Herbal remedies – efficacy and interactions with conventional medicine | Lesson

Introduction

Herbal medicinal products (HMPs) are defined in national and EU Medicines Law as any medicinal product exclusively containing, as active ingredients, one or more herbal substance or one or more herbal preparation.

Herbal substances are mainly dried whole, fragmented, or cut plants, or parts of plants (flowers, roots etc). However they can include lichen, fungi, or algae, and fresh material can also be used.

Herbal preparations are extracts made with a solvent such as water or more usually ethanol or methanol mixed with water (hydroalcoholic) to give tinctures, liquid extracts, or dry extracts where the solvent has been removed by evaporation to give a more concentrated extract. Herbal preparations can also be made by distilling the herbal substance to recover what is called the ‘essential’ or volatile oil e.g. peppermint oil, eucalyptus oil or tea tree oil.

You will be familiar with some plant-derived drugs such as morphine and codeine (opium poppy), digoxin (foxgloves), vincristine (Madagascan periwinkle) and paclitaxel (Yew trees) but while these are all ‘natura’ they are not considered to be herbal medicinal products because they have been extracted, isolated, purified and formulated in the same way as a synthetic drug. You will also be familiar with the laxative preparation Senokot which is made from the powdered fruits of the senna plant (senna pods) and also with the use of peppermint oil capsules used to treat irritable bowel syndrome.

The herbs that feature in this module are in general those with moderate levels of pharmacological activity which are used to self-medicate the symptoms of minor self-limiting conditions that do not require diagnosis or ongoing monitoring by a medical practitioner. Over 100 of these have been subjected to detailed scrutiny by a specialist committee at the European Medicines Agency (EMA) which assessed the pharmacology, clinical trial evidence and toxicology of each plant before drawing up what are called Community Herbal Monographs (CHMs). These monographs list the approved therapeutic indication claims under either the Well Established Use section of the Medicines Directive, based on adequate clinical trial evidence or under the Traditional Herbal Medicinal Products Directive where pharmacological plausibility is required. In addition, dosage information, contraindications, interactions and special precautions are listed for each herb. The monographs are a key resource accessible online through the EMA website www.ema.europa.eu. Click on Find Medicine, then Herbal Medicines and follow the instructions online. Most of what follows is based on information from those EMA Monographs and Assessment Reports.

Question 1: Is there a place for the use of herbal remedies in conjunction with conventional treatments?

Many herbal medicines are used as alternatives to conventional medicines as self-medication for minor self-limiting conditions (coughs and colds) or for conditions which do not require the intervention of a medical practitioner (e.g. agnus castus for PMS). Therefore there would not be many situations where they would be used in conjunction with conventional treatments, there are some however.

Perhaps the most important would be the use of any of the three echinacea species as adjuvant therapy along with antibiotics in cases of recurrent infections of the upper respiratory tract or of the urogenital tract. Others include cranberry juice in cases of recurrent cystitis, while nettle herb and blackcurrant leaf are used as adjuvants in arthritis and rheumatic conditions. Bitter herbs such as wormwood and gentian can be used to stimulate the appetite in those convalescing after illness or surgery.

In many cases the co-administration of conventional medicines and HMPs would be inadvisable due to the risk of interactions (see Question 3 below).

Question 2: Are there any instances where herbal remedies are totally contraindicated?

Obviously patients with life-threatening conditions such as cancer, major infections, cardiac, or cerebrovascular conditions, should not be given HMPs. Some HMPs are specifically contraindicated – St John’s wort (SJW) is expressly contraindicated in transplant patients on cyclosporine, in patients with HIV on anti-retroviral therapy with protease inhibitors such as amprenavir or indinavir, in cancer patients receiving irinotecan as part of their chemotherapy regime and in patients on warfarin. Echinacea is contraindicated in immunocompromised patients because of its documented immunomodulatory effect.

Most herbs are contraindicated in those with a known hypersensitivity to that plant. More specifically plants such as arnica, chamomile, echinacea, feverfew, marigold (calendula) and yarrow are those which are most likely to cause serious hypersensitivity reactions because they are all members of the plant family asteracae or compositae – the daisy/chyrsanthemum/aster family – which are notorious for causing allergic reactions up to and including serious anaphylaxis. Cross-sensitisation is an issue, whereby a person who is sensitive to chamomile will also react to arnica or echinacea. Those with a history of atopic reactions are particularly vulnerable.

