if symptoms of jaundice (e.g., dark urine, yellowing of the eyes) occur.

• Consumers should consult their primary healthcare provider if they have a history of liver problems or suspect possible liver problems before using kava or continuing its use.³⁷

The ABC precautions were published in the USA Today and Associated Press articles. ABC deemed these cautions reasonable and prudent, based on the information available at the time ABC issued the precautions (December). (The complete text, plus new information, can be viewed on the ABC website http://www.HerbalGram.org). AHPA issued similar information for consumers http://www.ahpa.org. As noted above, AHPA retained a board-certified toxicologist/pharmacologist from a leading university to ascertain the nature of the possible relationship between kava consumption and liver problems.³⁸ AHPA’s McGuffin was quoted as saying, "Despite the fact that the kava products under scrutiny are ones manufactured and sold in Europe, we believe that it is critical that kava’s long history of safe use be re-affirmed by a review of the information." In February 2002, AHPA posted a summary of this review on its website.

On March 25, the FDA released a Consumer Advisory and a notice to health professionals on its website,³⁹ which was promptly picked up by the Associated Press⁵ and other major news media, including CBS Evening News (March 25) and NBC Nightly News (March 26). The FDA Advisory contains most elements of ABC’s warning of December 20.

The following day (March 26), AHPA released its revised trade recommendations for kava product labels, revising its previous recommendations for kava labeling, originally issued in 1997:

AHPA recommends the following dosage and labeling policies for food and dietary supplement products containing kava (Piper methysticum):

• Products containing kava should be formulated and labeled to limit consumption of total kavalactones to 300 mg per day;

• Labels of food and dietary supplement products containing kava should bear the following or significantly similar statement:

Caution: Ask a healthcare professional before use if you have or have had liver problems, frequently use alcoholic beverages, or are taking any medication. Stop use and see a doctor if you develop symptoms that may signal liver problems (e.g., unexplained fatigue, abdominal pain, loss of appetite, fever, vomiting, dark urine, pale stools, yellow eyes or skin). Not for use by persons under 18 years of age, or by pregnant or breastfeeding women. Not for use with alcoholic beverages. Excessive use, or use with products that cause drowsiness, may impair your ability to operate a vehicle or heavy equipment.⁴⁰

The AHPA label recommendations were to be implemented in its member companies’ next round of printing of kava labels or within six months, whichever came first. CRN has also provided labeling guidelines to its members. The advisory to AHPA members also noted that, "It is important to emphasize that no actual relationship between the use of kava and any liver problem has been established, nor has FDA made any assertion to that effect. Rather, the revision to AHPA’s label is offered in order to assure that consumers of kava are informed of the potential risk identified by case reports."⁴⁰

Members of the U.S. herb industry have taken the kava issue very seriously. Because German regulators — Germany has led the world in clinical research and rational regulation of herbal medicines —believe that there may be a problem, U.S. industry members and the FDA have dealt with the situation carefully and forthrightly. AHPA sponsored the expert evaluation of the AERs and issued revised label warnings; the FDA issued its Consumer Advisory. At least three independent expert reviews have concluded that kava’s being anecdotally linked to cases of liver dysfunction lacks adequate scientific evidence to confirm a causal relationship.

Standard medical therapy for anxiety relies on conventional pharmaceutical drugs, particularly benzodiazepines, which are not generally deemed hepatotoxic. Kava preparations have demonstrated safety and efficacy in treating anxiety in controlled clinical trials and some analyses suggest that the incidence of hepatic risk with kava may be similar to benzodiazepines. It is highly possible that kava could be approved as a drug, if it were to undergo a formal risk-benefit assessment. However, the high cost of a new drug approval at FDA (estimated at $500 million per drug) and lack of patent protection basically preclude such attempts by anyone wishing to market kava as a drug in the U.S. Thus, at this time, dietary supplement status for kava appears to be one of the few options for anxiety sufferers wanting a relatively safe and effective natural product, so long as they follow reasonable precautions.

This is one of the most serious challenges ever experienced by the herb community. New developments on this issue will be posted on the ABC website (www.herbalgram.org) as they arise.

This article was revised and expanded from two previous versions published in Whole Foods (March 2002) and Texas Pharmacy (March 2002).

References:

1. Rubin R. Herb kava linked to liver problems. USA Today 2001 Dec 30.

2. Burros M. Raising new questions about kava’s safety. The New York Times 2002 Jan 16.

3. Packer-Tursman J. Anxiety over kava: FDA, others investigate reports of liver toxicity. Washington Post 2002 Jan 22.

4. Neergard L. FDA probes anti-anxiety herb. Associated Press 2002 Feb 11.

5. Neergard L. FDA: Herb kava may hurt liver. Associated Press 2002 Mar 26.

6. Strahl S, Ehret V, Dahm, HH, Maier, KP. Necrotizing hepatitis after taking herbal remedies. Dtsch Med Wochenschr 1998 Nov 20;123(47):1410-4.

