ArticlePDF AvailableLiterature Review

Herb-Drug Interactions: An Evidence Based Approach

  • The Daffodil Centre - University of Sydney - a joint venture with Cancer Council NSW

Abstract and Figures

The increasing use of herbal medicinal products (HMPs) in the community where people are also receiving prescription medicines suggests that adverse herb-drug interactions may be of significant public health consequence. The evidence available to guide practitioners in decision making is complex and consists of a range of sources including adverse event database entries, spontaneous or case reports, in vivo and in vitro drug metabolism studies, and in vivo drug interaction studies in healthy subjects and patients. In the absence of further rigorous studies to assess the clinical significance of herb-drug interactions, an evidence-based appraisal of the current literature is essential to guide practitioners involved in patient care.
Content may be subject to copyright.
Current Medicinal Chemistry, 2004, 11, 1513-1525 1513
0929-8673/04 $45.00+.00 © 2004 Bentham Science Publishers Ltd.
Herb-Drug Interactions: An Evidence Based Approach
P.D. Coxeter, A.J. McLachlan
, C.C. Duke and B.D. Roufogalis
Herbal Medicines Research and Education Centre, Faculty of Pharmacy, The University of Sydney, NSW 2006,
Abstract: The increasing use of herbal medicinal products (HMPs) in the community where people are also
receiving prescription medicines suggests that adverse herb–drug interactions may be of significant public
health consequence. The evidence available to guide practitioners in decision making is complex and consists
of a range of sources including adverse event database entries, spontaneous or case reports, in vivo and in
vitro drug metabolism studies, and in vivo drug interaction studies in healthy subjects and patients. In the
absence of further rigorous studies to assess the clinical significance of herb-drug interactions, an evidence-
based appraisal of the current literature is essential to guide practitioners involved in patient care.
Keywords: Herb-drug interactions, pharmacokinetics, pharmacodynamics, Hypericum perforatum, Panax ginseng, Ginkgo
biloba, Glycyrrhiza glabra.
The increasing use of herbal medicinal products (HMPs)
in the community where people are also receiving
prescription medicines suggests that adverse herb–drug
interactions may be of significant public health consequence.
Regulatory authorities worldwide (Therapeutic Goods
Administration, U.S. Food and Drug Administration,
Medicines Control Agency and the Medical Products
Agency) have highlighted herb-drug interactions as an
important safety issue. Despite these concerns, the evidence
available to guide practitioners in decision making is
complex, and encompasses a range of sources, inclusive of
adverse event database entries, spontaneous reports or case
reports/series, in vivo and in vitro drug metabolism studies,
and in vivo drug interaction studies in healthy subjects and
While the evidence relating to herb-drug interactions is
growing, interpretation of the clinical significance of
reported events or data to guide practitioners involved in
patient care is inadequate. Distinctions in the literature
between anecdotal reports, or those with a firm
pharmacological or clinical basis, are rare. In instances where
a hierarchy of evidence has been recognised [1], a thorough
critical analysis within each strata of report type has been
lacking. Such considerations are paramount for informing
appropriate, evidence-based, and clinically relevant
management decisions — and separating interaction from
The aim of this literature review is to provide critical
insight and commentary into the issues that need to be
considered in applying evidence based principles to assess
clinically important herb-drug interactions. The available
evidence supporting herb-drug interactions for several
commonly used HMPs, namely St. John’s wort (Hypericum
*Address correspondence to this author at the Faculty of Pharmacy,
Building A15, The University of Sydney, NSW 2006, Australia; Business
Tel: 61 - 2 - 9351 4452; Home Tel: 61 – 2 - 98092697; Fax: 61 - 2 - 9351
perforatum), Ginseng (Panax ginseng), Ginkgo biloba and
Licorice (Glycyrrhiza glabra) will provide salient examples
for discussion. There have been a number of systematic
reviews of the available evidence pertaining to herb-drug
interactions [2-5].
Separating Interaction from Over-Reaction
Unlike interactions between conventional medicines,
herb–drug interactions are difficult to interpret in the clinical
setting. The variable nature of HMPs (related to different
constituents, products, extracts, quality and dosage
regimens) makes it difficult to draw useful comparisons
between clinical studies or case reports. Correct
identification of botanical ingredients is crucial in assessing
the reliability of and generalizing suspected herb-drug
interactions. Adverse events attributed to HMPs where there
has been intentional or accidental substitution, or
contamination of botanical ingredients with a more toxic
botanical, poisonous metal, or potent non-herbal drug
substance, significantly complicates this interpretation [6].
The dose, duration, frequency, and route of administration,
can markedly influence the risk of serious drug interactions.
This is very important, because most drug interactions are
dose and frequency-dependent, and the heterogeneity of
dosing recommendations inherent with herbal extracts or
products can make it difficult to assess the clinical
significance of particular combinations. All of these factors
should be considered when assessing reports of potential
herb-drug interactions. Important criteria for assessing
potential interactions between HMPs and drugs are presented
in Table 1.
The scientific and clinical evidence supporting a herb-
drug interaction can vary significantly. This evidence may be
indirectly based on a mechanistic understanding of the
drug’s (or herbal constituent’s) action, or directly based upon
clinical observation or study outcomes. Controlled clinical
studies in healthy subjects, or in patients receiving a drug as
part of their therapy, may provide an insight into the clinical
1514 Current Medicinal Chemistry, 2004, Vol. 11, No. 11 McLachlan et al.
Table 1. Important Considerations for the Assessment of HMP-drug Interaction Reports
Quality of herb contamination/substitution of herbs or constituents
heterogeneity of active or other constituents
standardisation of constituents
stability and integrity of constituents during processing (e.g. extraction)
Dose variability in individual recommended dose
Dose form extraction procedure or dose form preparation may alter the activity of selected
Duration acute and/or chronic interaction effects
Frequency dosing intervals (multiple or single dosing of herb and/or drug)
Route of administration systemically administered herbs are more likely to lead to serious interactions
Evidence for herb–drug interaction(s) type and quality of studies available to allow informed clinical decision making for potential
Table 2. Comparison of the evidence related to herb-drug interactions
Study design Advantages Disadvantages
Controlled clinical trial in
o High internal validity and generalisability to
patients of interest
o Controlled trial may not reflect clinical reality
o Difficult to implement (logistics, cost, ethics)
o Increased sample size may be required to account for
expected variability
Controlled clinical trial in
healthy subjects
o High internal validity
o Allows a full and rigorous assessment of herb–drug
interaction mechanisms
o As above
o May not predict significance of interaction in patients
with disease, organ dysfunction, and/or other medications
Open label clinical trial o Controlled patient group for comparison o As above
o Confounded by subjective endpoints
Case reports or series o Informative and realistic information obtained in
the patient group of interest
o Generates hypotheses
o Uncontrolled observational study which may
overestimate significance given the many possible
o May lead to over-reporting of cases with no insight into
the clinical significance
Animal studies o Allows a rigorous investigation of integrated
pharmacological effects
o Some interspecies differences in drug response and
metabolism may confound
o Not possible to assess clinical significance
In vitro studies in animal
or human tissues
o Provides mechanistic information and allows
fundamental under-standing of interaction cause
o Isolated tissues may not reflect in vivo response
o Concentrations of constituents may not reflect
concentrations or metabolites in vivo
Adverse event database
o Provides immediate reporting of potentially
dangerous interactions
o Provides an international perspective on the
incidence of serious but rare interactions
o Hypothesis generating
o Often can’t be evaluated due to non-existent or
inaccessible extraneous data for critical interpretation
significance of an interaction. These studies need to be
adequately designed with appropriate numbers of subjects (or
patients) and an apposite control group. Case reports may be
suggestive of a clinically important interaction, however,
they do require adequate disclosure of the extent and nature
of the HMP and its ingestion to better assess the clinical
importance and risk to the patient, or indeed as the
cautionary basis for population alerts. Observational
evidence may also be confounded by other medicines a
patient is receiving or other diseases a patient may have. In
many circumstances there has been no systematic
investigation of the potential interaction between commonly
used HMPs and conventional drugs. Animal or in vitro data
may infer the potential for an interaction but cannot
conclusively inform of significant interactions in humans.
The levels of evidence that may support herb-drug
interactions and their clinical utility are presented in Table 2.
Often the health care practitioner is left to speculate on a
possible or probable interaction between a herb and drug.
While preventive strategies may usefully be employed to
minimise potential herb–drug interactions, overcoming the
deficiencies identified in the current herb–drug literature
should be a priority within the field of complementary
medicine research. Hypericum perforatum (H.perforatum),
Panax ginseng (P.ginseng), Ginkgo biloba (G.biloba) and
Glycyrrhiza glabra (G.glabra) are commonly suspected of
interacting with a spectrum of conventional drugs. Utilising
an evidence-based approach, the existing herb-drug
Herb-Drug Interactions: An Evidence Based Approach Current Medicinal Chemistry, 2004, Vol. 11, No. 11 1515
Table 3. Evidence Related to Hypericum Perforatum-Interaction Studies
Level of evidence Drugs/mechanism Clinical significance
Controlled clinical trial in patients o - o -
Controlled clinical trial in healthy subjects o Phenprocoumon
o Digoxin
o Substrates for P-glycoprotein
o Reduced therapeutic effect
o Reduced therapeutic effect
o Increased expression of P-gp
human study
o Cyclosporin/Tacrolimus
o Indinavir
o Irinotecan
o Amitriptyline
o Oral contraceptives
o Cytochrome P450
(substrates for isoforms 1A2, 2D6, 2E1, 3A4)
o Substrates for P-glycoprotein
o Reduced therapeutic effect
o Reduced therapeutic effect
o Reduced therapeutic effect
o Reduced therapeutic effect
o Contradictory findings
o No effect/CYP enzyme induction
o Inhibition/induction
Case reports
or series
o Cyclosporin/Tacrolimus
o Antidepressants (SSRIs, TCAs)
o Oral contraceptives
o Reduced therapeutic effect
o Potential pharmaco-dynamic effects
o Suspected unwanted pregnancy/
breakthrough bleeding
Animal studies o Cytochrome P450 (substrates for isoforms 1A, 3A,
o Substrates for P-glycoprotein
o No effect/induction
o Increased expression of P-gp
In-vitro studies in animal or human tissue o Cytochrome P450 (substrates for isoforms 1A2,
2C9, 2C19, 2D6, 3A4; Pregnane X receptor
o Substrates for P-glycoprotein
o CYP enzyme inhibition
o Increased expression of P-gp
interaction literature associated with each of these herbs will
be discussed according to therapeutic class in the proceeding
sections with a particular emphasis on their clinical
Hypericum Perforatum-Drug Interactions
Government warnings [7] on interactions between
H.perforatum and drugs reflect a recent intensification in the
scientific literature of published case reports, in vitro and in
vivo investigations, monographs and reviews. These
publications have implicated a broad range of drug classes.
However, in many instances, the evidence from which to
establish credible clinical guidelines is unconvincing and not
systematically applied. The following section examines the
evidence for clinically relevant H.perforatum interactions,
and highlights the need for their full assessment to guide the
quality use of complementary and conventional medicines.
The evidence related to H.perforatum–drug interactions is
summarised in Table 3.
In an open label experiment, 11 renal transplant patients
received H.perforatum (600 mg daily) for 14 days
concomitantly with cyclosporin. A significant decrease
(46%) in cyclosporin plasma concentrations and its
metabolites were observed. Cyclosporin dose adjustment
was required from 2.7 mg/day at baseline to 4.2 mg/day at
day 15, with the first dose adjustment required just 3 days
after H.perforatum co-medication [8]. Fifty-five cases of
suspected interactions between H.perforatum and
cyclosporin have been described [9-19]. In all cases, co-
medication with H.perforatum was associated with a
significant reduction in cyclosporin whole blood trough
levels below therapeutic goals. Organ rejection was
confirmed in four cases. No further evidence from controlled
clinical trials currently exists to support these findings, and
the quality of H.perforatum-containing products were often
not reported.
