Executive
Overview
Background
The use
of herbal products in the United States has increased dramatically,
as evidenced by sales trends. Consumers annually spend approximately
$14 billion on dietary supplements. The dietary supplement
industry estimates that as many as 2 to 3 billion doses of
dietary supplements containing ephedrine alkaloids as part
of botanical ingredients are consumed each year in the United
States (GAO, 1999; AHPA, 2000) primarily for weight loss and/or
energy enhancement. Many botanical ingredients found in herbal
products are generally regarded as safe; however, when not
used properly, they can result in effects that are distinct
from expected physiological responses to the agent.
Ephedra
refers to a plant genus containing approximately 40 species
throughout regions of Europe, Asia, and America. Only a few
Ephedra species contain the alkaloid ephedrine, which
was first isolated in 1885. In most species used commercially,
the dominant alkaloid is ephedrine, which usually comprises
between 40 to 90% of total alkaloids in the plant, depending
on the species and other factors. Other related alkaloids
are also present, such as pseudoephedrine, N-methylephedrine,
N-methylpseudoephedrine, norpseudoephedrine and norephedrine
(phenylpropanolamine). These alkaloids have been collectively
termed ephedrine-type alkaloids, or simply ephedrine alkaloids.
Proportions and total levels can vary from one species to
another, time of year of harvest, weather conditions and altitude.
Ephedrine content generally is 4 to 5 times greater than pseudoephedrine,
but some sources of ephedra contain a 2:1 ratio of ephedrine
and pseudoephedrine.
In general,
all the ephedrine-type alkaloids (also referred to herein
as "ephedrine alkaloids") contained in Ephedra
species show significant differences between diastereomers
(e.g., ephedrine and pseudoephedrine) with regard to
pharmacokinetic and pharmacodynamic effects. All have effects
on the cardiovascular and respiratory system, but not to the
same degree. It is important to note that the pharmacokinetic
and toxicokinetic behavior of any isomer cannot be used with
precision to predict that of any other ephedrine alkaloid
isomers. In the literature, statements regarding ephedrine
alkaloids sometimes consider them to be synonymous, which
implies that the pharmacological activity of a particular
alkaloid is equipotent and that the toxicity of all diasteromers
is equivalent, which is not the case. In the assessment of
all the evidence relevant to the safety of ephedra, literature
on ephedrine is evaluated and the differences or similarities
are recognized.
The physiological
characteristics of Ephedra species are dependent upon
its chemical composition. Since ephedrine is the dominant
ephedrine alkaloid isomer of most Asian Ephedra species,
the characteristics of ephedrine would provide a good indicator
of the expected chemistry, pharmacology, and toxicology. Ephedra,
for the purposes of this report, will generally refer to the
complex mixtures that are extracts of the branchlets of Asian
Ephedra species known as ma huang, or products containing
these extracts. These ephedra extracts typically contain 6-8%
ephedrine alkaloids. As with any mixture, the characteristics
of only one, albeit major, component cannot account for all
the constituents of ephedra, but ephedrine represents
a significant portion of ephedra’s activity. Furthermore,
since the effects of pseudoephedrine are somewhat weaker with
respect to hypertensive effects and stimulation of the central
nervous system, an assessment based on the ephedrine as a
surrogate for total ephedrine alkaloid content provides a
conservative evaluation of risk. Recently released data from
the six-month, randomized, placebo-controlled clinical trial
on ephedra performed by Columbia and Harvard universities
(Boozer et al., 2000) made a major contribution to
the database. Thus, information on ephedra itself, in addition
to that on ephedrine, is relied upon in the risk assessment.
Objective
of Risk Assessment
The purpose
of the present report is to critically review the available
information related to the safety of ephedra/ephedrine alkaloids.
Nonclinical and clinical studies, published case reports,
and animal data, along with adverse event reports (AERs) from
the medical literature and the voluntary reporting system
called Special Nutritionals/Adverse Event Monitoring System
(SN/AEMS) under the direction of the Center for Food Safety
and Applied Nutrition (CFSAN), were evaluated. The objective
of this review was to establish a safe upper intake based
on the National Academy of Sciences Upper Limit Model for
nutrients. This Upper Limit (UL) is intended to provide a
safety standard for dietary supplements containing ephedrine
such that no significant or unreasonable risk of illness or
injury would arise at or below this intake level. No attempt
was made to review or comment on findings related to the potential
benefits of ephedra/ephedrine alkaloids, or any risk versus
benefit considerations.
