Anxiolytic Effects of a Combination of Melissa officinalis and Valeriana officinalis during Laboratory induced Stress
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
PHYTOTHERAPY RESEARCH
Phytother. Res. 20, 96–102 (2006)
96 Published online in Wiley InterScience (www.interscience.wiley.com).
D. O. KENNEDY ET DOI: AL. 10.1002/ptr.1787
Anxiolytic Effects of a Combination of Melissa
officinalis and Valeriana officinalis during
Laboratory induced Stress
David O. Kennedy,* Wendy Little, Crystal F Haskell and Andrew B. Scholey
Human Cognitive Neuroscience Unit, Division of Psychology, Northumbria University, Newcastle upon Tyne, NE1 8ST, UK
Objective: Melissa officinalis (lemon balm) and Valeriana officinalis (valerian) have been used both tradi-
tionally and contemporaneously as mild sedatives, anxiolytics and hypnotics. Recent research has suggested
that both may attenuate laboratory induced stress. As the two herbs are most often sold in combination with
each other the current study assessed the anxiolytic properties of such a combination during laboratory-
induced stress.
Methods: In this double-blind, placebo-controlled, randomized, balanced cross-over experiment, 24 healthy
volunteers received three separate single doses (600 mg, 1200 mg, 1800 mg) of a standardized product contain-
ing M. officinalis and V. officinalis extracts, plus a placebo, on separate days separated by a 7 day wash out
period. Modulation of mood and anxiety were assessed during pre-dose and 1 h, 3 h and 6 h post-dose com-
pletions of a 20 min version of the Defined Intensity Stressor Simulation (DISS) battery. Cognitive perform-
ance on the four concurrent tasks of the battery was also assessed.
Results: The results showed that the 600 mg dose of the combination ameliorated the negative effects of the
DISS on ratings of anxiety. However, the highest dose (1800 mg) showed an increase in anxiety that was less
marked but which reached significance during one testing session. In addition, all three doses led to decre-
ments in performance on the Stroop task module within the battery, and the two lower doses led to decre-
ments on the overall score generated on the DISS battery.
Conclusions: These results suggest that a combination of Melissa officinalis and Valeriana officinalis
possesses anxiolytic properties that deserve further investigation. Copyright © 2006 John Wiley & Sons, Ltd.
Keywords: Melissa officinalis; Valeriana officinalis; stress; anxiety; lemonbalm; valerian; cognitive.
2002). Additionally a number of studies have reported
INTRODUCTION a benefit to sleep following a valerian/hops combina-
tion (e.g. Schmitz and Jackel 1998; Fussel et al., 2000).
The herbal species Melissa officinalis (Lemon Balm) It is also interesting to note that, whereas no effect
and Valeriana officinalis (Valerian) have both been in of valerian on psychomotor performance, alertness
use as pan-cultural medicinal treatments for more than or concentration has been reported the morning after
two millennia (Perry et al., 1999; Hobbs, 1989). They administration (Gerhard et al., 1996; Kuhlmann et al.,
share common traditional and contemporary indications 1999), acute doses of valerian and a valerian/hops com-
as mild sedatives, anxiolytics and hypnotics (Bisset and bination led to a modest, but statistically significant,
Wichtl, 1994; Blumenthal et al., 1998; Kommission E deterioration in performance in vigilance and informa-
Monograph, 1984). tion processing tasks 1–2 h following acute day-time
In the case of valerian, research in humans has tended administration (Gerhard et al., 1996). In contrast to this
to concentrate on its role in disturbed sleep. Stevinson latter finding, Gutierrez et al. (2004) reported no mood
and Ernst (2000) reviewed the 19 trials that assessed its or psychomotor effects of three single doses of a stand-
effects on sleep and insomnia that were extant by May ardized valerian extract (as opposed to significant effects
1999, and concluded from the nine trials which reached for 10 mg diazepam) in a small sample of 10 healthy
their inclusion criteria, that the efficacy of valerian in volunteers.
this respect was ‘promising but not fully conclusive’. The putative sedative and hypnotic effects of valerian
The potential for valerian in this regard has more may be attributable to a number of potentially active
recently been bolstered by a report of similar posi- constituents. These include a range of monoterpenes
tive effects of 6 weeks administration of 600 mg/day and sesquiterpenes, including the genus-specific valepo-
of valerian extract as that produced by 10 mg/day of triates and valerenic acid (Houghton, 1999). Specific
oxazepam on sleep quality and waking symptoms in GABAergic properties have also been demonstrated.
