Content uploaded by Harriet de Wit
Author content
All content in this area was uploaded by Harriet de Wit on Feb 23, 2018
Content may be subject to copyright.
Psychopharmacology (1999) 145 : 67–75 © Springer-Verlag 1999
ORIGINAL INVESTIGATION
Angela J.H. Justice · Harriet de Wit
Acute effects of
d
-amphetamine during the follicular and luteal phases
of the menstrual cycle in women
Received: 26 August 1998/Final version: 19 February 1999
Abstract Rationale: Little is known about the interac-
tions between ovarian hormones across the menstrual
cycle and responses to psychoactive drugs in humans.
Preclinical studies suggest that ovarian hormones such
as estrogen and progesterone have direct and indirect
central nervous system actions, and that these hor-
mones can influence behavioral responses to psy-
choactive drugs. Objectives: In the present study, we
assessed the subjective and behavioral effects of
d-amphetamine (AMPH; 15 mg orally) at two hor-
monally distinct phases of the menstrual cycle in
women. Methods: Sixteen healthy women received
AMPH or placebo capsules during the follicular and
mid-luteal phases of their cycle. During the follicular
phase, estrogen levels are low initially and then rise
while progesterone levels remain low. During the mid-
luteal phase, levels of both estrogen and progesterone
are relatively high. Dependent measures included self-
report questionnaires, physiological measures and
plasma hormone levels. Results: Although there were
no baseline differences in mood during the follicular or
luteal phase, the effects of AMPH were greater during
the follicular phase than the luteal phase. During
the follicular phase, subjects reported feeling more
“High”, “Energetic and Intellectually Efficient”, and
“Euphoric” after AMPH than during the luteal phase,
and also reported liking and wanting AMPH more.
Further analyses showed that during the follicular
phase, but not the luteal phase, responses to AMPH
were related to levels of estrogen. Higher levels of estro-
gen were associated with greater AMPH-induced
increases in “Euphoria” and “Energy and Intellectual
Efficiency”. During the luteal phase, in the presence of
both estrogen and progesterone, estrogen levels were
not related to the effects of AMPH. Conclusions: These
findings suggest that estrogen may enhance the sub-
jective responses to a stimulant drug in women, but
that this effect may be masked in the presence of prog-
esterone.
Key words Menstrual cycle · Estrogen · Progesterone ·
d-Amphetamine · Subjective effect
Introduction
Stimulant drugs such as amphetamine and cocaine are
used by a growing percentage of women. Estimates
from the National Household Survey indicate that, in
1996, almost half of all cocaine users were women (US
DHHS 1996). This represents a considerable rise from
1993, when only one-third of all cocaine users were
women (US DHHS 1993). However, to date, few stud-
ies have investigated variables that influence responses
to stimulant drugs in women. One class of variables
that may influence how women respond to these drugs
is the variations in levels of ovarian hormones that
occur across the menstrual cycle. Recently, preclinical
data has shown that ovarian hormones, such as estro-
gens and progesterone, interact directly with specific
neurotransmitter systems in the central nervous sys-
tem, including neurotransmitter systems that also
mediate the effects of stimulant drugs (e.g., McEwen
and Parsons 1982). For example, both ovarian hor-
mones and stimulant drugs have direct or indirect
actions on dopaminergic activity in the brain. Notably,
the reinforcing effects of stimulant drugs, in particular,
are believed to be mediated by dopamine (e.g., Randrup
and Munkvad 1966; Banerjee and Lin 1973; Wise 1978;
Koob and Bloom 1988; Seiden et al. 1993).
In general, estrogen increases dopamine (DA) activ-
ity and enhances the neurochemical and behavioral
responses to stimulant drugs. Estrogen receptors are
A.J.H. Justice · H. de Wit (*)
Department of Psychiatry MC3077, The University of Chicago,
5841 S. Maryland Ave, Chicago, IL 60637, USA
e-mail: hdew@midway.uchicago.edu
Fax: +1-773-702-6454
co-localized on some dopaminergic neurons in the dien-
cephelon (McEwens and Parsons 1982) and several
studies have shown that dopaminergic activity
fluctuates across the estrous cycle in rats. DA activity
increases when estrogen levels are high, and decreases
when estrogen levels are low (Dluzen and Ramirez
1985; Xiao and Becker 1994). DA activity also increases
in response to exogenous estradiol in ovariectomized
rats. For example, injections of the estrogen estradiol
increase DA synthesis (Pasqualini et al. 1995, 1996),
turnover and release (e.g., Becker and Ramirez 1981;
Di Paolo et al. 1985; Becker and Beer 1986), and
increase DA receptor density (Hruska and Silbergeld
1980; Rance et al. 1981). Estrogen also decreases
monoamine oxidase activity, thereby regulating the
degradation of dopamine (Luine et al. 1975).
Consistent with these neurochemical findings, estrogen
enhances the behavioral and neurochemical responses
to dopamine agonists such as d-amphetamine in rats.
For example, AMPH-stimulated DA release is greatest
during phases of the estrous cycle when estrogen lev-
els are high (Becker and Ramirez 1981; Becker and Cha
1989) and these neurochemical responses correspond
to enhanced behavioral responses to AMPH, such as
locomotor activity, rotational behavior, and stereotypy
(Becker et al. 1982; Joyce and Van Hartesveldt 1984;
Becker and Cha 1989).
Progesterone, on the other hand, appears to decrease
the activity of the DA system (Fernandez-Ruiz et al.
