Flaws in the KADSS study
From Mr A. R. Cox, MRPharmS
The Pharmaceutical Journal recently reported (PJ, 16 May 2009, p577) that the herb kava was “safe and effective at reducing anxiety” on the basis of the small (n=60) three-week placebo-controlled, crossover trial “the kava anxiety depression spectrum study” (KADSS).1
The Journal reported “the study authors note that the aqueous extract was safe, with no liver toxicity or other serious adverse effects”.
Kava was prohibited in the UK in January 2003. The Committee on Safety of Medicines (CSM) judged that it “posed a rare but serious risk to public health” following reports of idiosyncratic liver damage. A July 2006 report of the CSM’s Expert Working Group on the safety of kava concluded that the prohibition order was justified and proportional.2
Liver damage ranged from mild jaundice or changes in liver function to more serious cases of hepatitis and liver failure. Eleven patients required liver transplants and nine individuals died as a result of the hepatotoxicity. No specific susceptibility factor was identified, such as gender or age group.
There is debate about the relative safety of differing cultivars of kava and the differing extraction methods (aqueous, acetonic and alcoholic).3 A World Health Organization assessment of the risk of hepatotoxicity associated with kava noted that acetonic and alcoholic extracts may be risk factors for hepatotoxicity and should be avoided.4 The KADSS study used an aqueous extract of the “noble” cultivar of kava.
However, the WHO assessment notes five cases of hepatotoxicity associated with aqueous extracts of kava,4 and there are published case reports of hepatotoxicity with traditional aqueous preparations.5
Additionally, given the onset time of kava-associated hepatotoxicity in reported cases, with most occurring within three months of treatment initiation (ranging from immediate to many months), a three-week study does not give opportunity for such reactions to develop.
More importantly, even if an event has not occurred, this does not mean it cannot happen. Using the rule of three,6 we can calculate that we can be 95 per cent confident the incidence of an adverse event that did not occur within the trial would lie between zero in 60 cases and three in 60 cases. This means that hepatotoxicity occurring with a frequency as high as one in 20 patients could remain undetected by the KADSS study.
Given that kava-associated hepatotoxicity is a rare reaction, the KADSS study does not provide any meaningful safety information about kava or the relative safety of differing extraction methods. Headlines suggesting kava is safe on the basis of this study are, therefore, misleading and dangerous.
Anthony R. Cox
Pharmacovigilance Pharmacist
West Midlands Centre for Adverse Drug Reactions
City Hospital
Birmingham
References
1. Sarris J, Kavanagh DJ, Bryne G, Bone K, Adams J, Deed G. The kava anxiety depression spectrum study (KADSS): a randomised, placebo-controlled crossover trial using an aqueous extract of piper methysticum. Psychopharmacology 2009;DOI 10.1007/s00213-009-1549-9.
2. Medicines and Healthcare products Regulatory Agency. Report of the CSM’s Expert Working Group on the safety of kava July 2006. Available at www.mhra.gov.uk (accessed on 26 May 2009).
3. Tesche R, Schwarzenboeck A, Akinci A. Kava hepatotoxicity: a European view. The New Zealand Medical Journal 2008;121:90–8.
4. Coulter D. Assessment of the risk of hepatotoxicity with kava products. World Health Organization Appointed Committee 2007.
5. Russman S, Barguil Y, Cabalion P, Kritsanida M, Duhet D, Lauterburg B. Hepatic injury due to traditional aqueous extracts of kava root in New Caledonia. European Journal of Gastroenterology and Hepatology 2003; 5:1033–6.
6. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything alright? JAMA 1983;259:1743–5.
