Herb-Drug Interactions: St John’s wort

Contents

Hypericum perforatum L. (Clusiaceae)

Synonym(s) and related species

Hypericum, Millepertuis.

Hypericum noeanum Boiss., Hypericum veronense Schrank.

Pharmacopoeias

St John’s Wort (British Ph 2009, European Ph 2008, US Ph 32); St John’s Wort Dry Extract, Quantified (British Ph 2009, European Ph, 6th ed., 2008 and Supplements 6.1, 6.2, 6.3 and 6.4).

Constituents

The main groups of active constituents of St John’s wort are thought to be the anthraquinones, including hypericin, isohypericin, pseudohypericin, protohypericin, protopseudohypericin and cyclopseudohypericin, and the prenylated phloroglucinols, including hyperforin and adhyperforin. Flavonoids, which include kaempferol, quercetin, luteolin, hyperoside, isoquercitrin, quercitrin and rutin; biflavonoids, which include biapigenin and amentoflavone, and catechins are also present. Other polyphenolic constituents include caffeic and chlorogenic acids, and a volatile oil containing methyl-2-octane.

Most St John’s wort products are standardised at least for their hypericin content (British Pharmacopoeia 2009), even though hyperforin is known to be a more relevant therapeutic constituent, and some preparations are now standardised for both (The United States Ph 32). It is important to note that there will be some natural variation, and as both hypericin and hyperforin are sensitive to light, they are relatively unstable, so processes used during extraction and formulation, as well as storage conditions, can affect composition of the final product. Therefore different preparations of St John’s wort have different chemical profiles and they may not be equivalent in effect.

Use and indications

St John’s wort is widely used to treat mild-to-moderate depression, seasonal affective disorder, low mood, anxiety and insomnia, particularly if associated with menopause. It has also been used topically for its astringent properties.

Pharmacokinetics

St John’s wort has been implicated in numerous clinical interactions with conventional drugs and has therefore been extensively studied. Alongside the extensive clinical studies and case reports, there is also a plethora of in vitro and animal experimental data regarding its interactions and pharmacokinetics. This monograph will discuss the clinical evidence in preference to experimental data, where extensive literature is available and the clinical data are conclusive.

The main constituent found to be responsible for the activity of St John’s wort is hyperforin, but other constituents are considered to contribute to its antidepressant activity, such as hypericin and pseudohypericin, the flavonoid quercetin and its glycosides, and rutin. Bioavailability from varying formulations and extracts appears to be low, giving variable steady-state plasma concentrations. For information on the pharmacokinetics of individual flavonoids present in St John’s wort, see under flavonoids.

(a) Cytochrome P450 isoenzymes

St John’s wort is known to affect several cytochrome P450 isoenzymes and this accounts for the wide range of drugs with which St John’s wort has been reported to interact. It is thought to exert a biphasic effect on these isoenzymes, with inhibition occurring in in vitro studies with the initial exposure, and induction following long-term use. Therefore, predicting the overall effect from in vitro and animal experiments may not always be reliable.

The following is a list of cytochrome P450 isoenzymes that have been assessed with St John’s wort in a clinical setting:

• CYP3A4: the main clinically relevant effect of St John’s wort on cytochrome P450 is the induction of CYP3A4. This has been shown to be related to the constituent, hyperforin. Products vary in their hyperforin content; preparations with a high-hyperforin content, given for a long period of time, will induce CYP3A4 activity, and therefore decrease the levels of drugs metabolised by CYP3A4, by a greater extent than preparations containing low-hyperforin levels taken for a shorter period of time.

Conventional drugs are often used as probe substrates in order to establish the activity of another drug on specific isoenzyme systems. For CYP3A4 the preferred probe drug is midazolam, because it has no effects of its own on CYP3A4, and is metabolised almost exclusively by CYP3A4, with no known interference from other metabolic processes, such as transporter proteins. See St John’s wort + Benzodiazepines, for an example of the effects of St John’s wort on CYP3A4.

Studies have assessed the duration of the effects of St John’s wort on CYP3A4. One study found that CYP3A4 activity returned to baseline in about one week after St John’s wort was taken for 14 days. This may provide an indication of how long to leave between using St John’s wort and starting another drug, and therefore avoiding clinically important interactions. However, another study found that the effects of St John’s wort lasted for more than 2 weeks in some patients. See the table Drugs and herbs affecting or metabolised by the cytochrome P450 isoenzyme CYP3A4 for a list of known CYP3A4 substrates.

• CYP2C19: there are some clinical reports suggesting that St John’s wort induces CYP2C19. See St John’s wort + Proton pump inhibitors, for a clinically relevant example of this.

• CYP2C8: St John’s wort may induce CYP2C8, see repaglinide, under St John’s wort + Antidiabetics for a clinically relevant example of this.

• CYP2C9: St John’s wort may induce CYP2C9 (see St John’s wort + Warfarin and related drugs), but the mechanism for these interactions is not conclusive because not all CYP2C9 substrates have been found to interact (see tolbutamide, under St John’s wort + Antidiabetics).

• CYP2E1: St John’s wort may induce CYP2E1 but the general clinical importance of this is unclear, see St John’s wort + Chlorzoxazone.

• CYP1A2: St John’s wort is also thought to be an inducer of CYP1A2 as levels of caffeine, and theophylline, both of which are CYP1A2 substrates, have been reduced by St John’s wort. However, the general clinical importance of this is unclear as other studies have found no clinically significant effect on these drugs. This may be because St John’s wort only has a minor inducing effect on CYP1A2, which may depend on the level of exposure to hyperforin.

• CYP2D6: St John’s wort does not appear to affect the activity of CYP2D6 to a clinically relevant extent, see St John’s wort + Dextromethorphan, and St John’s wort + Tricyclic antidepressants.

(b) P-glycoprotein

St John’s wort is known to affect P-glycoprotein activity, especially intestinal P-glycoprotein, and it is generally thought that inhibition takes place initially, and briefly, but is followed by a more potent and longer-acting induction. It is the induction that leads to the clinically relevant drug interactions of St John’s wort that occur as a result of this mechanism. Hyperforin is implicated as the main constituent responsible for the effect, see St John’s wort + Digoxin.

(c) Serotonin syndrome

St John’s wort inhibits the reuptake of 5-hydroxytryptamine (5-HT, serotonin) and this has resulted in a pharmacodynamic interaction, namely the development of serotonin syndrome (see under Pharmacodynamics) with conventional drugs that also have serotonergic properties. These include bupropion, SNRIs, SSRIs, and triptans.

Interactions overview

St John’s wort is known to interact with many conventional drugs because of its ability to induce the activity of CYP3A4 and P-glycoprotein, which are involved in the metabolism and distribution of the majority of drugs. CYP2C19, CYP2C8 and CYP2E1 may also be induced by St John’s wort, although the evidence is not conclusive and further study is needed. In general, CYP2C9 and CYP1A2 do not appear to be significantly affected by St John’s wort; however, isolated reports of an interaction have still occurred. Hyperforin is the active constituent believed to be central to the inducing effects of St John’s wort. As St John’s wort preparations and dose regimens are varied, the amount of hyperforin exposure will also vary a great deal, which makes predicting whether an interaction will occur, and to what extent, difficult. For more information concerning the pharmacokinetic and pharmacodynamic properties of St John’s wort that are relevant to drug interactions, see under Pharmacokinetics, and for detail on the interactions of St John’s wort, see the sections that follow page 361. For information on the interactions of individual flavonoids present in St John’s wort, see under flavonoids.

St John’s wort + 5-Aminolevulinic acid

An isolated case report describes a severe phototoxic reaction attributed to a synergistic effect of 5-aminolevulinic acid and St John’s wort.

Clinical evidence

A 47-year-old woman who was taking St John’s wort (Hyperiforce, dose not stated), experienced a phototoxic reaction on skin areas exposed to light 6 hours after receiving 5-aminolevulinic acid 40mg/kg. She developed a burning erythematous rash and severe swelling of the face, neck and hands. Treatment with oral corticosteroids resulted in complete resolution after skin desquamation.

Experimental evidence

An in vitro study using human cell lines found that the combination of 5-aminolevulinic acid and an extract of St John’s wort (Hyperiforce) increased light-induced toxicity by up to 15%.!

Mechanism

It was suggested that there was a synergistic photosensitivity reaction between the two drugs.

Importance and management

This appears to be the only report of such an effect, but bear it in mind in the event of an unexpected adverse reaction to 5-aminolevulinic acid.

St John’s wort + Anaesthetics, general

It has been predicted that St John’s wort may prolong the effects of anaesthetics, which is supported by an isolated case. A case of profound hypotension during anaesthesia following the long-term use of St John’s wort has also been reported. The American Society of Anesthesiologists recommends that all herbal medicines should be stopped two weeks prior to elective surgery.

Clinical evidence

Prolonged anaesthesia has been reported in a 21-year-old woman who had been taking St John’s wort lg threetimes daily for 3 months before general anaesthetics were given for the surgical removal of an abscess. Anaesthesia was induced by intravenous fentanyl citrate 1 microgram/kg and propofol 3 mg/kg, and maintained throughout the procedure by sevoflurane and nitrous oxide using a facemask.

Another case report describes a healthy 23-year-old woman, who had been taking St John’s wort on a daily basis for 6 months, who developed severe hypotension (BP 60/20 mmHg) during general anaesthesia, which responded poorly to ephedrine and phenylephrine (BP increased to 70/40 mrnHg);

Experimental evidence

No relevant data found.

Mechanism

It has been suggested that St John’s wort may prolong anaesthesia, but there are no reports of this occurring. This appears to have been based on the possibility that St John’s wort acts as an MAOPA(although this has been disputed), and the limited evidence that MAOIs may cause hepatic enzyme inhibition and potentiate the effects of barbiturates.’.

However, there is now increasing evidence that St John’s wort induces hepatic enzymes, and might therefore increase the metabolism of barbiturates, which suggests that it could increase requirements for thiopental anaesthesia. The possible MAOI activity of St John’s wort has led to the recommendation that the same considerations apply as for other MAOIs and general anaesthetics.The authors of the second case report suggest that St John’s wort might have caused adrenergic desensitisation with decreased responsiveness to the vasopressors.

Importance and management

Not established. The evidence presented suggests that some caution may be warranted in patients using St John’s wort if they are given general anaesthetics. Because of the limited information, the American Society of Anesthesiologists have recommended discontinuation of all herbal medicines 2 weeks before an elective anaesthetic and, if there is any doubt about the safety of a product, this may be a prudent precaution.

St John’s wort + Antidiabetics

St John’s wort modestly decreases the AUC of gliclazide and rosiglitazone. Pioglitazone and repaglinide are similarly metabolised and may therefore be expected to interact similarly. St John’s wort does not affect the metabolism of tolbutamide.

Clinical evidence

(a) Gliclazide

In a study in 21 healthy subjects, a 300-mg dose of a St John’s wort preparation with a high hyperforin content (LI 160, Lichtwer Pharma) was given 3 times daily for 15 days. On the last day of treatment, a single 80-mg dose of gliclazide was given, followed 30 minutes later by glucose 75 g. St John’s wort reduced the maximum levels and AUC of gliclazide by 22% and 35%, respectively. The clearance was increased by 47%. No statistically significant changes were found in the AUC0-4 or blood levels of glucose or insulin.

(b) Rosiglitazone

A preliminary report of a pharmacokinetic study states that St John’s wort 900 mg daily decreased the AUC of a single dose of rosiglitazone by 26% and increased its clearance by 35%.