The
UL Model
The method
used to establish a UL for ephedra/ephedrine alkaloids intake
is the Tolerable Upper Intake Level risk assessment model
(Food and Nutrition Board, 1998). The term Tolerable Upper
Intake Level is defined as the maximum level of total chronic
daily intake of a substance judged unlikely to pose a risk
of adverse health effects to the most sensitive members of
the healthy population. Although the model was developed for
application to nutrients, these food components are like all
chemical agents in that they can produce adverse health effects
if intakes are excessive. In the UL model, as in all other
risk assessment models, it is not possible to identify a single,
realistic "risk-free" intake level for a nutrient
that can be applied with certainty to all members of a population.
It is possible to develop intake levels that are unlikely
to pose risks of adverse health effects to most members of
the healthy population, including sensitive individuals, throughout
the life stage, except in some discrete subpopulations (for
example, those with genetic predispositions or certain disease
states) that may be especially vulnerable to one or more adverse
effects.
The UL
for ephedrine alkaloids in ephedra does not apply to specific
groups of persons. In particular, ephedrine and related agents
should not be taken by individuals with coronary thrombosis,
diabetes, glaucoma, heart disease, hypertension, thyroid disease,
impaired circulation of the cerebrum, pheochromocytoma (a
type of adrenal cancer that releases epinephrine), or enlarged
prostate. Patients with renal impairment may be at special
risk for toxicity. Persons taking ephedrine alkaloid drugs,
due to cumulative intake, should not consume ephedra-containing
dietary supplements, and ephedra is contraindicated for persons
taking monomine oxidase inhibitor drugs. Furthermore, ephedrine
is not intended for use in infants, children, adolescents
younger than 18 years, and pregnant or lactating women.
Data
Evaluation
The data
evaluation process for the UL method, as well as other common
risk assessment techniques, requires the selection of the
most appropriate or critical dataset(s) for deriving the UL.
In the data evaluation process, high quality human data are
generally preferable to animal data; however, in the absence
of appropriate human data, information from an animal species
whose biological responses are most like those of humans is
used. The available human data provide the most relevant kind
of information for hazard identification of ephedrine. Although
the typical focus of the majority of clinical studies was
efficacy, taken collectively with the recent clinical trial
on safety and benefit (Boozer et al., 2000) they are
of sufficient quality and extent to draw conclusions on the
safety of ephedra. Observational data in the form of case
reports were evaluated for their usefulness in developing
hypotheses/relationships between exposure and effect.
The analysis
of the clinical database involved a review of published case
reports and of clinical trials and investigations involving
normal healthy individuals, under special conditions (i.e.,
exercise) and special populations (e.g., obese, asthmatic).
Spontaneous adverse events captured and reported by FDA were
analyzed in Appendix A of the full report. Examination of
clinical trials involving the use of ephedrine was limited
to studies that investigated safety parameters. While valid
and pertinent human data were considered superior to data
derived from animals when assessing the potential risks to
humans from exposure to chemicals, any data deficiencies in
the human data must be explicitly considered.
In terms
of assessing the health effects of chemicals in humans, controlled,
prospective clinical investigations provide the most reliable
source of information. For example, studies of this type are
used to provide information related to the efficacy and safety
of new pharmaceuticals (after appropriate animal testing has
defined potential risks and supported the investigational
dosing of humans). For most chemicals to which humans are
exposed, prospective human studies are unavailable, and as
a result, relevant information related to their health effects
must be obtained retrospectively through the use of epidemiological
methods, using standard principles, in an attempt to establish
causation and dose-response relationships. Another source
of information on the adverse effects of agents in humans
is case reports. These are typically based upon observations
in individuals or small groups and they serve the critical
function of alerting the medical/scientific community to possible
adverse events. Individual case reports generally cannot be
relied upon to establish a cause-effect relationship, but
confidence in the findings of individual reports increases
when there is consistency in the observations published by
different authors. As with all scientific investigations,
case-reports must be carefully reviewed for limitations in
methodology and the findings interpreted in the light of the
weight-of-evidence. Finally, with respect to the FDA AER database
for ephedra, reviewed in Appendix A, the reliability of the
reported information was a major concern, (e.g., missing
information, elapsed time before reporting). Only 10% of the
reported AERs contained a minimally sufficient quantity of
information, and no conclusive determination of unexpected
effects or causality was possible.