1990; Shimizu and Bray 1993) and the behavioral
responses to stimulant drugs (Michanek and Meyerson
1982; Dluzen and Ramirez 1987). Across the estrous
cycle, DA activity is greatest during estrus, which is
characterized not only by high levels of estrogen, but
also by low levels of progesterone (Dluzen and Ramirez
1985; Xiao and Becker 1994). DA activity also
decreases in response to exogenous progesterone.
However, the nature of these effects appear to be crit-
ically dependent on the timing and dose characteris-
tics of the progesterone administration. For example,
continuous 40-min infusions of 2 ng / ml progesterone
inhibit AMPH-stimulated DA release up to 12 h post-
progesterone. Conversely, pulsatile infusions of the
same dose of progesterone (2 ng/ml) may increase
AMPH-stimulated DA release and pulsatile infusions
of either a higher (50 ng/ml) or lower (0.2 ng/ ml) dose
of progesterone have no effect (Dluzen and Ramirez
1984). It has been suggested that the effects of proges-
terone on DA release are biphasic, with progesterone
initially increasing but ultimately decreasing basal and
AMPH-stimulated DA release (Dluzen and Ramirez
1984). When progesterone is administered to estrogen-
primed rats, some investigators have reported that it
has no effect (Bazzett and Becker 1994), whereas oth-
ers have reported that it decreases AMPH-induced
stereotypy (Michanek and Meyerson 1982).
These preclinical findings suggest that ovarian hor-
mones may influence the subjective or mood-altering
responses to AMPH in women. If estrogen enhances,
and progesterone reduces responses to AMPH in
women as it does in rats, then the subjective and behav-
ioral effects of AMPH may vary depending on the rel-
ative levels of estrogen and progesterone at a particular
menstrual cycle phase. Based on the preclinical
findings, one would expect the effects of AMPH to be
greater when estrogen levels are high relative to prog-
esterone levels. The present study was designed to char-
acterize the subjective and behavioral effects of AMPH
during the follicular phase and the luteal phase of the
menstrual cycle in women. The follicular phase is char-
acterized by low (early) or moderate (later) levels of
estrogen and very low levels of progesterone, whereas
the luteal phase is characterized by moderate levels of
estrogen with high levels of progesterone. Thus, based
on the relative levels of estrogen and progesterone, it
was postulated that the effects of AMPH would be
greater during the follicular phase than during the
luteal phase. It was further postulated that within the
follicular phase, responses to AMPH would be greater
in subjects whose estrogen levels were higher. We tested
these hypotheses by administering AMPH (15 mg) to
normal healthy women during the follicular and mid-
luteal phases of the menstrual cycle. Plasma estradiol
and progesterone levels were determined prior to drug
administration, and subjective, behavioral and physio-
logical data were obtained before drug administration
and at regular intervals thereafter.
Materials and methods
Design
Subjects participated in a within-subject study consisting of four
laboratory sessions, conducted across two consecutive menstrual
cycles. Two sessions were scheduled to occur during the follicular
and two during the luteal phase of each subject’s cycle. On two ses-
sions subjects received d-amphetamine (AMPH; 15 mg) and on the
other two sessions they received placebo (PL), in a quasi-random
order so that they received drug and PL once at each phase. Drugs
were administered under double-blind conditions. This dose of
AMPH was chosen because it is known to produce reliable, but
modest, subjective effects thereby allowing us to detect phase-depen-
dent increases or decreases in the magnitude of response. Subjective,
behavioral and physiological responses were assessed at regular
intervals during the session. These responses were analyzed in rela-
tion to menstrual cycle phase (follicular or luteal) and in relation
to plasma hormone levels obtained at the beginning of each
session.
Subject recruitment and screening
Sixteen women aged 18–35 years were recruited from the univer-
sity and surrounding community via posters, advertisements in
newspapers and by word of mouth referrals. Initial eligibility was
ascertained in a telephone interview. Eligible candidates reported
to the laboratory to complete standardized self-report question-
naires including the Symptom Checklist-90 (Derogatis 1983) and a
health questionnaire containing items of general health and drug
68
and alcohol use. Screening included a physical examination, an elec-
trocardiogram, a semi-structured psychiatric interview, and urine
pregnancy tests. Exclusion criteria were: irregular menstrual cycles,
menstrual cycles shorter than 25 days or longer than 35 days, amen-
orrhea, severe premenstrual syndrome diagnosed according to DSM
IV criteria (APA 1994) or any menstrual cycle dysfunction, use of
hormonal contraceptives, lactation, pregnancy or plans for preg-
nancy and endocrine, medical or axis I (APA 1994) psychiatric or
substance use disorders. The procedure was approved by the
Institutional Review Board at the University of Chicago Hospital.
During the initial screening interview and again during orienta-
tion before the study, subjects read the consent form and any ques-
tions about it were answered. The consent form outlined the
procedures to be followed and listed the classes and possible effects
of any drugs that subjects might receive. For blinding purposes,
subjects were told that on any session they might receive a stimu-
lant, tranquilizer, placebo or alcohol. Breath alcohol levels (BAL)
were determined prior to each session using an Intoximeter
Breathalyzer. No BAL reading was positive. Urine samples were
obtained prior to each session and screened for pregnancy. In addi-
tion, one of the urine samples was randomly selected for a urine
toxicology screen to verify the non-use of stimulants, barbiturates,
opioids and benzodiazepines. No urine toxicology screen was pos-
itive. Subjects agreed not to take any other drugs, other than their
usual amounts of caffeine and/or nicotine, for 12 h before and
6 h following each session. Subjects were not allowed to consume
caffeine or nicotine during the sessions.
