? ABSTRACT Herbal medicines are mixtures of more than one active ingredient. The multitude of pharmacologically active compounds obviously increases the likelihood of interactions taking place. Hence, the likelihood of herb–drug interactions is theoretically higher than drug–drug interactions, if only because synthetic drugs usually contain single chemical entities. Case reports and clinical studies have highlighted the existence of a number of clinically important interactions, although cause-and-effect relationships have not always been established. Herbs and drugs may interact either pharmacokinetically or pharmacodynamically.
Through induction of cytochrome P450 enzymes and/or P-glycoprotein, some herbal products (e. g. St John’s wort) have been shown to lower the plasma concentration (and/or the pharmacological effect) of a number of conventional drugs, including cyclosporine, indinavir, irinotecan, nevirapine, oral contraceptives and digoxin. The majority of such interactions involves medicines that require regular monitoring of blood levels. To date there is less evidence relating to the pharmacodynamic interaction.
However, for many of the interactions discussed here, the understanding of the mechanisms involved is incomplete. Taking herbal agents may represent a potential risk to patients under conventional pharmacotherapy. O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 1 Table I Herbal medicines involved in drug interactions.
Data extracted from Capasso et al. [1] and Hennessy et al. [34]. Herbal medicine: common name/Latin name/source Main constituent(s) Main pharmacological action(s) Condition(s) frequently treated Betel nut/Areca catechu/seeds Arecoline Direct-acting cholinergic agonist None (used as a relaxing/refreshing drug).
Boldo/Peumus boldus/leaves Boldine Choleretic/cholagogue, diuretic Indigestion, constipation, hepatic ailments Chinese wolfberry/Lycium barbarum/fruits Polysaccharides, glycoproteins, vitamin C Immunostimulant, hypoglycaemic Loss of energy. Diabetes, liver and kidney disorders Cranberry/Vaccinium macrocarpon/fruits Fructose, anthocyanins, ?
Avonoids Antibacterial Urinary tract infections Curbicin a Fatty acids, phytosterols, polysaccharides Antiandrogenic, anti-in? ammatory Benign prostatic hyperplasia Danshen/Salvia milthiorriza/roots Tanshinones, phenolic compounds Vasorelaxant, anti-ischaemic, antiplatelet, radical scavenger ngina, myocardial infarction, ischaemic diseases Dong quai/Angelica sinsensis/roots Phytoestrogens, ? avonoids, coumarins Estrogenic effects, anti-in? ammatory, vasorelaxant Gynaecological disorders, circulation conditions Devils claw//
Harpagophytum procumbens/tubers Harpagoside Anti-in? ammatory, anti-arrhythmic, positive inotropic Musculoskeletal and arthritic pain Echinacea/Echinacea species/roots and aerial parts Polysaccharides, phenols, alkamides Immunostimulant Infection of the upper respiratory tract Evening primrose/Oenothera biennis/seeds Essential fatty acids Anti-in? ammatory Dermatological conditions, rheumatoid arthritis.
Fenugreek/Trigonella foenum-graecum/seeds Alkaloids, ? avonoids, saponins Antilipidaemic, hypoglycaemic, cholagogue Diabetes mellitus, hypercholesterolemia, Ginger/Zingiber of? cinale/rhizome Pungent principles (gingerols, zingerone) Antiemetic, antiplatelet, anti-in? ammatory antiulcer Prevention of nausea, dyspepsia Ginseng/Panax ginseng/roots Triterpene saponins (ginsenosides) Immunomodulatory, anti-in? ammatory, antitumor, hypoglycaemic Loss of energy and memory, stress states, male sexual dysfunction Garlic /Allium sativum/bulb Alliins Antihypertensive, antidiabetic, antiplatelet, antilipidaemic.
Hypercholesterolemia, prevention of arteriosclerosis Ginkgo/Ginkgo biloba/leaves Ginkgolides, ? avonoids Increase of microcirculatory blood ? ow, antiplatelet, free radical scavenging Circulatory disorders Green tea/Camellia sinensis/leaves Polyphenols, caffeine Antioxidant, antilipidaemic, antitumoral, CNS stimulant Prevention of cancer, cardiovascular diseases Guar gum/Cyamopsis tetragonolobus/seeds Galactomannan, lipids, saponins Antihyperglycaemic, antilipidaemic Diabetes, obesity, hypercholesterolemia Kava/Piper methysticum/rhizome Kavapyrones Anxiolytic, anaesthetics, muscle relaxant Anxiety, insomnia.