Milk thistle (Silybum) used in liver conditions, is contraindicated in cases of obstruction of the bile duct, cholangitis, gallstones and other biliary disorders. In cases of benign prostatic hypertrophy (BPH) saw palmetto (Serenoa) should only be used after diagnosis by a medical practitioner and when prostate cancer has been ruled out.

Question 3: What are the major herbal-drug interactions doctors need to be aware of?

The first red flag signalling a need for your intervention is in patients on warfarin or other anti-coagulant/anti-platelet product. Such patients should be asked not only about their other conventional medicines but also about their use of herbal medicines and dietary supplements. It is well known that the bulk of patients do not tell their doctor about their use of such products and equally most practitioners, rarely ask. Thus any questions have to be put in a sensitive and non-judgemental manner to create an environment within which the patient feels comfortable in giving you a complete picture of what they are taking.

Apart from patients on warfarin which seems to interact with almost everything, notably garlic, ginseng, ginkgo, green tea and St Johns wort, other target groups are patients with chronic conditions such as arthritis, cancer, HIV, those from certain ethnic groups where herb use is traditional, and the ‘worried well’ who swear by herbals because they don’t contain any nasty ‘pharmaceutical drugs’.

The medicines most likely to be involved in herb-drug interactions (apart from warfarin) include HIV drugs, sedatives, antidepressants and anti-diabetics. The most common herb with interaction potential to watch out for is St Johns wort.

While this is the red flag product, other herbs have interacted with conventional medicines either by altering their pharmacokinetics or their pharmacodynamics. Garlic interacts with warfarin to increase the risk of bleeding because of its anti-platelet effects and it also reduces plasma concentrations of key HIV drugs such as saquinavir. Green tea and Korean ginseng conversely may reduce the INR. The evidence around ginger and ginkgo is more uncertain. In the case of ginkgo there are case reports of increased bleeding but formal studies in patients and healthy volunteers found no interaction. Equally, clinical studies of ginger have given contradictory results. This highlights the difficulty practitioners have in providing reliable advice.

It must also be borne in mind that some interactions could be said to be more food-drug interactions given that much larger amounts of say garlic and ginger are consumed in food than as herbal medicines. For warfarin patients advice about consistency in dietary intake of herbs and spices along with green vegetables may be the way forward. In the context of food-drug interactions, the impact of grapefruit on plasma concentrations of many drugs is well known. What may be less well-known is the fact that, the chemicals in grapefruit responsible for inhibiting drug metabolising liver enzymes, are also found in other citrus fruits, some of which (bitter orange, mandarin and tangerine) are frequently found in the complex formulations typical of traditional Chinese herbal medicine (TCM). Another component of complex TCM prescriptions is liquorice that contains the chemical glycyrrhizin. This chemical has clinically documented corticosteroid effects and will potentiate the action of steroids such as hydrocortisone, prednisolone and betamethasone.

But the biggie is of course SJW because hyperforin, one of its characteristic chemical constituents, induces CYP3A4 and other CYP450 enzymes in the liver leading to significant decreases in blood levels and therapeutic effectiveness of a huge number of drugs. The most notable are cyclosporine in transplant patients, warfarin, benzodiazepines, digoxin, methadone, finasteride, protease inhibitors (amprenavir, indinavir), chemotherapeutic agents (irinotecan, vinblastine, etoposide, etc) and low dose oral contraceptives. In the latter case women using any form of hormonal contraception including the morning after pill, hormonal creams or patches, or the IUD with hormones, should be advised to use another form of contraception such as condoms. This interaction also applies to HRT patches, gels, tablets and vaginal rings. Hypericum products (SJW) should be discontinued at least one week before surgery to avoid possible interactions with general and regional anaesthetics. SJW extracts may contribute to serotonergic effects if combined with other antidepressants such as SSRIs (sertraline, paroxetine and nefazodone), buspirone or with triptans.

References

Williamson E., Driver, S. Baxter, K. (2009) Stockleys Herbal Medicines Interactions. Pharmaceutical Press London.

Posadzki P. et al (2012). Herb-drug interactions: an overview of systematic reviews, British Journal of Clinical Pharmacology 75(3): 603-618.