(c) Tolbutamide

In a study using tolbutamide as a probe drug for CYP2C9 activity, St John’s wort 900 mg had no effect on the metabolism of a single dose of tolbutamide either after one day or after 2 weeks of use. The St John’s wort product used was from Sundown Herbals and provided about 33 mg of hyperforin daily. Similarly, in another study, a St John’s wort preparation with low hyperforin content (Esbericum) at a dose of 240 mg daily (which provided about 3.5 mg of hyperforin daily) had no effect on tolbutamide metabolism.

Experimental evidence

No relevant data found.

Mechanism

Gliclazide is a substrate of the cytochrome P450 isoenzyme CYP2C9 and the authors suggest that St John’s wort induces this isoenzyme, thereby increasing the metabolism of gliclazide and reducing its levels. The magnitude of this effect was not influenced by CYP2C9 genotype. The fact that tolbutamide, another CYP2C9 substrate, was unaffected by St John’s wort suggests that other factors may be involved.

Rosiglitazone is known to be metabolised principally by the cytochrome P450 isoenzyme CYP2C8, and it was therefore concluded that St John’s wort induces this isoenzyme. The magnitude of the effect of St John’s wort does not appear to be influenced by CYP2C8 genotype.

Importance and management

The clinical relevance of the modest reduction in rosiglitazone levels has not been assessed, but it would seem unlikely to be important. However, the authors state that concurrent use of St John’s wort should be monitored when patients are also given CYP2C8 substrates. Pioglitazone and repaglinide are also substrates of CYP2C8, and would therefore be expected to be similarly affected, although possibly not to a clinically relevant extent. A large decrease in pioglitazone levels would not be expected on the basis that rifampicin, a more potent enzyme inducer than St John’s wort, only caused a 54% reduction in the AUC of pioglitazone. However, the UK manufacturer recommends caution when prescribing pioglitazone with drugs that induce CYP2C8. All three drugs are also substrates for CYP3A4, of which St John’s wort is an established inducer. Further study is needed.

The small reduction in the levels of gliclazide do not appear to be clinically important as its blood-glucose-lowering effects were unaffected. Whether or not St John’s wort induces CYP2C9 to a clinically relevant extent remains inconclusive.

No special precautions appear to be necessary if tolbutamide and St John’s wort are used together.

Note that, regardless of whether an interaction occurs, the dose of antidiabetics should be titrated to achieve adequate diabetic control.

St John’s wort + Antiepileptics

St John’s wort modestly increased the clearance of single-dose carbamazepine, but had no effect on multiple-dose carbamazepine pharmacokinetics. Carbamazepine does not appear to significantly affect the pharmacokinetics of hypericin or pseudohypericin (constituents of St John’s wort). St John’s wort increased the clearance of mephenytoin by about 3-fold and is predicted to reduce the blood levels of phenytoin and phenobarbital, but this awaits clinical confirmation.

Clinical evidence

In a multiple-dose study in 8 healthy subjects, St John’s wort had no effect on the pharmacokinetics of carbamazepine or its metabolite (carbamazepine-10,ll-epoxide). In this study, subjects took carbamazepine 200 mg increased to 400 mg daily alone for 20 days, then with St John’s wort 300 mg (standardised to hypericin 0.3%) three times daily for a further 14 days. In contrast, the AUC of a single 400-mg dose of carbamazepine was reduced by 21% after St John’s wort 300 mg was given three times daily for 14 days, and the AUC of the 10,11-epoxide metabolite was increased by 26%.

A double-blind, placebo-controlled study in healthy subjects found that, apart from a modest 29% decrease in the AUC of pseudohypericin, carbamazepine did not significantly affect the pharmacokinetics of either hypericin or pseudohypericin, which are both constituents of St John’s wort.

In another placebo-controlled study in 6 extensive metabolisers of CYP2C19, St John’s wort 300 mg threetimes daily for 14 days increased the clearance of a single oral dose of mephenytoin 100 mg given on day 15, by about 3-fold. There were no significant effects when mephenytoin was given to 6 poor metabolisers of CYP2C19. Each St John’s wort tablet contained 0.3% hypericin and 4% hyperforin.

Experimental evidence

Because of the quality of the clinical evidence (controlled pharmacokinetic studies), experimental data have not been sought.

Mechanism

St John’s wort is a known inducer of CYP3A4, and the results with single-dose carbamazepine are as predicted. However, carbamazepine is also an inducer of CYP3A4, and induces its own metabolism (autoinduction). It is suggested that St John’s wort is not sufficiently potent an inducer to further induce carbamazepine metabolism when autoinduction has occurred, and therefore a small interaction is seen with single doses but no interaction is seen with multiple doses. However, the lack of effect seen in some of these studies may also be due to the different preparations used, and therefore differing levels of hyperforin.

Mephenytoin is a substrate of CYP2C19 and St John’s wort appears to induce this isoenzyme.

Importance and management

The available evidence suggests that a clinically significant interaction between carbamazepine and St John’s wort is unlikely. Before the publication of the above reports, the CSM in the UK had advised that patients taking a number of drugs including the antiepileptics carbamazepine, phenytoin and phenobarbital should not take St John’s wort. This advice was based on predicted pharmacokinetic interactions. In the light of the above studies, this advice may no longer apply to carbamazepine, although further study is needed. Until more is known, it would probably be prudent to avoid concurrent use in patients taking mephenytoin, phenytoin and phenobarbital (and therefore primidone), especially as phenytoin is also a substrate of CYP2C19, which St John’s wort also appears to induce. The situation with carbamazepine is less clear. As the pharmacokinetic effects reported were modest, it may not be necessary to avoid St John’s wort; however, concurrent use should probably still be monitored to ensure adequate carbamazepine levels and efficacy.

Note that St John’s wort does not appear to interfere with laboratory assays for carbamazepine, phenytoin, phenobarbital or valproate. See St John’s wort + Laboratory tests.

St John’s wort + Benzodiazepines

Long-term use of St John’s wort decreases the plasma levels of alprazolam, midazolam and quazepam. St John’s wort preparations taken as a single dose, or containing low-hyperforin levels, appear to have less of an effect.

Clinical evidence

(a) Alprazolam

In a study in 12 healthy subjects, St John’s wort (LI 160, Lichtwer Pharma, 0.12 to 0.3% hypericin) 300 mg three times daily for 16 days with a single 2-mg dose of alprazolam on day 14. The AUC of alprazolam was halved by St John’s wort and the clearance was increased by about twofold.

In another study, alprazolam 1 or 2mg was given to 7 healthy subjects on the third day of a 3-day treatment period with St John’s wort (Solaray; hypericin content standardised at 0.3%) 300 mg three times daily. The pharmacokinetics of alprazolam were unchanged by St John’s wort, but the authors note that 3 days may have been an insufficient time for St John’s wort to fully induce cytochrome P450 isoenzymes. In another study, 16 healthy subjects were given St John’s wort extract 120 mg (Esbericum capsules; corresponding to 0.5 mg total hypericins and 1.76 mg hyperforin) twice daily for 10 days. A single 1-mg dose of alprazolam was given on the day before treatment with St John’s wort and on the last day of treatment. St John’s wort extract at this low dosage and low hyperforin content had no clinically relevant effects on the pharmacokinetics of alprazolam, when compared with 12 subjects given placebo.

(b) Midazolam

An open-label study in 12 healthy subjects found that a single 900-mg dose of St John’s wort had no significant effect on the pharmacokinetics of single doses of either oral midazolam 5 mg or intravenous midazolam 0.05 mg/kg, although there was a trend for increased oral clearance. However, St John’s wort 300 mg three times daily for 14 or 15 days decreased the AUC and maximum plasma concentration of oral midazolam by about 50% and 40%, respectively. Intravenous midazolam was not significantly affected. Similar results were found in another six studies. In one of the studies, although no serious adverse events occurred, 3 subjects reported that the sedative effects of midazolam were less noticeable when St John’s wort was taken at the same time.

(c) Quazepam

In a placebo-controlled study, 13 healthy subjects were given St John’s wort (TruNature; hypericin content standardised at 0.3%) 300 mg three times daily for 14 days with a single 15-mg dose of quazepam on day 14. St John’s wort modestly decreased the AUC and maximum plasma levels of quazepam by 26% and 29%, respectively, but the pharmacodynamic effects of quazepam were not affected.

Experimental evidence

No relevant data found.

Mechanism

Alprazolam, midazolam and quazepam are substrates of the cytochrome P450 isoenzyme CYP3A4. St John’s wort appears to induce CYP3A4, thus increasing the metabolism of oral midazolam, alprazolam and quazepam, and reducing the bioavail-ability of these benzodiazepines.

Hyperforin appears to be the main active constituent that induces CYP3A4, because high-hyperforin extracts have more of an inducing effect than low-hyperforin extracts.

Importance and management

Although not all the studies found an interaction between St John’s wort and alprazolam or midazolam, those that did found a reduction in levels, which is in line with the known CYP3A4-inducing effects of St John’s wort. The variable findings reported in the studies (some found no interaction) could be due to the preparation of St John’s wort used and the duration of treatment. Until more is known about the interacting constituents of St John’s wort, and the amount necessary to provoke an interaction, it would seem prudent to monitor patients receiving alprazolam and oral midazolam concurrently for any signs of reduced efficacy. Single doses of intravenous midazolam do not appear to be significantly affected. Note that triazolam is also a substrate of CYP3A4 and is likely to be affected in the same way as alprazolam and midazolam.

The modest reduction in quazepam levels did not reduce its efficacy; however, it may be prudent to bear the potential for an interaction in mind should a patient taking St John’s wort have a reduced response to quazepam.

Those benzodiazepines that undergo glucuronidation, such as lorazepam, oxazepam and temazepam, would not be expected to be affected by St John’s wort, and may be useful alternatives.

St John’s wort + Bupropion

Two cases describe symptoms indicative of serotonin syndrome when bupropion was taken with long-term St John’s wort.

Clinical evidence

A 58-year-old woman who had been taking St John’s wort 300 mg daily for several years and receiving HRT (estradiol and medroxy-progesterone) developed acute facial dystonia affecting the right side of her face, neck and right arm when she started taking bupropion 150 mg daily for 4 days. The episodic spasms were completely resolved after 5 months of treatment with oral chorpheniramine, procyclidine, diazepam and carbamazepine.

A brief report describes the development of mania in one patient, which was associated with the concurrent use of St John’s wort and bupropion.

Experimental evidence

No relevant data found.

Mechanism

A pharmacodynamic interaction may occur between St John’s wort and bupropion because they can both inhibit the reuptake of 5-hydroxytryptamine (serotonin). Serotonin syndrome has been seen with St John’s wort alone, and so additive serotonergic effects appear to be the explanation for what occurred in the cases described here.

Importance and management

Information appears to be limited to these two reports, one of which is lacking detail. Nevertheless because of the potential severity of the reactions it would seem prudent to monitor concurrent use closely for an increased incidence of adverse reactions.

St John’s wort + Buspirone

Two patients taking buspirone developed marked CNS effects after starting to take herbal medicines including St John’s wort.

Clinical evidence

A 27-year-old woman who had been taking buspirone 30 mg daily for over one month started to take St John’s wort (Hypericum 2000 Plus, Herb Valley, Australia) three tablets daily. After 2 months she complained of nervousness, aggression, hyperactivity, insomnia, confusion and disorientation, which was attributed to serotonin syndrome. The St John’s wort was stopped, the buspirone was increased to 50 mg daily and her symptoms resolved over a week. A 42-year-old woman who was taking fluoxetine 20 mg twice daily and buspirone 15mg twice daily started to develop symptoms of anxiety, with episodes of over-sleeping and memory deficits. It was discovered that she had been self-medicating with St John’s wort, ginkgo biloba and melatonin. She was asked to stop the non-prescribed medication and her symptoms resolved.