Khat/Catha edulis/leaves Catinone Central stimulant and indirectly sympathomimetic Loss of energy (khat-chewing is also as a social event) Liquorice/Glycyrrhiza glabra roots Glycyrrhizinic acid Expectorant, anti-in? ammatory, antiulcer, aldosterone-like effects Gastric ulcer, catarrhs, in? ammation Papaya/Carica papaya/fruits Papain (enzyme) Proteolytic, lipolytic activity Indigestion, obesity PC-SPES b Polysaccharides, phytosterols, fatty acids, ? avonoids Immunostimulant, cytotoxic Prostate cancer Psyllium/Plantago spp. /seeds Mucilages Modi? es gut viscosity Constipation, overweight Red yeast rice c.
Monacolins Hypocholesterolaemic Hypercholesterolaemia Siberian ginseng/Eleutherococcus senticosus/roots Eleutherosides Immunomodulatory, anti-in? ammatory, antitumor, Loss of energy and memory, stress states, male sexual dysfunction Soya/Glycine max/beans Phytoestrogens Hepatoprotective, anti-osteoporosis Treatment of menopausal symptoms, prevention of heart diseases and cancer St John’s wort /Hypericum perforatum/aerial parts Hypericin, hyperforin, ? avonoids Antidepressant, antiretroviral Mild to moderate depression a Curbicin contains Serenoa repens (saw palmetto) fruits, Cucurbita pepo (pumpkin) seeds and vitamin E.
b PC-PCS is a mixture of eight herbal drugs, namely Dendrathema morofolium (chrysanthemum), Isatis indigotica (dyer’s woad), Glycyrrhiza glabra (liquorice), Ganoderma lucidum (reishi), Panax pseudoginseng (san-qui ginseng), Rabdosia rubescens (rubescens), Serenoa repens (saw palmetto), Scutellaria bacicalensis (Baikal skullcap). c Red yeast rice is produced by fermentation of cooked rice using the fungus Monascus purpureus. 2A. A. Izzo O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 MECHANISMS OF HERB–DRUG INTERACTIONS Herbal medicines follow modern pharmacological prin- ciples.
Hence, herb–drug interactions are based on the same pharmacokinetic and pharmacodynamic mecha- nisms as drug–drug interactions [15]. Pharmacokinetic interactions have been more extensively studied and in vitro and in vivo studies indicated that the altered drug concentrations by co-administered herbs may be attrib- utable to the induction (or inhibition) of hepatic and intestinal drug-metabolizing enzymes [particularly cyto- chrome P450 (CYP)], and/or drug transporters such as P-glycoprotein [16,17]. The CYP is the most important phase I drug-metabo- lizing enzyme system, responsible for the metabolism of a variety of drugs.
Many herbs (e. g. St John’s wort, echinacea, kava and garlic) and natural compounds isolated from herbs (e. g. ?avonoids, coumarins, fur- anocoumarins, anthraquinones, caffeine and terpenes) have been identi? ed as substrates, inhibitors and/or inducers of various CYP enzymes [18].
Speci? cally, clinical studies have shown that long-term (2 weeks) St John’s wort administration signi? cantly induced intestinal and hepatic CYP3A4 and possibly other CYP enzymes involved in drugs metabolism [19–26]. More- over, a clinical study performed on 12 healthy subjects showed that echinacea modulated the catalytic activity of CYP3A at hepatic and intestinal sites (induction of hepatic CYP3A4 and inhibition of intestinal CYP3A4) [27].
By contrast, a number of herbal medicines, including green tea [28], ginkgo [29], garlic [30] saw palmetto [31] and Siberian ginseng [32] did not affect CYP3A4 and CYP2D6 activities in normal volunteers. P-glycoprotein in the intestine, liver and kidney may play an important role in the absorption, distribution, or excretion of drugs. P-glycoprotein appears to limit the cellular transport from intestinal lumen into epithelial cells and also enhances the excretion of drugs out of hepatocytes and renal tubules into the adjacent luminal space [33].
Like CYP, P-glycoprotein is vulnerable to inhibition, activation, or induction by herbs and herbal constituents. Curcumin, ginsenosides, piperine, sylima- rin and catechins may affect P-glycoprotein-mediated drug transport [17]. St John’s wort induces the intestinal expression of P-glycoprotein [34–36] both in isolated cells [35] and in healthy volunteers [34,36]. Hyperforin, a major ingredient of St John’s wort, binds to orphan pregnane X receptor [37,38] resulting in a series of intracellular events leading to the expression of CYP3A4 and P-glycoprotein [16].
