Download Hypothalamo-pituitary-adrenal function in patients with depressive

Journal of Psychiatric Research 37 (2003) 463–470
www.elsevier.com/locate/jpsychires
Hypothalamo-pituitary-adrenal function in patients with depressive
disorders is correlated with baseline cytokine levels, but not with
cytokine responses to hydrocortisone
Andreas Schuld*, Dagmar A. Schmid, Monika Haack, Florian Holsboer,
Elisabeth Friess, Thomas Pollmächer
Max Planck Institute of Psychiatry, Kraepelinstrasse 10, D-80804 Munich, Germany
Received 10 December 2002; received in revised form 10 March 2003; accepted 27 March 2003
Abstract
Dysfunction of the hyopthalamo-pituitary adrenal (HPA) system is frequently found in major depression. In addition, signs of
non-specific inflammatory system activation have been reported. However, very little is known about interactions between the HPA
and immune systems in depressive patients. To assess HPA system function, we performed a combined dexamethasone suppression
and corticotropin-releasing hormone stimulation (DEX/CRH) test in 14 depressive patients. Moreover, baseline nocturnal plasma
levels of the inflammatory cytokines interleukin-6 (IL-6) and tumor necrosis factor (TNF)-a were measured. In addition, the system
was challenged with an intraveneous pulsatile injection of hydrocortisone (1 mg/kg body weight in total) and again cytokine levels
were measured across one night. Baseline TNF-a levels were negatively correlated with the amount of ACTH released upon CRH
stimulation during the DEX/CRH test. Acute hydrocortisone administration suppressed TNF-a and IL-6 levels independently of
baseline HPA system activity. We conclude that chronic HPA system overactivity in depressed patients might compromise the
production of inflammatory cytokines under baseline conditions. However, the responsivity of the cytokine production to acutely
administered glucocorticoids does not seem to correlate with the state of the HPA system.
# 2003 Elsevier Ltd. All rights reserved.
Keywords: Stress; Cortisol; Interleukins; Tumor necrosis factor
1. Introduction
In patients suffering from major depression, dysregulation of the hypothalamo-pituitary-adrenal (HPA) system is one of the most consistent neurobiological
findings (Gold & Chrousos, 1999; Holsboer, 2000).
Though, in general, basal cortisol or adrenocorticotropic hormone (ACTH) plasma levels are normal or
only slightly enhanced, but increased 24-h excretion of
cortisol in urine (Rubinow et al., 1984) and increased
levels of corticotropin-releasing hormone (CRH) in cerebrospinal fluid (Nemeroff et al., 1984) have been
demonstrated. In vivo the extent of HPA system overactivity can be assessed by neuroendocrine functioning
tests such as the combined dexamethasone (DEX)
* Corresponding author. Tel.: +49-89-30622-1; fax: +49-8930622-562.
E-mail address: [email protected] (A. Schuld).
0022-3956/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0022-3956(03)00054-2
suppression and CRH stimulation (DEX/CRH) test: in
depressive patients, CRH administration at 15:00 h the
day following oral intake of 1.5 mg DEX at 23:00 h
results in an enhanced release of cortisol and ACTH,
whereas in most healthy people the stimulation of
ACTH and cortisol release is still suppressed. This is
thought to be due to a reduced hypothalamic feedback
sensitivity for glucocorticoids at the glucocorticoid
receptor and an increased release of endogeneous CRH
(Heuser et al., 1994). This neuroendocrine pathology is
closely linked to the neurobiological background of
various aspects of anxious and depressive symptomatology (Holsboer, 2000). The crucial role of HPA
abnormalities in major depression is strongly supported
by recent studies reporting beneficial effects of CRH-1
receptor antagonists on depression-like symptoms in
animals and healthy humans (Baram et al., 1996; Keck
et al., 2001) as well as in patients with major depression
(Zobel et al., 2000).