Experimental evidence

No relevant data found.

Mechanism

The exact mechanism of these interactions is not clear, but it seems most likely they were due to the additive effects of the buspirone and the herbal medicines, either through their effects on elevating mood or through excess effects on serotonin. Fluoxetine may have had a part to play in one of the cases. See St John’s wort + SSRIs.

Importance and management

The clinical significance of these cases is unclear, but they highlight the importance of considering adverse effects from herbal medicines when they are used with conventional medicines.

St John’s wort + Caffeine

Two studies suggest that St John’s wort increases the metabolism of caffeine. However, four other studies using preparations of varying hyperforin content suggest that the metabolism of caffeine is not affected by St John’s wort.

Clinical evidence

A study in 16 healthy subjects given a single 200-mg dose of caffeine before and after St John’s wort 300 mg (containing 900 micrograms of hypericin) three times daily for 14 days found no overall change in the pharmacokinetics of caffeine. However, when the subset of 8 female patients was considered, it was found that there was an induction of CYP1A2 in this group of patients resulting in an increase in the production of caffeine metabolites.

In another study, St John’s wort 300 mg given to 12 healthy subjects three times daily for 28 days, modestly increased the metabolism of caffeine 100 mg (a CYP1A2 probe substrate) to paraxanthine by about 26%, although no serious adverse events occurred. The St John’s wort preparation used was standardised to 0.3% hypericin and provided each subject with about 12.2 mg of hyperforin daily. However, a later study using the same criteria in 12 elderly healthy subjects found that St John’s wort 300 mg three times daily for 28 days (standardised to hypericin 0.3%) generally had no statistically significant effect on the metabolism of a single 100-mg dose of caffeine to paraxanthine taken on day 28, although some individuals showed moderate changes.

Similarly, another study in 28 healthy subjects found no significant change in caffeine pharmacokinetics when a low-hyperforin (about 3.5 mg daily) St John’s wort extract (Esbericum) 120 mg was given twice daily for 11 days to patients who had received a single caffeine dose of 100 mg before St John’s wort was started, and on the last day of the study. These findings were also reported in two other studies using caffeine as a probe drug for CYP1A2 activity and a St John’s wort regimen that provided a high-hyperforin dose. One study gave hyperforin 33 mg and hypericin 2.5 mg daily, and the other gave a minimum of hyperforin 36 mg and hypericin 2.7 mg daily.

Experimental evidence

No relevant data found.

Mechanism

These studies investigated whether St John’s wort had any effect on the cytochrome P450 isoenzyme CYP1A2 by which caffeine is metabolised.

Importance and management

The extent of any interaction with caffeine may depend on the St John’s wort preparation involved and dose used and may be correlated with the dose of hyperforin. However, these studies generally suggest that an interaction between St John’s wort and caffeine is unlikely to be clinically important. Caffeine is often used as a probe drug for CYP1A2 activity, and therefore these results also suggest that a significant pharmacokinetic interaction as a result of this mechanism between St John’s wort and other CYP1A2 substrates is unlikely. Note that St John’s wort has only minimal effects on the metabolism of theophylline, which is also a CYP1A2 substrate, consider also St John’s wort + Theophylline.

St John’s wort + Calcium-channel blockers

St John’s wort significantly reduces the bioavailability of nifedipine and verapamil. Other calcium-channel blockers would be expected to interact similarly.

Clinical evidence

(a) Nifedipine

In a study in 10 healthy subjects, St John’s wort 900 mg daily for 14 days decreased the maximum levels and AUC of a single oral dose of nifedipine 10 mg by about 38% and 45%, respectively. The maximum levels and AUC of the active metabolite of nifedipine, dehydronifedipine, were raised by about 45% and 26%, respectively. The St John’s wort preparation used was standardised to contain hypericin 0.3% and hyperforin 5%.

(b) Verapamil

In a study in 8 healthy subjects, verapamil 24 mg was given as a jejunal perfusion over 100 minutes both before and after treatment with St John’s wort tablets (Movina; containing 3 to 6% hyperforin) 300 mg three times daily for 14 days. St John’s wort did not affect jejunal permeability or the absorption of either R- or S-verapamil. The AUCs of R- and S-verapamil were decreased by 78% and 80%, respectively, and the peak plasma levels were decreased by 76% and 78%, respectively. The terminal half-life was not changed significantly. The AUC for i?-verapamil was sixfold higher than that of S-verapamil and St John’s wort did not change this ratio.

Experimental evidence

No relevant data found.

Mechanism

It appears that St John’s wort decreased the bioavailability of both nifedipine and verapamil by inducing their metabolism by the cytochrome P450 isoenzyme CYP3A4 in the gut. An effect on P-glycoprotein-mediated transport is not likely, as intestinal permeability was not significantly altered.

Importance and management

The general importance of this interaction is unclear, as neither study reported on the clinical outcome of these reductions in calcium-channel blocker levels. Patients taking St John’s wort with nifedipine or verapamil should have their blood pressure and heart rate monitored to ensure that they are still effective, and the dose should be adjusted if needed. There appears to be no information about other calcium-channel blockers, but as they are all metabolised by CYP3A4, to a greater or lesser extent, it would seem prudent to monitor concurrent use carefully

St John’s wort + Chlorzoxazone

St John’s wort increases the clearance of chlorzoxazone.

Clinical evidence

In a study in 12 healthy subjects, St John’s wort 300 mg three times daily for 28 days more than doubled the clearance of chlorzoxazone 500 mg (a CYP2E1 probe substrate). The St John’s wort preparation used was standardised to a concentration of hypericin 0.3% and provided each subject with about 12.2 mg of hyperforin daily. A later study using the same criteria in 12 healthy subjects (aged between 60 and 76 years) found that St John’s wort 300 mg three times daily for 28 days increased the metabolism of a single 500-mg dose of chlorzoxazone to hydroxychorzoxazone by only 26%. The St John’s wort preparation used in this study provided a daily dose of 4.8 mg of hyperforin.

Experimental evidence

No relevant data found.

Mechanism

It appears that St John’s wort increases the clearance of chlorzoxazone by inducing its metabolism by the cytochrome P450 isoenzyme CYP2E1.

Importance and management

The clinical relevance of this interaction is unclear but, as no adverse events were reported in the studies, it is unlikely to be of general importance. The authors of the second study suggested that the lower induction of CYP2E1 may have occurred because of inter-individual variability or because there is an age-related reduction in CYP2E1 activity. However, the dose of hyperforin was only about one-third of the dose used in the first study, and this may also account for the difference in degree of induction seen. Nevertheless, because only a single dose of chlorzoxazone was used in these studies, if a patient is taking St John’s wort, it may be prudent to monitor for any signs of reduced chlorzoxazone efficacy. As chlorzoxazone is used as a probe drug for CYP2E1 activity, be aware that St John’s wort may induce the metabolism of other CYP2E1 substrates.

St John’s wort + Ciclosporin

Marked reductions in ciclosporin blood levels and transplant rejection can occur within a few weeks of starting St John’s wort

Clinical evidence

A marked drop in ciclosporin blood levels was identified in one kidney transplant recipient as being due to the addition of St John’s wort extract 300 mg three times daily. When the St John’s wort was stopped the ciclosporin levels rose. The authors of this report identified another 35 kidney and 10 liver transplant recipients whose ciclosporin levels had dropped by an average of 49% (range 30 to 64%) after starting St John’s wort. Two of them had rejection episodes. In addition, subtherapeutic ciclosporin levels in 7 kidney transplant recipients, one liver transplant recipient, and 6 heart transplant recipients have been attributed to self-medication with St John’s wort. Acute graft rejection episodes occurred in 7 cases, and one recipient subsequently developed chronic rejection, requiring a return to dialysis. Another case of subtherapeutic ciclosporin levels occurred in a kidney transplant recipient during the concurrent use of a herbal tea containing St John’s wort. The recipient’s levels remained subtherapeutic despite a ciclosporin dose increase from 150 to 250 mg daily. The levels recovered within 5 days of stopping the herbal tea and the ciclosporin dose was reduced to 175 mg daily.

These case reports are supported by a small study in which 11 renal transplant recipients, with stable dose requirements for ciclosporin, were given St John’s wort extract (Jarsin 300) 600 mg daily for 14 days. Pharmacokinetic changes were noted 3 days after the St John’s wort was added. By day 10 the ciclosporin dose had to be increased from an average of 2.7 to 4.2mg/kg daily in an attempt to keep ciclosporin levels within the therapeutic range. Two weeks after the St John’s wort was stopped, only 3 patients had been successfully re-stabilised on their baseline ciclosporin dose. Additionally, the pharmacokinetics of various ciclosporin metabolites were substantially altered.

Another study in 10 kidney transplant recipients stable taking ciclosporin found that the content of hyperforin in the St John’s wort affected the extent of the interaction with ciclosporin. In patients taking St John’s wort with a high hyperforin content (hyperforin 7mg, hypericin 0.45 mg) the reduction in the AUC0-12 of ciclosporin was 45% greater than that in patients taking St John’s wort with a low hyperforin content (hyperforin 0.1 mg, hypericin 0.45 mg). The maximum blood ciclosporin level and the trough ciclosporin level were also reduced by 36% and 45%, respectively, in the patients taking the higher hyperforin-containing St John’s wort preparation, when compared with the patients taking the preparation with a lower hyperforin content. The patients taking the high-hyperforin preparation required a mean ciclosporin dose increase of 65% whereas the patients taking the low-hyperforin preparation did not require any ciclosporin dose alterations.

Experimental evidence

Because of the extensive clinical evidence available, experimental data have not been sought.

Mechanism

St John’s wort is a known inducer of the cytochrome P450 isoenzyme CYP3A4 by which ciclosporin is metabolised. Concurrent use therefore reduces ciclosporin levels. It has also been suggested that St John’s wort affects ciclosporin reabsorption by inducing the drug transporter protein, P-glycoprotein, in the intestine.

Importance and management

An established and clinically important interaction. The incidence is not known, but all patients taking ciclosporin should avoid St John’s wort because of the potential severity of this interaction. Transplant rejection can develop within 3 to 4 weeks. It is possible to accommodate this interaction by increasing the ciclosporin dosage(possibly about doubled) but this raises the costs of an already expensive drug. Also, the varying content of natural products would make this hard to monitor. The advice of the CSM in the UK is that patients receiving ciclosporin should avoid or stop taking St John’s wort. In the latter situation, the ciclosporin blood levels should be well monitored and the dosage adjusted as necessary. The study described above suggests that increased monitoring will be needed for at least 2 weeks after the St John’s wort is stopped.

Note that St John’s wort does not appear to interfere with laboratory assays for ciclosporin, see St John’s wort + Laboratory tests.

St John’s wort + Cimetidine

Cimetidine does not significantly alter the metabolism of the constituents of St John’s wort, hypericin and pseudohypericin.

Clinical evidence

A placebo-controlled study in healthy subjects taking St John’s wort (LI160, Lichtwer Pharma) 300 mg three times daily found that, apart from a modest 25% increase in the AUC of pseudohypericin, cimetidine 1 g daily (in divided doses) did not significantly affect the pharmacokinetics of either the hypericin or the pseudohypericin constituents of St John’s wort.

Experimental evidence

No relevant data found.

