Aug 10 2009

Botanicals and Menopause – No Effect

Black Cohosh is a common herbal or botanical treatment for the hot flashes and other symptoms of menopause. This treatment appeals to women who want a more natural or gentle approach to their symptoms. Unfortunately, as the latest research shows, it doesn’t work.

Interest in herbal medicine has increased in the last 15 years as a response to weakened regulation and a successful marketing strategy by the supplement industry in the context of a cultural phenomenon of changing attitudes toward medicine. This has mainly led to, however, confusion about what herbal therapy really is. Most people I talk to think of herbal treatments as natural supplements – but they are not supplements and being “natural” is irrelevant. In fact it is much more accurate to think of them as unpurified drugs.

Botanical research is a legitimate and useful branch of pharmacology, with its own experts. The plant kingdom is a vast chemistry laboratory, specializing in chemicals that have some effect on other living things. The process of experimentation through evolution is slow, but has had hundreds of millions of years to operate. As a result we are surrounded by plant-based chemicals – some benign, most toxic, and a few with pharmacological properties we can exploit to our advantage if we’re careful.

That these chemicals are “natural” is meaningless. Beyond the fact that the term in this context has no useful operational definition, that these chemicals exist in nature should give us no comfort. They exist for the survival of the plant, mostly at the expense of animals who might want to eat it. No plant evolved to be a useful medicinal for one egocentric mammalian species. Toxicity, however, is all about dose, and in essence all useful drugs are simply toxins with a dose range in which they have an effect that can be exploited with acceptable side effects.

Also, plants have dozens or hundreds of chemicals in them, and these vary in different parts of the plant, in different times of the season, in different regions, and from year to year. They are very dirty drugs, with variable doses of potentially useful but also strictly harmful substances. Botanical pharmacology focuses on identifying useful plants, and then figuring out which chemical substances in them are useful and which are simply adding toxicity. Useful chemicals can then be purified, quantified, and studied for their biochemistry and net health effects.

I still have a hard time understanding why someone would want to treat themselves with a dirty herb containing highly variable doses of multiple chemicals, rather than a precisely known quantity of a specific chemical with known properties. The justification I am almost always given is because they are “natural” – but as I pointed out, this is no justification at all, but clever marketing bordering on mythology.

The one saving grace of herbs is that the doses of most chemicals within them is likely to be low and therefore below the level of toxicity – or any useful effect, either. This does not mean their safety can be assumed, and many herbal remedies have been found to have toxicity, (like kava kava and liver toxicity or kidney failure from a variety of botanicals) or to interfere with prescription drugs (so-called drug-drug interactions – because they’re drugs).

Once modern pharmacology came into its own our accumulated wealth of botanical knowledge was raided for useful drugs, and they now comprise the majority of the drugs we use today. We have already plucked the low hanging fruit. There are likely to still be useful drugs out there in the plant kingdom, but it will take some investigation and research to find them. Such efforts are ongoing, and will likely continue to feed pharmacy shelves for years to come.

However, the herbal remedies that have come into popular use are generally not the result of investigation and careful research, but either tradition (read anecdotes) or modern marketing masquerading as tradition. For example, echinacea may have been used by Native Americans for many things (as was tobbaco, incidentally) but they did not use it for the common cold. That use was an invention of a German herbalist who used the Native American angle for marketing purposes.

It is therefore not surprise that the herbs most commonly marketed today for common ailments (read large customer base) have mostly failed in clinical trials. Gingko does not work for memory or mental focus, echinacea does not treat the common cold or its symptoms,  and St. John’s Wort has minimal to no effect on depression. Black cohosh for menopausal symptoms was held up as an herbal remedy that actually works, but now we have evidence from a double-blind trial that is clearly negative.

This trial had four arms – including black cohosh, red clover, placebo, and standard hormonal therapy. They followed women with menopausal symptoms for 12 months and found that, over time, they all had a decrease in symptoms by 34, 57, 63  and 94 percent respectively. The black cohosh and red clover were not statistically significantly different from placebo, while the hormonal therapy was significantly improved from placebo. The primary weakness of the study, which was generally well designed, was that it enrolled only 89 women – this is enough to run meaningful statistics, but is smallish for a clinical study of this type.

