Reboxetine in a neuroendocrine challenge paradigm: evidence for

Reboxetine in a neuroendocrine challenge
paradigm : evidence for high cortisol responses in
healthy volunteers scoring high on subclinical
depression
A RT I C L E
International Journal of Neuropsychopharmacology (2000), 3, 193–201. Copyright # 2000 CINP
Juergen Hennig, Natalie Lange, Anja Haag, Sonja Rohrmann and Petra Netter
Center for Psychobiology and Behavioral Medicine, Department of Psychology, University of Giessen, Germany
Abstract
Received 16 February 2000 ; Reviewed 30 April 2000 ; Revised 11 June 2000 ; Accepted 13 June 2000
Key words : Reboxetine, cortisol, depression, heart rate, healthy subjects.
Introduction
Reboxetine [(RS)-2-[(RS)-α(2-ethoxyphenoxy)benzyl]morpholine methanesulphonate], has been introduced as a
highly specific noradrenaline reuptake inhibitor (SNRI) for
treatment of depression with a similar effectiveness as
imipramine but with fewer side-effects (Berzewski et al.,
1997 ; Mucci, 1997), less frequent drug–drug interactions
(Dostert et al., 1997) and no influence on vital signs or
laboratory parameters (Burrows et al., 1998). Moreover, it
has been shown to be superior to fluoxetine especially in
Address for correspondence : Dr J. Hennig, Department of
Psychology, University of Giessen, Otto-Behaghel-Str. 10, D-35394
Giessen, Germany.
Tel. : j49 641 99 26 154 Fax : j49 641 99 26 159.
E-mail : juergen.hennig!psychol.uni-giessen.de
severely depressed patients (Massana, 1998), and effective in long- and short-term treatment of depression
(Berzewski et al., 1997 ; Montgomery, 1997). It reduces
relapse rates in patients with recurrent depression
(Versiani et al., 1999). Its specificity has been demonstrated by the lowest IC for norepinephrine followed by
&!
serotonin and then dopamine (Massana, 1998). In the rat
brain there is no effect on MAO-A and very low influence
on MAO-B activity (Dostert et al., 1997). Furthermore, no
or very little receptor binding (Ki 10 000 n) has been
demonstrated for α -, α -, β -, β -, NMDA, cholinergic,
" # " #
dopaminergic, or serotonergic receptors.
In contrast to studies relating to the serotonin system,
neuroendocrine challenge tests specific for the norepinephrine system with cortisol as an indicator of responsiveness have not been performed with any frequency.
However, several approaches to increase central nor-
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
This paper investigates if the highly selective norepinephrine reuptake inhibitor reboxetine leads to a dosedependent cortisol release and if this response depends on personality dimensions related to clinical depression
in healthy volunteers. Twenty-four male subjects received placebo, 2 mg, or 4 mg reboxetine in a balanced,
randomized cross-over study. Cortisol was measured in saliva at six different time-points according to the
kinetics of the drug. Furthermore, several measurements of cardiovascular parameters, emotional states, and
possible side-effects were obtained. Subjects were divided into two groups scoring above or below the median
of a depressiveness questionnaire scale [n l 11, low (Dk) ; n l 13, high (Dj)]. Results clearly demonstrated,
that reboxetine stimulates cortisol release. Whereas blood pressure was not affected, heart rate increased after
2 and 4 mg but not dose dependently. Subjects reported more non-specific arousal while the dimensions of
tiredness–wakefulness and positive–negative emotional states were not affected by the drug. Somatic
complaints were low and only non-specific complaints were statistically elevated but of negligible amount.
Subjects classified as Dj can be characterized as high responders to the drug. This is especially true not only
for cortisol increases but also for changes in heart rate and some ratings on physical complaints. Hot flushes,
sweating and a throbbing sensation in blood vessels in the head were observed in Dj but only with the 4 mg
dose. The results clearly demonstrate that reboxetine stimulates cortisol release and heart rate and that this is
particularly pronounced in subjects scoring high on depression-related personality dimensions. Reboxetine,
therefore, is a promising tool for investigating neuroendocrine response to noradrenergic challenge tests. The
question whether increased responses in Dj are due to an up-regulation of receptor sensitivity as a
consequence of low norepinephrine supply is discussed.
194
J. Hennig et al.
related to depression. Moreover, it was of interest to
investigate reboxetine-induced changes in ratings on
emotional states and physical complaints as well as on
peripheral sympathetic activation (heart rate, blood
pressure) in relation to personality.