Toxicology
Studies
The nonclinical
toxicology of ephedrine and ephedra was reviewed to assess
its consistency with data obtained from clinical studies.
The studies evaluated addressed the acute, subchronic and
chronic safety, carcinogenicity, reproductive toxicity, and
mutagenicity of ephedrine. Where available, data related to
ephedra are emphasized; however, the database on ephedrine
makes a significant contribution to the total evidence relevant
to ephedra, and thus was considered in the risk assessment.
It is
interesting to note that one study in the nonclinical literature
compared the acute toxicity of ephedrine to that of botanical
ephedra extract. Although only one study was located in the
literature which compared the effects of ephedra versus ephedrine,
these results support the conservative assumption that ephedrine
can be used in a safety assessment as a surrogate for ephedra,
since the potency of ephedrine overestimates the potential
potency of ephedra itself.
The National
Toxicology Program (NTP) studies, given the quality of the
investigations, were used to support for the derivation of
a UL based on the clinical data. Rat carcinogenicity data
(103-week duration) were used, since mice were less sensitive
to the effects of ephedrine. Thus, the use of the rat species
conservatively estimates a lifetime No-Observed-Adverse-Effect
Level (NOAEL) value. A NOAEL value was obtained in male rats,
at an average daily consumption of approximately 9 mg/kg
body weight/day. A dose of 9 mg/kg body weight/day from the
male rat data extrapolated to a 60 kg (132 lb) person
would be 540 mg/day.
Clinical
Studies
Nine studies
in normal healthy individuals investigated the effects of
ephedrine intake (Bye et al., 1974; Drew et al.,
1978; Kuitunen et al., 1984; Astrup et al.,
1991; Astrup and Toubro, 1993; Liu et al., 1995; White
et al., 1997; Gurley et al., 1998a; Shannon
et al., 1999). Ephedrine exposures involved oral administration
over a short-duration such as 24-hours. The range of total
doses within these 9 studies was from 10 to 150 mg/day, given
at a frequency of 1 to 3 times/day to achieve daily maximum
specified. The foremost weakness in this healthy population
dataset was its limited duration (<24 h) that reduces the
utility of these data in the assessment of the UL. Nevertheless,
these data are used to support the database in obese but healthy
subjects.
Five studies
in healthy normal individuals investigated the effects of
exercise/physical parameters together with ephedrine use (Sidney
and Lefcoe, 1977; Strömberg et al., 1992; Vanakoski
et al., 1993; Bell et al., 1998; Bell and Jacobs,
1999). The range of total doses within these 5 studies was
from 24 to 81 mg/day together with exercise, or some physical
parameters, over a short duration of exposure (typically 24
hours).
Twenty
studies in obese, but otherwise reportedly healthy individuals
(19 in adults and 1 in children), investigated the effects
of ephedrine intake (Astrup et al., 1985, 1992; Pasquali
et al., 1985, 1987, 1992; Krieger et al., 1990;
Daly et al., 1993; Molnár, 1993; Toubro et
al., 1993a,b; Breum et al., 1994; Buemann et
al., 1994; Kaats and Adelman, 1994; Moheb et al.,
1998; Waluga et al., 1998; Nasser et al., 1999;
Huber, 1999,2000; Boozer et al., 2000; Molnár
et al., 2000 ). Ephedrine exposures involved oral administration
over durations from 10 days to 26 months. The range of total
doses within these studies was from 50 to 150 mg/day, given
at frequencies of 1 to 3 times/day to achieve the daily maximum
specified. Based on the strengths of the study design, duration
of study, number of subjects enrolled and endpoints evaluated,
studies conducted in obese individuals were determined to
be of sufficient quality and extent for inclusion as the critical
dataset for the determination of a UL. In particular, given
the quality of design and protocol, the Columbia/Harvard clinical
trial conducted by Boozer et al. (2000), was used to
derive the UL which is supported by several other studies
from the clinical literature.