202 non-organic insomnia outpatients (Ziegler et al., Indeed, root extracts contain appreciable levels of
GABA (Houghton, 1999). Other specific components
* Correspondence to: David. O. Kennedy, Human Cognitive Neuroscience bind to various neurotransmitter receptors (Marder
Unit, Division of Psychology, University of Northumbria, Newcastle upon
Tyne NE1 8ST, UK.
et al., 2003) including GABAA (Wasowski et al., 2000),
E-mail: david.kennedy@unn.ac.uk 5HT1A (Bodesheim and Holzl, 1997) and adenosine A1
Contract/grant sponsor: Pharmaton SA, Lugano, Switzerland. receptors (Muller et al., 2002).
Copyright © 2006 John Wiley & Sons, Ltd. Received
Phytother. Res. 25 May(2006)
20, 96–102 2005
Accepted 18 August 2005
Copyright © 2006 John Wiley & Sons, Ltd.ATTENUATION OF STRESS BY MELISSA/VALERIAN 97 With regard to Melissa officinalis, several studies study, Kennedy et al. (2004) used double-blind, have also confirmed its mild sedative or anxiolytic placebocontrolled, counterbalanced cross-over meth- effects. In mice, behavioural parameters have been odology to assess the mood and performance effects of reduced following administration of volatile oils, iso- two separate doses (300 mg, 600 mg) of a methanol lated terpenes (Wagner and Sprinkmeyer, 1973) and a extract of M. officinalis during the performance of the hydroalcohol extract of M. officinalis (Soulimani et al., ‘Defined Intensity Stress Simulator’ (DISS) laboratory 1991). In humans, a double-blind, placebo-controlled stressor battery by 18 healthy young participants. Whilst study demonstrated a significant reduction in agitation the battery itself led to increased subjective ratings of and social withdrawal, and an increase in constructive alertness and reduced ratings of calmness, the 600 mg activities, during 4 weeks of aromatherapy with M. dose of the methanol extract led to a direct significant officinalis essential oil in a group of patients suffering attenuation of these mood effects during completion of from severe dementia (Ballard et al., 2002). Three stud- the stress inducing battery. ies have also assessed mood and cognitive performance Interestingly, whilst both species under investiga- following acute oral ingestion of a standardized com- tion here would seem to exert sedative or hypnotic mercial extract (Kennedy et al., 2002; 2004 – see below) effects in themselves they are more commonly sold and encapsulated dried leaf (Kennedy et al., 2003) of in combination with other products. One of the most M. officinalis by healthy young participants. In two of common such combinations is that of M. officinalis these studies (Kennedy et al., 2002, 2003), self-ratings and V. officinalis. Whilst such preparations are not of calmness were increased, with this effect apparent well researched, a valerian/melissa treatment has pre- for the lowest dose (300 mg) of methanol extract and viously been shown to improve the sleep quality the middle (1000 mg) and highest (1600 mg) doses of of healthy normal sleepers (Cerny and Schmid, 1999) dried leaf. The highest dose of extract (900 mg) also and to have a similar effect on sleep parameters led to significantly reduced alertness. in poor sleepers as 0.125 mg of triazolam (Dressing The mechanisms underlying the behavioural actions et al., 1992). of M. officinalis are poorly understood. It has been The current double-blind, placebo-controlled, bal- suggested that the active components include mono- anced cross-over study extended previous research terpenoid aldehydes, flavonoids, polyphenolic com- examining the stress reducing effects of both M. pounds including rosmarinic acid (Carnat et al., 1998) officinalis and V. officinalis with an investigation of their and monoterpene glycosides (Mulkens et al., 1985). combined effects on laboratory induced stress. The These components may well underlie a number of effects of single doses (600 mg, 1200 mg, 1800 mg) of a effects seen in vitro, which include potent antioxidant product combining the two extracts on mood, anxiety properties (Mantle et al., 2000; Hohmann et al., 1999) and cognitive performance were assessed whilst par- and an affinity for binding to both nicotinic and ticipants completed the defined intensity stressor simu- muscarinic cholinergic receptors in human brain cortex lation (DISS) computerized battery. The DISS has tissue (Wake et al., 2000; Kennedy et al., 2003). This previously been shown to increase negative ratings of latter mechanism is of specific interest as modulation mood and engender physiological responses concomi- of the cholinergic system may well be beneficial to cogni- tant with increased stress (Wetherell and Sidgreaves, tive function. In line with this, the acute dose studies 2005; Kennedy et al., 2004). described above demonstrated that the methanol ex- tract led to decrements on the more difficult timed tasks from within a computerized cognitive battery (Kennedy et al., 2002), whereas the dried leaf led to similar de- METHODS AND MATERIALS crements for lower doses, but also enhanced memory performance for the highest dose utilized (Kennedy Participants. Twelve female and 12 male under- et al., 2003). One potential explanation for these dif- graduate volunteers (mean age 23.48 years, SD 5.3 ferential cognitive effects is that the methanol extract years) took part in the