Professional Documents
Culture Documents
www.elsevier.com/locate/ijcard
Review
a
Department of Experimental Pharmacology, University of Naples Federico II, via D. Montesano 49, 80131 Naples, Italy
Complementary Medicine, Penninsula Medical School, Universities of Exeter and Plymouth, 25 Victoria Park Road, EX2 4NT, UK
Received 18 April 2003; received in revised form 10 June 2003; accepted 14 June 2003
Available online 21 February 2004
Abstract
Use of herbal medicines among patients under cardiovascular pharmacotherapy is widespread. In this paper, we have reviewed the
literature to determine the possible interactions between herbal medicines and cardiovascular drugs. The Medline database was searched for
clinical articles published between January 1996 and February 2003. Forty-three case reports and eight clinical trials were identified.
Warfarin was the most common cardiovascular drug involved. It was found to interact with boldo, curbicin, fenugreek, garlic, danshen,
devils claw, don quai, ginkgo, papaya, lycium, mango, PC-SPES (resulting in over-anticoagulation) and with ginseng, green tea, soy and St.
Johns wort (causing decreased anticoagulant effect). Gum guar, St. Johns wort, Siberian ginseng and wheat bran were found to decrease
plasma digoxin concentration; aspirin interactions include spontaneous hyphema when associated with ginkgo and increased bioavailability if
combined with tamarind. Decreased plasma concentration of simvastatin or lovastatin was observed after co-administration with St. Johns
wort and wheat bran, respectively. Other adverse events include hypertension after co-administration of ginkgo and a diuretic thiazide,
hypokalemia after liquorice and antihypertensives and anticoagulation after phenprocoumon and St. Johns wort. Interaction between herbal
medicine and cardiovascular drugs is a potentially important safety issue. Patients taking anticoagulants are at the highest risk.
D 2004 Elsevier Ireland Ltd. All rights reserved.
Keywords: Cardiovascular pharmacotherapy; Herbal medicines; Drug interaction
1. Introduction
Interest in alternative medicine including plant-derived medications is growing. Self-administration of herbal
medicines is among the most popular of alternative therapies
[1,2]. In the US, the market for herbal medicinal products
(usually sold as food supplements or nutraceuticals)
amounted to US$590.9 million [3]. These sales figures
relate only to food stores, drug stores and mass market
and would obviously be larger if buying clubs, convenience
stores, natural food markets, multilevel marketing companies, health professionals, mail or Internet order had been
considered. The relevance of alternative therapies for cardiovascular medicine is highlighted by the recent workshop
on the use of herbal medicines in cardiovascular, lung and
blood research sponsored by the US National Heart, Lung,
and Blood Institute [4].
Table 1
Clinical interactions between herbal medicines and conventional cardiovascular drugs
Herbal
medicine
Result of
interaction
Possible
mechanism
Pharmacological
comment
Clinical
comment
No. of cases
Comment on report
reliability
Ref
Decreased
plasma digoxin
concentration
Reduced absorption
Similar amount
of digoxin was
found in 24-h urine
whether given with
or without guar gum.
St. Johns wort may
reduce efficacy of
digoxin and make a
patient a nonresponder.
The patient was
asymptomatic for
digoxin toxicity
despite a level of
2.5 ng/l.
Not applicablea
[27]
Not applicablea
[28]
[29]
Digoxin levels
were still within
the therapeutic range.
This interaction is
surprising as Ginkgo
is a peripheral
vasodilatator.
Some antihypertensive
drugs induce
hypokalemia; liquorice
has mineralcorticoid
effects which may
cause potassium
excretion.
If confirmed, the
interaction is
potentially dangerous.
[31]
Serum potassium
levels should be
monitored closely
in patients who are
predisposed to
cardiac arrhythmias
and who are
concurrently treated
with digitalis
glycosides.
[32]
Digoxin
St. Johns
wort
Decreased
plasma digoxin
concentration
Induction
of P-glycoprotein
Digoxin
Siberian
ginseng
Increased
plasma digoxin
concentration
Digoxin
Wheat bran
Decreased
plasma digoxin
concentration
Some component
of Siberian ginseng
might impair digoxin
elimination or
interfere with the
digoxin assay.
