Phytotherapy is the study of the use of extracts from natural origin as medicines of health-promoting agents. Even though phytotherapy is usually regarded as "alternative medicine" in the Western countries, it is as well, when critically carried out, a essential part of modern pharmacognosy.
- 1 Points to consider in phytotherapy
- 2 Issues related to clinical trials
- 3 See also
- 4 References
Points to consider in phytotherapy
In herbal medicine standardization refers to providing processed plant material that meets a specified concentration of a specific marker constituent. However plant constituents have synergy and even active constituent concentrations may be misleading measures of potency if cofactors are not present. A further problem is that the important constituent is often unknown. For instance St. Johnswort is often standardized to the antiviral constituent hypericin which is now known not to be the "active ingredient" for antidepressant use. Other companies standardize to hyperforin or both, although there may be some 24 known possible constituents. Only a minority of chemicals used as standardization markers are known to be active constituents. Standardization has not been standardized yet: different companies use different markers, or different levels of the same markers, or different methods of testing for marker compounds. Herbalist and manufacturer David Winston points out that whenever different compounds are chosen as "active ingredients" for different herbs, there is a chance that suppliers will get a substandard batch (low on the chemical markers) and mix it with a batch higher in the desired marker to compensate for the difference.
Quality in the use of crude drugs or plant medicines depends upon a variety of factors: genetically strong seed, correct species, maturity of the plant at harvest, good soils, air quality, climate, organoleptic factors such as intensity of color, flavor and odor, processing after harvest and a variety of other factors. These conditions have been noted in historical herbals such as Culpepper's Complete Herbal  or the The Shen Nong or Divine Farmer's Materia Medica This was standard pharmacognosy curriculum for many years. Storage after collection is a factor worthy of study: Researchers in Nara, Japan have stored samples of ginseng root (Panax ginseng), licorice root (Glycyrrhiza glabra) and rhubarb root (Rheum emodi) that have been shown to retain their active properties for over 1,200 years. 
In modern times the foregoing aspects are no less important, but have been neglected with the advent of laboratory testing, although it generally is true that only certain constituents are identified and measured. Processes like HPLC (High performance liquid chromatography), GC (gas chromatography), UV/VIS (Ultraviolet/Visible spectrophotometry) or AA (Atomic Absorption spectroscopy)are used to identify species, measure bacteriological contamination, assess potency and eventually creating Certificates of Analysis for the material.
Quality should be overseen by either authorities ensuring Good Manufacturing Practices or regulatory agencies by the US FDA. In the United States one frequently sees comments that herbal medicine is unregulated, but this is not correct since the FDA and GMP regulations are in place. In Germany, the Commission E has produced a book of German legal-medical regulations which includes quality standards . 
The political issues around the safety of crude drugs vary from considering natural remedies "safe" regardless of potential dangers to considering them a dangerous unknown. In fact, noted USDA researcher and scientist Dr. James Duke estimates the dangers of herbal medicine relative to other dangers in the United States as follows:
- Herbs 1 in 1,000,000
- Supplements 1 in 1,000,000
- Poisonous Mushrooms 1 in 100,000
- NSAID's 1 in 10,000
- Hospital Surgery 1 in 10,000
- Car Accident 1 in 5,000
- Improper Use of Medication 1 in 2,000
- Angiogram 1 in 1,000
- Alcohol 1 in 500
- Cigarettes 1 in 500
- Properly Prescribed Medications 1 in 333
- Medical Mishap 1 in 250
- Iatrogenic Hospital Infection 1 in 80
- Bypass Surgery 1 in 20
In a 19 year analysis of poison control calls in the United States, 44.6% were due to pharmaceuticals and only 2.4% were due to plants. This includes not only ingestion of herbs but exposure to poison ivy, children eating houseplants and allergic responses. The vast majority did not involve herbal medicines.  More recent compilations of calls to poison control centers in the US found only 37 deaths in the highest year, but the data were not vetted in order to see whether herbs were actually involved: one death counted as "ephedra related" involved the death by automobile accident of a person with three times the legal blood alcohol level who had an unopened bottle of ephedra in the front seat, although there was no evidence that the drunk driver had ever taken ephedra. 
