SINCE the American debut of Prozac, in 1988, if we listen to Peter Kramer, the author of the best-selling Listening to Prozac, we have experienced a change in our understanding of what happens when we treat psychiatric illness. One aspect of that change is the notion that psychical benefit need not have much of a downside. Even the reversible side effects (drowsiness, constipation) of older drugs could put off ambulatory depressives seeking relief. But Prozac is a kinder, gentler antidepressant than those that came before it. Furthermore, Prozac works. And its effects go beyond the relief of depression. The Kramerian psychiatrist tells his patients that the drug will relieve their symptoms but adds, "I must warn you, you may feel more contented." Prozac appears to have had an effect on the zeitgeist, too: "We will come to discover," Kramer says, "that modern psychopharmacology has become, like Freud in his day, a whole climate of opinion under which we conduct our different lives." The thinly veiled indictment in his conclusion is that psychopharmacology, the study of drugs that influence mind and mood, is less than science, more like predicting the weather. In post-Prozac 1998 one wonders how the psychopharmacological climate will change with the advent of the latest, rather unlikely antidepressant: Hypericum perforatum, an herb known as St. John's wort ("wort" is Old English for "plant").
According to Jonathan Zuess's (1997), there is nothing particularly new about H. perforatum. Solomon's Song of Songs refers to it as the Rose of Sharon, and the physician Paracelsus (1493-1541) recorded its use for nervousness, skin wounds, and abdominal pain after listening to "old wives, gipsies, sorcerers ... and such outlaws" while wandering Europe and the Middle East in search of unconventional treatments. H. perforatum has a yellow five-petaled flower with numerous stamens which typically blooms on or about June 24, Saint John's birthday; legend has it that the plan first grew from the blood that fell at Saint John's beheading. It grows abundantly near the Klamath River, in California, and is thus sometimes called Klamath weed. Like a weed, it is common in sunny areas and dry soils throughout the world. Goats eat it; cattle demur. Zuess reports that "medical researchers have been astonished to discover that even though St. John's wort works as well as antidepressant drugs like Prozac and Tofranil [an older antidepressant, also known as imipramine], it has only a fraction of the number of side effects that they do."
flower buds exude a burgundy-colored oil when soaked in alcohol, vaguely reminiscent, one supposes, of Saint John's blood. This tincture contains high concentrations of its putatively active chemical, hypericin. Dried extracts from harvested buds, blooms, leaves, and stems contain variable percentages of hypericin. The British Medical Journal (BMJ) recently published an overview of clinical trials involving 1,757 patients. In those trials daily doses of extract given for "mild or moderately severe" depression have run 350-900 milligrams; total hypericin dose has varied from 0.4 to 2.7 mg a day. There is no specific recommendation as to how best to take it; Zuess suggests that it be taken at least twice a day, and says that it may be preferable to take it with meals, to minimize stomach upset, though that is unlikely to occur. The duration of treatment in trials has generally been at least four weeks, so relief of symptoms is possible but not anticipated in less time.
H preparations are marketed in the United States as dietary supplements. They cannot legally be labeled as effective against any specific disease, despite the 1994 passage of the Dietary Supplement Health and Education Act (DSHEA), which sought to liberalize the dissemination of information about natural products. Federal regulations have been slow to respond to the therapeutic claims made for St. John's wort. To date, clinical and research experience come largely from Germany, where the herb is approved for the treatment of depression and anxiety. In 1994, according to a BMJ editorial, German physicians prescribed 66 million daily doses of Hypericum preparations for psychological complaints. American physicians have traditionally been less accepting of herbal treatments than German physicians, but the American public appears to have its own opinion: the medicinal use of botanicals is increasing in the United States. The U.S. market in 1995 was approximately one third the size of Europe's, but some estimate that it is growing by 15 percent a year. DSHEA included an estimate that some 600 dietary-supplement manufacturers in the United States produce some 4,000 products with annual sales of $4 billion.
Animals have experienced skin reactions to light (even fatal sunburns) after ingesting large quantities of St. John's wort. Side effects in human beings are, reportedly, infrequent (stomach upset in 0.6 percent, allergic reactions in 0.5 percent, fatigue in 0.4 percent), and serious adverse events (such as reactions to sunlight or marked changes in blood pressure) extremely rare, but these data were collected in "open" rather than "blinded" (neither doctor nor patient knowing what treatment is being used) and controlled fashion. Peter de Smet, a Dutch clinical pharmacologist and "herbal pharmaco-epidemiologist," and a co-author of the BMJ editorial, has warned, however, that one serious adverse reaction in 10,000 users may be considered unacceptable in "official medicine" if there is no clear therapeutic need for the drug in question.
