Content uploaded by Louise Dye
Author content
All content in this area was uploaded by Louise Dye on Jun 05, 2016
Content may be subject to copyright.
Dietary supplements and herbal remedies for
premenstrual syndrome (PMS): a systematic
research review of the evidence for their efficacy
SARAH CANNING,MITCH WATERMAN &LOUISE DYE
Institute of Psychological Sciences, University of Leeds, UK
Abstract Many women with PMS use alternative therapies, although there has been little
research to demonstrate their efficacy. This systematic review provides a comprehensive
discussion of dietary supplements and herbal remedies commonly used for premenstrual
syndrome (PMS), including calcium, magnesium, vitamin B6, evening primrose oil, Vitex
agnus castus, ginkgo biloba and St John’s Wort. Randomized controlled trials of magnesium
and evening primrose oil have produced conflicting results, in contrast to the substantial
evidence for the efficacy of calcium and vitamin B6. There are insufficient data to advocate
the use of ginkgo biloba, Vitex agnus castus and St John’s Wort, although preliminary data
seem supportive. Greater standardization of PMS diagnosis and assessment, with
randomized, double-blind, placebo-controlled trials using larger, representative samples, strict,
prospectively confirmed diagnostic criteria and assessment of treatment efficacy, would help to
clarify the role of these alternative PMS treatments. Although much of the clinical research is
preliminary and/or inadequately controlled, this review will be relevant to the practicing
clinician looking for greater understanding of the alternative therapies available to their
patients with PMS.
Introduction
This paper presents a systematic research review of various dietary and herbal
interventions used by women suffering from premenstrual syndrome (PMS). PMS is a
cyclical condition occurring 7–10 days before the onset of menstruation, and is relieved
shortly after menstrual flow begins (Reid, 1991). The most commonly reported
symptoms include depression, anxiety, irritability and mood swings, as well as physical
symptoms such as breast tenderness and bloating. PMS involves milder symptoms than
Premenstrual Dysphoric Disorder (PMDD) (Ismail & O’Brien, 2001), although PMS
symptoms are severe enough to disrupt women’s lives (Shaw et al., 2003). PMDD
Address for correspondence: Sarah Canning, Institute of Psychological Sciences, University of Leeds, Leeds LS2
9JT, UK. Tel: +44 (0)113 3433198; Fax: +44 (0)113 3435749. Email: psc1sec@leeds.ac.uk
Received: 26 May 2006. Accepted: 17 August 2006.
JOURNAL OF REPRODUCTIVE AND INFANT PSYCHOLOGY,
VOL. 24, NO. 4, NOVEMBER 2006, pp. 363–378
ISSN 0264-6838/print/ISSN 1469-672X/online/06/040363-16
#2006 Society for Reproductive and Infant Psychology
DOI: 10.1080/02646830600974170
characterizes a subgroup of women whose symptoms are particularly severe and is
confirmed using DSM-IV-TR criteria.
PMS assessment
The diagnosis of PMS is based on the pattern of symptoms over the menstrual cycle.
Both retrospective and prospective measures are used for this purpose. Retrospective
questionnaires promote incorrect symptom timing, bias due to cultural expectations
and heavy reliance upon memory of symptoms (Connolly, 2001), resulting in an
inflated estimate of symptom severity (De Souza et al., 2000). Prospective daily self-
report instruments are the only accepted means of confirming PMS diagnosis (Steiner
& Wilkins, 1996). These are easy to administer, less reliant on memory (Haywood et al.,
2002) and highlight intercycle variability in symptom type and severity (Connolly,
2001). However, they are demanding to complete and may bias symptom patterns
though nonadherence (Connolly, 2001).
Various measures have been used to prospectively confirm PMS diagnosis. Budeiri
et al. (1994) identified over 65 instruments, but concluded that there is not yet
agreement about which is the most appropriate. At least five different daily diaries are
widely used but there is no consensus as to which one is best (Steiner et al., 2003).
The diagnosis of PMS is also based on symptom severity through the application of
severity criteria to symptoms, such that an increase of at least 30% from the follicular to
luteal phase scores is required to confirm PMS (NIMH, 1983).
Many women with PMS use alternative therapies (Girman et al., 2003), even though
the efficacies of these are not established (Domoney et al., 2003). This review will
evaluate evidence for the effectiveness of dietary and herbal interventions for PMS
assessed by randomized, double-blind, placebo-controlled trials, in women whose PMS
was diagnosed using validated measures. The studies considered in this review did not
distinguish between PMS and PMDD.
Methods
Selection criteria
For review inclusion, studies were required to include a placebo/comparison group, as
the placebo effect has been shown to be large in women with PMS (Freeman & Rickels,
1999). They were required to test the efficacy of one treatment, taken for at least one
cycle, taken throughout or premenstrually. Participants had to be randomized to
treatments in the case of parallel designs, or have treatment orders counterbalanced in
the case of crossover designs.
