Вы находитесь на странице: 1из 9

JOURNAL OF

PSYCHIATRIC
RESEARCH

Journal of Psychiatric Research 41 (2007) 472480

www.elsevier.com/locate/jpsychires

Ginkgo biloba special extract EGb 761 in generalized anxiety


disorder and adjustment disorder with anxious mood:
A randomized, double-blind, placebo-controlled trial
H. Woelk a, K.H. Arnoldt b, M. Kieser c, R. Hoerr

c,*

Klinik fur Psychiatrie und Psychotherapie, Gieen, Germany


b
CNS Consultants, Darmstadt, Germany
Dr. Willmar Schwabe GmbH & Co. KG, Willmar-Schwabe-Str. 4, 76227 Karlsruhe, Germany
Received 22 February 2006; received in revised form 16 May 2006; accepted 17 May 2006

Abstract
Ginkgo biloba special extract EGb 761, an anti-dementia drug, enhances cognitive functioning and stabilizes mood in cognitively
impaired elderly subjects. Moreover, EGb 761 had been found to alleviate symptoms of anxiety in people with mental decline, therefore
it was now tested for clinical ecacy in younger patients suering from anxiety. One hundred and seven patients with generalized anxiety
disorder (GAD, n = 82) or adjustment disorder with anxious mood (ADWAM, n = 25) according to the diagnostic and statistical manual of mental disorders, third edition revised (DSM-III-R) were randomized to daily doses of 480 mg EGb 761, 240 mg EGb 761 or
placebo for 4 weeks. Intention-to-treat (ITT) analyses were performed on the primary outcome measure, the Hamilton rating scale for
anxiety (HAMA), and the secondary variables, the clinical global impression of change (CGI-C), the Erlangen anxiety tension and
aggression scale (EAAS), the list of complaints (B-L 0 ), and the patients global rating of change. The HAMA total scores decreased
by 14.3 (8.1), 12.1 (9.0) and 7.8 (9.2) in the high-dose EGb 761, the low-dose EGb 761 and the placebo group, respectively.
Changes were signicantly dierent from placebo for both treatment groups with p = 0.0003 (high-dose group) and p = 0.01 (low-dose).
Regression analyses revealed a doseresponse trend (p = 0.003). EGb 761 was signicantly superior to placebo on all secondary outcome measures. It was safe and well tolerated and may thus be of particular value in elderly patients with anxiety related to cognitive
decline.
2006 Elsevier Ltd. All rights reserved.
Keywords: Ginkgo biloba; EGb 761; Randomized controlled trial; Anxiety; Generalized anxiety disorder; Adjustment disorder

1. Introduction
Symptoms of anxiety, usually not in appropriate
response to a real threat, are among the frequent causes
for seeking medical help. Estimated prevalence rates for
anxiety disorders in primary care settings range from
2.5% to as much as 19% (Sansone et al., 2004; Ansseau
et al., 2004). For generalized anxiety disorder (GAD) a
prevalence of 10.3% was found in an adult primary care
*

Corresponding author. Tel.: +49 721 4005 492; fax: +49 721 4005
8492.
E-mail address: robert.hoerr@schwabe.de (R. Hoerr).
0022-3956/$ - see front matter 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpsychires.2006.05.004

population (Ansseau et al., 2004), and a lifetime prevalence


of 47% in the general population (Allgulander et al.,
2003). From a survey in 88 outpatients of an internal medicine clinic Sansone et al. (2004) reported a 33% prevalence
of generalized anxiety symptoms. Among the anxiety disorders classied by the diagnostic and statistical manual of
mental disorders, third edition revised (DSM-III-R,
American Psychiatric Association, 1987) GAD is the second most frequent diagnosis found (Lepine and Lellouch,
1994).
Another disorder associated with symptoms of anxiety
which are generalized rather than phobic in nature is
adjustment disorder with anxious mood (ADWAM). Its

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

prevalence was 7.8% as a sole diagnosis and a further 4.2%


as comorbidity with other axis I and II diagnoses in a consultation-liaison psychiatry setting (Strain et al., 1998). In a
primary care setting, the prevalence of ADWAM was 1%
in the whole population of consecutive patients, 4.5%
among the patients with psychological complaints and a
further 9.2% in association with other psychiatric disorders
(Semaan et al., 2001).
GAD is amenable to psychotherapy (primarily cognitive-behavioral therapy) as well as pharmacotherapy. Current recommendations include benzodiazepines, buspirone
and antidepressants, above all serotonin and noradrenalin
reuptake inhibitors (SNRIs) and selective serotonin reuptake inhibitors (SSRIs) (Allgulander et al., 2003). Benzodiazepines have the advantage of quick symptom relief, but
side eects (sedation, psychomotor and cognitive impairment), physical dependence and withdrawal eects limit
their long-term use. Buspirone is eective in the short-term
treatment of GAD, but long-term studies are missing. All
types of drugs currently recommended for treating GAD
may adversely aect patients functioning at work, cardriving and social life. Symptoms of anxiety, although
often at a sub-syndromal level, are frequently found in subjects suering cognitive decline (Lyketsos et al., 2002; Chan
et al., 2003; Forsell et al., 2003). In such cases, anxiety may
be a symptom of the underlying disease itself or a reaction
to the perception of mental decline.
There appears to be a gap between the frequency of
ADWAM and high-grade evidence for the eectiveness
of treatments. In a cross-sectional study of a sample of
ADWAM patients drawn from general practitioners in
France (Semaan et al., 2001), 74% were prescribed drug
treatments and 90% received non-pharmacological treatments. Anxiolytics were prescribed most frequently
(60%), followed by antidepressants (11%) and hypnotics
(8%). A Dutch group of experts issued a discussion paper
suggesting practice guidelines based on existing evidence,
experience in adjacent elds and consensus procedures
(van der Klink and van Dijk, 2003) for the treatment of
work-related adjustment disorders. While existing evidence
was sparse, their proposed three-phase psychotherapeutic
approach was based mainly on experience and consensus.
Ginkgo biloba special extract EGb 761 is registered in
Germany and many other countries for the treatment of
dementia disorders. In two large trials the drug has been
shown to improve cognitive performance, activities of daily
living and social functioning as well as the overall condition
of patients with dementia (Kanowski et al., 1996; Le Bars
et al., 1997). The ecacy of EGb 761 in the treatment of
dementia and aging-related cognitive impairment has also
been demonstrated in further studies, reviewed by Oken
et al. (1998) and Birks et al. (2002). Cognition-enhancing
eects in mentally healthy elderly subjects have been
reported recently by Mix and Crews (2002) and Cieza et al.
(2003). In one of the early studies enrolling patients with cognitive impairment due to cerebrovascular disease, anxiety
was one of the non-cognitive symptoms relieved signicantly