Methods
Subjects
Twenty-four healthy male subjects aged between 20 and
38 yr (meanp.., 24n8p4n3) participated in the study.
Prior to the main experiment subjects were carefully
checked for their health status. Any kind of acute or
chronic disease led to rejection of the subject. Furthermore, all participants were drug-free, non-smokers,
and reported no regular intake of alcohol. They received
thorough information about the objectives and procedure
of the study and about possible side-effects of the drug.
According to ethical regulations subjects were allowed to
quit the experiment any time they wished. This study was
approved by the Ethics Committee of the German
Psychological Association. After all information was
presented and no further questions posed by the subjects
they signed informed consent ; three test dates, 1 wk
apart, were then fixed for the main experiment according
to the cross-over design of this study (see below).
Design, drug and experimental procedure
The study was performed as a randomized, double-blind
cross-over design with three treatment levels. Four
subjects were each randomly assigned to one of the
resulting six possible orders of drug condition (placebo,
2 mg, 4 mg reboxetine) (Edronax2, Pharmacia & Upjohn)
as a single oral dose. The time interval of exactly 1 wk
was chosen between each session as it is sufficient for
complete washout since the elimination half life of
reboxetine is 13 h (Edwards et al., 1995). All experiments
started at 15 : 00 hours to control for circadian rhythmicity
of cortisol. After arrival subjects were asked to go the rest
rooms. Afterwards they completed a questionnaire on
emotional states and physical complaints. The former is a
slightly modified form of the mood adjective check list by
Janke and Debus (1978) and consists of 18 items on
different emotional states. Since many items are correlated, a factor analysis was computed for the purpose of
data reduction. This led to three clearly separated factors :
tiredness–wakefulness, positive–negative emotional state,
and arousal–relaxation, explaining 77 % of the total
variance. The questionnaire on physical complaints consists of 34 items on different somatic areas related to
possible side-effects of drugs. This questionnaire is much
more heterogenous and has no clear factor structure. In
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
adrenergic function specifically by MAO inhibition,
uptake inhibition and α -adrenoreceptor antagonism have
#
been tried. However, the reversible MAO inhibitor
moclobemide did not yield cortisol responses after a
single oral dose in healthy subjects (Koulu et al., 1989)
supporting the results and conclusions of Slater et al.
(1977) and Steiger et al. (1993). Also, cyclic antidepressants such as nomifensine obviously fail to stimulate
the hypothalamic–pituitary–adrenal (HPA) axis (Alagna
and Masala, 1979 ; Culig et al., 1983). On the other hand,
tricyclic antidepressants, e.g. desipramine, have been
demonstrated to increase cortisol concentrations (Laakmann et al., 1990 ; Torpy et al., 1995) but are not specific
for noradrenaline reuptake. Finally, specific α -receptor
#
antagonists, e.g. mianserin, obviously seem to be ineffective in stimulating the HPA axis (Tormey et al.,
1980) while mirtazapine significantly reduces cortisol
levels compared to placebo (Laakmann et al., 1999). The
clonidine (α -agonist) challenge test has only been
#
evaluated with respect to growth hormone (GH) in
healthy subjects (Tancer et al., 1990) as well as in
depressed patients (Charney et al., 1982 ; Checkley et al.,
1984 ; Gann et al., 1995 ; Lesch et al., 1988 ; Matussek,
1986 ; Ryan et al., 1994 ; Siever et al., 1986) but it does
not lead to an activation of the HPA axis and, therefore,
no changes in cortisol are observed (Boyar et al., 1980 ;
Hoehe et al., 1988 ; Tancer et al., 1990).
Results from neurotransmitter challenge tests can be
extrapolated into the range of healthy personality traits.
In several studies related to the serotonin system we were
able to demonstrate that subjects with low scores on
questionnaires measuring low satisfaction with life, or
related dimensions such as harm avoidance, which can
both be regarded as indicators of subclinical depression,
show blunted cortisol responses to combined d-fenfluramine and ipsapirone challenge tests (Hennig et al.,
2000). There is much evidence form the literature that a
common biological basis may be assumed for personality
and related diseases. This, however, does not mean that
psychopathology can be predicted by personality
measurements exclusively, but a predisposition cannot be
ruled out. For instance, subjects scoring high on such
personality traits as aggressiveness, sensation seeking and
impulsiveness are characterized by blunted prolactin
responses after d-fenfluramine (Netter et al., 1996, 1999)
and therefore reflect what Cocarro (1989) has described for aggressive–impulsive personality disorder ;
this demonstrates that the extrapolation from personality
traits to clinical phenomena is not restricted to affective
disorders but also applies to personality disorders.