study which was approved by was found to have negligible cholinergic activity, whilst the Joint Ethics Committee of Newcastle and North the dried leaf had been chosen specifically on the basis Tyneside Health Authority and was carried out in of its relatively high cholinergic receptor binding in accordance with the Declaration of Helsinki. Prior to human brain tissue. This does suggest that the common participation each volunteer signed an informed con- cognitive decrements evinced by both, and potentially sent form and completed a medical health question- the mood effects, are underpinned by another, as yet naire. All participants reported that they were in good unidentified, mechanism. health, did not suffer from any diagnosed medical Separate single studies have also assessed the effects condition, and were taking no illicit social drugs. Addi- of both V. officinalis and M. officinalis on laboratory tionally they were free of any ‘over the counter’ or induced stress. Cropley et al., (2002) assessed the prescribed medications, with the exception, for some effects, in comparison with a control group, of both female volunteers, of the contraceptive pill. Smokers V. officinalis and Kava (n = 18 per group) on changes were excluded from the study. All participants abstained in blood pressure, heart rate and subjective feelings from alcohol for a minimum of 12 h prior to the first of ‘pressure’ due to completion of a mentally stressing testing session of the morning. task. Both active treatments led to reduced systolic blood pressure and ratings of subjective ‘pressure’, whilst Treatments. The coated tablets utilized in the study valerian also mitigated the increase in heart rate during contained either 120 mg of Valeriana officinalis extract the task. While these results are promising, the lack of (4.5:1) and 80 mg of Melissa officinalis extract (5:1) a placebo group means that they should be interpreted (Songha Night™, Pharmaton Natural Health Products, with some caution. Of particular relevance to the present Lugano), or an inactive placebo. Copyright © 2006 John Wiley & Sons, Ltd. Phytother. Res. 20, 96–102 (2006)
98 D. O. KENNEDY ET AL.
On each study day the participants received nine Stroop colour-word test. Words describing one of four
capsules of identical appearance, each containing either colours (‘RED’, ‘YELLOW’, ‘GREEN’, ‘BLUE’) are
the active treatment or placebo. Depending on the con- presented in different coloured fonts on the left of the
dition to which they were allocated on that particular module area. The participant clicks on colour panels
day the combination corresponded to a dose of either on the right of the module area in order to identify the
0 mg (placebo), 600 mg, 1200 mg or 1800 mg of the font colour (e.g. if the word ‘GREEN’ is presented in a
combination. blue font, the correct response would be to click on the
blue panel). Ten points are awarded for every correctly
The defined intensity stressor simulation (DISS) com- identified colour, with ten points deducted for each
puterized battery. The DISS computerized battery com- incorrect answer, or for not responding in the allotted
prises a set of four concurrent cognitive and psychomotor time.
tasks presented via a split screen. All responses are
made with an external mouse. In this instance a 20 min Highest number tap. A grid containing digits between 0
version of the DISS was employed. The modules and 9 appear. The participant identifies the numerically
selected were the ‘mathematical processing’, ‘Stroop’, highest digit and then clicks on all of the instances of the
‘high tap’ and ‘memory search’ tasks. Participants are digit presented on the screen in order to activate them.
instructed via on screen standard instructions to attend Once all instances of the highest digit have been suc-
simultaneously to all four tasks, whilst monitoring the cessfully activated the grid is replaced. Ten points are
central counter displaying their accumulated aggregate awarded for each grid cleared, with ten points deducted
score. Accuracy and speed of response dictate the score, for any grids not fully activated within the time limit.
with failure to respond resulting in negative scoring.
Previous research has shown that the DISS engenders Memory search task. Four letters appear for the par-
increases in self-ratings of negative mood, arousal and ticipants to remember. After 4 s the letters disappear
stress related physiological responses (Wetherell and but can be viewed again by clicking on the ‘retrieve
Sidgreaves, 2005; Kennedy et al., 2004). list’ button. Approximately every 10 s a single target
In the current study the ‘Stroop’ and ‘highest number letter appears. Participants are instructed to indicate
tap’ tasks were set at a high difficulty/intensity level whether the target letter had appeared in the original
and the ‘mathematical processing’ and ‘memory search’ list of four letters by clicking on the ‘yes’ or ‘no’ buttons.
tasks were set at a medium difficulty/intensity level. Ten points are awarded for a correct answer, ten points
The battery was performed for 20 min. The on-screen deducted for an incorrect answer or no response, and
layout of the battery is shown in Fig. 1. The four tasks five points are deducted every time the list is retrieved.
are described below.