473

by EGb 761 treatment (Eckmann and Schlag, 1982). Benecial eects of EGb 761 on behavioral and psychological
symptoms of dementia (BPSD), including anxiety, have been
shown in a series of further studies (summarized in Hoerr,
2003), yet proof of ecacy in primary anxiety disorders or
anxiety in response to stressful events has been lacking.
EGb 761 showed stress-alleviating and anxiolytic-like
activity in pre-clinical studies. In a rat model EGb 761 signicantly reduced the detrimental eects of learned helplessness on a subsequent conditioned avoidance task. It
also increased the intake of novel food in a new environment in a mouse model of emotional hypophagia (Porsolt
et al., 1992). On the elevated plus maze (EPM), senescent
mice treated with EGb 761 spent signicantly more time
on the open (unsafe) arms than those treated with
drug-free vehicle only. In general, animals prefer the closed
(safe) arms. After a forced-swim task in cold water, anxiety was increased, i.e. the time spent on the open arms was
decreased, in vehicle-treated, but not in EGb 761-treated
animals (Ward et al., 2002). Mice treated with a particular
fraction of the special extract EGb 761 spent more time in
the light part of a light-dark box than control animals.
Similar eects were demonstrated for anxiolytic drugs, such
as diazepam and buspirone (Noldner and Chatterjee, personal communication). Anxiety-related hyperthermia as
observed in the remaining animals after removing mice
one by one from a cage and then replacing them, is also signicantly attenuated by the same extract fraction (Noldner
and Chatterjee, personal communication).
Using a discrimination learning task in rats, Rapin and
co-workers (1994) demonstrated that auditory perturbation (stress) during the discriminative phase of learning
decreased the percentage of correct responses and increased
the number of errors. These changes were correlated with
increases in plasma concentrations of epinephrine, norepinephrine and corticosterone. Both detrimental eects of
auditory perturbation and rises in plasma hormones were
suppressed by EGb 761. The drug eects were present
in both young and old animals. Long-term administration
of EGb 761 to rats resulted in a decreased basal corticosterone secretion and an attenuation of the related increase
in corticotropin releasing hormone (CRH) and arginine
vasopressin (AVP) gene expression (Marcilhac et al.,
1998). Under intense surgical stress CRH, ACTH and corticosterone plasma concentrations were markedly elevated
in control animals, but signicantly less so in EGb 761
treated animals. These ndings suggest that EGb 761
interferes with the regulation of the activity of the hypothalamicpituitaryadrenocortical (HPA) axis.
In a randomized and placebo-controlled double-blind
study in 70 healthy young volunteers, Jezova and co-workers (2002) used a stress model involving a combination of
static exercise (handgrip) and mental stimuli. A single dose
of 120 mg EGb 761 signicantly attenuated the stressinduced rise in systolic and diastolic blood pressure
without aecting the heart rate. A stress-related increase
in salivary cortisol levels was observed in placebo-treated