Studies required participants of reproductive age, with PMS or PMDD, diagnosed
prospectively or retrospectively. Few trials employed prospective diagnosis or
assessment of efficacy hence, in order to provide a comprehensive review, retrospective
measures were included. Women had to have no other pre-existing psychiatric
conditions (e.g. depression, anxiety), although studies including women presenting
with depression or anxiety only premenstrually were included.
Studies employing outcome measures which examined combined PMS symptoms,
global scores or specific symptoms, e.g. cyclical breast pain, were included. Trials
including women taking oral contraceptives were also included.
364 S. CANNING ET AL.
Search strategy
Electronic databases
Trials were identified by searching Embase (1980–2006), Medline (1966–2006),
AMED (1985–2006), Cinahl (1982–2006), PsycINFO (1967–2006), and the
Cochrane Controlled Trials Register database.
A general search on these databases revealed dietary supplements and herbal
remedies used for PMS, including vitamin B6, magnesium, calcium, Vitex agnus castus,
evening primrose oil, St John’s Wort and ginkgo biloba supplements. Databases were
searched using all Latin and English names for these supplements. Hence, the
following keywords were used:
NPms, pmt, pmdd, llpd, llpdd, premenstrual syndrome, pre menstrual syndrome, pre-
menstrual syndrome, premenstrual tension, pre menstrual tension, pre-menstrual
tension, premenstrual dysphoria, pre-menstrual dysphoria, pre menstrual dysphoria,
premenstrual dysphoric disorder, pre-menstrual dysphoric disorder, pre menstrual
dysphoric disorder, late luteal
NVitex, agnus castus, vitex agnus castus, vitex agnus-castus, chaste tree, chasteberry,
chaste-berry, monk’s pepper, monks pepper, hemp tree, agnolyt, agnufemil,
castufemin, cefanorm, femicur, gynocastus, hewekliman, kytta-femin, strotan,
agnomens
NEvening primrose oil, oenothera biennis, evening primrose, primrose oil, oenothera,
biennis, fever plant, oep, sundrop, essential fatty acids, efamol
NCalcium, calcium supplements, calcium therapy
NMagnesium, magnesium therapy
NVitamin B6, vitamin B6, vitamin b-6, vitamin B-6, vitamin therapy, vitamins,
pyridoxine, B-vitamins, B vitamins, pyridoxine hydrochloride
NSt john’s wort, st johns wort, hypericum perforatum, hypericum, perforatum,
hypericin, hypericins, kira
NGingko, gingko biloba, ginko, ginko biloba, biloba, living fossil, Japanese Silver
Apricot, Kew Tree, Maidenhair Tree, Yinsing
NAlternative therapy, alternative therapies, herbal therapy, nutritional supplements
NClinical trial, trial, randomized controlled trial, controlled trial, randomized,
randomized controlled trial, blind, double blind, double-blind, doubleblind,
crossover, cross-over, parallel, prospective, retrospective
There were 113 articles remaining when duplicates were removed, with 32 articles kept
as trials relevant to the research area. Seven articles did not meet the selection criteria,
five of which had no placebo or comparison group (Berger et al., 2000; Brush et al.,
1988; Larsson et al., 1989; Loch et al., 2000; Stevinson & Ernst, 2000). Two articles
were excluded as they tested the efficacy of more than one treatment. Krutan Berman
et al. (1990) tested the efficacy of pyridoxine for PMS, in combination with a dietary
intervention, while Callender et al. (1988) tested evening primrose oil using efamol
tablets which also included efavit (containing vitamin C, pyridoxine, niacin, and zinc
sulfate). Hence, 25 studies meeting the selection criteria were retained (see Table 1).
One additional article (Ockerman et al., 1986) was identified from the reference lists
from review articles in this area.
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 365
Data synthesis for the dietary supplements and herbal remedies for PMS
Two studies of calcium, four of magnesium, 10 of B6, four of evening primrose oil, four
of Agnus castus, one of St. John’s Wort and one of gingko biloba met the inclusion
criteria. Fifteen trials suggested some benefit of the treatment under investigation,
while 11 trials found no such benefit.
Table 1 describes and evaluates these studies. The trials are ordered alphabetically
and sub-divided into studies finding positive and negative effects for each treatment.
Aspects of the methodological quality (e.g. sample size, design, dose and duration of
intervention, screening and assessment tools employed) are considered in order to
provide a context for discussion of the reliability of the results. The table assumes that
studies excluded women taking the oral contraceptive pill, unless otherwise stated.
Two well-designed trials rigorously assessed the efficacy of calcium supplementation
for PMS (Thys-Jacobs et al., 1989, 1998). The similarities between their findings
suggest that calcium supplementation of at least three cycles may be of benefit to
women suffering from PMS.