The study also examined memory in the subjects and found no difference again among black cohosh, red clover, and placebo and a slight negative effect from hormonal therapy. So – standard hormonal therapy works, with a slight side effect on memory. The botanicals black cohosh and red clover had no detectable effects or side effects – not surprising. No one study is ever the final word on questions of clinical safety and efficacy – but the trend for research into popularly marketed herbs is definitely negative.

This, of course, raises many questions about regulation and research. The current regulation in the US is abysmal – botanical drugs are treated as if they were vitamins, and companies can even make pseudo-health claims for them (so-called structure-function claims), while having no obligation to prove safety. Research is now often funded by the NCCAM, which recently came under criticism for spending 2.5 billion dollars of tax payer money over the last decade and having nothing to show for it. This might have something to do with using politics to create a center specifically to bypass the usual methods of deciding what is useful to research.

As I said – botanical research is already a separate and useful branch of pharmacology that can be funded through other centers at the NIH. In fact this legitimate research is marginalized when brought under the umbrella of “alternative medicine” – another marketing term with no useful definition.

The good news is that quality clinical research into the effectiveness and safety of commonly used herbal remedies is being done. This research recently has largely been negative, which is not surprising given the history and nature of many of the products being studied. The bad news is that common use, regulation, and marketing are almost completely divorced from the scientific research. What, then, one might wonder, is the purpose of the research?

16 responses so far

16 thoughts on “Botanicals and Menopause – No Effect”

  1. juga says:

    I would appreciate your comments, Steven, on the placebo effect. So many articles like this say “such and such is no more effective than a placebo” but what exactly does that mean? In this case, the figures show that a placebo reduced symptoms by 63%. That sounds like a reasonably effective remedy but no one seems interested in what is happening here.

    How precisely have the women in this trial had their condition improved by taking a placebo? As by definition a placebo is inactive, the improvement has presumably come from the act of taking it. i.e. Because the subjects think they are taking a remedy, their symptoms are reduced.

    Is this not a hugely significant finding? i.e. That menopause symptoms can be reduced by 63% by a woman’s thought or mental state. If this level of improvement can be achieved by unknowingly taking a sugar pill, might it not be possible to learn how to achieve an even better effect by taking more precise actions on the mental state or by adopting other thought patterns? If not, why not? If some women can have their menopause symptoms completely removed without needing any chemical, why can’t all women?

    Why is no one interested in this? Is it because the drug companies have so much to lose if an alternative to drugs is found?

    As a further speculation, it seems plausible that one way of boosting the placebo effect could be homeopathic treatment. There is obviously nothing active in homeopathic remedies so what is left but the mental effect on the patient of the incantations?

  2. juga – Here is an overview of placebo effects I wrote:

    And here is a report of a study of placebo effects:

    Placebo effects are studied. They are not as useful as you make out, however. You are assuming that all of the measured placebo effect is due to mind over matter. However, there are many other effects as well. Some are non-specific effects of the therapeutic intervention. But others are simple reporting or observation bias. You also have to consider the natural course of symptoms – maybe hot flashes decrease over time in menopausal women.

    There is also a huge ethical and practical issue with lying to patients. Using homeopathy, for example, to get a placebo effect for a subjective symptom in a self-limiting condition creates unscientific notions in the patient and the public. Then, when they get cancer, they may rely on homeopathic treatments which do nothing. This happens – it is not a theoretical concern.

    Finally – you don’t need magical nonsense to get a placebo effect. You will get it from a real treatment. What this does mean is that in clinical practice, for symptomatic treatment (not preventive or disease modifying) it makes sense to begin with the most benign treatment, and only go to more invasive or toxic treatments as necessary. This effectively will see how much of a placebo effect you will get, but using a real if gentle treatment.

  3. Draal says:

    I’m unfamiliar with Black Cohosh but the choice of red clover is pretty clear. Red clove can have up to 30mg/DW of methylated versions of genistein, primarily Biochanin A and formononetin. Genistein, commonly associated with soy beans, has an IC50 value of about 10 nM for Estrogen receptor beta and about 40 nM for Estrogen receptor alpha. Isoflavones have a very similar chemical structure to estradiol (IC50 value of <0.5 nM for ER-alpha and beta). Biochanin A and formononetin have at least an order of magnitude less binding affinity for the ERs than genistein. There are plenty of study on soy (genistein) and its affect on breast cancer, menopause, ect. All very inconsistent though.

