A DV E RSE EVE NT S ASSOCIAT E D WIT H D IETA RY SUPPLEMENTS CONTA INING EPH ED RA A LK A LOID S

ADVERSE CARDIOVASCULAR AND CENTRAL NERVOUS SYSTEM EVENTS

ASSOCIATED WITH DIETARY SUPPLEMENTS CONTAINING
EPHEDRA ALKALOIDS

CHRISTINE A. HALLER, M.D., AND NEAL L. BENOWITZ, M.D.

ABSTRACT growing concern about the safety of ephedra alkaloids

Background Dietary supplements that contain in dietary supplements, the Food and Drug Admin-
ephedra alkaloids (sometimes called ma huang) are istration (FDA) requested an independent review of
widely promoted and used in the United States as a reports of adverse events related to the use of ephedra
means of losing weight and increasing energy. In the alkaloids to assess causation and determine the level
light of recently reported adverse events related to of risk these products pose to consumers.
use of these products, the Food and Drug Adminis- We conducted an in-depth review of 140 reports
tration (FDA) has proposed limits on the dose and of adverse events involving dietary supplements con-
duration of use of such supplements. The FDA re-
quested an independent review of reports of adverse
taining ephedra alkaloids that were submitted to the
events related to the use of supplements that con- FDA between June 1, 1997, and March 31, 1999,
tained ephedra alkaloids to assess causation and to and applied a standardized rating system for assessing
estimate the level of risk the use of these supple- causation. We also evaluated factors that might increase
ments poses to consumers. the risk to consumers and the adequacy of warnings
Methods We reviewed 140 reports of adverse events about potential risks included on product labels. The
related to the use of dietary supplements containing full report of our review of adverse events is available
ephedra alkaloids that were submitted to the FDA be- elsewhere.7 Here, we summarize our findings.
tween June 1, 1997, and March 31, 1999. A standardized
rating system for assessing causation was applied to METHODS
each adverse event.
The objective of the review was to determine the likelihood that
Results Thirty-one percent of cases were consid- ephedra alkaloids (which were usually combined with caffeine)
ered to be definitely or probably related to the use of caused the reported adverse events on the basis of the information
supplements containing ephedra alkaloids, and 31 per- provided in the FDA MedWatch report, along with supplemental
cent were deemed to be possibly related. Among the medical records. We independently reviewed each of the 140 cases.
adverse events that were deemed definitely, probably, Causation was assessed according to the criteria described by Blanc
or possibly related to the use of supplements contain- et al.8 and included an evaluation of the timing of the event in re-
ing ephedra alkaloids, 47 percent involved cardiovas- lation to the dose and duration of use of a product; an assessment
cular symptoms and 18 percent involved the central of the pattern of response to determine whether it constituted a
recognized reaction to the substance on the basis of previous re-
nervous system. Hypertension was the single most ports of ephedrine or similar stimulants in the medical literature;
frequent adverse effect (17 reports), followed by pal- and a determination of the contribution of any underlying diseas-
pitations, tachycardia, or both (13); stroke (10); and es or medical conditions.
seizures (7). Ten events resulted in death, and 13 events In general, we defined an adverse event as definitely related to the
produced permanent disability, representing 26 per- use of supplements containing ephedra alkaloids only if the symp-
cent of the definite, probable, and possible cases. toms recurred with the reintroduction of ephedra alkaloids or when
Conclusions The use of dietary supplements that the onset of symptoms coincided with the expected peak plasma
contain ephedra alkaloids may pose a health risk to concentration of the drug and resolved within an interval that was
consistent with the expected duration of the effect of ephedrine.
some persons. These findings indicate the need for a
An adverse event was defined as probably related to the use of sup-
better understanding of individual susceptibility to the plements containing ephedra alkaloids when the majority of the
adverse effects of such dietary supplements. (N Engl evidence supported the existence of a causal link but one or more
J Med 2000;343:1833-8.) aspects of the case, such as time since the last dose, were unknown
©2000, Massachusetts Medical Society. or there was a minor inconsistency in the supporting evidence, such
as a low reported dose. An adverse event was designated as possi-
bly related to the use of supplements containing ephedra alkaloids
when it was equally likely that the adverse event was not related to
the use of ephedra alkaloids; for example, in the case of effects that