A few pharmacodynamic inter- actions have also been described. Pharmacodynamic interactions may be additive (or synergetic), whereby the herbal medicine potentiates the action of synthetic drugs (e. g. interaction between the anticoagulant warfarin with antiplatelet herbs), or antagonistic, whereby the herbal medicine reduces the ef? cacy of synthetic drugs (e. g. kava possesses dopaminergic antagonistic properties and hence might reduce the pharmacological activity of the anti-parkinson drug levodopa) [15].
LIMITATIONS Much of the available information about the interaction between herbal products and prescribed drugs is gleaned from case reports, although clinical studies are now also beginning to appear in the literature. The published case reports are often incomplete as they do not allow us to conclude that a causal relationship exists. Even docu- mented case reports have to be interpreted with great caution, as causality is not usually established beyond reasonable doubt.
According to the scoring system described by Fugh-Berman and Ernst [7], 68. 5% of the cases reported were classi? ed as ‘unevaluable’ (i. e. reports contained inadequate information to assess the likelihood of an interaction), 18. 5% were classi? ed as ‘possible’ (i. e.reports provided some evidence for an interaction, but there may be other causes of the event) and 13% as ‘well documented’ (reports appeared to provide reliable evidence for an interaction).
INTERACTIONS WITH CARDIOVASCULAR PHARMACOTHERAPY Interaction between herbal remedy and conventional cardiovascular drugs is a potentially important safety issue, particularly for patients taking anticoagulants. The majority of reports concern drugs with a narrow therapeutic index such as warfarin and digoxin. These interactions have been systematically reviewed [10] and summarized here in Table II. Interactions with anticoagulant drugs Warfarin owes its action to its ability to antagonize the cofactor function of vitamin K.
Theoretically, increased anticoagulant effects could be expected when warfarin is combined with coumarin-containing herbs (e. g. boldo, fenugreek, don quai) or with antiplatelet herbs (e. g. danshen, garlic, ginkgo) [10]. Naturally occurring Herb–drug interactions 3 O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 Table II Clinical interactions between herbal medicines and conventional cardiovascular drugs. Drug Herb Result of interaction Possible mechanism Clinical comment Source of evidence Reference.
Digoxin Gum guar, St John’s wort, wheat bran Decreased plasma digoxin concentration Multiple mechanisms: (i) gum guar delays gastric emptying and hence may reduce digoxin absorption (ii) St John’s wort induces P-glycoprotein which is involved in digoxin absorption/excretion (iii) Fibres in bran may trap digoxin in the gut Digoxin has a narrow therapeutic index Clinical studies 66–69 Aspirin Ginkgo Spontaneous hyphema Additive effect on platelet aggregation (ginkgolides have antiplatelet activity) Hyphema is a clinical rare problem A case report 143 Lovastatin Pectin or oat bran Decreased absorption of lovastatin Pectins or bran ?
bres may bind or trap lovastatin in the gut The therapeutic manifestation of this interaction remains to be determined A clinical study 75 Simvastatin St John’s wort Decreased plasma digoxin concentration Simvastatin is a substrate of P-glycoprotein and is metabolized by CYP enzymes. Both CYP enzymes and P-glycoprotein are induced by St John’s wort The therapeutic manifestation of this interaction remains to be determined A clinical study 74 Phenprocoumon Ginger (i) Over-anticoagulation (i) Additive effect on coagulation (ginger inhibits platelet aggregation) The patient developed an elevated INR and epistaxis A case report 65.
St John’s wort (ii) decreased anticoagulant effect (ii) Phenprocoumon is metabolised by cytochrome enzymes which are induced by St John’s wort Phenprocoumon has a narrow therapeutic index A clinical study 64,67 Verapamil St John’s wort Decreased bioavailability of verapamil Induction of intestinal CYP3A4 by St John’s wort The clinical signi? cance remains to be determined.
A clinical study 26 Warfarin (i) Boldo/fenugreek, curbicin, cranberry danshen, devils claw, dong quai, garlic, ginkgo, ginseng, Chinese wolfberry, mango, papaya, quilinggao, PC-SPES (i) Over-anticoagulation (i) In general, additive effects on coagulation mechanisms. Such herbs may have anticoagulant (boldo, fenugreek, don quai, PC-SPES) or antiplatelet (i. e. danshen, garlic, ginseng, ginkgo) properties.