Reduced absorption
Antihypertensives
Liquorice
Hypokalemia
Not known
Additive effect
on potassium
excretion
[30]
Conventional
drug
Aspirin
Tamarind
Additive
inhibition
of platelet
aggregation
Ginkgolides from
ginkgo have
antiplatelet activity
and are PAF receptor
antagonists.
Increased
bioavailability
of aspirin
Not known
Increased
anticoagulant
effect
Additive effect
on coagulation
mechanisms
Warfarin
Curbicin
Increased
anticoagulant
effect
Additive effect
on coagulation
mechanisms
Warfarin
Danshen
Increased
anticoagulant
effect
Warfarin
Devils claw
Increased
anticoagulant
effect, purpura
Additive effect
on coagulation
mechanisms
and/or increased
plasma warfarin
concentration
Unknown
Warfarin
Dong quai
Increased
anticoagulant
effect
Additive effect
on coagulation
mechanisms
Warfarin
Garlic
Increased
anticoagulant
effect; increase
in clotting time
Additive effect
on coagulation
mechanisms
In contrast to NSAIDs,
devils claw does not
affect platelet function.
Spontaneous bleeding
from the iris into
the anterior chamber
of the eye is a rare
problem.
[33]
Uncertain
Risk of bleeding;
given the narrow
therapeutic index of
warfarin, vigilance
is needed.
Cases of coagulation
disorders related to
vitamin E have
been reported.
Interaction confirmed
by rechallenge
[35]
[36]
Risk of bleeding;
given the narrow
therapeutic index
of warfarin, vigilance
is needed.
Although neither of
the two patients has
been re-exposed to
curbicin, the causal
relation is quite strong.
Reports provide
reliable evidence for
an interaction.
Risk of bleeding;
given the narrow
therapeutic index
of warfarin, vigilance
is needed.
Risk of bleeding;
given the narrow
therapeutic index of
warfarin, vigilance
is needed.
Garlic treatment has
been associated
with bleeding even in
the absence of
warfarin or other
anticoagulant
treatment.
[31]
Reports provide
reliable evidence
for an interaction.
[40,41]
[42]
[34]
[37 39]
Spontaneous
hyphema
Table 1 (continued)
Conventional
drug
Herbal
medicine
Possible
mechanism
Pharmacological
comment
Clinical
comment
No. of cases
Comment on report
reliability
Ref
Intracerebral
hemorrhage
Additive effect
on coagulation
mechanisms
Ginkgolides from
ginkgo have antiplatelet
activity and are PAF
receptor antagonists.
[43]
Antiplatelet activity
of ginseng has been
reported but would
not seem to explain
this case of decreased
anticoagulation;
a pharmacokinetic study
in rats did not reveal a
significant interaction
between warfarin and
ginseng.
Warfarin produces
anticoagulation
by inhibiting
production of the
vitamin-K dependent
clotting factors.
Green tea contains
vitamin K and
thus antagonize the
effect of warfarin.
The weak inhibition
of Lycium on
hepatic enzyme
could not explain
such interaction.
Mango contains high
amounts of vitamin
A and human studies
have shown that
vitamin A (retinol)
inhibits CYP2C19
enzymes.
Spontaneous bilateral
subdural haematomas
associated with
long-term ginkgo
ingestion have been
reported (even in the
absence of
anticoagulants).
Potential seriousness
of thrombotic
complications
[44]
Patients receiving
warfarin need
to be routinely
questioned about
their intake of
vitamin K-containing
foods and beverages.
[45]
Risk of bleeding;
given the
narrow therapeutic
index of warfarin,
vigilance is needed.
Risk of bleeding;
given the narrow
therapeutic index
of warfarin,
vigilance is needed.
[46]
13
[47]
Warfarin
Ginseng
Decreased
anticoagulant
effect
Unknown
Warfarin
Green tea
Decreased
anticoagulant
effect
Pharmacological
antagonism
Warfarin
Lycium
Increased
anticoagulant
effect
Unknown
Warfarin
Mango
Increased
anticoagulant
effect
Hepatic enzyme
inhibition
Result of
interaction
Warfarin
Papaya
Unknown
Warfarin
PC-SPES
Warfarin
Risk of bleeding;
this interaction is
potentially fatal.
The thromboembolic
side effects of PC-SPES
are potentially fatal;
individuals at risk should
be strongly advised
against using PC-SPES
and warfarin or aspirin.