464
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
Cytokines are the most important humoral mediators
of the non-specific host response system. Moreover,
cytokines link the non-specific immune system and the
HPA system: inflammatory cytokines released during
infection or inflammation activate the HPA system at
the hypothalamic, pituitary, and adrenal level resulting
in the release of cortisol as the most important negative
feedback signal to prevent an overstimulation of ongoing host defense (Tilders et al., 1994; Bornstein &
Chrousos, 1999). Glucocorticoids strongly suppress the
production of proinflammatory cytokines in response to
various immune challenges in vitro and in vivo (Bratts
& Linden, 1996; Barber et al., 1993). Because alterations
of the non-specific immune system, in particular changes in the plasma levels of inflammatory cytokines such
as IL-6, have repeatedly been found in depressed
patients (Haack et al., 1999; Maes, 1999), it seems reasonable to hypothesize that immunological and neuroendocrine changes in depression might interact
(Leonard, 2000).
To investigate this interaction we studied cytokine
levels under baseline conditions and in response to the
acute anti-inflammatory actions of hydrocortisone. In
addition, we assessed the relationship of both the baseline levels and hydrocortisone-induced changes to the
function of the HPA system as characterized by the
combined DEX/CRH test. We hypothesized that cytokine baseline levels are lower in patients with a more severe
overactivity of the HPA system. It moreover seemed to us
very likely that a hydrocortisone injection might have
smaller immunosuppressive effects in these patients.
2. Materials and methods
2.1. Study sample
Fourteen patients (eight male, six female patients,
mean age 47.9 15.3 years) suffering from depressive
disorders based on DSM-IV criteria (American Psychiatric Association, 1994) were included after written
informed consent had been obtained. All subjects were
inpatients at the Max Planck Institute of Psychiatry.
They were severely depressed at the time of examination, as indicated by a mean Hamilton depression
(HAM-D) score (Hamilton, 1960) of 29.4 6.8. Nine of
the patients suffered from their first depressive episode,
whereas in the remaining five patients the index episode
was a recurrent major depressive episode. None of the
patients had received any psychotropic medication
within 7 days before they participated in the study protocol. Medical or neurological diseases were ruled out
by clinical examination and medical history, clinical
chemistry,
electrocardiogram
and
electroencephalogram. For more detailed description of the sample
see Table 1.
2.2. Experimental procedure
The study protocol was approved by an independent
ethics committee in accordance to the declaration of
Helsinki. After written informed consent has been
obtained from the patients, HPA function was assessed
by performing a combined DEX/CRH test using 1.5 mg
DEX at 23:00 h followed by CRH stimulation with 100
mg CRH (purchased from Clinalfa, Läufelfing, Switzerland) the following afternoon at 15:00 h. Then, the
patients spent three consecutive nights in the sleep
laboratory: after one night of adaptation to the laboratory conditions, the patients underwent two experimental nights. During these experimental nights, an
intravenous catheter was placed into an antecubital
forearm vein at 18:30 h. The line was kept patent with
0.9% saline solution containing heparin (400 IU/l).
Lights were turned off at 23:00 until 07:00 h and standard polysomnography was performed. The results of
sleep recording will be reported elsewhere. During the
experimental session, the patients were under continuous observation. An experienced physician was permanently on call and performed the hydrocortisone
injections in a single-blinded fashion. To avoid carryover effects, all patients received placebo injections during the first experimental night. During the second night
hydrocortisone (1 mg/kg body weight) was administered
in a pulsatile manner between 19:00 and 06:00 h. At
19:00 h the patient received a bolus containing 20% of
the respective hydrocortisone dosage, followed by consecutive hourly bolus injections in equal doses. Blood
samples were collected every hour between 19:00 and
07:00 h. Blood was stabilized with Na-EDTA (1 mg/ml
blood) and aprotinine (300 KIU/ml blood). Following
immediate centrifugation and aliquotation, plasma was
frozen to 20 or 80 C, respectively.