Positive evidence was also found for vitamin B6. A systematic review to evaluate
the efficacy of vitamin B6 for the treatment of PMS identified 25 published trials and
included nine trials, representing 940 patients (Wyatt et al., 1999). The overall odds
ratio for efficacy of B6 was 2.32 (1.95–2.54). Wyatt et al. suggest that doses of
vitamin B6 up to 100 mg/day may benefit premenstrual depression and other
symptoms. However, conclusions from the meta-analysis were limited by the
methodological weaknesses of some of the trials included. Findings from studies
assessing vitamin B6 in this review are also mixed, although studies which
demonstrated benefits appeared to be methodologically stronger. The evidence
suggests that continuous vitamin B6 treatment at doses of 50 to 150 mg/day may be
beneficial for some PMS symptoms, since the intermittent treatment (at 50 to
300 mg/day) did not prove effective (Diegoli et al., 1998; Mal mgren et al., 1987;
Stokes & Mendels., 1972). However, more trials, using stricter diagnostic criteria, are
required to confirm its benefit.
Trials looking at magnesium supplementation produced mixed findings and many
methodological limitations were apparent. Although most trials evaluating Agnus castus
treatment reported positive effects, many studies also suffered methodological
problems. Though some studies suggest evening primrose oil may benefit PMS
symptoms (Ockerman et al., 1986; Puolakka et al., 1985), currently the evidence is not
convincing, especially as the most methodologically sound study (Collins et al., 1993)
found no benefit for mood or physical symptoms. Therefore, trials with longer
treatment durations, tighter controls and larger samples are required to evaluate Agnus
castus, evening primrose oil and mg supplementation in PMS.
Only single studies have been performed to test the efficacy of St John’s Wort and
gingko biloba, so further investigation of these therapies is needed. Hicks et al. (2004)
suggested that their study had insufficient statistical power to demonstrate the efficacy
of the dose of St John’s Wort that they used (600 mg, standardized to 0.3% of
hypericin/day). Most studies have used a dose of 900 mg/day standardized to 0.3%
hypericin for depression (Szegedi et al., 2005), and this is the dose recommended for
depression by most manufacturers of St John’s Wort. There is currently no
recommended dose for PMS. Tamborini and Taurelle (1993) found gingko biloba
to benefit congestive PMS symptoms, particularly breast tenderness. However, the
sample of PMS sufferers studied was atypical in that women were required to report
366 S. CANNING ET AL.
Table 1. Summary of study characteristics.
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Calcium (Ca)
Thys-Jacobs et al.
(1989)
N560, 33
completed; 45%
dropout
R, DB
CO, PC
2 tablets providing
1000 mg elemental
Ca/d; 3 cycles
14 symptoms
rated daily on a
4-pt scale;
1 cycle
Daily symptom
change; global
assessment
measured
retrospectively
3negative affect,
water retention,
pain; retrospective
assessment
Well designed trial.
No washout.
Prospective
diagnosis but for
only 1 cycle.
Thys-Jacobs et al.
(1998)
N5479, 441
completed; 8%
dropout
R, DB
PGs, PC, MC
262, 750 mg CaCO3
tablets/d (1200 mg
elemental Ca/d);
3 cycles
17 symptoms
(4 factors:
negative affect,
water retention,
food and pain)
rated daily on a
4-pt scale;
2 cycles
Symptom complex
score during luteal
and menstrual
phases; factor
scores; rescue
medication used
3symptom
complex scores;
all symptom
factors
Well designed
trial.
Placebo treatment
not specified.
Strict diagnostic
criteria.
OC users included,
although no
difference between
users/non-users
found.
Magnesium (mg)
Facchinetti et al.
(1991)
N532, 28
completed; 13%
dropout
R, DB
PGs, PC
3 tablets providing
360 mg mg ion/d from
cycle day 15; 2 cycles
DSM III-R,
MDQ
MDQ daily during
the 2nd and 4th
cycles
3total MDQ
score, especially
negative affect
and arousal
Small sample.
Walker et al.
(1998)
N554, 24
completed; 56%
dropout
R, DB
CO, PC
200 mg of mg as mgO; 2
cycles
MHQ 22 symptoms daily
on a 4-pt scale,
covering 6
categories
(anxiety, craving,
depression,
hydration, other
and total)
3hydration
symptoms only
(weight gain,
swelling, breast
tenderness and
abdominal
bloating)
No washout.
Retrospective diag-
nosis.
Low dose.
OC users included.
No baseline diary
measurements
taken.
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 367
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
De Souza et al.
(2000)
N544, 37
completed;
16% dropout
R, DB
CO, PC
Each daily for 1 cycle: MHQ; 1 mth
menstrual diary
30 symptoms daily
on a 5-pt scale,
covering 6
categories (anxiety,
craving, depression,
hydration, other and
total)
7except combined
treatment, which
reduced mild
anxiety related
symptoms
OC users included
(18%).1) 200 mg mg, as mgO
Diagnosis methods
not specified.
2) 50 mg B6
Treatment for only
1mth.
3) 200 mg mg+50 mg B6
No washout.
4) placebo
Walker et al.
(2002)
N585 R, DB
CO, PC,
sorbitol
2 of the following for 2
cycles each, with 1 mth
washout: placebo, mg
(200, 350 or 500 mg/d,
containing 1050, 813
and 717 mg of sorbital)
MHQ 20 symptoms daily
on a 5-pt scale,
covering 5
categories (anxiety,
craving, depression,
hydration, total)
7Sorbitol
significantly better
than mg for
anxiety and total
PMS scores
OC users included
(28%).