Shi  S., Klotz U. (2012) Drug interactions with herbal medicines. Clinical Pharmacokinetics 51(2): 77-104.

O’Connell K. et al (2014 Posterior Reversible Encephalopathy  Syndrome (PRES) associated with liquorice consumption. Irish Journal of Medical Science online October 24^th 2014.

Question 4: Should blood monitoring be carried out for people who take herbal remedies? If so, what blood tests are required?

The obvious answer here is patients on warfarin, other anticoagulants or anti-platelet medication because of the ability of virtually every herb to either increase INR or to reduce it. The most likely culprits have been listed under Question 3.

Liver function tests may be required in patients who have used, or who are still using, herbs which are known or suspected to be hepatotoxic. These include comfrey and coltsfoot, particularly if they are used as herbal teas and if the toxic alkaloid-free variety of plant (in the case of comfrey) has not been used by the manufacturer/producer.

There are over 30 case reports of hepatotoxicity associated with green tea particularly involving non-traditional consumption of large amounts of concentrated green tea extracts rather than the leaf tea. According to the Cochrane Review, desirable green tea consumption is 3-5 cups providing a minimum of 250 mg per day of chemicals called catechins responsible for the beneficial effects in cancer and cardiovascular prevention. There appears to be individual differences in sensitivity to green tea toxicity so patients who admit to green tea use should be carefully questioned about the extent of that use and the nature of the green tea product used (extract versus traditional leaf).

Another popular herb which has been the subject of concern and controversy over its alleged effects on the liver is black cohosh about which the EMA issued a public statement advising patients to stop taking black cohosh root “and consult their doctor immediately if they developed signs and symptoms suggestive of liver injury (tiredness, loss of appetite, yellowing of the skin and eyes, or severe upper stomach pain with nausea and vomiting or dark urine)”. The Committee on Herbal Medicinal Products (HMPC) at the EMA issued that advice after evaluating 42 cases of hepatotoxicity associated with black cohosh use. It noted that in only four cases was there an association between the start of treatment with the root for peri- and post-menopausal symptoms such as hot flushes, sweating etc and the occurrence of liver problems. A subsequent causality assessment of these four cases concluded that there was no evidence for a causal relationship between treatment with the herb and the observed liver disease. The CHM on black cohosh published by EMA states in the section “Special warnings and precautions for use” that “patients with a history of liver disorder should take Cimicifuga (the scientific botanical name for black cohosh) preparations with caution” and repeats the advice to stop using the preparation if any signs of liver disease show up.

In patients with a history of use of saw palmetto for the symptom of BPH, platelet function should be assessed prior to surgery because of the risk of intra-operative bleeding as a result of platelet dysfunction which may be caused by the herb.

Question 5: Do herbal remedies interact with one another?

There does not appear to be any reports of serious adverse events arising from the use of combinations of herbal materials. Combinations of herbs are more common in TCM. Combinations are also used in medical herbalism – a system of alternative medicine where the concept of illness is totally different to rational phytotherapy (the name given nowadays to scientific evidence-based use of HMPs in conventional medicine).

Combinations of sedative/anxiolytic herbs are common e.g. valerian and hops but there is little or no evidence that such commercially available combinations are more effective than the individual herbal drugs. The same applies to combinations of laxative herbs such as senna (leaf or pod) plus frangula bark. The use of liquorice with stimulant laxatives such as senna may aggravate the sodium retention and hypokalaemia associated with liquorice use, leading to hypertension and oedema.

There are examples of, within-herb interactions due to the presence of different phytochemical (plant chemical) components. One example is liquorice which owes its anti-inflammatory and anti-ulcer properties to both glycyrrhizin and liquiritin which are very different chemically. In the case of the newly licensed medicine based on extracts from the cannabis plant used to treat some of the spasticity symptoms associated with multiple sclerosis, the product contains strictly defined and controlled amounts of two separate cannabinoids (cannabis-like compounds) namely THC and CBD. This has safety implications because THC is increasingly linked to the development of psychotic symptoms, whereas CBD can antagonise the effects of THC and go some way to protect patients from the risk of psychosis.

Question 6: Is there a role for St John’s wort in the treatment of mild depression?

