Content uploaded by Eva Heuberger
Author content
All content in this area was uploaded by Eva Heuberger
Content may be subject to copyright.
Introduction
The essential oil of East Indian Sandalwood (Santalum album L.,
Santalaceae) contains more than 90% of sesquiterpenic alcohols,
the santalols. The major component, with approximately 50±
60 %, among these alcohols is the tricyclic
a
-santalol.
b
-Santalol
makes up for about 20±25 % [1]. In traditional Oriental medicine
it is utilized for the treatment of various somatic disorders and
psychic syndromes. Recently, Sandalwood oil and
a
-santalol have
been associated with chemopreventive activity in animal models
of carcinogenesis [2], [3], [4]. In addition, antiviral [5], [6] and an-
ticandidal effects of Sandalwood oil [7] have been demonstrated.
It has long been observed that inhalation of essential oils may
cause physiological and behavioral changes in humans [8]. Clini-
cal evidence in aromatherapy suggests that beneficial effects are
also exerted by application of fragrances to the skin. It is assumed
that the effects of fragrances are elicited by psychological and
pharmacological mechanisms, i. e., via the sense of smell and
due to the uptake of fragrance molecules into the organism [9].
In aromatherapy, Sandalwood oil is deemed aphrodisiac, antide-
pressant, relaxing, and sedative. Buchbauer et al. [10] reported
that exposure to Sandalwood vapor reduced the motility of
mice. Recent investigations [11], [12] have shown that inhalation
of as well as massage with Sandalwood oil have an excellent
East Indian Sandalwood and
aa
-Santalol Odor Increase
Physiological and Self-Rated Arousal in Humans
Eva Heuberger
1, 2
Tapanee Hongratanaworakit
3
Gerhard Buchbauer
1
Affiliation
1
Department of Clinical Pharmacy and Diagnostics, Faculty of Life Sciences, University of Vienna, Vienna,
Austria
2
Smell and Taste Center, Department of Otorhinolaryngology ± Head and Neck Surgery, School of Medicine,
University of Pennsylvania, Philadelphia, USA
3
Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Srinakharinwirot University, Nakorn-nayok,
Thailand
Correspondence
Dr. Eva Heuberger ´ Department of Clinical Pharmacy and Diagnostics ´ University of Vienna ´ Althanstr. 14 ´
1090 Vienna ´ Austria ´ Phone: +43-1-4277-55560 ´ Fax: +43-1-4277-9555 ´
E-mail: eva.heuberger@univie.ac.at
Received January 20, 2006 ´ Accepted April 25, 2006
Bibliography
Planta Med 2006; 72: 792±800 Georg Thieme Verlag KG Stuttgart ´ New York
DOI 10.1055/s-2006-941544 ´ Published online June 19, 2006
ISSN 0032-0943
Abstract
In Ayurvedic medicine, East Indian Sandalwood is an important
remedy for the treatment of both somatic and mental disorders.
In this investigation, the effects of inhalation of East Indian San-
dalwood essential oil and its main compound,
a
-santalol, on hu-
man physiological parameters (blood oxygen saturation, respira-
tion rate, eye-blink rate, pulse rate, skin conductance, skin tem-
perature, surface electromyogram, and blood pressure) and self-
ratings of arousal (alertness, attentiveness, calmness, mood,
relaxation and vigor) were studied in healthy volunteers. Com-
pared to either an odorless placebo or
a
-santalol, Sandalwood
oil elevated pulse rate, skin conductance level, and systolic blood
pressure.
a
-Santalol, however, elicited higher ratings of atten-
tiveness and mood than did Sandalwood oil or the placebo. Cor-
relation analyses revealed that these effects are mainly due to
perceived odor quality. The results suggest a relation between
differences in perceived odor quality and differences in arousal
level.
Key words
Arousal ´ East Indian Sandalwood oil ´
a
-santalol ´ inhalation ´
physiological monitoring ´ odor quality
Original Paper
792
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
sleep-inducing effect on rats and mice which might be triggered
by
a
- and
b
-santalol since these alcohols have proven to be neu-
roleptic agents with a similar pharmacological activity as chlor-
promazine [13]. After percutaneous absorption, deactivating ef-
fects of East Indian Sandalwood oil and
a
-santalol were demon-
strated on human physiological and behavioral measures [14].
However, it remains to be shown whether these effects are also
elicited in humans when these fragrances are inhaled.