474

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

male subjects in the afternoon. This eect was absent in


volunteers treated with EGb 761. In a pilot study with a
similarly standardized Ginkgo extract, the sleep pattern of
depressed inpatients receiving trimipramine monotherapy
was examined (Hemmeter et al., 2001). Under adjuvant
Ginkgo treatment slow-wave sleep was enhanced, REMdensity was decreased in sleep stage 1 and was increased
in sleep stage 2. Sleep eciency was improved and awakenings were reduced during Ginkgo intake. Most eects were
lost after discontinuation of adjuvant Ginkgo treatment.
The authors concluded that the improved sleep pattern
was secondary to a weakening of tonic CRF-activity.
The objective of the present study was to establish
whether the stress-alleviating and anxiolytic-like eects
found in animal and volunteer studies translate into clinically meaningful anxiolytic eects in patients suering from
specied non-psychotic and non-phobic types of anxiety.
2. Patients and methods
The study was conducted in accordance with the current
versions of the Declaration of Helsinki and the Good Clinical Practice guidelines. The protocol was reviewed by the
independent ethics committees of the Chamber of Physicians of Baden-Wurttemberg and the Faculty of Medicine
of the University Erlangen-Nuremberg. Informed consent
was obtained from all patients after the nature of the study
and the procedures to be followed had been fully explained
and before any study-related procedures commenced. An
audit was performed by an independent expert for quality
assurance to verify that the study was conducted according
to the laws and ethical standards in force.
Patients with generalized anxiety disorder (GAD) or
adjustment disorder with anxious mood (ADWAM) were
enrolled by 15 private practices of specialists in neurology/
psychiatry, specialists in internal medicine or general practitioners and by the outpatient clinic of a psychiatric university
hospital. The diagnoses were established in accordance with
the criteria of the DSM-III-R (300.02 and 309.24, respectively). The latter diagnosis was chosen on the basis of the
assumption that such patients might be encountered frequently in an outpatient setting and that they predominantly
suer from generalized symptoms of anxiety similar to those
seen in GAD. Although all investigators were trained by a
psychiatrist and a psychologist in applying the diagnostic criteria, all diagnoses made by internists or general practitioners had to be veried by a psychiatrist before enrollment
of a patient. Patients were eligible if they were between 18
and 70 years of age, had a total score on the German version
of the Hamilton Rating Scale for Anxiety (HAMA, CIPS,
1996) above 20 both at enrollment and at the end of the qualication period, and had given informed consent to participation in the trial and access to their clinical data.
Patients had to be excluded if there was a perceived risk
of suicide or if they were too severely ill to justify their
enrollment in a placebo-controlled study. Patients with
other anxiety disorders, anxiety related to other psychiatric

disorders, obsessivecompulsive disorder, suspected dementia or severe somatic disorders were also excluded. Further
exclusion criteria were substance abuse, lack of cooperation,
inability to complete self-rating questionnaires, or treatment
with psychoactive drugs.
After general physical and neurological examination,
including laboratory tests, and a psychiatric interview,
patients were admitted to the study and underwent a
seven-day qualication period during which they received
placebo (2 lm-coated tablets t.i.d.) in a single-blind manner. Patients were eligible for randomization, if they were
compliant and if their HAMA total score did not change
by more than 7 points in either direction during the qualication period. Patients who responded to placebo treatment and those whose symptoms worsened progressively
were thus excluded.
Randomization was performed by the biometric unit of
Dr. Willmar Schwabe Pharmaceuticals. Using a validated
computer program each randomization number was assigned
to one of three treatments: high-dose EGb 761, low-dose
EGb 761 or placebo. The sealed randomization code was
stored safely and inaccessible to any person involved in the
study. Study drug was distributed to the sites in full blocks
of six successive randomization numbers. At the sites the randomization numbers, and thus the corresponding treatments,
were allocated to the patients in ascending order. This procedure ensured concealment and precluded any interference
with random treatment allocation.
During the four-week treatment period, all patients took
two lm-coated tablets t.i.d containing either 40 mg or
80 mg EGb 761 or no active drug (placebo). EGb 761
is a special extract (drugextract ratio 3567:1) of Ginkgo
biloba leaves, standardized to contain 2227% of avone
glycosides and 57% terpene lactones (ginkgolides, bilobalide). Active drug and placebo tablets were of the same size
and shape and coated with the same surface lm for identical appearance. To maintain high compliance, detailed
instructions for proper use of the study drug were provided
and repeated at every visit. Drug packages with unused
tablets were collected by the sites for compliance testing
at the end of the treatment period.
The standard instrument for the assessing of anxiety, the
Hamilton rating scale for anxiety (HAMA, CIPS, 1996) was
specied by the protocol as primary outcome measure. Secondary measures were the list of complaints (Beschwerdeliste) version B-L (CIPS, 1996), the Erlangen anxiety
tension and aggression scale (Erlanger Angst-Spannungsund Aggressivitatsskala EAAS; Junger, 1982), the clinical
global impression (CGI; CIPS, 1996) and a global self-rating
of change as perceived by the patient. The B-L and the
EAAS are self-rating instruments whereas the HAMA and
the CGI are rated by the investigators. Assessments took
place before the rst intake of the study drug (baseline),
and thereafter on days 4, 8, 15, and 29. The patients were
asked about adverse events at all visits, and laboratory tests
were done at screening and at the end of the treatment period, in order to assess safety and tolerability of the treatment.

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

Since there were hardly any data available to estimate the


potential eect size for EGb 761 in the treatment of anxiety, a reliable sample size calculation was dicult. An adaptive interim analysis was therefore planned after completion
of data collection from 90 patients. This allowed for sample
size re-calculation if the study was to be continued after the
interim analysis. The trial could be stopped after the interim
analysis with proof of ecacy of EGb 761 if the p-value fell
below the critical boundary a1 = 0.0207 (overall type I error
rate a = 0.05; Bauer and Kohne, 1994). The two nullhypotheses concerning the comparisons of the EGb 761
doses to placebo with respect to the primary outcome variable were ordered a priori, based on the assumption that the
higher dose of EGb 761 is not less eective than the lower
one. In conrmatory analysis, the corresponding sequentially rejecting test procedure was applied, which thus controlled the multiple level (Kieser et al., 1999). Sample size
calculation was based on the signicance level a1 = 0.0207
to be applied in the interim analysis and a desired power
of 1  b = 0.80, under the assumption that with a maximum
dropout rate of 20%, stochastic superiority of at least 0.74
can be detected using the MannWhitney U-test. All statistical analyses were performed using the software SPSS for
Windows, version 6.1; the exact U-tests stratied by centers
that were pre-specied for the conrmatory analysis were
computed using the software BMDP StatXact Turbo.
3. Results
3.1. Patient disposition and characteristics
Rapid enrollment of patients at 16 sites resulted in a
slight over-recruitment with a total sample of 109 patients.