D
IETARY supplements that contain ephedra have not been reported in the literature in association with ephedra
alkaloids (also known as ma huang) and alkaloids but that are pharmacologically plausible. Reports of ad-
guarana-derived caffeine are widely con-
sumed in the United States for purposes
From the Division of Clinical Pharmacology and Experimental Thera-
of weight reduction and energy enhancement. A peutics, Departments of Medicine and Biopharmaceutical Sciences, Univer-
number of reports of adverse reactions to dietary sity of California, San Francisco, and the California Poison Control System,
supplements that contain ephedra alkaloids, some of San Francisco Division — both in San Francisco. Address reprint requests
to Dr. Benowitz at the Division of Clinical Pharmacology and Experimen-
which resulted in permanent injury or death, have tal Therapeutics, University of California, San Francisco, Box 1220, San
appeared in the medical literature.1-6 In response to Francisco, CA 94143-1220, or at nbeno@itsa.ucsf.edu.

Vol ume 343 Numb e r 25 · 1833

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on December 8, 2024. For personal use only.
No other uses without permission. Copyright © 2000 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

verse events that included scant medical history and incomplete in-
formation about the product involved were usually considered to TABLE 1. AGE AND SEX OF USERS OF
have insufficient information to be assessed. This category was re- SUPPLEMENTS CONTAINING EPHEDRA
served for events in which the evidence was not substantial enough ALKALOIDS AND REASONS FOR USE,
to consider them as possibly related to the use of supplements con- ACCORDING TO ADVERSE-EVENT REPORTS.
taining ephedra alkaloids. Adverse events were defined as probably
unrelated to the use of supplements containing ephedra alkaloids
if the evidence that ephedra alkaloids were the cause was weak or NO. OF
if the likelihood was strong that there was some other cause, either VARIABLE USERS (%)
medical or toxicologic. When the scientific evidence or course of
Age
events was highly inconsistent with the known effects of ephedra <18 yr 10 (7)
alkaloids, the event was considered definitely unrelated; for exam- 18–29 yr 34 (24)
ple, in the case of symptoms that persisted long after the use of 30–45 yr 45 (32)
ephedra alkaloids had been discontinued or in the case of symp- >45 yr 32 (23)
toms that had no association with the known pharmacodynamic Unknown 19 (14)
effects of ephedra alkaloids. However, the event was considered re- Sex
lated if the patient had a preexisting condition such as hypertension Male 56 (40)
Female 84 (60)
that could have been aggravated by the use of ephedra alkaloids and
Reason for use
if the pattern of use met the criteria for causation. Weight loss 83 (59)
In determining the likelihood of a causal link, we evaluated as- To increase athletic performance 23 (16)
pects of the medical history, dietary patterns, and social habits as To increase energy 9 (6)
possible contributing or causative factors. For example, we noted Unknown 24 (17)
when events occurred while patients were fasting or in conjunction Intentional misuse 1 (1)
with high intakes of caffeine. We recognized that in the case of ad-
verse events that were most likely not related to the use of supple-
ments containing ephedra alkaloids one or more of the other ingre-
dients in the supplement may have been causally related to the event.