Ten renal transplant patients, stable on their usual
tacrolimus and mycophenolic acid regimen, concomitantly
received H.perforatum (Jarsin300®, 600 mg daily) for 14
days in an uncontrolled open label study. A decrease (>50%)
in the bioavailability of tacrolimus (and not mycophenolic
acid) was observed after 2 weeks of H.perforatum co-
medication, requiring significant adjustment to tacrolimus
dose (mean 4.5 mg/day at baseline to 8.0 mg/day with co-
treatment) [20]. A single case of an interaction with
tacrolimus has been reported [21].
A controlled clinical study [22] investigated the effects of
H.perforatum (3 x 300 mg tablets standardised to 0.3%
hypericin) and indinavir co-medication in eight healthy
volunteers. The area under the curve (AUC; a measure of
drug exposure) of indinavir was reduced by 57%, and the
1516 Current Medicinal Chemistry, 2004, Vol. 11, No. 11 McLachlan et al.
extrapolated 8-h indinavir trough levels by 81%, after 14
days administration of H.perforatum. To date, there is no
report to support these findings.
Anticancer Drugs
The effects of 14 days pre-treatment and 4 days co-
medication of H.perforatum (300 mg 3 times daily) on the
metabolism of irinotecan was investigated in an unblinded,
randomised, crossover study design in 5 cancer patients. The
plasma concentration of the active metabolite of irinotecan
(SN-38), responsible for the drug’s chemotherapeutic effect,
was reduced by 42% (range 14 - 70%).
Interactions between H.perforatum and digoxin have
been investigated in a placebo-controlled, parallel study
design in 25 healthy volunteers [23]. Effects of single and
multiple-dose (for 10 days) administration of H.perforatum
(3 x 300 mg LI160 extract) on digoxin pharmacokinetics
were considered. Digoxin area-under-the concentration-time
curve (influence by clearance and bioavailability) decreased
by 25% in subjects receiving multiple doses of
H.perforatum. H.perforatum led to a reduction in trough
concentrations (33%) and peak plasma concentrations (26%)
of digoxin. Currently, there are no reports to support these
In an open-label clinical study on 12 patients with
depression [24], the pharmacokinetics of amitriptyline
following multiple dosing (14 days) were compared before
and after dosing with H.perforatum (14 days). Significant
reduction in the AUC of amitriptyline (22%) and
nortriptyline (41%) was observed, in addition to all but one
of the measured hydroxylated metabolites. There is no
evidence to correlate these pharmacokinetic changes with
clinical outcomes. Pharmacodynamic interactions between
H.perforatum and tricyclic antidepressants and selective
serotonin re-uptake inhibitors have been proposed. A
published case series [25] (n=4) presents a possible
pharmacodynamic interaction between H.perforatum (600–
900 mg of unspecified H.perforatum preparations) and
sertraline. In elderly patients, symptoms were identified to
be consistent with serotonin syndrome. The symptoms
corresponded to treatment with H.perforatum. A single case
involving H.perforatum and nefazodone suggesting a
pharmacodynamic interaction has also been reported [25].
Another case reported the interaction of H.perforatum with
fluoxetine [26], in which the patient developed a mixed
hypomanic episode. A clinical syndrome resembling
sedative/hypnotic intoxication has been observed in one case
following co-medication of H.perforatum and paroxetine
[27]. There has been no systematic investigation of
pharmacodynamic interactions between H.perforatum and
Oral Contraceptives
One clinical study [28] investigated the effects of
H.perforatum on the pharmacokinetics of a combination
low-dose OC (Lovelle) containing ethynylestradiol and 3-
ketodesogestral. Healthy female volunteers (n=16) took
Lovelle for three months prior to the study and for the study
period (21 days). H.perforatum (2 x 250 mg dry extract
Ze117, drug extract ratio 4-7:1, ethanol 50% v/v,
standardised hypericin content 0.2% and hyperforin ≤ 0.2%)
was co-administered for 14 days. The authors concluded that
H.perforatum did not interact with the OC steroids. In 12
healthy female volunteers, long term administration of
H.perforatum (unknown extract, 3 x 300 mg daily)
significantly increased the oral clearance of norethindrone,
with 7 of 12 volunteers experiencing related breakthrough
bleeding (compared to 2 of 12 volunteers prior to SJW
dosing interval) [29]. Only one case report of an unwanted
pregnancy possibly associated with H.perforatum co-
medication has been reported [30]. Reports of breakthrough
bleeding and unexpected pregnancy have been recorded by
German and Swiss authorities [31,32], but original reports
are not being critiqued. Despite conflicting clinical evidence
and inadequately reported cases the interaction between
H.perforatum and OCs is still widely reported.
A placebo-controlled study investigated the effects of
H.perforatum on phenprocoumon [33]. Following 10 days’
pre-treatment with H.perforatum, the AUC of
phenprocoumon was reduced by 17%. Widely cited reports
[32,34] of cases (including warfarin) submitted to official
Swedish and German adverse event databases are unable to
be validated. Jiang et. al., [35] investigated the effect of
H.perforatum on warfarin pharmacokinetics and
pharmacodynamics in a randomised cross-over clinical trial
in healthy male subjects. The trial examined the effect of 14
days pre-treatment with Commission E recommended doses
of H.perforatum in a single ingredient HMP (whose quality
was assessed against the British Pharmacopeial monograph).
The major finding was that H.perforatum reduced the
effectiveness of warfarin (as monitor by INR) by induction
of the metabolism of warfarin enantiomers. H.perforatum
did not have a direct effect on the clotting status of healthy
subjects [35]. While this finding provides evidence to
support the recommendation to monitor INR in patients
receiving this herb-drug combination, further research is
needed to clarify the implications for elderly patients (who
are likely to receive warfarin).
Cytochrome P450 and P-Glycoprotein Substrates
H.perforatum, either as commercially prepared extracts
and/or purified components, has been found to inhibit
several cytochrome P450 (CYP 450) enzymes in vitro [36-
39]. In vivo studies, however, have either failed to find
significant effects on CYP isoenzymes in animal [40] or
human subjects [41-43], or confirmed that H.perforatum is a
potent inducer of intestinal and hepatic CYP enzymes
[16,17,19,22,43-46]. Induction of these enzymes can reduce
the effectiveness of co-administered drugs that are known
substrates for these enzymes. Studies utilising short
H.perforatum dosing regimens have generally failed to
observe effects on CYP enzymes [42,43]. A recent study
utilising improved methodology has confirmed no effect on
CYP isozymes following single-dose H.perforatum, but
Herb-Drug Interactions: An Evidence Based Approach Current Medicinal Chemistry, 2004, Vol. 11, No. 11 1517
Table 4. Evidence Related to Ginseng-drug Interaction Studies
Level of evidence Drug/mechanism Clinical significance
Controlled clinical trial in patients o - o -
Controlled clinical trial in healthy subjects o Warfarin o No effect
human study
o Cytochrome P450 substrates o No effect
Case reports
or series
o Digoxin (or digoxin assay)
o Phenelzine
o Warfarin
o Unclear, apparent assay interference
o Possible manic symptoms
o Decreased INR
Animal studies o Cisplatin/Mytomycin C
o Warfarin
o Possible synergistic effect
o No effect
In-vitro studies in animal or human tissue o Mytomycin C
o P-glycoprotein substrates
o Cytochrome P450 substrates
o Warfarin
o Possible synergistic effect
o Modulation of MDR
o Conflicting results
o Anticoagulant effect
significant and selective induction of intestinal CYP3A
activity following chronic (14 days) dosing of the herb [47].
H.perforatum has been found to activate Pregnane X
Receptor (PXR) [48,49]. Hyperforin was identified as a
potent PXR ligand [48] and responsible for mediating PXR
induction [49]. This activity leads to increased expression of
CYP3A4 and an increase in metabolism CYP3A4 substrates
(e.g. cyclosporin). H.perforatum has been shown to alter
expression and/or function of P-glycoprotein (P-gp) in
animal [45], and human [45,50] subjects, resulting in
reduced drug concentrations of drugs such as digoxin. This
effect on P-gp following chronic H.perforatum and hypericin
(another constituent of H. perforatum) exposure has been
observed in vitro [51], though interestingly, the authors
noted mild inhibition of P-gp following acute H.perforatum
dosing. Inhibitory and inductive effects of H.perforatum on
P-gp following acute and chronic dosing, respectively, have
been validated by a clinical study [52]. Mechanistic
investigation of H.perforatum–drug interactions is presently
inconclusive. Variation in study design, H.perforatum
containing product or purified constituent tested, and
H.perforatum dosing intervals are likely to contribute to
discrepant findings.
Activity of Hypericum Perforatum Constituents
Constituent variability in the extracts of H.perforatum
employed in herb–drug interaction studies may account for
variation in study findings. H.perforatum products are
commonly standardised to a fixed content of hypericin [37],
the constituent initially thought responsible for the herb’s
antidepressant activity [53]. Evidence now suggests that
hyperforin may be the primary contributor to this activity
[32,48,53], though an in vivo animal study has shown no
change in antidepressant effect of a H. perforatum extract
following exclusion of hyperforin [54]. Hyperforin is also
currently considered the main contributor to H.perforatum
drug interactions [28].
Adverse interactions between H.perforatum and many
drugs are widely reported. Despite a clear risk of serious and
life threatening H.perforatum-drug interaction effects in
discrete patient groups, the evidence supporting negative
health outcomes associated with many reported drugs is
compromised by variance in the levels of presented evidence
and inconsistent findings. The clinical significance of
H.perforatum-drug interactions most appropriately should be
based on careful appraisal of the literature (including study
design and H.perforatum formulation), and the potential risk
to patients by considering their characteristics and the safety
margin of the interacting drugs.
‘Ginseng’-Drug Interactions
Considerable speculation regarding drug interactions with
‘ginseng’ frequents the literature. ‘Ginseng’ refers to several
related (Panax ginseng M., Panax quinquefolius L., Panax
notoginseng B., Panax japonicus M.), or distinct
(Eleutherococcus senticosus) botanical species with
comparable traditional indications. A summary of the
evidence supporting ginseng-drug interactions with an
emphasis on correct botanical identification is presented
below and in Table 4.
Cardiovascular Drugs
Elevation of serum digoxin levels was associated with
co-ingestion of Siberian ginseng (Eleutherococcus
senticosis) by a 74 year old man [55]. Digoxin levels
remained high despite discontinuation of digoxin, but
returned to normal when E.senticosis was ceased. Analysis
confirmed the absence of digoxin or digitoxin in the
E.senticosis preparation [56]. In vivo conversion of
eleutherosides (glycosides with aglycones related to cardiac
glycosides), present in E.senticosis, into digoxin, or
impairment of the renal elimination of digoxin, was
excluded in the absence of drug toxicity signs. It was
concluded that constituents of E.senticosis may have
interfered with the digoxin assay. The interaction between
ginseng and anti-hypertensives is unclear. Two case reports
inadequately describe hypertension following ingestion of
‘ginseng’ products [57,58]. Clinical studies report that
ginseng has been shown to both normalise high and low
1518 Current Medicinal Chemistry, 2004, Vol. 11, No. 11 McLachlan et al.
blood pressure [59]. A similar paradox occurred in an animal
study, where ginseng administered intravenously induced
significant hypotension (and bradycardia), and the
vasoconstriction of renal, mesenteric and femoral arteries at
higher doses [60]. Individual ginseng saponins have
exhibited similarly contradictory effects on cardiac
haemodynamics [61].