The decrease in INR
was thought to be
clinically relevant
[31]
[48]
Soy
Decreased
anticoagulant
effect
Not known
Warfarin is metabolised
by CYP 1A2 in the
liver, which is induced
by St. Johns wort.
Although none
of the patients
developed
thromboembolic
complications,
the decrease in
INR was thought
to be clinically
relevant.
Hepatic enzyme
induction
Phenprocoumon
has a narrow
therapeutic window;
possible loss
of activity.
1a
An objective causality
assessment in this case
revealed that INR
decline was in the
range of possible to
probable.
These cases are
reported in a single
publication, which
contains insufficient
information; however,
a clinical study showed
that St. Johns wort
decreased the plasma
concentration of
phenprocoumon (which is
chemically related to
warfarin).
The report provides some
evidence for interaction.
Moreover, a clinical study
confirm such an
interaction.
Warfarin
St. Johns
wort
Decreased
anticoagulant
effect
Hepatic enzyme
induction
Phenprocoumon
St. Johns
wort
Phenprocoumon
Wheat bran
Increased
Quick-Wert
test (indicating
decreased
anticoagulant
effect)
Decreased plasma
level of
phenprocoumon;
increase in the
free plasma
phenprocoumon
fraction
Decreased absorption
can explain the
decreased plasma
level; however, the
mechanism of the
increase of free plasma
phenprocoumon
fraction is unknown.
In view of the
different effects
on phenprocoumon
pharmacokinetics,
the clinical
significance is
unpredictable.
Additive effect
on coagulation
mechanisms
PC-SPES contains
anticoagulant
coumarins.
Not applicablea
[50]
[51,52]
[53]
Increased
anticoagulant
effect
Increased
anticoagulant
effect
Table 1 (continued)
Conventional
drug
Herbal
medicine
Result of
interaction
Lovastatin
Oat bran
Decreased
lovastatin
absorption
Lovastatin
Pectin
Decreased
lovastatin
absorption
Pharmacological
comment
Hepatic enzyme
induction
Simvastatin is
extensively
metabolised by CYP
3A4 in the intestinal
wall and liver, which
are induced by
St. Johns wort.
Bran contains fibers
which can trap digoxin.
Interaction revealed by a clinical study. Clinical studies are more rigorous than case reports.
Clinical
comment
The decreased
absorption of
lovastatin resulted
to an increase in
LDL levels which
led to the abortion
of the trial. Lovastatin
pharmacokinetics
and LDL returned
normal after bran
discontinuation.
The decreased
absorption of
ovastatin resulted
to an increase in
LDL levels which
led to the abortion
of the trial. Lovastatin
pharmacokinetics
and LDL returned
normal after pectin
discontinuation.
No. of cases
Comment on report
reliability
Ref
Not applicable
[54]
[55]
Not applicable
[55]
Possible
mechanism
2. Methods
Systematic literature searches were made using Medline
(via PubMed, from January 1966 to February 2003). The
search terms were herbal medicine, botanical medicine,
phytotherapy, drug interaction, adverse effects, side effects,
adverse drug reaction, safety and toxicity. Recent books on
herb drug interactions or herbalism [21 26] were also
searched for further relevant information. Additional publications were identified by checking all reference lists and
by searching our files. No language restrictions were imposed. All clinical reports on interactions were read and
relevant data were extracted by the first three authors into
predefined tables and validated by the senior author. In vitro
experiments have been excluded.
3.1.1. Digoxin
Digoxin is a cardiac glycoside which originates from the
digitalis (foxglove) plant. As other cardiac glycosides, it can
increase the contractility of the heart muscle and is therefore
used in treating heart failure [56].
A single-blind, placebo-controlled study with two parallel groups showed that St. Johns wort reduced digoxin
through levels after 10 days of co-medication [28]. Intestinal
absorption, distribution and renal excretion of digoxin are
mediated by the multiple-drug-resistance gene product Pglycoprotein, which has been shown to be induced by St.
Johns wort [58,59]. Through this mechanism, St. Johns
wort may reduce efficacy of digoxin and make a patient a
nonresponder, whereas increased toxicity may be anticipated after withdrawal of the herb.
Increased levels of digoxin have been associated with
ingestion of Siberian ginseng [29]. The patient was asymptomatic for digoxin toxicity despite a level of 5.2 ng/l.