Table 1
Characteristics of the study sample
Patient
Age
Gender
BMI
DSM-IV diagnosis
HAM-D score
1
2
3
4
5
6
7
8
9
10
11
12
13
14
39
44
33
26
57
55
64
35
51
40
32
79
47
68
Male
Male
Male
Male
Male
Female
Female
Male
Male
Female
Female
Female
Male
Female
26.4
29.4
24.8
21.8
24.0
31.2
23.4
25.4
28.1
21.8
20.3
23.3
22.9
26.8
296.22
296.23
296.32
296.70
296.23
296.22
296.33
296.22
296.32
296.22
296.22
296.21
296.32
296.23
27
44
35
28
30
38
19
26
28
34
20
24
26
32
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
2.3. Determination of hormone and cytokine levels
The plasma levels of cortisol and ACTH were determined using coated-tube radioimmunoassays (cortisol:
ICN Biomedicals, Carson, California; ACTH: Nichols
Institute Diagnostics, San Juan Capistrano, California).
The limit of detection was 1.0 ng/ml for cortisol and 4.0
pg/ml for ACTH; the intra- and inter-assay coefficients
of variation were below 8%. IL-6 and TNF-a were
determined by enzyme-linked immunosorbent assays
(Medgenix Diagnostics, Brussels, Belgium). The limit of
detection was 2.0 and 3.0 pg/ml, respectively, for both
cytokines; the intra- and inter-assay coefficients of variation were below 8%.
2.4. Statistical methods
Data analysis was performed using commercially
available personal computer software (SPSS for Windows 10.0). For statistical analysis the AUC of ACTH
secretion following the injection of CRH during the
465
combined DEX/CRH test was computed as a parameter
representing HPA-function of the respective patient (see
Fig. 1). Because explorative analysis showed that comparable results were found by using the release of cortisol to CRH, just the results based on ACTH-release will
be described. For nocturnal hormone and cytokine
measurements the area under the curve (AUC) for the
various parameters was computed according to the trapezoid method. In addition, differences in the AUCs
between the placebo and verum conditions were computed. To reduce the amount of variables, the AUCs
were divided into three blocks of 4 h each, one covering
the time the patients stayed awake (from 19:00 to 23:00
h) and two equal periods of nocturnal sleep (from 23:00
to 03:00 and from 03:00 to 07:00 h).
For explorative analyses, Pearson’s or Spearman’s
correlation coefficients were used as appropriate. Characteristics of the patients like age, gender, or HAM-D
scores, the results of DEX/CRH-test and the amount of
plasma cortisol increase during hydrocortisone infusion
were correlated with the changes cytokine levels. Significant correlations identified were analyzed in more
detail by using analysis of variance (ANOVA) for repeated measures in a second step. For this purpose, the
sample was divided at the median of the respective
variable. Huynh–Feldt procedure was used for a-correction in every ANOVA model, post-hoc testing was
performed by students’ t-test, and P-values below 0.05
were considered significant. All data reported in the figures
show means standard error of the mean (SEM).
3. Results
3.1. Interactions of HPA function and cytokine levels
during baseline condition
Fig. 1. Secretion of cortisol (lower panel) and ACTH (upper panel) in
response to CRH during the combined DEX/CRH test (for methodological details see text). The sample was splitted based on the median
of ACTH release following CRH, values from the seven patients with
low ACTH-release are depicted with closed circles, whereas the values
of the patients with high ACTH-release are depicted with open circles.
Baseline TNF-a levels were significantly and negatively correlated with the amount of ACTH released
following CRH (Pearsons’s correlation coefficients:
r= 0.542, P < 0.05 for TNF-a levels between 19:00 and
23:00 h; r= 0.559, P < 0.05 between 23:00 and 03:00 h;
r= 0.522, P=0.055 between 03:00 and 07:00 h). The
patients who, following CRH, secrete an amount of
ACTH above the median displayed significantly lower
TNF-a levels during the placebo night than the patients
who secreted less. In contrast, the levels of cortisol and
IL-6 did not differ between these two groups (see Fig. 2).