Vitamin B6 (B6)
Abraham &
Hargrove (1980)
N525 R, DB
CO, PC
16B6 tablet/d
containing 500 mg of
pyridoxine HC1;
3 cycles
MSQ (Abraham,
1980)
19 symptoms daily
on a 4-pt scale
3total symptom
score; premenstrual
weight gain
Only 3 women with
premenstrual weight
gain included.
Completer rate
inconsistent.
Unclear when
symptoms were
rated for diagnosis.
Barr (1984) N548, 36
analysed; 25%
dropout assumed
R, DB
CO, PC
16100 mg pyridoxine
Hcl/d from cycle
days 10–3; 2 cycles
Not specified 8 symptoms daily
for 2 wks prior to
menstruation
3overall scores OC users included.
Entry criteria not
specified.
Symptoms only
recorded with posi-
tive/negative
responses.
Table 1. (Continued.)
368 S. CANNING ET AL.
Table 1. (Continued.)
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Doll et al. (1989) N568, 32
completed; 53%
dropout; moderate
to severe PMS
R, DB
CO, PC
50 mg of pyridoxine/d;
3 cycles
9 symptoms
(emotional,
somatic and
menstrual) rated
on a 4-pt scale;
1 cycle
9 symptoms on a
4-pt scale
throughout
treatment
3emotional type
symptoms
only—depression,
irritability,
tiredness
OC users included.
No washout.
It was not specified
how the scale was
used for diagnosis or
treatment
assessment.
Kendall & Schnurr
(1987)
N574, 55
completed;
(B6, 29; Pl, 26);
26% dropout
R, DB
PGs, PC
3650 mg B6 tablets/d;
2 cycles
PAF (Halbreich
et al., 1982)
MDQ, every other
day
3autonomic
reactions
(dizziness,
vomiting);
behaviour changes
(poor performance,
decreased social
activities) only
Baseline MDQ not
used for exclusion.
Half the women in
each condition
received capsules
and half tablets.
Williams et al.
(1985)
N5617, 434
completed; 30%
dropout
R, B PGs, PC 100 mg pyridoxine
tablet/d; 3 cycles. The
dose could be doubled/
halved by the patients.
Through the
general
practitioner, if
menstruation
relieved 1 or more
symptom
Symptom rating
during the last
week of each cycle
3for final
assessment
compared to entry
but not for
individual
symptoms
Retrospective
diagnosis and
treatment
assessment.
‘Other medication’
could be taken.
Many women
changed their dose.
Diegoli et al.
(1998)
N5120 (30 in
each group)
R, DB
CO, PC
One for 3 cycles: 24 symptoms
rated
retrospectively
on a 4-pt scale
24 symptoms on
a 4-pt scale at
the end of each
cycle
7only for the
pyridoxine group
Doses and
frequency of
treatment
1) 300 mg pyridoxine/d
from day 15
differed between
groups.
2) 360.25 mg of
alprazolam/d from day 15
Retrospective diag-
nosis and treatment
assessment.
3) 10 mg fluoxetine/d
It was unclear when
treatment effect was
assessed.
4) 20 mg propanlol/d,
40 mg when menstruating
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 369
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Hagen et al.
(1985)
N542, 34
completed; 19%
dropout
R, DB
CO, PC
2650 mg pyridoxine
tablets/d; 2 cycles
Interview
conducted by the
research team
VAS at baseline
and after each
treatment; ranking
of 6 symptoms,
adding others
7Treatment order not
fully counter-
balanced.
When rating,
women were
allowed to see their
previous ratings, and
it was unclear when
this was performed.
Malmgren et al.
(1987)
N519 DB,
CO PC
300 mg pyridoxine/d
from cycle day 15;
1 cycle
MDQ; STAI
(Spielberger et al.,
1970) on cycle days
5–7 and 25–27
Not specified 7No washout.
B6 and placebo were
only given for one
cycle each.
Smallwood et al.
(1986)
N542, severe
cyclical
mastalgia
R, DB
CO, PC
200 mg B6 (benadon)/d;
2 cycles
Not specified Monthly, by a
clinician within the
last 5 days of the
cycle; daily VAS for
breast pain and
tenderness;
paracetamol
requirements
7for all measures Only women having
severe pain
premenstrually for
at least 6
consecutive mths
included.
Exclusion of OC
users not stated.
Method of diagnosis
not specified.
Only a small
proportion of
symptoms assessed.
Table 1. (Continued.)
370 S. CANNING ET AL.
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Stokes & Mendels
(1972)
N513,
premenstrual
tension,
depression
R, DB
CO, PC
50 mg B6 or placebo/d
for 18 days/mth, for
8–12 mths, with the
order of placebo and
B6 months being
random
Not specified MDQ 7Method of diagnosis
not mentioned.
Tiny sample size.
It was unclear for
how long each
treatment was taken.
Treatment
assessment method
not explicit.
Evening Primrose
Oil (EPO)
Ockerman et al.