the definite, probable, and possible cases. Nine serious

RESULTS
adverse events occurred in persons who were taking
Features of the Cases relatively low doses of ephedra alkaloids (range, 12 to
The age and sex of the users of products containing 36 mg per day) and who had no important medical
ephedra aklaloids and the reported reasons for use are risk factors. Features of the definite or probable and
shown in Table 1. Although the labels of most such possible cases that resulted in death or permanent
products state that they are not intended for use by impairment or that necessitated substantial medical
persons less than 18 years of age, adverse events were intervention are given in Tables 4 and 5, respectively.
recorded in at least 10 persons under this age. The Of the sudden catastrophic cerebrovascular and car-
youngest was 15 years old. Overall, 43 cases (31 per- diovascular events, 11 occurred in previously healthy
cent) were considered to be definitely or probably re- persons. Some of these cases, which were definitely
lated to the use of supplements containing ephedra or probably related to the use of supplements contain-
alkaloids, 24 cases (17 percent) were considered to be ing ephedra alkaloids, are described in detail in the
unrelated to the use of such supplements, 44 cases following sections.
(31 percent) were deemed possibly related to the use Examples of Severe Cerebrovascular Adverse Events
of such supplements, and in 29 cases (21 percent)
Patient 1
the information provided was insufficient to assess
causation. The types of adverse events that were def- Patient 1 was a healthy 35-year-old woman who had
initely or probably related to the use of supplements taken aerobic-exercise classes for several years without
containing ephedra alkaloids and those that were pos- incident (Table 4). In July 1997, she began taking
sibly related are summarized in Table 2. one capsule of Shape-Fast Plus (according to the label,
Cardiovascular symptoms made up 47 percent of each capsule contained 15 mg of ephedra alkaloids and
the adverse events that were definitely, probably, or 40 mg of caffeine) three times a day before meals for
possibly related to the use of supplements containing weight loss; she was taking no other medications. She
ephedra alkaloids. Hypertension was the single most had been taking the product for one week when she
frequent adverse effect, followed by palpitations, tach- collapsed during an aerobics class. Bystanders observed
ycardia, or both. Eighteen percent of related and pos- that her arms and legs were flexing and tensing. In
sibly related adverse events involved the central nerv- the emergency department, her blood pressure was
ous system. Strokes (n=10) and seizures (n=7) were 110/38 mm Hg and the heart rate was 104 beats per
the most frequent type of central nervous system event minute. A computed tomographic scan of the head
reported. The clinical outcomes of the definite, prob- showed a subarachnoid hemorrhage. Cerebral angi-
able, and possible cases are listed in Table 3. ography showed no evidence of a vascular aneurysm.
Ten events resulted in death (including 1 neonatal A urine toxicology screen was positive for amphet-
death and 1 fetal death), and 13 events resulted in per- amine, a result presumed to reflect a cross-reaction
manent impairment, which represented 26 percent of with the ephedrine and therefore to be false positive.

1834 · Dec em b er 2 1 , 2 0 0 0

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on December 8, 2024. For personal use only.
No other uses without permission. Copyright © 2000 Massachusetts Medical Society. All rights reserved.
 A DV ERSE EVE NT S ASSOCIAT E D WIT H D IETA RY SUPPLEMENTS CONTA INING EPH ED RA A LK A LOID S

TABLE 2. TYPES OF ADVERSE EVENTS THAT WERE DEFINITELY, TABLE 3. CLINICAL OUTCOMES OF ADVERSE EVENTS
PROBABLY, OR POSSIBLY RELATED TO THE USE OF SUPPLEMENTS THAT WERE DEFINITELY, PROBABLY, OR POSSIBLY
CONTAINING EPHEDRA ALKALOIDS. RELATED TO THE USE OF SUPPLEMENTS CONTAINING
EPHEDRA ALKALOIDS.

DEFINITELY OR
PROBABLY POSSIBLY DEFINITELY OR
RELATED RELATED TOTAL PROBABLY POSSIBLY
ADVERSE EVENT (N=43) (N=44) (N=87) RELATED RELATED TOTAL
OUTCOME (N=43) (N=44) (N=87)
no. of events (%)
no. of events (%)
Cardiovascular
Hypertension 10 (21) 7 (14) 17 (17) Death 3 (7) 7 (16) 10 (11)
Palpitations, tachycardia, or both 8 (17) 5 (10) 13 (13) Permanent impairment 7 (16) 6 (14) 13 (15)
Arrhythmia 3 (6) 3 (6) 6 (6)
Myocardial infarction 2 (4) 0 2 (2) Ongoing medical treat- 4 (9) 4 (9) 8 (9)
Cardiac arrest or sudden death 5 (10) 3 (6) 8 (8) ment
Central nervous system Full recovery 29 (67) 13 (30) 42 (48)
Stroke 4 (8) 6 (12) 10 (10) Unknown 0 14 (32) 14 (16)
Transient ischemic attack 1 (2) 0 1 (1)
Seizure 1 (2) 6 (12) 7 (7)
Other 14 (29) 20 (40) 34 (35)
Total no. of events* 48 50 98