Therefore, in the present study the effects of inhalation of East
Indian Sandalwood oil and
a
-santalol on parameters of the peri-
pheral nervous system as well as on subjective feelings of arousal
were investigated. We aimed at assessing the impact of these
odors on the level of physiological arousal and on subjective
behavioral activation of healthy human subjects and at elucidat-
ing the mechanisms underlying these effects.
Materials and Methods
East Indian Sandalwood essential oil (SEO) and
a
-santalol were
used as fragrances in this study.
a
-Santalol was preferred over
b
-santalol since, from a pharmacodynamic viewpoint, we ex-
pected it to be the most effective constituent among the santa-
lols. SEO was obtained from Dragoco GmbH, Vienna, Austria
(product No. 16589). A voucher specimen (No. 4/942130) is
stored at 48C at the Department of Clinical Pharmacy and Diag-
nostics.
a
-Santalol was separated and isolated as described in [14].
SEO (containing 41.1 %
a
-santalol, [15]),
a
-santalol (GC-MS pur-
ity), and a placebo substance, i. e., odorless water, were adminis-
tered via a fragrance delivery system developed by our group.
The apparatus consists of an air tank, a drug nebulization device
(Inhalette
; Dräger; Vienna, Austria) and a breathing mask (B+P
Beatmungsprodukte GmbH; Neunkirchen, Germany). The nebu-
lizer was filled with 250 mg of fragrance or water and attached to
the system so that air (12 L/min) streamed over the surface of the
substance in the nebulizer. The so generated vapors were direct-
ed to the breathing mask and inhaled by the subject. All sub-
stances were administered to the subjects throughout 20 min-
utes. The amount of SEO and
a
-santalol, respectively, which had
evaporated during this period (ªconsumptionº), was calculated
by subtracting the mass of the nebulizer after inhalation from
the mass before inhalation.
Thirty-six young, healthy human subjects (18 males and 18 fe-
males; mean age 23.77 2.73 years) who were recruited by ad-
vertisement at the University of Vienna and paid for their parti-
cipation were randomly assigned to a control group or one of 2
experimental groups, i.e., the SEO group or the
a
-santalol group.
Finally, each group consisted of 12 subjects (6 males and 6 fe-
males). At 48 hours prior to testing subjects had to abstain from
food, beverages and toiletries containing SEO or santalols, as well
as from tea, coffee, and alcohol. Subjects gave written informed
consent to all aspects of the study (Viennese ethic commission's
permissions No. 324/96, 419/98) and were free to withdraw at
any time.
Subjects were tested in individual sessions. One session consist-
ed of two trials of 20 minutes. At the beginning as well as at the
end of each trial, subjective feelings of arousal, i. e., relaxation,
vigor, calmness, attentiveness, mood, and alertness, were asses-
sed by means of visual analogue scales (VAS). Subjective ratings
of the odor of the substance presented, i.e., pleasantness, inten-
sity, and effect, were measured on a VAS at the end of each trial.
Seven physiological parameters, i. e., blood oxygen saturation
(BOS), eye-blinks (EB), pulse rate (PR), respiration rate (RR), skin
conductance (SC), skin temperature (ST), and surface electro-
myogram (EMG), were recorded continuously during each trial
using MP100WSW hardware (Biopac Systems, Inc.; Santa Bar-
bara, CA, USA) and AcqKnowledge
software (V3.2.6, 1992±
1997; Biopac Systems, Inc.; Santa Barbara, CA, USA). In addition,
systolic (SBP) and diastolic blood pressure (DBP) were measured
at the beginning as well as at the end of each trial. Details on the
recording equipment, electrode positioning, and VAS are given
elsewhere [16], [17]. In the first trial, which served as an individ-
ual baseline, odorless water was administered to all subjects. In
the second trial, subjects in the control group again received the
placebo, whereas the appropriate fragrance was administered in
the experimental groups.
Procedure
The experiments were carried out in a bright and quiet room.
Ambient temperature was 21 ± 24 8C. Subjects were seated in a
semi-reclined position. They were told that a fragrance would
be presented, but were unaware of the kind of fragrance and the
exact time of administration.
All subjects were interviewed about their name, age, weight and
height, and about their dominant hand by means of the Edin-
burgh Handedness Inventory [18]. After the proceedings had
been explained to them, subjects rated their feelings of arousal
on the VAS. Subsequently, SBP and DBP were measured. Electro-
des were attached at suitable positions. The breathing mask was
fixed to the subject's face, comfortably covering nose and mouth
and allowing unhindered respiration. Subsequently, recording of
the physiological parameters was started. After completion of
the first trial subjects were allowed to remove the breathing
mask and VAS for subjective arousal and odor ratings were pres-
ented. SBP and DBP were measured at the end of the first trial. In
the second trial, which followed consecutively after the first trial,
this procedure was repeated.
