nature publishing group
From 1994 to 2004, ephedra-containing dietary supplements
were marketed and sold as weight-loss aids, “energy boosters,”
and exercise performance enhancers. Incorporating natural
sources of the drugs ephedrine, pseudoephedrine, norephed-
rine, and caffeine, ephedra-containing supplements resulted in
thousands of serious adverse event reports being submitted to
the US Food and Drug Administration (FDA), and more than
70 case reports appeared in the medical literature documenting
heart attacks, strokes, seizures, and psychoses among users of
these products.1–5 During this time frame, the FDA’s ability to
quickly remove these products from the market was limited by
the Dietary Supplement Health and Education Act of 1994. In
the interim, a combined outcry from various health-care organi-
zations, numerous court litigations against ephedra supplement
manufacturers, and the outright banning of such products in
several states led many supplement manufacturing companies
to develop ephedra-free formulations. In April 2004, the FDA
removed ephedra-containing supplements from the market,
citing unacceptable risks of adverse health effects;5 however,
many ephedra-free supplements containing ingredients similar
in chemical structure and pharmacologic activity to those found
in their controversial predecessors continue to be marketed. An
examination of ingredients commonly found in ephedra-free
supplements reveals that many substitutes for ephedra (e.g.,
extracts from Citrus aurantium, cocoa, and Coleus forskolii)
contain phytochemicals (e.g., synephrine, phenylethylamine,
and forskolin) that have pharmacologic properties similar to
those of the ephedrine alkaloids. When formulated with natural
sources of caffeine (e.g., guarana, green tea, and Yerba maté)—
a process common to most ephedra-free supplements—a new
category of natural stimulants emerges.
Since 2004, numerous case reports have appeared in the medi-
cal literature linking ephedra-free supplements to ischemic
stroke,6,7 myocardial infarction,8,9 hypertensive emergencies,10,11
Four popular ephedra-free dietary supplements were evaluated for their effects on heart rate (HR), blood pressure
(BP), and electrocardiographic (ECG) parameters. Twelve healthy men participated in a study randomized for product
sequence, with a 21-day washout period between supplement-administration phases. Throughout the study, Holter
monitors were used to assess ECG and HR activity. BP was assessed automatically on multiple occasions. The supplements
were ingested three times daily for 3 days. Caffeine content, microbial load, and serum caffeine concentrations were
determined. Mean systolic (SBP) and diastolic BP (DBP) readings showed significant increases relative to baseline
(10.8 ± 2.5 and 5.3 ± 3.1 mm Hg, respectively; P < 0.05). All supplements significantly increased HR and decreased
bradycardia runs; abnormal atrial/ventricular events were frequently noted. Gastrointestinal and sympathomimetic
symptoms were also common. Two supplements were heavily contaminated with Bacillus species. In light of these
findings, the use of ephedra-free dietary supplements should be discouraged in individuals with hypertension, diabetes,
or other cardiovascular diseases.
Received 24 August 2012; accepted 4 December 2012; advance online publication 30 January 2013. doi:10.1038/clpt.2012.241
Multiple Dosing of Ephedra-Free
Dietary Supplements: Hemodynamic,
Electrocardiographic, and Bacterial
L Foster1, MC Allan2,6, A Khan3, P Moore4, DK Williams4, M Hubbard5, L Dixon5,7 and BJ Gurley5
1Department of Dietetics and Nutrition, College of Health Professions, University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA; 2Department of
Cardiology, College of Medicine, University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA; 3Division of Microbiology, National Center for Toxicological
Research, US Food and Drug Administration, Jefferson, Arkansas, USA; 4Department of Biostatistics, College of Public Health, University of Arkansas for Medical
Sciences, Little Rock, Arkansas, USA; 5Department of Pharmaceutical Sciences, College of Pharmacy, University of Arkansas for Medical Sciences, Little Rock,
Arkansas, USA; Current affiliations: 6Parkridge Medical Center, Diagnostic Cardiology Group, Chattanooga, Tennessee, USA; 7University of Mississippi, College of
Pharmacy, University, Mississippi, USA. Correspondence: BJ Gurley (email@example.com)
ClInICAl PHARMAColoGy & THERAPEuTICS | VOLUME 93 NUMBER 3 | MARCH 2013
rhabdomyolysis,12 psychosis,13 and altered electrocardiographic
(ECG) findings, both at rest14,15 and during exercise.16 These
findings closely mimic those reported for ephedra-containing
supplements, suggesting that substitution of other sympathomi-
metic agents for ephedrine alkaloids may not have lessened the
risk of adverse health effects. Although several small prospec-
tive studies using ephedra-free supplements in healthy volun-
teers have found either no significant changes in blood pressure
(BP), heart rate (HR),17–19 or ECG readings20 or only modest
increases in either systolic (SBP)21,22 or diastolic BP (DBP),23,24
it should be noted that many of these studies investigated the
outcomes of only a single, one-time dose17,20,21,25 or of doses
that were only half of those in the label recommendation.19,20
Our purpose was to integrate label recommendations from four
different ephedra-free formulations—all of which contain natural
sources of caffeine—into a common, three-times-daily regimen
and to determine their effects on various hemodynamic and ECG
parameters over a period of 3 days.
General experimental observations
No serious adverse health events occurred during the study;
however, several side effects were reported, the most prevalent
being nausea, vomiting, and sympathomimetic-related symp-
toms (Table 1). Interestingly, these observations appeared
to correlate to caffeine/paraxanthine serum concentrations
(Table 2). On caffeine analysis, two products (Guarana and
Xenadrine EFX) were found to be heavily contaminated with
Bacillus species. Of these, Xenadrine EFX was associated with
more frequent episodes of nausea, a finding that may be related,
in part, to bacterial contamination.
Caffeine analysis: product content and serum concentrations
Product ingredients and caffeine content are listed in Table 3.
Zantrex 3 contained the most caffeine (183 mg per capsule),
followed by Xenadrine EFX, Guarana, and Metabolift, contain-
ing 123, 116, and 90 mg of caffeine, respectively, per capsule.