ACTH release following CRH administration during
the combined DEX/CRH test was not significantly correlated with the placebo levels of cortisol and IL-6 nor
has a significant association of age, gender and HAM-D
score and the levels of circulating cortisol or cytokines
during the baseline night been found in correlation
analysis using Pearson’s or Spearman’s correlation
coefficients as appropriate (data not shown).
466
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
Fig. 2. Plasma levels of cortisol, TNF-a and IL-6 during placebo condition depending on the amount of ACTH secreted in response to CRH during
the combined DEX/CRH test. Left panel: Nocturnal plasma levels of cortisol, TNF-a and IL-6 in 14 depressive patients measured between 19:00
and 07:00 h during the placebo night. The patient sample is divided in two subgroups of seven subjects each, separated according to a median-split
of the ACTH response to CRH during the combined DEX/CRH test (closed squares—ACTH response below median; open squares—ACTHresponse above median). Right panel: For statistical analysis the area under the curve for the respective variables was computed between 19:00 and
23:00 h, 23:00 and 03:00 h and 03:00 and 07:00 h (white columns—ACTH response above median; black columns—ACTH response below median).
ANOVA for repeated measures revealed significant time effects for cortisol (F[2;24]=43.091, P<0.001) and TNF-a (F[2;24]=4.109, P <0.05), but
not for IL-6 (F[2;24]=1.528). No significant difference between the two subgroups has been found with respect to cortisol (F[1;12]=0.967) or IL-6
(F[1;12]=0.025) levels, whereas TNF-a levels significantly differ between the groups (F[1;12]=5.724, P< 0.05). Asterisks indicate significant differences in post-hoc t-test comparison.
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
3.2. Influences of hydrocortisone on cortisol, TNF- and
IL-6 plasma levels
As shown in Fig. 2, the pulsatile infusion of hydrocortisone caused robust increases in the plasma levels of
cortisol. The mean levels of cortisol remained fairly
stable between 19:00 and 07:00 h and were 1.5–2 times
higher than the morning levels present after the placebo
night. In parallel to increases in cortisol levels, hydrocortisone infusion significantly decreased the levels of
both TNF-a and IL-6. TNF-a levels were significantly
decreased during each time period analyzed. The significant overall time effect on IL-6 levels only was significant between 23:00 and 03:00 h (see Fig. 3).
3.3. Factors influencing the immunomodulatory effects
of hydrocortisone infusion
The increase in cortisol levels induced by hydrocortisone was positively correlated to the decrease in IL6 levels occuring during the first four hours of the
experimental period (Pearson’s correlation coefficient
r=0.670, P < 0.01). During the rest of the night this
correlation did not reach statistical significance (23:00–
03:00 h: r= 0.337, n.s.; 03:00–07:00 h: r= 0.272, n.s.).
In contrast to changes in IL-6 levels, the changes in
TNF-a levels induced by hydrocortisone were not significantly influenced by the amount of cortisol increase
(Pearson’s correlation coefficients 19:00–23:00 h:
r= 0.002, n.s.; 23:00–03:00 h: r= 0.384, n.s.; 03:00–
07:00 h: r= 0.240, n.s.). A comparison of the time
course of TNF-a and IL-6 in the subgroups of patients
with cortisol levels during hydrocortisone infusion
above or below the median again revealed that the suppressing effect of hydrocortisone on IL-6 levels was
more pronounced in those patients with a stronger cortisol increase. In contrast, the reduction in TNF-a levels
did not differ between these two subgroups (see Fig. 4).
Pearson’s or Spearman’s correlation coefficients did
not reveal any significant association of age, gender and
HAM-D score with changes in cytokine levels induced
by hydrocortisone. In addition, ACTH release following
CRH during the combined DEX/CRH test did not correlate with the changes in TNF-a and IL-6 levels during
hydrocortisone infusion (data not shown).