As Paul Bergner, Editor of the journal Medical Herbalism and author of several articles on herbal toxicity recently pointed out:
Approximately 8% of all hospital admissions in the U.S. are due to adverse reactions to synthetic drugs. That's a minimum of 2,000,000. At least 100,000 people a year die from them. That's just in the U.S., and that's a conservative estimate. That means at least three times as many people are killed in the U.S. by pharmaceutical drugs as are killed by drunken drivers. Thousands die each year from supposedly "safe" over-the-counter remedies. Deaths or hospitalizations due to herbs are so rare that they're hard to find. The U.S. National Poison Control Centers does not even have a category in their database for adverse reactions to herbs.
Similar figures apply in the United Kingdom. Hepatoxicity, where perhaps the strongest case against some herbs lies, the statistics show that over 80% of cases of fulminant hepatic failure presenting for liver transplant (or death) over ten years in the UK were due to poisoning by freely available non-prescription NSAID's, such as paracetomol and aspirin. Not one case was due to ingestion of medicinal herbs.
While there are certainly poisonous plants which have limited medicinal use, most are not sold in material doses in the United States or are available only to trained practitioners. These include Aconite, Arnica, Belladonna, Bryonia, Datura, Gelsemium, Henbane, Male Fern, Phytolacca, Podophyllum, and Veratrum. Secondly, are herbs with powerful actions, often causing nausea, sweating vomiting, (that usually were traditionally prized for this action)which are perfectly safe used under appropriate conditions, like Lobelia, Ephedra and Eonymus. Third are plants with specific toxicity like hepatotoxic pyrrolizidine alkaloids like Comfrey or Petasites. There are other plant medicines which require caution or can interact with medications including St. Johnswort or grapefruit. 
Studies done Without Proper Identification of Assay of Plant Materials
With the decline in pharmacognosy education in the United States, it has been common for herbal research to be done on plants that have not been botanically identified, which have not been assayed for strength and which do not allow for proper understanding of the herb named in the research. For instance, Eleutherococcus senticosus is frequently identified as "ginseng" although it is not part of the Panax species and has significantly different medicinal characteristics. Echinacea has been "debunked" in studies where no attempt had been made to ensure that the species was proper, the correct part of the plant had been used or that a standard dosage had been used.  As Jonathan Treasure, NIMH, has written, "Inaccurate spelling of herb names is not uncommon in medical literature as all herbalists can attest, but more frequently there is a complete failure to use proper scientific nomenclature for botanicals. This is not a trivial issue, because without a validated description of a herb by its binomial name (and preferably the naming Authority) the identity of the herb cannot be established."  Studies have often been done on herbs with adulterants or where long storage has weakened constituent levels. It is not uncommon for scientists completely lacking clinical experience in using medicinal herbs to design studies to test the efficacy of herbs, without the expertise of a consulting herbalist and thus to fail at basics like using a species correctly for the use intended, chosing medicinally appropriate forms of the herb, chosing high quality herb or using a medicinally active dosage.
Studies conducted with non-inert placebo
The loss of knowledge about the effects of plant materials has led to clinical trials being done with plant material designated as "placebo" when in fact it is not inert. In some cases the placebo was chosen to match the color, aroma or flavor of an herb when color, aroma or flavor may be correlated with biochemical constituents of similar activity. (For example yellow color correlates with the presence of berberine, an antimicrobial compound found in plants ranging from goldenseal to phellodendron to yellow root.) Since all plant-based medicinal systems tend to stereotype the actions of a plant, based upon local availability or the presence of a more potent alternative, the researcher with a casual knowledge of herbal medicine may be unaware of less common uses of plants designated as placebos. And there are medicinal aspects to sugar pills, recognized from western medicine to Ayurveda and Traditional Chinese medicine.
Standards for Random Clinical Trials
The Consolidated Standards of Reporting Trials (CONSORT) group came up with standards for random clinical trials (RTC) of herbs in 1996, revised in 2002. This statement comprises a 22-item checklist and flow diagram to guide authors, peer reviewers, editors, and readers on the essential information required in reports of two-group parallel RCTs of natural products. The CONSORT statement is endorsed by leading medical journals, editorial groups, professional societies, and funding bodies.