Accordingly, the largest daily dose of caffeine was from Zantrex
3 (1,098 mg), and Metabolift accounted for the smallest daily
dose (540 mg). The caffeine content in Guarana, Metabolift, and
Xenadrine EFX was similar to that reported by Andrews et al. for
53 different caffeine-containing dietary supplements.26 Zantrex
3, however, exceeded the upper limit reported by Andrews et al.
(1,098 vs. 829 mg/day). Xenadrine EFX, the most complex of the
formulations, also contained measurable amounts of the meth-
ylxanthines theobromine (4.6 mg per capsule) and theophyl-
line (1.7 mg per capsule), amounting to daily doses of 27.6 and
10.2 mg, respectively. Only trace amounts of theobromine and
theophylline were present in the other products. An absence of
measurable serum caffeine and paraxanthine concentrations at
baseline before each study phase confirmed subject compliance
with respect to abstinence from caffeine. Serum caffeine and par-
axanthine concentrations, obtained 2 h after the first dose on days
1 and 3, are listed in Table 2. These concentrations appeared to
correlate with the administered caffeine doses, with Zantrex 3
producing the highest concentrations and Metabolift the lowest.
Hemodynamic effects of supplement administration
As compared with Guarana, a significant increase (P = 0.01) in
mean SBP (7.0 ± 2.7 mm Hg) was noted for Xenadrine EFX; this
occurred on day 3. However, when compared with baseline assess-
ments, significant changes in mean SBP and DBP were observed
for all the supplements (Table 4). A mean increase in SBP of
10.8 ± 2.5 mm Hg was noted on day 1 after ingestion of ephedra-
free supplements. Lower mean increases in SBP (7.5 ± 3.2 mm Hg)
were observed on day 3. Xenadrine EFX produced the greatest
average increase in SBP on day 1 (13.1 ± 3.9 mm Hg) and on day 3
(9.7 ± 2.2 mm Hg). Interestingly, SBP readings of 151 and 153 mm
Hg—the highest values measured in the study—were observed in
subject 4, 2 hours after the first doses of Metabolift and Guarana,
respectively, on day 1 of each of these phases. Nineteen SBP meas-
urements of >140 mm Hg were observed over the course of the
study, most of these on day 1. The DBP readings increased by an
average of 5.3 ± 3.1 mm Hg on day 1 and were slightly lower by
day 3 (4.7 ± 2.8 mm Hg). The greatest increases in DBP on day
1 were noted for Xenadrine EFX and Zantrex 3. On day 3, only
Xenadrine EFX produced statistically significant increases in
DBP after the first and second doses. During the study, 10 DBP
Table 1 Self-reported side effects of ephedra-free supplements
Zantrex 3 Xenadrine EFXMetaboliftGuarana
Insomnia (5) Insomnia (2)Insomnia (3)Insomnia (4)
Nausea (6)Nausea (6)Nausea (3)Nausea (1)
Indigestion (2) Indigestion (1)Indigestion (1)
Vertigo (1) Vertigo (1) Vertigo (1)
Flushing (2) Heart
Headache (2)Muscle aches (1) Headache (1)
Tinnitus (1) Vomiting (1)
Loss of appetite (2)
Table 2 Serum caffeine and paraxanthine concentrations (μg/ml; mean ± SD)
Guarana (mean ± SD) Metabolift (mean ± SD)Xenadrine EFX (mean ± SD) Zantrex 3 (mean ± SD)
Day 1 Day 3Day 1 Day 3Day 1Day 3Day 1 Day 3
Caffeine 3.72 ± 0.956.85 ± 2.182.61 ± 0.56 5.51 ± 2.143.46 ± 0.84 7.71 ± 3.395.53 ± 1.36 14.92 ± 6.40
Paraxanthine0.64 ± 0.263.49 ± 0.700.57 ± 0.21 3.11 ± 1.320.59 ± 0.25 3.71 ± 1.151.04 ± 0.53 6.11 ± 2.27
268 VOLUME 93 NUMBER 3 | MARCH 2013 | www.nature.com/cpt
readings of >90 mm Hg were noted. The interproduct differ-
ences in effects on SBP and DBP also appear to correlate with the
caffeine content of the doses; for instance, two capsules of Zantrex
3 contained 366 mg of caffeine whereas two capsules of Xenadrine
EFX contained 246 mg (Table 1). The effects of the supplements
on BP appeared to be somewhat tempered by day 3 of dosing, in
accordance with the development of pharmacologic tolerance to
caffeine.27 No significant differences were noted between base-
line SBP and DBP assessments on days 1 and 3, indicating that
hemodynamic effects were temporal and had dissipated during
the 12-h interval between daily regimens. Moreover, four sepa-
rate baseline assessments for SBP and DBP were obtained in each
subject, and no statistically significant differences were observed
within or between subjects.
Holter monitor findings
As compared with the parameters related to the intake of
Guarana, no significant changes in ECG parameters were
observed with respect to any of the multicomponent supple-
ments; however, when compared with baseline, several signifi-
cant changes were observed. Selected data collected from 24-h
Holter monitors are depicted in Figures 1–3. Figure 1 denotes
the effect of ephedra-free-supplement administration on 24-h
maximum HR readings at baseline, day 1, and day 3. Significant
increases in maximum HR were noted in association with the
administration of each of the supplements except Xenadrine
EFX. Maximum HR readings of >140 beats per minute (bpm)
were noted on 17 occasions; the highest (177 bpm) occurred
in subject 6 on day 1 of Zantrex 3 administration. Mean HR
increased after ingestion of each of the supplements evaluated
(data not shown); however, only Zantrex 3 produced statisti-
cally significant increases in mean HR (4.0 ± 1.4 bpm) on both
days 1 and 3, whereas Metabolift produced significant increases
on day 3 only (4.9 ± 2.1 bpm). No significant increases in 24-h
minimum HR were noted after administration of any of the sup-
plements. Increases in the mean number of supraventricular
events were observed relative to presupplement levels during
each of the ephedra-free-supplement administration periods,
but statistically significant increases were noted only during
administration of Xenadrine EFX (both days) and Metabolift
(only on the first day) (Figure 2). The most common supraven-
tricular events were short runs (<10 beats) of supraventricular
tachycardia; these were observed for all the products studied,
with 12 episodes occurring primarily in two of the subjects.
