Herbal Medicine for the Treatment

of Cardiovascular Disease
Clinical Considerations
Nick H. Mashour, MD; George I. Lin, MD; William H. Frishman, MD
H
erbs have been used as medical treatments since the beginning of civilization and
some derivatives (eg, aspirin, reserpine, and digitalis) have become mainstays of hu-
man pharmacotherapy. For cardiovascular diseases, herbal treatments have been used
in patients with congestive heart failure, systolic hypertension, angina pectoris, ath-
erosclerosis, cerebral insufficiency, venous insufficiency, and arrhythmia. However, many herbal
remedies used today have not undergone careful scientific assessment, and some have the poten-
tial to cause serious toxic effects and major drug-to-drug interactions. With the high prevalence of
herbal use in the United States today, clinicians must inquire about such health practices for car-
diac disease and be informed about the potential for benefit and harm. Continuing research is nec-
essary to elucidate the pharmacological activities of the many herbal remedies now being used to
treat cardiovascular diseases. Arch Intern Med. 1998;158:2225-2234
Since the beginning of human civiliza-
tion, herbs have been an integral part of
society, valued for both their culinary and
medicinal properties. Herbal medicine has
made many contributions to commercial
drug preparations manufactured today in-
cluding ephedrine from Ephedra sinica
(ma-huang), digitoxin fromDigitalis pur-
purea (foxglove), salicin (the source of as-
pirin) from Salix alba (willow bark), and
reserpine from Rauwolfia serpentina
(snakeroot), to name just a few. A natu-
rally occurring -adrenergic blocking agent
with partial agonism has been identified
in an herbal remedy.
1
The recent discov-
ery of the antineoplastic drug paclitaxel
from Taxus brevifolia (pacific yew tree)
stresses the role of plants as a continuing
resource for modern medicine.
However, with the development of
patent medicines in the early part of the
20th century, herbal medicine has been
losing ground to new synthetic medi-
cines touted by scientists and physicians
to be more effective and reliable. Never-
theless, about 3% of English-speaking
adults in the United States still report
having used herbal remedies in the pre-
ceding year.
2
This figure is probably
much higher for nonEnglish-speaking
Americans. Despite this heavy use of
herbal medicines in the United States,
health practitioners often fail to ask
about their use when taking clinical his-
tories. It is imperative that physicians
become more aware of the wide array of
herbal medicines available, as well as
learning more about their beneficial and
adverse effects.
3
Part of the problemfor both consum-
ers and physicians has been the paucity of
scientific data on herbal medicines used in
the United States.
4
As a result, those who
wish to obtain factual information regard-
ing the therapeutic use or potential harm
of herbal remedies would have to obtain it
frombooks and pamphlets, most of which
base their information on traditional repu-
tation rather than relying on existing sci-
entific research. One may wonder why the
herbal industry never chose tosimply prove
its products safe and effective. The answer
is primarily economical. With the slim
chance of patent protection for the many
herbs that have been in use for centuries,
pharmaceutical companies have not pro-
vided financial support for research on the
merits of herbal medicine.
5
At the same
time, the National Institutes of Healthhave
only been able to offer limited funding for
this purpose.
From the Department of Medicine, The Albert Einstein College of Medicine, Bronx, NY
(Dr Mashour); the Department of Family Medicine, Columbia Presbyterian Medical
Center, New York, NY (Dr Lin); and the Departments of Medicine and Pharmacology,
New York Medical College/Westchester Medical Center, Valhalla, NY (Dr Frishman).
REVIEW ARTICLE
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2225
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
This review examines herbal
medicines that affect the cardiovas-
cular system both in terms of effi-
cacy and safety as gleaned from the
scientific literature that is avail-
able. These herbs are categorizedun-
der the primary diseases they treat.
However, most herbal medicines
have multiple cardiovascular ef-
fects that frequently overlap. The
purpose of this organization is to
simplify, not topigeonhole herbs un-
der specific diseases. In general, the
dilution of active components in
herbal medicines results infewer ad-
verse and toxic effects in compari-
son with the concentration of ac-
tive components in the allopathic
medicines. However, these adverse
effects and drug interactions should
not be overlooked; cardiovascular
disease is a serious healthhazardand
no herbal remedy regimenshould be
initiated without careful consider-
ationof its potential impact (Table).
CONGESTIVE HEART FAILURE
A number of herbs contain potent
cardioactive glycosides, which have
positive inotropic actions on the
heart. The drugs digitoxin, derived
fromeither Dpurpurea (foxglove) or
Digitalis lanata, and digoxin, de-
rived fromDlanata alone, have been
used in the treatment of congestive
heart failure for many decades. Car-
diac glycosides have a lowtherapeu-
tic index, and the dose must be ad-
justed to the needs of each patient.
The only way to control dosage is to
use standardized powdered digi-
talis, digitoxin, or digoxin. When
12 different strains of D lanata
plants were cultured and exam-
ined, their total cardenolide yield
ranged from 30 to almost 1000
nmol/1 g.
6
As is evident, treating
congestive heart failure with non-
standardized herbal drugs would be
dangerous and foolhardy.
Some common plant sources of
cardiac glycosides include D pur-
purea (foxglove, already men-
tioned), Adonis microcarpa and Ado-
nis vernalis (adonis), Apocynum
cannabinum (black Indian hemp),
Asclepiascurassavica (redheadedcot-
ton bush), Asclepias friticosa (bal-
loon cotton), Calotropis precera
(kings crown), Carissa spectabilis
(wintersweet), Cerebra manghas (sea
mango), Cheiranthus cheiri (wall-
flower), Convallaria majalis (lily of
the valley, convallaria), Cryptoste-
gia grandiflora (rubber vine), Hel-
leborus niger (black hellebore), Hel-
leborus viridus, Nerium oleander
(oleander), Plumeria rubra (frangi-
pani), Selenicerus grandiflorus (cac-
tus grandiflorus), Strophanthus his-
pidus and Strophanthus kombe
(strophanus), Thevetia peruviana
(yellowoleander), and Urginea mar-
itima (squill).
5,7-15
Even the venom
glands of the animal Bufo marinus
(cane toad) contain cardiac glyco-
sides.
8
Recently, the digitalislike ste-
roid in the venom of the B marinus
toad was identified as a previously
described steroid, marinobufa-
genin. Marinobufagenin demon-
strated high digoxinlike immuno-
reactivity and was antagonized with
an antidigoxin antibody.
16
Accidental poisonings andeven
suicide attempts with ingestion of
cardiac glycosides are abundant in
the medical literature.
17-21
Some
herbal remedies (eg, Siberian gin-
seng) can elevate synthetic digoxin
drug levels and cause toxic ef-
fects.
22
Inthe United States, there are
about 15000intoxications due toac-
cidental or intentional ingestion of
poisonous plants annually.
23
In1993,
2388 toxic exposures in the United
States were reported to be due to
plant glycosides. Of these, the larg-
est percentage were attributedto ole-
ander (ie, 25%).
24
In the case of ole-
ander, all plant tissues, including the
seeds, roots, stems, leaves, berries,
and blossoms, are considered ex-
tremely toxic.
19
In fact, death in hu-
mans has beenreportedfollowing in-
gestionof as little as 1 oleander leaf.
25
The clinical manifestations of ole-
ander intoxication, as well as other
natural glycosides, is virtually iden-
tical to digoxin overdose. Morbid-
ity and mortality are mainly related
to cardiotoxic adverse effects that
usually include life-threatening ven-
tricular tachyarrhythmias, brady-
cardia, and heart block. The diag-
nosis should rely on the clinical
presentation of unexplained hyper-
kalemia, and cardiac, neurologic,
and gastrointestinal symptoms.
19
The diagnosis can be further
supported by the detection of the
substance digoxin in a radioimmu-
noassay for digoxin. However, the
extent of cross-reactivity betweenthe
cardiac glycosides from herbal
sources and antibodies used in the
radioimmunoassays has not been
clearly defined.
26
For this reason, di-
goxin assays may serve to confirm
the suspected diagnosis but not to
quantify the severity. Once the di-
agnosis has beenestablished, the use
of digoxin-specific Fab antibody
fragments may be helpful in the
treatment of severe intoxication.
Other modalities, such as dialysis,
cannot be easily facilitated be-
cause, like digoxin, natural glyco-
sides are distributed extensively into
peripheral tissues.
HYPERTENSION
The root of R serpentina (snake-
root), the natural source of the alka-
loid reserpine, has been a Hindu
Ayurvedic remedy since ancient
times. In 1931, Indian literature first
describedthe use of Rserpentina root
for the treatment of hypertension
and psychoses; however, the use of
Rauwolfia alkaloids in Western
medicine did not begin until the
Herbs for Cardiovascular Conditions With Severe Adverse Reactions
or Notable Drug Interactions*
Herbal Medicine Adverse Reaction/Drug Interaction Treatment
Natural cardiac glycosides
(20 plant sources)
Ventricular tachyarrhythmia, bradycardia,
and heart block
Digoxin-specific
Fab antibody
Veratrum
(hellebore)
Bradycardia, A-V dissociation, hypotension,
and (rarely) seizures
ECG changes responsive
to atropine
Crataegus
(hawthorn)
Potentiates digitalis activity NA
Salvia miltiorrhiza
(dan-shen)
Potentiates warfarin activity NA
Aesculus hippocastanum
(horse chestnut)
Renal and hepatic toxic effects Dialysis to reduce
toxic levels
*A-V indicates arteriovenous anastomosis; ECG, electrocardiographic; and NA, data not applicable.