475

Table 1
Patient characteristics at baseline, means standard deviations, n = 107

Gender m/f (n)


Age (years)
Height (cm)
Weight (kg)
HAMA total score
GAD/ADWAM (n)

EGb 761
480 mg

EGb 761
240 mg

Placebo

12/22
47.5 11.0
170.6 8.8
77.1 15.8
30.7 5.2
27/7

14/22
47.6 12.5
170.1 9.3
70.6 12.0
29.7 5.5
25/11

15/22
46.7 13.0
171.7 7.4
73.0 11.1
29.5 5.5
30/7

Patient disposition is depicted in Fig. 1, baseline characteristics for the randomized patients are provided in Table 1.
All randomized patients had at least one examination after
baseline and could be included in the ITT analysis. The
treatment groups were comparable with respect to demographic and anthropometric data as well as severity of anxiety. The most frequent concomitant diseases were reported
for the musculoskeletal system (24%), the cardio-vascular
system (17%), the gastro-intestinal tract (15%), and metabolic functions (15%). This pattern is reected by the drugs
taken most frequently: cardiac medications, blood pressure
regulating drugs, non-steroidal anti-inammatory drugs,
and drugs for gastro-intestinal disorders. The groups were
well balanced with respect to concomitant diseases, prior
and concomitant drug intake. Seventeen patients had been
taking psychoactive drugs before enrollment: eight patients
of the placebo group (zolpidem, depot uspirilene, carbamazepine, one patient each; various herbals, ve patients),
ve patients of the low-dose EGb 761 group (doxepin,
two patients; various herbals, three patients) and four
patients of the high-dose EGb 761 group (benzodiazepines, four patients; valeriana, one patient). These treatments were discontinued before enrollment in the

Screened
109
Dropped:
Placebo Response (1)
Withdrawal of consent (1)
Randomized
107

EGb 761 480 mg


34
(ITT)

Dropped:
Relocation (1)
Adverse Event (1)
EGb 761 480 mg
32
Completed

EGb 761 240 mg


36
(ITT)

Placebo
37
(ITT)

Dropped:
Vacation (1)

EGb 761 240 mg


35
Completed

Fig. 1. Patient disposition.

EGb 761 240 mg


37
Completed

476

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

qualication period (except carbamazepine, which was


taken for trigeminal neuralgia throughout the study).
3.2. Primary outcome measure
In the interim analysis the decrease in the HAMA total
score was signicantly larger for each of the EGb 761
Table 2
HAMA total scores at baseline and changes from baseline to day 29,
means standard deviations [numbers]
(Sub-)sample/treatment

Baseline

Changea

p-Value

All patients
EGb 761 480 mg
EGb 761 240 mg
Placebo

30.7 5.2 [34]


29.7 5.5 [36]
29.5 5.5 [37]

14.3 8.1 [34]


12.1 9.0 [36]
7.8 9.2 [37]

0.0003
0.01

GAD subsample
EGb 761 480 mg
EGb 761 240 mg
Placebo

30.5 5.4 [27]


30.0 5.8 [25]
29.6 5.3 [30]

13.7 7.0 [27]


11.8 9.0 [25]
7.6 8.8 [30]

0.004
0.04

ADWAM subsample
EGb 761 480 mg
EGb 761 240 mg
Placebo

31.6 4.9 [7]


29.0 5.0 [11]
28.7 6.5 [7]

16.4 11.9 [7]


12.8 9.3 [11]
8.4 11.4 [7]

n.a.
n.a.

HAMA P30 at baseline


EGb 761 480 mg
EGb 761 240 mg
Placebo

34.9 4.1 [17]


35.0 3.2 [16]
34.3 2.9 [18]

17.9 8.0 [17]


15.4 10.5 [16]
11.2 11.2 [18]

0.04
0.16

HAMA <30 at baseline


EGb 761 480 mg
EGb 761 240 mg
Placebo

26.5 1.6 [17]


25.5 2.5 [20]
24.8 2.3 [19]

10.7 6.6 [17]


9.4 6.7 [20]
4.6 5.1 [19]

0.002
0.006

p-values for comparisons between drug and placebo groups regarding


changes (exact U-test stratied by center).
n.a., not appropriate because of small size of subsample.
a
ITT with last observation carried forward (LOCF).

treated patient groups than for the placebo group


(p < 0.0207). Both null hypotheses could therefore be
rejected and the study could be stopped with the proof of
ecacy of EGb 761. Details concerning the changes in
HAMA scores are provided in Table 2 and Fig. 2. A reduction in HAMA total score of at least 50% (response) was
found in 44% of high dose EGb 761 patients, in 39% of
low-dose EGb 761 patients and in 22% of placebo
patients. Total remission, dened as a HAMA total score
67 at day 29, was reported for 9%, 8% and 5% of patients,
respectively. No change or a worsening in HAMA scores
was seen in 6%, 8% and 19%, respectively. Response and
remission rates must be interpreted in the context of the
rather short treatment period of 4 weeks. The results for
both diagnostic groups look similar, yet the small numbers
of patients with ADWAM do not allow conclusive statistical comparisons. Changes upon treatment appear to be larger in subjects whose baseline HAMA was 30 or higher.
However, this does not only apply to the EGb 761-treated
but also to the placebo-treated patients. Drug-placebo differences are similar to those found for the patients with
baseline HAMA below 30. Regression to the mean in subjects with high scores at baseline may account for this phenomenon. Patients who discontinued doxepin or
benzodiazepines showed improvement upon subsequent
treatment with EGb 761, except one patient who became
worse after a slight initial improvement.
3.3. Secondary outcome measures
According to the clinicians global judgement, as operationalized by the CGI-C, the patients treated with either
dose of EGb 761 fared signicantly better than those taking placebo (Table 3). According to the most frequently
used denition of a clinically signicant response, i.e.