Ventricular events marked by short runs of ventricular tachycar-
dia (<5 beats) and premature ventricular contractions increased
during all the supplement-administration phases (data not
shown), but statistically significant increases were noted only for
Zantrex 3 on day 3. However, the increase in ventricular events
approached significance on day 1 for both Xenadrine EFX (P =
0.05) and Zantrex 3 (P = 0.06). Statistically significant decreases
Table 3 Product ingredients and caffeine content
(mg per capsule/daily caffeine dose)
Xenadrine EFX Zantrex 3
Green tea Green tea
(116/696 mg) Yerba matéYerba matéCitrus
Salvia sclarea extract
Calcium tribasic Damiana
Grape seed extract
Table 4 Effect of ephedra-free supplementation on systolic and diastolic blood pressure
(mm Hg, mean ± SD)
(mm Hg, mean ± SD)
(mm Hg, mean ± SD)
(mm Hg, mean ± SD)
Day 1 (presupplementation) SBP/DBP118.9 ± 7.7/72.9 ± 5.4 119.9 ± 10.3/73.1 ± 7.3117.4 ± 10.5/72.6 ± 9.5119.8 ± 6.1/72.4 ± 4.1
Day 1 (2 h after first dose) SBP/DBP129.4 ± 9.8*/79.3 ± 12.1* 128.3 ± 9.0*/76.1 ± 10.0 128.8 ± 8.7*/79.1 ± 9.6*131.6 ± 10.9*/77.9 ± 8.2*
Day 1 (2 h after second dose) SBP/DBP 129.2 ± 8.5*/75.9 ± 11.8 127.9 ± 11.8*/77.2 ± 12.5 133.2 ± 9.4*/79.4 ± 12.9*129.8 ± 12.3*/78.6 ± 8.3*
Day 3 (presupplementation) SBP/DBP 119.3 ± 6.7/73.8 ± 6.7 118.4 ± 7.8/72.1 ± 7.8117.4 ± 9.7/72.7 ± 7.9 119.9 ± 5.2/73.4 ± 6.6
Day 3 (2 h after first dose) SBP/DBP124.9 ± 11.2/75.5 ± 10.6 126.0 ± 8.1*/75.9 ± 11.7129.6 ± 9.8*/80.5 ± 9.4*127.5 ± 13.2*/78.3 ± 7.5*
Day 3 (2 h after second dose) SBP/DBP 124.9 ± 7.4/75.9 ± 7.9 126.0 ± 11.9*/77.8 ± 11.3*125.9 ± 6.7*/77.0 ± 9.6125.4 ± 12.6/76.4 ± 6.7
DBP, diastolic blood pressure; SBP, systolic blood pressure.
*P < 0.05.
ClInICAl PHARMAColoGy & THERAPEuTICS | VOLUME 93 NUMBER 3 | MARCH 2013
in bradycardia runs on days 1 and 3 were observed during each
of the supplement-administration phases (Figure 3).
Some subjects appeared to be more susceptible than others to
the cardiovascular effects of these products. Pressor effects were
most notable in subjects 3, 7, and 12, and ECG events were more
pronounced in subjects 3, 7, 9, and 11.
During sample preparation for determining caffeine content,
a putrid odor emanated from Xenadrine EFX, indicating
possible microbial contamination. Samples of each prod-
uct were further analyzed for bacterial contamination using
standard microbiological methods. Two products were heavily
contaminated with Bacillus species: Bacillus lentus, Bacillus
pumilus, Bacillus megaterium, Paenibacillus thiaminolyticus,
and Paenibacillus amylolyticus. The microbial burden (colony-
forming units per gram, cfu/g) for each product was as fol-
lows: Xenadrine EFX (3.8 × 106 cfu/gm), Guarana (2.8 × 104
cfu/gm), Metabolift (no growth), Zantrex 3 (no growth). PCR
detection of enterotoxin and emetic-toxin genes revealed no
evidence of the human pathogen Bacillus cereus in the two
contaminated products. However, follow-up PCR analysis of
a separate lot of Xenadrine EFX (lot no. 82968) revealed the
presence of Bacillus cereus.
The advent of “ephedra-free” dietary supplements has brought
with it a connotation of “trouble-free” with regard to health con-
cerns of consumers in the United States. Somewhat troubling
to the medical community, and eerily reminiscent of the early
stages of the ephedra controversy, is the increasing number of
case reports documenting significant sympathomimetic-related
side effects associated with the use of such supplements.6–16 To
date, only a few prospective studies have investigated the safety
aspects of these products, and among these studies there is disa-
greement as to what potential health risks these entities might
pose.17–22,25,28 Concern about the safety of using these products
is well founded, given that most “ephedra-free” dietary supple-
ments incorporate a multitude of botanical extracts, many of
which contain significant amounts of caffeine and “natural”
sympathomimetic agents. Although the pharmacology of meth-
ylxanthines—particularly caffeine—has been thoroughly inves-
tigated, the cardiovascular effects of multicomponent dietary
supplement formulations containing guarana, green tea, yerba
maté, and kola nut extracts appear to differ from those of tradi-
tional dosage forms of caffeine.
One of the most noteworthy observations during this inves-
tigation was the significant increase in both SBP and DBP pro-
duced by each of the ephedra-free products throughout the
3-day study period, especially after the initial dose on day 1.
Given that the principal phytochemical components of these
Figure 1 Effect of ephedra-free supplements on 24-h maximum heart rate.
Error bars = SD. *Statistically significant difference from baseline. bpm, beats
Xenadrine EFXZantrex 3
24-h Maximum heart rate (bpm)
Day 1 Day 3 Baseline Day 1Day 3
BaselineDay 1 Day 3BaselineDay 1 Day 3
Figure 2 Effect of ephedra-free supplements on 24-h supraventricular
cardiac events as recorded by Holter monitoring. Error bars = SD. *Statistically
significant difference from baseline.