St. John's wort holds promise of therapeutic merit, but in April of last year the influential newsletter Biological Therapies in Psychiatry recommended against its use in the absence of large clinical trials. Among the twenty-three studies in the BMJ overview, patients with mild or moderately severe depressive disorders fared significantly better with St. John's wort than with placebos in fifteen studies, with a response rate of 55 percent (as against 22 percent). In eight studies comparing St. John's wort with standard antidepressants, including imipramine at conservative doses, efficacy was slightly higher (64 percent, as against 58 percent). The dropout rate -- a reflection of drug toxicity that may limit treatment -- was lower among those using St. John's wort than among those using standard antidepressants, and lower even than among those who received placebos, suggesting that the paucity of side effects may be H. perforatum's greatest attraction. (Its overall frequency of any side effects was 20 percent, as compared with 53 percent for standard antidepressants in blinded studies.)
The newsletter Biological Therapies in Psychiatry seems harsh in its judgment on the herb, given the preliminary evidence suggesting its efficacy. "There is a lack of sufficient data on safety (particularly long-term), efficacy, potential interactions, and product purity of products sold as nutritional supplements rather than as drugs, such as St. John's wort," the newsletter said. The problem of impurity is especially germane to herbal combinations. A 1981 British report attributed liver damage to a mistletoe preparation that included skullcap (Scutellaria lateriflora). By the late 1980s, after additional experience with Scutellaria species had surfaced, the skullcap (famous as an eighteenth-century treatment for canine rabies), rather than the mistletoe, was deemed to have been the likely culprit. Even herbal "monopreparations" are not exactly pure: in H. perforatum extracts some ten chemicals may play a role in the therapeutic effect, though hypericin is thought to be principally responsible. But how St. John's wort works chemically to mitigate depression is unclear, a point that Michael Jenike, of the Harvard Medical School, makes in a generally favorable preface to a section dedicated to the herb in the Journal of Geriatric Psychiatry and Neurology.
No well-informed clinician would disparage medicinal plants in general (a short list of well-established pharmaceuticals that derive from plants would include digitalis, morphine, reserpine, pilocarpine, and curare), but there is skepticism in the United States about the comparatively liberal use of them in Europe. Though the observation smacks of scientific xenophobia, the German experience with St. John's wort is not the English-speaking experience, and until recently the English-speaking world had no experience. The authors of the BMJ review observed that had they restricted their search for published studies to those in English, not a single clinical trial of St. John's wort would have been found as of 1994. The journals in which one might find clinical research on herbal remedies (for example, Phytomedicine, which in 1995 published what was probably the first English-language review of St. John's wort as an antidepressant) are rare in medical-school libraries.
Jonathan Zuess describes St. John's wort as the "most-used antidepressant in the history of humanity" and "truly a healer of both the body and the mind." But one ought to proceed with caution. The characterization of it as "the natural Prozac" is not, alas, supported by any published clinical research comparing the two substances, although the National Institute of Mental Health (including the Offices of Alternative Medicine and Dietary Supplements) is currently embarking on a comparison of St. John's wort and the Prozac-like drug Zoloft.
LIKE Prozac, St. John's wort promises much, and perhaps also like Prozac, it has effects that go beyond countering depression. Those familiar with the range of Prozac's influence (on, as Kramer has posited, "affect tolerance, autonomy and coercion, cultural expectations, evolutionary fitness, transcendence") may be unsurprised to find a similar breadth of powers attributed to St. John's wort. The hope, of course, is that someday we will find an antidepressant of great power with little or no downside -- but then the question will be, What are we treating? "Mild to moderately severe depression" -- the proper province of Prozac and, the available data suggest, of St. John's wort -- has come to mean many things. Assuming that I make a decision to take an antidepressant, am I clear about what it is specifically that I am trying to assuage in my life?
Since the vast majority of studies quoted in the recent literature on H. perforatum measure its effects according to changes in symptoms that appear on a standard clinical rating scale for depression, I applied the scale to myself, a busy clinician and researcher. I answered affirmatively to thirteen of its seventeen items, each of which is to be graded in terms of severity. Of those thirteen, two relate to problems with sleep, two to guilt and mood, three to anxiety, four to weight loss and bodily symptoms (especially muscle aches), and one to feelings of fatigue or incapacity in work and hobbies. If I explain that many of my symptoms stem from being an anxious, guilty insomniac with neck pain, I have to rate myself higher on the last remaining item, which questions my insight into my problem. Attributing depressive symptoms to the climate, a virus, or overwork is not the same as, and perhaps more pathological than, admitting that I am depressed, according to this scale. My total score is 16, out of a maximum of 52; the mean rating for patients enrolled in sixteen recent H. perforatum studies was comparable, at 20. But the point is that "depression" as measured by this yardstick is no motivation to seek out my psychopharmacological options. I would feel much the same need, perhaps even more acutely, if I scored a mere 3 owing to one symptom alone: "a decrease in productivity."
I do not know quite how I will fare with St. John's wort. With little risk, what harm in experimentation? I can explore what effect the drug has, which is the next best thing to understanding what it is that makes me take it.
Edison Miyawaki teaches neurology and psychiatry at the University of Kansas Medical Center and at the Brigham and Women's Hospital, in Boston.
Illustration by Cynthia von Buhler
The Atlantic Monthly; May 1998; Listening to St. John's Wort; Volume 281, No. 5; pages 40 - 43.
We want to hear what you think. Submit a letter to the editor or write to firstname.lastname@example.org.