Evaluation of Documented Drug Interactions and Contraindications Associated With Herbs and Dietary Supplements

A Systematic Literature Review

H.-H. Tsai; H.-W. Lin; A. Simon Pickard; H.-Y. Tsai; G. B. Mahady


Int J Clin Pract. 2012;66(11):1056-1078. 

In This Article


Literature Search

Finally, 461 articles of primary literature were initially identified. Eighty-five articles with full text, including 54 review articles, other than the 6 books and 2 web sites were selected for further review (Figure 1). The summaries of the animal studies, observational studies and clinical trials to retrieve information about HDS–drug interactions and contraindications for the original studies are listed in Table 1 , Table 2 and Table 3 , respectively. The summaries of the retrieved books and reviewed articles to retrieve information about HDS–drug interactions and contraindications were listed in Appendix 1 , Appendix 2 and Appendix 3 , respectively. Among the original studies (n = 31), more than half (n = 16) were clinical trials. All of these articles contained information about HDS–drug interactions,[12,30–113] but only five articles provided descriptive information about HDS contraindications.[55–57,59,102]

Figure 1.

Flow chart of primary literature search

Quantity of Retrieved Evidence

After excluding the evidence regarding HDS not recommended for human use (i.e. anvirzel, belladonna, chaparral, comfrey, ephedra and pennyroyal)[16,19,21–23] and the duplicates, a total of 1491 unique pairs of documented interactions between HDS and individual drugs were identified. Among these pairs, 814 pairs (54.6%) were retrieved from the primary literature, 1018 pairs (68.3%) from books and only 23 pairs of interactions were identified in the two reviewed web sites. Among these interactions, the corresponding mechanism and severity was determined for 507 pairs (34.0%) using MicroMedex® and 763 pairs (51.2%) in the NMCD® online database. In total, 882 pairs (59.2%) of documented HDS–drug interactions were identified for their potential mechanism and severity. In terms of contraindications, there were 128, 15 and 9 documented HDS contraindications retrieved from books, primary articles and web sites, respectively, for a total of 152.

HDS–drug Interactions

Among all included interactions between HDS and individual drugs, 166 different herbs/botanical products, 28 VMA and 19 other supplements accounted for 890 pairs (59.7%), 529 pairs (35.5%) and 72 pairs (4.8%) of documented interactions, respectively (Figure 2). The top five herbs/botanical products, which were documented to have the most interactions with individual medications, were St John's Wort (Hypericum perforatum), ginkgo (Ginkgo biloba), kava (Piper methysticum), digitalis (Digitalis purpurea) and willow (Salix alba). For example, St John's Wort, magnesium, calcium, iron and ginkgo have been documented to interact with 147, 102, 75, 71 and 51 individual medications, respectively. Furthermore, a total of 509 unique drugs contributed to the 1491 documented pairs of interactions with HDS. The majority of these medications (n = 100) were categorised as treatment for central nervous system (CNS), second were those medications affecting the cardiovascular system and then systemic anti-infective drugs (n = 90 and 75, respectively) (Figure 3). The medications that most contributed to documented interactions with HDS were warfarin, insulin, aspirin, digoxin and ticlopidine. Not surprisingly, warfarin was documented to have interactions with over 100 HDS entities (Figure 4).

Figure 2.

Herbs and dietary supplements tended to have documented interactions with medications in each caterory. VMA, vitamin/mineral/amino acid; DS, dietary supplements; DHEA, dehydroepiandrosterone

Figure 3.

Distribution of medications that might have interactions with herbs and dietary supplements. ATC, anatomical therapeutic chemical. The number of total medications was 509

Figure 4.