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2226
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
mid1940s.
27
Bothstandardizedwhole
root preparations of Rserpentina and
its reserpine alkaloid are officially
monographed in the United States
Pharmacopeia.
28
A powdered whole
root of 200to300mgorallyis equiva-
lent to 0.5 mg of reserpine.
29
Reserpine was one of the first
drugsusedonalargescaletotreat sys-
temichypertension. It actsbyirrevers-
ibly blocking the uptake of biogenic
amines (norepinephrine, dopamine,
andserotonin) inthestoragevesicles
of central andperipheral adrenergic
neurons, thus leaving the catechol-
amines to be destroyed by the intra-
neuronal monoamine oxidase inthe
cytoplasm. Thedepletionof catechol-
amines accounts for reserpines sym-
patholytic and antihypertensive ac-
tions. Reserpineseffectsarelonglast-
ing, since recovery of sympathetic
function requires synthesis of new
storage vesicles, whichtakes days to
weeks. Reserpine lowers bloodpres-
surebydecreasingcardiacoutput, pe-
ripheral vascularresistance, heart rate,
and renin secretion. With the intro-
duction of other antihypertensive
drugs withfewer central nervous sys-
temadverse effects, the use of reser-
pine has diminished. The daily oral
dose of reserpine shouldbe 0.25 mg
or less, andas littleas 0.05mgif given
withadiuretic. Usingthewholeroot,
theusual adult doseis 50to200mg/d
administered once daily or in 2 di-
vided doses.
27-29
Rauwolfia alkaloids are contra-
indicated for use in patients with
previously demonstrated hypersen-
sitivity to these substances, in pa-
tients with a history of mental de-
pression (especially with suicidal
tendencies), in patients with active
peptic ulcer disease or ulcerative co-
litis, and in patients receiving elec-
troconvulsive therapy. The most
common adverse effects are seda-
tion and inability to concentrate and
perform complex tasks. Reserpine
may cause mental depression, some-
times resulting insuicide, andits use
must be discontinuedat the first sign
of depression. Reserpines sympa-
tholytic effect and its enhancement
of parasympathetic actions account
for its well-describedadverse effects:
nasal congestion, increased gastric
secretion, and mild diarrhea.
27-30
Stephania tetrandra is an herb
sometimes used in traditional Chi-
nese medicine to treat hyperten-
sion. Tetrandrine, an alkaloid ex-
tract of S tetrandra, has been shown
to be a calcium ion channel antago-
nist, paralleling the effects of verap-
amil. Tetrandrine blocks TandLcal-
cium channels, interferes with the
bindingof diltiazemandmethoxyver-
apamil at calcium-channel binding
sites, andsuppresses aldosterone pro-
duction.
31,32
Aparenteral dose(15mg/
kg) of tetrandrine in conscious rats
decreases mean, systolic, and dias-
tolic blood pressures for more than
30 minutes; however, an intrave-
nous 40-mg/kg dose killedthe rats by
myocardial depression. In stroke-
prone hypertensive rats, anoral dose
of 25or 50mg/kg produceda gradual
and sustained hypotensive effect af-
ter 48hours without affecting plasma
renin activity.
33
In addition to its
cardiovascular actions, tetrandrine
has reported antineoplastic, immu-
nosuppressive, and mutagenic
effects.
31
Tetrandrine is 90% protein-
bound with an elimination half-life
of 88 minutes, according to dog
studies; however, rat studies have
shown a sustained hypotensive ef-
fect for more than 48 hours after a
25- or 50-mg oral dose. Tetran-
drine causes liver necrosis in dogs
orally administered 40 mg/kg of tet-
randrine 3 times weekly for 2
months, reversible swelling of liver
cells with a 20-mg/kg dose, and no
observable changes with a 10-
mg/kg dose.
31
Given the evidence of
hepatotoxicity, many more studies
are necessary to establish a safe dos-
age of tetrandrine in humans.
More recently, tetrandrine has
been implicated in an outbreak of
rapidly progressive renal failure,
termed Chinese herb nephropathy.
Numerous individuals developedthe
condition after using a combina-
tion of several Chinese herbs as part
of a dieting regimen. It has been hy-
pothesized that the cause may be at-
tributed to misidentification of S tet-
randra; nonetheless, questions still
remain as to the role of tetrandra in
the development of this serious toxic
effect.
34-37
Theroot of Lingusticumwallichii
is used in traditional Chinese medi-
cineas acirculatorystimulant, hypo-
tensive drug, and sedative.
38
Tetra-
methylpyrazine, the active constitu-
ent extractedfromLwallichii, inhibits
platelet aggregationinvitroandlow-
ers bloodpressure byvasodilationin
dogs. With its actions independent
of the endothelium, tetramethylpyr-
azines vasodilatoryeffect is mediated
by calciumchannel antagonismand
nonselective antagonism of -
adrenergicreceptors. Someevidence
suggeststhat tetramethylpyrazineacts
onthepulmonaryvasculature.
31
Cur-
rently, there is insufficient informa-
tiontoevaluatethesafetyandefficacy
of this herbal medicinal.
Uncaria rhynchophylla is some-
times used in traditional Chinese
medicine to treat hypertension. Its
indole alkaloids, rhynchophylline
and hirsutine, are thought to be the
active principles of U rhynchophyl-
las vasodilatory effect. The mecha-
nismof Urhynchophyllas actions is
unclear. Some studies point to an al-
teration in calcium ion flux in re-
sponse to activation, whereas oth-
ers point to hirsutines inhibition of
nicotine-induced dopamine re-
lease.
31
One invitro study has shown
Urhynchophylla extract relaxes nor-
epinephrine-precontracted rat aorta
through endothelium-dependent
and -independent mechanisms. For
the endothelium-dependent com-
ponent, Urhynchophylla extract ap-
pears to stimulate endothelium-
derived relaxing factor and/or nitric
oxide release without involving mus-
carinic receptors.
39
Also, in vitro and
in vivo studies have shown that
rhynchophylline can inhibit plate-
let aggregation and reduce platelet
thromboses induced with collagen
or adenosine diphosphate plus epi-
nephrine.
31
Safety and efficacy can-
not be evaluated at this time be-
cause of a lack of clinical data.
Veratrum (hellebore) is a pe-
rennial herb grown in many parts of
the world. Varieties include Vera-
trum viride from Canada and the
eastern United States, Veratrumcali-
fornicum from the western United
States, VeratrumalbumfromAlaska
and Europe, and Veratrum japoni-
cum from Asia. All Veratrum plants
contain poisonous alkaloids known
to cause vomiting, bradycardia, and
hypotension. Most cases of Vera-
trum poisonings are due to mis-
identification with other plants.
Although once a treatment for hy-
pertension, the use of Veratrum al-
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2227
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
kaloids has lost favor owing to a low
therapeutic index and unaccept-
able toxicity, as well as the intro-
duction of safer antihypertensive
drug alternatives.
40
Veratrum alkaloids enhance
nerve and muscle excitability by in-
creasing sodium ion conductivity.
They act on the posterior wall of the
left ventricle and the coronary si-
nus baroreceptors, causing reflex hy-
potension and bradycardia via the
vagus nerve (Bezold-Jarisch re-
flex). Nausea and vomiting are sec-
ondary to the alkaloids actions on
the nodose ganglion.
40
The diagnosis of Veratrumtox-
icity is established by history, iden-
tificationof theplant, andstrongclini-
cal suspicion. Clinical symptomsusu-
ally occur quickly, often within 30
minutes.
41
Treatment is mainly sup-
portive and directed at controlling
bradycardia and hypotension.
Veratrum-inducedbradycardia usu-
allyresponds totreatment withatro-
pine; however, thebloodpressurere-
sponse to atropine is more variable
andrequires theadditionof pressors.
Other electrocardiographicchanges,
suchas atrioventricular dissociation,
mayalsobereversiblewithatropine.
42
Seizures are a rare complicationand
maybetreatedwithconventional an-
ticonvulsants. For patients withpre-
existingcardiac disease, the use of -
agonists or pacingmaybe necessary.
Nausea may be controlledwithphe-
nothiazineantiemetics. Recoveryusu-
ally occurs within 24 to 48 hours.
40
Evodia rutaecarpa (wu-chu-
yu) is a Chinese herbal drug that has
been used as a treatment for hy-
pertension. It contains an active
vasorelaxant component called ru-
taecarpine that can cause endothe-
lium-dependent vasodilation in
experimental models.
43
ANGINA PECTORIS
Crataegus hawthorn, a name encom-
passing many Crataegus species
(such as Crataegus oxyacantha and
Crataegus monogyna inthe West and
Crataegus pinnatifida in China) has
acquired the reputation in modern
herbal literature as an important
tonic for the cardiovascular system
that is particularly useful for an-
gina. Crataegus leaves, flowers, and
fruits contain a number of biologi-
cally active substances, suchas oligo-
meric procyanins, flavonoids, and
catechins. From current studies,
Crataegus extract appears tohave an-
tioxidant properties and can in-
hibit the formation of throm-
boxane as well.