4. Discussion
The present study was performed to explore the
interactions between the HPA and immune systems in
patients with major depression. To this end, we measured baseline nocturnal IL-6 and TNF-a plasma levels
and the respective changes induced by the administration of hydrocortisone. We hypothesized that in vivo
cytokine production in depressive patients would corre-
467
late with their HPA system function as assessed by the
combined DEX/CRH test, and, moreover, that the sensitivity of cytokine levels to hydrocortisone infusion
might also depend on the activity of the HPA-system.
We found that cytokine levels were correlated with
HPA-function at baseline, but not following the
administration of hydrocortisone. In particular, baseline
TNF-a levels were negatively correlated to the amount
of ACTH secreted in response to CRH after DEX-pretreatment, indicating that HPA system overactivity goes
along with a reduced production of this inflammatory
cytokine. This is in line with in vivo and in vitro studies
in healthy subjects and depressive patients demonstrating that the administration of glucocorticoids suppresses TNF-a production (Bratts & Linden, 1996;
Schuld et al., 2001). In the present study, we showed
here for the first time that even under baseline conditions circulating cytokines in depressive patients co-vary
with the endogeneous activity of the HPA system.
Interestingly, this was the case for TNF-a, but not for
IL-6 levels. This discrepancy is somewhat surprising
because oral intake of low-dose dexamethasone suppresses the production of both of these cytokines in
parallel (Schuld et al., 2001).
This discrepancy might be explained by a difference in
sensitivity to the in vivo administration of glucocorticoids between IL-6 and TNF-a: We found that pulsatile
administration of hydrocortisone induced a clearcut and
stable suppression of TNF-a production throughout the
night. In contrast, IL-6 levels were only transiently suppressed during the second third of the night suggesting
that IL-6 production is less sensitive to hydrocortisone
administration than TNF-a. Moreover, baseline TNF-a
but not IL-6 levels were related to the HPA system
activity in the patients. Thus, TNF-a was also influenced by quite subtle, chronic changes in neuroendocrine secretion.
In the present study, the suppressive effect of acute
hydrocortisone administration on cytokine levels was
independent from HPA system function. This finding
suggests that the negative feedback of cortisol on cytokine release during infection and inflammation might
not be compromised by chronic HPA overactivity associated with depressive disorder. This conclusion is in
line with lacking evidence for an increased frequency or
more severe course of infectious diseases in depressive
patients.
IL-6 and TNF-a levels have been measured in
numerous studies comparing depressive patients with
healthy controls. However, the results are conflicting
and increased as well as decreased levels of inflammatory cytokines have been reported in depression (Haack
et al., 1999; Maes, 1999). In view of the suppressive
action of glucocorticoids on cytokine production and
the results of the present study which demonstrate a
negative correlation between TNF-a plasma levels and
468
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
Fig. 3. Changes in cortisol, TNF-a and IL-6 levels during pulsatile infusion of 1.0 g/kg body weight of hydrocortisone in patients with depression.
Left panel: Influence of hydrocortisone-infusion on the plasma levels of cortisol, TNF-a and IL-6 in 14 depressive patients administered between
19:00 and 07:00 h (open circles—placebo night; closed circles—verum night). Right panel: For statistical analysis the area under the curve for the
respective variables was computed between 19:00 and 23:00 h, 23:00 and 03:00 h and 03:00 and 07:00 h (white columns—placebo night; black columns—verum night). ANOVA for repeated measures revealed a significant time factor for cortisol (F[2;26]=11.799, P <0.01), but not for TNF-a
(F[2;26]=2.321) or IL-6 (F[2;26]=0.756). A significant condition effect was found with respect to all parameters (cortisol: F[2;13]=112.749,
P <0.001; TNF-a: F[2;13]=53.031, P<0.001; IL-6 F[2;13]=4.983, P <0.05). Time-by-condition interaction effects were significant for cortisol
(F[2;26]=16.403, P<0.001) and TNF-a (F[2;26]=8.466, P <0.01), but not for IL-6 (F[2;26]=0.756). Asterisks indicate significant differences in
post-hoc t-test comparison.