*The total number of events exceeds the total number of cases, since
some cases involved more than one adverse event.

sis of the Ultimate Orange product confirmed the

presence of ephedrine, as well as of pseudoephedrine,
norephedrine, and norpseudoephedrine.
Examples of Severe Cardiovascular Adverse Events
Neurogenic pulmonary edema rapidly developed, ne-
Patient 2
cessitating endotracheal intubation and mechanical
ventilation. Electrocardiographic findings and cardiac- Patient 2 was a 22-year-old man with a history of
enzyme levels were consistent with the occurrence of asthma who collapsed while lifting weights at a gym
a small myocardial infarction. The treating cardiologist on March 31, 1998 (Table 4). His medications in-
and neurologist thought that ephedrine induced the cluded theophylline (Theo-Dur; 300 mg twice daily),
subarachnoid hemorrhage. The finding of amphet- albuterol (Ventolin; administered as necessary through
amine on the urine toxicology test supports the pres- a metered-dose inhaler), and a combination of chlor-
ence of ephedrine at the time of the event. Labora- pheniramine maleate, phenylephrine hydrochloride,
tory analysis of the supplement determined that the and phenylpropanolamine hydrochloride (Atrohist
ephedrine content was 12.0 mg per capsule. At that Plus SR). According to friends, he had consumed one
time, the FDA’s recommendation was a maximal dose 18-oz bottle of Ripped Force (which is listed as con-
of 8 mg per serving.7 taining 20 mg of ephedrine alkaloids, 100 mg of caf-
feine, 250 mg of L-carnitine, and 240 µg of chromi-
Patient 10
um) before working out and was regularly drinking
Patient 10 was an apparently healthy 39-year-old three bottles of Ripped Force per day. He also took
man who experienced numbness of the right arm creatine and protein supplements. Witnesses report-
and leg on March 17, 1998, 90 minutes after drink- ed that he had a seizure. Paramedics initially found
ing Ultimate Orange, which according to the label him apneic and in ventricular fibrillation. He was suc-
contained 415 mg of ma huang (ephedra alkaloids) cessfully resuscitated. Computed tomography of the
per serving as well as guarana (a source of caffeine), head showed cerebral edema but no hemorrhage or
and 5 minutes after running 3 miles (4.8 km) (Table masses. An initial electrocardiogram showed atrial flut-
4). He also regularly took multivitamins and amino ter, which subsequently converted to sinus rhythm.
acid supplements, but no other medications. On pres- An echocardiogram revealed mild left ventricular hy-
entation at a nearby hospital, his blood pressure was pertrophy. The plasma theophylline level was 11 µg
140/78 mm Hg and his pulse was 60 beats per min- per milliliter (therapeutic range, 10 to 20), and uri-
ute. A computed tomographic scan of the head re- nalysis revealed 12 µg of ephedrine per milliliter, 0.38
vealed a left-sided intrathalamic hemorrhage. Cerebral µg of pseudoephedrine per milliliter, and 0.41 µg of
angiography showed no evidence of vascular anom- phenylpropanolamine per milliliter. The treating car-
alies. The patient had gradual clinical improvement, diologist thought that the combination of ephedra
and his symptoms resolved except for persistent sen- alkaloids and caffeine in Ripped Force and the the-
sory loss on the right side of his face. Chemical analy- ophylline and albuterol medications caused a ventric-

Vol ume 343 Numb e r 25 · 1835

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on December 8, 2024. For personal use only.
No other uses without permission. Copyright © 2000 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

TABLE 4. OUTCOME IN 11 PATIENTS WITH ADVERSE EVENTS THAT WERE DEFINITELY OR PROBABLY RELATED TO THE USE
OF SUPPLEMENTS CONTAINING EPHEDRA ALKALOIDS.