44,45
Also, Crataegus extract antago-
nizes the increases incholesterol, tri-
glyceride, andphospholipidlevels in
low-density lipoprotein (LDL) and
very low-density lipoprotein in rats
fed a hyperlipidemic diet; thus, it
may inhibit the progression of ath-
erosclerosis.
46
This hypocholester-
olemic action may be due to an up-
regulation of hepatic LDL receptors
resulting in greater influx of plasma
cholesterol into the liver. Cratae-
gus also prevents cholesterol accu-
mulation in the liver by enhancing
cholesterol degradation to bile ac-
ids, as well as suppressing choles-
terol biosynthesis.
47
Accordingtoanother study, Cra-
taegus extract, in high concentra-
tions, has a cardioprotective effect on
ischemic-reperfused hearts without
causinganincrease incoronaryblood
flow.
48
On the other hand, oral and
parenteral administration of oligo-
meric procyanins of Crataegus has
been shown to lead to an increase in
coronary blood flowin both cats and
dogs.
49,50
Double-blind clinical trials
have demonstratedsimultaneous car-
diotropic and vasodilatory actions of
Crataegus.
51
Inessence, Crataegus in-
creases coronaryperfusion, has a mild
hypotensive effect, antagonizes ath-
erogenesis, and has positive inotro-
pic and negative chronotropic ac-
tions.
46,52
In a recent multicenter,
placebo-controlled, double-blind
study, an extract of Crataegus was
shown to clearly improve the car-
diac performance of patients with
New York Heart Association class II
heart failure. In this study, the pri-
mary parameter analyzed was the
heart rate product (systolic blood
pressure heart rate).
53
Recent stud-
ies have suggested that the mecha-
nism of cardiac action for Crataegus
species may be due to the inhibition
of the 3, 5-cyclic adenosine mono-
phosphate phosphodiesterase.
54
Hawthorn is relatively devoid
of adverse effects. In fact, in com-
parison with other inotropic drugs
such as epinephrine, amrinone, mil-
rinone, and digoxin, Crataegus has
a potentially reduced arrhythmo-
genic risk because of its ability to
prolong the effective refractory pe-
riod, while the other drugs men-
tioned previously all shortenthis pa-
rameter.
55,56
Also, it should be noted
that concomitant use of hawthorn
with digitalis can markedly en-
hance the activity of digitalis.
5,57
Un-
doubtedly, more studies are needed
to show that hawthorn can be used
safely and effectively.
Because of its resemblance to
Panax ginseng (Asianginseng), Panax
notoginseng has acquired the com-
mon name of pseudoginseng, espe-
cially since it is oftenanadulterant of
P ginseng preparations. In tradi-
tional Chinese medicine, the root of
P notoginsengis usedfor analgesia and
hemostasis. It is alsooftenusedinthe
treatment of patients withangina and
coronary artery disease.
38
Panax no-
toginseng has beendescribed as a cal-
cium ion channel antagonist in vas-
cular tissue. More specifically, its
pharmacological action may be as a
novel and selective calcium ion an-
tagonist that does not interact withthe
L-type calciumionchannel but rather
may interact with the receptor-
operated calcium ion channel.
58
Althoughclinical trials are lack-
ing, in vitro studies using P notogin-
seng suggest possible cardiovascular
effects. One study that used purified
notoginsenoside R1, extracted from
P notoginseng, on human left umbili-
cal vein endothelial cells showed a
dose- and time-dependent synthesis
of tissue-type plasminogen activat-
ing factor without affecting the syn-
thesis of plasminogen activating in-
hibitor. Thus, fibrinolytic parameters
were enhanced.
59
Another study sug-
gests that Pnotoginseng saponins may
inhibit atherogenesis by interfering
with the proliferation of smooth
muscle cells.
60
In vitro and in vivo
studies using rats and rabbits dem-
onstrate that P notoginseng may be
useful as an antianginal drug, since
it dilates coronary arteries in all con-
centrations. The role of P notogin-
seng in the treatment of hyperten-
sionis less certain, since Pnotoginseng
causes vasodilation or vasoconstric-
tion depending on the concentra-
tion and target vessel.
61
The results
of these in vitro and in vivo studies
are encouraging; however, clinical tri-
als will be necessary to make a more
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2228
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
informed decision regarding the use
of P notoginseng.
Sal vi a mi l t i orrhi za (dan-
shen), a relative of the Western sage
Salvia officinalis, is native to China.
In traditional Chinese medicine, the
root of S miltiorrhiza is used as a cir-
culatory stimulant, sedative, and
cooling drug.
38
Salvia miltiorrhiza
may be useful as an antianginal drug
because it has been shown to dilate
coronary arteries in all concentra-
tions, similar to P notoginseng. Also,
S miltiorrhiza has variable action on
other vessels depending on its con-
centration, so it may not be as help-
ful in treating hypertension.
61
In
vitro, S miltiorrhiza, in a dose-
dependent fashion, inhibits plate-
let aggregation and serotonin re-
lease induced by either adenosine
diphosphate or epinephrine, which
is thought to be mediated by an in-
crease in platelet cyclic adenosine
monophosphate caused by S milti-
orrhizas inhibition of cyclic aden-
osine monophosphate phosphodi-
esterase.
62
Salvia miltiorrhiza appears
to have a protective action on ische-
mic myocardium, enhancing the
recovery of contractile force on
reoxygenation.
63
More recently, S
miltiorrhiza has been shown to pro-
tect myocardial mitochondrial mem-
branes fromischemia-reperfusionin-
jury and lipid peroxidation because
of its free radicalscavenging ef-
fects.
64
Qualitatively and quantita-
tively, a decoction of S miltiorrhiza
was as efficacious as the more ex-
pensive isolated tanshinones.
59
Clinical trials will be neces-
sary to evaluate the safety and effi-
cacy of S miltiorrhiza. Of note, it has
been observed clinically that when
S miltiorrhiza and warfarin sodium
are coadministered, there is an in-
creased incidence in warfarin-
relatedadverse effects; inrats S milti-
orrhiza was shown to increase the
plasma concentrations of warfarinas
well as the prothrombin time.
65
ATHEROSCLEROSIS
In addition to its use in the culi-
nary arts, garlic (Alliumsativum) has
been valued for centuries for its me-
dicinal properties. Garlic is one of
the herbal medicines that has been
examined more closely by the sci-
entific community. In recent de-
cades, research has focused on
garlics use in preventing atheroscle-
rosis. Garlic, like many of the other
herbal medicines discussed previ-
ously, has demonstrated multiple
beneficial cardiovascular effects. A
number of studies have demon-
stratedthese effects that include low-
ering blood pressure, inhibiting
platelet aggregation, enhancing fi-
brinolytic activity, reducing serum
cholesterol and triglyceride levels,
and protecting the elastic proper-
ties of the aorta.
Consumption of large quanti-
ties of fresh garlic (0.25 to 1.0 g/kg
or about 5-20 average sized 4-g
cloves in a person weighing 78.7 kg)
has been shown to produce the ben-
eficial effects mentioned earlier.
66
In
support of this, a recent double-
blind cross-over study was con-
ducted on moderately hypercholes-
terolemic men that compared the
effects of 7.2 g of aged garlic ex-
tract withplacebo onblood lipid lev-
els. This study found that there was
a maximal reduction of 6.1% in to-
tal serumcholesterol levels and4.6%
in LDL cholesterol levels with gar-
lic compared with placebo.
67
However, despite positive evi-
dence from numerous trials, some
investigators have been hesitant to
outright endorse the routine use of
garlic for cardiovascular disease be-
cause many of the published stud-
ies had methodological shortcom-
i ngs,
6 6 , 6 8 - 7 2
perhaps because
constituent trials were small, lack-
ing statistical power. Also, inappro-
priate methods of randomization,
lack of dietary run-in period, short
duration, or failure to undertake in-
tention-to-treat analysis may ex-
plain the cautious acceptance of pre-
vious meta-analyses.
73
In fact, one
recent study found no demon-
strable effect of garlic ingestion on
lipid and lipoprotein levels. This
study used a cross-over design pro-
tected by a washout period to re-
duce between-subject variability as
well as close assessment and report-
ing of dietary behavior, which had
been lacking in previous trials.
74
An-
other study found no effect of gar-
lic on cholesterol absorption, cho-
lesterol synthesis, or cholesterol
metabolism.
71
As is evident, the pre-
cise extent of garlics impact on ath-
erosclerosis remains controversial;
larger, more rigorously designed tri-
als may be necessary to better de-
termine its utility in preventing car-
diovascular disease.
Garlic has also been studied in
hypertensive patients as a blood
pressurelowering agent. Similar to
its lipid effects, no conclusive stud-
ies have been conducted and many
methodological shortcomings exist
in study designs. The results of one
meta-analysis that considered 8 dif-
ferent trials suggest some clinical use
for patients with mild hyperten-
sion, but there is insufficient evi-
dence to recommend its use as rou-
tine clinical therapy.
68
Garlic has also
been shown to possess antiplatelet
activity. In the past, this action was
mostly documentedinvitro.