Xenadrine EFX Zantrex 3
24-h Superventricular events
Day 1 Day 3BaselineDay 1Day 3
Baseline Day 1Day 3 BaselineDay 1 Day 3
Figure 3 Effect of ephedra-free supplements on 24-h bradycardia runs
as recorded by Holter monitoring. Error bars = SD. *Statistically significant
difference from baseline.
Baseline Day 1 Day 3
Bradycardia runs per day
GuaranaMetabolift Xenadrine EFXZantrex 3
270 VOLUME 93 NUMBER 3 | MARCH 2013 | www.nature.com/cpt
products are methylxanthines, particularly caffeine, statistically
significant increases in BP were not unexpected in response to
these doses of caffeine.23,24,29 Such increases in BP are probably
attributable to the relatively high doses of caffeine ingested per
day, and the resultant serum levels of caffeine and paraxanthine
observed on day 3. Paraxanthine, the principal metabolite of
caffeine, elicits hemodynamic effects similar to those of its par-
ent compound.30 Moreover, caffeine exhibits dose-dependent
metabolism31 and, given the large doses of caffeine inherent in
products such as Zantrex 3, it is conceivable that the disposition
of caffeine in the subjects in our study was dissimilar to that
resulting from products with lower caffeine content.
Our findings also corroborate those of Haller et al., who
observed significant elevations in SBP and DBP after a sin-
gle dose of an “ephedra-free” dietary supplement (Xenadrine
EFX).22 Increases in SBP and DBP of a magnitude similar to
that observed by us have been reported for 400-mg doses of pure
caffeine.24 In the current study, typical caffeine doses ranged
from 180 mg (Metabolift) to 366 mg (Zantrex 3). Therefore, caf-
feine may not have been the only ingredient responsible for the
hemodynamic changes observed. Several of the product labels
listed other sympathomimetics (e.g., synephrine and phenyleth-
ylamine) and methylxanthines (e.g., theobromine and theophyl-
line) as part of their “proprietary blends.” This was corroborated
by our analysis of Xenadrine EFX, which revealed measurable
levels of theobromine and theophylline. The additive effects of
these other constituents cannot be overlooked.
Other phytochemicals that may have contributed to the
observed effects include the catechin polyphenols (e.g.,
epicatechin gallate and epigallocatechin gallate). It has been
demonstrated that catechin polyphenols found in green tea
exacerbate the sympathomimetic effects of ephedrine when
administered concomitantly, an effect apparently related to cat-
echin-mediated inhibition of catechol-O-methyltransferase.32
Like caffeine and the other methylxanthines, catechins are bio-
available33; consequently, additive or even synergistic hemo-
dynamic effects are not unfeasible when significant doses of
caffeine from natural sources are consumed. Guarana, an ingre-
dient common to all the products examined in this study, is
another source of catechin polyphenols, and such phytochemical
combinations (i.e., catechins/methylxanthines) may have con-
tributed to the effects observed in our subjects.
It must also be emphasized that botanical extracts are con-
centrated collections of numerous phytochemicals whose
individual pharmacologic activities are not well understood,
let alone their combined effects. Therefore, the mechanism(s)
underlying the effects brought about by these multi-ingredient
supplements may be thought of as a collection of herb–herb
Our findings also suggest that the prolonged use of certain
“ephedra-free” supplements may alter cardiac electrophysiol-
ogy. These findings tend to support those reported earlier by
Gurley et al.28 but are at variance with those of Min et al.20
This discrepancy may stem from the fact that Min et al. admin-
istered only a single tablet (half the recommended dose) of
ephedra-free Metabolife and monitored their subjects for only
5 h. Although the consumption of caffeine (i.e., coffee) alone
is not thought to pose a significant risk for the development
of arrhythmias,34 it is important to remember that “ephedra-
free” supplements are more than just sources of natural caf-
feine. Moreover, prolonged ingestion of high doses of caffeine
may induce hypokalemia.35 Although none of the ECG-related
anomalies reported here was considered clinically significant,
consumers with underlying cardiovascular illnesses or those
taking other arrhythmogenic medications should either con-
sult their health-care provider before using these products or
avoid them altogether.
It must be noted that, in order to balance the dosing regimens
in this study, six capsules of Xenadrine EFX were administered
daily, exceeding the label recommendation by two capsules.
Nevertheless, most of the significant cardiovascular effects
observed in this study appear attributable to the first two doses
(four capsules). Therefore, our findings lend credence to the
relevance of the extensive warning on the Xenadrine EFX label.
A potential shortcoming of the study was the absence of a
placebo arm in the design. Ideally, a decaffeinated Guarana
supplement could have served as a placebo; however, no such
products were commercially available at the time. A placebo may
also have aided in discriminating between some of the reported
side effects; however, all the reported side effects have also been
described in other placebo-controlled studies of caffeine-con-
An unexpected finding in this study was the heavy bacterial
contamination of some of the products. Given the nature and
source of botanical extracts, some microbial contamination is
anticipated. For dietary supplements, the US Pharmacopoeia sets
acceptable limits for nonpathogenic bacteria in botanical dietary
supplements at 1 × 104 cfu/g of material.36 In two of the four prod-
ucts, this limit was exceeded. (An additional lot of Xenadrine EFX
that was not used in this study did contain the pathogen B. cereus.)
It is conceivable that the presence of these organisms contrib-
uted to the gastrointestinal side effects seen in association with
Xenadrine EFX and Guarana. Microbial contamination of dietary
supplements37,38 and their potential for toxicity39 are not unex-
pected, given their “natural” origins and the less rigorous regula-
tion of dietary supplements by the FDA. Hopefully, the recent
implementation of the FDA’s Good Manufacturing Practices for
Dietary Supplements will curb this problem.
