Get Involved
About Us
Our Members
Herbal-Drug Interactions and Adverse Effects: an evidence-based quick reference guide
Herbal-Drug Interactions and Adverse Effects: an evidence-based quick reference guide

Herbal-Drug Interactions and Adverse Effects: an evidence-based quick reference guide by Richard B. Philp. New York: McGraw-Hill, 2004. 335 pp. soft-cover. ISBN:0-07-142153-X. $29.95.

Due to growing concerns over problems that may arise from using herbal preparations in combination with pharmaceutical drugs, the number of articles and books on the subject has been steadily increasing. The addition of this text begs the question of how it distinguishes itself from the rest. The intent is indicated clearly in the subtitle and further outlined in the 8-page Overview and General Principles which ends by stating: “This text is intended to be a quick source of information for physicians and other health care professionals who wish to know something of the potential hazards associated with a particular herb or nutriceutical, especially as they relate to the concomitant use of conventional pharmaceuticals.”

The author is a professor in pharmacology and toxicology at the University of Western Ontario in Canada. In the overview he makes his preference for drugs as therapeutic agents clearly known by emphasizing the comparative lack of scientific evidence for botanical safety and efficacy. The reliance on evidence as the determining factor for valid assessment is the stated raison d’être for this text, yet the standards constituting reliable evidence are not given. While Philp utilizes a limited number of references from scientific and medical literature for each herb, suggestive evidence seems to predominate. Most herb experts would agree that there is limited value from data derived from in vitro laboratory cell cultures or animal studies utilizing isolated pure compounds or fractions (multiple compounds of related chemical structure) from herbs. The same is true of isolated case reports involving uncharacterized (i.e., poorly described or lacking documentation of identity and/or content) products. Whether these are useful for predicting clinical outcomes for other preparations of the herb is debatable. The author frequently fails to characterize the type of research or study being cited, while seldom noting the botanical form and its dosage and duration, thereby rendering the “evidence” nebulous and its significance further suspect. The reader is expected to accept that such limited research findings provide scientific credibility, but extrapolating the data is based largely on supposition and can lead to questionable speculation.

The best points about this book are those aspects described as the author’s intent. It is a concise, easy-to-access accumulation of information with a listing of the reference citations accompanying each herb. The 160 herbs and several nutriceuticals are alphabetically listed by common names, with a botanical scientific name cross-referencing list in appendix XV. Some of the nutriceuticals, described as relatively pure dietary ingredients consumed in large quantities, seem appropriate for inclusion in an “herbal” book (guar gum, avocado/soybean unsaponifiables). Others can arguably be classified as drugs (gossypol from cotton [Gossypium spp.] seed, capsaicin from Capsicum spp. peppers) or are not derived from plants (L-carnitine, chondroitin and glucosamine sulfate, L-tryptophan, melatonin, SAMe, DHEA). Vitamins and minerals are excluded. Some overtly toxic and generally unavailable plants used as the source of conventional drugs are listed together with herbs commonly sold and used for self-treatment. The more toxic botanicals include the anticholinergics European mandrake (Atropa mandragora, syn. Mandragora officinarum L., Solanaceae) and henbane (Hyoscyamus niger L., Solanaceae) and the cardiotonics white squill (Scilla maritima, syn. Urginea maritima L., Liliaceae) and foxglove (Digitalis lanata Ehr., D. purpurea L., Scrophulariaceae).

Summaries addressing the pharmacology of extracts, fractions, and components can at times provide a useful window into the probable action of the whole herb or certain extracts. A section on adverse effects and interactions covers both issues under the same heading in an unusual choice of formatting. A clear distinction usually exists between an herb’s inherent risks and those which may occur only in conjunction with simultaneous exposure to certain medications. As with the pharmacology section, this information often fails to clearly and specifically detail the type of evidence and the character and dosage of the preparations. The implied generalization of research data derived from narrow or undefined parameters taints the evidence with even greater uncertainty in an attempt to simplify unavoidable botanical complexity.

The assumption that botanical activity is comparable to isolated drugs is one unreliable aspect of this text. Such conjecture based on a partial and limited knowledge of botanical pharmacology can be misleading. For example, the author relays medical concerns that saw palmetto (Serenoa repens Small., Arecaceae) could mask prostate cancer because pharmaceutical 5-alpha-reductase inhibitors are known to lower PSA (prostate-specific antigen) levels, resulting in false-negative cancer screening tests for this marker. In fact, evidence from clinical trials have shown this is not the case; two different commercial saw palmetto preparations in daily doses of 320 mg did not change PSA levels after 26 weeks.1, 2 These results are common knowledge among those familiar with clinical research on saw palmetto preparations in the treatment of benign prostatic hyperplasia.

Facts can be misrepresented by failing to fully disclose them. An example where this results in an exaggeration of probable risk is found in the case of milk thistle (Silybum marianum Gaertn., Asteraceae) and its silymarin-standardized extract. The author noted with two citations that major silymarin flavonolignans in micromolar concentrations were shown to inhibit various cytochrome P450 (CYP) isozymes (2C9, 2D6, 2E1, 3A4)—enzymes in the human gut and liver known to metabolize specific drugs. Dr. Philp speculates that this “could lead to unexpected increases in plasma levels of many drugs.” The studies actually found the flavonolignans had relatively weak effects that required in vitro concentrations from 10-200 mcM for CYP 2C9, 2D6, 2E1, and 3A4 substrates. Physiological concentrations of the individual flavonolignans peak at 0.6 mcM in the plasma and 150 mcM in lipophilic bile secretions. The authors of the two cited studies separately concluded: “No drug interactions should be expected using therapeutic doses of silymarin,” though with a limited risk, i.e., “Metabolic interactions with xenobiotics metabolized by CYP3A4 and CYP2C9 cannot be excluded.”