75
Anew
study examinedthe effect of the con-
sumption of a fresh clove of garlic
on platelet thromboxane produc-
tion and showed that after 26 weeks,
serumthromboxane levels were re-
duced about 80%.
76
This may prove
to be beneficial in the prevention of
thrombosis in the future. Recently,
the effect of long-term garlic intake
on the elastic properties of the aorta
was also studied. Participants in the
trial (limited to those aged 50-80
years) consumed 300 mg/d of stan-
dardizedgarlic powder for more than
2 years. The results showed that the
pulse-wave velocity and standard-
ized elastic vascular resistance of the
aorta were lower in the garlic group
than in the control group. Conse-
quently, long-termgarlic powder in-
take may have a protective effect on
the elastic properties of the aorta re-
lated to aging.
77
In these ways, gar-
lic has shown numerous beneficial
cardiovascular effects that need to be
investigated further to determine its
therapeutic utility.
Intact cells of garlic bulbs in-
clude an odorless, sulfur-contain-
ing amino acid known as allinin.
When garlic is crushed, allinin
comes into contact with allinase,
which converts allinin to allicin. Al-
licin has potent antibacterial prop-
erties, but it is also highly odorifer-
ous and unstable. Ajoenes, self-
condensation products of allicin,
appear to be responsible for garlics
antithrombotic activity. Most au-
thorities now agree that allicin and
its derivatives are the active con-
stituents of garlics physiological ac-
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2229
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
tivity. Fresh garlic releases allicin in
the mouth during the chewing pro-
cess. Dried garlic preparations lack
allicin but contain allinin and allin-
ase. Since allinase is inactivated in
the stomach, dried garlic prepara-
tions should be coated with enteric
so that they pass through the stom-
achinto the small intestine where al-
linin can be enzymatically con-
verted to allicin. Few commercial
garlic preparations are standard-
ized for their allicin yield based on
allinin content, hence making their
effectiveness less certain.
5
How-
ever, one double-blind, placebo-
controlled study involving 261 pa-
tients for 4 months using one
800-mg tablet of garlic powder daily,
standardized to 1.3% allinin con-
tent, demonstrated significant re-
ductions in total cholesterol (12%)
and triglyceride levels (17%).
78
Aside froma garlic odor on the
breathandbody, moderate garlic con-
sumption causes few adverse ef-
fects. However, consumption in ex-
cess of 5 cloves daily may result in
heartburn, flatulence, and other gas-
trointestinal disturbances. Some
people have reported allergic reac-
tions to garlic, most commonly al-
lergic contact dermatitis. Patch test-
ing with 1% diallyl disulfide is
recommended when garlic allergy is
suspected.
79
Because of its antithrom-
botic activity, garlic should be used
withcautioninpeople taking oral an-
ticoagulants concomitantly.
5,80
The resinof Commiphora mukul
(gugulipid), a small, thorny tree na-
tive to India, has long been used in
Ayurvedic medicine totreat lipiddis-
orders. The primary mechanism of
action of gugulipid is through an in-
crease in the uptake and metabo-
lismof LDLcholesterol by the liver.
81
In a double-blind, cross-over study
completed in125 patients taking gu-
gulipid compared with 108 patients
taking clofibrate, the average de-
crease in serum cholesterol and tri-
glyceride levels was 11%and 16.8%,
respectively, with gugulipid com-
pared with 10% and 21.6%, respec-
tively, withclofibrate. Ingeneral, hy-
per chol est er ol emi c pat i ent s
responded more favorably to gugu-
lipid therapy thanhypertriglyceride-
mic patients.
82
Moreover, it was
shown in another randomized,
double-blind trial that C mukul also
decreased LDL cholesterol levels by
12.5%andthe total cholesterolhigh-
density lipoprotein cholesterol ratio
by11.1%, whereas the levels were un-
changed in the placebo group.
83
Besidesbeingpotentiallyaseffec-
tive in lowering blood lipid levels as
modernhyperlipidemicdrugs, gugu-
lipid may even be safer. In the trial
mentionedpreviously,compliancewas
greater than 96%, with only the ad-
verse effects of headache, mild nau-
sea, andhiccups noted.
83
However, it
has been shown that gugulipid may
affect the bioavailabilityof other car-
diovasculardrugs, namely, proprano-
lol hydrochloride and diltiazem hy-
drochloride. Gugulipidsignificantly
reduced the peak plasma concentra-
tionandarea under the curve of both
thesedrugs, whichmayleadtodimin-
ishedefficacyornonresponsiveness.
84
Undoubtedly, gugulipid is a natural
lipid-loweringdrugwithpotential for
therapeutic use, but rigorous, larger
clinical trials will be necessarytofur-
ther evaluateitssafetyandefficacybe-
fore it can be endorsed as an alterna-
tive therapy for hyperlipidemia and
prevention of atherosclerosis.
Maharishi amrit kalash-4 and
Maharishi amrit kalash-5 are 2 com-
plex herbal mixtures with signifi-
cant antioxidant properties that have
been shown to inhibit LDL oxida-
tion in patients with hyperlipid-
emia. In experimental studies, the
herbal mixtures havealsobeenshown
to inhibit enzymatic- and nonenzy-
matic-inducedmicrosomal lipidper-
oxidation and platelet aggregation.
85
CEREBRAL AND PERIPHERAL
VASCULAR DISEASE
Having existed for more than 200
million years, Ginkgo biloba (maid-
enhair tree) was apparently saved
fromextinction by human interven-
tion, surviving inFar Easterntemple
gardens while disappearing for cen-
turies in the West. It was reintro-
duced to Europe in 1730 and be-
came a favorite ornamental tree.
38,86
Although the root and kernels of G
biloba have long been used in tradi-
tional Chinese medicine, the tree
gained attention in the West dur-
ing the 20th century for its medici-
nal value after a concentrated ex-
t ract of G bi l oba l eaves was
developed in the 1960s. At least 2
groups of substances within G bi-
loba extract (GBE) demonstrate ben-
eficial pharmacological actions. The
flavonoids reduce capillary perme-
ability as well as fragility and serve
as free radical scavengers. The ter-
penes (ie, ginkgolides) inhibit plate-
let-activating factor, decrease vas-
cular resistance, and improve
circulatory flow without apprecia-
bly affecting bloodpressure.
57,87
Con-
tinuing research appears to sup-
port the primary use of GBE for
treating cerebral insufficiency andits
secondary effects on vertigo, tinni-
tus, memory, and mood; also, GBE
appears to be useful for treating pe-
ripheral vascular disease, includ-
ing diabetic retinopathy and inter-
mittent claudication.
5,57,87-91
In a randomized, placebo-
controlled, double-blind study, EGb
761, which is a standardized ex-
tract of G biloba with respect to its
flavonol glycoside and terpene lac-
tone content, was shown to signifi-
cantly decrease the areas of ische-
mia as measured by transcutaneous
partial pressure of oxygenduring ex-
ercise. Because of its rapid anti-
ischemic action, EGb 761 may be
valuable in the treatment of inter-
mittent claudication and periph-
eral artery disease in general.
92
Also, studies have been exam-
ining the cardioprotective efficacy of
EGb 761 in regard to its antifree
radical action in myocardial isch-
emiareperfusion injury. In vitro
studies with animal models have
shown that this compound may ex-
ert suchaneffect.
93,94
Aclinical study
of 15 patients undergoing coronary
bypass surgery demonstrated that
oral EGb 761 therapy may limit free
radicalinduced oxidative stress oc-
curring in the systemic circulation
and at the level of the myocardium
during these operations.
95
It re-
mains to be studied whether ex-
tracts of G biloba may be used as
pharmacological adjuvants to limit
tissue damage and metabolic alter-
ations following coronary bypass
surgery, coronary angioplasty for
acute myocardial infarctions, or even
in managing coronary thrombosis.
Although approved as a drug
in Europe, Ginkgo is not approved
in the United States and is instead
marketed as a food supplement,
usually supplied as 40-mg tablets
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2230
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
of extract. Since most of the inves-
tigations examining the efficacy of
GBEs used preparations such as
EGb 761 or LI 1370, the bio-
equivalence of other GBE prod-
ucts has not been established. The
recommended dosage in Europe is
one 40-mg tablet taken 3 times
daily with meals (120 mg/d).
5,87
Adverse effects due to GBE are
rare but can include gastrointesti-
nal disturbances, headache, and
allergic skin rash.
5,87
Known mostly as a culinary
spice and flavoring agent, Rosmari-
nus officinalis (rosemary) is listed in
many herbal sources as a tonic and
all-around stimulant. Tradition-
ally, rosemary leaves are said to en-
hance circulation, aid digestion, el-
evate mood, andboost energy. When
applied externally, the volatile oils
are supposedly useful for arthritic
conditions and baldness.
5
Although research on rose-
mary is scant, some studies have fo-
cused on antioxidant effects of di-
terpenoids, especially carnosic acid
and carnosol, isolated from rose-
mary leaves. In addition to having
antineoplastic effects, antioxidants
in rosemary have been credited with
stabilizing erythrocyte membranes
and inhibiting superoxide gener-
ation and lipid peroxidation.
96,97
Essential oils of rosemary have dem-
onstrated antimicrobial, hypergly-
cemic, and insulin-inhibiting prop-
erties.