Somewhat related to the absence of information regarding
microbial contamination was the lack of label information on
caffeine content. For consumers who are sensitive to caffeine or
are looking to restrict their caffeine intake, the labels for these
products may be difficult to interpret. Xenadrine EFX provided
no indication of the caffeine content per dose other than “the
recommended dose of this product contains about as much
caffeine as a cup of coffee.” Our analysis revealed that a rec-
ommended dose of two capsules contained 246 mg of caffeine,
about twice the amount in an average cup of coffee.40 The label
for Zantrex 3 stated only that a dose of two capsules contains “in
excess of 300 mg.” In reality, two capsules of Zantrex contained
366 mg of caffeine, and the recommended daily dose would
exceed 1,000 mg of caffeine. The Metabolift label stated that its
ClInICAl PHARMAColoGy & THERAPEuTICS | VOLUME 93 NUMBER 3 | MARCH 2013
800 mg of guarana extract was standardized to 22%; however,
given that green tea was also a component in the “proprietary
blend,” the consumer would not have been able to determine
This ambiguity in label information is disconcerting given
that many individuals are sensitive to the effects of caffeine.
In this study, five subjects accounted for a preponderance of
the more remarkable hemodynamic and ECG findings, with
one (subject 7) exhibiting the most such effects. Such sensi-
tivities may stem from genetic polymorphisms in adenosine
A2a receptors,41 CYP1A2,42 or catechol-O-methyltransferase
enzymes,43 or they may be related to the effects of other medi-
cations (e.g., oral contraceptives)44 and/or exercise45 on caf-
feine pharmacokinetics. Exercise can increase serum caffeine
concentrations,45 and this effect may be magnified in individu-
als consuming ephedra-free supplements because the product
labels often recommend that they be taken before exercising.
Given the cardiovascular effects noted in this study, and the
fact that these products contain myriad botanical extracts, the
possibility of herb–drug interactions must also be considered.46
Consumers with hypertension who ingest an “ephedra-free”
supplement along with conventional antihypertensive medica-
tions may not achieve adequate control of their BP. Also, St John’s
wort and Piper nigrum are listed as ingredients on the labels for
Metabolift and Zantrex 3, respectively. St John’s wort is one of the
most problematic botanicals and is known to be involved in phar-
macokinetic herb–drug interactions.47 Hyperforin, a phytochem-
ical component in St John’s wort, is a potent inducer of many
cytochrome P450 drug metabolizing enzymes (e.g., CYP3A4,
CYP2C9) as well as the drug efflux pump, P-glycoprotein.
Prolonged ingestion of St John’s wort, therefore, renders many
medications ineffective. Conversely, piperine, a phytochemical
found in the pepper species P. nigrum and Piper longum, is a
potent inhibitor of human CYP3A4. P. nigrum extracts have been
shown to enhance the bioavailability and potential toxicity of
several drugs.47 It is therefore conceivable that consumers taking
Metabolift or Zantrex 3 with other conventional medications may
be at risk for serious herb–drug interactions.
In summary, our findings suggest that significant hemodynamic
and ECG effects may occur when “ephedra-free” dietary supple-
ments are consumed for prolonged periods. Although these find-
ings were obtained with healthy young adult subjects, the effects
may be especially detrimental in consumers who are sensitive to
caffeine or who have underlying medical conditions. The possibil-
ity of additional health risks such as microbial contamination and
the potential for herb–drug interactions must also be considered
before these products are consumed. Currently, product labels
of most ephedra-free supplements warn against possible adverse
health effects related to methylxanthine ingestion, especially in
persons with underlying cardiovascular conditions. This is cer-
tainly a prudent practice, given that the connotation “ephedra-
free” may not necessarily mean “trouble-free.”
Subjects. Attempts were made to recruit subjects of both sexes, but
only male subjects responded. Twelve healthy male volunteers (age,
mean ± SD = 31.2 ± 7.8 years; weight, mean ± SD = 88.2 ± 11.8 kg) were
enrolled in the study. Exclusion criteria were a history of consump-
tion of any botanical dietary supplement, any known medical condi-
tion, smoking, or current use of any prescription or nonprescription
medication. The subjects were required to abstain from caffeine for 5
days prior to and throughout each active study phase. During recruit-
ment, questions regarding daily caffeine consumption revealed mild
to moderate intakes (no more than two servings of coffee/tea/cola
daily). The need for abstention from caffeine-containing beverages
was emphasized through e-mail and telephone reminders. Food/
medication diaries were maintained to document any breaches of this
requirement. The subjects were also required to abstain from rigorous
exercise throughout the study period. Before entering the study, all
subjects underwent 12-lead ECG and a physical examination. Baseline
chemistry panels and complete blood counts were obtained to rule out
any baseline biochemical abnormalities. Subjects who showed normal
findings at the baseline visit returned for placement of a Holter moni-
tor. The University of Arkansas for Medical Sciences Human Research
Advisory Committee approved the protocol, and participants provided
written informed consent before any study-related procedures were
Dosing and ECG and hemodynamic monitoring. The study had an open-
label crossover design with randomization for the sequence of adminis-
tration of the supplements. Each supplement administration phase was
3 days long and was separated from the subsequent phase by a 3-week
washout period. Ephedra-free supplements were obtained from local
retailers in Little Rock, AR: Zantrex 3 (Zoller Laboratories, Salt Lake
City, UT; lot no. 053191), Xenadrine EFX (Cytodyne, Hicksville, NY;
lot no. 81498), Metabolift (Twin Laboratories, American Fork; UT, lot
no. 205112473), and Guarana (Solaray, Park City, UT; lot no. 090911).
Label recommendations for dosing Guarana, Metabolift, and Zantrex 3
were two capsules three times daily; for Xenadrine EFX it was two cap-
sules twice daily. To have practical, replicable, and comparable dosing
regimens, two capsules of each product were administered at 9:00 am,
3:00 pm, and 9:00 pm during each 3-day supplementation phase. Blood
samples (7 ml) were drawn from an antecubital vein for serum caffeine
and paraxanthine analysis at baseline and at 2 h after the first dose (i.e.,
at 11 am) on day 1 and again at a similar time point on day 3. The 2-h
time point was selected to correspond with peak caffeine concentra-
tions and maximum SBP readings after the administration of an acute
dose of Xenadrine EFX, as previously reported by Haller et al.22
Holter monitors (Aria Digital Recorders; Del Mar Reynolds Medical,
Irvine, CA) were placed on each subject for 24 h to assess ECG activity.
