Kava: Information for sponsors - safety of kava-containing medicines

Related information

28 August 2002
  • The Therapeutic Goods Administration (TGA) has viewed with increasing concern international reports of liver damage (hepatotoxicity) associated with the use of kava-containing medicines. In January this year the TGA wrote to all Australian sponsors of kava-containing medicines on the Australian Register of Therapeutic Goods (ARTG) requesting them to remain vigilant in regard to possible adverse reactions to kava-containing medicines and reminding them of their responsibility to forward all adverse reaction reports to the TGA (as required in the condition of listing imposed under Section 28 of the Therapeutic Goods Act 1989 (the Act)).
  • The TGA also issued safety alerts to Australian consumers and health professionals about concerns following these international case reports, the most serious of which involved death or liver transplantation.
  • The TGA has continued to monitor closely both the Australian and international situation regarding suspected adverse events associated with kava-containing medicines. Until recently, there has been no suggestion that kava-containing medicines in Australia have presented a risk. However, the TGA has now become aware of the death from complications of liver failure of a woman in Australia who had been taking, among other complementary medicines, a kava-containing medicine.
  • The TGA believes that, as a precaution, a voluntary recall to consumer level of all medicines containing kava on sale in Australia should commence immediately. The TGA has consulted with the complementary medicines industry over its concerns and the need for a recall.
  • The recall has the support of the Complementary Healthcare Council (CHC, see Media Release on website http://www.cmaustralia.org.au/). The CHC has agreed to co-ordinate the recall on behalf of the complementary medicines industry.
  • The Recalls Unit of the TGA has written to all sponsors of kava-containing medicines to assist them in conducting the recall in accordance with the procedures laid down in the Uniform Recall Procedure for Therapeutic Goods (the URPTG).
  • In addition to this letter concerning the voluntary recall, TGA is also writing to all sponsors of kava-containing medicines in Australia on a related matter. This second letter is in the form of a 'proposal to cancel' notice under Section 30(2) of the Act and it advises sponsors that the TGA is providing a 28 day period in which sponsors of kava-containing medicines are to provide to the TGA evidence that their product(s) is safe. At the end of this period, the TGA will consider the information it has received, make a decision as to the acceptability for kava-containing medicines to remain on the Australian market, and will inform sponsors accordingly. Sponsors are to note that the provision of this data in response to a proposal to cancel by sponsors is a separate and additional process to their participation in the voluntary recall.
  • In the event that the TGA decides that the availability of kava-containing medicines in Australia presents an unacceptable safety risk, all kava-containing medicines on the Australian Register of Therapeutic Goods will be cancelled. Sponsors who have not, up until that time, carried out a recall of all their kava-containing medicines will be contacted by the TGA and required to institute an urgent consumer level recall in accordance with the Uniform Recall Procedure for Therapeutic Goods (URPTG). Sponsors who have already carried out a recall of their products will be required to demonstrate that they have done so in compliance with the URPTG.
  • Sponsors should be aware that consumers are also being advised to discontinue use of kava-containing medicines, to return any unused medicines containing kava to the point of sale or take them to their local pharmacy for safe disposal. Reimbursement by sponsors for returned unused products is a commercial decision on the part of the sponsor and not something in which the TGA has any involvement.

Frequently asked questions

What does a voluntary recall mean?

A voluntary recall in this case is a recall of a medicine carried out voluntarily by the sponsor. This is in contrast to a sponsor being compelled to conduct a recall in response to the TGA acting under its legislative power to institute a recall. In the case of kava-containing medicines, this recall is being conducted as a co-operative effort between the complementary medicines industry and the TGA in recognition of the potential risk to consumers posed by kava remaining on the market. The recall aims to ensure that products are removed from sale and consumers advised to discontinue taking kava-containing medicines and return them to point of sale or to a pharmacy for safe disposal.

What if I don't respond to the TGA in the 28 day 'proposal to cancel' period?

Then you are subject to any decision the TGA may make with respect to information it receives from other sponsors regarding the acceptability of kava-containing medicines remaining on the Australian market, including having to conduct a full consumer level recall of your product (s).

What type of data do I send to the TGA in response to the 'proposal to cancel' letter?

The TGA is looking for sponsors to provide any data supporting the ongoing safety of kava-containing medicines for supply in Australia. This should include information which adequately addresses the safety concerns raised by the existing reports associating kava-containing medicines with liver damage. These case reports together with the regulatory action taken or proposed in other countries played a significant role in the TGA's proposal to cancel. The TGA will also take into consideration any other information it receives which is relevant to the safety of kava-containing medicines.