- The title and/or abstract should include the Latin binomial for the plant species from which the herbal medicine(s) originated, the part(s) of the plant used in the preparation, and the type of preparation (e.g., dried crude herb, ethanolic extract). The background ...should include a statement explaining the rationale for investigation of the specific herbal medicinal product and whether the indication for which it is being tested is new or is based on traditional use. Participant eligibility criteria ...in a trial testing a traditional indication (e.g., in traditional Chinese herbal medicine, a trial may test the effects of an herbal medicine intervention for liver chi (Qi) deficiency) should describe the theories and concepts underlying this indication.
- The description of the intervention ...must include the herbal medicinal product name, manufacturer, plant part used, type of preparation, source and authentication of the herbal material, pharmaceutical quality (e.g., herbal drug-to-extract ratio, type and concentration of the extraction solvent, quantity of known active constituents per unit dose), and dosage regimen and qualitative testing (purity). Also, reporting of the rationale for the control/placebo used in the trial is recommended. For studies involving herbal medicine practitioners as part of the intervention, details of practitioners (e.g., training, registration status) should be reported. Not all recommendations are relevant for all types of herbal medicine interventions. Therefore, we begin this section with the words "where applicable." For example, a report of an RCT of an herbal medicinal product comprising crude herbal material (e.g., leaves, stems, root) prepared as a tea or decoction does not require reporting of the "type and concentration of solvent used and the plant to plant extract ratio"... In addition, herbal interventions made by the investigators specifically for the study will not have a finished product or extract name or manufacturer ...For such products, all methods used in preparing and formulating the product must be reported. Similarly, allegiance is only relevant for studies in which the practitioner is a part of the intervention. In other studies, the practitioner may serve a more neutral role and thus their characteristics need not be reported. With these exceptions, all information outlined in these recommendations are suggested to be reported for all herbal medicine interventions.
- Also, outcome measures ...should reflect the intervention and indications tested while considering their underlying theories and concepts. For the results section, it is recommended that in addition to other baseline data..., RCTs of herbal medicine interventions report any concomitant medication, herbal medicinal product, or other CAM use. It is recommended that when interpreting the results ... there be consideration of the specific herbal product and dosage regimen tested. This includes an overview of evidence on this particular herbal medicinal product. When considering generalizability ...), it is suggested that authors report how the product used in the trial generalizes to products used in self-care and/or in clinical practice. Finally, when interpreting the results in the context of the evidence it is recommended that a general discussion in relation to trials of other available products should be reported.
- Alan Tillotson Growth, Maturity, Quality
- Culpeper's Complete Herbal by Nicholas Culpeper reprinted in 2003 by Kensington Arts Press
- The Divine Farmer's Materia Medica: A Translation of the Shen Nong Ben Cao (Blue Poppy's Great Masters Series) by Yang Shou-Zhong and Bob Flaws (translator) Blue Poppy 1998
- Making Sense of Commission E, review by Jonathan Treasure, 1999-2000.
- [Jane Brody. Taking Stock of Mysteries of Medicine]http://query.nytimes.com/gst/fullpage.html?res=9A05EEDC1F3EF936A35756C0A96E958260&sec=health&spon=&pagewanted=2
- Herbs: Panaceas or Poisons by David Winston 
- David Winston
- Treasurer, Treasure,Jonathan. Side effects/Toxicity of Herbs. 1995
- Treasurer, Treasure, Jonathan. Side effects/Toxicity of Herbs. 1995
- Winston, David. Herbal Medicine Introduction
- Is there a need to control the placebo in placebo controlled trials?De Craen et al. Heart.1997; 77: 95-96
- Moher D, Altman DG, Schulz KF, Elbourne DR, for the CONSORT Group. Opportunities and challenges for improving the quality of reporting clinical research: CONSORT and beyond. Canadian Medical Association Journal 2004;171:349-350.
- http://www.herbalgram.org/herbalgram/articleview.asp?a=3008/ Improving the Quality of Reporting Randomized Controlled Trials Evaluating Herbal Interventions: Implementing the CONSORT Statement. HerbalGram. 2006;71:50-56 © American Botanical Councilby Joel J. Gagnier, ND, MSc, PhD (Candidate); Heather Boon, PhD; Paula Rochon, MD, MPH; David Moher, PhD; Joanne Barnes, PhD, MRPharmS, FLS; and Claire Bombardier, MD