ESTIMATED
DAILY
DOSE OF
PATIENT AGE (YR)/ EPHEDRA DURATION PREEXISTING CONDITIONS
NO. SEX NAME OF SUPPLEMENT ALKALOIDS OF USE ADVERSE EVENT OUTCOME OR CONCURRENT RISKS

mg

1 35/F Shape-Fast Plus 45 1 wk Subarachnoid hemorrhage Permanent disability None

2 22/M Ripped Force 20–60 Unknown Arrhythmia, cardiac arrest Permanent disability Asthma
3 28/F Herbalife’s Thermo- 21 1 day Cardiac arrest Permanent disability None
jetics
4 43/M Ripped Fuel 60 7 mo Cardiac arrest Death Family history of coronary
artery disease
5 37/F Metabolife 356 36 1 wk Severe hypertension, cardiac Death None
arrest, hypokalemia
6 59/F OmniTrim Extra 36 3 wk Acute myocardial infarction Coronary bypass surgery Hypertension
Vitamin-Fortified
tea
7 38/M Ripped Fuel 20 1 yr Arrhythmia, cardiac arrest Death None
8 47/F Total Control 44–66 9 mo Hypertension, bilateral lacunar Permanent disability Concomitant ingestion of
infarctions caffeine and ethanol
9 29/M Ultimate Orange 30 2 wk Stroke Permanent disability Concomitant use of dehy-
droepiandrosterone and
androstenedione
10 39/M Ultimate Orange Unknown Unknown Hemorrhagic stroke Permanent disability None
11 47/M Purple Blast Unknown 3 wk Hemorrhagic stroke Permanent disability Possible hypertension

ular arrhythmia that resulted in cardiac arrest. The pa- to the autopsy report included the comment, “ephe-
tient suffered anoxic encephalopathy and remained in drine is a stimulant medication, and as such may have
a vegetative state for several weeks. After one month contributed to a fatal arrhythmia in the decedent.”
in an acute care facility and six weeks at a rehabilitation
facility, he was discharged with substantial residual DISCUSSION
neurologic impairment. Ephedrine and related alkaloids have been associat-
ed with adverse cardiovascular events, including acute
Patient 7 myocardial infarction, severe hypertension, myocardi-
Patient 7 was an apparently healthy 38-year-old tis, and lethal cardiac arrhythmias.10,11 Constriction of
man who had been taking two capsules of Ripped coronary arteries and, in some cases, vasospasm are
Fuel (according to the label each capsule contains 10 believed to be the mechanisms of myocarditis and my-
mg of ephedrine and 100 mg of caffeine) each morn- ocardial infarction. The adrenergic effects of ephed-
ing for one year as directed on the product label (Ta- rine shorten cardiac refractory periods, permitting
ble 4). On June 6, 1996, he took his usual dose the development of reentrant cardiac arrhythmias.
along with a cup of coffee and went jogging for 20 Ephedrine can predispose patients to both hemor-
minutes. After returning home, he was talking with rhagic and ischemic stroke.12 Subarachnoid hemor-
his family when he suddenly collapsed and appeared rhage is thought to be a result of the hypertensive ac-
to have a tonic–clonic seizure. He had not reported tion of ephedrine, which can be short lived, or of
any symptoms before collapsing. He was in full car- cerebral vasculitis, which has been described in asso-
diac arrest when paramedics arrived and could not be ciation with a variety of sympathomimetic drugs.13,14
resuscitated. Autopsy showed mild cardiomegaly with Thrombotic stroke is presumably related to vasocon-
four-chamber dilatation and coronary artery disease, striction of large cerebral arteries, which leads to local
with narrowing of 50 to 75 percent in four vessels. thrombosis as a result of stasis and sympathomimetic-
The cause of death was acute arrhythmia resulting induced platelet activation.
from atherosclerotic cardiovascular disease. Subse- Caffeine is present in many products that contain
quent toxicology testing showed blood levels of 110 ephedra alkaloids, and those who take these products
ng of ephedrine per milliliter (the therapeutic range might also be consuming considerable quantities of
used for bronchodilation is 20 to 80).9 An addendum caffeine in coffee, tea, and soft drinks. Caffeine is like-

1836 · Decem b er 2 1 , 2 0 0 0

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on December 8, 2024. For personal use only.
No other uses without permission. Copyright © 2000 Massachusetts Medical Society. All rights reserved.
 A DV E RSE EVE NT S ASSOCIAT E D WIT H D IETA RY SUPPLEMENTS CONTA INING EPH ED RA A LK A LOID S

TABLE 5. OUTCOME IN 15 PATIENTS WITH ADVERSE EVENTS THAT WERE POSSIBLY RELATED TO THE USE
OF SUPPLEMENTS CONTAINING EPHEDRA ALKALOIDS.