(1986)
N536, severe
PMS
DB,
PGs PC,
olive oil
860.5g Efamol
capsules/d;
3 cycles
Not specified Not specified 3Treatment resistant
sample. Extremely
brief report failing to
specify diagnosis
methods, symptoms
measured and
measures used.
Puolakka et al.
(1985)
N530,
incapacitating
PMS
R, CO PC 263 Efamol capsules
from cycle day 15;
4 cycles
19 symptoms
rated on a
3-pt scale
19 symptoms
on a 3-pt scale
on the last
treatment day
3depression; total
score; responder
rate: Efamol (62%),
placebo (40%)
Placebo treatment
not defined.
Double-blinded?
Retrospective
measurement?
Exclusion of OC
users not specified.
Data for only 22
women analysed?
Analysis unclear.
Table 1. (Continued.)
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 371
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Collins et al.
(1993)
N568, 38
completed; 44%
dropout
R, DB
CO, PC,
paraffin oil
364 Efamol capsules/d
for 4 cycles
DSM-IIIR criteria;
prospective 16
symptom VAS
(Hammarbeck
et al., 1989)
VAS 7all mood and
physical symptoms
Well controlled
study.
Khoo et al.
(1990)
N538; no
dropouts;
moderate PMS
R, DB
CO, PC,
liquid paraffin
8 EPO capsules/d;
3 cycles
Retrospective 10
symptom scale
10 symptoms
(4 psychological,
6 physical)
completed
retrospectively
each cycle
7total PMS score;
psychological, fluid
retention, breast,
and menstrual
symptoms
Dose not stated.
Diagnosis process
unclear and seemed
to be retrospective.
It was unclear when
symptoms were
recorded in each
cycle.
Vitex agnus castus
(AC)
Atmaca et al. (2003) N541, 38
completed; 7%
dropout
(fluoxetine, 21;
AC, 20) PMDD
R, SB
PGs, PC
Fluoxetine or AC
(20–40 mg/d);
2 cycles
DSM-IV; Penn
daily symptom
reports; 2 cycles
DSM-IV criteria
for PMDD;
premenstrual
score of the DSR,
HAM-D (Hamilton,
1960), CGI-I;
agreement of
improvement by
the authors
3No significant
difference between
groups on DSR,
CGI-SI scores or
responder rates
AC compared to
fluoxetine – no
placebo group.
Participants and
raters blinded but
prescribing
physician not.
Halaska et al.
(1998)
N5100,
completion: AC
48, placebo 49,
3% dropout
DB, PGs PC 2630 AC drops
(1.8ml)/d; 3 cycles
Not specified VAS scale 3breast pain Only women
suffering from
mastalgia.5 days/
cycle included.
Table 1. (Continued.)
372 S. CANNING ET AL.
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Schellenberg
(2001)
N5170 (active,
86; placebo 84)
R, DB
PGs, PC
1620 mg AC
tablet/d; 3 cycles
DSM-III-R 6 symptoms
(irritability, mood
alteration, anger,
headache, other
menstrual
symptoms including
bloating, breast
fullness) on a VAS
compared to
previous 3 cycles;
responder rate
3combined and
individual
symptoms (not
bloating);
responder rates
52% AC v 24%
placebo
Assessment scale
and timing of
symptom rating was
unclear.
OC users included.
Turner & Mills
(1993)
N5600,
completion:
AC, 105;
placebo 112,
64% dropout
R, DB
PGs, PC
Treatment; 3 mths MDQ MDQ at end of
treatment; a
shortened version
administered at
the end of cycles
1 and 2
7except for the
symptom ‘feel
jittery or restless’
and responder
rates 25% AC v
16% placebo
Dose and frequency
of administration
was unclear.
Only women high in
negative affect
included.
Huge drop out rate,
but evenly
distributed across
groups.
St John’s Wort
(SJW)
Hicks et al. (2004) N5169, 125
completed;
26% dropout
R, DB
PGs, PC,
lactose/
cellulose
26300 mg tablets
standardized to 900 mg
hypericin/d; 2 cycles
Retrospective
assessment; 25
symptoms rated
daily; 1 cycle
VAS assessing 25
symptoms grouped
into 6 categories
(anxiety, craving,
depression,
hydration, other
and total)
7all symptom
subgroups
Although diagnosis
was prospectively
confirmed, this
process was unclear.
Table 1. (Continued.)
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 373
Reference Sample size Study design Dose and duration Diagnosis Assessment measures Outcome Comments
Ginko Biloba
Tamborini &
Taurelle (1993)
N5165,
congestive PMS
symptoms
R, DB
PGs, PC MC
EGb 761 from cycle
day 16; 2 cycles
Observation of 1
menstrual cycle
Daily scale
measuring
congestion, breast
tenderness and
mood; practitioner
observation
premenstrually before
and after treatment
3congestive
symptoms,
particularly breast
symptoms
Method of diagnosis
not specified.