Change in HAMA Total Score

Placebo
EGb 761 240mg
EGb 761 480mg
0

-5

-10

-15

-20

Day -7

Day -1

Day 4

Day 8

Day 15

Day 29

Fig. 2. Changes in HAMA total scores relative to baseline (day 1); means and standard errors.

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480


Table 3
Clinical global impression of change (CGI-C) as rated on day 29, number
(%) of patients
Description of change

EGb 761
480 mg

EGb 761
240 mg

Placebo

Very much improved


Much improved
Minimally improved
Unchanged
Minimally worse
Much worse
Total
p-Value

11 (34)
15 (47)
4 (13)
1 (3)
0 (0)
1a (3)
32 (100)
0.0001

10 (28)
14 (39)
5 (14)
7 (19)
0 (0)
0 (0)
36 (100)
0.008

5 (14)
9 (24)
9 (24)
11 (30)
3 (8)
0 (0)
37 (100)

p-values for comparisons between drug and placebo groups (U-test); ITT
analysis.
a
Severe depression in reaction to death of father.

ratings of much improved or very much improved in


the CGI-C, response rates are 81%, 67% and 38% for
high-dose EGb 761, low-dose EGb 761 and placebo,
respectively (p < 0.05). The numbers needed to treat
(NNT) to achieve one clinically signicant response are
therefore 2.3 and 3.5 for high-dose and low-dose EGb
761, respectively. This pattern of response is also reected
in the patients global perception of change with 78% of the
high-dose EGb 761 group, 61% of the low-dose EGb 761
group and 30% of the placebo group rating their condition
as at least markedly improved on a 9-level categorical scale
extending from complete remission to marked deterioration. On the contrary, 6%, 19% and 46%, respectively, rated
themselves as unchanged or worse (p < 0.005). On the
EAAS as well as on the B-L 0 the patients of both active
treatment groups experienced greater improvements than
the placebo-treated patients, with improvements being most
marked in the high-dose EGb 761 group (Table 4).
3.4. Doseresponse trend
Numerically, changes in all outcome measures were larger for the high-dose group than for the low-dose group.
An additional regression analysis revealed a signicant
association between dose per kg body weight and decrease
in HAMA total score (p = 0.003).
Table 4
EAAS and B-L 0 total scores at baseline and changes from baseline to day
29, means standard deviations [numbers]
Changea

Scale/treatment

Baseline

EAAS
EGb 761 480 mg
EGb 761 240 mg
Placebo

23.7 8.7 [34]


24.7 7.2 [36]
24.0 7.8 [37]

9.9 8.0 [34]


9.2 9.7 [36]
5.2 9.6 [37]

0.004
0.02

B-L 0
EGb 761 480 mg
EGb 761 240 mg
Placebo

37.6 10.2 [34]


35.8 11.0 [36]
36.6 11.0 [37]

15.3 10.1 [34]


10.4 11.1 [36]
7.9 12.9 [37]

0.0006
0.07

p-Value

p-values for comparisons between drug and placebo groups regarding


changes (U-test).
a
ITT with last observation carried forward (LOCF).

477

3.5. Safety and tolerability


No adverse events were observed upon discontinuation
of psychoactive drugs at enrollment. During the course of
the treatment period, nine adverse events (AE) were
reported, three for each treatment group. No serious AE
occurred. The AEs emerging under placebo were mild gastroenteritis, moderate pain in the context of pre-existing
osteoarthritis of the lumbar spine and moderate to severe
tachycardia with hypertension. In the low-dose EGb
761 group moderate to severe gastroenteritis, moderate
sinusitis and mild pruritus, the latter persisting after study
completion, were documented. A causal relationship with
the study drug was excluded for the rst and considered
unlikely for the second and third event. Of the patients
treated with the higher dose of EGb 761, one female
patient developed severe depression as a reaction to the
death of her father. This event was considered unrelated
to the study drug. One case each of mild spastic bronchitis
and a moderately severe allergic rash were observed in the
high-dose group. While a causal relationship with the
study drug was considered unlikely for the former, it was
considered probable for the latter, because the symptoms
ceased a few days after premature termination of
treatment.
The systolic blood pressure dropped from a median 130
to 122.5 mmHg and from 137.5 to 130 mmHg in the highdose and low-dose EGb 761 groups, respectively, whereas
there was no obvious change in the placebo group. In the
laboratory tests there were neither systematic changes nor
single cases of marked deviations related to EGb 761
treatment.
4. Discussion
In this study, we enrolled patients who fullled either the
diagnostic criteria for generalized anxiety disorder (GAD)
or those for adjustment disorder with anxious mood
(ADWAM). Irrespective of the inclusion diagnosis, distinct
improvements compared to placebo were seen in the clinicians structured rating, the HAMA, a standard measure of
the severity of anxiety. The ndings are corroborated by
the patients structured self-ratings, EAAS and B-L 0 as well
as by the patients and physicians overall judgement of
change. Improvements in psychopathology, as indicated
by the HAMA, were associated with an alleviation of
somatic symptoms as reected by the B-L 0 . In all outcome
measures, the patients treated with the higher dose of EGb
761 tended to fare better than those taking the lower dose.
In fact, a regression analysis revealed a signicant association between dose per body weight and decrease in HAMA
score. This dose-dependency of eects together with the
striking concordance of ndings in all outcome measures
strongly supports the notion that EGb 761 is eective in
the treatment of anxiety.
The 14.3 and 12.1 point improvements in HAMA total
scores found for 480 mg and 240 mg EGb 761 are of a