98,99
Rosemary leaves contain
high amounts of salicylates, and its
flavonoid pigment diosmin is re-
ported to decrease capillary perme-
ability and fragility.
57,100,101
Despite the conclusions de-
rived fromin vitro and animal stud-
ies, the therapeutic use of rosemary
for cardiovascular disorders re-
mains questionable, because few, if
any, clinical trials have been con-
ducted using rosemary. Because of
the lack of studies, no conclusions
can be reached regarding the use of
the antioxidants of rosemary in in-
hibiting atherosclerosis. Although
external application may cause cu-
taneous vasodilation fromthe coun-
terirritant properties of rosemarys
essential oils, there is no evidence to
support any prolonged improve-
ment in peripheral circulation.
5
While rosemary does have some car-
minative properties, it may also
cause gastrointestinal and kidney
disturbances in large doses.
5,101
Un-
til more studies are done, rosemary
should probably be limited to its use
as a culinary spice and flavoring
agent rather than as a medicine.
VENOUS INSUFFICIENCY
The seeds of horse chestnut, Aescu-
lus hippocastanum, have long been
used in Europe to treat venous dis-
orders suchas varicose veins. The sa-
ponin glycoside aescin from horse
chestnut extract (HCE) inhibits the
activityof lysosomal enzymes thought
to contribute to varicose veins by
weakening vessel walls and increas-
ing permeability, which result in di-
lated veins and edema.
5
In fact, re-
cent research has shown that A
hippocastanum inhibits only against
hyaluronidase but not elastase, and
this activity is linkedmainly tothe sa-
ponin escin.
102
In animal studies,
HCE, inadose-dependent fashion, in-
creases venous tone, venous flow, and
lymphatic flow. It also antagonizes
capillary hyperpermeability in-
duced by histamine, serotonin, or
chloroform. This extract has been
shown to decrease edema forma-
tion of lymphatic and inflamma-
tory origin. Horse chestnut extract
has antiexudative properties, sup-
pressing experimentally induced
pleurisyandperitonitis byinhibiting
plasma extravasationandleukocyte
emigration, and its dose-dependent
antioxidant properties caninhibit in
vitro lipid peroxidation.
103,104
Ran-
domized, double-blind, placebo-
controlledtrials withHCEshoware
eduction in edema, measured using
plethysmography.
105,106
In another recent random-
ized, placebo-controlled study, the
efficacy and safety of class 2 com-
pression stockings and dried HCE
were compared. Both HCE and the
compression stockings decreased
lower leg edema after 12 weeks of
therapy; the results showed an av-
erage 43.8-mL reduction with HCE
and 46.7-mL with compression
stockings, while the placebo group
showed an increase of 9.8 mL. Both
HCEand compression therapy were
well tolerated, with no serious ad-
verse effects. This study may indi-
cate that both of these modalities are
reasonable alternatives for the effec-
tive treatment of patients with
chronic venous insufficiency.
107
Also,
HCE has been shown to markedly
improve other symptoms associ-
ated with chronic venous insuffi-
ciency, such as pain, tiredness, itch-
ing, and tension in the swollen leg,
in a case-observation study.
108
Aside
from effects on venous insuffi-
ciency, prophylactic use of HCE has
been thought to decrease the inci-
dence of thromboembolic compli-
cations of gynecological surgery.
However, since this issue is still con-
troversial,
109
this does not appear to
be the case.
109
StandardizedHCEispreparedas
anaqueous alcohol extract of 16%to
21% of triterpene glycosides, calcu-
latedas aescin. The usual initial dos-
ageis 90to150mg/dof aescin, which
may be reduced to 35 to 70 mg/d if
clinical benefit is seen.
5
Standardized
HCEpreparations arenot availablein
the United States, but nonstandard-
ized products may be available.
Some manufacturers promote
the use of topical preparations of HCE
for treatment of varicose veins as well
as hemorrhoids; however, at least one
study has demonstrated poor aescin
distributionat sites other thantheskin
andmuscle tissues underlyingthe ap-
plication site.
110
Moreover, the in-
volvement of arterioles and veins in
the pathophysiology of hemor-
rhoids makes theeffectiveness of HCE
doubtful, since HCE has no known
effects onthe arterial circulation. For
now, researchstudies have yet tocon-
firmany clinical effectiveness of topi-
cal HCE preparations.
Althoughadverse effects are un-
common, HCE may cause gastroin-
testinal irritation. Parenteral aescin
has produced isolated cases of ana-
phylactic reactions, as well as he-
patic and renal toxic effects.
5,111-113
In
the event of toxicity, aescin can be
eliminated via dialysis, withelimina-
tion dependent on protein-bind-
ing.
114
Horse chestnut extract is also
one of the components of veno-
curan, a drug marketed as a treat-
ment for venous disorders. In 1975,
venocuran was determined to cause
a pseudolupus syndrome character-
ized by recurrent fever, myalgia, ar-
thralgia, pleuritis, pulmonary infil-
trates, pericarditis, myocarditis, and
mitochondrial antibodies in the ab-
sence of nuclear antibodies after pro-
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2231
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
longedtreatment.
115,116
Venocuranhas
since been withdrawn fromthe mar-
ket; however, the nature of its patho-
physiologic action is still unknown.
Like A hippocastanum, Ruscus
aculeatus (butchers broom) is also
known for its use in treating ve-
nous insufficiency. Ruscus aculea-
tus is a short evergreen shrub found
commonly in the Mediterranean re-
gion. Two steroidal saponins, rus-
cogenin and neurogenin, extracted
fromthe rhizomes of R aculeatus are
thought to be its active compo-
nents.
101
In vivo studies on hamster
cheek pouch reveal that topical Rus-
cus extract dose dependently antago-
nizes histamine-inducedincreases in
vascular permeability.
117
More-
over, topical Ruscus extract causes
dose-dependent constriction of ven-
ules without appreciably affecting ar-
terioles.
118
Topical Ruscus extracts
vascular effects are also tempera-
ture dependent and appear to
counter the sympathetic nervous
systems temperature-sensitive vas-
cular regulation: venules dilate at a
lower temperature (25C), con-
strict at near physiologic tempera-
tures (36.5C), and further con-
strict at higher temperatures (40C);
arterioles dilate at 25C, are unaf-
fected at 36.5C, and remain unaf-
fected or constrict at 40C, depend-
ing onRuscus concentration.
119
Based
on the influence of prazosin, diltia-
zem, and rauwolscine, the periph-
eral vascular effects of Ruscus ex-
tract appear to be selectively
mediatedby effects oncalciumchan-
nels and
1
-adrenergic receptors
with less activity at
2
-adrenergic re-
ceptors.
117,118
Also, R aculeatus ex-
hibits strong antielastase activity and
has little effect on hyaluronidase in
direct contrast to A hippocastanum.
This activity may contribute to their
efficacy in the treatment of venous
insufficiency since these enzyme sys-
tems are involved in the turnover of
the main components of the peri-
vascular amorphous substance.
102
Several small clinical trials us-
ing topical Ruscus extract support its
role in treating venous insuffi-
ciency. One randomized, double-
blind, placebo-controlled trial in-
volving 18 volunteers showed a
beneficial decrease in femoral vein
diameter (median decrease, 1.25
mm) using duplex B-scan ultrason-
ography. The decrease was mea-
sured 2.5 hours after applying 4 to
6 g of a cream containing 64 to 96
mg of Ruscus extract.
120
In another
small trial (N= 18) it was shownthat
topical Ruscus extract may be help-
ful in reducing venous dilation dur-
ing pregnancy.
121
Oral agents may be
useful as topical drugs for venous in-
sufficiency, although the evidence is
less convincing.
122
Although capsule, tablet, oint-
ment, and suppository (for hemor-
rhoids)preparationsof Ruscus extract
areavailableinEurope, onlycapsules
are available in the United States.
Thesecapsules contain75mgof Rus-
cusextract and2mgof rosemaryoil.
101
Asidefromoccasional nauseaandgas-
tritis, adverseeffectsfromusing Racu-
leatus haverarelybeenreported, even
in high doses.
57
Nevertheless, one
should be wary of any drug that has
not beenthoroughlytested. Although
thereisampleevidencetosupport the
pharmacological activityof Raculea-
tus, there is still a relative deficiency
of clinical data to establishits actual
safety andefficacy. Until more stud-
ies are completed, no recommenda-
tionsregardingdosagecanbeoffered.
ARRHYTHMIA
In traditional Chinese medicine, ar-
rhythmias are categorized by the
characteristic symptoms of palpita-
tions and abnormal pulse. Numer-
ous Chinese herbal medicines are
identified to have antiarrhythmic ef-
fects, such as xin bao, ci zhu wan,
bu xin dan, and several others.
123
However, few clinical trials have
been conducted to study their ef-
fects and safety. Xin bao is one agent
that has begun to be examined. The
mechanism of action of xin bao is
thought to be through its stimula-
tion and increased excitability of the
sinuatrial node.
124
In one observa-
tional study, the effects of xin bao
were documentedin87patients with
sick sinus syndrome. Xin bao was
administered orally 2 to 3 times per
day for 2 months. Patients with ma-
jor symptoms of sick sinus syn-
drome, which included dizziness,
palpitations, and chest pressure, im-
proved significantly after treat-
ment.