The present study was mainly conducted to investigate
whether reboxetine leads to cortisol responses and
whether these depend, in part, on personality dimensions
Reboxetine and neuroendocrine challenge
195
Table 1. Means, standard errors, t values and levels of significance for personality dimensions according to the classification of
Dk and Dj
Dj
Dk
Trait
Mean
...
Mean
...
t
p
Extraversion
Neuroticism
Altruism
Ambitiousness
Inhibition
Irritability
Aggression
Feelings of being stressed
Psychosomatic complaints
Worries about health
Openness
6n53
8n30
5n76
7n00
6n07
6n30
4n92
7n23
3n30
4n15
8n53
1n08
0n87
0n92
0n85
0n97
0n97
0n64
0n87
0n71
0n69
0n65
8n72
3n00
6n54
7n36
2n90
3n45
3n36
4n00
1n09
3n45
8n36
1n00
0n57
1n07
0n93
0n84
0n49
0n76
0n82
0n36
0n75
0n49
k1n45
4n85
k0n55
k0n28
2n41
2n46
1n59
2n66
2n59
0n79
0n20
ns
0n05
0n05
ns
0n05
0n05
ns
ns
discrimination between subjects characterized by traits
which had previously been demonstrated to be associated
with vulnerability to depression (Van Praag, 1998). The
comparison between two groups of depressives was
chosen to demonstrate possible differences in the timecourse of responses to reboxetine which could not be
demonstrated by using correlations between personality
and mean changes of response measures.
At 15 : 45 hours subjects received placebo or reboxetine
(2 or 4 mg) which was administered in identical capsules
to maintain blindness of the subjects and the experimenters (N. L. and A. H.). Ratings on emotional states
were obtained at j45, j105 and j135 min after drug
intake. At the expected reboxetine plasma peak
(j105 min) subjects rated physical complaints again.
Saliva was sampled at j25, j55, j85, j105 and
j135 min after drug intake and cardiac parameters were
measured at j105 and j135 min after drug intake. At
the end of the experiment (18 : 15 hours) all saliva samples
were centrifugated and frozen at k20 mC until assayed
by a commercial radioimmunoassay (RIA) (ChironDiagnostics, Fernwald, Germany) as described elsewhere
(Kirschbaum et al., 1989). The coefficients of variance for
intra- and inter-assay precision were lower than 5 % and
lower than 8 %, respectively.
Statistical evaluation
Since the study was conducted as a cross-over design, 3factor ANOVAs were computed with the two repeated
measurement factors ‘ treatment ’ (three levels : placebo,
2 mg, and 4 mg reboxetine), ‘ time ’ (six measurements)
and a third independent factor (Dj and Dk). Dependent
variables were cortisol, emotional states and cardiac
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
addition saliva samples collected with the Sarstedt salivette (Sarstedt, Rommelsbach, Germany) were obtained.
The first saliva sample and questionnaire scores were
defined as baseline measurements, Immediately afterwards, blood pressure and heart rate were measured by an
automatic device (Tonomed\Tonoprint, Speidel & Keller,
Germany). To enhance the reliability of the measurements
three consecutive measures of blood pressure and heart
rate were obtained, the means of which were used for
further analyses. Subjects were then asked to complete
personality questionnaires. One was the Freiburg Personality Inventory (Fahrenberg et al., 1984) which contains
the major dimensions of neuroticism and extraversion.
However, both secondary factors consist of several
primary factors. According to the therapeutical use of
reboxetine we were interested in differential effects of the
drug related to personality traits associated with clinical
depression. The one of interest in this study is ‘ satisfaction
with life ’, which is an important correlate of neuroticism.
Low satisfaction with life (e.g. ‘ I do not feel optimistic
when viewing the future ’ or ‘ I am deeply unsatisfied with
my social relations ’ or ‘ I often feel sad and depressed ’)
represents a form of subclinical depression in healthy
samples. Subjects were divided into high (Dj, n l 13)