Mechanism

Cimetidine is an inhibitor of the cytochrome P450 isoenzymes CYP3A4, CYP1A2 and CYP2D6. This study suggests that St John’s wort is not significantly metabolised by these isoenzymes.

Importance and management

The available evidence suggests that cimetidine is unlikely to affect the dose requirements of St John’s wort.

St John’s wort + Dextromethorphan

St John’s wort does not affect the pharmacokinetics of dextromethorphan or debrisoquine.

Clinical evidence

In a study in 12 healthy subjects, St John’s wort (LI160, Lichtwer Pharma, 0.12 to 0.3% hypericin) 300 mg three times daily was taken for 16 days with a single 30-mg dose of dextromethorphan on day 14. There was no consistent change in the urinary dextromethorphan to dextrorphan metabolic ratio: 6 subjects had an increase in the production of dextrorphan while the other 6 subjects had a reduction in dextrorphan production. This finding was within the normal inter-patient variation in dextromethorphan metabolism.Similar findings were reported in another study in 16 healthy subjects given a single 25-mg dose of dextromethorphan on the last day of a 14-day course of St John’s wort (Jarsin; 900 micrograms of hypericin) 300 mg three times daily. Similarly, the metabolism of dextromethorphan was not significantly affected by St John’s wort when 12 healthy subjects were given a single 30-mg dose of dextromethorphan after 14 days of St John’s wort (Jarsin, Lichtwer Pharma) 300 mg three times daily. In yet another study in 12 healthy subjects, St John’s wort 300 mg three times daily for 14 days had no significant effect on the urinary excretion of a single 30-mg oral dose of dextromethorphan either after one day or after 2 weeks of use. The St John’s wort product used was from Sundown Herbals and provided about 33 mg of hyperforin daily.

Three further studies found that St John’s wort 300 mg three times daily (containing up to 24 mg of hyperforin) for 14 or 28 days had no clinically relevant effect on the pharmacokinetics of a single 5-mg dose of debrisoquine.

Experimental evidence

An in vitro study found that a St John’s wort extract (Hypericum Stada) inhibited the metabolism of dextromethorphan when used as a probe substrate for cytochrome P450 isoenzyme CYP2D6.

Mechanism

St John’s wort taken as a multiple-dose regimen does not appear to have a clinically relevant effect on the metabolism of dextromethorphan or debrisoquine, both of which are used as substrates to assess the activity of the cytochrome P450 isoenzyme CYP2D6. Inhibition of CYP2D6, seen in the single dose in vitro study, is therefore not expected to be clinically relevant.

Importance and management

St John’s wort is unlikely to interact with dextromethorphan to a clinically relevant extent. Dextromethorphan and debrisoquine are used as a probe drugs for CYP2D6 activity, and therefore these results also suggest that a pharmacokinetic interaction as a result of this mechanism between St John’s wort and other CYP2D6 substrates is unlikely.

St John’s wort + Digoxin

Digoxin toxicity occurred in a patient taking digoxin when he stopped taking St John’s wort. There is good evidence that some preparations of St John’s wort can reduce the levels of digoxin by about one-quarter to one-third.

Clinical evidence

An 80-year-old man taking long-term digoxin and St John’s wort herbal tea (2 litres daily) developed symptoms of digoxin toxicity (nodal bradycardia of 36 bpm and bigeminy) when he stopped taking the herbal tea.

In a study 13 healthy subjects were given digoxin for 5 days until steady state had been achieved, and then St John’s wort extract (LI 160, Lichtwer Pharma) 300 mg three times daily for a further 10 days. The AUC and trough level of digoxin decreased by 28% and 37%, respectively. When compared with a parallel group of 12 subjects taking digoxin and placebo, the St John’s wort group had 26.3% lower maximum plasma digoxin levels, 33.3% lower trough digoxin levels and a 25% lower AUC.

In another study, 8 healthy subjects pretreated with St John’s wort 300 mg 3 times daily for 14 days were given a single 500-microgram dose of digoxin. St John’s wort decreased the AUC0_7 of digoxin by 18%.3

Another study in 18 healthy subjects found that St John’s wort 300mg 3times daily (Nature’s Way, containing a daily dose of 24 mg of hyperforin) for 14 days reduced the maximum levels and AUCo-24 of a single 250-microgram dose of digoxin by 36% and 23%, respectively. These findings were comparable to rifampicin (an established P-glycoprotein inducer) 600 mg daily for 7 days. No significant adverse effects were reported.

In a further randomised placebo-controlled study, 93 healthy subjects were given digoxin alone for 7 days and then with one of ten St John’s wort preparations for 14 days. The extract used in the earlier study (LI 160, Jarsin 300, Lichtwer Pharma) 300 mg three times daily similarly reduced the digoxin AUC, peak and trough plasma levels by 25%, 37%, and 19%, respectively. Comparable results were found with hypericum powder containing similar amounts of hyperforin (about 21 mg daily), while hypericum powder with half the hyperforin content (about 10 mg daily) reduced the AUC, peak and trough plasma levels by about 18%, 21% and 13%, respectively. Some St John’s wort products, including tea, juice, oil extract and powder with low-dose hyperforin (all 5 mg daily or less), did not significantly affect the pharmacokinetics of digoxin.Similarly, a further study in 28 healthy subjects found no statistically significant change in digoxin pharmacokinetics when another low-hyperforin (about 3.5 mg daily) St John’s wort extract (Esbericum) 120 mg was given twice daily for 11 days to patients who had received a digoxin loading dose of 750 micrograms daily for 2 days before starting St John’s wort, and then received digoxin 250 micrograms daily each day during the study.

Experimental evidence

In a study using human cell lines, St John’s wort and hyperforin, a major active constituent, were found to induce P-glycoprotein transport of digoxin out of the cells in a reversible manner, which was comparable to rifampicin, a known inducer of P-glycoprotein. When treated with hypericin, another active constituent of St John’s wort, the transport of digoxin out of the cells was not increased.

Mechanism

St John’s wort, and specifically hyperforin, a major active constituent, has been shown to increase the activity of the P-glycoprotein drug transporter protein in the intestines, which reduces the absorption of digoxin.

Importance and management

Information seems to be limited to these reports, but the interaction would appear to be established. The extent of the interaction may depend on the St John’s wort preparation involved and dose used and seems to be correlated with the dose of hyperforin. Reductions in serum digoxin levels of the size seen with LI 160 could diminish the control of arrhythmias or heart failure. Digoxin serum levels should therefore be closely monitored if St John’s wort is either started or stopped, and appropriate dosage adjustments made if necessary. The recommendation of the CSM in the UK is that St John’s wort should not be used by patients taking digoxin.

Note that St John’s wort does not appear to interfere with various immunoassays used for therapeutic drug monitoring of digoxin, see St John’s wort + Laboratory tests.

St John’s wort + Eplerenone

St John’s wort slightly decreases the AUC of eplerenone.

Clinical evidence

St John’s wort caused a small 30% decrease in the AUC of a single 100-mg dose of eplerenone.

Experimental evidence

No relevant data found.

Mechanism

Eplerenone is metabolised by the cytochrome P450 isoenzyme CYP3A4, and therefore inducers of this isoenzyme, such as St John’s wort, would be expected to decrease its levels.

Importance and management

Because of the possibility of decreased efficacy of eplerenone, the UK manufacturers do not recommend the concurrent use of potent CYP3A4 inducers with eplerenone and they specifically name St John’s wort. However, it is unlikely that the decrease seen with St John’s wort is clinically relevant. Further study is needed to demonstrate the clinical significance.

St John’s wort + Etoposide

The interaction between St John’s wort and etoposide is based on experimental evidence only.

Clinical evidence

No interactions found.

Experimental evidence

In vitro studies suggest that hypericin, a component of St John’s wort, may antagonise the effects of etoposide. It may also stimulate the hepatic metabolism of etoposide by the cytochrome P450 isoenzyme CYP3A4.

Mechanism

Etoposide is metabolised by the cytochrome P450 isoenzyme CYP3A4, and therefore inducers of this isoenzyme, such as St John’s wort, would be expected to decrease its levels.

Importance and management

Information is very limited but it seems that it would be prudent to avoid St John’s wort in patients taking etoposide or related drugs. More study is needed.

St John’s wort + Fexofenadine

Pretreatment with St John’s wort had no clinically relevant effect on the plasma levels of single-dose fexofenadine in one study, but markedly reduced fexofenadine levels in two others.

Clinical evidence

In a study in 12 healthy subjects, a single 900-mg dose of St John’s wort (Hypericum perforatum) increased the maximum plasma level and AUC of a single 60-mg dose of fexofenadine by 45% and 31%, respectively.

Conversely, St John’s wort 300 mg three times daily for 14 days caused a slight 5 to 10% decrease in the maximum level and AUC of a single dose of fexofenadine 60 mg in the same subjects. In contrast, in another study, 12 days of pretreatment with St John’s wort increased the oral clearance of a single dose of fexofenadine by about 1.6-fold in healthy subjects. Similarly, a study in 30 healthy subjects found that 10 days of pretreatment with St John’s wort 300 mg three times daily almost doubled the oral clearance of a single 60-mg dose of fexofenadine.

Experimental evidence

No relevant data found.

Mechanism

In these studies St John’s wort was thought to be interacting via its effects on P-glycoprotein.

Importance and management

The findings from these multiple-dose studies suggest that St John’s wort either has no clinically relevant effect on fexofenadine, or a decrease occurs that is possibly clinically important. It may be prudent to monitor closely for signs of reduced fexofenadine efficacy in a patient taking regular St John’s wort and, if this is the case, consider St John’s wort as a possible cause. Further study is needed.

St John’s wort + Food; Tyramine-rich

An isolated report describes a patient taking St John’s wort who experienced a hypertensive crisis after consuming tyramine-rich food and drink.

Clinical evidence

A man who had taken a St John’s wort supplement for 7 days (preparation and dose not stated) was admitted to hospital with confusion and disorientation. He was unable to recall events after eating aged cheeses and pouring a glass of red wine 8 hours earlier. On examination he had a pulse rate of 115 bpm, a respiratory rate of 16 breaths per minute and his blood pressure was 210/140 mmHg. He was treated with intravenous phentolamine and oral labetalol and his blood pressure decreased to 160/100 mmHg after 2 hours and the delirium also resolved. Extensive laboratory investigations did not find any cause for the hypertension and delirium.

Experimental evidence

No relevant data found.

Mechanism

It was suggested that the time scale of starting to regularly take St John’s wort and the onset of delirium and hypertension after the consumption of tyramine-rich food and drink was suggestive of hypertension associated with MAOIs. Normally any ingested tyramine is rapidly metabolised by the enzyme monoamine oxidase in the gut wall and liver before it reaches the general circulation. However, if the activity of the enzyme at these sites is inhibited (by the presence of an MAOI), any tyramine passes freely into the circulation, causing not just a rise in blood pressure, but a highly exaggerated rise due to the release from the adrenergic neurones of the large amounts of noradrenaline that accumulate there during inhibition of MAO. Although St John’s wort is a potent inhibitor of MAO, this effect has not been demonstrated at recommended doses. It was concluded that the hypertensive crisis in this patient may have been mediated by MAO inhibition, but there was also a possibility of another, as yet unknown, pharmacological action of St John’s wort being involved.

Importance and management

Given the widespread use of St John’s wort, this case would seem to be unusual, and there is currently little grounds for suggesting any dietary restriction in those taking St John’s wort.

St John’s wort + Herbal medicines

For a case report describing delirium following the use of St John’s wort, valerian and loperamide, see under St John’s wort + Loperamide.