124
No serious adverse effects
were noted. This study suggests a
possible role of xin bao in the treat-
ment of sick sinus syndrome. How-
ever, more scientific research on xin
bao and other antiarrhythmic Chi-
nese herbs mentioned previously are
necessary before any recommenda-
tions can be made for their routine
use in patients with sick sinus syn-
drome or other arrhythmias.
COMMENT
With the high prevalence of herbal
medicine use in the United States,
health practitioners should remem-
ber toinquire about suchhealthprac-
tices whentakingclinical histories and
remain informed of the beneficial or
harmful effects of these treatments.
Continuing research is necessary to
elucidate the pharmacological activi-
ties of the many cardiopotent herbal
medicines and to stimulate future
pharmaceutical development of thera-
peutically beneficial herbal drugs.
However, such research is currently
lacking in the United States and re-
quires more support from govern-
ment agencies before the full poten-
tial of these types of treatments can
be determined. At the same time, le-
gal surveillanceof herbal medicineuse
withlowsafety margins shouldbe in-
stituted for the sake of public health;
this is especially imperative for those
herbs withadverse cardiovascular re-
actions
125
and drug interactions. As
more information becomes available
regarding the safety and efficacy of
herbal medicines through newclini-
cal trials, research-supported claims
mayone daybecome available tocon-
sumers and physicians in a manner
similar to the allopathic medicines.
Accepted for publication July 23, 1998.
Reprints: William H. Frishman,
MD, New York Medical College,
Munger Pavilion, Valhalla, NY10595.
REFERENCES
1. Wu B-N, Huang Y-C, Wu H-M, Hong S-J, Chiang
L-C, Chen I-J. A highly selective (
1
-adrenergic
blocker with partial
2
agonist activity derived
fromferulic acid, an active component of Ligus-
ticum wallichii Franch. J Cardiovasc Pharma-
col. 1998;31:750-757.
2. Eisenberg DM, Kessler RC, Foster C, et al. Un-
conventional medicine in the United States. N
Engl J Med. 1993;328:246-252.
3. Astin JA. Why patients use alternative medi-
cine: results of a national study. JAMA. 1998;
279:1548-1553.
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2232
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
4. Ernst E: Harmless herbs? a review of the recent
literature. Am J Med. 1998;104:170-178.
5. Tyler VE. Herbs of Choice: The Therapeutic Use
of Phytomedicinals. NewYork, NY: Pharmaceu-
tical Product Press; 1994.
6. Stuhlemmer U, Kreis W, Eisenbeiss M, Reinhard
E. Cardiac glycosides in partly submerged shoots
of Digitalis lanata. Planta Med. 1993;59:539-545.
7. Dickstein ES, Kunkel FW. Foxglove tea poison-
ing. Am J Med. 1980;69:167-169.
8. Radford DJ, Gillies AD, Hinds JA, Duffy P. Natu-
rally occurring cardiac glycosides. Med J Aust.
1986;144:540-544.
9. CheungK, HindsJA, DuffyP. Detectionof poison-
ing by plant-origin cardiac glycoside with the Ab-
bott Tdx analyzer. Clin Chem. 1989;35:295-297.
10. Moxley RA, Schneider NR, Steinegger DH, Carl-
son MP. Apparent toxicosis associated with lily-
of-the-valley (Convallaria majalis) ingestion in a
dog. J Am Vet Med Assoc. 1989;195:485-487.
11. Bossi M, Brambilla G, Cavilla A, et al. Threaten-
ing arrhythmia by uncommon digitalic toxicosis
[in Italian]. G Ital Cardiol. 1981;11:2254-2257.
12. Hayes BE, Bessen HA, Wightman WD. Oleander
tea: herbal draught of death. Ann Emerg Med.
1985;14:350-353.
13. Ansford AJ, Morris H. Fatal oleander poisoning.
Med J Aust. 1981;1:360-361.
14. Shaw D, Pearn J. Oleander poisoning. Med J
Aust. 1979;2:267-269.
15. Tumcok Y, Kozan O, Cadvar C, et al. Urginea mar-
itima (squill) toxicity. J Toxicol Clin Toxicol. 1995;
33:83-86.
16. BagrovAY, RoukoyatkinaNI, PinaevAG, Dmitrieva
RI. Effects of two endogenous Na+, K(+)-ATPase
inhibitors, marinobufageninandouabain, oniso-
latedrat aorta. EurJPharmacol. 1995;274:151-158.
17. Nishoka SA, Resende ES. Transitory complete atri-
oventricular blockassociatedtoingestionof Nerium
oleander. Rev Assoc Med Bras. 1995;41:60-62.
18. Rich SA, Libera JM, Locke RJ. Treatment of fox-
glove extract poisoning with digoxin-specific Fab
fragments. AnnEmergMed. 1993;22:1904-1907.
19. Safadi R, Levy I, Amitai Y, Caraco Y. Beneficial
effect of digoxin-specific Fab antibody frag-
ments in oleander intoxication. Arch Intern Med.
1995;155:2121-2125.
20. Galey FD, Holstege DM, Plumlee KH, et al. Di-
agnosis of oleander poisoning in livestock. J Vet
Diagn Invest. 1996;8:358-364.
21. Langford SD, Boor PJ. Oleander toxicity: an ex-
amination of human and animal toxic exposure.
Toxicology. 1996;109:1-13.
22. McRae S. Elevated serumdigoxin levels in a pa-
tient taking digoxin and Siberian ginseng. CMAJ.
1996;155:293-295.
23. Geehr E. Common toxic plant ingestions. Emerg
Med Clin North Am. 1984;2:553-563.
24. Litovitz TL, Clark LR, Soloway RA. 1993 Annual
Report of the American Association of Poison
Control Centers Toxic Exposure Surveillance Sys-
tem. Am J Emerg Med. 1993;12:546-584.
25. Szabuniewicz M, McGrady JD, Camp BJ. Treat-
ment of experimentally induced oleander poison-
ing. Arch Int Pharmcodyn. 1971;189:12-21.
26. Osterloh J, Herold S, Pond S. Oleander interfer-
ence in the digoxin radioimmunoassay in a fatal
ingestion. JAMA. 1982;247:1596-1597.
27. Oates JA. Antihypertensive agents and the drug
therapy of hypertension. In: Hardman JG, Lim-
bird LE, Molinoff PB, et al, eds. Goodman and
Gilmans The Pharmacological Basis of Thera-
peutics. 9th ed. NewYork, NY: McGraw-Hill Book
Co; 1996:781-808.
28. Rauwolfiaalkaloids. USPDI: DrugInformationfor
theHealthCareProfessional. Vol 1. 18thed. Rock-
ville, Md: USPharmacopeial Convention; 1998:2486.
29. Physicians Desk Reference. 43rd ed. Montvale,
NJ: Medical Economics Co; 1989:1648.
30. BruntonLL. Agents affectinggastrointestinal wa-
ter flux, emesisandantiemetics, bileacidsandpan-
creaticenzymes. In: HardmanJG, LimbirdLE, Mo-
linoff PB, et al, eds. Goodman and Gilmans The
Pharmacological Basisof Therapeutics. 9thed. New
York, NY: McGraw-Hill Book Co; 1996:917-936.
31. Sutter MC, Wang YX. Recent cardiovascular
drugs from Chinese medicinal plants. Cardio-
vasc Res. 1993;27:1891-1901.
32. Rossier MF, PythonCP, Capponi AM, et al. Block-
ing T-type calcium channels with tetrandrine
inhibitssteroidogenesisinbovineadrenal glomeru-
losa cells. Endocrinology. 1993;132:1035-1043.
33. Kawashima K, Hayakawa T, Miwa Y, et al. Struc-
ture and hypotensive activity relationships of tet-
randrine derivatives in stroke-prone spontane-
ously hypertensive rats. Gen Pharmacol. 1990;
21:343-347.
34. Vanherweghem JL. A new form of nephropathy
secondary to the absorption of Chinese herbs [in
French]. Bull MemAcad R Med Belg. 1994;149:
128-135.
35. Schmeiser HH, Bieler CA, Wiessler M, Van Yper-
sele DE, Strihou C, Cosyns JP. Detection of DNA
adducts formed by aristolochic acid in renal tis-
sue frompatients with Chinese herbs nephropa-
thy. Cancer Res. 1996;56:2025-2028.
36. Bieler CA, Stiborova M, Wiessler M, et al. 32P-
post-labelling analysis of DNA adducts formed
by aristolochic acid in tissues frompatients with
Chinese herbs nephropathy. Carcinogenesis.
1997;18:1063-1067.
37. Violin C. Belgian (Chinese herb) nephropathy:
why? J Pharm Belg. 1997;52:7-27.
38. Ody P. The Complete Medicinal Herbal. NewYork,
NY: Dorling Kindersley; 1993.
39. Kuramochi T, Chu J, Suga T. Gou-teng (fromUn-
caria rhynchophyllaMiquel)-inducedendothelium-
dependent and-independent relaxations inthe iso-
lated rat aorta. Life Sci. 1994;54:2061-2069.
40. Jaffe AM, Gephardt D, Courtemanche L. Poison-
ing due to ingestion of Veratrumviride (false hel-
lebore). J Emerg Med. 1990;8:161-167.