In regard to actual human effects when using milk thistle extracts, the book acknowledges that one study using the antiretroviral drug indinavir, metabolized by CYP3A4, with 459 mg silymarin (from 525 mg extract) daily for 13 [sic; actually 3] weeks showed modest though insignificant reductions in the blood. Based on the in vitro data, theoretically indinavir would be expected to increase. Speculative risk should not be given undue emphasis if it conflicts with documented human outcomes. Study results with healthy humans published too recently for inclusion in the book confirm silymarin’s lack of significant effect at 480 mg daily for 2 weeks on the metabolism of indinavir3 (summarized in HerbalGram 63 Research Reviews). Likewise, an uncharacterized milk thistle preparation given for 4 weeks to human subjects failed to influence drug metabolism significantly when taken with other substrates of CYP isozymes 2D6, 2E1, and 3A4.4 With the ongoing rapid growth of herb-drug interaction research data, these examples illustrate the problem of relying exclusively on a printed reference source that in some aspects is inevitably outdated by the time it is published.

The appendices list short collections of agents with similar potential interactions, adverse effects, or therapeutic effects and uses. These lists are not intended to be exhaustive and are accompanied by only a few major reference citations. Except for hypoglycemic herbs, the information provided in the appendices does not clearly distinguish between documented clinical and theoretical interactions but simply lists herbs and drugs according to categories. Interactions described (though not specified) as either confirmed or theoretical are given for pharmacological categories in Appendix I, coagulation (presuming incorrectly that all coumarins are anticoagulant and all salicylates are anti-platelet); Appendix II, psychotropics; Appendix VI, drug metabolism; and for potential interactions Appendix XI, diuretics. Similar adverse effects are used for grouping herbs in appendices III, hepatotoxins; V, dermatitis; VIIA, cardiovascular dysfunctions; IX, Asteraceae allergies; XI, nephrotoxins; XIIA, carcinogens; and XIII, photodermatitis. Possible therapeutic activity groups are given in appendices IV, phytoestrogens; VIIB, hypolipidemics; VIII, anti-inflammatories; X, antihyperglycemics; and XIIB, antineoplastics.

Though presumably implicated by their inclusion as likely causing additive adverse interactions, many herbs in the therapeutic effects appendices have a reasonable potential as beneficial combinations with similar drugs, e.g., as has been demonstrated with anti-inflammatory drugs.5 Failing to address the evidence for positive interactions by focusing only on the “dark side” of herbal preparations creates an impression of lack of balance. However, the author states in the Overview: “It is not intended to be a reference work for those contemplating the therapeutic use of herbs. Nor is it meant to provide a source of in-depth information for those seeking details concerning all aspects of a particular herb.”

Unfortunately, treating herbs as simple drugs that can be explained primarily on the basis of limited and fragmented scientific evidence is usually unsatisfactory. Informed guidance involves acknowledging the complexity that necessarily complicates botanical issues. Until patterns of influence become established in humans for specific preparations from each herb, we need to know more, not less, about pertinent details to more accurately assess information obtained from both empirical and analytical sources. Full disclosure is necessary to effectively compare and contrast data and to obtain a more complete appreciation of the facts and possibilities.

The value of scientific research data and mechanistic explanations as it applies to human biological interactions seems greatest when considered in conjunction with the larger context of empirical observation found in human traditional medical systems and common contemporary use. One informs the other, but where there is apparent conflict, I prefer human experience over conceptual speculation based on fragmentary laboratory “evidence.”


—Francis Brinker, N.D. Clinical Assistant Professor Program for Innovative Medicine College of Medicine,University of Arizona, Tucson

Author of Herb Contraindications & Drug Interactions 3d ed. and Complex Herbs – Complex Medicines: A Merger of Eclectic & Naturopathic Visions of Botanical Medicine



1. Carraro J-C, Raynaud J-P, Koch G, et al. Comparison of phytotherapy (Permixon®) with finasteride in the treatment of benign prostate hyperplasia: a randomized international study of 1,098 patients. Prostate. 1996;29:231-240.

2. Gerber GS, Zagaja GP, Bales GT, et al. Saw palmetto (Serenoa repens) in men with lower urinary tract symptoms: effects on urodynamic parameters and voiding symptoms. Urology. 1998;51(6):1003-1007.

3. DiCenzo R, Shelton M, Jordan K, et al. Coadministration of milk thistle and indinavir in healthy subjects. Pharmacother. 2003;23(7):866-870.

4. Gurley BJ, Gardner SF, Hubbard MA, et al. Assessment of botanical supplementation on human cytochrome P450 phenotype: Citrus aurantium, Echinacea, milk thistle, saw palmetto. Clin Pharmcol Ther. 2004;75(2):P35.

5. Brinker F. Enhancing anti-inflammatory and analgesic drug effects while reducing risks with herbs and their derivatives. Integrat Med. 2004;3(1):24-42.