Other mechanisms include: cranberry ? avonoids may inhibit CYP enzymes responsible of warfarin metabolism; curbicin contain high amount of vitamin E which can antagonize vitamin K. Mechanism not known for devils claw, quilinggao and Chinese wolfberry (i) Risk of bleeding. Given the narrow therapeutic index of warfarin, vigilance is needed. Caution for ginkgo (a case reported intracerebral haemorrhage) and cranberry (fatal haemorrhage) (i).
Single case reports for boldo/fenugreek, devil’s claw, ginkgo, ginseng, Chinese wolfberry, papaya, PC-SPES. Multiple case for cranberry curbicin, danshen, dong quai, garlic 40–53,61 (ii) Ginseng a , green tea, soya, St John’s wort (ii) Decreased anticoagulant effect (ii) Green tea contain vitamin K and thus may antagonize the effect of warfarin; warfarin is metabolized by CYP enzymes which are induced by St John’s wort. Mechanism not known for ginseng and soya (ii).
Potential thrombotic complications (ii) Single case reports for green tea, ginseng and soya. Multiple cases for St John’s wort. A clinical study for ginseng and St John’s wort 54–58,60,62,63 4A.
A. Izzo O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 coumarins are only weak anticoagulant, but if the plant material is not stored properly dicoumarol may be formed by microbial transformation, and this compound is much more potent [1,39]. Conversely, vitamin K containing herbs (e. g. green tea) can antagonized the anticoagulant effect of warfarin. Case reports indicate over-anticoagulation [generally revealed by increased international normalized ratio (INR) values] when war- farin is combined to boldo/fenugreek [40], curbicin (a preparation containing saw palmetto, pumpkin and vitamin E, see Table I) [41], danshen [42–44], devil’s claw [45], dong quai [46,47], garlic [48], ginkgo [49].
Chinese wolfberry (Go-Qi-Zi) [50], papaya [45], mango [51], PC-SPES (a mixture of eight herbs) [52], quilinggao (a Chinese herbal combination) [53] and decreased anticoagulant effect when the drug is co-administered with cranberry [54,55], green tea [56], soya [57] and St John’s wort [58]. Analysis of case reports revealed reliable evidence for an interaction only for the boldo/ fenugreek, danshen, cranberry, dong quai and ginkgo interactions; some evidence for an interaction (but theremay be other causes of the event) was provided for curbicin, quilinggao, mango, Chinese wolfberry, green tea, PC-SPES and soya, while warfarin interactions caused by devil’s claw, papaya and garlic were published in reports containing inadequate information to assess the likelihood of an interaction [7].
PC-SPES has been found to contain adulterated indomethacin, warfarin and diethylstilbestrol [59]. In 2003, it has been suspen- ded from sale from the FDA [9]. A case of fatal interaction has been also reported [54]. Six weeks after starting cranberry juice a 70-year-old man under warfarin was admitted to hospital with an INR >50. Before, his control of INR had been stable.
He died of a gastrointestinal and pericardial haemorrhage. Moreover, the Committee on Safety of Medicines (UK) reported seven other reports about a possible interaction between warfarin and cranberry juice leading to changes in INR or bleeding [55]. Cranberry juice contains antioxidants, including ? avonoids, which are know to inhibit CYP enzymes responsible of warfarin metabolism [16,18]. The reported ginseng–warfarin interaction is some- what puzzling as case reports have shown that the herb may either decrease [60] or increase [61] the antico- agulant effect of warfarin.
Two recent randomized clinical trials have investigated the effect of ginseng on the pharmacokinetics and pharmacodynamics of war- farin in healthy subjects [62,63]. Jiang et al. showed a 7-day treatment with ginseng (Panax ginseng) did not affect the pharmacokinetics or pharmacodynamics (platelet aggregation and INR) of the anticoagulant in healthy subjects [62]; by contrast, Yuan et al. reported that the administration of American ginseng for 2 weeks reduced warfarin’s anticoagulant activity (decreased INR and warfarin AUC) [63].