and low (Dk, n l 11) depressives according to below
( 7) or above ( 7) median values. Scores obtained by
our subjects ranged from 0 to 12 which represents the
full range of the scale. As expected, results of t test
comparisons depicted in Table 1 show that subjects
differing in the particular subscale of satisfaction with life
were also markedly different from each other with respect
to other personality dimensions related to neuroticism
and depression. Therefore, dividing our sample into high
and low scorers on satisfaction with life leads to a valid
0n001
ns
ns
196
J. Hennig et al.
parameters. However, levels for the factor ‘ time ’ were
different for cortisol (6), emotional states (4), and heart
rate and blood pressure (3). Since physical complaints
were rated only twice, difference scores between the
measurements obtained at j105 min minus baseline
values were computed and statistically tested by 2-factor
ANOVA for repeated measures (treatment) and an
independent factor Dj vs. Dk. All effects with an error
probability (p 0n05) were considered significant. All
tests of significance were two-tailed. Furthermore, effect
sizes [eta# (η#)] are given for each significant effect to
demonstrate the amount of variance explained by the
specific effect (ranging between 0 and 1).
Results
2 mg
4 mg
Cortisol in saliva (nmol/I)
Placebo
Time (min) after drug intake (0)
Figure 1. Means and standard errors of saliva cortisol levels according to treatment and high (
) vs. low () depression scores
as a personality trait.
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
A single oral dose of reboxetine induced a highly
significant increase in cortisol concentrations (F l 11n64,
d.f. l 2, p 0n0001, η# l 0n35) which started 60 min
after drug intake irrespective of dose (Figure 1).
Comparison of means and standard errors for the mean
values across all measurements within each condition
reveals that the increases (nmol\l) after 2 and 4 mg do not
differ (placebo, 2n92p0n41 ; 2 mg, 5n98p0n8 ; 4 mg,
6n84p0n85) from each other. Although the mean increase
after the 4 mg dose tends to be higher a statistically
conclusive dose–response relationship cannot be demonstrated. Of more interest, subjects scoring high on
dispositional depression can be identified as high
responders in both drug conditions. While no difference
between the Dj and Dk groups is observed within
the placebo condition, Dj reveals higher cortisol
levels (nmol\l) than Dk across all measurements in
both reboxetine conditions [2 mg (Dj, 6n21p1n01 ;
Dk, 5n79p1n18) ; 4 mg (Dj, 7n94p1n15 ; Dk,
5n68p1n24)]. However, the most striking finding is that
Dj shows a much faster and more pronounced response
than Dk after 2 and 4 mg reboxetine. While Dk does
not respond at all before 85 min, cortisol levels are both
clearly and highly significantly increased in Dj at that
time-point. This different time-course and the size of the
response leads to a highly significant interaction effect
between time-course (measurements) and the group factor
Dj\Dk (F l 4n58, d.f. l 5, p 0n01, η# l 0n17).
Since this pattern of responses of Dj and Dk can be
obtained after both doses the interaction effect between
dose, measurements and the group factor is not significant
(F l 1n153, d.f. l 10, p l ns, η# l 0n05). In summary,
one can clearly conclude that reboxetine stimulates
cortisol release, and that this is much more pronounced in
subjects scoring high on depression.
Evaluation of blood pressure results in a lack of any
reboxetine effects. Means of systolic blood pressure
were neither different between the active drug and
placebo nor between the two doses of reboxetine
(placebo, 123n72p2n07 ; 2 mg, 124n05p1n82 ; 4 mg,
123n45p2n47) leading to a non-significant F value (F l
0n056, d.f. l 2). Moreover, no significant main effect or
interaction with drug treatment or time-course could be
found for personality. The same is true for diastolic blood
pressure which does not change (F l 1n64, d.f. l 2,
p l ns) after treatment with reboxetine (placebo,
Reboxetine and neuroendocrine challenge
2 mg
4 mg
Heart rate (beats/min)
Placebo
197
Figure 2. Means and standard errors of heart rate according to treatment and high (
) vs. low () depression scores as a
personality trait.
Score in questionnaire
(0, not at all; 6, extremely)
6
Placebo
2 mg
4 mg
5
4
3
2
1
0
Tiredness
Negative
emotions
Arousal
Factors
Figure 3. Means and standard errors of ratings on emotional
states according to treatment.
75n79p1n95 ; 2 mg, 78n68p2n17 ; 4 mg, 77n79p2n60)
and does not produce any significant effects for timecourse or interaction between dose, time-course and
personality.
However, reboxetine leads to a small but significant
increase in heart rate (F l 8n91, d.f. l 2, p 0n01, η# l
0n28). The overall means (across all measurements) are
65n30p1n93, 70n08p1n89, 70n91p1n94 for placebo, 2
and 4 mg, respectively. However, more interestingly,
heart rate increases are dependent on personality within
the 4 mg dose condition. As indicated in Figure 2, Dj
reveals heart rate accelerations while Dk remains
approximately constant. This interaction between group,
drug condition and time-course leads to a significant F
value of 2n56 (d.f. l 4, p 0n05, η# l 0n10). Again, Dj
reveals a higher responsiveness after treatment with
reboxetine, but in contrast to the results demonstrated for
cortisol, higher doses are needed for changes in heart rate.