Bond-Lader visual analogue mood scales (Bond and
Mathematical processing task. A series of calculations Lader, 1974). Mood was assessed before and after
are presented. The participant adds two numbers, each completion of the DISS battery using the 16 visual
entering the three figure answer via an onscreen number analogue scales of Bond-Lader, which were combined
pad. On completion of each calculation, the participant as recommended by the authors to form three mood
clicks on the ‘done’ button which cues the next calcula- factor scores: ‘alert’, ‘content’ and ‘calm’. Factor scores
tion. Ten points are awarded for each correct answer can vary between 0 and 100 (average millimetres along
and ten points deducted from the running total for each 100 mm visual analogue line). For each completion of
incorrect answer. the stress battery the pre-DISS factor scores were sub-
tracted from the post-DISS factor scores to give single
scores representing the change in mood engendered by
completion of the DISS battery.
State trait anxiety inventory (STAI) (Speilberger et al.,
1969). Participants’ level of anxiety was assessed before
and after each completion of the DISS using the STAI.
The STAI is a widely used instrument consisting
of two subscales assessing ‘State’ and ‘Trait’ anxiety
respectively. Each subscale contains 20 statements (e.g.
‘I am calm’) each with a 4-point Likert scale, giving a
range of potential scores from 20 to 80. Participants
rate how much they feel like each statement at the time
of making the response (State subscale), and how much
they generally feel like each statement (Trait subscale).
Higher scores indicate greater anxiety. For each com-
pletion of the stress battery the pre-DISS factor scores
were subtracted from the post-DISS factor scores to
give single scores representing the change in mood
engendered by completion of the DISS battery.
Figure 1. The on-screen layout of the defined intensity stressor
simulation battery. The concurrent tasks comprise (clockwise Procedure. Each participant was required to attend a
from top left) ‘mathematical processing’, ‘highest number tap’,
‘memory search’ and ‘Stroop colour–word’. The participant’s
total of 5 study days that were conducted 7 days apart,
aggregate score for the four tasks is shown at the centre of the to ensure a sufficient wash-out between conditions.
display. All responses are made via a standard computer mouse. Testing took place, commencing at the same time on
Copyright © 2006 John Wiley & Sons, Ltd. Phytother. Res. 20, 96–102 (2006)ATTENUATION OF STRESS BY MELISSA/VALERIAN 99
each day, in a suite of laboratories with participants reduction in ‘calmness’ as assessed by the Bond-Lader
visually isolated from each other. scales [F (1,23) = 4.76, p = 0.04]. There was no significant
On arrival at their first session on the first day par- effect on ratings of either ‘contentedness’ or ‘alertness’.
ticipants were randomly allocated to a treatment regime Completion of the DISS also led to increased anxiety
using a Latin square design which counterbalanced the as assessed by the STAI ‘state’ scale [F (1,23) = 9.63,
order of treatments across the 4 active days of the study. p = 0.005], and a trend towards the same effect on the
The first day was identical to the following four, ex- ‘trait’ scale [F (1,23) = 3.72, p = 0.07].
cept that no treatment (active or placebo) was offered,
to allow familiarization with the test battery and pro-
cedure. Data from the four sessions of this practice day Treatment effects on mood during the DISS
were not included in any analysis.
Immediately prior to and following each completion Bond-Lader scales. The initial ANOVA suggested that
of the DISS battery participants filled out Bond-Lader scores on the ‘Alert’, ‘Content’, and ‘Calm’ factors were
mood scales and the STAI. not significantly affected by the treatments.
Each day of the study comprised an initial pre-dose
completion of the 20-min DISS battery (plus mood/ STAI state anxiety. Planned comparisons indicated that
anxiety scales before and after), followed by ingestion of the 600 mg dose of the combination led to an attenua-
the day’s treatment. At 1 h post-dose, 3 h post-dose and tion of the anxiety inculcated by the DISS at the 3 h
6 h post-dose participants made further completions of [t (138) = 2.28, p = 0.02] and 6 h [t (138) = 2.21, p = 0.03]
the DISS battery (plus mood scales before and after). testing sessions. In contrast to this, the highest dose
(1800 mg) led to increased ratings of anxiety through-
Statistics: mood effects of DISS. Prior to the analysis of out the post-dose period, with this effect reaching
treatment effects (see below) the effects of the DISS on significance at the 1 h testing session [t (138) = 2.6, p =
mood in the absence of any treatment were assessed utilis- 0.01], with a trend towards the same at 6 h [t (138) =
ing a two-way repeated measures ANOVA (Condition X 1.71, p = 0.09].
pre-post DISS) of pre-dose data from all four conditions.