Holter assessments were determined at baseline (24 h before each sup-
plementation phase) and again on days 1 and 3 of each supplementa-
tion phase. ECG data from each Holter assessment were analyzed using
the Impresario and CardioNavigator Plus software (Del Mar Reynolds
Medical). The following software analysis settings were used to generate
ECG parameters: minimum ventricular tachycardia rate (100 bpm); R
on T interval (370 ms); minimum supraventricular ectopic rate (80 bpm);
pause interval (2 s); supraventricular ectopic prematurity (20%); T-wave
amplitude (0.7 mV); bradycardia rate (60 bpm); bradycardia interval
(300 ms); blanking period (250 ms); form sensitivity;5 and HR min/max
(1-min intervals). A cardiologist evaluated all Holter event reports and
selected ECG recordings. BP was measured in each subject at baseline
and again at 2 h after the first dose (i.e., at 11 am) and at 2 h after the
second dose (i.e., at 5 pm) during each phase. Casual cuff measurements
of BP were carried out using a Dinamap Pro 100 automated blood pres-
sure monitor (MEDWorldwide, Miami, FL) on the dominant arm of the
subject after a minimum of 10 min of rest in a seated position.
Analyses for caffeine. Twelve capsules of each ephedra-free supplement
were analyzed for caffeine content using the high-performance liquid
chromatographic method described by Andrews et al.26 Serum caf-
feine and paraxanthine concentrations were determined in duplicate
using the high-performance liquid chromatography method described
272 VOLUME 93 NUMBER 3 | MARCH 2013 | www.nature.com/cpt
by Holland et al.48 The limits of quantitation for both serum caffeine
and serum paraxanthine were set at 0.10 μg/ml, with accuracy and pre-
cision assessments <10%.
Isolation and identification of bacteria from ephedra-free diet supple-
ments. Three capsules of each ephedra-free supplement were broken
under sterile conditions and dissolved by occasional vortex mixing
(15 min) in a dilution of Butterfield’s phosphate-buffer (pH 7.2) (1:10
wt/vol). For enrichment, 1-ml inocula of 10−2, 10−3, and 10−4 dilutions
of the samples were added to three tubes of 15 ml trypticase soy-pol-
ymyxin broth. The inocula were incubated at 30 °C for 48 h. A loop-
ful of inocula from positive tubes and/or lowest dilution tubes was
transferred and streaked on mannitol-egg yolk-polymyxin agar plates
and Bacara plates (AES Chemunex, Cranbury, NJ).49 After the plates
were incubated at 30 °C overnight, samples positive for B. cereus were
selected based on the color of the colonies and formation of white pre-
cipitates (which indicates the production of lecithinase). To confirm
other bacterial isolates at the genus and species level, the VITEK2
System was used (BioMerieux VITEK, Hazelwood, MO).50
PCR amplification of genes producing B. cereus enterotoxin and emetic
toxins. B. cereus type strain ATCC 14579 was used as a positive refer-
ence control for eight enterotoxin genes.51 DNA templates were pre-
pared from 4- to 6-h culture grown in nutrient broth at 30 °C. The cells
were collected from 1 ml of culture broth by centrifugation at 6,000g
for 5 min. The cell pellet was washed once with 1 ml of sterile water,
suspended in 100 μl of sterile water, and boiled for 10 min. The boiled
samples were centrifuged and the supernatant was used as the tem-
plate DNA. DNA was isolated using the Qiagen chromosomal DNA
isolation kit (Qiagen, Valencia, CA). The primers for eight enterotoxin
genes, including those for emetic toxins,51 were synthesized and used
to detect enterotoxin genes in Bacillus strains isolated from dietary
Statistical methods. Nine individual repeated-measures analysis of
variance models were used to test the differences between the four sup-
plements with respect to nine different cardiovascular measurements
(24-h mean HR, minimum HR, maximum HR, HR, bradycardia runs,
SBP, DBP, ventricular events, and supraventricular events) across
multiple time points. Each model included a single cardiovascular
measurement as the response variable and the following as independ-
ent variables: supplement (Guarana, Metabolift, Xenadrine EFX, or
Zantrex 3), time, and time by supplement interaction effect. The 24-h
mean HR, minimum HR, maximum HR, bradycardia runs, ventricu-
lar events, and supraventricular events were measured across three
time points (baseline, day 1, and day 3); SBP, DBP, and HR were meas-
ured at a total of six time points, three each on days 1 and 3 ( baseline,
2 h, and 8 h). A compound symmetry covariance structure was used in
all repeated-measures analysis of variance models. For models with an
insignificant interaction effect and a significant main effect, Tukey’s
post hoc multiple comparisons were made. A significance level of 0.05
was set for all the analyses results. All the analyses were conducted
using SAS 9.3 (SAS Institute, Cary, NC).
Funding was provided by the Gustavus & Louise Pfeiffer Medical Research
Foundation. The views presented in this article do not necessarily reflect
those of the US Food and Drug Administration.
B.J.G., L.F., and P.M. wrote the manuscript. B.J.G. and D.K.W. designed the
research. L.F., M.C.A., A.K., M.H., and L.D. performed the research. B.J.G., L.F.,
M.C.A., A.K., P.M., D.K.W., M.H., and L.D. analyzed the data. A.K. and M.H.
contributed new reagents/analytical tools.
CONFLICT OF INTEREST
The authors declared no conflict of interest.
© 2013 American Society for Clinical Pharmacology and Therapeutics
1. Haller, C.A. & Benowitz, N.L. Adverse cardiovascular and central nervous
system events associated with dietary supplements containing ephedra
alkaloids. N. Engl. J. Med. 343, 1833–1838 (2000).