Key:
Study design
R Randomized
DB Double-blind
SB Single-blind
PC Placebo-controlled
PGs Parallel groups
CO Crossover
MC Multi-centred
OC Oral contraceptives
Measures
MHQ Mental Health Questionnaire (Warner & Bancroft, 1990)
MDQ Menstrual Distress Questionnaire (Moos, 1968)
VAS Visual analogue scale
Table 1. (Continued.)
374 S. CANNING ET AL.
congestive premenstrual symptoms for 7 days per cycle for the three cycles prior to
recruitment.
Discussion
Many women with PMS use alternative therapies, despite the lack of established
efficacy (Domoney et al., 2003). This review included 26 trials that assessed the efficacy
of seven different dietary supplements and herbal remedies for PMS. The most
substantial positive evidence was found for calcium and continuous vitamin B6
treatment. Trials assessing magnesium and evening primrose oil produced conflicting
findings, whilst insufficient data were found to advocate the use of vitex agnus castus,
gingko biloba or St John’s Wort.
The studies considered in this review differed greatly in the diagnostic methods
they used. It is generally accepted that prospective daily self-report measures are
needed to confirm PMS (Steiner & Wilkins, 1996). Some studies did diagnose by
use of the DSM criteria, and confirmed their diagnoses prospectively. However,
others relied upon retrospective diagnosis, which has been criticized (Connolly,
2001), since these often result in inflated estimates of symptom severity (De Souza
et al., 2000).
The methods used for the assessment of treatment efficacy also differed. Many
studies used the total symptom score of a rating scale as their primary outcome
measure, and simultaneously considered symptom clusters, and some also considered
individual symptoms. This increases the chances of finding symptom effects.
Assessment measures were used prospectively with daily symptom ratings or by
averaging these ratings across phases in some trials. Others assessed treatments
retrospectively, at the end of each cycle or at the end of treatment, using a variety of
methods, including questionnaires, interviews (Kendall & Schnurr, 1987; Loch et al.,
2000) and general practitioner assessments (Smallwood et al., 1986; Williams et al.,
1985). Some authors did not specify their treatment assessment methods (Mal mgren
et al., 1987; Ockerman et al., 1986).
Women taking the oral contraceptive, which has previously been used as a PMS
treatment (Girman et al., 2003), were not excluded from some studies, while
others focused on specific groups of women, including women with ‘premenstrual
tension depression’ (Stokes & Mendels, 1972), severe cyclical mastalgia
(Smallwood et al., 1986) and congestive PMS symptoms (Tamborini & Taurelle,
1993). It is difficult to compare such studies with those examining a more
representative sample.
Conclusion
The variations apparent in diagnostic procedures, assessment methods, and outcome
measures make it difficult to assess treatment efficacy (Connolly, 2001). More
consensus about the diagnosis, measurement and assessment of PMS is required, as are
randomized, double-blind, placebo-controlled trials. Such carefully controlled trials
with strict diagnostic criteria, prospective confirmation of PMS and prospective
assessment of symptoms in response to treatment would help to clarify the efficacy of
the many alternative treatments used for PMS. At the moment, calcium and
continuous vitamin B6 treatment seem likely to be beneficial.
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 375
References
ABRAHAM, G.E. (1980). The premenstrual tension syndromes. In: L.K. McNall (Ed.), Contemporary obstetric
and gynecologic nursing. Vol. III (p. 170). C. V. Mosby: New York.
ABRAHAM, G.E. & HARGROVE, J.T. (1980). Effect of vitamin B-6 on premenstrual tension syndromes: a
double blind crossover study. Infertility,3, 155–165.
AMERICAN PSYCHIATRIC ASSOCIATION. (2000). Diagnostic and statistical manual of mental disorders (4th ed.
rev.). Washington, DC: American Psychiatric Association.
ATMACA, M., KUMRU,S.&TEZCAN, E. (2003). Fluoxetine versus Vitex agnus castus extract in the treatment
of premenstrual dysphoric disorder. Human Psychopharmacology,18, 191–195.
BARR, W. (1984). Pyridoxine supplements in the premenstrual syndrome. The Practitioner,228, 425–427.
BERGER, D., SCHAFFNER, W., SCHRADER,E.&MEIER, B. (2000). Efficacy of Vitex agnus castus L. extract
Ze440 in patients with pre-menstrual syndrome (PMS). Archives of Gynecology & Obstetrics,264, 150–153.
BRUSH, M.G., BENNETT,T.&HANSEN, K. (1988). Pyridoxine in the treatment of premenstrual syndrome: a
retrospective survey in 630 patients. British Journal of Clinical Practice,42, 448–452.
BUDEIRI, D.J., LIWANPO,A.&DORNAN, J.C. (1994). Clinical trials of treatment of premenstrual syndrome:
entry criteria and scales for measuring treatment outcomes. British Journal of Obstetrics and Gynaecology,
101, 689–695.
CALLENDER, K., MCGREGOR, M., KIRK,P.&THOMAS, C.S. (1988). A double-blind trial of evening primrose
oil in the premenstrual syndrome: nervous symptom subgroup. Human Psychopharmacology,3, 57–61.
COLLINS, A., CERIN, A., COLEMAN,G.&LANDGREN, B-M. (1993). Essential fatty acids in the treatment of
premenstrual syndrome. Obstetrics and Gynecology,81, 93–97.