Emotional states are only marginally affected by
reboxetine. While no treatment effect could be found for
either the tiredness–wakefulness dimension or the
negative–positive effect, arousal is increased after treatment with reboxetine (F l 3n23, d.f. l 2, p 0n05, η# l
0n12). Figure 3 demonstrates that 2 and 4 mg reboxetine
only slightly increase arousal ratings. Moreover, a
dose–response relationship cannot be demonstrated. With
respect to the whole range of the scale used, the effect
described should not be overestimated (placebo,
0n89p0n15 ; 2 mg, 1n16p0n18 ; 4 mg, 1n13p0n16). No
further significant main effects or interactions between
drug, measurements and personality could be observed
for emotional states.
Finally, ratings on physical complaints were very low
after treatment with reboxetine (see Figure 4). The only
differences found were for hot flushes, sweating, a
throbbing sensation in blood vessels in the head, and nonspecific complaints (feeling uncomfortable).
Analysis of variance revealed a significant interaction
between treatment and group for hot flushes (F l 4n85,
d.f. l 2, p 0n05, η# l 0n18) whereas the pure treatment
effect is not significant. This result can be explained by the
fact that only Dj complains about hot flushes after 4 mg
of reboxetine which significantly differs from the ratings
obtained after placebo and the 2 mg dose. The same is
true for sweating (F l 3n68, d.f. l 2, p 0n05, η# l
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
Time (min) after drug intake (0)
198
J. Hennig et al.
Difference score
(a)
Total group
(b)
High
Low
(c)
High
Low
High
Low
(d)
High
Low
Total group
Figure 4. Means and standard errors of ratings on physical complaints according to placebo (P), 2 and 4 mg reboxetine for the
total group and in high vs. low depression scores as a personality trait. (a) Hot flushes ; (b) sweating ; (c) throbbing sensation in
blood vessels in the head ; (d) feeling of non-specific complaints.
0n14), and a throbbing sensation in blood vessels in the
head (F l 5n00, d.f. l 2, p 0n05, η# l 0n18). In contrast, feelings of non-specific complaints were increased
after 2 and 4 mg irrespective of personality (F l 3n71,
d.f. l 2, p 0n05, η# l 0n14) [see Figure 4(a–d)]. Taken
together, reboxetine has very few side-effects after an
acute dose of 2 or 4 mg. It should be noted that the
absolute amount of changes is low with respect to the
range of the scale (0, not at all ; 6, extremely). Moreover,
a higher responsiveness to the drug in Dj can also be
observed for some of these items.
Discussion
The present study was conducted to investigate whether
a single oral dose of reboxetine stimulates the HPA axis.
Both doses of 2 and 4 mg led to a highly significant
increase in cortisol. Moreover, this response could be
related to the personality dimension of ‘ low satisfaction
with life ’ which may reflect a subclinical model of
depression. However, since the kinetics of change in
cortisol after reboxetine were not previously described,
one must realize that the period of observation we used
was too short to draw the conclusion that subjects scoring
high on depression are characterized by higher responses
within the 4 mg condition. In other words, the differences
observed could be due to differences in pharmacokinetics.
To gain further insight into this matter we analysed
personality differences according to time-point of peak
cortisol concentrations and cortisol responses not affected
by different time-courses [area under the response curve
(AURC) ; measurements 2–6, minus 1]. Figure 5 illustrates
the results of these additional analyses.
As can be seen in Figure 5(a) the time-points of cortisol
peaks differ considerably between the drug conditions
(F l 3n09, d.f. l 2, p l 0n059, η# l 0n12) but not between personality groups within each condition (F l 0n36,
d.f. l 2, p l ns, η# l 0n01). This indicates that the timepoint of responses is not different between subjects
scoring high and low on ‘ low satisfaction with life ’ ; this
makes it unlikely that the differences described above are
purely due to differences in pharmacokinetics. An analysis
of the AURC confirms clear effects for dose (F l 5n79,
d.f. l 2, p 0n01, η# l 0n21) but fails to demonstrate a
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
Total group
Total group
(a)
199
(b)
Area under the curve for
cortisol responses
Time of peak cortisol responses
(min/after drug intake)
Reboxetine and neuroendocrine challenge
Figure 5. Means and standard errors of (a) times of peak cortisol responses and (b) area under the response curve, for subjects
scoring high (
) and low () on depression.
satisfaction with life ’ is one of the most potent primary
factors of neuroticism and from early clinical studies it is
known that some depressed patients also exhibit low
levels of MHPG (Maas, 1968). In addition, low levels of
MHPG predicted responses to imipramine (Beckmann and
Goodwin, 1975) whereas normal or high levels were
predicted for the clinical outcome after amitriptyline
(Modai et al., 1979).