The use of different species (Panax ginseng vs. Panax quinquefolius) and a different time of administration (1 week vs. 2 weeks) may explain such apparent discrepancy. Phenprocoumon is an anticoagulant chemically rela- ted to warfarin. A clinical study [64] highlighted the possible reduced ef? cacy of this anticoagulant if co- administered with St John’s wort. Another clinical trial showed that coadministration of St John’s wort increased the apparent clearance of warfarin, leading to a reduc- tion in the pharmacodynamic effect of the anticoagulant [62].
Consistently, seven cases of INR associated with concomitant use of warfarin and St John’s wort were reported by the Swedish Medical Product Agency [58].
Lastly, a 76-year-old woman on long-term phenprocou- mon therapy developed an elevated INR and epistaxis after taking the herb ginger. The INR returned to the normal range after ginger was stopped and vitamin K1 given [65]. An objective causality assessment revealed that the adverse drug event as a result of the phenpro- coumon and ginger interaction was probable.
Interactions with cardiac inotropic drugs There is clinical evidence that blood levels of digoxin can be reduced by the concurrent administration of some herbal products, although to date, no therapeutic interactions between herbs and digoxin have been reported. St John’s wort was shown to reduce digoxin through level after 10 days of co-medication in a single- blind, placebo-controlled study [66].
It was suggested that the underlying mechanism involves induction (by St John’s wort) of the P-glycoprotein drug transporter, facilitating the ef? ux of the drug from the enterocytes to the intestinal lumen [18]. The interaction between St John’s wort and digoxin seems to be correlated with the dose, particularly of the active ingredient hyperforin [67]. Other clinical trials showed that serum digoxin concentration may be reduced by concomitant admin-istration of gum guar (which may reduce digoxin absorption by delaying gastric emptying) [68] or whet bran (digoxin may be trapped by ? bres contained in wheat bran) [69].
However these interactions have likely no clinical relevant in? uence on therapeutic digoxin [10]. By contrast, two herbal medicines, namely haw- thorn and ginkgo, did not change the pharmacokinetics Herb–drug interactions 5 O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 of orally administered digoxin in healthy volunteers [70,71]. Increased levels of digoxin have been associated with ingestion of Siberian gingeng [72].
The patient was asymptomatic for digoxin toxicity despite a level of 5. 2 ng/mL. Siberian ginseng contains glycosides with structural similarities to digoxin which interfere with digoxin assay [73]. Interactions with antihyperlipidaemic drugs Simvastatin, pravastatin and lovastatin are inhibitors of hydroxymethylglutaryl (HMG)-CoA reductase, the rate limiting step in cholesterol synthesis. Two clinical studies reported in a single publication showed that St John’s wort decreased plasma concentrations of simvastatin but not of pravastatin [74].
The difference probably lies in the different metabolic pro?les of the two drugs. Simvastatin, in contrast to pravastatin, is metabolized through a CYP-dependent pathway, and it is also a substrate of P-glycoprotein. The therapeutic manifesta- tion of this interaction remains to be determined. A decrease of absorption of lovastatin [associated to increased low-density lipoprotein (LDL) levels] was observed in patients who took this drug concomitantly with pectin or oat bran [75].
Lovastatin pharmaco- kinetics and LDL returned normal after bran discontinu- ation. The interaction is likely due to the ability of pectins or bran ? bres to bind or trap concurrently administered lovastatin. Interactions with antihypertensive drugs Few (and poorly documented) cases of herb–drug inter- actions with antihypertensive drugs have been reported. Surprisingly, an elderly patient was found to have an increase in blood pressure after taking ginkgo (a periph- eral vasodilator) while receiving a thiazide diuretic (not speci? ed in the original paper) [45].
There is no rational pharmacological mechanism to explain this unusual interaction. In addition, hypokalemia, associated to ?accid quadriplegia, has been reported after the ingestion of small amounts of liquorice (which possess mineral-corticoid effects) in combination with antihypertensive treatment (not speci? ed in the original paper) [76]. Verapamil reduces arterial pressure by inhibiting calcium ion in? ux into the vascular smooth muscle cells, which results in a decrease in smooth muscle tone and vascular resistance.
Tannergreen et al. [26] reported that repeated administration of St John’s wort decreased verapamil bioavailability in healthy volunteers. This effect is likely caused by induction of ? rst-pass CYP3A4 metabolism in the gut, because the jejunal permeability and the terminal half-life of verapamil were unchanged.