St John’s wort + Hormonal contraceptives

St John’s wort may affect the pharmacokinetics of desogestrel, ethinylestradiol and norethisterone. Both breakthrough bleeding and, more rarely, combined oral contraceptive failure have been reported in women taking St John’s wort Two cases describe the failure of emergency hormonal contraception, which was attributed to the use of St John’s wort.

Clinical evidence

(a) Combined hormonal contraceptives

A study in 17 healthy women taking ethinylestradiol/desogestrel 20/150 micrograms daily found that St John’s wort (300 mg twice or three times daily) did not affect the AUC or maximum levels of ethinylestradiol. However, the AUC and maximum levels of the active metabolite of desogestrel were significantly decreased by about 40% and 20%, respectively. There was no evidence that ovulation occurred. However, the frequency of breakthrough bleeding increased significantly from 35% to around 80%, which may affect compliance. Another study in 12 healthy women taking ethinylestradiol/norethisterone 35 micrograms/lmg (Ortho-Novum) found that St John’s wort 300 mg three times daily for 8 weeks increased the oral clearance of norethisterone and reduced the half-life of ethinylestradiol, but the serum levels of LH, FSH and progesterone were unaffected. However, of more importance, was the increase in breakthrough bleeding, which the authors state as a major cause of patients stopping hormonal contraceptives. A further study in 16 subjects also found reductions in the levels of low-dose ethinylestradiol/norethisterone 20 micrograms/1 mg. Furthermore, they found increased progesterone levels of more than 3nanog-rams/mL (an indication that ovulation occurred) in 3 patients who also took St John’s wort compared with one subject who took placebo. Breakthrough bleeding was also increased. In a secondary analysis of this study, the anti-androgenic effects of ethinylestradiol/norethisterone, utilised in the treatment of hirsutism and acne, were not significantly affected by St John’s wort.

The Adverse Drug Reactions Database of the Swedish Medical Products Agency has on record 2 cases of pregnancy due to the failure of a combined oral contraceptive, which was attributed to the use of products containing St John’s wort (Esbericum and Kira). One woman was taking ethinylestradiol and norethisterone and the other was taking ethinylestradiol and levonorgestrel. This follows an earlier report from the Swedish Medical Products Agency of 8 cases of breakthrough bleeding in women aged 23 to 31 taking long-term oral contraceptives and St John’s wort. Breakthrough bleeding occurred within about a week of starting St John’s wort in 5 of the cases, and was known to have resolved in 3 cases when the St John’s wort was stopped. The CSM in the UK has on record a further 7 cases of pregnancy in women taking St John’s wort and oral contraceptives in the two-year period from February 2000 to February 2002. Another earlier brief report describes 3 women taking a combined oral contraceptive (ethinylestradiol/desogestrel 30/150 micrograms) who developed breakthrough bleeding one-week (2 cases) and 3 months (1 case) after starting to take St John’s wort. A single case of pregnancy has also been reported in a patient taking St John’s wort with ethinylestradiol/dienogest (Vedette)? The German Federal Institute for Drugs and Medical Devices has received a total of 8 case reports of ineffective contraception with St John’s wort.

In contrast, in a study, 16 healthy women took ethinylestradiol/ desogestrel 20/150 micrograms daily on days 1 to 21 of a 28-day cycle, and an extract of St John’s wort with a low hyperforin content of 650 micrograms, (Zell7, standardised to 0.2% hypericin), 250 mg twice daily on days 7 to 21. The plasma levels of ethinylestradiol and the active metabolite of desogestrel were not significantly altered by St John’s wort. None of the women experienced any breakthrough bleeding or spotting, and measurements of plasma hormone levels indicated that the contraceptive efficacy was unchanged.

(b) Emergency hormonal contraceptives

The CSM in the UK has received reports of 2 women taking St John’s wort who became pregnant despite taking emergency hormonal contraception. One of them was also taking an oral contraceptive.

Experimental evidence

No relevant data found.

Mechanism

It is believed that St John’s wort can induce the metabolism of the contraceptive steroids by the cytochrome P450 isoenzyme CYP3A4, thereby reducing their serum levels and their effects. This can lead to breakthrough bleeding and, in some cases, contraceptive failure. This is consistent with the way that St John’s wort appears to lower the serum levels of some other drugs. Note that, although hyperforin is the most likely constituent responsible for enzyme induction (supported by the study that found no interaction with a low-hyperforin preparation), others may contribute and the levels of individual constituents can vary between different preparations of the herb.

Importance and management

Information appears to be limited to these reports but the interaction between hormonal contraceptives and St John’s wort appears to be established. Its incidence is not known but the evidence so far suggests that breakthrough bleeding may be a problem, although pregnancy resulting from this interaction appears to be uncommon. Only two cases of emergency hormonal contraceptive failure attributed to an interaction with St John’s wort have so far been reported, but the effects of any interaction here would be very difficult to assess. Since it is not known who is particularly likely to be at risk, the recommendation of the CSM/MCA and the Faculty of Family Planning and Reproductive Health Care (FFPRHC) in the UK. is that women taking oral contraceptives (both combined and progestogen-only pills) should either avoid St John’s wort or they should use an additional form of contraception. The FFPRHC Clinical Effectiveness Unit is in agreement with the CSM advice but recommends that, if St John’s wort must be continued, the following general guidelines for the use of liver enzyme inducers with hormonal contraceptives should be followed:

• Women taking combined oral contraceptives should use an ethinylestradiol dose of at least 50 micrograms daily. The dose may be increased further above 50 micrograms if breakthrough bleeding occurs. Omitting or reducing the pill-free interval has not been shown to reduce the risk of ovulation with liver enzyme inducers. Additional non-hormonal methods of contraception, such as condoms, should also be used by patients using combined hormonal contraceptives, both when taking the liver enzyme inducers and for at least 4 weeks after stopping the drug. Alternatives to all forms of combined hormonal contraceptives should be considered with long-term use of liver enzyme inducers.

• The combined contraceptive patch may be continued in the usual manner. Additional, non-hormonal methods of contraception, such as condoms, should also be used by patients using the combined contraceptive patch, both when taking the liver enzyme inducers and for at least 4 weeks after stopping the drug. Using more than one patch is not recommended.

• The progestogen-only oral contraceptive is not recommended for use with liver enzyme inducers. Alternative methods of contraception are advised.

• The progestogen-only implant may be continued with short courses of enzyme inducers. Additional non-hormonal methods of contraception, such as condoms, should also be used by patients using the progestogen-only implant, both when taking the liver enzyme inducers and for at least 4 weeks after stopping the drug. Alternatives to the progestogen-only implant should be considered with long-term use of liver enzyme inducers.

• The effectiveness of both combined and the progestogen-only emergency hormonal contraceptive will be reduced in women taking liver enzyme inducers. The FFPRHC Clinical Effectiveness Unit states that there appears to be no good evidence on how to manage the interaction between emergency hormonal contraception and enzyme inducers such as St John’s wort, but current clinical practice is to increase the contraceptive dose by approximately 50%. The British National Formulary recommends giving a single 3-mg dose of levonorgestrel, although this is unlicensed. A copper intrauterine device (IUD) may also be used as an effective alternative. In the UK it is possible to buy the progestogen-only emergency hormonal contraception without a prescription; however, it has been advised that patients taking enzyme inducers should not be supplied the emergency hormonal contraceptive but should be referred to a doctor or family planning service. Given the potential consequences of an unwanted pregnancy, these seem sensible precautions.

• The depot progestogen-only injection, copper and levonorgestrel-releasing IUDs do not appear to be affected by enzyme-inducing drugs, such as St John’s wort, and may be used as alternative contraceptive methods, particularly for women requiring hormonal contraception who are likely to be taking the enzyme inducer in the long term, as these are unaffected by liver enzyme inducers.

Although the considerable worldwide popularity of St John’s wort is fairly recent, it is currently the most widely used antidepressant in Germany and has been used for very many years in both Germany and Austria. Yet, there seems to be no published evidence that oral contraceptive failure in those countries is more frequent than anywhere else. This would seem to confirm that contraceptive failure leading to pregnancy occurring as a result of this interaction is very uncommon, or perhaps that it has failed to be identified as a possible cause.

The anti-androgenic effects of ethinylestradiol/norethisterone, utilised in the treatment of hirsutism and acne, do not appear to be significantly affected by St John’s wort. However, as this was a small study, it may be prudent to still monitor the effectiveness of the combined hormonal contraceptive for this indication until further evidence is available.

St John’s wort + Ibuprofen

St John’s wort does not affect the pharmacokinetics of ibuprofen.

Clinical evidence

Eight healthy male subjects were given an oral dose of ibuprofen 400 mg before, and at the end of, a 21-day course of St John’s wort 300 mg three times daily. The pharmacokinetics of ibuprofen were unaffected by St John’s wort. The St John’s wort extract was standardised to contain hypericin (probably 0.3%) and a minimum of 4% hyperfbrin.

Experimental evidence

No relevant data found.

Mechanism

As ibuprofen is a substrate for the cytochrome P450 isoenzymes CYP2C9 and CYP2C8, the authors of the study suggest that the lack of interaction is evidence that St John’s wort has no significant effects on these isoenzymes. Minor or no significant effects on pharmacokinetics have similarly been reported for rosiglitazone, a substrate for CYP2C8, and gliclazide and tolbutamide, both of which are substrates for CYP2C9. See St John’s wort + Antidiabetics.

Importance and management

St John’s wort does not appear to interact with ibuprofen and therefore no special precautions seem necessary on concurrent use.

St John’s wort + Imatinib

St John’s wort lowers serum imatinib levels.

Clinical evidence

In a study in 12 healthy subjects, the pharmacokinetics of a single dose of imatinib was determined before, and on day 12, of two weeks of treatment with St John’s wort extract (Kira (LI 160), Lichtwer Pharma) 300 mg three times daily. The AUC and maximum plasma level of imatinib was decreased by 30% and 15%, respectively. Imatinib clearance was increased by 43% and its half-life was decreased from 12.8 to 9 hours. Similar results were found in another study.

Experimental evidence

No relevant data found.

Mechanism

St John’s wort induces intestinal CYP3A4 and it therefore also reduces imatinib levels.

Importance and management

This study suggests that St John’s wort may modestly reduce the exposure to imatinib, which could result in a reduction in its efficacy. The manufacturers suggest that concurrent use of imatinib and potent enzyme-inducing drugs should be avoided. St John’s wort has smaller effects than other known CYP3A4 inducers, but, nevertheless, some suggested that concurrent use should also be avoided. However, if this is not possible it would be prudent to monitor the outcome of concurrent use closely, and increase the imatinib dose as necessary.

St John’s wort + Irinotecan

St John’s wort increases the metabolism of irinotecan, which may decrease its activity.

Clinical evidence

In a randomised, crossover study St John’s wort decreased the plasma levels of the active metabolite of irinotecan, SN-3 8, by 42%. Myelosuppression was also reduced; with irinotecan alone the leucocyte and neutrophil counts decreased by 56% and 63%, respectively, but in the presence of St John’s wort the decreases were only 8.6% and 4.3%, respectively. In this study, irinotecan was given as a single 350-mg/m intravenous dose every 3 weeks, and during one cycle a St John’s wort preparation was given three times daily, beginning 14 days before and stopping 4 days after the irinotecan.

Experimental evidence

In an experimental study in rats, St John’s wort 400mg/kg given daily for 14 days reduced the maximum levels of irinotecan and its active metabolite, SN-38, by 39.5% and 38.9%, respectively. The AUC of SN-38 was also reduced by 26.3%.