STAI trait anxiety. The 600 mg dose also led to an
Statistics: Treatment effects on mood and DISS per- attenuation of anxiety on the trait subscale during the
formance. Scores for the treatment related change in 1 h testing session [t (138) = 2.07, p = 0.04] with a trend
STAI and Bond-Lader scores during DISS completion towards the same effect during the 3 h completion of
and the individual DISS task performance measures the DISS [t (138) = 1.93, p = 0.06].
were analysed as ‘change from baseline’ (i.e. change Baseline and change from baseline scores for the
during 1 hr, 3 hr and 6 hr post-dose DISS minus change Bond-Lader and STAI scales, with a graphical repres-
during baseline DISS). Prior to carrying out planned entation of the measures that reached significance, are
comparisons, an ANOVA (General Linear Model), with shown in Fig. 2.
terms fitted to the model for dose, visit, dose × visit and
subject (Kirk, 1968), was carried out to identify main
effects and interaction effects on change from baseline Treatment effects on performance of the DISS
data for each measure. The primary statistical analysis
of the ‘change from baseline’ data for each measure Memory search. There was no significant effect on
was undertaken using planned comparisons, utilizing memory search performance.
t-tests with the MSError from an omnibus ANOVA as
an error term (Keppel, 1991). For each measure data Maths. There was no significant effect on memory search
from the placebo condition were compared to that for performance.
each of the three doses of the melissa/valerian com-
bination (600 mg, 1200 mg, 1800 mg). To ensure the High tap. There was no significant effect on memory
overall Type I error protection level only those planned search performance.
comparisons associated with measures that generated a
significant main effect or interaction effect or trend Stroop. Performance on the Stroop task was signific-
towards the same on the initial ANOVA are reported. antly disturbed by all three treatments. This effect was
Furthermore, all testing was two-tailed, comparisons evident at the 3 h [t (138) = 2.47, p = 0.02] and 6 h
were strictly planned prior to the study, were restricted [t (138) = 2.53, p = 0.01] time-points for the lowest
to the number of conditions minus one at each time- dose, with a trend towards the same at 1 h [t (138) =
point, and only probabilities associated with these pre- 1.86, p = 0.07]; at all time points for the middle dose
planned comparisons were calculated. (1 h [t (138) = 2.39, p = 0.02], 3 h [t (138) = 2.44, p =
0.02], 6 h [t (138) = 4.06, p < 0.001]) and at a single time
point for the highest dose (1 h [t (138) = 3.26, p = 0.001])
with a trend towards the same effect at 3 h [t (138) =
RESULTS 1.71, p = 0.09].
Effect of DISS on mood and anxiety (in the absence Total DISS score. Reduced performance across the
of treatment) whole battery reached significance for both the 600 mg
[t (138) = 2.17, p = 0.03] and 1200 mg [t (138) = 3.62,
Analysis of variance of the pre-dose baseline data p < 0.001] doses at the 1 h testing session and at the
from all four conditions showed that completion of the 6 h testing session for the latter dose [t (138) = 2.17,
DISS in the absence of treatment led to a significant p = 0.03].
Copyright © 2006 John Wiley & Sons, Ltd. Phytother. Res. 20, 96–102 (2006)100 D. O. KENNEDY ET AL.
Figure 2. Effects of melissa/valerian combination on the change in Bond-Lader and State-Trait Anxiety Inventory scores during
completion of the DISS battery (in comparison to change during pre-dose baseline DISS). Factor scores from the Bond-Lader scales
can vary between 0 and 100, and STAI subscale scores can vary between 20 and 80. Graphs represent the measures that reached
significance on the initial ANOVA (t, p < 0.1, *, p < 0.05, **, p < 0.001 on planned comparisons).
Figure 3. Effects of the melissa/valerian combination on the total score and module scores from the DISS battery. Graphs represent
the measures that reached significance or a trend on the initial ANOVA (t, p < 0.1, *, p < 0.05, ***, p < 0.001 on planned comparisons).
Baseline and change from baseline scores for the colour–word task, and following the two lower doses
DISS measures, with a graphical representation of the on the overall aggregate battery score.
measures that reached significance, are shown in Fig. 3. With regards the modulation of mood, whilst the
results for the lowest and highest dose appear to be
somewhat contradictory, it must be borne in mind that
the lowest (600 mg) dose represents the highest daily
DISCUSSION dose recommended by the manufacturers (i.e. two or
three 200 mg tablets), and is equal to the dose of the
The most notable finding of the current study was that same product that Cerny and Schmid (1999) found to
the 600 mg dose of the melissa/valerian combination improve sleep parameters in healthy volunteers. The
led to decreases in task related anxiety throughout the negative mood effects evinced here for the highest dose
post-dose testing sessions, with this effect apparent at were only seen following three times this ‘normal’ dose.