41. Quatrehomme G, Bertrand F, Chauvet C, Ollier
A. Intoxication from Veratrum album. Hum Exp
Toxicol. 1993;12:111-115.
42. Festa M, Andreeto B, Ballaris MA, Panio A, Pier-
vittori R. Un caso di avvelenamento da Veratro.
Minerva Anestesiol. 1996;62:195-196.
43. Chiou W-F, ShumAY-C, Liao J-F, Chen C-F. Stud-
ies of the cellular mechanisms underlying the va-
sorelaxant effects of rutaecarpine, a bioactive
component extracted froman herbal drug. J Car-
diovasc Pharmacol. 1997;29:490-498.
44. Bahorun T, Trotin F, Pommery J, et al. Antioxi-
dant activities of Crataegus monogyna ex-
tracts. Planta Med. 1994;60:323-328.
45. Vibes J, Lasserre B, Gleye J, et al. Inhibition of
thromboxane A2 biosynthesis in vitro by the main
components of Crataegus oxyacantha (Haw-
thorn) flower heads. Prostaglandins Leukot Es-
sent Fatty Acids. 1994;50:173-175.
46. Shanthi S, Parasakthy K, Deepalakshmi PD, De-
varaj SN. Hypolipidemic activity of tincture of Cra-
taegus in rats. Indian J BiochemBiophys. 1994;
31:143-146.
47. Rajerdan S, Deepalakshmi PD, Parasakthy K, De-
varaj H, Devaraj SN. Effect of tincture of Cratae-
gus on the LDL-receptor activity of hepatic
plasma membrane of rats fed an atherogenic diet.
Atherosclerosis. 1996;123:235-241.
48. Nasa Y, Hashizume H, Hoque AN, Abiko Y. Pro-
tective effect of Crataegus extract on the the car-
diac mechanical dysfunction in isolated per-
fused working rat heart. Arzneimittelforschung.
1993;43:945-949.
49. Roddweig C, Hensel H. Reaction of local myo-
cardial blood flowin nonanesthesized dogs and
anesthesized cats to the oral and parenteral ad-
ministration of a Crataegus fraction (oligomere
procyanidines) [in German]. Arzneimittelforsch-
ung. 1977;27:1407-1410.
50. Taskov M. On the coronary and cardiotonic ac-
tion of crataegus. Acta Physiol Pharmacol Bulg.
1997;3:53-57.
51. Blesken R. Crataegus in cardiology [in Ger-
man]. Fortschr Med. 1992;110:290-292.
52. Petkov V. Plants and hypotensive, antiathero-
matous and coronarodilating action. Am J Chin
Med. 1979;7:197-236.
53. Weikl A, Assmus KD, Neukum-Schmidt A, et al.
Crataegus special extract WS 1442: assessment
of objective effectiveness inpatients withheart fail-
ure (NYHA II). Fortschr Med. 1996;114:291-296.
54. Schussler M, Holzl J, Fricke U. Myocardial ef-
fects of flavonoids fromCrataegus species. Arz-
neimittelforschung. 1995;45:842-845.
55. JosephG, ZhaoY, KlausW. Pharmacologicaction
profile of Crataegus extract in comparison to epi-
nephrine, amirinone, milrinone, anddigoxininthe
isolatedperfusedguineapigheart [inGerman]. Arz-
neimittelforschung. 1995;45:1261-1265.
56. PoppingS, RoseH, IonescuI, Fischer Y, Kammer-
meier H. Effect of a hawthorn extract on contrac-
tionandenergyturnover of isolatedrat cardiomyo-
cytes. Arzneimittelforschung. 1995;45:1157-1161.
57. Mawrey DB. Herbal Tonic Therapies. NewCanaan,
Conn: Keats Publishing Inc; 1993.
58. Kwan CY. Vascular effects of selected antihy-
pertensive drugs derived from traditional me-
dicinal herbs. Clin Exp Pharmacol Physiol. 1995;
22:297-299.
59. Zhang W, Wojta J, Binder BR. Effect of notogin-
senoside R1 on the synthesis of tissue-type plas-
minogen activator and plasminogen activator in-
hibitor-1 in cultured human umbilical vein
endothelial cells. Arterioscler Thromb Vasc Biol.
1994;14:1040-1046.
60. Lin SG, Zheng XL, Chen QY, Sun JJ. Effect of
Panax notoginseng saponins on increased pro-
liferation of cultured aortic smooth muscle cells
stimulated by hypercholestoremic serum. Chung
Kuo Yao Li Hsueh Pao. 1993;14:314-316.
61. Lei XL, Chiou GC. Cardiovascular pharmacology
of Panax notoginseng (Burk) FHChen and Salvia
miltiorrhiza. Am J Chin Med. 1986;14:145-152.
62. Wang Z, Roberts JM, Grant PG, et al. The effect
of a medicinal Chinese herb on platelet func-
tion. Thromb Haemost. 1982;48:301-306.
63. Yagi A, Fujimoto K, Tanonaka K, et al. Possible
active components of tan-shen (Salvia miltior-
rhiza) for protection of the myocardium against
ischemia-induced derangements. Planta Med.
1989;55:51-54.
64. Zhao BL, Jiang W, Zhao Y, Hou JW, Xin WJ. Scav-
enging effects of Salvia miltiorrhiza on free radi-
cals and its protection for myocardial mitochon-
drial membranesfromischemia-reperfusioninjury.
Biochem Mol Biol Int. 1996;38:1171-1182.
65. Chan K, Lo AC, Yeung JH, Woo KS. The effects
of Danshen (Salvia miltiorrhiza) on warfarin phar-
macodynamics and pharmacokinetics of warfa-
rin enantiomers in rats. J Pharm Pharmacol.
1995;47:402-406.
66. Kleijnen J, Knipschild P, Ter Riet G. Garlic, on-
ions and cardiovascular risk factors: a reviewof
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2233
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014
the evidence fromhuman experiments with em-
phasis on commercially available preparations.
Br J Clin Pharmacol. 1989;28:535-544.
67. Steiner M, Khan AH, Holbert D, Lin RI. A double-
blind cross-over study in hypercholesterolemic
men that compared the effect of aged garlic ex-
tract and placebo administration on blood lip-
ids. Am J Clin Nutr. 1996;63:866-870.
68. Silagy CA, Neil HA. A meta-analysis of the ef-
fect of garlic on blood pressure. J Hypertens.
1994;12:463-468.
69. Silagy C, Neil A. Garlic as a lipid lowering agent:
a meta-analysis. J RColl Physicians Lond. 1994;
28:39-45.
70. Kendler BS. Garlic(Alliumsativum) andonion(Al-
liumcepa): areviewof their relationshiptocardio-
vascular disease. Prev Med. 1987;16:670-685.
71. Isaacsohn JL, Moser M, Stein EA, et al. Garlic
powder andplasma lipids andlipoproteins: a mul-
ticenter, randonized, plcebo-controlled trial. Arch
Intern Med. 1998;158:1189-1194.
72. Jain AK, Vargas R, Gotzkowsky S, McMahon FG.
Cangarlicreducelevelsof serumlipids?acontrolled
clinical study. Am J Med. 1993;94:632-635.
73. Neil HA, Silagy CA, Lancaster T, et al. Garlic pow-
der in the treatment of moderate hyperlipidae-
mia: a controlled trial and meta-analyses. J RColl
Physicians Lond. 1996;30:329-334.
74. Simons LA, BalaSubramaniam S, Von Konigs-
mark M, Parfitt A, Simons J, Peters W. On the
effect of garlic on plasma lipids and lipopro-
teins in mild hypercholesterolemia. Atheroscle-
rosis. 1995;113:219-225.
75. Bordia A, Verma SK, Srivastava KC. Effect of gar-
lic on platelet aggregation in humans: a study in
healthy subjects and patients with coronary ar-
tery disease. Prostaglandins Leukot Essent Fatty
Acids. 1996;55:201-205.
76. Ali M, Thomson M. Consumption of a garlic clove
a day could be beneficial in preventing throm-
bosis. Prostaglandins Leuko Essent Fatty Ac-
ids. 1995;53:211-212.
77. Breithaupt-Grogler K, Ling M, Boudoulas H, Be-
liz GG. Protective effect of chronic garlic intake
on elastic properties of aorta in the elderly. Cir-
culation. 1997;96:2649-2655.
78. Mader FH. Treatment of hyperlipidaemia with gar-
lic powder tablets: evidence fromthe German As-
sociation of General Practitioners Multicentric
Placebo-Controlled Double-Blind Study. Arznei-
mittelforschung. 1990;40:1111-1116.
79. Delaney TA, Donnelley AM. Garlic dermatitis. Aus-
tralas J Dermatol. 1996;37:109-110.
80. Rose KD, Croissant PD, Parliament CF, Levin MB.
Spontaneousspinal epidural hematomawithasso-
ciatedliverdysfunctionfromexcessivegarlicinges-
tion: acasereport. Neurosurgery. 1990;26:880-882.
81. Singh V, Kaul S, Chander R, Kapoor NK. Stimu-
lation of low density lipoprotein receptor activ-
ity in liver membrane of guggulsterone treated
rats. Pharmacol Res. 1990;22:37-44.