Data reduction
The physiological recordings of each subject were computed
using AcqKnowledge
software. Mean values of ST, level of SC
(SCL), PR, and BOS within each trial were obtained for each sub-
ject. Mean values of RR and eye-blink rate (EBR) were deter-
mined by counting the number of inspirations and EB per min-
ute, respectively, and calculating the arithmetic mean for each
trial. The raw EMG signals were converted to the root-mean-
square (rms) values, and mean rms values were obtained. For
every subject and every parameter the mean value in the second
trial was subtracted from the mean value in the first trial. Differ-
ences between blood pressure measurements 2 and 4, and 3 and
4, respectively, were calculated for each subject.
Feelings of arousal as well as fragrance ratings were processed by
measuring the distance of the mark from the left-hand side in
mm on each scale. Individual differences between the second
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
793
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
rating and the fourth rating as well as between the third rating
and the fourth rating were calculated for each item. For the fra-
grance ratings, individual differences between the first and the
second rating were computed.
Statistical analysis
Systat 9.0 (SPSS Inc., 1999) was used for statistical analyses. The
effects of SEO and
a
-santalol on the physiological parameters
and feelings of arousal were determined by comparing the differ-
ences between the control group and the experimental groups by
means of Mann-Whitney-U-Test and Kruskal-Wallis one-way
analysis of variance. Furthermore, the differences of the SEO
group were compared to those of the
a
-santalol group in order
to detect differences in the effectiveness of the fragrances.
To test for the overall influence of the fragrances on the physiol-
ogical arousal level, the 11 two-sided probabilities P resulting
from the univariate between-group comparisons were combined
by means of the transformation given below [19, pp. 46 ± 55], and
the resulting
c
2
-value was tested for significance:
The same analysis was performed on the 12 univariate between-
group comparisons of the parameters of subjective arousal to de-
termine the effectiveness of SEO and
a
-santalol at the behavioral
arousal level.
Spearman rank order correlation analyses were performed
within each group on the inter-trial differences of the physiol-
ogical parameters, the differences of SBP and DBP between the
second and the fourth measurement, the differences between
the second and the fourth rating of subjective arousal, and the
differences between the first and the second odor rating. With a
sample size of N = 12, Spearman's
r³
|0.591| is considered sta-
tistically significant at the 0.05 level [19, p. 749]. Correlations be-
tween the consumption of fragrance and physiological para-
meter differences in the experimental groups were tested with
the Bravais-Pearson product moment correlation. Significant
correlations were inspected visually for outliers which might ac-
count for statistical significance.
Results
In Table 1, medians and mean inter-quartile ranges (MIQR) of the
physiological parameters in the first and the second trial as well
as of the self-ratings of arousal on the second, the third and the
forth rating are presented for all groups. Medians and MIQR of
the inter-trial differences of EBR and PR are depicted in Fig. 1.
Univariate comparisons of the physiological parameters showed
a trend toward an increase of EBR in the SEO group compared
with the control group (U = 72.0, P = 0.064). Comparison of the
inter-trial differences of PR revealed a significant increase of PR in
the SEO group compared to both the control group (U = 114.0,
P =0.000)andthe
a
-santalol group (U = 15.0, P =0.003).
Medians and MIQR of the inter-trial differences of ST and SCL are
shown in Fig. 2. Comparison of the inter-trial differences of ST re-
vealed a trend toward a decrease of ST in the second trial in the
SEO group compared with the control group (U = 40.0,
P = 0.065). A significant decrease of ST in the
a
-santalol group
compared to the control group (U = 38.0, P = 0.050) was ob-
served. Comparison of the inter-trial differences of SCL showed
a significant increase of SCL in the second trial in the SEO group
compared with both the control group (U = 91.0, P = 0.011) and
the
a
-santalol group (U = 20.0, P = 0.008).
Medians and MIQR of SBP and DBP differences between the sec-
ond and the fourth measurement (difference 2± 4) and between
the third and the fourth measurement (difference 3 ± 4) are
shown in Fig. 3. SBP significantly decreased in the
a
-santalol
group compared to the SEO group on the fourth in relation to
the third measurement (U = 38.0, P = 0.048). Comparison of
DBP differences 2± 4 revealed a trend toward a smaller increase
in the SEO group than in both the control group (U = 42.0,
P = 0.081) and the
a
-santalol group (U = 104.0, P = 0.063).