478

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

similar magnitude to those found after a 6-week treatment


with daily doses of 300 mg of the Kava-Kava extract WS
1490 (11.7 points), 5 mg of oxazepam (11.2 points) or
3 mg of bromazepam (13.9 points) in non-psychotic outpatients with slightly milder symptoms of anxiety, tension
and restlessness (Woelk, 1993). In another sample of
slightly less aected patients, most of whom had GAD,
de Jonghe et al. (1989) found decreases of approximately
10 points on the HAMA during a four-week course of
treatment with suriclone or lorazepam. In their study, both
active treatments relieved symptoms more quickly than
placebo, but the eects tended to level o later during the
course of treatment in a similar way to the placebo group.
In our study, on the contrary, the decrease in HAMA
scores was similar in all three treatment groups during
the early phase of treatment and then leveled o in the placebo group but continued in both actively treated groups.
This suggests that it might require more than 4 weeks of
treatment to see the full therapeutic potential of EGb
761 unfold.
EGb 761 also compares well with modern anti-depressant drugs in the treatment of GAD. For the serotonin and
noradrenalin reuptake inhibitor, venlafaxine, and the selective serotonin reuptake inhibitor, sertraline, mean improvements on the HAMA of approximately 11 points and 8
points, respectively, were found after treatment periods of
4 weeks in patients with slightly lower baseline scores
(Meoni et al., 2004; Allgulander et al., 2004). The NNTs
of 4.0 for imipramine, 5.0 for venlafaxine and 6.7 for paroxetine, as reported in a Cochrane Review on antidepressants for GAD (Kapczinski et al., 2003), do not seem to
support the notion that these drugs might be more potent
than EGb 761 with NNTs of 3.5 for low-dose and 2.3
for high-dose treatment.
Although symptom reduction was obvious as early as 4
days after the start of treatment, there appears to be no
immediate strong anxiolytic eect as is expected for benzodiazepines. Over the course of four weeks, however, the
relief provided by EGb 761 was of the same magnitude
as that achieved with benzodiazepines in other studies. In
the long run, the advantages of EGb 761 compared to
other anxiolytics, benzodiazepines in particular, are of considerable interest. EGb 761 does not impair vigilance and
cognitive functioning. On the contrary, studies in healthy
volunteers and in cognitively impaired subjects found that
it has vigilance-enhancing and cognitively activating eects
(Mix and Crews, 2002; Luthringer et al., 1995; Schulz
et al., 2000). It may therefore be particularly suitable for
the treatment of patients who wish to continue driving
vehicles or who need to operate dangerous machines, as
well as for patients undergoing cognition-based psychotherapy. Moreover, neither clinical studies involving several thousands of patients nor extensive long-term use of
marketed products have revealed a potential in EGb
761 to cause dependence.
The molecular basis of the eects of EGb 761 on
anxiety remains to be elucidated. Suppression of CRH

secretion (Rapin et al., 1994; Marcilhac et al., 1998; Jezova et al., 2002; Hemmeter et al., 2001) seems to be a
feasible candidate mechanism. Central administration or
transgenic overproduction of CRH leads to increased
anxiety-like behavior in rodents (van Gaalen et al.,
2002), whereas genetic disruption of the CRH/CRHR1
system in rodents (Deussing and Wurst, 2005) or oral
administration of a CRH receptor antagonist to primates
(Habib et al., 2000) attenuates stress-related anxiety.
There is increasing evidence that elevated CRH and
overactivity of the HPA axis contribute substantially to
symptoms of depression and anxiety in humans (Strohle
and Holsboer, 2003) and that CRH-antagonistic treatment strategies may be clinically eective (Holsboer,
2001).
Whether it is appropriate to enroll patients with GAD
and ADWAM in the same study may be questioned, as
the latter is not an anxiety disorder by classication. In
fact, the etiologies appear to be dierent. However, this
does not necessarily mean that the biochemical and pathophysiological correlates of anxiety are also dierent.
Abnormal activation of the HPA axis has been found in
several disorders associated with anxiety (Strohle and
Holsboer, 2003). In subgroup analyses we found similar
baseline scores and similar dose-related eects of EGb
761 for both diagnostic groups. Statistical testing revealed
that EGb 761 was signicantly superior to placebo in the
GAD subgroup. However, testing was inappropriate for
the small ADWAM sub-sample.
EGb 761, even at a daily dose of 480 mg, was well tolerated. Adverse events were equally distributed across the
treatment groups, and laboratory tests did not reveal any
systematic changes. The clear-cut drop in systolic blood
pressure found in both treatment groups is in striking contrast to the ndings of other trials in which EGb 761 was
administered at daily doses up to 240 mg. The fact that the
size of change is exactly the same for both dosage groups
renders a high-dose eect unlikely. The nding can be better explained as a consequence of the profound relief of
anxiety and anxiety-related stress provided by EGb 761.
This also appears to be in line with the ndings by Jezova
et al. (2002) who found a lack of stress-related increase in
blood pressure.
Altogether, the results of this study suggest that EGb
761 has a specic anxiolytic eect that is dose-dependent
and signicantly exceeds the placebo eect commonly seen
in trials of psychoactive drugs. Taking into account that
anxiety in early stages of dementing disorders with perceived cognitive decline may to some extent be related to
activation of the HPA axis, as in GAD, and is reactive in
nature to some extent, as in ADWAM, it appears to be
worth considering EGb 761 for the treatment of this type
of anxiety. With its excellent tolerability, the absence of a
risk for dependence, lack of adverse impact on vigilance
and cognitive functioning (according to general experience), the drug is not only suitable for elderly patients,
but also for young people at work.