Ginseng has been speculated to stimulate breast cancer
growth and counter the actions of oestrogen receptor
antagonists (e.g. tamoxifen) [62]. Ginseng’s oestrogenic
activity is supported by 7 cases of mastalgia [63,64], 3
examples of uterine bleeding [65,66] and one case of
gynecomastia [67]. The majority of reports lack adequate
detail, or involved patients taking ginseng of unknown
quality, or at higher than recommended dosages. Clinical
studies demonstrate that a standardised ginseng extract did
not effect male or female hormonal status [68]. Furthermore,
no in vitro interaction was observed between ginseng extract
and either cytosolic oestrogen receptors, isolated from
mature rat uterus or progesterone receptors from human
myometrium [68].
Anticancer Drugs
Prolonged survival and synergism with cisplatin
treatment was observed in tumour bearing mice following
oral administration of a ginsenoside [69]. In another study,
survival was significantly greater in tumour bearing mice
orally administered ginseng and exposed to radiation,
compared to those receiving the treatments separately.
Ginseng also appeared to assist the recovery of healthy liver
cells injured from radiation exposure. Other experiments in
mice have shown a reduction in tumours and metastases,
when ginseng was administered in conjunction with
chemical carcinogens [70]. Potentiation of cytotoxic activity
of mitomycin C has been demonstrated in vivo and appears
to occur without increasing host toxicity [69]. These
findings are yet to be confirmed in human trials.
Protopanaxatriol ginsenoside (PTG) constituents of Korean
red ginseng demonstrated a chemosensitising effect on P-
glycoprotein (Pgp)-mediated multidrug resistance cells, by
increasing the intracellular accumulation of drugs via direct
interaction with Pgp. The PTGs were also considered a
beneficial long term adjunct to cancer chemotherapy due to
absence of associated Pgp activation [71].
Insomnia, headache and tremor, followed the ingestion of
ginseng (Natrol High and ginseng tea) and the MAOI
phenelzine by a 64 year-old woman. While it is difficult to
make a full assessment of this case, because a combination
of herbs were ingested, it was noted that symptoms recurred
3 years later following repeat co-medication [72,73]. A 43
year old depressed female experienced manic-like symptoms
after concomitant phenelzine, ginseng and bee pollen
ingestion. Symptoms resolved following discontinuation of
all medications and diminished response to phenelzine was
observed on retrial of the drug [74]. The additive
psychoactive effect of ginseng ginsenosides and the MAOI
was the proposed mechanism. Mania is described in a 35
year old depressed woman, 10 days following the
withdrawal of her usual lithium carbonate and amitriptyline
treatment, and beginning once daily supplementation of
ginseng tablets. Her symptoms improved following
cessation of ginseng and a return to her previous medication
[75]. However, the effect of lithium cessation cannot be
excluded. Irritability and insomnia was reported in 5
schizophrenic patients smoking ginseng-containing
cigarettes. Behaviour improved after smoking was
discontinued [76]. The relevance of this report may be
questioned as the recognised active constituents are not
volatile and readily undergo pyrolysis. ‘Ginseng Abuse
Syndrome’ (nervousness, sleeplessness, hypertension,
diarrhoea, skin eruptions) were described in a study of 133
ginseng users [72], although the study inadequately reported
details relevant to the herb, and unusually large ginseng
doses were being consumed [53,68].
The hypoglycaemic activity of ginseng has been
documented in diabetic and non-diabetic subjects [78,79].
Hypoglycaemic activity occurred independently of dose in
medically (n=7) or diet controlled (n=3) type-2 diabetic
patients. The authors suggested co-administration of ginseng
and hypoglycaemic agents might improve mealtime
glycaemia, and/or create undesired postprandial
hypoglycaemia, however, the patient numbers in this study
are very small preventing generalisation of the results [80].
In 36 NIDDM patients, ginseng (200 mg daily for 8 weeks)
reduced fasting blood glucose and serum aminoterminal
propeptide of type III procollagen concentrations, and
lowered glycated haemoglobin [79]. An antidiabetic effect
was observed in rats following intravenous or intraperitoneal
administration of ginseng polypeptides81. Ginseng extract
increased in the number of insulin receptors in bone marrow,
and reduced glucocorticoid receptors in rat brain homogenate
[81]. Ginsenosides promoted the release of insulin from
isolated rat pancreatic islets [82]. Hypoglycaemic activity of
Siberian ginseng, administered intraperitoneally in mice,
was attributed to the eleutheran (polysaccharide)
constituents. In vitro experiments linked hypoglycaemic
activity to panaxan [68] (not considered active following oral
administration), ginsenoside [82], and polysaccharide
constituents of ginseng. Polysaccharides isolated from
Korean or Chinese ginseng confer greater hypoglycaemic
activity than Japanese species [83]. The conclusion from
examining this evidence is that ginseng appears to alter
glycaemic response, but the herb’s role in the management
of patients with diabetes and the risk of clinically significant
interactions is yet to be clarified.
Decreased INR followed the ingestion of ginseng
capsules by a 47 year-old male patient with a mechanical
heart valve previously stable on warfarin. The INR returned
to the therapeutic range following the herb’s discontinuation
[84]. In a separate case, a 58 year old male with a prosthetic
Herb-Drug Interactions: An Evidence Based Approach Current Medicinal Chemistry, 2004, Vol. 11, No. 11 1519
aortic heart valve developed thrombosis, as a consequence of
subtherapeutic anticoagulation control, temporally associated
with use of an undisclosed commercial ginseng product [85].
This is supported by studies involving ginseng
administration to rats which did not effect warfarin
pharmacokinetics. Vitamin K, which may have decreased
warfarin activity, was not identified in ginseng extracts [86].
Conversely, increased anticoagulant activity was observed in
rats, fed the lipophilic fraction of panax ginseng in corn oil
for 3 weeks, compared with controls [87]. In vitro studies
[88,89] propose that several components of Panax ginseng
(panaxynol and some ginsenosides) can inhibit platelet
aggregation and thromboxane formation and may antagonise
PAF [90]. The mechanism of this effect is unclear, but may
relate to the antiplatelet components of ginseng [91] or
vitamin K antagonism [92]. Adverse bleeding, due to
antithrombotic action of ginseng, has not been reported, and
therefore the clinical significance of the latter observations is
unclear. A randomised cross over study in healthy male
subject by Jiang et. al. [35] investigated the effect of Panax
ginseng on warfarin pharmacokinetics and pharmacodyna-
mics. These researchers used Commission E recommended
doses for 7 days from a HMP of known quality, and found
no effect of the herb on action or disposition in healthy
subjects. In summary, it is appropriate to suggest increased
vigilance on patients receiving anti-coagulants and ‘ginseng’
This recommendation is made, based on the narrow safety
margin of warfarin and the significant interaction potential
for this drug, and is made despite the lack of conclusive
evidence of a herb-drug interaction between warfarin and
Cytochrome P450 Substrates
Data related to the effect of ginseng and its constituents
on drug metabolism enzymes is conflicting. P.ginseng (4%
ginsenosides, 100 mg twice daily for 14 days) did not
induce CYP3A4 as measured by 6-beta-
hydroxycortisol/cortisol ratios in 20 human subjects [94]. E.
senticosus (485 mg twice daily for 14 days) did not
influence the pharmacokinetics of probe substrates for
CYP3A4 and CYP2D6 in 12 healthy volunteers [95].
Standardised ginseng extracts (G115, 4% ginsenosides;
P.quinquefolius, NAGE, 10% ginsenosides) significantly
inhibited CYP1A1, CYP1A2, and CYP1B1 in vitro, with
NAGE found to be 45-fold more potent than G115 in
inhibiting CYP1A2. However, constituent variation between
the extracts may have contributed to this variability. CYP1
inhibition by ginsenosides occurred only at greater
concentrations than in tested extracts [96]. Conversely,
ginsenoside and elutheroside constituents of ginseng did not
significantly inhibit a number of cDNA-expressed
cytochrome P450 enzymes in vitro [97]. ‘Red ginseng’
(steamed P. ginseng) total saponins inhibited cytochrome
P450 activity in rat liver microsomes in a concentration
dependent manner. The inhibitory effect was most marked
for CYP2E1 and CYP3A [98].
Ginseng is widely considered to be safe at recommended
dose regimens, and there are few accounts of ginseng-drug
interactions. The majority of ginseng-drug interactions are
theoretically derived from pharmacological or adverse event
data. Low generalisibility, heterogenous administration
routes, and isolation of single constituents for investigation,
limit the clinical utility available pharmacological and
experimental findings. Inadequate analysis of species or
product authentication, processing method, dose and
duration of ginseng administration is a frequent limitation of
available adverse event and ginseng-drug interaction
Ginkgo biloba – Drug Interactions
G. biloba–drug interaction reports frequent the scientific
literature. Proposed interactions have conferred negative and
positive health outcomes for those concerned. Widespread
caution regarding the herb’s safety, however, appears
founded on limited evidence from either isolated case reports
of spontaneous bleeding in the absence of concomitant drug
therapy, or extrapolation from in-vitro findings. Few
genuine G.biloba–drug interaction reports or studies are
extant in the literature. The evidence for G.biloba-drug
interactions is summarised in Table 5.
Spontaneous and/or protracted bleeding has been
observed in several patients taking G.biloba products [99-
101]. While these reports represent adverse events, rather
than genuine drug interactions, they provide theoretical
support for possible interactions with anticoagulant,
antiplatelet or antithrombitic agents. Interestingly, these
cases have also attracted criticism [102,103] for inadequately
demonstrating a causal association with the herb’s ingestion.
Moreover, the type of G.biloba product (and its
constituents) were either not reported or unclear. Ginkgolide
constituents of G.biloba (particularly ginkgolide B) have
demonstrated potent and specific platelet activating factor
(PAF) inhibition, based on in vitro studies [104]. A
ginkgolide mixture (BN 52063) was found to significantly
inhibit PAF-induced platelet aggregation in six healthy
volunteers [105], although no clinical effects were observed
[106]. A G.biloba extract (EGb761) did not effect bleeding
times in healthy subjects [107]. Relatively few cases of herb-
drug interactions involving G.biloba have been published.
One case involved left parietal haemorrhage [108] in a 78-
year-old woman, who had previously been stable on warfarin
for five years. She began co-medicating with G.biloba
(extract type, daily dose and regimen not detailed) two
months prior to her haemorrhagic event. G.biloba extracts
have been reported to inhibit S-warfarin metabolism (via the
enzyme CYP2C9) by human liver microsomes [109].
Ginkgolic acids I and II have demonstrated significant and
potent inhibition of one or more cDNA-expressed P450
isozymes (particularly CYP 1A2, 2C9, and 2C19) in vitro
[39]. However, no effect on CYP activity (assessed using a
range of specific substrates) was observed, in a clinical study
of 12 healthy volunteers, following 28 days of G.biloba
administration [47]. Another case described spontaneous
hyphema [110] confirmed in a 70-year-old male taking
aspirin (325 mg daily) for three years following heart bypass
surgery. Self-medication with G.biloba (80 mg daily, 50:1
1520 Current Medicinal Chemistry, 2004, Vol. 11, No. 11 McLachlan et al.
Table 5: Evidence Related to Ginkgo biloba - drug Interaction Studies
Level of evidence Drug/mechanism Clinical significance
Controlled clinical trial in patients o Haloperidol
o Antidepressants
o May enhance efficacy
o No benefit on drug-
induced sexual dysfunction
Controlled clinical trial in healthy
o Glycaemic control
o Cytochrome 450
o May interfere with diabetic screening
o Effect on PAF antagonism inconclusive
o No effect in vivo
human study
o Antidepressants o Effect on drug induced sexual dysfunction
Case reports
or series
o G.biloba alone/
warfarin, aspirin,
o Trazodone
o Sodium valproate
o Spontaneous bleeding
o Coma
o Loss of seizure
Animal studies o Amikacin
o Ticlopidine
o Potentiates
ototoxicity of
o Enhances
In-vitro studies in animal or human
o Cytochrome P450
o GABA/glutamic
o PAF membrane antagonism
o Potent inhibition of CYP 450 isozymes
o Flavonoids exhibit GABAergic activity
herb extract) occurred one week prior to the onset of
symptoms. Recently, the effect of G.biloba co-
administration on warfarin effects (assessed by monitoring
INR) was investigated using a randomized, double-blind
placebo-controlled, cross-over trial in patients taking 100 mg
of G.biloba extract daily over four weeks. These researchers
concluded that ginkgo did not influence warfarin
pharmacodynamics at this dose [121].