Laboratory environment
This study was conducted in the recreational laboratory environ-
ment in the Human Behavioral Pharmacology Laboratory (HBPL)
in the Department of Psychiatry. The recreational environment con-
sists of three rooms each furnished to resemble a living room. The
rooms have incandescent lighting, couches and upholstered chairs,
casual tables with magazines and board games, posters on the walls,
televisions and VCRs with a choice of movies. Subjects were tested
individually and were allowed to bring in their own recreational
materials.
Procedure
Subjects participated in a total of four, 4.5-h sessions over the course
of two menstrual cycles in the HBPL. The sessions were scheduled
such that the subject came in twice per cycle: once during the fol-
licular phase, and once during the luteal phase. When a session was
missed due to illness or scheduling problems, subjects made up the
appropriate session in their next cycle. Half of the subjects started
during the follicular phase and the other half started during the
luteal phase, and half received PL first while the other half received
AMPH first. Subjects were trained to use the OvuQuick kit (Quidel
Corp.) to detect urinary levels of luteinizing hormone (LH). The
initial rise in urine LH levels is one of the best predictors of the
time of ovulation (Stern and McClintock 1996).
Follicular phase sessions
Subjects telephoned the researcher on day 1 of menstruation and
scheduled their follicular phase sessions 2–10 days from day 1 of
menstruation. This range of times in the follicular phase provided
a range of estrogen levels, across subjects, allowing us to address
the second hypothesis. Early in the follicular phase the levels of
estrogen are low (i.e., around 50 pg / ml; Griffin and Ojeda 1996),
whereas later in this phase estrogen levels begin to rise (i.e., up to
200 pg /ml; Griffin and Ojeda 1996). Throughout the follicular phase
progesterone levels are consistently very low.
Luteal phase sessions
Subjects measured their urinary LH every day at 6 p.m. beginning
9–15 days after onset of menses, depending on the length of their
cycle. Luteal phase sessions were scheduled 6–10 days after the onset
of the luteinizing hormone (LH) surge as detected in urine. At this
time, levels of estradiol are at a moderate level (140 pg/ ml; Griffin
and Ojeda 1996), whereas progesterone levels are very high (between
7.5 and 14.0 ng/ ml; Griffin and Ojeda 1996).
Session protocol
On each session, subjects reported to the Clinical Research Center
(CRC) at 7:30 a.m., after fasting overnight. Upon arrival, they pro-
vided a blood sample for estradiol and progesterone assays, and
urine samples for pregnancy and toxicology tests. Blood samples
were centrifuged and the serum was frozen at 970°C until the hor-
mone assays were conducted. Immediately after the blood and urine
samples were obtained and a negative pregnancy test was confirmed,
subjects reported to the HBPL where they stayed for the remain-
der of the session. There, subjects completed baseline questionnaires
to assess their mood (see below) and their heart rate, blood pres-
sure and temperature were recorded. At 8:00 a.m. they ingested
two capsules containing AMPH (total = 15 mg) or PL with 100 ml
water. AMPH and PL capsules were opaque, colored gelatin cap-
sules (size 00) and were identical in appearance. AMPH capsules
were packed with dextrose filler; placebo capsules contained only
dextrose. Subjects received drug and placebo each during the
follicular and luteal phases of their menstrual cycles. Drug admin-
istrations were double-blind. After taking the capsules, subjects
repeated the mood and performance tests and vital signs were
recorded at 0.5, 1, 1.5, 2, 2.5, 3, 3.5, and 4 h after ingesting the cap-
sules. Subjects left the laboratory shortly after 12 p.m.
Dependent measures
The primary measures were the measures of subjective state as mea-
sured by an experimental version of the Profile of Mood States
(POMS; McNair et al. 1971), the Addiction Research Center
Inventory (ARCI; Martin et al. 1971), and two visual analog ques-
tionnaires, the VAS and the DEQ. The POMS consists of 72 adjec-
tives commonly used to describe momentary mood state. Subjects
indicate how they feel at that moment in relation to each of the
adjectives on a 5-point scale ranging from “not at all” (0) to
“extremely” (4). The 49-item ARCI is a true-false questionnaire
with five empirically derived scales: A (amphetamine-like, stimu-
lant effects), BG (benzedrine group, energy and intellectual
efficiency), MBG (morphine-benzedrine group, euphoric effects),
LSD (lysergic acid diethylamide, dysphoric effects, somatic com-
plaints), and PCAG (pentobarbital-chlorpromazine-alcohol group,
sedative effects) (Haertzen 1974). The VAS assessed subjective states
including “Stimulated”, “Hungry”, “Anxious”, “Sedated”, “High”,
and “Down”. The DEQ assessed subjective states such as “Feel
drug”, “Like drug”, “Feel high”, and “Want more drug”. These
measures have been shown to be sensitive to the effects of a vari-
ety of psychoactive drugs, including stimulants (Fischman and
Foltin 1991).