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
469
Fig. 4. Changes in the plasma levels of cortisol, TNF-a and IL-6 in response to pulsatile hydrocortisone injection in patients with high versus low
hydrocortisone-induced increases in cortisol plasma levels. Left panel: Hydrocortisone-induced changes in the plasma levels of TNF-a and IL-6 in 14
depressive patients measured between 19:00 and 07:00 h. The patient sample is divided in two subgroups of seven subjects each, separated according
to a median-split of the hydrocortisone-induced increase in cortisol levels (open squares—cortisol increase below median; closed squares—cortisol
increase above median). Right panel: For statistical analysis the area under the curve for the respective variables was computed between 19:00 and
23:00 h, 23:00 and 03:00 h and 03:00 and 07:00 h (white columns—cortisol increase below median; black columns—cortisol increase above median).
As expected, ANOVA for repeated measures revealed significant time (F[2;24]=16.280, P<0.001) and time-by-group interaction effects
(F[1;12]=17.259, P=0.001) for cortisol. Moreover, a significant time effect for TNF-a (F[2;24]=10.276, P <0.01), but not for IL-6 (F[2;24]=2.152)
was found. In contrast, a significant difference between groups was found with respect to IL-6 levels (F[1;12]=6.254, P<0.05), but not for TNF-a
(F[1;12]=1.583). Asterisks indicate significant differences in post-hoc t-test comparison.
470
A. Schuld et al. / Journal of Psychiatric Research 37 (2003) 463–470
HPA system activity one would expect lower levels of
inflamatory cytokines in patients compared to controls.
However, the group of Maes and coworkers reported
the opposite, i.e. increased levels of IL-6 which were
even positively correlated to HPA system activity (Maes
et al., 1993 and 1995). IL-6 indeed plays an important
role in neuroendocrine activation, but much higher IL-6
plasma levels are needed to significantly stimulate the
HPA system during experimental immune stimulation
in vivo (Späth-Schwalbe et al., 1998; Schuld et al.,
2000). The reason for such conflicting findings remains
obscure and might only be discovered by future studies
in this field.
In summary, in the present study circulating TNF-a,
but not IL-6 levels in a small sample of patients suffering from depressive disorders were negatively correlated
with the activity of the HPA system. Moreover, IL-6
and TNF-a levels were differentially modulated by exogeneous glucocorticoids. Because both neuroendocrine
and immune systems have been suggested to play a
crucial role in the pathophysiology of depression, the
interaction of these systems is of particular interest for
further research in the field and for the development of
future treatment strategies. Thus, further studies should
be undertaken to replicate these findings in independent
samples of depressive patients, but also in healthy people and in patients whose HPA system is disturbed in a
much more pronounced manner as for example, in
patients suffering from Cushing’s disease.
Acknowledgements
We acknowledge the help and skillful technical assistance of Gabi Kohl and Johannes Huber.
References
American Psychiatric Association. Diagnostic and statistical manual
of mental disorders. IVth ed. Washington DC: American Psychiatric Press; 1994.
Baram TZ, Mitchell WG, Halden E. Inhibition of pituitary-adrenal
secretion by a corticotropin releasing hormone antagonist in
humans. Molecular Psychiatry 1996;1:320–4.
Barber AE, Coyle SM, Marano MA, Fischer E, Calvano SE, Fong Y,
Moldawer LL, Lowry SF. Glucocorticoid therapy alters hormonal
and cytokine responses to endotoxin in man. Journal of Immunology 1993;150:1999–2006.
Bornstein SR, Chrousos GP. Adrenocorticotropin (ACTH)- and nonACTH-mediated regulation of the adrenal cortex: neural and
immune inputs. Journal of Clinical Endocrinology and Metabolism
1999;84:1729–36.
Bratts R, Linden M. Cytokine modulation by glucocorticoids:
mechanisms and actions in cellular studies. Alimentation, Pharmacology and Therapeutics 1996;10(Suppl 2):81–90.