1 h post-dose on the ‘trait’ subscale of the STAI and at Interestingly, previous in vitro research has suggested
3 and 6 h on the ‘state’ subscale. This modulation of dose-dependent biphasic interactions between valerian
mood represents a direct attenuation of the negative and both GABA receptors and synaptosomal GABA
anxiety effects associated with completing the DISS uptake (Ortiz et al., 1999). The authors attribute this
battery. In contrast to this, the highest dose of the com- to the presence of at least two different biological act-
bination (1800 mg) led to (smaller) increases in anxiety ivities. Effects such as these could certainly account for
across all three post-dose sessions, with this effect the differing pattern of mood effects seen with lower
reaching significance at a single time point on the ‘state’ and higher doses. It is also worth noting that previous
sub-scale. research assessing the cognitive and mood effects of
DISS performance was also modulated, with decre- the same extract of M. officinalis in isolation (Kennedy
ments seen following all three doses on the Stroop et al., 2002) utilized a not dissimilar range of doses
Copyright © 2006 John Wiley & Sons, Ltd. Phytother. Res. 20, 96–102 (2006)ATTENUATION OF STRESS BY MELISSA/VALERIAN 101
(300 mg, 600 mg and 900 mg compared with a melissa activities, it is difficult to delineate any underlying
component of 240 mg, 480 mg and 720 mg here) and mechanism(s) with any confidence. Certainly the anxio-
demonstrated markedly different mood effects for each. lytic effects evinced here for the lowest dose (and the
In that case the lowest dose led to increased calmness, opposite effects for the highest) could well be as a re-
the middle dose had no effect and highest dose led to sult of GABA-ergic modulation, for which there is
reduced alertness. mounting evidence with regards valerian (Houghton
Interestingly, both completion of the DISS (with only 1999; Marder et al., 2003; Wasowski et al., 2002; Ortiz
a trend for ‘trait’) and the administration of valerian/ et al., 1999). However, they could equally well be as a
melissa modified anxiety as assessed by both the ‘state’ consequence of serotonergic (Bodesheim and Holzl,
and ‘trait’ subscales of the STAI. Whilst it would seem 1997) or adenosinergic activity (Muller et al., 2002),
more likely that both would have their impact on the or as a result of any number of as yet undelineated
more malleable ‘state’ subscale, it is noteworthy that actions. With regards mechanisms, M. officinalis is of
the authors of the instrument (Speilberger et al., 1969) particular interest, as it has been shown to exert marked
intended both subscales to assess the long-term and behavioural and mood effects in the presence (Kennedy
immediate effects of interventions (including medic- et al., 2003) and absence (Kennedy et al., 2002) of nico-
inal). Given the exploratory nature of this study it is tinic and muscarinic receptor binding properties. Whilst
difficult to assess the relevance of specific modulation this latter mechanism has been suggested to underlie
of the ‘trait’ subscale. improved memory following administration of the dried
The mild decrements seen in task performance are leaf (Kennedy et al., 2003), this does still leave a plethora
also not out of keeping either with the actions of a mild of mood and cognitive effects as yet unattached to any
anxiolytic/sedative or with past research into both treat- identified mechanism.
ments. Whilst the cognitive effects of valerian are little It is also notable that whilst participants were not
researched outside of a ‘morning after’ context, a single required to complete side effect questionnaires for
study (Gerhard et al., 1996) assessed daytime acute methodological reasons, they were asked to verbally
effects and found a slight, but significant, deterioration report any instances to the investigators both during
in performance in vigilance and information processing and at the end of the day. No deleterious side effects
tasks 1–2 h following administration. Similarly, both the were reported. Similarly, an analysis of the pre-DISS
previous studies of the cognitive effects of M. officinalis mood data throughout the day (not shown) suggested
(Kennedy et al., 2002; 2003) demonstrated decrements that the treatments did not have any impact on any
for all doses on several of the more difficult timed tasks measure of mood (including alertness, calmness and
from within the computerized battery that was utilized. levels of anxiety) in the absence of the stressing effects
These effects are in keeping with a decrement on Stroop of the DISS. However, this latter finding should also be
performance. Although this task is self-paced, it is viewed on the basis that the cohort was one of healthy
associated with a high processing load. Indeed, in the undergraduates, who were not suffering pathological
current colour-word incongruence form, the Stroop task levels of ‘day to day’ stress.
