June 2000; Volume 3; 64-69
Herb-Drug Interactions: An Evidence-Based Table
June 1, 2000
Herb-Drug Interactions: An Evidence-Based Table
June 2000; Volume 3; 64-69
By Mary L. Hardy, MD
As patients’ use of herbal products increases, so do physicians’ concerns regarding the possibility of herb-drug interactions. Uncertainty in this area is rife (e.g., active constituents, mechanisms of action, consistency of products) and complicates the assessment of available data. Current literature consists mainly of case reports that often are not adequately investigated. There are few clinical trials and the pharmacologic data available have not been assessed for clinical relevance.
Further, patients have been reluctant to disclose fully their use of natural products to their physicians, so interactions generally only come to light when a serious problem occurs. Most doctors, without adequate training in this area, feel uncomfortable commenting on or even reporting cases that involve the use of herbal medications. Too much of our experience is theoretical or anecdotal.
For commonly used herbs and commonly prescribed drugs, I have assembled a detailed compendium of herb-drug interactions. This table is designed to provide clinicians with guidance in assessing the potential for interaction. It cites a mechanism for interaction where one is known or postulated. It reports the level of evidence for that interaction. It also offers a clinically useful scale for the evidence, outlined in Table 1.
| Table 1-Level of evidence to support use | |
| CT = Controlled trial | CS = Case series |
| CR = Case report | AS = Animal study |
| TU = Traditional use | P = Pharmacology |
| TH = Theoretical | |
The drug-herb interaction table is not exhaustive and should be considered a work in progress. New data will become available, and as we learn more about which herbs interact with which drugs, we will report significant findings in future newsletters. A limited bibliography is available on request.
Dr. Hardy is Medical Director at Cedars-Sinai Integrative Medicine Medical Group in Los Angeles.
| Table 2-Herb-Drug Interactions | |||
| Drug Category | Herbs | Herb Effect | Mechanism (Evidence Type) |
| Alkaloids | High tannin-containing | Decreased plasma levels | Precipitation of alkaloids by (e.g., caffeine-containing tannins (TU) herbs, cat’s claw, tea, uva ursi) |
| Anesthetics | Kava, valerian | Prolongation of sedation | Additive effect (CR) time |
| Antihypertensives | a. Licorice | Decreased therapeutic effect | a. Increased salt and water retention (CR) |
| b. Sympathomimetic herbs (e.g., ephedra) | b. Opposition of therapeutic action (P) | ||
| Antiarrhythmics | Cathartic laxatives (e.g., aloe, cascara, senna, yellow dock), diuretics (e.g., celery seed, corn silk, horsetail, juniper), licorice | Increased side effects (arrhythmia) | Increased potassium loss (P) |
| Antiarrhythmics | Anticholinergic herbs (not generally used clinically, e.g., belladonna) | Decreased therapeutic effect | Decreased absorption (P, TH) |
| Anticoagulants | Antiplatelet-aggregating (e.g., Panax ginseng, feverfew, garlic, ginkgo) | Increased side effect (bleeding) | Inhibition of platelet aggregation through inhibition of thromboxane synthetase (ginger) (P); arachadonic acid production (feverfew) (P); inhibition of epinephrine induced in vitro (garlic) (P); platelet thromboxane synthetase aggregation (garlic) (P, CR); inhibition of platelet activating factor (ginkgo) (CR) |
| Anticoagulants: Warfarin | Panax ginseng, St. John’s wort | Opposition of therapeutic effect; decreased enzyme bioavailability | Unknown (CR); hepatic induction (CS) |
| Anticoagulants: Warfarin | Coumarin-rich herbs, (e.g., sweet clover, danshen), white clover | Increased therapeutic effect | Only danshen has been observed to do this clinically. Increased maximum concentration and decreased volume of distribution (CR, P) |
| Anticoagulants: Warfarin | Vitamin K-rich herbs (e.g., collard, kale, spinach) | Decreased therapeutic effect | Opposes activity (CR, P) |