CONNOLLY, M. (2001). Premenstrual syndrome: an update on definitions, diagnosis and management.
Advances in Psychiatric Treatment,7, 469–477.
DESOUZA, M.C., WALKER, A.F., ROBINSON, P.A. & BOLLAND, K. (2000). A synergistic effect of a daily
supplementation for 1 month of 200 mg magnesium plus 50mg vitamin B6 for the relief of anxiety-related
premenstrual symptoms: a randomized, double-blind, crossover study. Journal of Women’s Health and
Gender-Based Medicine,9, 131–139.
DIEGOLI, M.S.C., DA FONSECA A.M., DIEGOLI, C.A. & PINOTTI, J.A. (1998). A double-blind trial of four
medications to treat severe premenstrual syndrome. International Journal of Gynecology and Obstetrics,62,
63–67.
DOLL, H., BROWN, S., THURSTON,A.&VESSEY, M. (1989). Pyridoxine trial. Journal of the Royal College of
General Practitioners,39, 364–368.
DOMONEY, C.L., VASHISHT,A.&STUDD, J.W.W. (2003). Premenstrual syndrome and the use of alternative
therapies. Annals of the New York Academy of Sciences,997, 330–340.
FACCHINETTI, F., BORELLA, P., SANCES, G., FIORONI, L.P., NAPPI, R.E. & GENAZZANI, A.R. (1991).
Oral magnesium successfully relieves premenstrual mood changes. Obstetrics and Gynaecology,78,
177–181.
FREEMAN,E.&RICKELS, K. (1999). Characteristics of placebo responses in medical treatment of
premenstrual syndrome. American Journal of Psychiatry,156, 1403–1408.
GIRMAN, A., LEE,R.&KLIGLER, B. (2003). An integrative medicine approach to premenstrual syndrome.
Obstetrics and Gynecology,188, S56–S65.
HAGEN, I., NESHEIM,B.&TUNTLAND, T. (1985). No effect of vitamin B-6 against premenstrual tension: a
controlled clinical study. Obstetricia et Gynecologica Scandinavica,64, 667–670.
HALASKA, M., RAUS, K., BELES, P., MARTAN,A.&PAITHNER, K.G. (1998). Treatment of cyclical
mastodynia using an extract of Vitex agnus castus: results of a double-blind comparison with a placebo.
Ceska Gynekologie,63, 88–92.
HALBREICH, U., ENDICOTT,J.&SCHACHT, S. (1982). Premenstrual syndromes: a new instrument for their
assessment. Journal of Psychiatric Treatment & Evaluation,4, 161–164.
HAMILTON, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery and Psychiatry,23,
56–62.
HAMMARBACK, S., BACKSTROM,T.&MACGIBBON-TAYLOR, B. (1989). Diagnosis of premenstrual tension
syndrome: description and evaluation of a procedure for diagnosis and differential diagnosis. Journal of
Psychosomatic Obstetrics and Gynecology,10, 25–42.
HAYWOOD, A., SLADE,P.&KING, H. (2002). Assessing the assessment measures for menstrual cycle
symptoms. a guide for researchers and clinicians. Journal of Psychosomatic Research,52, 223–237.
HICKS, S.M., WALKER, A.F., GALLAGHER, J., MIDDLETON, R.W. & WRIGHT, J. (2004). The significance of
‘nonsignificance’ in randomized controlled studies: a discussion inspired by a double-blinded study on St.
376 S. CANNING ET AL.
John’s Wort (Hypericum perforatum L.) for premenstrual symptoms. Journal of Alternative and
Complementary Medicine,10, 925–932.
ISMAIL, K.M.K. & O’BRIEN P.M.S. (2001). Premenstrual syndrome: problem-based learning in gynaecology.
Current Obstetrics and Gynaecology,11, 251–255.
KENDALL, K.E. & SCHNURR, P.P. (1987). The effects of vitamin B6 supplementation on premenstrual
symptoms. Obstetrics and Gynecology,70, 145–149.
KHOO, S.K., MUNRO,C.&BATTISTUTTA, D. (1990). Evening primrose oil and treatment of premenstrual
syndrome. Medical Journal of Australia,153, 189–192.
KRUTAN BERMAN, M., TAYLOR, M.L. & FREEMAN, E. (1990). Vitamin B-6 in premenstrual syndrome.
Journal of the American Dietetic Association,90, 859–861.
LARSSON, B., JONASSON,A.&FIANU, S. (1989). Evening primrose oil in the treatment of premenstrual
syndrome. Current Therapeutic Research,46, 58–63.
LOCH, E., SELLE,H.&BOBLITZ, N. (2000). Treatment of premenstrual syndrome with a phytopharma-
ceutical formulation containing Vitex agnus castus. Journal of Women’s Health and Gender-Based Medicine,
9, 315–320.
MALMGREN, R. (1987). Platelet serotonin uptake and effects of vitamin B6-treatment in premenstrual
tension. Neuropsychobiology,18, 83–88.