The clinical
database that has been considered involved administration
of ephedrine singly, ephedrine together with other components
such as caffeine and/or acetylsalicylic acid (ASA) or ephedra
with caffeine. Since many dietary supplements which contain
ephedrine often contain other ingredients, including these
data in the safety assessment of ephedrine is relevant, especially
given that these other ingredients are major components of
ephedra preparations. The pharmacology of individual ephedrine-type
alkaloids has been well characterized, but the effects of
combinations of these other compounds are less well known.
In addition, interactions between ephedrine-type alkaloids
and xanthine alkaloids (e.g., caffeine), as well as
biologically active compounds in other plant species that
are constituents of many dietary supplements, have yet to
be fully characterized. This risk assessment assumes that
a combination product (i.e., ephedra together with
a caffeine-containing ingredient) would be no more or less
active than an equivalent dose of ephedrine singly. Since
combination products were given in many of the clinical studies,
this report evaluated the contribution/interaction of other
ingredients typically contained in ephedrine preparations,
insofar as they contribute to the analysis of ephedra itself.
Determination
of Upper Limit (UL)
Following
the assessment of the most appropriate or critical dataset(s),
a NOAEL dose or intake level for humans is identified. In
the absence of a NOAEL determination, a Lowest Observed Adverse
Effect Level (LOAEL) is chosen. In principle, the primary
aim of safety studies is to recognize the potential hazards
associated with a particular chemical and identify a NOAEL
or LOAEL from the dose-response data. Monitoring data for
adverse effects following ephedra intake in obese individuals
were used to identify the NOAEL of 90 mg/day, with the ephedra
clinical trial by Boozer et al. (2000) identified as
the critical study.
Following
characterization of the NOAEL, safety factors or uncertainty
factors are typically applied. Judgments are made regarding
uncertainties associated with extrapolating from the observed
data to the healthy population. The UF is typically applied
to a NOAEL or LOAEL to derive the UL, which generally represents
a lower estimate of the threshold above which the risk of
adverse effects may increase. The application of safety factors
or uncertainty factors has been used for over 30 years in
the determination of a safe level of exposure to chemicals
based on the studies in experimental animals and humans (Renwick,
1995). The UFs allocated are dependent on the nature and extent
of the toxicity database.
A UF of
1 was judged appropriate, based on considerations of pharmacokinetics
of ephedrine, use patterns, duration of expected use, and
animal studies and the strong scientific findings reported
by Boozer et al. (2000) and supported by the clinical
findings from Pasquali et al. (1985), Krieger et
al. (1990), Astrup et al. (1992), Quaade et
al., (1992), Daly et al. (1993), Toubro et al.
(1993a,b), Nasser et al. (1999) and Molnár et
al., 2000. Given the quality of the long-term investigation
of ephedrine alkaloids in an herbal ephedra supplement by
Boozer et al. (2000), this study represents a pivotal
clinical study in the safety evaluation of ephedra.
Application
of the UF of 1 to the NOAEL of 90 derived a UL of 90 mg of
ephedrine alkaloids in ephedra per day for a generally healthy
population. This daily level of intake is unlikely to pose
a risk of adverse health effects. Label instructions together
with considerations of pharmacokinetics of ephedrine, use
patterns, duration of expected use, and supportive animal
studies further support a UF of 1. Label instructions would
include statements that (1) consumers should check with their
healthcare provider about taking the product; (2) use is contraindicated
for certain people; (3) direct the consumer to split the daily
dose into at least three parts, so that no dose exceeds 30
mg; (4) the product is intended for use of not more than 6
months; (5) persons younger than 18 years should not use the
product; (6) pregnant and lactating women should not use the
product; and (7) provide information to facilitate post-market
monitoring.
Conclusion
For healthy
adults:
NOAEL
= 90 mg/day, and UF =1
UL = 90
mg/day total ephedrine alkaloids from ephedra