Medians and MIQR of the differences of subjective attentiveness
and relaxation between the second and the fourth rating (differ-
ence 2 ±4) as well as between the third and the fourth rating (dif-
ference 3±4) are depicted in Fig. 4. Univariate comparisons of
the self-ratings of arousal indicated that subjective attentiveness
on rating 4 compared with rating 2 increased significantly more
in the
a
-santalol group than in the SEO group (U = 36.0,
P = 0.038). At the end of the second trial in relation to the begin-
ning of that trial, subjects in the SEO group felt marginally more
attentive than those in the control group (U = 41.5, P = 0.078).
Subjects in the
a
-santalol group rated themselves as significantly
more attentive than those in both the control group (U = 26.5,
P = 0.009) and the SEO group (U = 36.5, P = 0.040). Comparison
of the differences 2±4 revealed a trend toward a larger decrease
of subjective relaxation in the SEO group than in the control
group (U = 101.0, P = 0.094).
Medians and MIQR of the differences of subjective alertness and
mood between the second and the fourth rating (difference 2 ± 4)
as well as between the third and the fourth rating (difference 3 ±
4) are presented in Fig. 5. Comparison of the differences 2 ± 4
showed a weak trend toward a larger increase of subjective alert-
ness in the SEO group than in the
a
-santalol group (U = 100.5,
P = 0.100). A trend toward higher subjective alertness on rating
4 compared to rating 2 was found between the SEO group and
the control group (U = 42.0, P = 0.083). Comparison of the dif-
ferences 3± 4 revealed a significant increase of subjective mood
in the
a
-santalol group compared with the control group
(U = 34.0, P = 0.028).
No significant effects of the fragrances were found on BOS, RR,
EMG, subjective vigor, and calmness (P > 0.1 for all, data not
shown).
Comparison of the combined measures of physiological and
behavioral effects between the control group and the SEO group
resulted in a significant effect of SEO at the physiological level
(
c
2
= 51.859, df = 22, P < 0.001) and a trend toward significance
at the level of feelings of arousal (
c
2
= 33.714, df = 24, P < 0.1).
Comparison of the SEO group with the
a
-santalol group showed
significant differences at both the physiological and the behav-
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
794
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
ioral level (
c
2
= 46.838, df = 22, P < 0.005;
c
2
= 36.909, df = 24,
P < 0.05). No overall effects were found for
a
-santalol compared
to the placebo substance (
c
2
= 17.802, df = 22, P >0.1;
c
2
= 29.632, df = 24, P > 0.1).
Medians and MIQR of subjective odor intensity and effect on the
first and second rating for the control group and the experimen-
tal groups are given in Table 1. Medians and MIQR of the differen-
ces of subjective odor intensity and effect are presented in Fig. 6.
In the experimental groups, the odor of the substance in the sec-
ond trial compared to water in the first trial was experienced as
significantly more intense than in the control group (SEO vs. con-
trol: U = 123.0, P = 0.003;
a
-santalol vs. control: U = 112.0,
P = 0.021). In addition, the odor intensity of SEO was rated sig-
nificantly higher than that of
a
-santalol (U = 32.0, P = 0.021).
Comparison of the effect differences showed a trend toward
higher ratings at the end of trial 2 in relation to the end of trial 1
in the SEO group than in the control group (U = 41.0, P = 0.073).
The ratings of odor pleasantness did not differ significantly be-
tween groups (P > 0.1, data not shown).
Mean mass and SEM of evaporated SEO and
a
-santalol are shown
in Fig. 7. The consumption of SEO was significantly higher than
that of
a
-santalol (t = 3.834, P = 0.001).
Significant correlations were observed in all groups. The less in-
tense water was rated at the end of the second trial, the more at-
tentive felt subjects in the control group (
r
= + 0.804); and the
more cheerful subjects felt at the end of the second trial, the
more rose BOS (
r
= -0.648).
The more stimulating subjects in the SEO group rated the odor in
the second trial, the more increased SCL (
r
= ±0.691) and the
more vigorous they felt in the second trial (
r
= + 0.615). Also,
the more vigorous subjects felt on rating 4, the more rose SCL
(
r
= ±0.642) and EBR in the second trial (
r
= ±0.547); the more
attentive subjects felt at the end of trial 2, the more increased PR
(
r
= ±0.717); the more restless subjects rated themselves, the
more rose EMG in the second trial (
r
= + 0.609); the more cheer-
ful subjects were on rating 4, the less increased RR (
r
= + 0.718);
and the higher the consumption of SEO was, the more cheerful
felt subjects (
r
= + 0.610) and the less increased RR
(r = + 0.605, P = 0.037).