Andraws, R., Chawla, P. & Brown, D.L. Cardiovascular effects of ephedra
alkaloids: a comprehensive review. Prog. Cardiovasc. Dis. 47, 217–225
Maglione, M., Miotto, K., Iguchi, M., Hilton, L. & Shekelle, P. Psychiatric
symptoms associated with ephedra use. Expert Opin. Drug Saf. 4, 879–884
Haller, C.A., Meier, K.H. & Olson, K.R. Seizures reported in association with use
of dietary supplements. Clin. Toxicol. (Phila). 43, 23–30 (2005).
Federal Register. 21 CFR Part 119. Final rule declaring dietary supplements
containing ephedrine alkaloids adulterated because they present an
unreasonable risk; final rule. 11 February 2004.
Bouchard, N.C., Howland, M.A., Greller, H.A., Hoffman, R.S. & Nelson, L.S.
Ischemic stroke associated with use of an ephedra-free dietary supplement
containing synephrine. Mayo Clin. Proc. 80, 541–545 (2005).
Holmes, R.O. Jr & Tavee, J. Vasospasm and stroke attributable to ephedra-free
xenadrine: case report. Mil. Med. 173, 708–710 (2008).
Nykamp, D.L., Fackih, M.N. & Compton, A.L. Possible association of acute
lateral-wall myocardial infarction and bitter orange supplement. Ann.
Pharmacother. 38, 812–816 (2004).
Thomas, J.E., Munir, J.A., McIntyre, P.Z. & Ferguson, M.A. STEMI in a 24-year-old
man after use of a synephrine-containing dietary supplement. Tex. Heart J.
36, 586–590 (2009).
10. Ahmed, I. Malignant hypertension and acute aortic dissection associated
with caffeine-based ephedra-free dietary supplements: a case report. Cases J.
2, 6612 (2009).
11. Moaddeb, J., Tofade, T.S. & Bevins, M.B. Hypertensive urgency associated with
Xenadrine EFX use. J. Pharm. Pract. 24, 400–403 (2011).
12. Burke, J., Seda, G., Allen, D. & Knee, T.S. A case of severe exercise-induced
rhabdomyolysis associated with a weight-loss dietary supplement. Mil. Med.
172, 656–658 (2007).
WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?
3 To date, a very small number of clinical studies have
attempted to assess the cardiovascular effects of
ephedra-free supplementation. Because of design
issues, these earlier studies have provided inadequate
and sometimes misleading assessments of the safety
of ephedra-free supplements.
WHAT QUESTION DID THIS STUDY ADDRESS?
3 This study characterizes the cardiovascular effects
produced by multicomponent ephedra-free
supplements administered in multiple doses for
multiple days. To date, this is the first study of its kind.
WHAT THIS STUDY ADDS TO OUR KNOWLEDGE
3 With the ban on ephedra-containing supplements in
2004, the medical community has perceived ephedra-
free supplements as innocuous from the point of view
of safety. A growing number of serious case reports
suggests otherwise, and our findings support these
HOW THIS MIGHT CHANGE CLINICAL PHARMACOLOGY
3 The findings of this study will prompt the clini-
cal pharmacology community to refocus on safety
concerns regarding ephedra-free supplements and
dietary supplements in general.
ClInICAl PHARMAColoGy & THERAPEuTICS | VOLUME 93 NUMBER 3 | MARCH 2013
13. Retamero, C., Rivera, T. & Murphy, K. “Ephedra-free” diet pill-induced psychosis.
Psychosomatics 52, 579–582 (2011).
14. Gange, C.A., Madias, C., Felix-Getzik, E.M., Weintraub, A.R. & Estes, N.A. 3rd.
Variant angina associated with bitter orange in a dietary supplement. Mayo
Clin. Proc. 81, 545–548 (2006).
15. Karth, A., Holoshitz, N., Kavinsky, C.J., Trohman, R. & McBride, B.F. A case report
of atrial fibrillation potentially induced by hydroxycut: a multicomponent
dietary weight loss supplement devoid of sympathomimetic amines.
J. Pharm. Pract. 23, 245–249 (2010).
16. Nasir, J.M., Durning, S.J., Ferguson, M., Barold, H.S. & Haigney, M.C. Exercise-
induced syncope associated with QT prolongation and ephedra-free
Xenadrine. Mayo Clin. Proc. 79, 1059–1062 (2004).
17. Min, B. et al. Hemodynamic impact of an ephedra-free multicomponent
weight-loss supplement. Am. J. Health. Syst. Pharm. 62, 1582–1585 (2005).
18. Sale, C., Harris, R.C., Delves, S. & Corbett, J. Metabolic and physiologic effects of
ingesting extracts of bitter orange, green tea and guarana at rest and during
treadmill walking in overweight males. Int. J. Obes. 30, 764–773 (2006).
19. Seifert, J.G., Nelson, A., Devonish, J., Burke, E.R. & Stohs, S.J. Effect of acute
administration of an herbal preparation on blood pressure and heart rate in
humans. Int. J. Med. Sci. 8, 192–197 (2011).
20. Min, B. et al. Electrocardiographic effects of an Ephedra-Free, multicomponent
weight-loss supplement in healthy volunteers. Pharmacotherapy 25, 654–659
21. Bui, L.T., Nguyen, D.T. & Ambrose, P.J. Blood pressure and heart rate effects
following a single dose of bitter orange. Ann. Pharmacother. 40, 53–57 (2006).
22. Haller, C.A., Benowitz, N.L. & Jacob, P. 3rd. Hemodynamic effects of ephedra-
free weight-loss supplements in humans. Am. J. Med. 118, 998–1003 (2005).
23. Benowitz, N.L. Clinical pharmacology of caffeine. Annu. Rev. Med. 41, 277–288
24. Ammar, R., Song, J.C., Kluger, J. & White, C.M. Evaluation of
electrocardiographic and hemodynamic effects of caffeine with acute dosing
in healthy volunteers. Pharmacotherapy 21, 437–442 (2001).
25. Haller, C.A., Duan, M., Jacob, P. 3rd & Benowitz, N. Human pharmacology of
a performance-enhancing dietary supplement under resting and exercise
conditions. Br. J. Clin. Pharmacol. 65, 833–840 (2008).