In addition, verapamil is also a substrate of P-glyco- protein [77], which can be induced by St John’s wort [16,37,38]. INTERACTIONS WITH DRUGS AFFECTING THE CENTRAL NERVOUS SYSTEM The interactions between herbal remedies and conven- tional drugs affecting the central nervous system are summarized in Table III. Patients mixing synthetic and herbal anxiolytic or antidepressant drugs are at the highest risk. Interactions with antidepressant drugs The herb St John’s wort is effective in the treatment of mild to moderate depression [24].
Both St John’s wort and synthetic antidepressants have a high probability of concomitant use. St John’s wort and serotonin re-uptake inhibitors (i. e. sertraline, paroxetine, nefazodone and venlafaxine) may result in symptoms characteristic of central serotonin excess (e. g. mental status changes, tremor, autonomic instability, gastrointestinal upset, headache, myalgias and motor restlessness) as highligh- ted by case series and case reports [78–80].
These effects could be the result of an additive effect on 5-hydroxy- tryptamine (5-HT) because hyperforin in St John’s wort inhibit the re-uptake of several brain neurotransmitters including 5-HT [24]. Concomitant use of St John’s wort and sertraline has been reported to cause a manic episode in a 28-year-old man [81]. According to the report reliability scale for drug interaction [7], the case was classi? ed as possible, although it was complicated by concomitant testosterone replacement therapy following bilateral orchidectomy.
Finally, a clinical study showed that co-medication with St John’s wort decreased plasma and urine concentration of amitriptyline in 12 patients [82]. In addition to being a P-glycoprotein substrate, the demethylation and subsequent hydroxylation of ami- triptyline is catalysed by CYP2C19 and CYP3A4 which may be induced by St John’s wort. Other herbs which may interact with conventional antidepressants include ginkgo and ginseng. A 80-year- old woman with Alzheimer’s disease fell into a coma after taking a low dose of the atypical antidepressant trazodone with ginkgo [83]. The case was classi? ed as ‘possible’ [7].
Another report described a patient who experienced insomnia, headache, tremulousness and 6A. A. Izzo O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 Table III Clinical interactions between herbal medicines and conventional drugs affecting the central nervous system.
Drug Herb Result of interaction Possible mechanism Clinical comment Source of evidence Reference Alprazolam St John’s wort Decreased plasma levels of alprazolam Alprazolam is a speci? c probe for CYP 3A4, which is induced by St John’s wort The therapeutic manifestation of such interaction was not determined Clinical studies 23,25 Kava Semicomatose state Additive effect on GABA receptors.
Moreover, kava inhibits CYP3A4 in humans The patient was lethargic and disoriented for several hours A case report 93 Amitriptyline St John’s wort Decreased plasma levels of amitriptyline Amitriptyline is a substrate of both CYP2C19 and P-glycoprotein which are induced by St John’s wort The clinical signi? cance of such interaction was not determined A clinical study 82 Buspirone a St John’s wort Hypomania Synergistic effect on 5-HT receptors.
This is the ? rst case of hypomania following brain injury involving herb–drug interaction A case report 95 Caffeine Echinacea Reduction of caffeine oral clearance Caffeine is a substrate of CYP1A2 which is inhibited by echinacea This interaction is not clinically relevant A clinical study 27 Fluphenazine Evening primrose oil Seizures Gamalenic acid from evening primrose oil lowers the seizure threshold Phenothiazines are known to be epileptogenic themselves Two cases in a clinical study 90.
Levodopa Kava Reduced ef? cacy of levodopa Kava possesses dopaminergic antagonistic properties Increase in the duration and number of ‘off’ periods have been reported A case report 97 Lithium (i) Psyllium, hispagula (i) Decreased plasma lithium concentration (i) Hydrophilic psyllium may prevent lithium from ionizing The therapeutic index of lithium salts are extremely low A case report 86 (ii) Herbal diuretics b (ii) Decreased plasma lithium concentration (ii) Unknown A case report 87.
Midazolam Echinacea Increased (oral midazolam) or decreased (systemic midazolam) clearance Midazolam is a substrate of CYP3A4. Echinacea inhibits intestinal CYP3A4, while it induces hepatic CYP3A The therapeutic manifestation of such interaction was not determined A clinical study 27 St John’s wort Decreased plasma levels of midazolam Induction of CYP3A4 by St John’s wort The therapeutic manifestation of such interaction was not determined Clinical studies 20,22,36 Phenelzine Ginseng Sleeplessness, tremor and headaches Unknown The psychoactive effects were considered to be relevant Two cases 84.