Table 1 Medians and mean inter-quartile ranges (MIQR) of skin temperature (ST), skin conductance level (SCL), eye-blink rate (EBR), pulse rate
(PR), systolic blood pressure (SBP), subjective relaxation, subjective attentiveness, subjective mood, and subjective alertness for the
control group (C), the Sandalwood oil group (SEO) and the
a
-santalol group (S)
Median MIQR
C SEO S C SEO S
ST Trial 1 31.69 32.38 33.29 1.26 1.72 0.96
Trial 2 31.23 31.26 32.03 1.74 1.71 1.60
SCL Trial 1 2.51 2.34 3.93 0.93 0.85 1.13
Trial 2 3.13 4.20 3.77 1.09 0.79 1.97
EBR Trial 1 10.88 11.75 16.90 4.59 4.58 5.91
Trial 2 15.25 17.05 17.35 6.09 7.71 4.42
PR Trial 1 67.78 64.12 71.19 6.91 5.73 4.24
Trial 2 64.84 63.43 72.05 4.92 6.34 6.44
SBP Measurement 2 116.00 115.50 115.50 7.00 10.00 11.00
Measurement 3 118.50 118.50 113.50 6.00 10.25 13.25
Measurement 4 119.00 120.00 114.00 6.25 10.25 11.25
DBP Measurement 2 68.50 71.00 71.50 3.00 6.25 5.50
Measurement 3 68.50 67.50 73.00 4.00 6.25 8.00
Measurement 4 69.50 70.00 76.00 2.50 7.25 7.50
Relaxation Rating 2 9.50 14.25 20.25 8.75 6.00 11.50
Rating 3 11.75 16.75 18.00 5.63 5.63 5.50
Rating 4 18.50 30.00 21.75 11.13 11.88 15.63
Attentiveness Rating 2 49.00 29.75 31.25 10.75 9.00 21.25
Rating 3 25.50 23.25 27.00 18.63 9.38 15.63
Rating 4 41.50 25.25 19.75 15.13 6.50 9.88
Mood Rating 2 16.50 21.00 24.50 10.75 11.38 11.88
Rating 3 14.75 24.75 25.75 8.38 12.63 12.00
Rating 4 30.00 26.00 22.75 18.25 8.50 9.13
Alertness Rating 2 56.75 52.25 47.00 11.88 13.88 24.25
Rating 3 44.50 37.50 49.75 12.38 12.38 17.88
Rating 4 51.25 24.25 32.75 20.38 8.63 25.50
Intensity Rating 1 14.75 14.50 16.00 14.00 8.25 14.38
Rating 2 11.75 63.75 36.50 17.50 14.00 22.13
Effect Rating 1 66.00 58.25 55.00 10.25 12.75 12.63
Rating 2 55.25 33.00 36.25 9.25 11.63 16.13
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
795
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
In the
a
-santalol group, the odor was rated the more stimulating,
the more attentive (
r
= + 0.677) and alert subjects felt at the end
of trial 2 (
r
= + 0.706); the more pleasant the odor was judged
on the second rating, the more cheerful felt subjects
(
r
= + 0.683); the lower the consumption of
a
-santalol was, the
more cheerful (
r
= ±0.729) felt subjects on the fourth rating and
the more pleasant they judged the odor (
r
= ±0.791).
Discussion
In our study, inhalation of Sandalwood oil, compared to odorless
water, increased ANS arousal, specifically pulse rate and skin
conductance level, in healthy human subjects. In contrast, the
test for effects on overall self-rated arousal failed to reach statis-
tical significance. Inhalation of
a
-santalol, compared to odorless
Fig. 2 Medians and mean inter-quartile
ranges of skin temperature and skin conduc-
tance level differences for the control group
and the experimental groups (Sandalwood
oil and santalol); H on the top of the bars in-
dicates significant differences (P
£
0.050)
between the control group and the experi-
mental groups, * on top of the bars indi-
cates significant differences (P
£
0.050) be-
tween the experimental groups.
Fig. 1 Medians and mean inter-quartile
ranges of eye-blink rate and pulse rate dif-
ferences for the control group and the ex-
perimental groups (Sandalwood oil and san-
talol); H on the top of the bars indicates sig-
nificant differences (P
£
0.050) between the
control group and the experimental groups,
* on top of the bars indicates significant dif-
ferences (P
£
0.050) between the experi-
mental groups.