St John’s wort (SJW) is the plant Hypericum perforatum. The Cochrane Review states that “the available evidence suggests that hypericum extracts tested are, a) superior to placebo in patients with major depression; b) are similarly effective as standard antidepressants; c) have fewer side effects than standard antidepressants.”

Based on the clinical trial data the EMA has published a CHM on St John’s wort which must be taken into account by medicines regulators such as the HPRA. The CHM distinguishes between SJW products made from extracts using high strengths (68%, 80%) of either methanol or ethanol which are recognised as medicines under the Well Established Use section of the EU Medicines Directive. These have accepted therapeutic indications as “herbal medicinal products for the treatment of mild to moderate depressive episodes (according to ICD 10)” or “for the short term treatment of symptoms in mild depressive disorders”. These products are likely to be licensed as full medicines with a PA (Product Authorisation) number on the package with Prescription Only Medicine status because they deliver more than 1 mg per day of a plant chemical called hyperforin which is believed to be responsible for the induction of cytochrome P450 isozymes 3A4, 2C9, 2C19 as well as P-glycoprotein, leading to the interactions listed in answers to Q.2 and Q.3.

Other extracts of SJW made with lower strengths of alcohol or with lower amounts of plant material (usually labelled as the DER – Drug:Extract Ratio) deliver less than 1 mg of hyperforin and are considered unlikely to cause significant interactions. They are also less potent, so they are eligible for registration as Traditional Herbal Medicinal Products, therefore packs will have a TR number rather than a PA number and they can be sold OTC. The approved therapeutic indication is as a “traditional herbal medicinal product for the relief of temporary mental exhaustion” or low mood.

Reference

Linde K, et al. 2008. “St John’s wort for major depression”. Cochrane Database of Systematic Reviews Issue 4. Art no. CD000448. DOI: 10.1002/14651858. CD000448.pub3.

Question 7: Many women don’t wish to use HRT. What herbal remedies are beneficial for managing menopause symptoms?

Much of the use of agnus castus for the menopause has to do with the similarities between the symptoms experienced during the peri-menopause and those of PMS for which it is more widely used. Evidence from randomised clinical trials of vitex agnus castus (which is often referred to as just agnus castus) in the menopause is lacking, but pharmacological evidence is said to support a role for the berries in reducing the symptoms. This view is based on the well-recognised dopaminergic activity of extracts, reported binding to mu-opioid receptors and enhancement of melatonin secretion. It is known that endorphins decrease as the menopause progresses and this is associated with mood disorders, migraines and fluid retention. The dopaminergic activity is relevant to reductions in hot flushes and of the emotional symptoms associated with the menopause. The effect on melatonin secretion relates to the possible role declining melatonin production has in menopause-related sleep disturbance because vitex can increase levels by up to 60%.

There have been a number of trials with mono-preparations of agnus castus for PMS which show effects superior to placebo, whereas trials in the menopause have tended to use complex formulations which make it impossible to draw conclusions about the effectiveness of any single herb in the mixture. In both sets of trials small numbers were involved and there was a very high placebo response. The PMS studies do show evidence of beneficial effects and the HMPC at the EMA has published a monograph on agnus castus which describes the material as a “traditional herbal medicinal product for the relief of minor symptoms in the days before menstruation (premenstrual syndrome)”.

Black cohosh or cimicifuga is widely used in menopausal women, and the EMA monograph on the plant states that black cohosh preparations come within the scope of the “well established use” Article (10a) of the EU Medicines Directive, which means that the level of evidence from standard clinical trials is deemed to be sufficient. The therapeutic indication accepted by the EMA is “Herbal medicinal product for the relief of menopausal complaints such as hot flushes and profuse sweating.”

That indication is based on a detailed assessment of 27 clinical trials involving approximately 5300 women, which concluded that there was sufficient evidence from 23 of those trials to use preparations of black cohosh to treat minor menopausal symptoms. The assessment also noted that there are no alternative therapeutic options, especially in peri-menopausal women for whom HRT would be inappropriate and in women with breast cancer. However, the monograph states categorically that patients who have been treated or who are undergoing treatment for breast or any other cancer should not use the plant because the data on effects on hot flushes in patients with breast cancer is conflicting and there are no randomised controlled trials assessing the efficacy of the plant for breast cancer.