Plasma estradiol and progesterone levels were determined in
duplicate at the University of Chicago Endocrinology Laboratory
from plasma samples obtained before capsule ingestion on amphet-
amine and placebo sessions. Estradiol (pg/ml) was measured by
radioimmunoassay with the Pantex extraction based estradiol kit
(Pantex, Santa Monica, Calif., USA). The radioimmunoassay
method was performed according to the manufacturer’s directions
modified by a staggered incubation where sample and first antibody
were incubated for 1 h at 23°C, followed by addition of the
125
I-
estradiol analogue and incubation for 20 min at 37°C. This assay
69
has an average sensitivity of 2 pg/ ml, an 11 % interassay coefficient
of variation and a 5.1% intraassay coefficient of variation, for con-
trol at 36 pg / ml. Plasma progesterone levels were measured by a
previously described radioimmunoassay procedure (Rosenfield
et al. 1994) after a single thin layer chromatographic purification
step and correction for procedural losses. This procedure has a sen-
sitivity of 0.25 ng / ml, an interassay coefficient of variation of less
than 10% and a 7% intra-assay coefficient of variation.
Data analysis
The analyses addressed two main questions: i) were the responses
to AMPH versus placebo different during the follicular and luteal
phases of the menstrual cycle, and ii) was there a relationship
between hormone levels and magnitude of response to AMPH?
Prior to addressing the first question, repeated measures univariate
ANOVAs were conducted on each dependent measure at baseline
with factors Phase (follicular and luteal) and Drug (AMPH or
placebo) to ensure there were no differences in mood across the two
phases. Then, to address the first question, repeated measures uni-
variate ANOVAs were conducted on each dependent measure with
factors Phase (follicular or luteal), Drug (AMPH or placebo), and
Time (baseline through 4-h post-capsule administration). For all
analyses, Fvalues were considered significant at P< 0.05. Fisher-
Hayter post hoc comparisons were conducted when significant
Phase by Drug by Time interactions were observed. To address the
second question, peak change scores (from pre-capsule to the largest
positive or negative value post-capsule) on POMS, VAS, and ARCI
were calculated. For the DEQ, which was not administered before
capsule administration due to the nature of the questions, only peak
values were used. The plasma estradiol and progesterone levels that
were measured prior to capsule administration were correlated with
the peak change score or peak score for each dependent measure.
Correlations between subjective responses and estradiol level were
conducted within the follicular phase for AMPH and placebo ses-
sions, separately. Correlations between subjective responses and
progesterone levels were not conducted on follicular phase sessions
because progesterone levels were near zero. For luteal phase ses-
sions, linear regression analyses were performed with estradiol and
progesterone as predictors. In addition, correlations were also con-
ducted between the ratio of estrogen to progesterone and response
to AMPH and placebo for luteal phases sessions. All analyses were
conducted with SPSS for Macintosh v. 6.1.1. It should be noted
that because of the number of correlations calculated, we risked
finding significant correlations by chance. However, this was an
exploratory and descriptive study, and we were interested in iden-
tifying any trends on dependent measures that were related to hor-
mone levels. Therefore, while recognizing that these do not meet
the standards of statistical significance, we will discuss any corre-
lations with a significance level of P< 0.06.
Results
Subject characteristics
Sixteen women completed the study. Their mean age
was 25 years (SD 5.1), mean height 166.6 cm (SD 6.4),
mean weight 61.3 kg (SD 7.0). Twelve were Caucasian,
three were African American and one was Asian. They
reported consuming a mean of two alcoholic drinks and
6.7 caffeinated beverages each week. Eight subjects
reported some cigarette use, smoking a mean of 2.4 cig-
arettes a day (SD 3.5). No subjects smoked more than
ten cigarettes a day. They reported minimal use of other
recreational drugs. Subjects had regular menstrual
cycles with an average length of 29.6 days (range 27–33
days) and an intercycle variability of 2 or 3 days at most.
Hormone levels
Table 1 shows the mean, SD, minimum and maximum
levels of estradiol and progesterone during the AMPH
and placebo sessions at the follicular and luteal phases
of the cycle. Both estradiol and progesterone levels were
significantly greater during the luteal phase than dur-
ing the follicular phase [F(15.1) = 32.68, P< 0.001;
F(13,1) = 127.38, P< 0.001, respectively]. Figure 1
shows the estradiol and progesterone levels for the 16
individual subjects during the follicular phase sessions
and the luteal phase sessions expressed as a function
of days from the LH surge. Expected values are also
shown (Griffin and Ojeda 1996). All plasma samples
were taken prior to drug administration. As expected,
during the follicular phase, estrogen levels were low ini-
tially, but rose towards the end, whereas progesterone
levels were low throughout. During the luteal phase,
both estrogen and progesterone levels were high.
Effects of menstrual cycle phase
There were no differences on any dependent measure
at baseline. Furthermore, only one mood variable,
“Down” (VAS), showed an overall difference between
70
Follicular Luteal
Pre-placebo Pre-amphetamine Pre-placebo Pre-amphetamine
E2 Prog E2 Prog E2 Prog E2 Prog
(pg/ml) (ng/ml) (pg/ml) (ng/ml) (pg/ml) (ng/ml) (pg/ml) (ng/ml)
n=16 n=16 n=16 n=16 n=16 n=16 n=16 n=15
Mean 61.69 0.21 67.63 0.21 177.88 10.50 152.50 11.50
SD 12.21 0.01 12.05 0.01 12.15 0.71 16.23 2.90
Min 27.00 0.10 28.00 0.11 90.00 5.74 73.00 6.37
Max 228.00 0.27 191.00 0.27 244.00 15.72 290.00 15.30
Table 1 Plasma levels of
estradiol (E2) and
progesterone (Prog)
determined from pre-capsule
plasma samples on placebo
and AMPH sessions. Lowest
values for progesterone
represent lower limits of
detection
the follicular and luteal phases. Regardless of drug,
subjects rated themselves as slightly less “Down” (VAS)
during the follicular phase than the luteal phase
[F(1,15) = 4.61, P< 0.05]. However, other measures
that generally correlate with self-reported ratings
of “Down”, such as “Depression” (POMS) and
Sedation (ARCI PCAG scale and VAS ratings of
“Sedation”) did not show any menstrual cycle phase
effects. Regardless of drug, diastolic blood pressure was
slightly higher during the follicular phase than the
luteal phase [F(1,15) = 4.96, P< 0.05]. No other
differences in physiological measures were observed
between the follicular and luteal phases. In addition,
no relationships were observed between plasma hor-
mone levels and mood on placebo sessions.