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
796
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
water, did not significantly alter overall physiological or behav-
ioral arousal, although a decrease of skin temperature, higher at-
tentiveness and better mood were observed. These findings con-
trast results of previous investigations in which Sandalwood oil
showed deactivating effects when it was applied percutaneously
[11], [14]. The disagreement between these data possibly reflects
the impact of the route of administration of odorants. Although
in experimental animals Sandalwood oil showed sleep-inducing
and sedating effects also upon inhalation [12], [13], our results
are supported by a recent investigation in humans [20]. These
authors reported that a number of odors which were perceived
as pleasant and comfortable significantly decreased alpha
Fig. 3 Medians and mean inter-quartile
ranges of systolic and diastolic blood pres-
sure differences for the control group and
the experimental groups (Sandalwood and
santalol); difference 2 ±4: difference be-
tween measurements 2 and 4, difference
3 ± 4: difference between measurements 3
and 4; * on top of the bars indicates signifi-
cant differences (P
£
0.050) between the ex-
perimental groups.
Fig. 4 Medians and mean inter-quartile
ranges of subjective attentiveness and re-
laxation differences for the control group
and the experimental groups (Sandalwood
oil and santalol); difference 2 ± 4: difference
between ratings 2 and 4, difference 3 ±4:
difference between ratings 3 and 4; H on
the top of the bars indicates significant dif-
ferences (P
£
0.050) between the control
group and the experimental groups, * on
top of the bars indicates significant differen-
ces (P
£
0.050) between the experimental
groups.
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
797
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
rhythm, but that Sandalwood odor which was experienced as
unpleasant, induced uncomfortable feelings and did not sup-
press alpha 1 activity. We found significant differences between
the intensity ratings of the control group and the experimental
groups, which demonstrate that the odor concentrations of both
Sandalwood oil and
a
-santalol were well above the detection
threshold, even after an inhalation period of 20 minutes. Many,
if not all odorants have the tendency to stimulate the trigeminal
nerve, and this propensity gets stronger with increasing concen-
tration [21]. Thus, one explanation for the results of the present
study could be that Sandalwood oil stimulated trigeminal nerve
endings located in the nasal mucosa. Activation of the trigeminal
system may have elevated physiological arousal levels, e. g., via
the ascending reticular activating system. Reports of pungent
and itchy nasal sensations by our subjects are in agreement
with this hypothesis. In a follow-up study using ten-fold dilu-
tions of Sandalwood oil and
a
-santalol, both odors induced tri-
geminal stimulation [22]. However, concentrations used in
Fig. 5 Medians and mean inter-quartile
ranges of subjective alertness and mood dif-
ferences for the control group and the ex-
perimental groups (Sandalwood oil and san-
talol); difference 2±4: difference between
ratings 2 and 4, difference 3 ± 4: difference
between ratings 3 and 4; H on the top of
the bars indicates significant differences (P
£
0.050) between the control group and
the experimental groups.
Fig. 6 Medians and mean inter-quartile
ranges of odor intensity and odor effect dif-
ferences for the control group and the ex-
perimental groups (Sandalwood oil and San-
talol); H on the top of the bars indicates sig-
nificant differences (P
£
0.050) between the
control group and the experimental groups,
* on top of the bars indicates significant dif-
ferences (P
£
0.050) between the experi-
mental groups.
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
798
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
more natural settings, such as in aromatherapy practice, are
probably much lower than in our experiment, and it remains un-
clear whether arousing effects would also be observed under
these conditions.
Comparison of the experimental groups showed differences in
the stimulant properties of the fragrances. Subjects who had in-
haled essential Sandalwood oil showed higher overall physiol-
ogical and behavioral arousal levels than those who had inhaled
a
-santalol, although subjects in the
a
-santalol group felt more
attentive than those in the Sandalwood oil group. Differences be-
tween the fragrances were also obvious in terms of intensity rat-
ings, the odor of Sandalwood oil being experienced as more in-
tense than that of
a
-santalol. Sandalwood odor, but not
a
-santa-
lol odor, tended to be rated more stimulating than the odor of the
placebo substance. It may be speculated that differences in sti-
mulus characteristics between Sandalwood oil and
a
-santalol
are associated with differences in their arousing properties.