Mechanism

St John’s wort induces the cytochrome P450 isoenzyme CYP3A4 and P-glycoprotein, which are both involved in the metabolism of irinotecan. The evidence suggests that St John’s wort increases the metabolism of irinotecan to an unknown inactive metabolite, rather than the active SN-38, thereby reducing its effects.

Importance and management

The evidence appears to be limited. Irinotecan has a narrow therapeutic range and, as irinotecan is a prodrug that is metabolised to its active metabolite SN-3 8, the lower levels of SN-3 8 suggest that its activity will be reduced in the presence of St John’s wort. It would therefore seem sensible to warn patients who are about to receive irinotecan to avoid St John’s wort. It seems likely that topotecan, a related drug that is also a substrate for CYP3A4, will be similarly affected, but evidence for this is currently lacking.

St John’s wort + Ivabradine

The metabolism of ivabradine is increased by St John’s wort

Clinical evidence

Twelve healthy subjects were given a single oral dose of ivabradine 10 mg 24 hours before St John’s wort (Jarsin tablets) 300 mg three times daily was given for 14 days. On day 16, they were given a further dose of ivabradine 10 mg with a single 300-mg dose of St John’s wort. The maximum levels and AUC of ivabradine were reduced by more than half by St John’s wort. The maximum levels and AUC of its active metabolite were reduced by 25% and 32%, respectively. No adverse effects were reported, and the heart rate and blood pressure remained unchanged. Similar findings are also reported by the manufacturers of ivabradine.

Experimental evidence

No relevant data found.

Mechanism

St John’s wort is a known inducer of the cytochrome P450 isoenzyme CYP3A4, by which ivabradine is metabolised. Concurrent use therefore increases the metabolism of ivabradine, which results in a reduction in its plasma levels, and a potential reduction in effects.

Importance and management

Evidence is limited to the study above and, despite the lack of change in pharmacodynamic effects seen in this study, the pharmacokinetic changes may be significant to affect individual patients. Monitor concurrent use for ivabradine efficacy and adjust the dose as necessary. Remember to re-adjust the dose of ivabradine if concurrent use of these drugs is stopped. The UK manufacturer suggests that the use of St John’s wort should be restricted in patients taking ivabradine.

St John’s wort + Laboratory tests

St John’s wort does not interfere with in vitro assays for carbamazepine, ciclosporin, digoxin, phenobarbital, phenytoin, procainamide, quinidine, tacrolimus, theophylline, tricyclic antidepressants and valproate.

Clinical evidence

No interactions found.

Experimental evidence

In in vitro experiments, St John’s wort added to serum samples did not interfere with a fluorescence polarisation immunoassay (FPIA, Abbott Laboratories) for carbamazepine, digoxin, phenytoin, quinidine, theophylline, tricyclic antidepressants and valproate. It also did not interfere with a serum sample microparticle enzyme immunoassay (MEIA, Abbott Laboratories) for digoxin, or other assays (exact assays not specified, Roche Diagnostics/Hitachi) for phenobarbital or procainamide. Whole-blood FPIA analysis of riclosporin levels and whole-blood MEIA analysis of tacrolimus levels were also not affected by the addition of St John’s wort.

Mechanism

No mechanism expected.

Importance and management

St John’s wort does not appear to interfere with various immunoassays used for therapeutic drug monitoring of carbamaze-pine, ciclosporin, digoxin, phenobarbital, phenytoin, procainamide, quinidine, tacrolimus, theophylline, tricyclics and valproate.

St John’s wort + Lithium

A brief report describes mania in a patient taking lithium who also took St John’s wort.

Clinical evidence

A search of Health Canada’s database of spontaneous adverse reactions identified one case in which St John’s wort was suspected of inducing mania in a patient also taking lithium.

Experimental evidence

No relevant data found.

Mechanism

Unknown, although it seems likely that the symptoms could be due to the effects of both lithium and St John’s wort on serotonin.

Importance and management

No general conclusions can be drawn from this case as no further details were given.

St John’s wort + Loperamide

A case report describes delirium in a woman taking St John’s wort and valerian root who also took loperamide.

Clinical evidence

A 39-year-old woman who had been taking two tablets of St John’s wort with valerian root daily for 6 months (exact products and doses not specified) was hospitalised after becoming disorientated, agitated and confused. The patient had also recently started loperamide for diarrhoea prior to admission. The delirium subsided within two days of stopping these drugs.

Experimental evidence

No relevant data found.

Mechanism

Unclear. A MAOI-induced reaction caused by the combination of St John’s wort and loperamide was suggested as a possible cause for the delirium. However, an interaction between St John’s wort and valerian, or valerian and loperamide, cannot be ruled out.

Importance and management

This appears to be the only report of delirium associated with the combination of St John’s wort, valerian and loperamide. Its general relevance is therefore unclear.

St John’s wort + Methylphenidate

St John’s wort may decrease the efficacy of methylphenidate in the treatment of attention deficit hyperactivity disorder.

Clinical evidence

A 22-year-old man who had been successfully treated with methylphenidate 20 mg daily for attention deficit hyperactivity disorder (ADHD) for 6 months started to take St John’s wort 600 mg daily. Over the next 4 months the efficacy of the methylphenidate decreased, but, 3 weeks after the St John’s wort was stopped, the methylphenidate became more effective. No adverse effects were seen during the concurrent use of the herbal medicine and drug.

Experimental evidence

No relevant data found.

Mechanism

Unknown.

Importance and management

This is an isolated case report and therefore no general recommendations can be made. However, if the efficacy of methylphenidate becomes reduced, it may be worth questioning the patient about St John’s wort use, and giving consideration to stopping the herb.

St John’s wort + Mycophenolate

St John’s wort does not appear to alter the pharmacokinetics of mycophenolate.

Clinical evidence

In a pharmacokinetic study, 8 stable kidney transplant recipients taking mycophenolate 1 g to 2 g daily and tacrolimus were given 600 mg of St John’s wort extract (Jarsin 300) daily for 14 days. The levels of mycophenolic acid, the main metabolite of mycophenolate, were measured before St John’s wort was started, on day 14, and two weeks after St John’s wort was stopped. The pharmacokinetics of mycophenolic acid were unchanged throughout the study, and no dosage adjustments were needed in any of the 8 patients.

Experimental evidence

No relevant data found.

Mechanism

No mechanism. St John’s wort is an inducer of the cytochrome P450 isoenzyme CYP3A4 and P-glycoprotein. As mycophenolate is not significantly metabolised or transported by these routes, an interaction would not be expected.

Importance and management

St John’s wort does not appear to affect the pharmacokinetics of mycophenolate and therefore no additional precautions seem necessary on concurrent use.

St John’s wort + NNRTIs

There is some evidence to suggest that St John’s wort may decrease the levels of nevirapine. Delavirdine and efavirenz would be expected to be similarly affected.

Clinical evidence

Nevirapine levels, obtained by routine monitoring, were noted to be lower in 5 men who were also taking St John’s wort. Based on a pharmacokinetic modelling analysis, it was estimated that St John’s wort increased the oral clearance of nevirapine by about 35%.

Experimental evidence

No relevant data found.

Mechanism

This finding supports predictions based on the known metabolism of the NNRTIs delavirdine, efavirenz and nevirapine by the cytochrome P450 isoenzyme CYP3A4, of which St John’s wort is a known inducer.

Importance and management

The interaction between St John’s wort and nevirapine confirms advice issued by the CSM in the UK, that St John’s wort may decrease blood levels of the NNRTIs with possible loss of HIV suppression. Therefore concurrent use should be avoided.

St John’s wort + Opioids

St John’s wort reduces the plasma concentrations of methadone and withdrawal symptoms may occur.

Clinical evidence

In a study in 4 patients taking methadone, St John’s wort (Jarsin) 900 mg daily for 14 to 47 days decreased methadone plasma concentration-to-dose ratios (indicating decreased methadone levels) by 19 to 60%. Two patients reported symptoms that suggested a withdrawal syndrome.

Experimental evidence

No relevant data found.

Mechanism

St John’s wort is metabolised in the liver and induces the cytochrome P450 enzyme CYP3A4, and so could affect plasma levels of drugs metabolised in this way, such as methadone.

Importance and management

St John’s wort appears to reduce the plasma levels of methadone causing withdrawal symptoms in some patients. Therefore, concurrent use should be avoided. It may be prudent to follow the same advice for other opioids that are mainly metabolised by CYP3A4, such as buprenorphine, fentanyl and alfentanil.

St John’s wort + Prednisone

St John’s wort does not appear to affect the pharmacokmetics of prednisone.

Clinical evidence

Eight healthy male subjects were given a single oral dose of prednisone 20 mg before, and at the end of, a 28-day course of St John’s wort 300 mg three times daily. The pharmacokinetics of prednisone, and its metabolite prednisolone, were not significantly affected by St John’s wort. The St John’s wort extract was standardised to contain hypericin 0.3% and a minimum of 4% hyperforin.

Experimental evidence

No relevant data found.

Mechanism

It was thought that St John’s wort, a known inducer of the cytochrome P450 isoenzyme CYP3A4, would increase the metabolism of prednisone and prednisolone and reduce their levels. While prednisone and prednisolone are substrates of CYP3A4, it is not a major metabolic pathway as they have been shown to be relatively unaffected by potent CYP3A4 inhibitors in healthy subjects.

Importance and management

St John’s wort does not appear to induce the metabolism of a single dose of prednisone, or its metabolite prednisolone, in healthy male subjects; however, further study is needed to clarify significance of this in patients receiving long-term prednisone.

St John’s wort + Procainamide

The interaction between St John’s wort and procainamide is based on experimental evidence only.

Clinical evidence

No interactions found.

Experimental evidence

In a study in mice, a single dose of St John’s wort extract significantly raised the bioavailability of procainamide 100 mg/kg for a period of up to 4 hours. A trend towards an increase in procainamide levels was seen in the mice given St John’s wort for 2 weeks, with the procainamide dose given the day after St John’s wort was stopped; however, this was not statistically significant. Other pharmacokinetic parameters remained unaffected by both single-dose and long-term use of St John’s wort.

Mechanism

Not understood.

Importance and management

The evidence for any significant effect of St John’s wort on the pharmacokinetics of procainamide is extremely limited and, although the bioavailability of procainamide may have been raised slightly in mice, its metabolism was unchanged. The clinical significance of this in humans is unknown and further study is needed.

St John’s wort also does not interfere with laboratory assays for procainamide, see St John’s wort + Laboratory tests.

St John’s wort + Protease inhibitors

St John’s wort causes a marked reduction in the serum levels of indinavir, which may result in HIV treatment failure. Other protease inhibitors, whether used alone or boosted by ritonavir, are predicted to interact similarly.

Clinical evidence

In a single-drug pharmacokinetic study, 8 healthy subjects were given three 800-mg doses of indinavir on day 1 to achieve steady-state serum levels, and then an 800-mg dose on day 2. For the next 14 days they were given St John’s wort extract 300 mg three times daily. Starting on day 16, the indinavir dosing was repeated. It was found that the St John’s wort reduced the mean AUC of indinavir by 54% and decreased the 8-hour indinavir trough serum level by 81%.

Experimental evidence

No relevant data found.

Mechanism

Not fully understood, but it seems highly likely that St John’s wort induces the activity of the cytochrome P450 isoenzyme CYP3A4, thereby increasing the metabolism of indinavir and therefore reducing its levels.