is widely used as a psychosocial stressor. It is a measure Whilst this study does provide a first demonstration
of selective attention since the task requires attention of anxiolytic effects for this combination in the face of
to one feature (colour) while ignoring competing a laboratory stressor, which is in line with similar
features (word) (Scholey, 2002). In previous studies findings for both valerian (Cropley et al., 2002) and
utilizing Melissa officionalis alone, improvement in melissa (Kennedy et al., 2004), it did not address
attentional accuracy in one study (Kennedy et al., 2002) the treatment’s concomitant effects on physiological
and no effect on attention in a second (Kennedy et al., indices of stress. A useful next step might be to
2003) were found. It seems possible therefore that extend this line of enquiry with the inclusion of
the effects observed here on performance might be heart rate, blood pressure and salivary cortisol
attributable to the addition of valerian to the prepara- measures.
tion used. Further, well-controlled studies into the Given that neither valerian nor melissa are associated
behavioural effects of valerian alone are clearly needed with the deleterious side effects of currently prescribed
before drawing any conclusions. anxiolytics, this study does suggest that there might be
Naturally, as the treatment used here consisted some merit in further investigating the potential for
of two separate herbal extracts, both of which can these herbal products to provide safe, side effect free
be assumed to exert a range of specific physiological relief of stress.
REFERENCES
Ballard C, O’Brien J, Reichelt K, Perry E. 2002. Aromatherapy Bond A, Lader M. 1974. The use of analogue scales in rating
as a safe and effective treatment for the management of subjective feelings. Br J Psychol 47: 211–218.
agitation in severe dementia: The results of a double blind, Carnat AP, Carnat A, Fraisse D, Lamaison JL. 1998. The aro-
placebo controlled trial. J Clin Psychiatry 63: 553–558. matic and polyphenolic composition of lemon balm (Melissa
Bisset NG, Wichtl M. 1994. Herbal Drugs. Stuttgart: Medpharm. officinalis L. subsp. officinalis) tea. Pharm Acta Helv 72:
Blumenthal M, Gruenwald J, Hall T (eds). 1998. The Complete 301–305.
German Commission E Monographs: Therapeutic Guide to Cerny A, Schmid K. 1999. Tolerability and efficacy of valerian
Herbal Medicines. Integrative Medicine Communications: and lemon balm in healthy volunteers a double-blind,
Boston, MA. placebo-controlled, multicentre study Fitoterapia 70: 221–
Bodesheim U, Holzl J. 1997. Isolierung, Strukturaufkla¨rung und 228.
Radioreceptor-assays von Alkaloiden und Lignan aus Cropley M, Cave Z, Ellis J, Middleton RW. 2002. Effect of kava
Valeriana officinalis L. Pharmazie 52: 386–391. and valerian on human physiological and psychological
Copyright © 2006 John Wiley & Sons, Ltd. Phytother. Res. 20, 96–102 (2006)102 D. O. KENNEDY ET AL.
responses to mental stress assessed under laboratory valeriana flavonoids with activity on the CNS. Pharmacol
conditions. Phytother Res 16: 23–27. Biochem Behav 75: 537–545.
Dressing H, Riemann D, Löw H, Schredl M, Reh C, Laux P, Mulkens A, Stephanou E, Kapetenadis I. 1985. Heterosides a
Müller WE. 1992. Insomnia: are valerian/balm combinations genines volatiles dans les feuilles de Melissa officinalis L.
of equal value to benzodiazepine? Therapiewoche. 42: (lamiaceae). Pharm Acta Helv 60: 276–278.
726 –736. Muller CE, Schumacher B, Brattstrom A, Abourashed EA, Koetter
Fussel A, Wolf A, Brattstrom A. 2000. Effect of a fixed valerian- U. 2002. Interactions of valerian extracts and a fixed valerian–
Hop extract combination (Ze 91019) on sleep polygraphy hop extract combination with adenosine receptors. Life Sci
in patients with non-organic insomnia: a pilot study. Eur J 71: 1939–1949.
Med Res 5: 385–390. Ortiz JG, Nieves-Natal J, Chavez P. 1999. Effects of
Gerhard U, Linnenbrink N, Georghiadou C, Hobi V. 1996. Valeriana officinalis extracts on [3H]flunitrazepam binding,
Vigilanzmindernde Effekte zweier pflanzlicher Schlafmittel synaptosomal [3H]GABA uptake, and hippocampal
[Vigilance-decreasing effects of 2 plant-derived sedatives]. [3H]GABA release. Neurochem Res 24: 1373–1378.
Revue Suisse Medecine Praxis 85: 473–481. Perry EK, Pickering AT, Wang WW, Houghton PJ, Perry NSL.