G.biloba (EgB 761) and ticlopidine co-treatment was
compared to ticlopidine alone in normal and thrombosis-
induced rats [111]. EGb761 (20 or 40 mg/kg) and
ticlopidine (50 mg/kg) co-medication increased antiplatelet
effects, prolonged bleeding time by approximately two-fold,
and increased recovery from acute thrombotic challenge
compared to ticlopidine alone. In an ex-vivo model, co-
medication for nine days decreased the weight of thrombus
formed, and reduced adenosine diphosphate-induced platelet
aggregation. The potential for this treatment combination to
reduce side effects and therapeutic dose of the ticlopidine
alone was postulated.
Loss of seizure control occurred [112] in two patients
well-controlled on sodium valproate. Both patients began
ingesting G.biloba (120 mg daily) in the two weeks prior to
the reoccurrence of seizures. The patients remained seizure-
free on their usual medication following cessation of
G.biloba. A neurotoxin, 4-O-methylpyridoxine, (which may
be present in leaf and seed preparations, although are
unlikely to exist in clinically significant amounts in
commercial extracts) has been implicated in cases of
convulsions and loss of consciousness [113]. Prior
administration of G.biloba to mice resulted in increased the
frequency of seizures provoked by picrotoxin, and reduced
the protective effect of sodium valproate and carbamazepine
[114,115]. G.biloba may also alter electro-encephalograph
activity, reducing
activity. The clinical relevance of this
observation is unclear [112].
Coma is documented in an 80-year-old female with
Alzheimer’s disease three days after being prescribed
G.biloba (EGb761, 80 mg bid) and trazodone (20 mg bid)
[116]. The coma was immediately reversed by flumazenil, a
benzodiazepine antagonist. Flavonoids (also present in
G.biloba) have exhibited GABAergic activity as partial
agonists at benzodiazepine-binding sites. Flavonoids may
also increase CYP3A4 activity. Trazodone is metabolised
into an active metabolite by CYP3A4, which enhances the
release of GABA through agonistic action on presynaptic
Herb-Drug Interactions: An Evidence Based Approach Current Medicinal Chemistry, 2004, Vol. 11, No. 11 1521
Table 6. Evidence Related to G. Glabra-drug Interaction Studies
Level of evidence Drug/mechanism Clinical significance
Controlled clinical trial in patients o - o -
Controlled clinical trial in healthy subjects o - o -
human study
o Prednisolone
o Aspirin
o Antibiotics
o Potentiates drug action
o Protects gastric mucosa
o Decreases side effects
Case reports
or series
o Diuretics
o Cardiac glycosides
o Antidiabetics
o ‘Serious adverse effects’
o Congestive heart failure
o Decrease blood glucose levels
Animal studies o NSAIDs/aspirin
o OC’s
o Anti-diabetics
o Anti-thrombotics
o Protects gastric mucosa
o Anti-oestrogen effects
o Inhibits aldose reductase
o May increase bleeding time
In-vitro studies in animal or human tissue o Anti-diabetics
o Anti-thrombotics
o Anti-depressants
o Inhibits aldose reductase
o May increase bleeding time
o MAO inhibitor
receptors 5-HT2 and α-2 located on GABAergic nerve
terminals [117]. The combination of these factors may
provide an explanatory mechanism for the case [116].
Aminoglycoside Antibiotics
In a controlled animal experiment [118], G.biloba
(EGb761, 100 mg/kg/day) was expected to reduce amikacin
toxicity in adult rats. Distortion product oto-acoustic
emissions instead confirmed that EGb761 facilitated the
development of amikacin toxicity through an earlier, and
more significant, ototoxic effect.
An open-label clinical study [119] investigated the role
of PAF in pancreatic β-cell insulin secretory response to a
glucose load in 20 healthy, glucose-tolerant volunteers
before and after three months’ administration of G.biloba
(120 mg 50:1 standardised extract). The herb was found to
accelerate pancreatic β-cell function. The author alerts of the
potential for chronic G.biloba ingestion to induce
hyperinsulinaemia, and alter Oral Glucose Tolerance Test
results during diabetes screening. The clinical significance of
this finding is unclear.
The effects of G.biloba and haloperidol co-treatment on
schizophrenia symptomology and blood superoxide
dismutase (SOD) levels (higher in the blood of
schizophrenic patients than the general population) were
evaluated in a clinical study [120]. Eighty-two confirmed
schizophrenic patients participated in this 12-week, parallel-
group, placebo-controlled, trial. Responding to a validated
positive and negative symptom questionnaire, statistically
significant improvement was observed in the co-treatment
for both assessments. Controls showed significant
improvement in the negative assessment only. SOD levels
were significantly reduced only in the co-treatment group.
Further research is needed to assess the possible clinical
implications of this observation.
The widespread reporting of G.biloba–drug interactions
appears exaggerated based upon the available evidence. The
conflicting observations, regarding the possible effects of
ginkgo constituents on cytochrome P450, could be a
consequence of variability in the concentration of different
constituents of ginkgo used in different studies. For
example, based on Commission E conformity, the widely
used EGb 761 ginkgo extract contains less than 0.5 ppm
ginkgolic acid, a constituent known to influence CYP
activity in vitro [39]. However, caution and close
monitoring of patients co-medicating with G.biloba is
appropriate. Further research to evaluate the clinical
significance of G.biloba–drug interactions, with respect to
beneficial and adverse effects is required.
Glycyrrhiza Glabra-Drug Interactions
Sodium and water retention, potassium loss and
suppression of the renin-angiotensin-aldosterone system
symptoms, attributed to G.glabra consumption (from foods,
products, or medicinals) are recurrent in published case
reports, and substantiated by clinical and experimental
observation [53,68,69,122,123]. Understanding the
1522 Current Medicinal Chemistry, 2004, Vol. 11, No. 11 McLachlan et al.
physiological effects of G.glabra has lead to speculation
regarding interactions between G.glabra and several other
drugs. However, there is a conspicuous absence of actual
G.glabra-drug interaction reports in the literature. Table 6
summarises the available evidence related to G.glabra-drug
Cardiovascular Drugs
G.glabra and its constituents reduce sodium and water
excretion, and may compromise the effectiveness of
diuretics, such as spironolactone and amiloride [63,123].
G.glabra ingestion may also increase the risk of developing
hypokalaemia, when taken concomitantly with potassium
depleting diuretics (e.g. thiazides or loop diuretics)
[53,68,69,122,123]. Two unvalidated case reports [69] have
described ‘serious side effects’, resulting from the co-
administration of diuretics with large amounts of G.glabra
confectionary [124,125]. Increased potassium depletion also
potentiates the action of, or increases sensitivity to, cardiac
glycosides [53, 68, 69, 122, 123]. Worsening congestive
heart failure in an elderly man taking digoxin was attributed
to ingesting a total 1200 mg G.glabra from a Chinese
herbal laxative formula (other constituents not identified)
[126]. The fact that these reports involve a combination
product, make it difficult to unambiguously assess these
observations, however, these interactions appear to be due to
inhibition of 11-beta-hydroxysteroid dehydrogenase by
constituents of G.glabra (glycyrrhizin and its aglycone,
glycyrrhetinic acid), leading to cortisol accumulation
producing a mineralocorticoid response [53,68,69,122,123,
127]. Although demonstrating relatively low affinity [128],
G.glabra metabolites may also bind directly to
mineralocorticoid receptors [129]. Furthermore, glycyrrhizin
and glycyrrhetinic acid may displace cortisol from binding
to transcortin [130]. These effects are yet to be rigorously
investigated to assess their clinical significance.
Anti-inflammatory Agents
G.glabra may potentiate the pharmacological effects of
corticosteroids. It is proposed that concomitant use of
G.glabra may reduce the necessary dose, and therefore side-
effects associated with these anti-inflammatory medications
[131]. A clinical study investigating prednisolone
pharmacokinetics in healthy male subjects, with or without
oral pre-treatment with 4 doses of 50 mg G.glabra,
demonstrated inhibition of the drug’s metabolism [132,133].
The effects of several G.glabra containing Chinese and
Japanese herbal mixtures on prednisolone concentrations
appear contradictory [134,135], and the relevance of the
findings with that observed for G.glabra alone is
questionable. Glycyrrhizin and glycyrrhetinic acid have been
observed to potentiate the activity of topical corticosteroids
on human skin [136]. The anti-inflammatory actions of
G.glabra have been attributed to the corticosteroid-like
activity of glycyrrhizin and glycyrrhetinic acid [68]. These
constituents have demonstrated no intrinsic glucocorticoid
action, instead acting indirectly by potentiating the activity
of corticosteroids. Studies in adrenalectomised animals
indicate these effects are due to cortisol [69]. Glycyrrhizin
may directly alter corticosteroid effects by inhibiting
thymolytic and adrenal suppressive effects, while
potentiating anti-inflammatory effects [137]. The observed
anti-thrombin action of glycyrrhizin [138] may also
contribute to anti-inflammatory activity. The
immunostimulating effects of G.glabra constituents,
however, may potentially offset the immunosuppressive
effects of corticosteroid treatment for patient’s with
autoimmune disease or transplant recipients [91].
The potential for G.glabra to reduce NSAID-induced
gastrointestinal ulceration has been investigated. In rats,
intra-peritoneal, intra-duodenal, or oral administration of
G.glabra and its derivatives, inhibited the formation of
gastric ulcers induced by aspirin [139] and ibuprofen
[140,141]. Deglycyrrhizinated G.glabra (DGL) prevented
ulcer development [142], inhibited gastric acid secretion
[143] and protected the gastric mucosa of animals treated
with aspirin and bile [144]. In a short-term study in healthy
male subjects, DGL prevented aspirin-induced gastric
damage [145]. The simultaneous administration of aspirin
and G.glabra was considered necessary to confer protective
activity [144,145]. This effect is thought to be due to
glycoprotein synthesis, in the gastric mucosa, prolonging
the life of epithelial cells and antipepsin activity [140]. This
mechanism is mediated by the inhibitory effects of G.glabra
constituents on enzymes, responsible for the breakdown of
cytoprotective prostaglandins [53, 68, 69, 122, 123]. The
clinical significance and the therapeutic potential of these
observations are yet to be established.
Oral Contraceptives
G.glabra may interfere with oestrogen activity [121], and
has the potential to decrease the effectiveness of oral
contraceptives at high doses [69]. Anti-oestrogenic action for
glycyrrhizin, at high concentrations in animals, is consistent
with antagonism at oestrogen receptors [146]. G.glabra may
influence oestrogen metabolism, causing inhibition, if
oestrogen concentrations are high, and potentiation when
concentrations are low [147]. Isoflavone constituents of
G.glabra have also demonstrated oestrogenic activity [147].
Phytoestrogenic constituents of G.glabra may theoretically
result in symptoms of oestrogen excess [91]. Co-medication
of G.glabra with oral contraceptives may increase sensitivity
to adverse effects of glycyrrhizin (hypertension, oedema and
hypokalaemia). Avoiding excessive doses of G.glabra seems
prudent in women taking oral contraceptives.