Medications with the largest number of interactions with herbs and dietary supplements. HDS, herbs and dietary supplements

Among 882 pairs of interactions with identified mechanisms, a total of 373 pairs (42.3%) were attributable to pharmacokinetic-related mechanisms, i.e. affected the absorption, distribution, metabolism or excretion of the HDS/drug. Approximately 40.1% of all interaction pairs accounted for pharmacodynamic-related mechanisms, and 8.5% were attributed to a combination of both mechanisms. No mechanism was identifiable for the remaining 9.1% of pairs. Among the 373 documented HDS interaction pairs that were pharmacokinetic-related, 87 pairs were associated with St John's Wort (23.3%), whereas calcium supplements were involved in 47 pairs of documented interactions (12.6%), and iron was involved in 42 pairs of interactions (11.3%). St John's Wort was documented to reduce the effectiveness of alprazolam, amitriptyline, imatinib, midazolam, nifedipine and verapamil via the CYP (Cytochrome P450) 3A4 pathway, and the plasma levels of fexofenadine and digoxin via PgP (p-glycoprotein) pathway. Some drugs (i.e. atorvastatin, cyclosporin, indinavir, nevirapine and simvastatin) were documented to interact with St John's Wort through both pathways.[37,99] Among the 354 documented interactions that were pharmacodynamic-related, kava accounted for 4.8% pairs of interactions (17 pairs). St John's Wort and ginkgo were both involved in 15 pairs of interactions (4.2%). Risk of additive serotonergic effects were increased when St John's Wort was used concurrently with monoamine oxidase inhibitors (MAOI), selective serotonin reuptake inhibitors (SSRI), or tryptamine-based drugs causing symptoms of anxiety, dizziness, restlessness, nausea and vomiting.[16–18,20] As a result of their pharmacological actions on the GABA receptor, synergism in CNS adverse events may result from taking barbiturates or benzodiazepines in combination with kava.[16,20,98] Furthermore, kava may worsen the extrapyramidal effects associated with the use of droperidol, haloperidol, metoclopramide or risperidone because of a dopamine.[21,98]

Among the 507 documented interaction pairs identified with a severity rating in MicroMedex®, 69.4% were categorised as the moderate interactions, 17.2% as major interactions, 10.3% as minor interactions and 3.1% were attributable to the contraindications. As for the 763 pairs of documented interactions being identified with the severity rating based on the NMCD®, the majority documented interaction pairs were categorised as moderate (69.2%), major (26.5%) and minor (4.3%). Approximately, 240 documented HDS–drug interactions were categorised as major severity in either database (Table 4 and Table 5). For example, the following pairs of interactions were considered as being contraindicated for concurrent use in MicroMedex®: l-Tryptophan vs. MAOI (i.e. isocarboxazid, phenelzine and tranylcypromine) or venlafaxine and St John's wort vs. protease inhibitors (i.e. amprenavir, fosamprenavir and indinavir), irinotecan, rasagiline or voriconazole, respectively. Among the 390 documented interaction pairs having severity ratings in both databases, 41.3% were inconsistent. For example, the combination of alfalfa (Medicago sativa) and warfarin were considered as the minor interaction in MicroMedex®; however, it was rated as the major interaction in NMCD®. The combination of St John's Wort with quetiapine, quinidine, risperidone or sildenafil gave severity ratings major according to NMCD®, and no interaction was reported in MicroMedex®.

HDS Contraindications

Fifty-nine HDS from 152 reports were contraindicated for use among patients with specific disease states. The reports were classified into 19 disease states, including gastrointestinal diseases, neurologic disorders, renal/genitourinary diseases, neoplastic disorders, diseases of the liver/gallbladder/bile ducts and cardiovascular diseases (Figure 5). Flaxseed (Linum usitatissimum), echinacea (Echinacea purpurea) and yohimbe (Pausinystalia yohimbe) had the highest number of documented contraindications. For example, flaxseed was documented to have contraindications associated with gastrointestinal disorders such as acute or chronic diarrhoea, oesophageal stricture, inflammatory bowel disease, hypertriglyceridemia and prostate cancer.[21] Echinacea was contraindicated for use among patients with rheumatoid arthritis, systemic lupus erythematosus, leukosis, multiple sclerosis, tuberculosis and HIV infection.[16,18] Yohimbe was contraindicated in patients with anxiety, bipolar disorder, depression, mania and schizophrenia, as well as benign prostate hypertrophy and kidney disease.[21,22]

Figure 5.

Common contraindications for HDS use. *Other contraindications of gastrointestinal diseases included fecal impaction for aloe vera and oesophageal stricture for flaxseed. †Other contraindications of neurologic disorders included multiple sclerosis for echinacea and posttraumatic stress disorder for yohimbe. HDS, herbs and dietary supplements