ESTIMATED
DAILY
DOSE OF
PATIENT AGE (YR)/ NAME OF EPHEDRA DURATION PREEXISTING CONDITIONS
NO. SEX SUPPLEMENT ALKALOIDS OF USE ADVERSE EVENT OUTCOME OR CONCURRENT RISKS

mg

1 46/M Diet Fuel Unknown 5–6 mo Stroke Death None

2 22/M Ripped Fuel Unknown Unknown Hyperthermia, abnormal elec- Death None
trolyte levels, cardiac arrest
3 64/F Fit America Natural Unknown 2 mo Atrial fibrillation, stroke Permanent disability Hypertension, tran-
Weight Control sient ischemic attack
Aid
4 47/F Per-Form Dieter’s Unknown 6 mo Rhabdomyolysis, hydronephro- Prolonged hospital care None
Natural Tea sis, hypokalemia
5 64/F Shape-Fast 20 Unknown Hemorrhagic stroke Permanent disability None
6 34/M Herbalife’s Thermo- Unknown >3 wk Atrial flutter, renal failure, hy- Death None
jetics pokalemia, rhabdomyolysis
7 32/F Ripped Fuel 20 4 yr Premature delivery (34 wk of Death of neonate Smoking
gestation)
8 29/M Ultimate Nutrition Unknown 6–7 mo Stroke Permanent disability None
Product Ma Huang
9 15/F Ripped Fuel Unknown 2–3 wk Arrhythmia, cardiac arrest Death None
10 41/F Diet-Phen 12 1 mo Hypertension, multiple brain- Permanent disability Possible hypertension
stem infarcts
11 22/F Magic Herb 72 3 mo Spontaneous abortion at 9 wk Death of fetus None
12 43/F Metabolife 356 12 6 mo Severe hypertension, hemor- Permanent disability None
rhagic stroke
13 18/M Ultimate Orange Unknown Unknown Seizure, hemorrhagic stroke Death None
14 61/F Metabolife 356 24; increased 1 mo Hypertension, unstable angina Coronary bypass surgery Asthma
to 60
15 26/M Ripped Fuel 20–60 3 yr Status epilepticus, hypokalemia Permanent disability None

ly to enhance the cardiovascular and central nervous as reported on package labels, is typically about 20 mg
system effects of ephedrine. Caffeine acts by compet- per serving, and the usual dose frequency is two to
itively antagonizing the receptors for adenosine, a hor- three times per day. These products may contain larg-
mone released by endothelial cells that dilates blood er or smaller amounts of ephedra alkaloids than are list-
vessels.15 By inhibiting adenosine-mediated dilatation ed on the product label. For example, 11 of 20 supple-
of blood vessels, caffeine constricts blood vessels and ments tested by Gurley et al.18 either failed to list the
may increase blood pressure in persons prone to hy- alkaloid content on the label or had more than a 20
pertension. Caffeine also augments the release of cat- percent difference between the amount listed on the
echolamines, an effect that, when combined with that label and the actual amount.
of ephedrine, could lead to increased stimulation of Often, the dose of ephedrine that was associated
the central nervous system and cardiovascular system.16 with an adverse event was less than a typical dose of
Phenylpropanolamine, another ephedrine alkaloid, ephedrine used for bronchodilation (25 to 50 mg).
was marketed with caffeine in various weight-reduc- Experimental studies show that ephedrine has only
ing aids until 1983, when the combination was banned moderate effects on heart rate and blood pressure at
by the FDA after numerous reports of adverse ef- these doses.19,20 The discrepancy between such data
fects. Several studies have shown that caffeine and and our findings of serious adverse events reported
phenylpropanolamine have an additive effect on blood with the use of dietary supplements containing ephed-
pressure.17 These interactions between phenylpropa- ra alkaloids may be due to individual susceptibility,
nolamine and caffeine support the idea that the com- the additive stimulant effects of caffeine, the variabil-
bination of ephedrine and caffeine in a dietary supple- ity in the contents of pharmacologically active chem-
ment could increase the risk of adverse effects. icals in the products, or preexisting medical conditions.
The quantity of ephedrine in dietary supplements, Many of the cases we reviewed involved side effects