    So, they could have chosen red clover since it's almost guaranteed that there would be no affect and they avoided soy so they wouldn't have to deal with the possibility that there would be a correlation.

    My perception on herb treatment is that people associate it with "safe".

  4. Draal – First, just to be clear, I agree that herbs and plants can have useful active ingredients. I am not against botanicals in any way – I just want them to be studied and regulated properly.

    But, I am confused by your penultimate point – are you saying the authors were trying to avoid a useful herb? They state themselves that they were looking for a useful alternative. Why would they avoid one that could work?

    Regarding soy, the inconsistent effects in trials is likely due to the variability of dose and bioavailability from unpurified sources. Also – if it can work through binding to the estrogen receptor, then wouldn’t we expect similar effects and side effects as from estrogen and other analogues? Maybe there is some advantage – so test the purified compound as a drug.

  5. Draal says:

    But, I am confused by your penultimate point – are you saying the authors were trying to avoid a useful herb? They state themselves that they were looking for a useful alternative. Why would they avoid one that could work?

    It’s just my opinion but red clover was a poor choice compared to soy. The isoflavone composition of red clover just contains very weak phytoestrogens (based of in vitro studies). Soy contains the most potent phytoestrogen, genistein, at about 5-10 mg per DCW (primarily as the glucoside, genistin, but the sugar group is easy to strip off). So, yes, I think the researches were avoiding soy and went with red clover b/c the both contain “isoflavones”.
    Hype in soy reasearch hit a fever pitch in the late 1990s and early 2000s. The inconsistent results coming in made the NIH to hugely scale back funding into isoflavones in 2005. Still, there is a lot, I mean a lot, of info on genistein (soy) and its use as a phytoestrogen. Frankly, they were avoiding navigating a mine field.

  6. Draal says:

    Regarding soy, the inconsistent effects in trials is likely due to the variability of dose and bioavailability from unpurified sources. Also – if it can work through binding to the estrogen receptor, then wouldn’t we expect similar effects and side effects as from estrogen and other analogues? Maybe there is some advantage – so test the purified compound as a drug.

    off the top of my head, I think there are 7 or 8 ERs. Estradiol has a preference towards ER-alpha and genistein has a preference towards ER-beta. I’d have to dig this up so my memory could be wroung, but ER-beta is not as widely expressed and different tissues vary in composition.
    There are bio-availability issues that need to be addressed with phytochemicals. They run the gauntlet which is your digestive tract and liver, eventually being metabolized. I don’t know the specifics or what the half life is, ect. but only <5% of consumed flavonoids can be detected in the blood.
    Any there is the reductionist idea that as a whole, there is synergy but as you purify plant extract fractions, the effects disappear.

  7. gfb1 says:

    as i recall, one of the early studies that SEEMED to validate the ‘black cohosh as menopausal treatment’ concept had positive results. later, and less publicized work, showed that the herbs used had been sprayed with DDT, were contaminated and had a wonderful, if predictable, estrogenic effect.

    when you talk about herbals being dirty; depending on where they are grown (and who grows them), they can be REALLY dirty…..

  8. Draal says:

    When submitting research, I generally thought that the paper must contain novel ‘ideas’, else you’re just redoing someone’s experiment. I can’t just rerun someone’s experiment and publish (assuming the same results). Soy/genistein has been used in so many studies that the authors may have run into a “novelty” issue. Red clover has not been studied as extensively but the CAM peeps do run with it.