AMPH effects
Regardless of phase, AMPH increased systolic blood
pressure [F(8,120) = 4.80, P< 0.001] and produced its
71
Fig. 1 Plasma levels of estradiol ( filled circles) and progesterone
(open squares), plotted as a function of the number of days before
(follicular phase) or after (luteal phase) the LH surge. Each sub-
ject (n= 16) was tested twice during each phase, so the total num-
ber of observations for each hormone is 32. Shaded areas show
expected values of estradiol (light area) and progesterone (dark area;
Griffen and Ojeda 1996)
Table 2 Summary of significant results
Drug Phase Time Drug*Time Drug*Phase Drug*Phase*Time
DEQ F(1,15)= F(1,15)= F(8,120)= F(8,120)= F(1,15)= F(8,120)=
Feel Drug 5.15* 5.68*** 2.26*
Like Drug 6.85*
Want More 2.50*
VAS F(1,15)= F(1,15)= F(8,120)= F(8,120) = F(1,15)= F(8,120) =
Feel High 14.82** 4.38* 4.53*** 4.69*** 4.35*
Stimulated 14.98** 2.65** 3.66***
Down 4.61*
Hunger 2.39*
ARCI F(1,14)= F(1,14)= F(8,112)= F(8,112) = F(1,14)= F(8,112) =
BG (Energy) 5.35* 2.83* 5.36*
MBG (Euphoria) 6.69** 2.11* 2.03*
A (AMPH-like) 20.92*** 5.12***
POMS F(1,15)= F(1,15)= F(8,120)= F(8,120) = F(1,15)= F(8,120) =
Arousal 11.55** 5.49*** 3.91***
Vigor 14.80** 6.11***
Elation 11.36** 3.86***
Positive Mood 7.52 3.29**
Friendliness 5.12*
Fatigue 11.36*** 2.41*
Physiological F(1,15)= F(1,15)= F(8,120)= F(8,120)= F(1,15) = F(8,120) =
Systolic BP 9.03** 4.80***
Diastolic BP 4.96*
Heart Rate 4.28, P< 0.06 3.34** 3.94***
*P< 0.05; **P< 0.01; ***P< 0.001
prototypic subjective effects such as increased ratings
of “Feel Drug” [DEQ; F(1,15) = 5.15, P< 0.05],
Arousal, Vigor, Elation, and Friendliness and
decreased ratings of Fatigue (POMS). These results are
summarized in Table 2 and Fig. 2.
Interactions between menstrual cycle phase
and AMPH
Several significant interactions were observed between
menstrual cycle phase and AMPH. Effects of AMPH
were greater, relative to placebo, during the follicular
than luteal phase on ratings of Like drug [DEQ; F(1,15)
= 6.85, P< 0.02], High [VAS; F(1,15) = 4.35, P< 0.05],
Want more drug [DEQ; F(8,120) = 2.5, P< 0.03], and
measures of Energy and Intellectual Efficiency [ARCI
BG scale; F(1,14) = 5.36, P< 0.04], and Euphoria
[ARCI MBG scale; F(8,120) = 2.03, P< 0.05]. These
results are summarized in Table 2 and Fig 3. Figure 3
shows subjective ratings of Energy and Intellectual
Efficiency, Euphoria and “High”, after AMPH (and
PL) during the follicular and luteal phases. Relative to
placebo, AMPH increased ratings of Energy and
Intellectual Efficiency (ARCI BG scale) and Euphoria
(ARCI MBG), respectively, during the follicular phase
but not during the luteal phase. During the follicular
phase, Energy and Intellectual Efficiency peaked at
120 min and Euphoria peaked at 90 min. Ratings of
“High” peaked 120 min after AMPH administration
during both phases, but were approximately twice as
72
Fig. 2 Mean (SEM) for subjective ratings of “Feel drug” (DEQ;
top panel) and systolic blood pressure (bottom panel) during the
follicular (circles; left) and luteal (squares; right) phases. Filled sym-
bols represent data from AMPH sessions. Open symbols represent
data from placebo sessions. The maximum score on the DEQ is
100
Fig. 3 Mean (SEM) subjective ratings of Energy and Intellectual
Efficiency (ARCI BG; top panel), Euphoria (ARCI MBG; middle
panel), and “High” (VAS; bottom panel) during the follicular (cir-
cles; left) and luteal (squares; right) phases. Filled symbols represent
data from AMPH sessions. Open symbols represent data from
placebo sessions. A significant Drug by Phase interaction was
observed on Energy and Intellectual Efficiency (ARCI BG) and
“High” (VAS), and a significant Drug by Phase by Time interac-
tion was observed on Euphoria (ARCI MBG). Fisher-Hayter post-
hoc tests were conducted on the ARCI MBG data and asterisks
indicate which points are different from placebo (Fisher-Hayter
post-hoc, P< 0.01). The maximum score on the ARCI BG and
MBG scale is 16, and on the VAS it is 100
high during the follicular phase than during the luteal
phase.