The TGA offers the following background information to assist sponsors to address TGA concerns over the safety of kava-containing medicines:

There are internationally 68 case reports of liver damage associated with the use of kava-containing medicines. Recently, the Adverse Drug Reactions Advisory Committee (ADRAC) received a report of death in Australia of a woman from complications of fulminant hepatic failure associated with the use of a kava-containing medicine (see Australian Case Report below for details). It should be noted that reporting of an adverse reaction does not necessarily imply a causal relationship. Case reports of liver damage associated with the use of kava-containing medicines have been given causality classifications ranging from probable, possible to unlikely. The case reported in Australia, involving a fatality (see below) was assigned a causality rating of possible. The severity of liver damage reported varies and includes abnormal liver function, jaundice, hepatitis, liver failure and death.

The mechanism of kava-related liver toxicity is not well understood and there are no clearly identifiable predictors of the likelihood of a person getting liver damage from a kava-containing medicine. The cases reported cannot be used to estimate the incidence of the potential for liver damage within the population, as it is likely that only a small proportion of adverse reactions are reported. It is not possible to determine which consumers are at greatest risk. On the evidence available at present, it is not possible to determine the relationship between the likelihood of liver damage and the strength, total dose, method of administration, degree of processing or the extraction method of the kava-containing medicine.

Sponsors are reminded that rare but serious adverse reactions associated with liver toxicity have previously been reported for herbal medicines. Some medicines have been used traditionally and generally considered to be devoid of serious side effects when taken over a long period of time. For example, the hepatotoxic potential of chaparral (Larrea tridentata) only became apparent after an estimated 500 million capsules had been used without concern over a period of 20 years (De Smet, P.A.G.M. (1998) Towards Safer Herbal Medicines, pp1 - 9. Edited Proceedings of the 5th International ESCOP Symposium, The European Phytojournal http://escop.com/.)

What if my product is a non-extracted kava product OR if my product contains a kava extract in low dose and/or mixed with other herbs?

Because of the uncertainty as to the form or dose of kava which might be associated with the liver damage, all kava-containing medicines on the Australian Register of Therapeutic Goods are subject to the proposal to cancel, regardless of the form or the amount of kava they contain, or the presence of other herbal or non-herbal materials.

Where can I get information about the case report of the fatality in Australia?

This information is provided in an extracted form to safeguard privacy under 'Australian Case Report' below.

When will I know a decision about the future ability to supply kava-containing medicines into the Australian market?

As soon as the TGA has evaluated the data on the safety of kava-containing medicines which it receives from sponsors or industry bodies on behalf of its members, the TGA will make a decision as to the suitability of the ongoing supply of kava-containing medicines into the Australian market. In the process it may decide to consult with expert groups such as the CMEC and the ADRAC.

What have other countries done?

Concerns over liver-related health risks associated with the use of kava-containing medicines have prompted regulatory agencies in other countries, including those in Germany, Singapore, Switzerland, France, Ireland, Portugal, Austria, United Kingdom, USA and Canada to take action ranging from warning consumers about potential risks, to removing kava-containing medicines from the marketplace.

Australian Case Report

Complementary medicine and fatal hepatic reaction

The Adverse Drug Reactions Advisory Committee (ADRAC) gives undertakings of confidentiality to reporters of suspected adverse drug reactions in Australia. In respect of this case, the reporting doctors have indicated that they wish to publish a case report. For this reason, not all details available to ADRAC are included in the following narrative. The Adverse Drug Reactions Unit, TGA, will deal with any requests for further information individually.

ADRAC has received a report of a 56 year old woman who died during a liver transplant operation for fulminant hepatic failure. Other than a twelve month history of a stable benign monoclonal gammopathy, she had been previously well. She had been admitted to the hospital 17 days earlier for the investigation of a two week history of fatigue and jaundice. Over the 3-4 months before admission, she had been taking five complementary medicines - Vitamin E 500IU (AUSTL 73328); HGF Tablet (AUSTR 19994); Mineral Matrix with Selenium (AUSTL 70158); Fibroplex Plus (AUSTL 80895); and Kava 1800 Plus (AUSTL 64198). She had been taking all of these for about four months. Before this, she had been taking several other complementary medicines (details of which are known) for about three months. She had not been taking any prescribed medicines, and drank only a minimal amount of alcohol. On admission, she had markedly abnormal liver function tests, with highly elevated transaminases, and elevations of bilirubin, alkaline phosphatase, and gamma GT. Liver biopsy showed a severe acute hepatitis with confluent necrosis, consistent with a viral or drug aetiology. Investigations performed at the hospital excluded known viral causes of hepatitis.

Consistent with ADRAC's published criteria, all five medicines have been coded as possible causes. At review, the Committee noted that of the ingredients of the five medicines, kava extracts have been most commonly associated with reports of severe hepatitis. The product being taken (Kava 1800 Plus) is stated also to contain Passiflora incarnata (passion flower), and Scutellaria lateriflora (skullcap). Testing by TGA has confirmed the presence of kava and Passiflora, but Scutellaria lateriflora was not detected. To date, the identity of the third ingredient remains to be established. The presence of some other herbs reported to be hepatotoxic has been excluded.