  9. Draal says:

    Me and my ignorance, but…
    Why, in this day and age, is there no (seemingly) minimum number of people that must be enrolled in a study to avoid the “but is smallish for a clinical study of this type” observations? I mean, it’s a very well known issue! Time after time, there’s a new “clinical study” coming out that proves this or doesn’t prove that based on a handful of people and inevitably the last paragraph of these papers contains some line like “further investigation is required”. Well $hit, really? How ’bout you get to work??
    For one thing, the credibility of the study is immediately in question, and another thing, it allows any yahoo and his/her grad student to publish just about anything.
    Ok, some situations like rare diseases do not allow large studies to be conducted. But the treatment of hot flashes in women and all they could lasso was 89 people?
    Without hard and fast standards, the literature is just polluted with endless crap. Today, there are a few distillation methods to shift through the junk. 1) Review papers 2) specialized journals that cherry pick the most noteworthy articles. The downside is that journal editors and reviewers are subject to biases and politics.
    If minimum standards were in place, I’d hope they would solve a few issues 1) CAM articles would largely disappear or be further marginalized 2) The quality of the work would be improved 3) I won’t have to read as much :P. The public doesn’t know if the study was good or bad. All they read or hear is ‘this study proved this or that” and make a binary conclusion as if it was a fact.
    Now I can see issues where $$ for small research groups is a significant obstacle in enrolling large groups. Therefore, I suggest the following idea: Segregate the literature by assigning a ‘power factor’ to articles to appear in the title (sort of like parental rating of movies); 100 or less people would get a lower rating than 1000 or 10,000 participants.

  10. Draal – I agree to a point. This is a complex issue. Smallish studies are OK to decide what will deserve the time and resources of a larger and more definitive study. Clinical trials take a lot of time and money, and patient populations may be limited. To get large numbers you may need to do a multi-center trial, which vastly increases the complexity and cost – so many people do a single center smallish study to convince their colleagues a multi-center trial is needed.

    Also, small studies are useful to hammer out the details of how a question should be studied. It’s probably a good idea for an expensive and time-consuming large trial to not be the first trial of a specific question.

    But – also, a lot of this is driven by the need for academics to publish or perish. Crank out a few smallish studies to get the requisite number of publications on your CV before you come up for promotion.

    I agree that many studies are just useless – and even worse they clog the literature with low grade and confusing studies. Poor design is worse than smallish size – I would take a tight but small study over a large sloppy or unblinded study any day. Researcher do need to really think – is this study powerful enough and tight enough to be worth it.

  11. Draal says:

    Quick background on Stacie Geller, the lead PI at UIC. She’s a CAM advocate for the treatment of menopause. She’s written a review paper saying, “The evidence to date suggests that black cohosh is safe and effective for reducing menopausal symptoms, primarily hot flashes and possibly mood disorders. Phytoestrogen extracts, including soy foods and red clover, appear to have at best only minimal effect on menopausal symptoms but have positive health effects on plasma lipid concentrations and may reduce heart disease.”
    I have to give it to her and Farnsworth. They both have been hyping up the effects of Black Cohosh in a number of reviews. I’m pleased that they put their money where their mouths were and let science function as it should. (I know a certain someone that jokes that when PhDs graduate and are handed their diplomas, the presenter’s hand anoints them with something that makes it forever impossible for them to say “I don’t know” or “I was wrong.”)

  12. HHC says:

    Why does a pharmacognosist continue to study plants? For the the same reasons that a neurologist continues to study the nervous system.

    Based on the symptoms that the studied botanical plants are treating, wouldn’t medical marijuana be a more effective drug for
    premenstrual or menopausal symptoms in women?

    And what about menopause and the male physiological equivalent? What are men’s issues?

  13. Draal says:

    And what about menopause and the male physiological equivalent? What are men’s issues?

    The Mayo clinic suggests exercise, Viagara and Prozac.

  14. SteveA says:

    Anyone wanting a demonstration of the variability of active compounds in plants should try growing chillies. All else being equal, a well fed and watered chilli plant growing in sunshine will produce dramatically hotter chillies than one grown in less optimimum conditions.

  15. Joe says:

    Is this study really large-enough? The HRT arm was a waste of time, unless it was needed as an inducement to women to enroll; in which case, the red clover arm only makes the results weaker. So, somebody can argue “more research is needed.”

    I like this analysis of the value of studying herbs absent evidence of active ingredients

    Basically, one can only conclude that this particular preparation did not work. The antimalarial drug, artemisinin, is instructive on two fronts- first, the plant (artemisia) does not always contain the drug (something implied by SteveA). Second, even when the drug is present, the method of preparation is critical to obtaining it.

  16. SteveA says:

    Just to clarify, my comment was intended to support Steve’s obvservation that it is impossible to get a ‘measured dose’ of a drug from a herbal preparation, which is dangerous. Depending on how it’s grown a preperation of foxglove leaves might contain a lethal dose of digitalis, or not. Let’s eat it and find out!

Leave a Reply