With respect to the higher responsiveness (cortisol,
heart rate, some ratings on physical complaints and
subjective arousal) to reboxetine in Dj, one can speculate
that low levels of norepinephrine in the CNS may have
resulted in an up-regulation of the sensitivity of postsynaptic adrenoreceptors. This idea receives some support, since it is a well-established fact that chronic
antidepressive treatment decreases β-adrenergic receptor
responsiveness, which has been claimed to refer particularly to the β -subtype (Zemlan and Garver, 1990).
"
Following this argument one would assume that the β "
receptor sensitivity is elevated in untreated depressed
patients and probably in healthy volunteers scoring high
on dispositional depression as well.
The highly selective norepinephrine reuptake inhibitor
reboxetine stimulates the HPA axis and seems to be
promising for neuroendocrine challenge paradigms since
a differential responsivity could be demonstrated for
subjects scoring high on personality traits related to
depression. Studies on patients are needed which should
obtain different indices for neurotransmitter responsivity,
availability, and receptor sensitivity because correspondence between results obtained from different approaches
are definitely needed for a better understanding of
underlying mechanisms.
Acknowledgements
This study was supported by Pharmacia & Upjohn. We
are greatly indebted to Dr Leroux and Dr Schu$ ler for
many constructive comments and their personal involvement in the study.
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
significant interaction between dose and personality
groups (F l 1n41, d.f. l 2, p l ns, η# l 0n06). However,
an isolated comparison (t test for independent groups) of
the AURC between Dj and Dk for the 2 mg dose only
marginally fails to reach statistical significance (t l 1n68,
d.f. l 1, p l 0n09). The results obtained so far seem to be
sufficient to reject the alternative hypothesis that pure
pharmacokinetics may explain the personality differences
in responses after 2 mg of reboxetine. However, in order
to reject this alternative conclusion for the 4 mg dose, a
longer period of sampling is required.
Increases in central availability of norepinephrine are
known to stimulate the HPA axis as demonstrated by
CNS infusion studies. For example, an i.c.v. injection of
phenylephrine (α -adrenergic receptor agonist) leads to a
"
marked corticosteroid response in the rat, which could be
antagonized by an i.c.v. treatment with the selective α "
antagonist prazosin. The same study demonstrates that
i.c.v. applications for isoproterenol (β-adrenergic receptor
agonist) also stimulate the HPA axis and that the βreceptor antagonist propanolol abolished this effect
(Bugajski et al., 1995). Adrenergic receptors therefore play
an important role in the regulation of the HPA axis.
Peripheral administration of epinephrine or norepinephrine was shown to be ineffective with respect to adrenocorticotropic hormone (ACTH) and cortisol responses in
healthy male volunteers (Oberbeck et al., 1996). It can
thus be assumed that inhibition of norepinephrine reuptake induced by reboxetine leads to a stimulation of
adrenoreceptors in the CNS which in turn affects the
HPA axis.
Cortisol responses after reboxetine were shown to be
dependent on personality traits. There is much evidence
from clinical and non-clinical studies that norepinephrine
relates to personality. For example, the early study by
Ballenger (1983) demonstrated that subjects scoring high
on neuroticism exhibit low levels of 3-methoxy-4hydroxy-phenylglycol (MHPG) in plasma, which indicates low levels of norepinephrine in the brain. ‘ Low
200
J. Hennig et al.
References
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
Alagna S, Masala A (1979). Cortisol secretion following
nomifensine administration in normal subjects. Biomedicine
31, 189–191.
Ballenger JC (1983). Biochemical correlates of personality
traits in normals : an exploratory study. Personality and
Individual Differences 4, 615–625.
Beckmann H, Goodwin FK (1975). Antidepressant response to
TCA’s and urinary MHPG in unipolar depression. Archives
of General Psychiatry 32, 17–21.