Vol ume 343 Numb e r 25 · 1837

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on December 8, 2024. For personal use only.
No other uses without permission. Copyright © 2000 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

such as anxiety, tremulousness, insomnia, palpitations, Abuse, National Institutes of Health. Dr. Haller was supported by a Veter-
ans Affairs Medical Toxicology fellowship and by a National Research Serv-
and personality changes that are well known to occur ice Award (T32GM07546) from the National Institutes of Health.
with the use of stimulant drugs. When ephedrine is
used for medical purposes, these types of reactions
are considered side effects and must be included in the We are indebted to Dr. Raymond Woosley for his critical review of
assessment of risks and benefits. In fact, ephedrine is the manuscript and to Kaye Welch for editorial assistance.
rarely prescribed today for medical purposes, because REFERENCES
newer drugs have more specific actions and fewer side 1. Adverse events associated with ephedrine-containing products — Texas,
effects. The risks of taking ephedra alkaloids as a di- December 1993-September 1995. JAMA 1996;276:1711-2.
etary supplement, however, are difficult to justify 2. Theoharides TC. Sudden death of a healthy college student related to
ephedrine toxicity from a ma huang-containing drink. J Clin Psychophar-
because the alkaloids have no demonstrated benefit. macol 1997;17:437-9.
Unlike vitamins and minerals, ephedra alkaloid sup- 3. Josefson D. Herbal stimulant causes US deaths. BMJ 1996;312:1378-9.
plements are not essential for proper nutrition. Peo- 4. Zahn KA, Li RL, Purssell RA. Cardiovascular toxicity after ingestion
of “herbal ecstasy.” J Emerg Med 1999;17:289-91.
ple who take these products to increase their exercise 5. Zaacks SM, Klein L, Tan CD, Rodriguez ER, Leikin JB. Hypersensi-
capacity or to lose weight place themselves at risk with- tivity myocarditis associated with ephedra use. J Toxicol Clin Toxicol 1999;
37:485-9.
out a substantial likelihood of benefit. 6. Vahedi K, Domigo V, Amarenco R, Bousser MG. Ischaemic stroke in a
A limitation to the use of reports of adverse events sportsman who consumed MaHuang extract and creatine monohydrate for
as an indicator of a product’s safety is that the number body building. J Neurol Neurosurg Psychiatry 2000;68:112-3.
7. Dietary supplements containing ephedrine alkaloids. FDA docket no.
of people at risk for the event is unknown. Manufac- 00N-1200. Rockville, Md.: Food and Drug Administration, 2000. (See
turers of dietary supplements that contain ephedra www.accessdata.fda.gov/scripts/oc/ohrms/index.cfm.)
alkaloids reported that 3 billion servings were sold 8. Blanc S, Leuenberger P, Berger JP, Brooke EM, Schelling JL. Judg-
ments of trained observers on adverse drug reactions. Clin Pharmacol Ther
in 1999.21 The number of servings that were actually 1979;25:493-8.
consumed is difficult to determine. Assuming that the 9. Costello JF, May CS, Paterson JW, Pickup ME. Pharmacokinetics of
ephedrine in asthmatics receiving acute and chronic treatment. Br J Clin
products were consumed as directed — three doses Pharmacol 1975;2:180-1. abstract.
per day for 12 weeks — then approximately 12 million 10. Wiener I, Tilkian AG, Palazzolo M. Coronary artery spasm and myo-
people used these supplements in 1999. cardial infarction in a patient with normal coronary arteries: temporal re-
lationship to pseudoephedrine ingestion. Cathet Cardiovasc Diagn 1990;
Another limitation is that adverse events are known 20:51-3.