With respect to autonomic arousal, this relation has been discuss-
ed by other authors who found that unpleasant odor stimuli in-
duced different patterns of ANS activation and, in general, higher
autonomic arousal than pleasant smells [23], [24], [25], [26]. In a
recent investigation [27], intensity and arousal ratings of the sti-
muli were positively correlated with skin conductance varia-
tions. The latter relation was also obvious in the Sandalwood oil
group in the present investigation. In this group, subjective rat-
ings of the stimulating effect of the odor were also correlated
with subjective vigor, and vigor ratings covaried with skin con-
ductance and eye-blink rate. These findings suggest that psycho-
logical mechanisms triggered by a scent mediated the increase of
arousal. Regarding single physiological and behavioral measures,
inhalation of
a
-santalol produced some stimulating effect which
seems to have occurred mostly via psychological mechanisms.
Correlations between odor ratings and behavioral measures sug-
gest that odor pleasantness was the determining factor for these
effects. On the other hand, the mass of evaporated Sandalwood
oil was significantly greater than that of
a
-santalol. While this
difference is very likely due to dissimilar physico-chemical prop-
erties, e. g., vapor pressure, of the two fragrances, it may have
evoked different levels of activation of the trigeminal system
and could therefore alternatively explain the difference in stimu-
lant effect between Sandalwood oil and
a
-santalol.
Animal studies have shown that, after i.p. administration, seda-
tive effects of Sandalwood oil may be attributed to both
a
- and
b
-
santalol [13]. A pilot investigation by our group (unpublished
data) in which
b
-santalol was applied to the skin of 4 healthy hu-
man subjects also indicated that this component contributes to
the effects of Sandalwood oil on human arousal. Although to
our knowledge there are no reports on the inhalation of
b
-santa-
lol in either experimental animals or humans and we can there-
fore only speculate about its impact, it cannot be ruled out that
the effects of Sandalwood oil in the present investigation were
in part elicited by
b
-santalol. Future studies with an adequate
sample size will have to prove this assumption.
The results of the present study supplement previous research
performed by this group [14], [16], [17] and indicate (i) that, in
humans, the effects of fragrances on arousal may depend on ad-
ministration route, and (ii) that, when fragrances are adminis-
tered by means of inhalation and olfactory processing takes
place, several mechanisms may be involved which may be active
simultaneously. To thoroughly understand the mode of action of
volatile chemicals it is necessary to study the full range of their
effects ± those that can be observed after inhalation as well as
those that are unrelated to sensory processing.
Acknowledgements
The authors are grateful to Sirindhorn Scholarship, Srinakharin-
wirot University, Bangkok, Thailand (grant for T.H.) and to Drago-
co (now Symrise) GmbH, Vienna, for financial support of this
study.
References
1
Krotz A, Helmchen G. Total syntheses, optical rotations and fragrance
properties of sandalwood constituents: (±)-(Z)- and (±)-(E)-
b
-santalol
and their enantiomers, ent-
b
-santalene. Liebigs Ann Chem 1994; 6:
601±9
2
Banerjee S, Ecavade A, Rao AR. Modulatory influence of sandalwood
oil on mouse hepatic glutathione S-transferase activity and acid solu-
ble sulphydryl level. Cancer Lett 1993; 68: 105 ±9
3
Dwivedi C, Zhang Y. Sandalwood oil prevents skin tumour develop-
ment in CD1 mice. Eur J Cancer Prev 1999; 8: 449 ± 55
4
Dwivedi C, Guan X, Harmsen WL, Voss AL, Goetz-Parten DE, Koopman
EM et al. Chemopreventive effects of alpha-santalol on skin tumor de-
velopment in CD-1 and SENCAR mice. Cancer Epidemiol Biomarkers
Prev 2003; 12: 151 ±6
5
Benencia F, Courreges MC. Antiviral activity of sandalwood oil against
Herpes simplex viruses-1 and -2. Phytomedicine 1999; 6: 119 ± 23
6
Imai S, Oosato T, Sugiura A, Tsunakawa M, Kawasaki Y, Nagai K. Anti-
viral agents containing
a
-or
b
-pinene, d-limonene, or sandalwood oil.
Chem Abstracts 1996; 124: 250907n
7
Hammer KA, Carson CF, Riley TV. In-vitro activity of essential oils, in
particular Melaleuca alternifolia (tea tree) oil and tea tree oil products,
against Candida spp. J Antimicrob Chemother 1998; 42: 591 ±5
Fig. 7 Mean mass and SEM of inhaled fragrance; * on top of the bars
indicates significant differences (P
£
0.050) between the Sandalwood
oil group and
a
-santalol group.