Berzewski H, Van-Moffaert M, Gagiano CA (1997). Efficacy
and tolerability of reboxetine compared with imipramine in
a double-blind study in patients suffering from major
depressive episodes. European Journal of
Neuropsychopharmacology 7, S37–S47.
Boyar RM, Fixler DF, Kaplan NM, Graham RM, Price KP,
Chipman JJ, Laird WP (1980). Effects of clonidine on 24hour hormonal secretory patterns, cardiovascular
hemodynamics, and central nervous function in
hypertensive adolescents. Hypertension 2, 83–89.
Bugajski, J, Gadek MA, Olowska A, Borycz J, Glod R,
Bugajski AJ (1995). Adrenergic regulation of the
hypothalamic-pituitary-adrenal axis under basal and social
stress conditions. Journal of Physiology and Pharmacology 46,
297–312.
Burrows GD, Maguire KP, Norman TR (1998).
Antidepressant efficacy and tolerability of the selective
norepinephrine reuptake inhibitor reboxetine : a review.
Journal of Clinical Psychiatry 59, 4–7.
Charney DS, Heninger GR, Sternberg DE, Hafstad KM,
Giddings S, Landis DH (1982). Adrenergic receptor
sensitivity in depression. Effects of clonidine in depressed
patients and healthy subjects. Archives of General Psychiatry
39, 290–294.
Checkley SA, Glass IB, Thompson C, Corn T, Robinson P
(1984). The GH response to clonidine in endogenous as
compared with reactive depression. Psychological Medicine
14, 773–777.
Coccaro EF (1989). Central serotonin and impulsive
aggression. British Journal of Psychiatry 8, 52–62.
Culig J, Ehsanullah RS, Hallett C, Iliopoulou A, Matheson I,
Turner P (1983). A clinical pharmacological comparison of
diclofensine (Ro 8-4650) with nomifensine and
amitriptyline in normal human volunteers. British Journal of
Clinical Pharmacology 15, 537–543.
Dostert P, Benedetti MS, Poggesi I (1997). Review of the
pharmacokinetics and metabolism of reboxetine, a selective
noradrenaline reuptake inhibitor. European Journal of
Neuropsychopharmacology 7, S23–S35.
Edwards DM, Pellizzoni C, Breuel HP, Berardi A, Castelli
MG, Frigerio E, Poggesi I, Rocchetti M, Dubini A, Strolin
BM (1995). Pharmacokinetics of reboxetine in healthy
volunteers. Single oral doses, linearity and plasma protein
binding. Biopharmaceutics and Drug Disposition 16, 443–460.
Fahrenberg J, Hampel R, Selg H (1984). Das Freiburger
PersoW nlichkeitsinventar (FPI-R) (4th edn). Go$ ttingen : Hogrefe.
Gann H, Riemann D, Stoll S, Berger M, Muller WE (1995).
Growth hormone response to growth hormone-releasing
hormone and clonidine in depression. Biological Psychiatry
38, 325–329.
Hennig J, Toll T, Schonlau P, Rohrmann S, Netter P (2000).
Endocrine responses after d-fenfluramine and ipsapirone
challenge : further support for Cloninger’s tridimensional
model of personality. Neuropsychobiology 41, 38–47.
Hoehe M, Valido G, Matussek N (1988). Growth hormone,
noradrenaline, blood pressure and cortisol responses to
clonidine in healthy male volunteers : dose–response
relations and reproducibility. Psychoneuroendocrinology 13,
409–418.
Janke W, Debus G (1978). Die EigenschaftswoW rterliste (EWL).
Go$ ttingen : Hogrefe.
Kirschbaum C, Strasburger CJ, Jammers W, Hellhammer DH
(1989). Cortisol and behavior : 1. Adaptation of a
radioimmunoassay kit for reliable and inexpensive salivary
cortisol determination. Pharmacology, Biochemistry and
Behavior 34, 747–751.
Koulu M, Scheinin M, Kaarttinen A, Kallio J, Pyykko K,
Vuorinen J, Zimmer RH (1989). Inhibition of monoamine
oxidase by moclobemide : effects on monoamine
metabolism and secretion of anterior pituitary hormones
and cortisol in healthy volunteers. British Journal of Clinical
Pharmacology 27, 243–255.
Laakmann G, Munz T, Hinz A, Voderholzer U (1990).
Influence of clenbuterol, a beta-adrenergic agonist, on
desipramine induced growth hormone, prolactin and
cortisol stimulation. Psychoneuroendocrinology 15, 391–399.