Relationship between plasma estradiol
and response to AMPH
To test the hypothesis that higher levels of estrogen dur-
ing the follicular phase would be associated with greater
subjective effects of AMPH, correlations were per-
formed between baseline (pre-AMPH) estradiol levels
and peak change from baseline scores on the subjec-
tive effects measures after AMPH. Positive correlations
were observed between baseline estradiol levels and
ratings of Energy and Intellectual efficiency (ARCI BG
scale; r= 0.64, P< 0.01; Fig. 4) and Euphoria (ARCI
MBG scale; r= 0.56, P< 0.04), and a marginally sig-
nificant correlation was observed with AMPH-like
effects (ARCI A scale; r= 0.49, P< 0.06). No
significant relationships were observed between either
estradiol or progesterone levels, and response to
AMPH during the luteal phase, or between the ratios
of estradiol to progesterone and response to AMPH
during the luteal phase. In addition, there were no con-
sistent relationships between plasma hormone levels
and self-reported mood states before drug administra-
tion.
Discussion
The present study demonstrated that the subjective and
behavioral effects of AMPH vary across the menstrual
cycle and appear to be related to levels of both estro-
gen and progesterone. First, it was found that the effects
of AMPH were greater during the follicular phase than
during the luteal phase. Subjects reported feeling more
energetic and intellectually efficient, euphoric, and high
after AMPH during the follicular phase than the luteal
phase. Moreover, subjects reported liking AMPH more
and wanting AMPH more during the follicular phase.
Second, it was found that within the follicular phase
higher levels of estrogen were associated with greater
subjective and behavioral responses to AMPH. During
the follicular phase, subjects with higher plasma estra-
diol levels reported larger increases in Energy and
Intellectual efficiency (ARCI BG) and Euphoria (ARCI
MBG). These results provide the first demonstration
that the subjective, or mood-altering, effects of AMPH
are influenced by menstrual cycle phase, and that the
effects of AMPH are enhanced when estrogen levels are
high and progesterone levels are low.
These findings are consistent with findings from lab-
oratory animals that during estrus when estrogen is
high and progesterone is low, rats show increased loco-
motor activity, rotational behavior and stereotypy in
response to AMPH (e.g., Becker et al. 1982; Becker
and Cha 1989). Likewise, in the present study, we found
that women reported feeling more energetic and
euphoric after AMPH during the follicular phase when
estrogen levels were high and progesterone levels were
low. Furthermore, those women with the highest estra-
diol levels reported feeling the greatest increases in
“energy” after AMPH. In rats, estrogen is known to
facilitate the release of DA, and increased DA activity
is believed to underlie the reinforcing effects of AMPH
in laboratory animals (e.g., Koob and Nestler 1997).
Although this is a descriptive study and the results are
correlational, it is tempting to speculate that the
enhanced effects of AMPH during the follicular phase
were related to an increased release of DA via interac-
tions with estrogen.
It is interesting to note that estradiol levels were not
related to AMPH response during the luteal phase. One
possible explanation is that the presence of proges-
terone during the luteal phase may have directly or indi-
rectly counteracted the effects of estrogen. Progesterone
has been shown to antagonize estrogen-dependent
increases in DA transmission (Fernandez-Ruiz et al.
1990; Shimizu and Bray 1993), AMPH-stimulated DA
release (Dluzen and Ramirez 1987), and behavioral
responses to AMPH in estrogen-primed animals
(Michanek and Meyerson 1982). Progesterone may
similarly block the subjective effects of AMPH during
the luteal phase in women. Another possible explana-
tion is that neuroadaptation to the elevated levels of
estrogen and/ or progesterone occurred during the
luteal phase, thus creating a situation in which the sub-
jects became “cross-tolerant” to the effects of AMPH
(Friedman et al. 1993).
One important limitation to this study is its descrip-
tive and correlational nature. First, a variety of hor-
monal and physiological changes occur concurrently
across the menstrual cycle making it impossible to
73
Fig. 4 Individual subjects’ estradiol levels (n= 16) plotted as a func-
tion of their peak change score on measures of Energy and
Intellectual Efficiency (ARCI BG; top panel) after AMPH during
the follicular phase (r= 0.64, P< 0.01). Plasma hormone levels were
obtained prior to drug administration. Peak change scores were
calculated by subtracting each subjects’ baseline (pre-capsule) score
from the highest or lowest score after capsule ingestion. The
Pearson’s rcorrelation is shown
identify individual causal variables (Leibenluft et al,
1994; Griffin and Ojeda 1996). Future studies in which
hormone levels are manipulated experimentally (e.g.,
exogenous hormone administration) would be valuable
to investigate these relationships further. Second, a
large number of correlations were conducted, and
although we discuss correlations with a probability of
less than 0.06 as potentially interesting, it should be
noted that these do not meet statistical significance.