Electrocardiogram, potassium level and serum creatine level
were normal. Digoxin levels decreased upon dechallenge
and increased upon rechallenge. The product was analyzed
3.2.1. Aspirin
Aspirin is currently employed in the prophylactic treatment of transient cerebral ischemia, to reduce the incidence
of recurrent myocardial infarction and to decrease mortality in postmyocardial infarction patients. Aspirin blocks
thromboxane A2 synthesis from arachidonic acid in platelets by irreversibly acetylating and thus inhibiting cyclooxygenase, a key enzyme in prostaglandin synthesis. The
aspirin-induced inhibition of thromboxane A2 synthesis
last for the life of the platelet (approximately 7 10 days)
[56].
A case report demonstrated a patient treated with aspirin
for 5 years experienced bleeding of the eye and blurred
vision after self-medication with ginkgo for 1 week [33].
After stopping ginkgo, there was no recurrence of bleeding
over a 3-month follow-up period. The interaction is likely
due to an additive effect on platelet function as ginkgolide B
from ginkgo is a potent PAF receptor antagonist [62].
A clinical study [34] performed on six healthy volunteers
showed that a tamarind extract incorporated in a traditional
meal increased plasma levels of aspirin and salicylic acid
3. Results
Table 2
Herbal medicines interacting with cardiovascular pharmacotherapy: source, main constituent(s), main pharmacological action(s), promoted use, clinical
evidence and adverse events
Herbal medicine
(Common name/
Latin name)
Source
Main
constituent(s)
Main
pharmacological
action(s)
Promoted use
Clinical
evidence
Adverse events
Boldo/Peumus boldus
Leaves
Boldine
Not expected
Fatty acids,
phytosterols,
flavonoids,
polysaccharides
Indigestion,
constipation,
hepatic ailments
Benign prostatic
hyperplasia
Specific studies
not available
Curbicina/Serenoa repens/
Cucurbita pepo
Choleretic/
cholagogue,
diuretic
Antiandrogenic,
anti-inflammatory
Danshen/Salvia
miltiorrhiza
Roots
Tanshinones,
phenolic
compounds
Vasorelaxant,
anti-ischemic,
antiplatelet;
radical scavenger
Angina, myocardial
infarction, ischemic
diseases
Dong quai/Angelica
sinensis
Roots
Phytoestrogens,
flavonoids,
coumarins
Estrogenic effects,
anti-inflammatory,
vasorelaxant
Devils claw/
Harpagophytum
procumbens
Root, tubers
Harpagoside
Gynecological
disorders,
circulation
conditions
Musculoskeletal
and arthritic pain
Fenugreek/Trigonella
foenum-graecum
Seeds
Alkaloids,
flavonoids,
saponins
Anti-inflammatory,
anti-arrhythmic,
positive inotropic,
negative chronotropic
Antilipidaemic,
Diabetes mellitus,
hypoglycemic,
hypercholesterolemia
cholagogue
Ginseng/Panax
ginseng
Roots
Triterpene
saponins
known as
ginsenosides
Immunomodulatory,
anti-inflammatory,
antitumor,
hypoglycemic
Garlic/Allium
sativum
Bulb
Alliins
Ginkgo/Ginkgo
biloba
Leaves
Ginkgolides,
flavonoids
Antihypertensive,
antidiabetic,
antiplatelet,
antilipidaemic
Increase of
microcirculatory
blood flow,
antiplatelet, free
radical scavenging
Green tea/Camellia
sinensis
Leaves
Polyphenols,
caffeine
Antimutagenic,
antioxidant,
antilipidaemic,
antitumoral, CNS
stimulant
Guar gum/Cyamopsis
tetragonolobus
Seeds
Galactomannan,
lipids, saponins
Antihyperglycemic,
antilipidaemic
Kava/Piper
methysticum
Rhizome
Kavapyrones
Anxiolytic,
anesthetic,
muscle relaxant
Serenoa repens is
effective in the
treatment of benign
prostatic hyperplasia.
Gastrointestinal
complaints,
constipation,
diarrhea,
decreased libido
Effectiveness not
Specific studies
proven. Most studies not available
are neither
placebo-controlled
nor blinded.