Hypoglycaemic Agents
Hypokalaemia is known to aggravate glucose intolerance
[91]. G.glabra ingestion may therefore affect blood glucose
levels, and interfere with hypoglycaemic therapy [121]. In a
patient presenting with hypoglycaemia and myopathy
secondary to G.glabra-induced hypokalaemia, an inverse
relationship was noted between concentrations of fasting
serum glucose and serum potassium [121]. Interestingly, the
G.glabra constituent, isoliquiritigenin, inhibits aldose
reductase, which reduces glucose to sorbitol [148], and has
inhibited sorbitol accumulation in tissues and cells
Herb-Drug Interactions: An Evidence Based Approach Current Medicinal Chemistry, 2004, Vol. 11, No. 11 1523
About 80% of patients previously unable to tolerate the
side effects of streptomycin treatment alone, were able to
persist with drug treatment when co-administered with
G.glabra [150]. Concomitant administration of G.glabra
also significantly reduced gastrointestinal symptoms induced
by nitrofurantoin without altering the drugs antibacterial
effects [151]. Original reports are not available to assess the
clinical significance (or potential) of these findings.
G.glabra has demonstrated inhibition of platelet activity
[152] in vitro and in vivo [153], and may potentially increase
bleeding time. This effect appears due to potent inhibition of
both cyclooxygenase and lipoxygenase enzymes [153].
Antiplatelet activity has also been documented for a 3-
arylcoumarin derivative, GU-7, isolated from G.glabra,
which inhibited platelet aggregation in vitro by increasing
intraplatelet cyclic AMP concentration [154]. The relevance
of these findings is yet to be determined clinically.
G.glabra may have MAO inhibitor activity and
concomitant use with known MAO inhibitors may be
contraindicated [91]. Glycyrrhizin is 10 times more active as
an MAO inhibitor as hypericin (a constituent of
H.perforatum) in vitro [91,155]. The clinical relevance of
this activity is yet to be established.
Cytochrome P450 Substrates
G.glabra extracts and glycyrrhizin altered the expression
of cytochrome P450 in mice, and elevated the metabolism of
substrates for CYP3A [38], and to a lesser extent CYP2B,
CYP1A2 and CYP2A1 [156]. These results were not
replicated in rats following ingestion of a single dose,
however, marked increases were observed following
administration of G.glabra for 4 days [157]. In healthy
subjects, daily administration of an aqueous extract of
G.glabra (1 g) for 7 days was found to have no effect on
midazolam pharmacokinetics, a CYP3A4 substrate [158].
G.glabra may potentially interact with drugs sharing similar
metabolic pathways, and further research is needed to clarify
the significance of these observations.
There is limited clinical evidence to support serious
G.glabra-drug interactions, but clear theoretical evidence for
this potential. The profound effect of G.glabra constituents
on the renin-angiotensin-aldosterone system indicates care is
required, when this herb is ingested by patients receiving
cardiovascular drugs, particularly the elderly. Health
professionals can assess the significance of G.glabra-drug
interactions, by examining the dose and frequency of
G.glabra ingestion, and the medical status of the patient.
Current recommendation for G.glabra ingestion are not to
exceed 50 g (equating to 100 mg glycyrrhizin) for longer
than 6 weeks [68]. Congenital variation of 11-beta-
hydroxysteroid dehydrogenase [159], or prolongated transit
times in frequent G.glabra users (delayed plasma clearance
of glycyrrhetinic acid occurs due to duodenal reabsorption
[160]), however, may contribute to dose and frequency-
dependent variations, and predispose certain individuals to
increased risk. Potentially beneficial G.glabra-drug
interactions are yet to be established. Careful monitoring of
all patients is therefore warranted.
This review has provided an overview of the issues that
contribute to the difficulty in assessing the significance of
herb-drug interactions. Using examples from four commonly
used herbal medicines, we have demonstrated that conclusive
evidence of herb-drug interactions is often lacking, and
where clinical observations have been made or studies
conducted, issues with respect to the type, quality and
content of HMPs is often not described (and may also be the
case for reports pertaining to conventional drug-drug
interactions). The available evidence (especially for
H.perforatum) indicates that herbal medicines have the
potential to cause clinically significant life threatening herb-
drug interactions. It is imperative that rigorous studies are
conducted to assess the mechanism and clinical significance
of potential herb-drug interactions, to guide health care
professionals in the selection, and consumers in the choice
of safe herbal medicines. Until a comprehensive evidence
base for herb-drug interactions is available, it is clear that
patients receiving drugs with a narrow safety margin (e.g.
immunosuppressants, anticoagulants, digoxin and others),
and those who are most at risk of serious drug interactions
(i.e. patients who are elderly, have chronic illness, have
organ dysfunction and those receiving multiple medicines),
should be closely monitored to avoid the possible adverse
consequences of herb-drug interactions.
The authors acknowledge the support of the Vincent
Fairfax Family Foundation and the National Health and
Medical Research Council of Australia.
[1] Brinker, F. Herb Contraindications and Drug Interactions, 2nd ed.;
Eclectic Medical Publications: Oregon, 1998.
[2] Izzo, A.A.; Ernst E. Drugs 2001, 61, 2163-2175.
[3] Ernst E. Perfusion 2000, 13, 4-15.
[4] Ernst E. Perfusion 2000, 13, 60-70.
[5] Thompson, C.J, Pittler, M.H.; Ernst, E. Arch. Intern. Med. 2003,
163, 1371.
[6] Blumenthal, M. Herbal Gram 2000, 49, 52.
[7] Briggs, D.R. Toxicology 2002, 181-182, 565.
[8] Bauer, S.; Stormer, E.; Johne, A.; Kruger, H.; Biudde, K.;
Neumayer, H.; Mai, I. Br. J. Clin. Pharmacol. 2003, 55, 203.
[9] Moschella, C.; Jaber, B. Am. J. Kidney, Dis. 2001, 38, 1105.
[10] Karliova, M.; Treichel, U.; Malago, M.; Frilling, A.; Gerkin, G.;
Broelsch, C. J. Hepatol. 2000, 33, 853.
1524 Current Medicinal Chemistry, 2004, Vol. 11, No. 11 McLachlan et al.
[11] Mai, I.; Kruger, H.; Budde, K.; Johne, A.; Brockmoller, J.;
Neumayer, H.; Roots, I. Int. J. Clin. Pharmacol. Ther. 2000, 38,
[12] Mandelbaum, A.; Pertzborn, F.; Martin-Facklam, M.; Weisel, M.
Nephrol. Dial. Trans. 2000, 15, 1473.
[13] Rey, J.; Walter, G. Med. J. Aust.. 1998, 169, 583.
[14] Turton-Weeks, S.; Barone, G.; Gurley, B.; Ketel, B.; Lightfoot,
M.; Abul-Ezz, S. Progress in Transplantation 2001, 11, 116.
[15] Barone, G.; Gurley, B.; Ketel, B.; Abul-Ezz, S. Transplantation
2001, 71, 239.
[16] Breidenbach, T.; Hoffmann, M.; Becker, T.; Schlitt, H.;
Klempnauer, J. Lancet 2000a, 355, 1912.
[17] Breidenbach, T.; Kliem, V.; Burg, M.; Radermacher, J.;
Hoffmann, M.; Klempnauer, J. Transplantation 2000b, 69, 2229.
[18] Ahmed, S.; Banner, N.; Dubrey, S. J. Heart. Lung. Transplant.
2001, 20, 795.
[19] Ruschitzka, F.; Meier, P.; Turina, M.; Lusher, T.; Noll, G. Lancet.
2000, 355, 548.
[20] Mai, I.; Stormer, E.; Bauer, S.; Kruger, H.; Budde, K.; Roots, I.
Nephrol. Dial. Transplant. 2003, 18, 819.
[21] Bolley, R.; Zulke, C.; Kammerl, M.; Fischereder, M.; Kramer, B.
Transplantation 2002, 73, 1009.
[22] Piscitelli, S.; Burstein, A.; Chaitt, D.; Alfaro, R.; Falloon, J. Lancet
2000, 355, 547.
[23] Johne, A.; Brockmoller, J.; Bauer, S.; Maurer, A.; Langheinrich,
M.; Roots, I. Clin. Pharmacol. Ther. 1999, 66, 338.
[24] Johne, A. J. Clin. Psychopharmacol. 2002, 22, 46.
[25] Lantz, M.; Buchalter, E.; Giambanco, V. J. Geriatric. Psych.
Neurol. 1999, 12, 7.
[26] Spinella, M.; Eaton, L. Brain. Injury 2002, 16, 359.
[27] Gordon, J. Am. Family. Phys. 1998, 57, 950.
[28] Will-Shahab, L.; Bauer, S.; Kunter, U.; Roots, I.; Brattstrom, A. [in
[29] Hall, S.D.; Wang, Z.; Huang, S.M.; Hamman, M.A.; Vasavada,
N.; Adigun, A.Q.; Hilligoss, J.K.; Miller, M.; Gorski, J.C. Clin.
Pharmacol. Ther. 2003, 74, 525.
[30] Schwarz, U.; Buschel, B.; Kirch, W. Br. J. Clin. Pharmacol. 2003,
55, 112.
[31] Ernst, E. Lancet 1999, 354, 2014.
[32] Schultz, V. Phytomed. 2001, 8, 152.
[33] Maurer, A.; Johne, A.; Bauer, S. Eur. J. Clin. Pharmacol. 1999,
55, A22.
[34] Yue, Q.; Bergquist, C.; Gerden, B. Lancet 2000, 355, 576.
[35] Jiang, X.; Williams, K.C.; Liauw, W.; Ammit, A.J.; Roufogalis,
B.D.; Duke, C.C.; Day, R.O.; McLachlan, A.J. Br. J. Clin.
Pharmacol. 2004 (Online publication date; 3-Feb-2004).
[36] Carson, S.; Hill-Zabala, C.; Roberts, S.; Hawke, R. Clin.
Pharmacol. Ther. 2000 [abstract].
[37] Obach, R. J. Pharmacol. Exp. Ther. 2000, 294, 88.
[38] Budzinski, J. W.; Foster, B. C.; Vandenhoek, S.; Arnason, J. T.
Phytomed. 2000, 7, 273.
[39] Zou, L.; Harkey, M.; Henderson, G. LIfe. Sci. 2002, 71, 1579.
[40] Noldner, M.; Chatterjee, S. Pharmacopsych 2001, 34, S108.
[41] Markowitz, J.S.; Donovan, J.L.; DeVane, C.L.; Taylor, R.M.;
Ruan, Y.; Wang, J.S.; Chavin, K.D. JAMA 2003, 290, 1500
[42] Wang, Z.; Gorski, J.C.; Hamman, M.A.; Huang, S.M.; Lesko, L.J.;
Hall, S.D. Clin. Pharmacol. Ther. 2001, 70, 317
[43] Markowitz, J. LIfe. Sci. 2000, 66, 133.
[44] Roby, C.; Anderson, G.; Kantor, E. Clin. Pharmacol. Ther. 2000,
67, 451.
[45] Durr, D.; Stieger, B.; Kullak-Ublick, G.; Rentsch, K.; Steinert, H.;
Meier, P.; Fattinger, K. Clin. Pharmacol. Ther. 2000, 68, 598.
[46] Bray, B.; Perry, N.; Menkes, D.; Rosengren, R. Toxicol. Sci. 2002,
66, 27.
[47] Gurley, B. J.; Gardner, S. F.; Hubbard, M. A.; Williams, D. K.;
Gentry, W. B.; Cui, Y.; Ang, C. Y. Clin. Pharmacol. Ther. 2002,
72, 276.
[48] Moore, L.; Goodwin, B.; Jones, S.; Wisely, G.; Serabjit-Singh, C.;
Willson, T.; Collins, J.; Kliewer, S. Proc. Natl. Acad. Sci. USA
2000, 97, 7500.
[49] Wentworth, J. J. Endocrinol. 2000, 166, R11.
[50] Hennessy, M.; Kelleher, D.; Spiers, J.; Barry, M.; Kavanagh, P.;
Back, D.; Mulcahy, F.; Feely, J. Br. J. Clin. Pharmacol. 2002, 53,
[51] Perloff, M.; von Moltke, L.; Stormer, E.; Shader, R.; Greenblatt,
D. Br. J. Pharmacol. 2001, 134, 1601.
[52] Wang, Z.; Hamman, M.; Huang, S.; Lesko, L.; Hall, S. Clin.