Importance and management

Direct information seems to be limited to this study, but the interaction would appear to be established. Such a large reduction in the serum levels of indinavir is likely to result in treatment failures and the development of viral resistance. Therefore St John’s wort should be avoided. There seems to be no direct information about other protease inhibitors, but since they are also metabolised by CYP3A4 it is reasonable to expect that they will be similarly affected by St John’s wort. The FDA in the US has suggested that concurrent use of St John’s wort and protease inhibitors is not recommended. Similarly, the CSM in the UK has advised that patients taking protease inhibitors should avoid St John’s wort and that anyone already taking both should stop the St John’s wort and have their HIV RNA viral load measured. The levels of the protease inhibitors are likely to increase as the induction effects of St John’s wort diminish, usually over one to two weeks. Therefore the dose of the protease inhibitor will probably need adjusting. The US and UK manufacturers of all protease inhibitors (amprenavir, atazana-vir, darunavir, fosamprenavir, lopinavir/ritonavir, nelfinavir, ritonavir, saquinavir, tipranavir) either contraindicate or advise against the use of St John’s wort.

St John’s wort + Proton pump inhibitors

St John’s wort induces the metabolism of omeprazole, and this might result in reduced efficacy. Other proton pump inhibitors are likely to be similarly affected.

Clinical evidence

In a crossover study, 12 healthy subjects (6 of the extensive CYP2C19 metaboliser phenotype and 6 of the poor CYP2C19 metaboliser phenotype) were given St John’s wort 300 mg three times daily or placebo for 14 days, followed by a single 20-mg dose of omeprazole on day 15. St John’s wort modestly decreased the AUC of omeprazole in all subjects (by 49% in extensive metabolisers and 41% in poor metabolisers), and also increased the plasma levels of hydroxyomeprazole by 35% in those who were extensive metabolisers. It also markedly increased the levels of the inactive CYP3A4 sulfone metabolite of omeprazole in both extensive and poor metabolisers (by 148% and 132%, respectively).

Experimental evidence

No relevant data found.

Mechanism

St John’s wort increases the metabolism of omeprazole by inducing both CYP2C19 and CYP3A4.1

Importance and management

This appears to be the only study examining the effects of St John’s wort on proton pump inhibitors. However, the reduction seen in the AUC of omeprazole (about 40%) suggest that there is a possibility that omeprazole will be less effective in patients taking St John’s wort. As all PPIs are metabolised by CYP2C19 to varying extents, it is likely that the effects of St John’s wort seen in these studies will be similar with other PPIs, although note that rabeprazole is much less dependent on this route of metabolism than other PPIs.

There is insufficient evidence to suggest that St John’s wort should be avoided in patients taking PPIs. However, the potential reduction in the efficacy of the PPI should be borne in mind, particularly where the consequences may be serious, such as in patients with healing ulcers.

St John’s wort + SNRIs

Serotonin syndrome has been reported in one patient taking venlafaxine and St John’s wort

Clinical evidence

An interaction between venlafaxine and St John’s wort was reported to the Centre Regional de Pharmacovigilance de Marseille involving a 32-year-old man who had been taking venlafaxine 250 mg daily for several months. He started taking St John’s wort at a dose of 200 drops three times daily (usual dose up to 160 drops daily) and on the third day felt faint and anxious, and had symptoms of diaphoresis, shivering and tachycardia. St John’s wort was stopped and his symptoms resolved in 3 days without altering the dose of venlafaxine. A search of Health Canada’s database of spontaneous adverse reactions for the period 1998 to 2003 also found one case of suspected serotonin syndrome as a result of an interaction between venlafaxine and St John’s wort.

Experimental evidence

No relevant data found.

Mechanism

A pharmacodynamic interaction may occur between St John’s wort and venlafaxine because they can both inhibit the reuptake of 5-hydroxytryptamine (serotonin). Serotonin syndrome has been seen with St John’s wort alone, and so additive serotonergic effects appear to be the explanation for what occurred in the cases described here.

Importance and management

Information appears to be limited to these reports. Duloxetine would be expected to interact similarly and the manufacturers of both duloxetine and venlafaxine generally advise caution if they are given with drugs that affect the serotonergic neurotransmitter systems, a similar caution with St John’s wort would be prudent.

St John’s wort + SSRIs

Cases of severe sedation, mania and serotonin syndrome have been reported in patients taking St John’s wort with SSRIs.

Clinical evidence

(a) Fluoxetine

For a report of hypomania that occurred when St John’s wort, ginkgo biloba and melatonin were added to treatment with fluoxetine and buspirone, see St John’s wort + Buspirone.

For a report of serotonin syndrome when eletriptan, fluoxetine and St John’s wort were used together, see St John’s wort + Triptans.

(b) Paroxetine

In one report, a woman stopped taking paroxetine 40 mg daily after 8 months, and 10 days later started to take 600 mg of St John’s wort powder daily. No problems occurred until the next night when she took a single 20-mg dose of paroxetine because she thought it might help her sleep. The following day at noon she was found still to be in bed, reusable but incoherent, groggy and slow moving, and almost unable to get out of bed. Two hours later she still complained of nausea, weakness and fatigue, but her vital signs and mental status were normal. Within 24 hours all symptoms had resolved.

(c) Sertraline

Four elderly patients taking sertralme developed symptoms characteristic of serotonin syndrome within 2 to 4 days of also taking St John’s wort 300 mg, either two or three times daily. The symptoms included dizziness, nausea, vomiting, headache, anxiety, confusion, restlessness and irritability. Two of them were treated with oral cyproheptadine 4mg either two or three times daily, and the symptoms of all of them resolved within a week. They later resumed treatment with sertralme without problems. A search of Health Canada’s database of spontaneous adverse reactions from 1998 to 2003 found two cases of suspected serotonin syndrome as a result of an interaction between sertralme and St John’s wort.

Mania developed in a 28-year-old man, who continued to take St John’s wort against medical advice while also receiving sertralme 50 mg daily for depression; he was also receiving testosterone replacement post-orchidectomy.

Experimental evidence

No relevant data found.

Mechanism

A pharmacodynamic interaction may occur between St John’s wort and SSRIs because they can both inhibit the reuptake of 5-hydroxytryptamine (serotonin). Serotonin syndrome has been seen with St John’s wort alone, and so additive serotonergic effects appear to be the explanation for what occurred in the cases described here.

Importance and management

Information appears to be limited to these reports, but interactions between SSRIs and St John’s wort would seem to be established. The incidence is not known but it is probably small; nevertheless because of the potential severity of the reaction it would seem prudent to avoid concurrent use. The advice of the CSM in the UK is that St John’s wort should be stopped if patients are taking any SSRI because of the risk of increased serotonergic effects and an increased incidence of adverse reactions.

St John’s wort + Statins

St John’s wort modestly decreases the plasma levels of atorvastatin and simvastatin, but not pravastatin.

Clinical evidence

In a placebo-controlled, crossover study, 16 healthy subjects took St John’s wort 300 mg three times daily for 14 days. On day 14 simvastatin 10 mg was given to 8 subjects and pravastatin 20 mg was given to the other 8 subjects. St John’s wort did not affect the plasma concentration of pravastatin, but it tended to reduce the simvastatin AUC and significantly reduce the AUC of its active metabolite, simvastatin hydroxyacid, by 62%.

In a crossover study in 24 patients with hypercholesterolemia taking long-term simvastatin 10 to 40 mg daily (an average dose of 20.8 mg daily), St John’s wort (Movina) 300 mg twice daily for 4 weeks significantly raised the levels of total cholesterol from 4.56mmol/L (pre-treatment) to 5.08mmol/L and LDL-cholesterol from 2.30mmol/L to 2.72mmol/L. The authors equate the magnitude of the increased LDL-cholesterol levels to a halving of the effects of simvastatin.

In a similar study by the same authors, 16 patients with hypercholesterolemia taking long-term atorvastatin 10 to 40 mg daily (an average dose of 14.4 mg daily) were given St John’s wort (Movina) 300 mg twice daily for 4 weeks. St John’s wort significantly raised the levels of total cholesterol from 4.76mmol/L (pre-treatment) to 5.1 mmol/L and LDL-cholesterol from 2.39mmol/L to 2.66 mmol/L. The levels of atorvastatin were not measured in this study. The authors equate the magnitude of the increased LDL-cholesterol levels to a loss of a third of the effects of atorvastatin. No adverse effects were reported.

Experimental evidence

No relevant data found.

Mechanism

The reason for this interaction is unknown, but St John’s wort may reduce the levels of simvastatin and its metabolite, and atorvastatin, by inducing the cytochrome P450 isoenzyme CYP3A4 or by having some effect on P-glycoprotein.

Importance and management

Although the evidence is limited, it appears that St John’s wort may reduce the efficacy of atorvastatin and simvastatin, which may result in a clinically relevant increase in total cholesterol and LDL levels, depending on the patient’s baseline result and medical history. It may be prudent to consider an interaction if lipid-lowering targets are not met, and advise the patient to stop taking St John’s wort or adjust the dose of statin, as needed.

No significant interaction would be expected with pravastatin as it is not primarily metabolised by CYP3A4, and this was demonstrated in the study above. As fluvastatin and rosuvastatin are not significantly metabolised by CYP3A4, a clinically relevant interaction would also not be expected.

St John’s wort + Tacrolimus

St John’s wort decreases tacrolimus levels.

Clinical evidence

In a clinical study, 10 healthy subjects were given a single 100-microgram/kg dose of tacrolimus alone, or after they took St John’s wort 300 mg three times daily for 14 days. On average St John’s wort decreased the maximum blood level of tacrolimus by 65% and its AUC by 32%. However, the decrease in AUC ranged from 15% to 64%, with one patient having a 31% increase in AUC. Similar results have been found in a study in 10 kidney transplant recipients given St John’s wort (Jarsin 300) 600 mg daily for 2 weeks. In order to achieve target levels, the tacrolimus dose was increased in all patients, from a median of 4.5 mg daily to 8 mg daily. Two weeks after stopping St John’s wort, tacrolimus doses were reduced to a median of 6.5 mg daily, and then to the original dose of 4.5 mg daily after about 4 weeks.

A case report describes a 65-year-old patient taking tacrolimus following a kidney transplantation. The patient started to take St John’s wort (Neuroplani) 600 mg daily, and after one month the tacrolimus trough blood levels had dropped from a range of 6 to 10 nanograms/mL down to 1.6nanograms/mL, with an unexpected improvement in creatinine levels. When the St John’s wort was stopped, tacrolimus levels and creatinine returned to the previous range. Subsequently a lower target range of tacrolimus was set.

Experimental evidence

No relevant data found.

Mechanism

St John’s wort induces the cytochrome P450 isoenzyme CYP3A4 and affects the transporter protein P-glycoprotein. CYP3A4 and P-glycoprotein are involved in the metabolism and clearance of tacrolimus, so an increase in their effects would be expected to result in a decrease in tacrolimus levels.

Importance and management

Although the evidence currently seems limited to these reports, the interaction between tacrolimus and St John’s wort has been predicted from the pharmacokmetics of these two drugs. Given the unpredictability of the interaction (and the variability in content of St John’s wort products) it would seem prudent to avoid St John’s wort in transplant recipients, and possibly other types of patient taking tacrolimus. If St John’s wort is started or stopped, monitor tacrolimus levels closely and adjust the dose accordingly.

St John’s wort also does not interfere with laboratory assays for tacrolimus, see St John’s wort + Laboratory tests.

St John’s wort + Talinolol

St John’s wort modestly decreases the plasma levels of talinolol.