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

References
Allgulander C, Bandelow B, Hollander E, Montgomery SA, Nutt DJ,
Okasha A, et al. World Council of Anxiety. WCA recommendations
for the long-term treatment of generalized anxiety disorder. CNS
Spectrums 2003;8(Suppl. 1):5361.
Allgulander C, Dahl AA, Austin C, Morris PLP, Sogaard JA, Fayyad R,
et al. Ecacy of sertraline in a 12-week trial for generalized anxiety
disorder. The American Journal of Psychiatry 2004;161:16429.
American Psychiatric Association. Diagnostic and statistical manual of
mental disorders (3rd ed. rev.) DSM-III-R. Washington (DC): American Psychiatric Association; 1987.
Ansseau M, Dierick M, Buntinkx F, Cnockaert P, De Smedt J, Van Den
Haute M, et al. High prevalence of mental disorders in primary care.
Journal of Aective Disorders 2004;78:4955.
Bauer P, Kohne K. Evaluation of experiments with adaptive interim
analyses. Biometrics 1994;50:102941.
Birks J, Grimley Evans J. Ginkgo biloba for cognitive impairment and
dementia. The Cochrane Database of Systematic Reviews 2002, Issue
4. Art. No.: CD003120. DOI: 10.1002/14651858.CD003120.
Chan DC, Kasper JD, Black BS, Rabins PV. Prevalence and correlates of
behavioral and psychiatric symptoms in community-dwelling elders
with dementia or mild cognitive impairment: the Memory and Medical
Care Study. International Journal of Geriatric Psychiatry
2003;18:17482.
Cieza A, Maier P, Poppel E. Eects of Ginkgo biloba on mental
functioning in healthy volunteers. Archives of Medical Research
2003;34:37381.
CIPS Collegium Internationale Psychiatriae Scalarum, editor. Internationale Skalen fur Psychiatrie [International scales for psychiatry]. Gottingen: Beltz Test; 1996.
de Jonghe F, Swinkels J, Tuynman-Qua H, Jonkers F. A comparative
study of suriclone, lorazepam and placebo in anxiety disorder.
Pharmacopsychiatry 1989;22:26671.
Deussing JM, Wurst W. Dissecting the genetic eect of the CRH system
on anxiety and stress-related behaviour. Comptes Rendus Biologies
2005;328:199212.
Eckmann F, Schlag H. Kontrollierte Doppelblind-Studie zum Wirksamkeitsnachweis von Tebonin forte bei Patienten mit zerebrovaskuarer Insuzienz [Controlled, double-blind study for
proving the ecacy of Tebonin forte in patients with cerebrovascular
insuciency]. Fortschritte der Medizin 1982;100:14748.
Forsell Y, Palmer K, Fratiglioni L. Psychiatric symptoms/syndromes
in elderly persons with mild cognitive impairment. Data
from a cross-sectional study. Acta Neurologica Scandinavica
2003;107(Suppl. 179):258.
Habib KE, Weld KP, Rice KC, Pushkas J, Champoux M, Listwak S,
et al. Oral administration of corticotropin-releasing hormone receptor
antagonist signicantly attenuates behavioral, neuroendocrine and
autonomic responses to stress in primates. Proceedings of the National
Academy of Sciences of the United States of America 2000;97:607984.
Hemmeter U, Annen B, Bischof R, Bruderlin U, Hatzinger M, Rose U,
et al. Polysomnographic eects of adjuvant Ginkgo biloba therapy in
patients with major depression medicated with trimipramine. Pharmacopsychiatry 2001;34:509.
Hoerr R. Behavioural and psychological symptoms of dementia (BPSD):
eects of EGb 761. Pharmacopsychiatry 2003;36(Suppl. 1):S5661.
Holsboer F. CRHR1 antagonists as novel treatment strategies. CNS
Spectrums 2001;6:5904.
Jezova D, Duncko R, Lassanova M, Kriska M, Moncek F. Reduction of
rise in blood pressure and cortisol release during stress by Ginkgo
biloba extract (EGb 761) in healthy volunteers. Journal of Physiology
and Pharmacology 2002;53:33748.
Junger A. Zur Gultigkeit von selbstabnehmbaren psychometrischen und
psychopathometrischen Verfahren in der zahnarztlichen Praxis. Vergleich von Trait- und State-Verfahren. Dissertation at the Faculty of
Medicine. Erlangen: University; 1982.