82. Nityanand S, Srivastava JS, Asthana OP. Clinical
trials with gugulipid: a newhypolipidemic agent.
J Assoc Physicians India. 1989;37:323-328.
83. Sing RB, Niaz MA, Ghosh S. Hypolipidemic and
antioxidant effects of Commiphora mukul as an
adjunct to dietary therapy in patients with hy-
percholesterolemia. Cardiovasc Drugs Ther.
1994;8:654-664.
84. Dalvi SS, Nayak VK, Pohujani SM, Desai NK,
Kshirsagar NA, Gupta KC. Effect of gugulipid on
bioavailability of diltiazemand propranolol. J As-
soc Physicians India. 1994;42:454-455.
85. Sundaram V, Hanna AN, Lubow GP, Koneru L,
Falko JM, Sharma HM. Inhibition of low-
density lipoprotein oxidation by oral herbal mix-
tures Maharishi Amrit Kalash-4 and Maharishi
Amrit Kalash-5 in hyperlipidemic patients. Am
J Med Sci. 1997;314:303-310.
86. Huxtable RJ. The pharmacology of extinction.
J Ethnopharmacol. 1992;37:1-11.
87. ZBrun A. Ginkgo: myth and reality [in German].
Schweiz Rundsch Med Prax. 1995;84:1-6.
88. Kleinjen J, Knipschild P. Ginkgo biloba for ce-
rebral insufficiency. Br J Clin Pharmacol. 1992;
34:352-358.
89. Allard M. Treatment of the disorders of aging with
Ginkgo biloba extract [in French]. Presse Med.
1986;15:1540-1545.
90. LeBarsPL, Katz MM, BermanN, et al, for theNorth
American Study Group. A placebo-controlled,
double-blind, randomizedtrial ofanextractofGinkgo
biloba for dementiaJAMA. 1997;278:1327-1332.
91. DolyM, Droy-LefaixMT, Braquet P. Oxidativestress
in diabetic retina. EXS. 1992;62:299-307.
92. Mouren X, Calliard P, Schwartz F. Study of the
antiischemic action of EGb 761 in the treatment
of peripheral arterial occlusive disease by TcPO
2
determination. Angiology. 1994;45:413-417.
93. Tosaki A, Droy-Lefaix MT, Pali T, Das DK. Ef-
fects of SOD, catalase, and a novel antiarrhyth-
mic drug, Egb 761, on reperfusion-induced ar-
rhythmias in isolated rat hearts. Free Radic Biol
Med. 1993;14:361-370.
94. Haramaki N, Aggarwal S, KawabataT, Droy-Lefaix
MT, Packer L. Effects of natural antioxidant EGb
761 on myocardial ischemia-reperfusion injury.
Free Radic Biol Med. 1994;16:789-794.
95. Pietri S, Seguin JR, DArbigny P, Drieu K, Cul-
casi M. Egb 761 pretreatment limits free radical-
induced oxidative stress in patients undergoing
coronary bypass surgery. Cardiovasc Drugs Ther.
1997;11:121-131.
96. Offord EA, Mace K, Ruffieux C, et al. Rosemary
components inhibit benzo[a]pyrene induced
genotoxicity in human bronchial cells. Carcino-
genesis. 1995;16:2057-2062.
97. Haraguchi H, Saito T, Okamura N, Yagi A. Inhi-
bition of lipid peroxidation and superoxide gen-
eration by diterpenoids from Rosmarinus offi-
cinalis. Planta Med. 1995;61:333-336.
98. Larrondo JV, Agut M, Calvo-Torras MA. Antimi-
crobial activity of essences from labiates. Mi-
crobios. 1995;82:171-172.
99. Al-Hader AA, Hasan ZA, Aqel MB. Hyperglycemic
and insulin release inhibitory effects of Rosmari-
nusofficinalis. JEthnopharmacol. 1994;43:217-221.
100. Swan AR, Dutton SP, Truswell AS. Salicylates
in foods. J Am Diet Assoc. 1985;85:950-960.
101. Tyler VE. The Honest Herbal: A Sensible Guide to
theUseof HerbsandRelatedRemedies. 3rded. New
York, NY: Pharmaceutical Product Press; 1993.
102. Facino RM, Carini M, Stefani R, Aldini G, Sai-
bene L. Anti-elastase and anti-hyaluronidase ac-
tivities of saponins and sapogenins from Hed-
era helix, Aesculus hippocastanum, and Ruscus
aculeatus: factors contributing to their efficacy
in the treatment of venous insufficiency. Arch
Pharm (Weinheim). 1995;328:720-724.
103. Guillaume M, Padioleau F. Veinotonic effect, vas-
cular protection, antiinflammatory and free radi-
cal scavenging properties of horse chestnut ex-
tract. Arzneimittelforschung. 1994;44:25-35.
104. Rothkopf M, Vogel G. New findings on the effi-
cacy and mode of action of the horse chestnut
saponin escin [in German]. Arzneimittelforsch-
ung. 1976;26:225-235.
105. Diehm C, Vollbrecht, Amendt K, Comberg HU.
Medical edema protection: clinical benefit in pa-
tients with chronic deep vein incompetence: a
placebo controlled double-blind study. Vasa.
1992;21:188-192.
106. Bisler H, Pfeifer R, Kluken N, Pauschinger P. Ef-
fects of horse chestnut seed extract on tran-
scapillary filtration in chronic venous insuffi-
ciency [in German]. Dtsch Med Wochenschr.
1986;111:1321-1329.
107. Diehm C, Trampisch HJ, Lange S, Schmidt C.
Comparisonof legcompressionstockingandoral
horse chestnut seed extract therapy in patients
with chronic venous insufficiency. Lancet. 1996;
347:292-294.
108. GreeskeK, PohlmannBK. Horsechestnut seedex-
tract: aneffectivetherapyprincipleingeneral prac-
tice: drugtherapy inchronic venous insufficiency
[in German]. Fortschr Med. 1996;114:196-200.
109. Schorr DM, Gruber UF. Prophylaxis of throm-
boembolic complications in gynecological sur-
gery [in German]. Geburtshilfe Frauenheilkd.
1977;37:291-296.
110. Lang W. Studies on the percutaneous absorp-
tion of 3H-aescin in pigs. Res Exp Med (Berl).
1977;169:175-187.
111. Takegoshi K, Tohyama T, Okunda K, et al. A case
of Venoplant-induced hepatic injury. Gastroen-
terol Jpn. 1986;21:62-65.
112. Voigt E, Junger H. Acute posttraumatic renal fail-
ure following therapy with antibiotics and beta-
aescin[inGerman]. Anaesthesist. 1978;27:81-83.
113. Hellberg K, Ruschewski W, De Vivie R. Drug in-
ducedacuterenal failureafter heart surgery[inGer-
man]. Thoraxchir Vask Chir. 1975;23:396-399.
114. Lang W. Dialysability of aescin [in German].
Arzneimittelforschung. 1984;34:221-223.
115. Grob PJ, Muller-Schoop JW, Hacki MA, Joller-
Jemelka HI. Drug-induced pseudolupus. Lan-
cet. 1975;2:144-148.
116. Walli F, Grob PJ, Muller-Schoop J. Pseudo-
(venocuran-) lupus: a minor episode in the his-
tory of medicine [in German]. Schweiz Med
Wochenschr. 1981;111:1398-1405.
117. Bouskela E, Cyrino FZ, Marcelon G. Possible
mechanisms for the inhibitory effect of Ruscus
extract on increased microvascular permeabil-
ity induced by histamine in hamster cheek pouch.
J Cardiovasc Pharmacol. 1994;24:281-285.
118. Bouskela E, Cyrino FZ, Marcelon G. Possible
mechanisms for the venular constriction elic-
ited by Ruscus extract on hamster cheek pouch.
J Cardiovasc Pharmacol. 1994;24:165-170.
119. Bouskela E, Cyrino FZ, Marcelon G. Effects of
Ruscus extract on the internal diameter of arte-
rioles and venules of the hamster cheek pouch
microcirculation. J Cardiovasc Pharmacol. 1993;
22:221-224.
120. Berg D. Venous constriction by local adminis-
tration of ruscus extract [in German]. Fortschr
Med. 1990;108:473-476.
121. BergD. Venoustonicityinpregnancyvaricoseveins
[inGerman]. FortschrMed. 1992;110:67-68,71-72.
122. Weindorf N, Schultz-Ehrenburg U. Controlled
study of increasing venous tone in primary vari-
cose veins by oral administration of Ruscus acu-
leatus and trimethylhespiridinchalcone [in Ger-
man]. Z Hautkr. 1987;62:28-38.
123. Zhou ZY, Jin HD. Clinical Manual of Chinese
Herbal Medicine. New York, NY: Churchill Liv-
ingstone Inc; 1997.
124. Chen ZY. Use of Xin Bao in the treatment of 87
patients with sick sinus syndrome. Chin Mod-
ern Develop Trad Med. 1990;10:529-531.
125. Jones TK, Lawson BM. Profound neonatal con-
gestive heart failure caused by maternal con-
cumption of blue cohosh herbal medication.
J Pediatr. 1998;132:550-552.
ARCH INTERN MED/ VOL 158, NOV 9, 1998
2234
1998 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ on 03/20/2014