No sufficient
Photosensitivity
evidence of
leading to mild
effectiveness
dermatitis,
bleeding
Promising to treat
Gastrointestinal
musculoskeletal
symptoms
and back pain
Promising in
reducing serum
cholesterol levels
Minor
gastrointestinal
symptoms,
allergic reactions
Loss of energy and
Not established
Insomnia, diarrhea,
memory; stress
for any indications
vaginal bleeding,
states; male sexual
mastalgia, possible
dysfunction
cause of
Stevens Johnson
syndrome
Allergic reactions,
Hypercholesterolemia, Small
nausea, heartburn,
prevention of
antihypertensive
flatulence, breath
arteriosclerosis
and antilipidaemic
and body odor
effect
Gastrointestinal
Circulatory disorders Favorable evidence
disturbances,
for the treatment
vomiting, allergic
of intermittent
claudication, tinnitus reactions, pruritus,
headache, dizziness,
and dementia
nose bleeding
(including
Alzheimers
dementia)
Prevention of
Cautiously positive
Insomnia
as anticancer; strong
cancer,
inverse associations
cardiovascular
diseases,
of tea intake with
aortic arteriosclerosis
adjuvant treatment
and cardiovascular
for AIDS
riskb.
Diabetes, obesity,
Small effect
Flatulence, diarrhea,
hypercholesterolemia cholesterol
abdominal
levels; ineffective
distension, nausea,
for obesity
hypoglycemic
symptoms
Stomach
Anxiety insomnia
Well documented
for the treatment
complaints,
of anxiety
restlessness,
mydriasis,
dermatomyositis,
hepatitis
Table 2 (continued)
Herbal medicine
(Common name/
Latin name)
Source
Main
constituent(s)
Main
pharmacological
action(s)
Promoted use
Clinical
evidence
Adverse events
Liquorice/Glycyrrhiza
glabra
Roots
Glycyrrhizinic
acid
Gastric ulcer,
catarrhs, cancer
prevention,
inflammation
Liquorice is an
effective anti-ulcer;
however, its use
has declined.
Adverse effect
consistent with
adrenocorticotropic
actions
Mango/Mangifera
indica
Fruits
Vitamins A
and C; fibers
Expectorant,
anti-inflammatory,
anti-ulcer,
aldosterone-like
effects
Analgesic;
anti-inflammatory;
antioxidant; laxative
Specific studies
not available
Not expected
Seeds
Fibers
Constipation
(also used as a
food and as a
source of vitamins)
Hypercholesterolemia,
prevention of
atherosclerosis
Not expected
Papaya/Carica papaya
Fruits
Papain (enzyme)
Promising in
reducing
cholesterol and
LDL blood levels
Specific studies
not available
Prostate cancer
Lack of
randomized
clinical studies
Reduced libido,
hot flashes,
diarrhea, dyspepsia,
leg cramps,
gynaecomastia,
nipple tenderness,
pulmonary emboli,
vein thrombosis
Pectins-based
home remedies
are useful in the
treatment of
diarrhea.
Not established
for any
indications
Not expected
d
PC-SPESd/Dendrathema
morofolium/Isatis
indigotica/Glycyrrhiza
glabra/Ganoderma
lucidum/Panax
pseudoginseng/Rabdosia
rubescens/Serenoa
repens/Scutellaria
bacicalensis
Pectinse
fleshy fruits
and storage
roots of many
plantse
Antilipidaemic,
anti-atherosclerotic
Proteolytic,
amylolytic,
lipolytic activity
Polysaccharides, Immunostimulant,
phytosterols, fatty cytotoxic
acids, flavonoids
Indigestion, obesity
Antidiarrheal
Diarrhea
Siberian ginseng/
Eleutherococcus
senticosus
Roots
Eleutherosides
Immunomodulatory,
anti-inflammatory,
antitumor,
hypoglycemic
Loss of energy
and memory;
stress states,
male sexual
dysfunction
Soy/Glycine max
Beans
Phytoestrogens
Hepatoprotective,
anti-osteoporosis
Treatment of
menopausal
symptoms;
prevention of
heart diseases
and cancer
Aerial parts
Hypericin,
hyperforin,
flavonoids
Antidepressant,
antiretroviral
Depression
Tamarind/Tamarindus
indica
Fruits
Sugars,
mucilages
Stimulates
intestinal
peristalsis
Constipation
(also used as
a food)
Promising for
treating
menopausal
symptoms;
case-control
studies suggest
a link between
soy phytoestrogen
consumption and
reduced risk of
breast and other
cancers.
