St. John's wort - quack medicine or novel antidepressant treatment?

Reports that extracts of the plant St. John's wort (Hypericum perforatum) possess an antidepressant property have generated considerable professional and lay interest. While some trials have reported an effect greater than placebo and comparable to antidepressants, the studies have significant methodological weaknesses. Investigations using animal models also suggest the plant has an antidepressant effect. A possible active principle, hyperforin – which inhibits uptake of monoamines such as serotonin - has been identified. These reports, and evidence of acceptable safety and tolerability, have resulted in a large multicentre study which should clarify if hypericum extract is a true novel antidepressant.

In the history of medicine, many active therapeutic compounds have been derived from plant products.¹ In Peru during the 17th century, the Augustinian monk, Father Antonio de la Calancha observed that the bark of the Cinchona tree could alleviate the fever of malaria. In Europe, where it was used medicinally, it was known as 'Jesuit's bark'. However, it was not until the 19th century that the active principle, quinine, was isolated.

Another example is acetylsalicylic acid. In the 18th century, Edward Stone discovered that the bark of the willow tree reduced fever. However, it was only in the early 19th century that German chemists isolated the active compound, salicin. This was improved later that century to salicylic acid, then finally acetylsalicylic acid, more commonly known as aspirin.

Do reports that the extract of St. John's wort possesses antidepressant properties similarly herald the discovery of a new active therapeutic compound,² or will it end up on the scrap heap of pharmacological history?

The herbal medication is an extract of the plant St. John's wort or Hypericum perforatum. The name of the plant derives from its flowering around the time of St. John's Day, June 24. It has been used mainly in Germany, where it was formally approved in 1984 for treating 'psychogenic disturbances, depressive states, anxiety or nervous excitement'. There are many different commercial preparations available. These are standardised by their content of hypericin, which was initially thought to be the active constituent. In recent years, German prescriptions of St. John's wort have outnumbered those for fluoxetine by at least four times. (One of the possible explanations for this is that German medical practitioners have a limited prescribing budget, so cheaper compounds are very popular.)

Interest in St. John's wort in the English-speaking medical world arose in 1996 with a report in the British Medical Journal.^3,4 Before then, most reports had been published in German, with many appearing in phytomedicine journals. The meta-analysis³ identified 23 randomised clinical trials that had compared extracts of hypericum with placebo or other antidepressant treatments. Fifteen of those trials (involving 1008 patients) were placebo-controlled, with the other eight (involving 749 patients) comparing hypericum with other antidepressant or sedative drugs. Of the 23 trials, 20 were double-blind, one was single-blind and two were open in design.

About 22% of patients responded to placebo, whereas 51% responded to the hypericum. In the active comparator studies, 64% responded to the hypericum, whereas about 59% responded to standard antidepressant treatments. ('Response' was most commonly assessed using the Hamilton depression rating scale, requiring a score of <10, or <50% of the baseline score.)

There were, however, significant methodological limitations with these studies. The classification of depression was not consistent, and most did not use structured interviews or operationalised diagnostic criteria. Recruitment was undertaken not only by psychiatrists, but also by a broad spectrum of medical practitioners including physicians and gynaecologists. (There was also a problem blinding patients taking hypericum, as the extracts tend to have a characteristic taste.)

Most trials lasted only four weeks, whereas 6-8 weeks is now the standard expectation for such studies. There were certainly no 6-12 month studies of either effectiveness or safety, and no studies of severe depression.

A further difficulty was the standardisation of dosage. Seven different preparations of hypericum were used, with the dose varying between 0.4 and 2.7 mg daily.

The meta-analysis concluded that, despite the methodological limitations, there was good evidence that hypericum was more effective than placebo in treating some depressive disorders. This is consistent with studies of animal models,⁵ which have shown that St. John's wort appears to possess antidepressant properties.

St. John's wort appears to be well tolerated. In the analysis,³ only 20% of patients reported adverse effects on St. John's wort, compared to 36% of those on standard preparations. Drop-out rates due to severe adverse effects were about 1% in the St. John's wort group compared to 3% of those on other treatments. There is a need to be aware of potential drug interactions. There has been already one report of an interaction with a selective serotonin reuptake inhibitor (SSRI) leading to serotonin syndrome.⁶

What is the active principle of the hypericum extract? Early reports suggested that it was one of the hypericins, but hypericum extract contains at least 10 different constituents or groups of components that may contribute to its action. Examples are the naphthodianthrons (such as the hypericins), flavonoids (such as quercetin), xanthones and the bioflavonoids. One particular derivative of hypericum extract, hyperforin, has recently been found⁵ to be not only the major lipophilic chemical constituent of this plant, but also a potent inhibitor of the uptake of several monoamines such as serotonin, dopamine and noradrenaline, and also of the amino acid glutamate. The potency of two different hypericum extracts in animal models has also been correlated with their content of hyperforin.⁵ The extract containing more hyperforin was more active. There is, therefore, considerable evidence that hyperforin may be the major active principle of St. John's wort.

(It was initially thought that hypericum acted as a monoamine oxidase inhibitor, but this effect now appears to be weak. There are also some effects of hypericum on the GABA-A receptor, although hyperforin does not account for this.)

The studies published to date are intriguing and suggest an antidepressant effect. While the efficacy data appear impressive, it should be emphasised that the methodological standards and sample sizes of the published trials would not satisfy the criteria for marketing approval required by government regulatory authorities such as the Therapeutic Goods Administration in Australia or the Food and Drug Administration in the U.S.A. The results, however, are of considerable interest and have led to the National Institutes of Health in the U.S.A. sponsoring a large multi centre study comparing hypericum with an SSRI and placebo in a large sample of patients with moderately severe depression. Recruitment for that study began in 1997 and the study should be completed in the near future.

The identification of a novel antidepressant would be an enormous boon for the treatment of this common and severe disorder. Most of the available antidepressants are derivative, being, in essence, modifications of the structure or function of the tricyclics or monoamine oxidase inhibitors. It would therefore be ironic if the mechanism of St. John's wort was found to be similar to that of 'designer antidepressants' such as the SSRIs.

The medical profession must remain circumspect in its judgement of St. John's wort until confirmatory data from large, methodologically strict trials become available.

Acknowledgements

The author is grateful to Mrs Kerrie Eyers for her helpful comments and to Mrs Zora Vuckovic for preparation of this paper.

The following statements are either true or false.

1. There have been no long-term studies of St. John's wort in severely

2. Extracts of St. John's wort may interact with selective serotonin

Answers to self-test questions

1. True

2. True