479

Kapczinski F, Lima MS, Souza JS, Cunha A, Schmitt R. Antidepressants


for generalized anxiety disorder. The Cochrane Database of Systematic Reviews 2003, Issue 2. Art. No.: CD003592. DOI: 10.1002/
14651858.CD003592.
Kanowski S, Herrmann WM, Stephan K, Wierich W, Horr R. Proof of
ecacy of the Ginkgo biloba special extract EGb 761 in outpatients
suering from mild to moderate primary degenerative dementia of the
Alzheimer type or multi-infarct dementia. Pharmacopsychiatry
1996;29:4756.
Kieser M, Bauer P, Lehmacher W. Inference on multiple endpoints in
clinical trials with adaptive interim analyses. Biometrical Journal
1999;41:26177.
Le Bars PL, Katz MM, Berman N, Itil TM, Freedman AM, Schatzberg
AF. for the North American EGb Study Group. A placebo-controlled,
double-blind randomized trial of an extract of Ginkgo biloba for
dementia. Journal of the American Medical Association
1997;278:132732.
Lepine JP, Lellouch J. Classication and epidemiology of anxiety
disorders. In: Darcourt G, Mendlewicz J, Racagni G, Brunello N,
editors. Current therapeutic approaches to panic and other anxiety
disorders. Karger: Basel; 1994. p. 114.
Luthringer R, dArbigny P, Macher JP. Ginkgo biloba extract (EGb 761),
EEG and event-related potentials mapping prole. In: Christen Y,
Courtois Y, Droy-Lefaix MT, editors. Advances in Ginkgo biloba
extract research, vol 4. Eects of Ginkgo biloba extract (EGb 761) on
aging and age-related disorders. Paris: Elsevier; 1995. p. 10718.
Lyketsos CG, Lopez O, Jones B, Fitzpatrick AL, Breitner J, DeKosky S.
Prevalence of neuropsychiatric symptoms in dementia and mild
cognitive impairment. Results from the cardiovascular health study.
Journal of the American Medical Association 2002;288:147583.
Marcilhac A, Dakine N, Bourhim N, Guillaume V, Grino M, Drieu K,
et al. Eect of chronic administration of Ginkgo biloba extract or
ginkgolide on the hypothalamicpituitaryadrenal axis in the rat. Life
Sciences 1998;62:232940.
Meoni P, Hackett D, Lader M. Pooled analysis of venlafaxine XR ecacy
on somatic and psychic symptoms of anxiety in patients with
generalized anxiety disorder. Depression and Anxiety 2004;19:12732.
Mix JA, Crews WD. A double-blind, placebo-controlled, randomized trial
of Ginkgo biloba extract EGb 761 in a sample of cognitively intact
older adults: neuropsychological ndings. Human Psychopharmacology Clinical & Experimental 2002;17:26777.
Oken BS, Storzbach DM, Kaye JA. The ecacy of Ginkgo biloba on
cognitive function in Alzheimer disease. Archives of Neurology
1998;55:140915.
Porsolt RD, Martin P, Fromage S, Lene`gre A, Drieu K. Anti-stress eects
of EGb 761 in rodent models. In: Cristen Y, Costentin J, Lacour M,
editors. Eects of Ginkgo biloba extract (EGb 761) on the central
nervous system. Paris: Elsevier; 1992. p. 13545.
Rapin JR, Lamproglou I, Drieu K, DeFeudis FV. Demonstration of the
anti-stress activity of an extract of Ginkgo biloba (EGb 761) using a
discrimination learning task. General Pharmacology 1994;25:100916.
Sansone RA, Hendricks CM, Gaither GA, Reddington A. Prevalence of
anxiety symptoms among a sample of outpatients in an internal
medicine clinic: a pilot study. Depression and Anxiety 2004;19:1336.
Schulz J, Halama P, Hoerr R. Ginkgo biloba extracts for the treatment of
cerebral insuciency and dementia. In: van Beek T, editor. Ginkgo
biloba. Amsterdam: Harwood; 2000. p. 34570.
Semaan W, Hergueta T, Bloch J, Charpak Y, Duburcq A, Le Guern ME,
tude transversale de la prevalence du trouble de ladaptation
et al. E
avec anxiete en medecine generale [Cross-sectional study of the
prevalence of adjustment disorder with anxiety in general practice].
LEncephale 2001;27:23844.
Strain JJ, Smith GC, Hammer JS, McKenzie DP, Blumeneld M, Muskin
P, et al. Adjustment disorder: a multisite study of its utilization and
interventions in the consultation-liaison psychiatry setting. General
Hospital Psychiatry 1998;20:13949.
Strohle A, Holsboer F. Stress responsive neurohormones in depression
and anxiety. Pharmacopsychiatry 2003;36(Suppl. 3):S20714.

480

H. Woelk et al. / Journal of Psychiatric Research 41 (2007) 472480

van der Klink JJL, van Dijk FJH. Dutch practice guidelines for managing
adjustment disorders in occupational and primary health care.
Scandinavian Journal of Work, Environment & Health 2003;29:47887.
van Gaalen MM, Stenzel-Poore MP, Holsboer F, Steckler T. Eects of
transgenic overproduction of CRH on anxiety-like behaviour. The
European Journal of Neuroscience 2002;15:200715.

Ward CP, Redd K, Williams BM, Caler JR, Luo Y, McCoy JG. Ginkgo
biloba extract: cognitive enhancer of antistress buer. Pharmacology,
Biochemistry and Behavior 2002;72:91322.
Woelk H, Kapoula O, Lehrl S, Schroter K, Weinholz P. Behandlung von
Angst-Patienten. Doppelblindstudie: Kava-Spezialextrakt versus Benzodiazepine. Zeitschrift fur Allgemeinmedizin 1993;69:2717.

Вам также может понравиться