Pharmacol. Ther. 2002, 71, 414.
[53] Barnes, J.; Anderson, L.; Phillipson, J. Herbal Medicines, 2nd ed.;
Pharmaceutical Press; London, 2002.
[54] Butterweck, V. CNS. Drugs. 2003, 17, 539.
[55] McRae, S. Can. Med. Ass. J. 1996, 155, 293.
[56] Awang, D. Can. Med. Assoc. J. 1996, 155, 1237.
[57] Hammond, T.; Whitworth, J. Med. J. Aust. 1981, 1, 492.
[58] Nielson, A. Ugeskrift for Laeger 1988, 150, 377.
[59] Baldwin, C. Pharm. J. 1986, 237, 583.
[60] Lei, X.L.; Chiou, G.C. Am. J. Chin. Med. 1986, 14, 145.
[61] Rao, M.R.; Shen, X.H.; Zou, X. J. Tradit. Chin. Med. 1987, 7, 127.
[62] Boyle, F. Med. J. Aust. 1997, 167, 286.
[63] Palmer, B.; Montgomery, A.; Montiero, J. Br. Med. J. 1978, 1,
[64] Koriech, O. Br. Med. J. 1978, 1, 1556.
[65] Greenspan, E. JAMA. 1983, 249, 2018.
[66] Palop-Larrea, V.; Gonzalvez-Perales, J.; Catalan-Oliver, C.;
Belenguer-Varea, A.; Martinez-Mir, I. Ann. Pharmacother.
2000, 34, 1347.
[67] Palop, V.; Catalan, C.; Rubio, E.; Martinez-Mir, I. Med. Clin.
(Barc.) 1999, 112, 46.
[68] World Health Organisation. WHO monographs on selected
medicinal plants; The World Health Organisation; Geneva, 1999;
Vol. 1.
[69] Mills, S.; Bone, K. Principles and Practice of Phytotherapy;
Churchill Livingstone; New York, 2000.
[70] Yun, T. Nutr. Rev. 1996, 54, S71.
[71] Choi, C.; Kang, G.; Min, Y. Planta. Med. 2003, 69, 235.
[72] Shader, R.; Greenblatt, D. J. Clin. Psychopharmacol. 1985, 5, 65.
[73] Shader, R. I.; Greenblatt, D. J. J. Clin. Psychopharmacol. 1988, 8,
[74] Jones, B. D.; Runikis, A. M. J. Clin. Psychopharmacol. 1987, 7,
[75] Gonzalez-Seijo, J.; Romas, Y.; Lastra, I. J. Clin.
Psychopharmacol. 1995, 15, 447.
[76] Wilkie, A.; Cordess, C. J. R.. Soc. Med. 1994, 87, 594.
[77] Siegel, R. JAMA 1979, 241, 1614.
[78] Vuksan, V.; Sievenpiper, J.; Koo, V.; Francis, T.; Beljan-
Zdravkovic, U.; Xu, Z.; Vidgen, E. Arch. Int. Med. 2000, 160,
[79] Sotaniemi, E.; Haapakoski, E.; Rautio, A. Diabetes. Care 1995, 18,
[80] Vuksan, V.; Stavro, M.; Sievenpiper, J.; Beljan-Zdravkovic, U.;
Leiter, L.A.; Josse, R.G.; Xu, Z. Diabetes. Care 2000, 23, 1221.
[81] Yushu, H.; Yuzhen, C. J. Tradit. Chin. Med. 1988, 8, 293.
[82] Guodong, L.; Zhongqi, L. Chin. J. Integrat. Tradit. West. Med.
1987, 7, 326.
[83] Konno, C.; Murakami, M.; Oshima, Y.; Hikino, H. J.
Ethnopharmacol. 1985, 14, 69.
[84] Janetzky, K.; Morreale, A. Am. J. Health. Sys. Pharm. 1997, 54,
[85] Rosado, M. Cardiology 2003, 99, 111.
[86] Zhu, M.; Chan, K. W.; Ng, L. S.; Chang, Q.; Chang, S.; Li, R. C. J.
Pharm. Pharmacol. 1999, 51, 175.
[87] Park, H.; Lee, J.; Song, Y.; Park, K. Biol. Pharmaceut. Bull. 1996,
19, 1434.
[88] Kuo, S.; Teng, C.; Lee, J.; Ko, F.; Chen, S.; Wu, T. Planta. Med.
1990, 56, 164.
[89] Kimura, Y.; Okuda, H.; Arichi, S. J. Pharm. Pharmacol. 1988, 40,
[90] Jung, K.Y.; Kim, D.S.; Oh, S.R.; Lee, I.S.; Lee, J.J.; Park, J.D.;
Kim, S.I.; Lee, H.K. Biol. Pharm. Bull. 1998, 21, 79.
[91] Miller, L. G. Arch. Int. Med. 1998, 158, 2200.
[92] Cheng, B.; Hung, C.; Chiu, W. HKMJ 2002, 8, 123.
[93] Gillis, C. Biochem. Pharmacol. 1997, 54, 1.
[94] Anderson, G.; Rosito, G.; Mohustsy, M.; Elmer, G. J. Clin.
Pharmacol. 2003, 43, 643.
[95] Donovan, J.; DeVane, C.; Chavin, K.; Taylor, R.; Markowitz, J.
Drug. Metab. Dis. 2003, 31, 519.
[96] Chang, T.; Chen, J.; Benetton, S. Drug. Metab. Dis. 2002, 30, 378.
[97] Henderson, C.L,; Harkey, M.; Gershwin, M.; Hackman, R.; Stern,
J.; Stresser, D. Life. Sci. 1999, 65, 209.
[98] Kim, H.; Chum, Y.; Park, J.; Kim, S.; Roh, J.; Jeong, T. Planta.
Med. 1997, 63, 415.
[99] Rowin, J.; Lewis, S. Neurol. 1996, 46, 1775.
[100] Lewis, S.; Rowin, J. Neurol. 1997a, 48, 789.
Herb-Drug Interactions: An Evidence Based Approach Current Medicinal Chemistry, 2004, Vol. 11, No. 11 1525
[101] Gilbert, G. Neurol. 1997, 48, 1137.
[102] Odawara, M.; Tamaoka, A.; Yamashita, K. Neurol. 1997, 48, 789.
[103] Lewis, S.; Rowin, J. Neurol. 1997b, 48, 1137.
[104] Lamant, V.; Mauco, G.; Braquet, P.; Chap, H.; Douste-Blazy, L.
Biochem. Pharmacol. 1987, 36, 2749.
[105] Chung, K.; McCusker, M.; Page, C.; Dent, G.; Guinot, P.; Barnes,
P. Lancet 1987, 1, 248.
[106] Vaes, L. P.; Chyka, P. A. Ann. Pharmacother. 2000, 34, 1478.
[107] Cott, J. CNS. Spectrums 2001, 6, 827.
[108] Matthews, M. J. Neurology 1998, 50, 1933.
[109] Scott, G.; Elmer, G. Am. J. Health. Sys. Pharm. 2002, 59, 339.
[110] Rosenblatt, M.; Mindel, J. New. Eng. J. Med. 1997, 336, 1108.
[111] Kim, Y.; Pyo, M.; Park, K.; Park, P.; Hahn, B.; Wu, S.; Yun-Choi,
H. Thromb. Res. 1998, 91, 33.
[112] Granger, A. Age. Ageing 2001, 30, 523.
[113] Wuda, K. et al. Chem. Pharm. Bull. 1998, 36, 1779.
[114] Manocha, A.; Pillai, K.; Husain, S. Indian J. Pharmacol. 1996, 28,
[115] Manocha, A.; Pillai, K.; Husain, S. Indian J. Pharmacol. 1997, 29,
[116] Galluzzi, S.; Zanetti, O.; Binetti, G.; Trabucchi, M.; Frisoni, G. B.
J. Neurol. Neurosurg. Psych. 2000, 68, 679.
[117] Koley, A.; Buters, J.; Robinson, R.; Markowitz, A.; Friedman, F. J.
Biol. Chem. 1995, 272, 3149.
[118] Miman, M.; Ozturan, O.; Iraz, M.; Erdem, T.; Olmez, E. Hearing
Res. 2002, 169, 121.
[119] Kudolo, G. J. Clin. Pharmacol. 2000, 40, 647.
[120] Zhang, X. Y.; Zhou, D. F.; Su, J. M.; Zhang, P. Y. J. Clin.
Psychopharmacol. 2001, 21, 85.
[121] Engelsen, J.; Nielsen, J.D.; Winther, K. Thromb. Haemost. 2002,
87, 1075.
[122] European Scientific Cooperative On Phytotherapy. ESCOP
Herbal Monographs 1999.
[123] Blumenthal, M. The Complete German Commission E
Monographs; Therapeutic Guide to Herbal Medicines; American
Botanical Council; Austin, 1998.
[124] Doll, R. Gut 1968, 9, 42.
[125] Heidemann, H.; Kreuzfelder, E. Klin. Wochenschr. 1983, 61, 303.
[126] Harada, T.; Ohtaki, E.; Misu, K.; Sumiyoshi, T.; Hosada, S.
Cardiol. 2002, 98, 218.
[127] Stormer, F.; Reistad, R.; Alexander, J. Food Chem. Toxicol. 1993,
31, 303.
[128] Armanini, D.; Karbowiak, I.; Funder, J.W. Clin. Endocrinol. 1983,
19, 609.
[129] Chandler, R. Glycyrrhiza Glabra. In Adverse Effects of Herbal
Drugs; De Smet, P., Keller, K., Hansel, R., Chandler, R., Eds.;
Springer Verlag; New York, 1997; Vol. 3.
[130] Forslund, T.; Fyhrquist, F.; Froseth, B.; Tikkanen, I. J. Intern.
Med. 1989, 225, 95.
[131] Yarnell, E.; Abascal, K. Alt. Complement Ther. 2002, April, 87.
[132] Chen, M.F.; Shimada, F.; Kato, H.; Yano, S.; Kanaoka, M.
Endocrinol. Jpn. 1990, 37, 331.
[133] Chen, M.F.; Shimada, F.; Kato, H.; Yano, S.; Kanaoka, M.
Endocrinol Jpn. 1991, 38,167.
[134] Shimizu, K.; Amagaya, S.; Ogihara, Y. J. Pharm. Dyn. 1984, 7,
[135] Homma, M.; Oka, K.; Ikeshima, K.; Takahashi, N.; Niitsuma, T.;
Fukuda, T.; Itoh, H. J. Pharm. Pharmacol. 1995, 47, 687.
[136] Teelucksingh, S.; Mackie, A.; Burt, D.; McIntyre, M.A.; Brett, L.;
Edwards, C.R. Lancet 1990, 335, 1060.
[137] Kumagai, A.; Nanaboshi, M.; Asanuma, Y.; Yagura, T.; Nishino,
K. Endocrinol. Japonica 1967, 14, 39.
[138] Francischetti, I.; Monteiro, R.; Guimaraes, J.; Francischetti, B.
Biochem. Biophys. Res. Comm. 1997, 235, 259.
[139] Dehpour, A.; Zolfaghari, M.; Samadian, T.; Vahedi, Y. J. Pharm.
Pharmacol. 1994, 46, 148.
[140] Dehpour, A.; Zolfaghari, M.; Samadian, T. Int. J. Pharm. 1995,
119, 133.
[141] Hikino, H. Recent Research on Oriental Medicinal Plants;
Academic Press; London, 1985; Vol. 1.
[142] Andersson, S.; Barany, F.; Caboclo, J.; Muzuno, T. Scand. J.
Gastroenterol. 1971, 6, 683.