26. Andrews, K.W. et al. The caffeine contents of dietary supplements commonly
purchased in the US: analysis of 53 products with caffeine-containing
ingredients. Anal. Bioanal. Chem. 389, 231–239 (2007).
27. Shi, J., Benowitz, N.L., Denaro, C.P. & Sheiner, L.B. Pharmacokinetic-
pharmacodynamic modeling of caffeine: tolerance to pressor effects.
Clin. Pharmacol. Ther. 53, 6–14 (1993).
28. Gurley, B.J., Gardner, S.F., Williams, D.K., & Grippo, A. Hemodynamic and
electrocardiographic effects of an “Ephedra-free” dietary supplement
(Metabolife Ultra™). Clin. Pharmacol. Ther. 79 (suppl. 2), P2 (2006).
29. Lovallo, W.R., Wilson, M.F., Vincent, A.S., Sung, B.H., McKey, B.S. & Whitsett, T.L.
Blood pressure response to caffeine shows incomplete tolerance after short-
term regular consumption. Hypertension 43, 760–765 (2004).
30. Benowitz, N.L., Jacob, P. 3rd, Mayan, H. & Denaro, C. Sympathomimetic effects of
paraxanthine and caffeine in humans. Clin. Pharmacol. Ther. 58, 684–691 (1995).
31. Denaro, C.P., Brown, C.R., Wilson, M., Jacob, P. 3rd & Benowitz, N.L. Dose-
dependency of caffeine metabolism with repeated dosing. Clin. Pharmacol.
Ther. 48, 277–285 (1990).
32. Dulloo, A.G., Seydoux, J., Girardier, L., Chantre, P. & Vandermander, J. Green tea
and thermogenesis: interactions between catechin polyphenols, caffeine and
sympathetic activity. Int. J. Obesity 24, 252–258 (2000).
33. Van Amelsvoort, J.M., Van Hof, K.H., Mathot, J.N., Mulder, T.P., Wiersma, A. &
Tijburg, L.B. Plasma concentrations of individual tea catechins after a single
oral dose in humans. Xenobiotica. 31, 891–901 (2001).
34. Katan, M.B. & Schouten, E. Caffeine and arrhythmia. Am. J. Clin. Nutr. 81,
35. Tajima, Y. Coffee-induced hypokalemia. Clin. Med. Insight: Case Reports 3, 9–13
36. US Pharmacopoeial Convention. Microbiological attributes of nonsterile
nutritional and dietary supplements. USP35-NF30 (U.S. Pharmacopoeia:
National Formulary). Rockville, MD (2012).
37. Raman, P., Patino, L.C. & Nair, M.G. Evaluation of metal and microbial
contamination in botanical supplements. J. Agric. Food Chem. 52, 7822–7827
38. Brown, J.C. & Jiang, X. Prevalence of antibiotic-resistant bacteria in herbal
products. J. Food Prot. 71, 1486–1490 (2008).
39. Stickel, F., Droz, S., Patsenker, E., Bögli-Stuber, K., Aebi, B. & Leib, S.L. Severe
hepatotoxicity following ingestion of Herbalife nutritional supplements
contaminated with Bacillus subtilis. J. Hepatol. 50, 111–117 (2009).
40. Heckman, M.A., Weil, J. & Gonzalez de Mejia, E. Caffeine (1, 3,
7-trimethylxanthine) in foods: a comprehensive review on consumption,
functionality, safety, and regulatory matters. J. Food Sci. 75, R77–R87 (2010).
41. Alsene, K., Deckert, J., Sand, P. & de Wit, H. Association between
A2a receptor gene polymorphisms and caffeine-induced anxiety.
Neuropsychopharmacology 28, 1694–1702 (2003).
42. Grosso, L.M. & Bracken, M.B. Caffeine metabolism, genetics, and perinatal
outcomes: a review of exposure assessment considerations during
pregnancy. Ann. Epidemiol. 15, 460–466 (2005).
43. Miller, R.J. et al. The impact of the catechol-O-methyltransferase genotype
on vascular function and blood pressure after acute green tea ingestion.
Mol. Nutr. Food Res. 56, 966–975 (2012).
44. Haller, C.A., Jacob, P. 3rd & Benowitz, N.L. Pharmacology of ephedra alkaloids
and caffeine after single-dose dietary supplement use. Clin. Pharmacol. Ther.
71, 421–432 (2002).
45. Collomp, K., Anselme, F., Audran, M., Gay, J.P., Chanal, J.L. & Prefaut, C. Effects of
moderate exercise on the pharmacokinetics of caffeine. Eur. J. Clin. Pharmacol.
40, 279–282 (1991).
46. Gurley, B.J. Pharmacokinetic herb-drug interactions (part 1): origins,
mechanisms, and the impact of botanical dietary supplements. Planta Med.
78, 1478–1489 (2012).
47. Gurley, B.J., Fifer, E.K. & Gardner, Z. Pharmacokinetic herb-drug interactions
(part 2): drug interactions involving popular botanical dietary supplements
and their clinical relevance. Planta Med. 78, 1490–1514 (2012).
48. Holland, D.T., Godfredsen, K.A., Page, T. & Connor, J.D. Simple high
performance liquid chromatography method for the simultaneous
determination of serum caffeine and paraxanthine following rapid sample
preparation. J. Chromatogr. B Biomed. Sci. Appl. 707, 105–110 (1998).
49. Tallent, S.M., Kotewicz, K.M., Strain, E.A. & Bennett, R.W. Efficient isolation
and identification of Bacillus cereus group. J. AOAC Int. 95, 446–451
50. Jin, W.Y. et al. Evaluation of VITEK 2, MicroScan, and Phoenix for identification
of clinical isolates and reference strains. Diagn. Microbiol. Infect. Dis. 70,
51. Hansen, B.M. & Hendriksen, N.B. Detection of enterotoxic Bacillus cereus
and Bacillus thuringiensis strains by PCR analysis. Appl. Environ. Microbiol.
188, 3382–3390 (2001).
274 VOLUME 93 NUMBER 3 | MARCH 2013 | www.nature.com/cpt