7. Singh YN, Blumenthal M. Kava: An overview. HerbalGram 1997;39:33-57.

8. Blumenthal M, senior editor. Kava. In: The ABC Clinical Guide to Herbs. Austin, TX: American Botanical Council, 2002. (in press).

9. Pittler M, Ernst E. Efficacy of kava extract for treating anxiety: systematic review and meta-analysis. J Clin Psychopharmacol 2000;20(1):84-9.

10. Pittler MH, Ernst E. Kava extract for treating anxiety (Cochrane Review). In: The Cochrane Library, Issue 1, 2002.

11. Blumenthal M, Busse WR, Goldberg A, et al, editors. Klein S, Rister RS, translators. The Complete German Commission E Monographs: Therapeutic Guide for Herbal Medicines. Austin: American Botanical Council, Boston: Integrative Medicine Communications; 1998. p. 156-7.

12. Hagemann U. Pharmaceutical products containing Kava kava (Piper methysticum) and kavain, including homeopathic preparations with a final concentration up to D6 [letter]. Berlin: German Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, or BfArM). 2001 Nov 8.

13. Stoller R. Liver damage and kava extracts. Schweizerische Ärztezeitung 2000;81(24):1335-6.

14. German Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, or BfArM). Introduction of a drug safety plan for pharmaceutical products containing Kava-Kava. 2000 July 24.

15. Cumbo J. Kava drug withdrawn. Financial Times 2001 Nov 29.

16. German Nonprescription Drug Manufacturers Association (Bundesverband der Arzneimittel-Hersteller, or BAH) and German Pharmaceutical Industry (Bundesverband der Pharmazeutischen Industrie, or BPI). Comments of BAH/BPI on the German Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte, or BfArM) letter of 2001 Nov 8 on Kava- and Kavain-containing Medicinal Products. 2001 Dec 18.

17. Schmidt M, Narstadt A. Is kava hepatotoxic? Deutsche Apotheker Zeitung 2002;142(9):58-63.

18. Schulz J, Siegers C-P. Toxicity of kava pyrones: A reappraisal. Br J Clin Pharmacol 2002 (submitted).

19. Steinhoff B. Personal communication, 2001 Dec 21.

20. Schmidt M. Analysis of Kava Side Effects Reports Concerning the Liver. 2001 Nov (Translated from German by Lindenmaier M. and Brinckmann J., 2001 Dec 31). Courtesy American Herbal Products Association.

21. Busse WR. Personal communication. 2002 Jan 23.

22. Taylor CL. Notice to Physicians re Hepatic Reactions to Kava. Washington, DC: U.S. Food and Drug Administration. 2001 Dec 18.

23. U.S. Food and Drug Administration. Kava AERs (CD-ROM) 2002 Mar 22.

24. Woodfield R. Safety of Kava-kava products: Temporary and voluntary suspension of sale and supply [letter]. London: Medicines Control Agency, 2001 Dec 19.

25. Anon. Pacific island beverage banned in France after hepatitis scare. Agence France-Presse. 2002 Jan 9.

26. Therapeutic Goods Administration (Australia). TGA Consumer Advice on Kava. 2002 Feb 26. http://www.health.gov.au:80/tga/docs/html/kavaca.htm.

27. Therapeutic Goods Administration (Australia). TGA Kava Fact Sheet. 2002 Feb 26. http://www.health.gov.au:80/tga/docs/html/kavafs.htm.

28. Therapeutic Goods Administration (Australia). TGA Practitioner Alert on Kava. 2002 Feb 26. http://www.health.gov.au:80/tga/docs/html/kavapa.htm.

29. Loew D. Kava kava extract: Risks, benefits or a problem of society? Deutsche Apotheker Zeitung 2002;142(9):64-74.

30. Waller DP. Report on Kava and Liver Damage. Silver Spring, MD: American Herbal Products Association. 2002.

31. Anon. "fX": Chemically adulterated product does not contain kava. HerbalGram 1997;39:9.

32. McGuffin M. Personal communication, 2002 Jan 21.

33. Cianci S. Personal communication, 2002 Jan 17.

34. McGuffin M. Personal communication, 2002 Jan 11.

35. Connor KM, Davidson JRT, Churchill LE. Adverse-effect profile of kava. CNS Spectrums 2001;6(10):848-53.

36. Anon, 2001. Govt rebukes kava scare. Vanuatu Trading Post, Dec. 8, 2001.

37. American Botanical Council. American Botanical Council Announces New Safety Information on Kava [press release]. Austin, TX, 2001 Dec 20.

38. AHPA et al. Dietary Supplement Associations Undertake Scientific Evaluation of Kava. Washington, DC: American Herbal Products Association, Council for Responsible Nutrition, National Nutritional Foods Association, Utah Natural Products Alliance. 2001 Dec 20.

39. U.S. Food and Drug Administration. Kava-containing dietary supplements may be associated with severe liver injury. Consumer Advisory. 2002 Mar 25. Available online at http://www.fda.gov.

40. AHPA. AHPA Update: AHPA Executive Committee Adopts Revised Kava Labeling. Silver Spring, MD: American Herbal Products Association, 2002 Mar 26.