Another limitation to this study is that we did not
measure plasma AMPH levels, leaving the possibility
that there were pharmacokinetic differences between
the follicular and luteal phases, or within the follicular
phase. Generally, factors that might be expected to
affect drug distribution such as weight, percentage body
fat, body volume (Byrd and Thomas 1983), urinary and
plasma volume and total body water (Hamilton and
Yonkers 1996) do not significantly change across the
menstrual cycle in most women. Furthermore, in the
present study, both systolic blood pressure and ratings
of “Feel drug” were similar between the two phases
suggesting that the variations in effects were specific to
certain measures. In particular, the effects of AMPH
that did vary with cycle phase or hormone levels may
be effects that are mediated, at least in part, by the DA
system. Therefore, while it cannot be ruled out that
women achieved higher plasma AMPH levels during
the follicular phase, it does not seem likely.
Another limitation of this study is that only a sin-
gle dose of AMPH was tested. This dose was chosen
because it is known to produce reliable, but modest,
subjective effects, thereby allowing us to detect phase-
dependent increases or decreases in the magnitude of
response (Brauer and de Wit 1997). In future studies,
it would be of interest to test a range of AMPH doses
to test for a leftward shift in the entire dose-reponse
curve for AMPH during different phases of the cycle.
Nevertheless, this was a descriptive study and the
results indicate that the effects of a relatively low dose
of AMPH are significantly affected by relative levels of
ovarian hormones.
In summary, these results provide evidence that the
effects of AMPH are influenced by menstrual cycle
phase, and perhaps by hormonal environment. This
confirms findings with laboratory animals, and sup-
ports cross-species inference, including between such
diverse measures as locomotor stimulation and sub-
jective reports of stimulant effects. The locomotor and
reinforcing effects of AMPH are known to be medi-
ated by DA and increased by estrogen in laboratory
animals, and the results of the present study are a first
step in determining if similar mechanisms may also
mediate the subjective effects of AMPH in humans.
Results of the present study may also have important
implications for the growing population of female stim-
ulant users. Of particular interest is the observation
that the euphoric effects of AMPH and the perception
of being “high” were greater during the follicular phase,
but physiological parameters such as systolic blood
pressure were similar during the two phases. If women
use drugs to achieve a certain level of feeling high, this
could potentially lead to an increased incidence of car-
diac problems when stimulants are consumed during
the luteal phase. These results also have important
implications for studies investigating gender differences
in the effects of stimulant drugs. Future studies com-
paring the effects of AMPH in men and women and
examining the role of hormonal state are crucial if we
are to understand the biological basis for gender
differences in stimulant use and determine why the
number of female stimulant users has increased almost
20% in three years (US DHHS 1993, 1996).
Acknowledgements This work was supported by grants from the
NIH (R01 DA02812 and M01 RR00055). Angela Justice was sup-
ported by T32 DA07255. The authors thank Christopher O’Connor
for excellent technical assistance and Dr. Margaret Rukstalis for
professional consultations.
References
American Psychiatric Press (1994) Diagnostic and Statistical
Manual of Psychiatry, 4th edn. Washington, D.C.
Banerjee U, Lin GS (1973) On the mechanisms of central action of
amphetamine: the role of catecholamines. Neuropharmacology
12:917–931
Bazzett TJ, Becker JB (1994) Sex differences in the rapid and acute
effects of estrogen on striatal D
2
dopamine receptor binding.
Brain Res 637:163–172
Becker JB, Beer ME (1986) The influence of estrogen on nigrostri-
atal dopamine activity: behavioral and neurochemical evidence
for both presynaptic components. Behav Brain Res 19:27–33
Becker JB, Cha JH (1989) Estrous cycle-dependent variation in
amphetamine-induced behaviors and striatal dopamine assessed
with microdialysis. Behav Brain Res 35:117–125
Becker JB, Ramirez VD (1981) Experimental studies on the devel-
opment of sex differences in the release of dopamine from
striatal tissue fragments in vitro. Neuroendocrinology 32:
168–173
Becker JB, Robinson TE, Lorenz KA (1982) Sex differences and
estrous cycle variations in amphetamine-elicited rotational
behavior. Eur J Pharmacol 80:65–72
Brauer LH, de Wit H (1997) High dose of pimozide does not block
amphetamine-induced euphoria in normal volunteers. Pharma-
col Biochem Behav 56:265–272
Byrd PJ, Thomas TR (1983) Hydrostatic weighing during different
stages of the menstrual cycle. Res Quart Exerc Sport 54:
296–298
Derogatis, L (1983) SCL-90-R Manual-II. Clinical psychometric
research. Towson, MD
Di Paolo T, Rouillard C, Bedard P (1985) 17-beta-Estradiol
at a physiological dose acutely increases dopamine turnover
in rat brain. Eur J Pharmacol 117:197–203
Dluzen DE, Ramirez VD (1984). Bimodal effect of progesterone on
in vitro dopamine function of the rat corpus striatum. Neuro-
endocrinology 39:149–155
Dluzen DE, Ramirez VD (1985) In vitro dopamine release from
the rat striatum: diurnal rhythm and its modification by the
estrous cycle. Neuroendocrinology 41:97–100
Dluzen DE, Ramirez VD (1987) Intermittent infusion of proges-
terone potentiates whereas continuous infusion reduces amphet-
amine-stimulated dopamine release from ovariectomized
estrogen-primed rat striatal fragments superfused in vitro. Brain
Res 406:1–9
74
Fernandez-Ruiz JJ, de Miguel R, Hernandez ML, Ramos JA (1990)
Time-course of the effects of ovarian steroids on the activity of
limbic and striatal dopaminergic neurons in female rat brain.