| Anticonvulsants | a. GLA-rich herbs | Decreased therapeutic effect | GLA (CR) and thujone may |
| b. Thujone-containing herbs (e.g., cedar, tansy, sage) | decrease seizure threshold; mechanism unknown | ||
| Drug Category | Herbs | Herb Effect | Mechanism (Evidence Type) |
| Anticonvulsants | Salicylate-rich herbs (e.g., cramp bark, willow, wintergreen) | Increased therapeutic effect | Transient; unknown mechanism (CR) |
| Anticonvulsants: Phenytoin | Shankapulshpi (Ayurvedic preparation with multiple herbs) | Opposition of therapeutic action | Decreased effectiveness of drug; decreased drug levels (CR) |
| Antiplatelet-aggregating | Antiplatelet-aggregating (e.g., Panax ginseng, feverfew, garlic, ginkgo) | Increased side effect (bleeding) | Similar therapeutic action (P, CR) |
| Barbiturates | Valerian | Increased therapeutic effect; increased side effects | Shown to prolong barbiturate-induced sleep (AS) |
| Benzodiazepines | St. John’s wort, kava | Decreased therapeutic efficacy; may increase side effects; increased sedation | Herb binds to GABA receptor site (AS, P) |
| Cardiac glycosides | Cardiac glycoside- containing herbs | a. Enhanced therapeutic effect | Same active constituents (TH) |
| (e.g., foxglove, lily of the (arrhythmia) | b. Increased side effects valley) | ||
| Cardiac glycosides | Cathartic laxative herbs (e.g., aloe, cascara, senna, yellow dock), licorice, diuretic herbs (e.g., celery seed, corn silk, horsetail, juniper) | Increased side effects (arrhythmia) | Increased potassium loss (TH) |
| Cardiac glycosides | Quinine-containing herb (e.g., cinchona bark) | Increased plasma levels | (TH) |
| Cholesterol-lowering drugs | Garlic, artichoke, ginger, fenugreek | Increased therapeutic effect | Similar clinical effect via different mechanism (TH) |
| Corticosteroids | Cathartic laxative herbs (e.g., aloe, cascara, senna, | r>r>Increased side effects yellow dock), diuretic herbs (e.g., celery seed, corn silk, horsetail, juniper) | Both cause increased potassium loss (TH) |
| Corticosteroids | Licorice | Increased plasma levels | Increased half-life (increased bioavailability) (CR); inhibition of ll-ß-dehydrogenase (P) |
| Corticosteroids | Panax ginseng | Increased side effects | Similar side effects of CNS stimulation and insomnia (CR) |
| Digoxin | Siberian ginseng | Increased plasma level | Mechanism unknown; validated by rechallenge (CR) |
| Drug Category | Herbs | Herb Effect | Mechanism (Evidence Type) |
| Digoxin | a. Kyushin (Chinese remedy containing the without toxic effects venom of the Chinese toad) | Increased serum levels | Interferes with assay (P, CR) |
| b. Panax ginseng | |||
| Diuretic: Lasix | Panax ginseng | Decreased therapeutic effect | Diuretic resistance with ginseng; unknown mechanism (CR) |
| Diuretic: Potassium sparing | Licorice | Decreased therapeutic effect | Interferes with potassium-sparing effects by wasting K+ |
| Estrogen replacement therapy | a. Herbs high in phytoestrogens (e.g., soy, fenugreek, licorice, black cohosh) | a. Increased therapeutic effect to excess | a. Never reported (TH) |
| b. Panax ginseng | b. Increased side effect (estrogen excess) | b. Reported in few cases to produce postmenopausal bleeding or mastalgia (CR) | |
| General medication | High-fiber herbs (e.g., flax, psyllium, acacia,slippery elm, marshmallow) | Decreased absorption | (P) |
| General medication | "Hot" remedies (e.g., ginger, garlic, black pepper, red pepper) | Increased absorption | Taken internally, "hot" remedies lead to vasodilatation of gut wall and increased absorption (TU) |
| GI motility drugs | Anticholinergic herbs (not generally used clinically, e.g., belladonna) | Decreased activity | Opposition of therapeutic activity |
| Hepatotoxic drugs | Hepatotoxic herbs (e.g., borage, coltsfoot, comfrey, rue, tansy) | Increased side effect (hepatotoxicity) | Additive toxicity from similar side effects (CR) |