Phenytoin Shankhapushpi Loss of seizure control Not know The interaction is potentially relevant Two cases 91 Procyclidine Betel nut Rigidity, bradykinesia, jaw tremors Antagonistic effect of arecoline from betel nut to the anticholinergic agent procyclidine Betel nut may reduce drug compliance in schizophrenic patients A case report 88 SRI c St John’s wort Serotonergic syndrome Synergistic effect on 5-HT uptake The syndrome could be fatal, particularly in elderly Case series and case reports 78–80 Trazodone Ginkgo Coma Not known The interaction is potentially relevant A case report 83 a.
A case of possible serotonergic syndrome has been also reported [96]. b The herbal formulation contained juniper, buchu, horsetail, corn silk, bearberry, parsley, bromelain and paprika. c Serotonin-reuptake inhibitors (i. e. sertraline paroxitine, nefazodone and venlafaxine). A case of mania in a patient taking St John’s wort and sertraline has been also reported [81]. Herb–drug interactions 7 O2004 Blackwell Publishing Fundamental & Clinical Pharmacology 19 (2004) 1–16 mania after co-administration of ginseng with phenel- zine [84]; here, causality is likely because inadvertent re-challenge resulted in similar symptoms [85].
The mechanisms of such interactions are not known. Interactions with neuroleptic drugs Lithium salts are used prophylactically in treating manic-depressive patients and in the treatment of manic episodes. Patients taking isphagula or psyllium were found to have lower blood levels of lithium, possibly because the plant product may trap lithium in the gut [86]. In addition, a 26-year-old woman stable on lithium for 5 months, experienced dizziness, grogginess and diarrhoea after taking a combination of herbal diuretics (juniper, buchu, horsetail, corn silk, bearberry, parsley, bromelain and paprika) [87].
The adverse events were associated to increased plasma lithium levels. As there were several herbal diuretics in the preparation used and the mechanism of action of each is elusive, it is impossible to determine which herb caused the lithium toxicity. Extrapyramidal symptoms occurring in a schizophre- nic patient who was maintained on depot neuroleptic medication following a period of heavy betel nut are described in a well-documented case report [88].
The underlying mechanism of this interaction is based on the pharmacological antagonism of the anticholinergic agent procyclidine (which is given to treat acute neurological adverse effects resulting from the use of neuroleptic drugs) by arecoline (an acetylcholine recep- tor agonist), the active ingredient of the betel nut [89]. Seizures were reported in two patients taking the phenothiazine ? uphenazine with evening primrose oil and in one patient taking placebo with evening primrose oil in a study of 23 patient with schizophrenia [90].
Evening primrose oil contains c-linolenic acid, which lowers the seizure threshold [34]. However, it should be noted that phenothiazines are known to be epileptogenic themselves. Interactions with antiepileptic drugs During the course of routine plasma drug level monit- oring an unexpected loss of seizure control and reduction in plasma phenytoin levels was noticed in two patients who were also taking the Ayurvedic multi-herb syrup Shankhapushpi [91].
Subsequent animal experiments con? rmed that the herbal remedy decreased the anti- epileptic activity of phenytoin without lowering its plasma level [91]. The mechanism of this interaction is not known. In addition, a clinical study showed that a 14-day treatment with St John’s wort did not alter carbamazepine pharmacokinetics in healthy volunteers [92].
Interactions with anxiolytic drugs The benzodiazepines alprazolam and midazolam are used experimentally as probe for CYP3A4 activity because they are entirely metabolized by intestinal and hepatic CYP 3A4. Consistently, clinical studies have shown that St John’s wort decreased alprazolam and midazolam plasma levels in healthy volunteers [20,22,23,25,36]. St John’s wort decreased plasma levels of midazolam the effect being considerably less after intravenous adminis- tration than after oral administration. These ? ndings indicate that enzymatic induction occurs both in the intestine and in the liver [36].
Moreover, a clinical study performed on 12 healthy subjects showed that echinacea increased the oral availability of midazolam (CYP3A probe), which is consistent with inhibition of intestinal CYP3A by the herb. By contrast the AUC of systemically given midazolam was reduced possibly due to induction of hepatic CYP3A [27]. The increase of oral availability of midazolam by echinacea mainly based on the inclu- sion of the reduced AUC of midazolam after intravenous dosing in the calculation of oral availability [27]. In addition, clinical studies showed the pharmacokinetics of alprazolam were not affected by.