Effective for mild
to moderate
depression; not
suited for major
depression
Effective
laxative
Not expected
Diarrhea,
dizziness,
hypertension,
pericardial pain,
tachycardia,
insomnia,
extrasystoles,
headaches
Occasional
gastrointestinal
effects, i.e.
stomach pain,
flatulence, loose
stool and diarrhea
Gastrointestinal
symptoms
Not expected
10
Table 2 (continued)
Herbal medicine
(Common name/
Latin name)
Source
Main
constituent(s)
Main
pharmacological
action(s)
Promoted use
Clinical
evidence
Adverse events
Wheat bran/Triticum
aestivum
Seeds
Indigestible
carbohydrates
(starch, cellulose,
hemicelluloses),
lignin
Stimulates
intestinal
peristalsis
Constipation,
obesity
Ineffective to
treat obesity
Bloating
11
12
4. Discussion
Herbal medicines follow modern pharmacological principles. Hence, herb drug interactions are based on the same
pharmacokinetic and pharmacodynamic mechanisms as
drug drug interactions [5]. Herbal medicines may affect
absorption (e.g. guar gum reduces digoxin absorption) [27],
metabolism (e.g. St. Johns wort increases warfarin metabolism, causing decreased anticoagulant effect) [50] or excretion (St. Johns wort increases digoxin renal excretion)
[28] of concurrently administered cardiovascular drugs.
Herb drug interactions that involve distribution mechanisms have not been reported. Moreover, interactions may
be additive or synergetic, whereby the herbal products
potentiate the action of the conventional cardiovascular
drug (e.g. ginkgo potentiates the antiplatelet effect of
aspirin) [33]. Conversely, the herb may be directly antagonistic to the action of the drug (e.g. green tea antagonizes the
anticoagulant effect of warfarin) [45].
Based on the above evidence, there can be little doubt
that interactions between herbal medicines and cardiovascular drugs exist. The real incidence of such interactions is
probably unknown, as is the likelihood that a patient will
have an adverse event when taking two drugs (i.e. herbal
and conventional medicines) with the potential to interact.
Much of the available information about the interaction
between herbal medicines and cardiovascular pharmacother-
Acknowledgements
This work was supported by the Enrico and Enrica
Sovena Foundation and SESIRCA (Regione Campania,
Italy).
References
[1] Ernst E, Pittler MH, Stevinson C, White A. The desktop guide to complementary and alternative medicine. An evidence-based approach.
Edinburg: Butterworth Heinemann; 2000.
[2] Eisenberg DM, Davis RB, Ettner SL, Apple S, Wilkey S, Van Rompay M, et al. Trends in alternative medicine use in the United States,
1990 1997: results of a follow-up national survey. JAMA 1998;
280:1569 75.
[3] US sales figures for herbal medicinal products. Fact 2001;6:231.
[4] Lin MC, Nahin R, Gershwin ME, Longhurst JC, Wu KK. State of
complementary and alternative medicine in cardiovascular, lung, and
blood research [executive summary of a workshop]. Circulation 2001;
103:2038 41.
[5] Izzo AA, Borrelli F, Capasso R. Herbal medicine: the risk of drug
interaction. Trends Pharmacol Sci 2002;23:358 9.
[6] Fugh-Berman A. Herb drug interactions. Lancet 2000;355:134 8.
[7] Miller LG. Herbal medicinals. Selected clinical considerations focusing on known or potential drug herb interactions. Arch Intern Med
1998;158:2200 11.
[8] Izzo AA, Ernst E. Interactions between herbal medicines and prescribed drugs. Drugs 2001;61:2163 75.
[9] Williamson EM. Synergy and other interactions in phytomedicines.
Phytomedicine 2001;8:401 9.
[10] Scott GN, Elmer GW. Update on natural product drug interactions.
Am J Health-Syst Pharm 2002;59:339 47.
[11] Klepser TB, Klepser ME. Unsafe and potentially safe herbal therapies. Am J Health-Syst Pharm 1999;56:125 37.
[12] Smolinske SC. Dietary supplement drug interactions. J Am Med
Womens Assoc 1999;54:191 5.
[13] Ang-Lee MK, Moss J, Yuan C-S. Herbal medicines and perioperative
care. JAMA 2001;286:208 16.
[14] Barrett B, Kiefer D, Rabago D. Assessing the risks and benefits of
herbal medicine: an overview of scientific evidence. Altern Therap
1999;5:40 9.
13
14