Clinical evidence

In a randomised study, a single oral dose of talinolol 50 mg was given to 9 healthy subjects after 12-days of St John’s wort (Jarsin, Lichtwer Pharma) 900 mg daily. St John’s wort was found to reduce the AUC and oral bioavailability of talinolol by about 31% and 25%, respectively. The non-renal clearance of a single dose of talinolol 30 mg given as a 30-minute infusion was increased by about 26%. Other pharmacokinetic parameters of both oral and intravenous talinolol were not significantly affected.

Experimental evidence

No relevant data found.

Mechanism

Talinolol is a known substrate for P-glycoprotein. This study found that the levels of intestinal P-glycoprotein in the duodenal biopsy samples of 9 subjects were raised by St John’s wort, leading to a reduction in the absorption of talinolol.

Importance and management

Information appears to be limited to this study but it is in line with the known effects of St John’s wort on substrates of P-glycoprotein, such as digoxin. See St John’s wort + Digoxin. The modest decrease in talinolol levels suggests that, in most patients, this interaction is unlikely to be clinically significant. Nevertheless, consider this interaction if blood pressure is difficult to control.

St John’s wort + Theophylline

A patient needed a marked increase in the dosage of theophylline while taking St John’s wort. In contrast, no pharmacokinetic interaction was found in a 2-week study in healthy subjects.

Clinical evidence

A study in 12 healthy subjects found that a standardised preparation of St John’s wort 300 mg (hypericin 0.27%) three times daily for 15 days had no significant effects on the plasma level of a single 400-mg oral dose of theophylline.

However, an isolated case has been reported of a woman, who had previously been stable for several months taking theophylline 300 mg twice daily, but was found to need a marked increase in her theophylline dosage to 800 mg twice daily to achieve serum levels of 9.2 mg/L. Two months previously she had started to take 300 mg of a St John’s wort supplement (hypericin 0.3%) each day. When she stopped taking the St John’s wort, her serum theophylline levels doubled within one week to 19.6 mg/L and her theophylline dosage was consequently reduced. This patient was also taking a whole spectrum of other drugs (amitriptyline, furosemide, ibuprofen, inhaled triamcinolone, morphine, potassium, prednisone, salbutamol (albuterol), valproic acid, zolpidem and zafirlukast) and was also a smoker. No changes in the use of these drugs or altered compliance were identified that might have offered an alternative explanation for the changed theophylline requirements.

Experimental evidence

In vitro data suggest that hypericin can act as an inducer of the cytochrome P450 isoenzyme CYP1A2.

Mechanism

Uncertain. It has been suggested that treatment with St John’s wort for 15 days was unlikely to induce the isoenzymes sufficiently to cause changes in plasma theophylline, and St John’s wort is thought to have a rather limited ability to induce CYP1A2. The patient in the case report had been taking St John’s wort for 2 months, although at a lower dose, therefore differences in duration of treatment may account for the discrepancy.

Importance and management

Direct information about this apparent interaction between theophylline and St John’s wort appears to be limited. Despite the isolated case report of a marked decrease in theophylline levels, no significant pharmacokinetic interaction was noted in healthy subjects, and any pharmacokinetic interaction appears likely to be minor. Mechanistic studies suggest a modest interaction at most. Furthermore, most clinically significant interactions with St John’s wort are mediated by the cytochrome P450 isoenzyme CYP3A4. However, until further evidence is available, it would be prudent to be aware of the possibility of an interaction. Patients should be warned of the possible effects of concurrent use. In 2000, the CSM in the UK recommended that patients taking theophylline should not take St John’s wort. In those patients already taking the combination, the St John’s wort should be stopped and the theophylline dosage monitored and adjusted if necessary. However, this guidance was issued before the pharmacokinetic study that suggests that an interaction is generally unlikely.

Note that St John’s wort does not appear to interfere with laboratory assays for theophylline, see St John’s wort + Laboratory tests.

St John’s wort + Tibolone

An isolated case describes liver damage in a woman taking tibolone and St John’s wort.

Clinical evidence

A 57-year-old woman who had been taking tibolone 2.5 mg daily for the past 2 years for postmenopausal symptoms, and hydroxychloroquine sulfate 200 mg daily for the past 7 years for rheumatoid arthritis, without complaint, was hospitalised for liver damage after taking a 2-g infusion of St John’s wort daily for 10 weeks for mild depression. The patient was suffering from fatigue, reduced appetite and jaundice. Her liver function normalised after about one year of taking ursodesoxycholic acid 250 mg twice daily.

Experimental evidence

No relevant data found.

Mechanism

Unknown.

Importance and management

The general clinical importance of this isolated report is uncertain. Both tibolone and hydroxychloroquine sulfate have been associated with liver toxicity alone but cases with hydroxychloroquine sulfate are quite rare. Therefore the authors of the report suggest that an interaction between tibolone and St John’s wort was to blame for the liver damage in this case; however, both drugs may cause liver damage alone. Both tibolone and St John’s wort are widely used long term, which suggests that this interaction is not common. Nevertheless, it may be prudent to be aware of a possible interaction if symptoms of liver toxicity (fatigue, reduced appetite, dark urine) become apparent.

St John’s wort + Tricyclic antidepressants

The plasma levels of amitriptyline and its active metabolite, nortriptyline, are modestly reduced by St John’s wort.

Clinical evidence

Twelve depressed patients were given amitriptyline 75 mg twice daily and St John’s wort extract (Jarsin, Lichtwer Pharma) 900 mg daily for at least 14days. The AUC0-12 of amitriptyline was reduced by about 22% and the AUC of nortriptyline (its metabolite) was reduced by about 41%.

Experimental evidence

No relevant data found.

Mechanism

Not fully understood. St John’s wort is known to induce the activity of the cytochrome P450 isoenzyme CYP3A4, which is a minor route of metabolism of the tricyclic antidepressants. However, the tricyclics are predominantly metabolised by CYP2D6, so an effect on CYP3A4 is unlikely to lead to a clinically relevant reduction in their levels. Induction of P-glycoprotein by St John’s wort may also contribute; however, the extent of its involvement in the transport of the tricyclics is unclear.

Importance and management

The evidence for an interaction is limited to this study, and based on the minor reduction in amitriptyline levels seen, it seems unlikely that a clinically significant reduction in efficacy would occur. Other tricyclics would be expected to interact similarly.

Both the tricyclics and St John’s wort are antidepressants, but whether concurrent use is beneficial or safe is not known, and it was not assessed in this study. Further study is needed.

Note that St John’s wort does not appear to interfere with laboratory assays for tricyclics, see St John’s wort + Laboratory tests.

St John’s wort + Triptans

Serotonin syndrome has been reported in a patient taking eletriptan and St John’s wort.

Clinical evidence

A 28-year-old woman who had been taking fluoxetine 60 mg daily for one year for an eating disorder, and St John’s wort (dose and frequency not stated) for one month, suffered a loss of consciousness, convulsions and mental confusion after eletriptan 40 mg daily was started 3 days earlier for a recurrent migraine. Previous use of eletriptan and fluoxetine had not resulted in any reported adverse effects. After admission to hospital, the patient developed acute rhabdomyolysis and transient mild acute renal failure. Serotonin syndrome was diagnosed, all medications were stopped and the symptoms gradually resolved over 10 days.

Experimental evidence

No relevant data found.

Mechanism

Additive serotonergic effects are the likely explanation for the case report above as serotonin syndrome has been reported with both St John’s wort and with triptans alone.

Importance and management

The CSM/MCA in the UK note that potentiation of serotonergic effects have been identified between triptans and St John’s wort, leading to an increased risk of adverse effects, and they advise that patients taking triptans should not take St John’s wort preparations. However, most UK manufacturers of triptans simply warn about the potential increase in undesirable effects. The possible concern is that concurrent use may result in the development of serotonin syndrome.

St John’s wort + Voriconazole

St John’s wort more than halves the AUC of a single dose of voriconazole.

Clinical evidence

In a study in 17 healthy subjects, a single 400-mg dose of oral voriconazole was given alone and on the first and last day of St John’s wort (Jarsin, Lichtwer Pharma) 300 mg three times daily for 15 days. Taking St John’s wort for one day had no effect on the voriconazole AUC0-∞, but slightly increased the maximum serum level and AUC0-10 by 22%. However, when voriconazole was given on day 15 of treatment with St John’s wort, the AUC of voriconazole was decreased by 59% and there was a 2.4-fold increase in oral clearance.

Experimental evidence

No relevant data found.

Mechanism

These results suggest that the short-term effect of St John’s wort is to slightly enhance the absorption of voriconazole, whereas the longer-term effect is to induce absorption-limiting transport proteins, such as P-glycoprotein, and intestinal metabolism via cytochrome P450 isoenzyme CYP3A4.

Importance and management

The slight increase in voriconazole absorption with a single dose of St John’s wort is not clinically relevant. However, the reduction in voriconazole levels after 15 days of St John’s wort could impact on clinical efficacy. This suggests that patients requiring voriconazole should be asked about current or recent use of St John’s wort, since this may indicate the need to use an increased voriconazole dose, at least initially. Patients taking voriconazole should be advised not to take St John’s wort. The manufacturers of voriconazole specifically contraindicate concurrent use of St John’s wort.

St John’s wort + Warfarin and related drugs

St John’s wort can cause a moderate reduction in the anticoagulant effects of phenprocoumon and warfarin.

Clinical evidence

(a) Phenprocoumon

In a randomised, placebo-controlled, crossover study in 10 healthy men, St John’s wort extract (LI 160, Lichtwer Pharma) 900 mg daily for 11 days reduced the AUC of a single 12-mg dose of phenprocoumon by a modest 17.4%. There is also a case report of a 75-year-old woman taking phenprocoumon who had a reduced anticoagulant response (a rise in the Quick value) 2 months after starting to take St John’s wort.

(b) Warfarin

In a randomised, crossover study in 12 healthy subjects, one tablet of St John’s wort three times daily for 3 weeks modestly decreased the AUC of both R- and S-warfarin by about 25% after a single 25-mg dose of warfarin taken on day 14. In this study, the brand of St John’s wort used was Bioglan tablets, each tablet containing an extract equivalent to 1 g of Hypericum perforatum flowering herb top containing 825 micrograms of hypericin and 12.5 mg of hyperforin.

The Swedish Medical Products Agency received 7 case reports over the 1998 to 1999 period of patients stabilised on warfarin who showed decreased INRs when St John’s wort was added. Their INRs fell from the normal therapeutic range of about 2 to 4 to about 1.5. Two patients needed warfarin dosage increases of 6.6% and 15%, respectively, when St John’s wort was added. The INRs of 4 patients returned to their former values when St John’s wort was stopped.

Experimental evidence

No relevant data found.

Mechanism

Uncertain, but it is suggested that the St John’s wort increases the metabolism and clearance of the anticoagulants possibly by induction of cytochrome P450 isoenzyme CYP3A4, and possibly also CYP2C9, as both R- and S-warfarin were affected. However, note that St John’s wort had no effect on the metabolism of tolbutamide, which is commonly used as a probe substrate for CYP2C9 activity. See St John’s wort + Antidiabetics.

Importance and management

Information seems to be limited to these reports, but a modest pharmacokinetic interaction is established, which might be clinically important in some patients. It would be prudent to monitor the INRs of patients taking phenprocoumon, warfarin or any other coumarin if they start taking St John’s wort, being alert for the need to slightly raise the anticoagulant dosage. However, note that the advice of the CSM in the UK is that St John’s wort should not be used with warfarin. They note that the degree of induction of warfarin metabolism is likely to vary because levels of active ingredients may vary between St John’s wort preparations. If a patient is already taking the combination, they advise checking the INR, stopping the St John’s wort and then monitoring the INR closely and adjusting the anticoagulant dosage as necessary.