Pharmacol Biochem Behav 36:603–606
Fischman MW, Foltin RW (1991) Utility of subjective-effects mea-
surements in assessing abuse liability of drugs in humans. Br J
Addict 86:1563–1570
Friedman L, Gibbs TT, Farb DH (1993) Gamma-aminobutyric
acid A receptor regulation: chronic treatment with preganolone
uncouples allosteric interactions between steroid and benzodia-
zepine recognition sites. Mol Pharmacol 44: 191–197
Griffin JE, Ojeda SR (eds) (1996) Textbook of endocrine physiol-
ogy, 3rd edn. Oxford University Press, New York, pp 3–374
Haertzen CA (1974) An overview of Addiction Research Center
Inventory scales (ARCI): an appendix and manual of scales.
National Institute on Drug Abuse, Addiction Research Center,
Lexington, Kentucky
Hamilton JA, Yonkers KA (1996) Sex differences in pharmaco-
kinetics of psychotropic medications, part I: physiological basis
for effects. In: Jensvold MF, Halbreich I, Hamilton JA (eds)
Psychopharmacology and women: sex, gender, and hormones.
American Psychiatric Press, Washington, D.C., pp 11–42
Hruska RE, Silbergeld EK (1980) Increased dopamine receptor sen-
sitivity after estrogen treatment using the rat rotation model.
Science 208:1466–1467
Joyce JN, VanHartesveldt C (1984) Behaviors induced by intras-
triatal dopamine vary independently across the estrous cycle.
Pharmacol Biochem Behav 20:551–557
Koob GF, Bloom FE (1988) Cellular and molecular mechanisms
of drug dependence. Science 242:715–723
Koob GF, Nestler EJ (1997) The neurobiology of drug addiction.
J Neuropsychiatr Clin Neurosci 9:482–497
Leibenluft E, Fiero PL, Rubinow DR (1994) Effects of the men-
strual cycle on dependent variables in mood disorder research.
Arch Gen Psychiatry 51:761–781
Luine E, Khylchevskaya RJ, McEwen BS (1975) Effect of gonadal
steroids on activities of monoamine oxidase and choline acety-
lase in rat brain. Brain Res 86:293–306
Martin WR, Sloan JW, Sapira JD, Jasinski DR (1971) Physiologic,
subjective and behavioral effects of amphetamine, metham-
phetamine, ephedrine, phenmetrazine and methylphenidate in
man. Clin Pharmacol Ther 12:245–258
McEwen BS, Parsons B (1982) Gonadal steroid action of the brain:
neurochemistry and neuropharmacology. Annu Rev Pharmacol
Toxicol 22:555–598
McNair D, Lorr M, Droppleman L (1971) Profile of mood states.
Educational and Industrial Testing Service, San Diego
Michanek A, Meyerson BJ (1982) Influence of estrogen and prog-
esterone on behavioral effects of apomorphine and ampheta-
mine. Pharmacol Biochem Behav 16:875–879
Pasqualini C, Olivier V, Guibert B, Frain O, Leviel V (1995) Acute
stimulatory effect of estradiol on striatal dopamine synthesis. J
Neurochem 65:1651–1657
Pasqualini C, Olivier V, Guibert B, Frain O, Leviel V (1996) Rapid
stimulation of dopamine synthesis by estradiol. Cell Mol
Neurobiol 16:411–415
Rance N, Wise PM, SelmanoffMK, Barraclough CA (1981)
Catecholamine turnover rates in discrete hypothalamic areas
and associated changes in median eminence luteinizing hor-
mone-releasing and serum gonadotropins on proestrus and die-
strous day 1. Endocrinology 108:1795–1802
Randrup A, Munkvad I (1966) Role of catecholamines in the
amphetamine excitatory response. Nature 24:540
Rosenfield RL, Barnes RB, Ehrmann DA (1994) Studies of thature
of 17-hydroxyprogesterone hyerresponsiveness to gonado-
tropic-releasing hormone agonist challenge in functional ovar-
ian hyperandrogenism. J Clin Endocrinol Metab 79:1686–1692
Seiden LS, Sabel GA, Ricuardi GA (1993) Amphetamine effects on
catecholaminergic systems and behavior. Annu Rev Pharmacol
Toxicol 362:639–677
Shimizu H, Bray GA (1993) Effects of castration, estrogen replace-
ment and estrus cycle on monoamine metabolism in the nucleus
accumbens, measured by microdialysis. Brain Res 621:200–206
Stern KN, McClintock MK (1996) Individual variation in biolog-
ical rhythms: accurate measurement of preovulatory LH surge
and menstrual cycle phase. In: Jensvold MF, Halbreich U,
Hamilton JA (eds) Psychopharmacology and women: sex,
gender, and hormones. American Psychiatric Press,
Washington, D.C., pp 11–42
US Department of Health and Human Services (1993) Preliminary
estimates from the 1993 national household survey on drug
abuse. Substance Abuse and Mental Health Services
Administration
US Department of Health and Human Services (1996) Preliminary
estimates from the 1996 national household survey on drug
abuse. Substance Abuse and Mental Health Services
Administration
Wise RA (1978) Catecholamine theories of reward: a critical review.
Brain Res 152:215–247
Xiao L, Becker JB (1994) Quantitative microdialysis determination
of extracellular striatal dopamine concentration in male and
female rats: effects of estrous cycle and gonadectomy. Neurosci
Lett 180:155–158
75