Gutierrez S, Ang-Lee MK, Walker DJ, Zacny JP. 2004. Assessing 1999. Medicinal plants and Alzheimer’s disease: from
subjective and psychomotor effects of the herbal medica- ethnobotany to phytotherapy. J Pharm Pharmacol 51: 527–
tion valerian in healthy volunteers. Pharmacol Biochem 534.
Behav 78: 57–64. Schmitz M, Jackel M. 1998. Comparative study for assess-
Hobbs C. 1989. Valerian: a literature review. HerbalGram 21: ing quality of life of patients with exogenous sleep dis-
19–34. orders (temporary sleep onset and sleep interruption
Hohmann J, Zupko I, Redei D et al. 1999. Protective effects disorders) treated with a hops-valerian preparation and
of the aerial parts of Salvia officinalis, Melissa officinalis a benzodiazepine drug. Wien Med Wochenschr. 148(13):
and Lavandula angustifolia and their constituents against 291–298.
enzyme-dependent and enzyme independent lipid Scholey AB. 2002. Attention. In Perry EK, Ashton H, Young AH
peroxidation. Planta Med 65: 576–578. (Eds) Neurochemistry of Consciousness. Benjamins: Am-
Houghton PJ. 1999. The scientific basis for the reputed activity sterdam, 43–63.
of Valerian. J Pharm Pharmacol 51: 505–512. Souliman R, Fleurentin J, Mortier F, Misslin R, Derrieu G,
Kennedy DO, Little W, Scholey AB. 2004. Attenuation of labora- Pelt JM. 1991. Neurotropic action of the hydroalcoholic
tory induced stress in humans following acute administra- extract of Melissa officinalis in the mouse. Planta Med 57:
tion of Melissa officinalis (Lemon Balm). Psychosom Med 105–109.
66: 607–613. Speilberger CD, Gorsuch RL, Lushene RE. 1969. The State Trait
Kennedy DO, Scholey AB, Tildesley NTJ, Perry EK, Wesnes KA. Anxiety Inventory Manual. Consulting Psychologists Press:
2002. Modulation of mood and cognitive performance Palo Alto.
following acute administration of Melissa officinalis (lemon Stevinson C, Ernst E. 2000. Valerian for insomnia: a systematic
balm). Pharmacol Biochem Behav 72: 953–964. review of randomized clinical trials. Sleep Med 1: 91–99.
Kennedy DO, Wake G, Savelev S et al. 2003. Modulation of Wagner H, Sprinkmeyer L. 1973. Uber die pharmakologische
mood and cognitive performance following administration Wirkung von Melissengeist. Dtsch Apoth Ztg 113: 1159–
of single doses of Melissa officinalis (Lemon balm) with 1166.
human CNS nicotinic and muscarinic receptor binding Wake G, Court J, Pickering A, Lewis R, Wilkins R, Perry E. 2000.
properties. Neuropsychopharmacology 28: 1871–1881. CNS acetylcholine receptor activity in European medicinal
Keppel G. 1991. Design and analysis. A Researcher’s Hand- plants traditionally used to improve failing memory. J
book. Prentice Hall, Upper Saddle River, New Jersey. Ethnopharmacol 69: 105–114.
Kirk RE. 1968. Experimental design: procedures for the behav- Wasowski C, Marder M, Viola H, Medina JH, Paladini AC. 2002.
ioural sciences. Brooks/Cole, Belmont, California. Isolation and identification of 6-methylapigenin, a competi-
Kommission E Monograph. Melissenbla’tter, Bundesanzeiger, tive ligand for the brain GABAA receptors, from Valeriana
1984, 05.12. wallichii D.C. Planta Med 68: 934–936.
Kuhlmann J, Berger W, Podzuweit H, Schmidt U. 1999. The Wetherell MA, Sidgreaves MC. 2005. Secretory immunoglobulin-
influence of valerian treatment on ‘reaction time, alertness A reactivity following increases in workload intensity using
and concentration’ in volunteers. Pharmacopsychiatry 32: the Defined Intensity Stress Simulator. Stress Health 21:
235–241. 99–106.
Mantle D, Eddeb F, Pickering AT. 2000. Comparison of relative Ziegler G, Ploch M, Miettinen-Baumann A, Collet W. 2002.
antioxidant activities of British medicinal plant species in Efficacy and tolerability of valerian extract LI 156 compared
vitro. J Ethnopharmacol 72: 47–51. with oxazepam in the treatment of non-organic insomnia –
Marder M, Viola H, Wasowski C, Fernandez S, Alejandro JH, a randomized, double-blind, comparative clinical study. Eur
Paladini MC. 2003. 6-Methylapigenin and hesperidin: new J Med Res 7: 480–486.
Copyright © 2006 John Wiley & Sons, Ltd. Phytother. Res. 20, 96–102 (2006)You can also read