Laakmann G, Schule G, Baghai T, Waldvogel E (1999). Effects
of mirtazapine on growth hormone, prolactin, and cortisol
secretion in healthy male subjects. Psychoneuroendocrinology
24, 769–784.
Lesch KP, Laux G, Erb A, Pfuller H, Beckmann H (1988).
Growth hormone (GH) responses to GH-releasing hormone
in depression : correlation with GH release following
clonidine. Psychiatry Research 25, 301–310.
Maas JW, Fawcett J, Dekirmenjian H (1968). 3-methoxy-4hydroxy-phenylglycol (MHPG) execretion in depressive
states. Archives of General Psychiatry 19, 129–134.
Massana J (1998). Reboxetine versus fluoxetine : an overview
of efficacy and tolerability. Journal of Clinical Psychiatry 59,
8–10.
Matussek N (1986). Biological aspects of depression.
Psychopathology 19, 66–71.
Modai I, Apter RC, Golomb M, Wijsenbeek J (1979).
Response to amitriptyline and urinary MHPG in bipolar
depression. Neuropsychobiology 5, 184.
Montgomery SA (1997). Reboxetine : additional benefits to
the depressed patient. Journal of Psychopharmacology 11,
S9–S15.
Mucci M (1997). Reboxetine : a review of antidepressant
tolerability. Journal of Psychopharmacology 11, S33–S37.
Netter P, Hennig J, Rohrmann S (1999). Psychobiological
differences between the aggression and psychoticism
dimension. Pharmacopsychiatry 32, 5–12.
Reboxetine and neuroendocrine challenge
Tancer ME, Stein MB, Uhde TW (1990). Growth hormone
(GH) response to clonidine and growth hormone releasing
factor (GRF) in normal controls. Psychoneuroendocrinology 15,
253–259.
Tormey WP, Buckley MP, O’Kelly DA, Conboy J, Pinder
RM, Darragh A (1980). Sleep-endocrine profile of the
antidepressant mianserin. Current Medical Research and
Opinion 6, 456–460.
Torpy DJ, Grice JE, Hockings GI, Crosbie GV, Walters MM,
Jackson RV (1995). The effect of desipramine on basal and
naloxone-stimulated cortisol secretion in humans :
interaction of two drugs acting on noradrenergic control of
adrenocorticotropin secretion. Journal of Clinical
Endocrinology and Metababolism 80, 802–806.
Van Praag HM (1998). Anxiety and increased aggression as
pacemakers of depression. Acta Psychiatrica Scandinavica
(Suppl.) 393, 81–88.
Versiani M, Mehilane L, Gaszner P, Arnaud CR (1999).
Reboxetine, a unique selective NRI, prevents relapse and
recurrence in long-term treatment of major depressive
disorder. Journal of Clinical Psychiatry 60, 400–406.
Zemlan FP, Garver DL (1990). Depression and antidepressant
therapy : receptor dynamics. Progress in
Neuropsychopharmacology and Biological Psychiatry 14,
503–523.
Downloaded from http://ijnp.oxfordjournals.org/ by guest on September 25, 2016
Netter P, Hennig J, Roed IS (1996). Serotonin and dopamine
as mediators of sensation seeking behavior.
Neuropsychobiology 34, 155–165.
Oberbeck R, Schurmeyer T, Hosch W, Jeschmann JU, Schmidt
RE, Schedlowski M (1996). Epinephrine or norepinephrine
fail to influence pituitary-adrenal secretion in man. Hormone
and Metabolic Research 28, 142–146.
Ryan ND, Dahl RE, Birmaher B, Williamson DE, Iyengar S,
Nelson B, Puig AJ, Perel JM (1994). Stimulatory tests of
growth hormone secretion in prepubertal major depression :
depressed versus normal children. Journal of the American
Academy of Child and Adolescent Psychiatry 333, 824–833.
Siever LJ, Coccaro EF, Benjamin E, Rubinstein K, Davis KL
(1986). Adrenergic and serotonergic receptor
responsiveness in depression. Ciba Foundation Symposium
123, 148–163.
Slater SL, Lipper S, Shiling DJ, Murphy DL (1977). Elevation
of plasma-prolactin by monoamine-oxidase inhibitors.
Lancet 2, 275–276.
Steiger A, von-Bardeleben U, Guldner J, Lauer C, Rothe B,
Holsboer F (1993). The sleep EEG and nocturnal hormonal
secretion studies on changes during the course of
depression and on effects of CNS-active drugs. Progress in
Neuropsychopharmacology and Biological Psychiatry 17,
125–137.
201

Download Report