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
799
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.
8
Tisserand R. The art of aromatherapy. Essex: CW Daniel, 1977
9
Jellinek JS. Psychodynamic odor effects and their mechanisms. Cosmet
Toiletries 1997; 112: 61± 71
10
Buchbauer G, Jirovetz L, Jager W, Plank C, Dietrich H. Fragrance com-
pounds and essential oils with sedative effects upon inhalation. J
Pharm Sci 1993; 82: 660 ± 4
11
Ide N, Hanata T. Cosmetics containing sleep-inducing fragrances.
Chem Abstracts 1997; 128: 53073u
12
Komori T, Tanida M, Nakamura S. Sleep-inducing perfume composi-
tions. Chem Abstracts 1996; 124: 155727v
13
Okugawa H, Ueda R, Matsumoto K, Kawanishi K, Kato A. Effect of
a
-
santalol and
b
-santalol from Sandalwood on the central nervous sys-
tem in mice. Phytomedicine 1995; 2: 119 ± 26
14
Hongratanaworakit T, Heuberger E, Buchbauer G. Evaluation of the ef-
fects of East Indian sandalwood oil and
a
-santalol on humans after
transdermal absorption. Planta Med 2004; 70: 3 ± 7
15
Braun NA, Meier M, Pickenhagen W. Isolation and chiral GC analysis of
b
-bisabolols-trace constituents from the essential oil of Santalum
album L. (Santalaceae). J Ess Oil Res 2003; 15: 63 ± 5
16
Heuberger E, Hongratanaworakit T, Bohm C, Weber R, Buchbauer G.
Effects of chiral fragrances on human autonomic nervous system
parameters and self-evaluation. Chem Senses 2001; 26: 281± 92
17
Heuberger E, Redhammer S, Buchbauer G. Transdermal absorption of
(±)-linalool induces autonomic deactivation but has no impact on rat-
ings of well-being in humans. Neuropsychopharmacology 2004; 29:
1925± 32
18
Oldfield RC. The assessment and analysis of handedness: the Edin-
burgh inventory. Neuropsychologia 1971; 9: 97 ± 113
19
Bortz J, Lienert GA, Boehnke K. Verteilungsfreie Methoden in der Bio-
statistik. Berlin: Springer, 1990
20
Masago R, Matsuda T, Kikuchi Y, Miyazaki Y, Iwanaga K, Harada H et al.
Effects of inhalation of essential oils on EEG activity and sensory eval-
uation. J Physiol Anthropol Appl Human Sci 2000; 19: 35 ± 42
21
Doty RL, Cometto-Muæiz JE. Trigeminal chemosensation. In: Hand-
book of Olfaction and Gustation Doty RL, editor. New York: Marcel
Dekker, 2003: p 981± 99
22
Heuberger E, Springsics C, Meddeb R, Buchbauer G. Trigeminal che-
mosensation and its relation to physiological odor effects. 35th ISEO,
Messina, Italy: 2004
23
Alaoui-Ismaili O, Vernet-Maury E, Dittmar A, Delhomme G, Chanel J.
Odor hedonics: connection with emotional response estimated by au-
tonomic parameters. Chem Senses 1997; 22: 237 ±48
24
Brauchli P, Ruegg PB, Etzweiler F, Zeier H. Electrocortical and auto-
nomic alteration by administration of a pleasant and an unpleasant
odor. Chem Senses 1995; 20: 505± 15
25
Robin O, Alaoui-Ismaili O, Dittmar A, Vernet-Maury E. Basic emotions
evoked by eugenol odor differ according to the dental experience. A
neurovegetative analysis. Chem Senses 1999; 24: 327± 35
26
Nagai M, Wada M, Usui N, Tanaka A, Hasebe Y. Pleasant odors attenu-
ate the blood pressure increase during rhythmic handgrip in humans.
Neurosci Lett 2000; 289: 227 ± 9
27
Bensafi M, Rouby C, Farget V, Bertrand B, Vigouroux M, Holley A. Influ-
ence of affective and cognitive judgments on autonomic parameters
during inhalation of pleasant and unpleasant odors in humans. Neuro-
sci Lett 2002; 319: 162 ± 6
Heuberger E et al. East Indian Sandalwood ¼ Planta Med 2006; 72: 792 ± 800
Original Paper
800
Downloaded by: Saarländische Universitäts- u. Landesbibliothek. Copyrighted material.