MOOS, R. (1968). The development of a menstrual distress questionnaire. Psychosomatic Medicine,30,
853–867.
NIMH. (1983). National Institute of Mental Health. NIMH premenstural syndrome workshop guidelines.
Rockville, MD: National Institute of the Mental Health.
OCKERMAN, P.A., BACKRACK, I., GLANS,S.&RASSNER, S. (1986). Evening primrose oil as a treatment of the
premenstrual syndrome. Recent advances in clinical nutrition,39, 411–414.
PUOLAKKA, J., MAKARAINEN, L., VIINIKKA,L.&YLIKORKALA, O. (1985). Biochemical and clinical effects of
treating the premenstrual syndrome with prostaglandin synthesis precursors. Journal of Reproductive
Medicine,30, 149–153.
REID, R. (1991). Premenstrual syndrome. New England Journal of Medicine,324, 1208–1210.
SCHELLENBERG, R. (2001). Treatment for the premenstrual syndrome with agnus castus fruit
extract: prospective, randomized, placebo controlled study. British Medical Journal,322,
134–137.
SHAW, S., WYATT, K., THOMPSON COON, J., CAMPBELL,J.&ERNST, E. (2003). Vitex agnus castus for
premenstrual syndrome. The Cochrane Database of Systematic Reviews,4, Art. No.: CD004632, DOI:
10.1002/14651858, CD004632.
SMALLWOOD, J., AH-KYE,D.&TAYLOR, I. (1986). Vitamin B6 in the treatment of pre-menstrual mastalgia.
British Journal of Clinical Practice,40, 532–534.
SPIELBERGER, C.D., GORSUCH,R.&LUSHENE, R. (1970). STAI manual. Palo Alto: Consulting Psychologists.
STEINER,M.&WILKINS, A. (1996). Diagnosis and assessment of premenstrual dysphoria. Psychiatric Annals,
26, 571–575.
STEINER, M., DUNN,E.&BORN, L. (2003). Hormones and mood: from menarche to menopause and
beyond. Journal of Affective Disorders,74, 67–83.
STEVINSON,C.&ERNST, E. (2000). A pilot study of Hypericum perforatum for the treatment of premenstrual
syndrome. An International Journal of Obstetrics and Gynaecology,107, 870–876.
STOKES,J.&MENDELS, J. (1972). Pyridoxine and premenstrual tension. The Lancet,1, 1177–1178.
SZEGEDI, A., KOHNEN, R., DIENEL,A.&KIESER, M. (2005). Acute treatment of moderate to severe
depression with hypericum extract WS 5570 (St John’s Wort): randomised controlled double blind non-
inferiority trial versus paroxetine. British Medical Journal,330, 503–506.
TAMBORINI,A.&TAURELLE, R. (1993). Value of standardized gingko biloba (EGb 761) in the management
of congestive symptoms of premenstrual syndrome. Revue Francaise de Gynecologie et d Obstetrique,88,
447–457.
THYS-JACOBS, S., CECCARELLI, S., BIERMAN, A., WEISMAN, H., COHEN, M-A. & ALVIR, J. (1989). Calcium
supplementation in premenstrual syndrome: a randomized crossover trial. Journal of General Internal
Medicine,4, 183–189.
THYS-JACOBS, S., STARKEY, P., BERNSTEIN,D.&TIAN, J. (1998). Calcium carbonate and the premenstrual
syndrome: effects on premenstrual and menstrual symptoms. American Journal of Obstetrics and Gynecology,
179, 444–452.
TURNER,S.&MILLS, S. (1993). A double-blind clinical trial on a herbal remedy for premenstrual syndrome:
a case study. Complementary Therapies in Medicine,1, 73–72.
DIETARY SUPPLEMENTS AND HERBAL REMEDIES FOR PMS 377
WALKER, A.F., DESOUZA, M.C., VICKERS, M.F., ABEYASEKERA, A., COLLINS, M.L. & TRINCA, L.A. (1998).
Magnesium supplementation alleviates premenstrual symptoms of fluid retention. Journal of Women’s
Health,7, 1157–1165.
WALKER, A.F., DESOUZA, M.C., MARAKIS, G., ROBINSON, P.A., MORRIS, A.P. & BOLLAND, K.M. (2002).
Unexpected benefit of sorbital placebo in mg intervention study of premenstrual symptoms: implications
for choice of placebo in RCTs. Medical Hypotheses,58, 213–220.
WARNER,P.&BANCROFT, J. (1990). Factors related to self-reporting of the premenstrual syndrome. British
Journal of Psychiatry,157, 249.
WILLIAMS, M.J., HARRIS, R.I., BERNARD, C.D. & DEAN, C. (1985). Controlled trial of pyridoxine in the
premenstrual syndrome. Journal of International Medical Research,13, 174–179.
WYATT, K.M., DIMMOCK, P.W. & O’BRIEN, P.W. (1999). Efficacy of vitamin B-6 in the treatment of
premenstrual syndrome: systematic review. British Medical Journal,318, 3175–1381.
378 S. CANNING ET AL.