Question 8: Can herbal remedies be used safely in children?

Most HMPs are restricted to adults because the EMA has determined that experience of efficacy and safety in children has not been sufficiently documented. There are however a number of exceptions. For example arnica tincture may be used topically in children for the relief of bruises, sprains and contusions. Under no circumstances should it be administered internally to children or indeed adults because of the risk of cardiac side effects. Equally it should not be used on broken skin due to the risk of absorption of toxic constituents.

A number of HMPs may be used as expectorants in children in cases of cough associated with the common cold including thyme, primrose root, ivy and Pelargonium sidoides from South Africa. In the case of thyme it can be used in children over the age of 4, for pelargonium the permitted age is 6 or over, while primula cannot be used in those under 12 and ivy is contraindicated in those under 2.

The best-known herb for use in the prevention and treatment of the common cold is undoubtedly echinacea which is the generic botanical name for three separate species the best known of which is Echinacea purpurea or purple coneflower. It has a Well Established Use according to EMA in the common cold, but is contraindicated in infants less than 1 year old and not recommended in those under 12 because safe use has not been documented in that age group and there is an elevated risk of atopic reactions.

Question 9: Are herbal remedies safe to use in pregnancy?

The international data shows that between 7% and 45% of women report consuming such products during pregnancy.

One detailed Canadian study found that the most widely used herbal materials during pregnancy were linseed, chamomile, peppermint and green tea. No association was found  between chamomile, peppermint or green tea use in the last two trimesters and the risk of premature birth. There was an increased risk with linseed.

Another study looked at raspberry leaf tea used by large numbers of women to ease childbirth but found no adverse events had been reported even though one authoritative British text cautions against its use. The Canadian researchers make the valid point that such overly cautious advice, which runs counter to the experience of thousands of women (that raspberry leaves are benign) undermines the credibility of healthcare professionals when they advise caution about herbs which truly are harmful during pregnancy.

Among the real no-nos are herbs which are known abortifacients, such as pennyroyal and blue cohosh. The latter should not be confused with black cohosh which is a totally unrelated plant. Blue cohosh is a Prescription Only Medicine in this country because of its effects.

Other plants to be avoided are the laxative form of aloes, related laxatives such as cascara, frangula (buckthorn) and senna where there is a risk of uterine stimulation. Parsley in concentrated form such as parsley oil or seed should also be avoided, as should goldenseal (Hydrastis), because of their uterine stimulant properties. Another plant with a history of traditional use as uterine stimulant/abortifacient includes yarrow, which the EMA states in its assessment is contraindicated during pregnancy because of its traditional use for stimulating uterine contractions. The EMA report notes that while three experimental studies on embryotoxicity and reproductive effects demonstrate relatively marginal effects, those tests are not considered adequate.

Some plants have been tested e.g. St John’s wort, used for mild to moderate depression, has shown no adverse effects in pregnant rats and dogs. In the case of echinacea, a carefully controlled prospective study of exposure during the first trimester found no statistically significant difference in the incidence of major malformations. This provides some level of confidence that the use of echinacea during pregnancy is likely to be safe.

In cases where there have been no formal studies it is sometimes possible to assess potential risks from knowledge of the chemical composition of the plant in question. For example, herbs containing thujone are inadvisable as is any plant containing hepatotoxins such as comfrey, coltsfoot and germander. In the case of thujone – found in large amounts in wormwood, tansy and sage – the EMA issued a Draft Public Statement in which it was noted that thujone is a neurotoxin causing epileptic-type convulsions in animals and humans. Because there are no preclinical or clinical studies which would allow for a scientific assessment of the consequences of exposure of pregnant women to thujone, the report states that such exposure should be minimised. Exposure to thujone from food could be of the order of 1 mg per day but this would not be considered as a specific cause of concern. A cause for concern would be exposure above a 6 mg daily limit whether someone is pregnant or not. For pregnant women the key message would be to avoid concentrated forms e.g. sage oil, cedar leaf oil (in aromatherapy perhaps?) but if they like sage and onion stuffing then there is no reason why they cannot continue to do so – assuming it doesn’t cause them heartburn.

So the advice is to watch out for obvious herbs such as those that stimulate the uterus and the hepatotoxic herbs which may harm the foetus.