Gold PW, Chrousos GP. The endocrinology of melancholic and atypical depression: relation to neurocircuitry and somatic consequences. Proceedings of the Association of American Physicians
1999;111:22–34.
Haack M, Hinze-Selch D, Fenzel T, Kraus T, Kühn M, Schuld A,
Pollmächer T. Plasma levels of cytokines and soluble cytokine
receptors in psychiatric patients upon hospital admission: effects of
confounding factors and diagnosis. Journal of Psychiatric Research
1999;33:407–18.
Hamilton M. A rating scale for depression. Journal of Neurology
Neurosurgery Psychiatry 1960;23:56–62.
Heuser I, Yassouridis A, Holsboer F. The combined dexamethasone/
CRH test: a refined laboratory test for psychiatric disorders. Journal of Psychiatric Research 1994;28:341–56.
Holsboer F. The corticosteroid receptor hypothesis of depression.
Neuropsychopharmacology 2000;23:477–501.
Keck ME, Welt T, Wigger A, Renner U, Engelmann M, Holsboer F,
Landgraf R. The anxiolytic effect of the CRH1 receptor antagonist
R121919 depends on innate emotionality in rats. European Journal
of Neuroscience 2001;13:373–80.
Leonard B. Stress, depression and the activation of the immune system. World Journal of Biological Psychiatry 2000;1:17–25.
Maes M. Major depression and activation of the inflammatory
response system. Advances in Experimental Medicine and Biology
1999;461:25–46.
Maes M, Bosmans E, Meltzer HY. Immunoendocrine aspects of
major depression. Relationships between plasma interleukin-6 and
soluble interleukin-2 receptor, prolactin and cortisol. European
Archives of Psychiatry and Clinical Neuroscience 1995;245:172–8.
Maes M, Scharpe S, Meltzer HY, Bosmans E, Suy E, Calabrese J,
Cosyns P. Relationships between interleukin-6 activity, acute phase
proteins, and function of the hypothalamic-pituitary-adrenal axis in
severe depression. Psychiatry Research 1993;49:11–27.
Nemeroff CB, Widerlov E, Bissette G, Walleus H, Karlsson I, Eklund
K, Kilts CD, Loosen PT, Vale W. Elevated concentrations of CSF
corticotropin-releasing factor-like immunoreactivity in depressed
patients. Science 1984;226:1342–4.
Rubinow DR, Post RM, Savard R, Gold PW. Cortisol hypersecretion
and cognitive impairment in depression. Archives of General Psychiatry 1984;41:279–83.
Schuld A, Kraus T, Haack M, Hinze-Selch D, Zobel AW, Holsboer F,
Pollmächer T. Effects of dexamethasone on cytokine plasma levels
and white blood cell counts in depressed patients. Psychoneuroendocrinology 2001;26:65–76.
Schuld A, Mullington J, Friess E, Hermann DM, Galanos C, Holsboer F, Pollmächer T. Changes in dehydroepiandrosterone (DHEA)
and DHEA-sulfate plasma levels during experimental endotoxinemia in healthy volunteers. Journal of Clinical Endocrinology and
Metabolism 2000;85:4624–9.
Späth-Schwalbe E, Hansen K, Schmidt F, Schrezenmeier H, Marshall
L, Burger K, Fehm HL, Born J. Acute effects of recombinant
human interleukin-6 on endocrine and central nervous sleep functions in healthy men. Journal of Clinical Endocrinology and Metabolism 1998;83:1573–9.
Tilders FJ, DeRijk RH, Van Dam AM, Vincent VA, Schotanus K,
Persoons JH. Activation of the hypothalamus-pituitary-adrenal axis
by bacterial endotoxins: routes and intermediate signals. Psychoneuroendocrinology 1994;19:209–32.
Zobel AW, Nickel T, Künzel HE, Ackl N, Sonntag A, Ising M,
Holsboer F. Effects of the high-affinity corticotropin-releasing
hormone receptor 1 antagonist R121919 in major depression: the
first 20 patients treated. Journal of Psychiatric Research 2000;34:
171–81.