[143] Hakanson, R.; Liedberg, G.; Oscarson, J.; Rehfeld, J.; Stadil, F.
Experientia 1973, 29, 570.
[144] Russell, R.; Morgan, R.; Nelson, L. Scand. J. Gastroenterol. 1984,
92 (suppl.), 97.
[145] Rees, W.; Rhodes, J.; Wright, J.; Stamford, L.F.; Bennett, A.
Scand. J. Gastroenterol. 1979, 14, 605.
[146] Tamaya, M.; et al. Am. J. Obstet. Gynecol. 1986, 155, 1134.
[147] Pizzorno, J.; Murray, A. A Textbook of Natural Medicine; John
Bastyr College Publications; Seattle, WA, 1985.
[148] Aida, K.; Tawata, M.; Shindo, H.; Onaya, T.; Sasaki, H.;
Yamaguchi, T.; Chin, M.; Mitsuhashi, H. Planta Medica 1990, 56,
[149] Yun-ping, Z.; Jia-qing, Z. Chin. Med. J. 1989, 102, 203.
[150] Xu, Y.-Z. Chin. J. Modern. Development. Trad. Med. 1987, 3,
[151] Liu, J.-B. Shandong. J. Trad. Chinese. Med. 1993, 6, 37.
[152] Yamamoto, S.; Aizu, E.; Jiang, H.; Nakadate, T.; Kiyoto, I.;
Wang, J.C.; Kato, R. Carcinogenesis 1991, 12, 317.
[153] Tawata, M.; Aida, K.; Noguchi, T.; Ozaki, Y.; Kume, S.; Sasaki,
H.; Chin, M.; Onaya, T. Eur. J. Pharmacol. 1992, 212, 87.
[154] Tawata, M.; Yoda, Y.; Aida, K.; Shindo, H.; Sasaki, H.; Chin, M.;
Onaya, T. Planta Medica 1990, 56, 259.
[155] Duke, J. Psychopharmacol, Bull. 1995, 31, 177.
[156] Paolini, M.; Pozetti, L.; Sapone, A.; Cantelli-Forti, G. Life. Sci.
1998, 62, 571.
[157] Paolini, M.; Barillari, J.; Broccoli, M.; Pozzetti, L.; Perocco, P.;
Cantelli-Forti, G. Cancer, Letters 1999, 145, 35.
[158] Shon, J.; Park, J.; Kim, M.; Cha, I.; Chun, B.; Shin, J. Clin.
Pharmacol. Ther. 2001, 69, 78.
[159] Gomez-Sanchez, C.; Gomez-Sanchez, C.; Yamakita, N. Semin.
Nephrol. 1995, 15, 1.
[160] Ploeger, B.; Mensinga, T.; Sips, A.; Seinen, W.; Meulenbelt, J.;
DeJongh, J. Drug Metab. Rev. 2001, 33, 125.
... John's Wort was shown to increase the metabolism o f a diverse range of medications such as cyclosporin (an immunosuppressant), indinavir (a protease inhibitor) and the oral contraceptive (Coxeter et al., 2004;Williamson, 2005). Further investigations revealed that hyperforin; the major active ingredient of St. John's Wort is a PXR agonist, resulting in significant induction o f the CYP 3A4 and MDRl genes (Staudinger, 2004;Wang et al., 2001). ...
... Induction of CYP2E1, CYP3A4, CYP1A2, CYP2D6, CYP2C19. (Diirr et al, 2000;Coxeter et al, 2004, Obach, 2000Wang et al, 2001;Wiliams on 2005) Hyperforin -agonist (Dürr et al, 2000;Staudinger, 2006) Some reports of induction, some reports of initial inhibition then induction (Rengelshausen et al, 2005;Williamson, 2005) Contra-Indicated with a number of medications, e.g. antocoagulants, immune suppressants, digoxin, protease inhibitors, statins, anti-depressants, oral contraceptives etc. ...
Despite the increasing use and popularity of Echinacea in the treatment of upper respiratory tract infections, little is known about its interactions with conventional medicines. Recent legislation changes have made it necessary to investigate potential interactions between herbal medicinal products and the CYP P450 system. To address this knowledge gap we measured the CYP P450 inhibition of Echinaforce® and nine more commercial Echinacea liquid preparations (ELP), along with selected constituents, with a modified fluorogenic assay. We demonstrated that all Echinacea preparations and all alkylamides assessed, directly inhibited CYP3A4. In addition Echinaforce® weakly inhibited CYP2D6, CYP2C19 and CYP1A2, while alkylamides 1 and 2 inhibited 2D6 and 2C19, but not 1A2. We observed no inhibition with caffeic acid but the results for three caffeic acid derivatives were inconclusive. Separation of six ELP (including Echinaforce®) into ethanol and water soluble components showed that most of the inhibitory activity resided in the ethanol fraction. Multivariate data analysis of 1H-NMR spectra of the ethanol fractions identified peaks linked with inhibitory activity. SPE fractionation of the ethanol fraction of Echinaforce® produced three active fractions. These were further analyzed by LC-MS (positive ion mode), revealing two major components, with molecular ion [M-H]+ masses of 282 and 248. Tandem MS and accurate mass analysis revealed that the 248 ion is in fact alkylamide 1, while ion 282 is most likely an unknown compound with molecular formula C18H36NO+ for which we have deduced a tentative structure. To assay for CYP3A4 induction we exposed HepG2 cells to relevant concentrations of Echinaforce® and alkylamides 1 and 2, but no significant changes in mRNA steady state levels were seen. Overall our results (in agreement with the available pharmacovigilance data) suggested that ELP are unlikely to cause clinically observable interactions with prescription medicines via the CYP P450 system, but that the observed effects vary widely in accordance with the products' chemical composition.
... With the significant prevalence of natural product use worldwide, Harnett et al. 2019) in some cases without the knowledge of the patient's primary practitioner (Xue et al. 2007), the risk of potential adverse events or drug interactions is of clinical concern. Evidence relating to nutraceutical/phytoceutical side effects and drug interactions can vary significantly and be challenging to interpret, due in part to variability in product standardisation, quality assurance processes, manufacturing methods, routes of administration, and dose (Coxeter et al. 2004). The nutraceuticals/phytoceuticals reviewed in these guidelines have undergone a consensus-based grading process that has taken into account currently available clinical and pharmacological evidence, along with the findings of several governmental regulatory agencies. ...
Full-text available
Objectives: The therapeutic use of nutrient-based 'nutraceuticals' and plant-based 'phytoceuticals' for the treatment of mental disorders is common; however, despite recent research progress, there have not been any updated global clinical guidelines since 2015. To address this, the World Federation of Societies of Biological Psychiatry (WFSBP) and the Canadian Network for Mood and Anxiety Disorders (CANMAT) convened an international taskforce involving 31 leading academics and clinicians from 15 countries, between 2019 and 2021. These guidelines are aimed at providing a definitive evidence-informed approach to assist clinicians in making decisions around the use of such agents for major psychiatric disorders. We also provide detail on safety and tolerability, and clinical advice regarding prescription (e.g. indications, dosage), in addition to consideration for use in specialised populations. Methods: The methodology was based on the WFSBP guidelines development process. Evidence was assessed based on the WFSBP grading of evidence (and was modified to focus on Grade A level evidence - meta-analysis or two or more RCTs - due to the breadth of data available across all nutraceuticals and phytoceuticals across major psychiatric disorders). The taskforce assessed both the 'level of evidence' (LoE) (i.e. meta-analyses or RCTs) and the assessment of the direction of the evidence, to determine whether the intervention was 'Recommended' (+++), 'Provisionally Recommended' (++), 'Weakly Recommended' (+), 'Not Currently Recommended' (+/-), or 'Not Recommended' (-) for a particular condition. Due to the number of clinical trials now available in the field, we firstly examined the data from our two meta-reviews of meta-analyses (nutraceuticals conducted in 2019, and phytoceuticals in 2020). We then performed a search of additional relevant RCTs and reported on both these data as the primary drivers supporting our clinical recommendations. Lower levels of evidence, including isolated RCTs, open label studies, case studies, preclinical research, and interventions with only traditional or anecdotal use, were not assessed. Results: Amongst nutraceuticals with Grade A evidence, positive directionality and varying levels of support (recommended, provisionally recommended, or weakly recommended) was found for adjunctive omega-3 fatty acids (+++), vitamin D (+), adjunctive probiotics (++), adjunctive zinc (++), methylfolate (+), and adjunctive s-adenosyl methionine (SAMe) (+) in the treatment of unipolar depression. Monotherapy omega-3 (+/-), folic acid (-), vitamin C (-), tryptophan (+/-), creatine (+/-), inositol (-), magnesium (-), and n-acetyl cysteine (NAC) (+/-) and SAMe (+/-) were not supported for this use. In bipolar disorder, omega-3 had weak support for bipolar depression (+), while NAC was not currently recommended (+/-). NAC was weakly recommended (+) in the treatment of OCD-related disorders; however, no other nutraceutical had sufficient evidence in any anxiety-related disorder. Vitamin D (+), NAC (++), methylfolate (++) were recommended to varying degrees in the treatment of the negative symptoms in schizophrenia, while omega-3 fatty acids were not, although evidence suggests a role for prevention of transition to psychosis in high-risk youth, with potential pre-existing fatty acid deficiency. Micronutrients (+) and vitamin D (+) were weakly supported in the treatment of ADHD, while omega-3 (+/-) and omega-9 fatty acids (-), acetyl L carnitine (-), and zinc (+/-) were not supported. Phytoceuticals with supporting Grade A evidence and positive directionality included St John's wort (+++), saffron (++), curcumin (++), and lavender (+) in the treatment of unipolar depression, while rhodiola use was not supported for use in mood disorders. Ashwagandha (++), galphimia (+), and lavender (++) were modestly supported in the treatment of anxiety disorders, while kava (-) and chamomile (+/-) were not recommended for generalised anxiety disorder. Ginkgo was weakly supported in the adjunctive treatment of negative symptoms of schizophrenia (+), but not supported in the treatment of ADHD (+/-). With respect to safety and tolerability, all interventions were deemed to have varying acceptable levels of safety and tolerability for low-risk over-the-counter use in most circumstances. Quality and standardisation of phytoceuticals was also raised by the taskforce as a key limiting issue for firmer confidence in these agents. Finally, the taskforce noted that such use of nutraceuticals or phytoceuticals be primarily recommended (where supportive evidence exists) adjunctively within a standard medical/health professional care model, especially in cases of more severe mental illness. Some meta-analyses reviewed contained data from heterogenous studies involving poor methodology. Isolated RCTs and other data such as open label or case series were not included, and it is recognised that an absence of data does not imply lack of efficacy. Conclusions: Based on the current data and clinician input, a range of nutraceuticals and phytoceuticals were given either a supportive recommendation or a provisional recommendation across a range of various psychiatric disorders. However several had only a weak endorsement for potential use; for a few it was not possible to reach a clear recommendation direction, largely due to mixed study findings; while some other agents showed no obvious therapeutic benefit and were clearly not recommended for use. It is the intention of these guidelines to inform psychiatric/medical, and health professional practice globally.
... The impact of the selected CMs on bleeding risks seemed various across different individual CMs. We assumed those CMs with multiple ingredients, with different concentrations of the active ingredients, and various processing procedures may be the reasons for the inconsistent findings [37]. Furthermore, the composition of a CM formula is usually based on the common, conventional principle of a drug remedy, that is, prescribing based on the roles of drugs as monarch, minister, adjuvant, or guide. ...