| Hypoglycemic agents: | Hypoglycemic | Enhanced therapeutic effect | a. Direct hypoglycemic |
| Oral and insulin | (e.g., Panax ginseng, garlic, | activity (CR, AS, P) | |
| fenugreek, bitter melon, | b. Decreased glucose | ||
| aloe, gymnema) | absorption | ||
| Hypoglycemic agents: Oral and insulin | Hyperglycemic (e.g., cocoa, rosemary, stinging nettle) | Decreased therapeutic effect | Direct opposition of therapeutic action (CS) |
| Immune suppressants | Echinacea, astragalus | Opposition of therapeutic action | General immune stimulation by these herbs may interfere with ability of immunosuppressive drugs to prevent tissue rejection; never reported (TH) |
| Drug Category | Herbs | Herb Effect | Mechanism (Evidence Type) |
| Iron | Tannin-rich herbs (e.g., caffeine-containing herbs, cat's claw, tea, uva ursi) | Decreased therapeutic effect | Tannin binds with iron, decreasing absorption (TH, P) |
| Lithium | Diuretic herbs (e.g., celery seed, corn silk, horsetail, juniper) | Increased side effects | Decreased sodium leads to increased lithium toxicity |
| Lower seizure threshold (drugs that) | GLA-rich herbs (e.g., evening primrose, borage, black currant) | Increased side effect to additive side effect | Decreased seizure threshold (CR) |
| Methotrexate and similar cytotoxic drugs | Salicylate herbs (e.g., cramp bark, willow, wintergreen) | Increased plasma levels (toxicity) | Decreased excretion (TH) |
| Minerals | Fiber-containing herbs (e.g., flax, psyllium, acacia, slippery elm, marshmallow) | Decreased bioavailability | Psyllium has been reported to decrease the absorption of Ca, Mg, Cu, Zn (CR) |
| Monoamine oxidase inhibitors (MAOIs) | Panax ginseng, bioactive amines, licorice | Increased side effects | Additive side effects may lead to toxicity; glycyrrhizin is reported to be a very potent MAOI (TH, CR) |
| Monoamine oxidase inhibitors (MAOIs) | Ginkgo | Increased therapeutic effect; increased side effects | Inhibition of monoamine oxidase (P) |
| Nonsteroidal anti-inflammatory drugs | Gastric irritant herbs (e.g., caffeine, rue, uva ursi) | Increased side effects | Similar side effects may increase risk of gastric erosion (NSAIDs) and bleeding (TH) |
| Nonsteroidal anti-inflammatory drugs (NSAIDs) | Nettles | Increased therapeutic effect | Potentiation of the anti-inflammatory activity of NSAIDs (CT) |
| Opioids | Panax ginseng | Decreased therapeutic effects | Animal model demonstrated the blunting of the analgesic effects of morphine via a non-opioid receptor-mediated mechanism (AS) |
| Photosensitizing drugs | Photosensitizing herbs (e.g., St. John's wort, angelica, rue, fennel) | Increased side effects | Furanocoumarins found often in umbelliferae resemble psoralens (P, AS, CR) |
| Salicylates | Herbs that alkalinize urine | Decreased plasma levels (e.g., uva ursi) | Increased urinary excretion (P) |
| Sedative hypnotics | Opioid herbs (e.g., opium poppy, California poppy) | Increased side effects (CNS depression) | Additive side effects |
| Drug Category | Herbs | Herb Effect | Mechanism (Evidence Type) |
| Sedative hypnotics including alcohol | Sedative herbs (e.g., hops, kava, valerian) | Increased therapeutic action; increased side effects (CNS depression) | Additive effects lead to CNS depression except valerian does not potentiate the effects of alcohol (AS, P) |
| SSRIs | St. John’s wort | Increased therapeutic activity; increased side | May contribute to serotonin syndrome—similar action effects (TH) |
| Statin drugs | Red yeast (Cholestin®) | Increased therapeutic effect | Similar active compounds; not known if taking both products simultaneously increases side effects of statin drugs (TH) |
| Thyroid hormone | a. Horseradish | a. Decreased therapeutic effect | a. Depressed thyroid function |
| b. Kelp | b. Increased therapeutic effect | b. Iodine in kelp may result in hyperthyroidism (TH) | |
Which class of herbs generally is not used clinically?
a. High-tannin containing
b. Cathartic laxatives
c. Anticholinergic
d. Coumarin-rich