to be underreported. Several studies have shown that 11. To LB, Sangster JF, Rampling D, Cammens I. Ephedrine-induced car-
spontaneous reporting of adverse events to MedWatch diomyopathy. Med J Aust 1980;2:35-6.
12. Bruno A, Nolte KB, Chapin J. Stroke associated with ephedrine use.
is not routine, and the rate of reporting may be less Neurology 1993;43:1313-6.
than 15 percent.22,23 The frequency of reports of ad- 13. Fallis RJ, Fisher M. Cerebral vasculitis and hemorrhage associated with
phenylpropanolamine. Neurology 1985;35:405-7.
verse reactions to herbal products is thought to be 14. Wooten MR, Khangure MS, Murphy MJ. Intracerebral hemorrhage
even lower.7,24 Therefore, the frequency of serious ad- and vasculitis related to ephedrine abuse. Ann Neurol 1983;13:337-40.
verse events associated with the use of supplements 15. Benowitz NL. Clinical pharmacology of caffeine. Annu Rev Med
1990;41:277-88.
containing ephedra alkaloids cannot be precisely deter- 16. Robertson D, Frölich JC, Carr RK, et al. Effects of caffeine on plasma
mined with the use of current reporting mechanisms. renin activity, catecholamines and blood pressure. N Engl J Med 1978;
Because of the severity of the adverse events that we 298:181-6.
17. Brown NJ, Ryder D, Branch RA. A pharmacodynamic interaction be-
reviewed and, in particular, the occurrence of events tween caffeine and phenylpropanolamine. Clin Pharmacol Ther 1991;50:
that caused permanent disability and death, we con- 363-71.
18. Gurley BJ, Gardner SF, Hubbard MA. Content versus label claims in
clude that dietary supplements that contain ephedra ephedra-containing dietary supplements. Am J Health Syst Pharm 2000;
alkaloids pose a serious health risk to some users. Al- 57:963-9.
though the incidence of serious adverse effects can- 19. Drew CD, Knight GT, Hughes DT, Bush M. Comparison of the ef-
fects of D-(¡)-ephedrine and L-(+)-pseudoephedrine on the cardiovascu-
not be determined from our analysis, our findings lar and respiratory systems in man. Br J Clin Pharmacol 1978;6:221-5.
arouse concern about the risks of these products, giv- 20. White LM, Gardner SF, Gurley BJ, Marx MA, Wang PL, Estes M.
en that they have no scientifically established benefits. Pharmacokinetics and cardiovascular effects of ma-huang (Ephedra sinica)
in normotensive adults. J Clin Pharmacol 1997;37:116-22.
Our findings indicate the need for a better understand- 21. Dietary supplement market view. Chevy Chase, Md.: FDC Reports,
ing of the determinants of individual susceptibility August 2000.
22. Chyka PA, McCommon SW. Reporting of adverse drug reactions by
to the serious adverse effects of dietary supplements poison control centres in the US. Drug Saf 2000;23:87-93.
containing ephedra alkaloids so that appropriate dos- 23. Spontaneous adverse event reports. (See http://www.fda.gov/ohrms/
ing guidelines and warnings can be devised. dockets/ac/99/slides/3558sle/tsld003.htm.)
24. Barnes J, Mills SY, Abbot NC, Willoughby M, Ernst E. Different
standards for reporting ADRs to herbal remedies and conventional OTC
Supported by funds from the Food and Drug Administration and by medicines: face-to-face interviews with 515 users of herbal remedies. Br J
grants (DA02277 and DA12393) from the National Institute on Drug Clin Pharmacol 1998;45:496-500.

1838 · Dec em b er 2 1 , 2 0 0 0

The New England Journal of Medicine is produced by NEJM Group, a division of the Massachusetts Medical Society.
Downloaded from nejm.org on December 8, 2024. For personal use only.
No other uses without permission. Copyright © 2000 Massachusetts Medical Society. All rights reserved.