Full-text available
Despite the evidence that some commonly used Chinese medications (CMs) have antiplatelet/anticoagulant effects, many patients still used antiplatelets combined with CMs. We conducted a nested case-crossover study to examine the associations between the concomitant use of antiplatelets and CMs and major bleeding using population-based health database in Taiwan. Among the cohort of 79,463 outpatients prescribed antiplatelets (e.g., aspirin and clopidogrel) continuously, 1,209 patients hospitalized with new occurring bleeding in 2012 and 2013 were included. Those recruited patients served as their own controls to compare different times of exposure to prespecified CMs (e.g., Asian ginseng and dong quai) and antiplatelet agents. The periods of case, control 1, and control 2 were defined as 1–4 weeks, 6–9 weeks, and 13–16 weeks before hospitalization, respectively. Conditional logistic regression analyses found that concurrent use of antiplatelet drugs with any of the prespecified CMs in the case period might not significantly increase the risks of bleeding over that in the control periods (OR = 1.00, 95% CI 0.51 to 1.95 and OR = 1.13, 95% CI 0.65 to 1.97). The study showed no strong relationships between hospitalization for major bleeding events and concurrent use of antiplatelet drugs with the prespecified CMs.
This chapter reviews current research, knowledge and practice in the field of herbal medicines (HMs). Its focus is on phytovigilance, on what has been achieved in the science and the urgent need for its development. As the use of HMs surges across the world, safety and rational use become ever-more pressing issues, especially with regard to the discovery of adverse effects from often lightly-regulated or unregulated HMs themselves and from interactions from concomitant use of Western medicine. The current state of research is reviewed, with illustrative examples from work on COVID-19 and cancer. The chapter proposes expansion and improvement in regulation, standardization, clinical trials, information for practitioners and users, monitoring and reporting systems and in communication at all levels. Future progress relies on phytovigilance being given a higher priority in public health policy, in professional training, in research and in public communication.
Purpose: This review provides a summary of the currently available clinical data on drug-drug interactions (DDIs) involving over-the-counter (OTC) medicines. It aims to educate and increase awareness among healthcare providers, and to support decisions in daily practice. Methods: An extensive literature search was performed using bibliographic databases available through An initial structured search was performed using the keywords "drug-drug-interaction AND (over-the-counter OR OTC)", without further restrictions except for the language. The initial results were screened for all described DDIs involving OTC drugs, and further information was gathered specifically on these drugs using dedicated database searches and references found in the bibliography from the initial hits. Results: From more than 1200 initial hits (1972-June 2021), 408 relevant publications were screened for DDIs involving OTC drugs, leading to two major findings: first, certain types of drug regimens are more prone to DDIs or have more serious DDI-related consequences such as antiretroviral, anti-infective, and oral anticancer therapies. Second, while most DDIs involve OTC drugs as the perpetrators, some prescription drugs (statins, phosphodiesterase-5 inhibitors) that currently have OTC status can be identified as the victims in DDIs. The following groups were identified to be frequently involved in DDIs: non-steroidal anti-inflammatory drugs, food supplements, antacids, proton-pump inhibitors, H2-antihistamines, laxatives, antidiarrheal drugs, and herbal drugs. Conclusion: The most significant finding was the lack of high-quality evidence for commonly acknowledged interactions. High-quality interaction studies involving different phenotypes in drug metabolism (cytochrome P450) and distribution (transporters) are urgently needed. This should include modern and critical drugs, such as oral anti-cancer medications and direct oral anticoagulants.
Herbal drugs are being in use for the management of human health and for prevention as well as to cure human diseases since ancient civilization. In recent days, the use of herbal drugs has been increased significantly in various forms such as herbal formulations, dietary supplements, and nutraceuticals in the global market. This growing demand undoubtedly proves the therapeutic claims of herbal drugs as biomedicines and/or functional foods. However, the safe use of herbal products/herbal medicines is still challenging due to the toxicity and regulatory issues. This review discusses toxicity-related and safety issues of herbal medicinal products, factors responsible for, and suitable remedial measures. Some challenges associated with monitoring the safety of herbal drugs are also discussed to ensure their effectiveness for adequate protection of public health and the relevant regulatory issues.
Purpose: The combination of warfarin and compound Danshen dripping pill (CDDP) is helpful for patients with both coronary heart disease (CHD) and atrial fibrillation (AF). The main adverse drug reaction of warfarin is bleeding because of its narrow therapeutic index. The safety of a combination therapy with warfarin and CDDP is always a concern. Our previous research showed that the combination of warfarin and CDDP improved the quality of life for patients with both CHD and AF. This study describes the changes in dose and concentration of warfarin necessary and evaluates bleeding risk when warfarin is given concomitantly with CDDP. Methods: An ultra-performance liquid chromatography-MS/MS method with a chiral column was developed to assay the concentration of S-warfarin and R-warfarin in human plasma simultaneously. The method was applied to compare the concentration of warfarin in patients taking warfarin combined with CDDP and without CDDP. International normalized ratio (INR) values were monitored to evaluate bleeding risk. Paired t tests were then used to compare the dose and the concentration in 2 periods. Moreover, patients with VKORC1, CYP2C9*3, CYP4F2, EPHX1, and PROC gene polymorphisms were evaluated to determine interactions. Findings: The results indicate that the dose of warfarin had no significant change with or without CDDP. Also, the peak concentrations of S-warfarin and total warfarin were significantly different in CYP4F2 C/C patients, but there was no significant difference identified in other genetic groups. No bleeding occurred in the study. Implications: The dose of warfarin would be sustainable when combined with CDDP, because CDDP did not affect concentration of warfarin significantly in most patients and the change of INR was not significant. China clinical trial registry identifier: ChiCTR-ONRC-13003523.
This chapter aims to construct a representative inhibitor pharmacophore model and apply it to identify Organic Anion Transporter 1 (OAT1) inhibitors from Traditional Chinese Medicine (TCM) that may induce herb-drug interactions clinically. The OAT1-mediated drug-drug interactions have received considerable interest for their potential influence on drug efficacy and toxicity. TCM computational toxicology is in its early phrase and more work is warranted to demonstrate its full potential in TCM-related studies, such as deciphering the toxic mechanisms and research and development of modern TCM patent prescriptions. Hits exhibiting a good fit to the generated pharmacophore model were retrieved from TCM compound database for further evaluation. The chapter also presents a case study on the application of computational techniques to evaluate the potential risk of transporter-mediated drug-TCM interactions. Literature mining and database searching were performed to identify small molecule compounds that have been described in the TCMs by the keyword-based approach.
Licorice, a core traditional herb medicine, is frequently coadministered with conventional medications in Korea. A few reports suggested that licorice induced the cytochrome P450 (CYP) in animals, especially for CYP3A isoform. To evaluate the effect of licorice on the PK and PD of midazolam, a known CYP3A4 substrate, 1g licorice (freeze dried water extract: extraction ratio-25%) or placebo were orally administered (bid) to 10 healthy male subjects for 1 days as double blind randomized crossover manner. After oral dose of 7.5mg midazolam at 8th day of pretreatment, multiple blood samples were drawn up to 24 hours and psychomotor performance tests (Digit span test and Digit symbol substitution test) were conducted for 4 hours. There was no significant difference of PK parameters of midazolam (AUC: 107.74±33.17 and 106.30±56.76 ng/ml·hr, t1/2:2.6±0.38 and 1.67±0.63 hr, and Cl/F: 1.26±0.42 and 1.46±0.70L/min) and 1-hydroxymidazolam (AUC: 54.22±14.83 and 63.20±18.29 ng/ml· hr), between two phases of placebo and licorice pretreatment. The effect of midazolam on psychomotor tests was not changed after licorice pretreatment. These results suggested that licorice seems not to induce CYP3A4 catalyzing midazolam hydroxylation, and drug interaction of licorice with CYP3A4 substrate drugs are less likely expected in humans.
Objective: The present study was designed to evaluate the effect of Ginkgo biloba (a PAF antagonist) on chemoshock in mice. Methods: Picrotoxin and strychnine were used for inducing chemoconvulsions in mice. The onset, duration, nature, severity of convulsions and mortality were observed. The severity of convulsions was assessed by scoring method. Results: Picrotoxin and strychnine produced convulsions in higher doses but not in lower doses. Prior administration of G. biloba extract potentiated the action of picrotoxin and strychnine respectively. Conclusion: The potentiation of picrotoxin action by G. biloba indicates the involvement of GABAergic system and chloride channel. Facilitation of strychnine action by G. biloba indicates the modulating effect of the extract on glycine.
Anticonvulsants, namely, sodium valproate and carbamazepine were found to exhibit significant protective effect against chemoshock induced by picrotoxin as well as strychnine in mice. Ginkgo biloba (a PAF antagonist) treatment decreased the protective effect of both sodium valproate and carbamazepine. The present study may have practical implications if a PAF antagonist is used in epileptic patients with coexisting symptoms of cerebral insufficiency.
Siberian ginseng ([SG]; Eleutherococcus senticosus) is a commonly used herbal preparation. The objective of this study was to assess in normal volunteers (n = 12) the influence of a standardized SG extract on the activity of cytochrome P450 CYP2D6 and 3A4. Probe substrates dextromethorphan (CYP2D6 activity) and alprazolam (CYP3A4 activity) were administered orally at baseline and again following treatment with SG (1 × 485 mg twice daily) for 14 days. Urinary concentrations of dextromethorphan and dextorphan were quantified, and dextromethorphan metabolic ratios (DMRs) were determined at baseline and after SG treatment. Likewise, plasma samples were collected (0–60 h) for alprazolam pharmacokinetics at baseline and after SG treatment to assess effects on CYP3A4 activity. Validated high performance liquid chromatography methods were used to quantify all compounds and relevant metabolites. There were no statistically significant differences between pre- and post-SG treatment DMRs indicating a lack of effect on CYP2D6 (P > 0.05). For alprazolam there also were no significant differences in the pharmacokinetic parameters determined by noncompartmental modeling (Cmax, Tmax, area under the curve, half-life of elimination) indicating that SG does not significantly induce or inhibit CYP3A4 (P > 0.05). Our results indicate that standardized extracts of SG at generally recommended doses for over-the-counter use are unlikely to alter the disposition of coadministered medications primarily dependent on the CYP2D6 or CYP3A4 pathways for elimination.
Herbal medicinals are being used by an increasing number of patients who typically do not advise their clinicians of concomitant use. Known or potential drug-herb interactions exist and should be screened for. If used beyond 8 weeks, Echinacea could cause hepatotoxicity and therefore should not be used with other known hepatoxic drugs, such as anabolic steroids, amiodarone, methotrexate, and ketoconazole. However, Echinacea lacks the 1,2 saturated necrine ring associated with hepatoxicity of pyrrolizidine alkaloids. Nonsteroidal anti-inflammatory drugs may negate the usefulness of feverfew in the treatment of migraine headaches. Feverfew, garlic, Ginkgo, ginger, and ginseng may alter bleeding time and should not be used concomitantly with warfarin sodium. Additionally, ginseng may cause headache, tremulousness, and manic episodes in patients treated with phenelzine sulfate. Ginseng should also not be used with estrogens or corticosteroids because of possible additive effects. Since the mechanism of action of St John wort is uncertain, concomitant use with monoamine oxidase inhibitors and selective serotonin reuptake inhibitors is ill advised. Valerian should not be used concomitantly with barbiturates because excessive sedation may occur. Kyushin, licorice, plantain, uzara root, hawthorn, and ginseng may interfere with either digoxin pharmacodynamically or with digoxin monitoring. Evening primrose oil and borage should not be used with anticonvulsants because they may lower the seizure threshold. Shankapulshpi, an Ayurvedic preparation, may decrease phenytoin levels as well as diminish drug efficacy. Kava when used with alprazolam has resulted in coma. Immunostimulants (eg, Echinacea and zinc) should not be given with immunosuppressants (eg, corticosteroids and cyclosporine). Tannic acids present in some herbs (eg, St John wort and saw palmetto) may inhibit the absorption of iron. Kelp as a source of iodine may interfere with thyroid replacement therapies. Licorice can offset the pharmacological effect of spironolactone. Numerous herbs (eg, karela and ginseng) may affect blood glucose levels and should not be used in patients with diabetes mellitus.