- Email: [email protected]
Pharmacological modulation of cortisol secretion and dexamethasone suppression in alzheimer's disease
BIOL PSYCHIATRY 1988:23: 13-24
13
Pharmacological Modulation of Cortisol Secretion and Dexamethasone Suppression in Alzheimer’s Disease P. Davous, M. Roudier, M. L. Piketty, C. Abramowitz, and Y. Lamour
We have investigated the dexamethasone suppression of cortisol release in a group of 28 patients with senile dementia of the Alzheimer type (SDAT) after stimulation by physostigmine and clonidine, as compared with basal conditions. All patients but one had previously been evaluated with a depression symptom checklist and had submitted to a standard Dexamethasone Suppression Test (DST). SDAT patients showed normal baseline cortisol values measured at 4:00 PM. DST was reproducible, but nonsuppression did not appear to be a feature of the disease, nor of the dementia syndrome, although a majority of the most demented patients were found to be nonsuppressors. Physostigmine stimulated cortisol secretion in 20 of 24 cases, irrespective of the severity of dementia. Clonidine induced a secretion in 12 of 15 cases, but this was less than that observed after cholinergic stimulation, Physostigmine made cortisol release significantly less sensitive to the suppressive effect of dexamethasone than clonidine in SDAT. This double response should be tested as a possible predictor of a cholinergic therapeutic effect.
Introduction The Dexamethasone Suppression Test (DST) has now been widely studied in dementia, and conflicting results have been reported (Raskind et al. 1982; Spar and Gemer 1982; Balldin et al. 1983; Cames et al. 1983; Coppen et al. 1983; Jenike and Albert 1984; MacKeith 1984; Pomara et al. 1984; Charles et al. 1986; Davis et al. 1986; Greenwald et al. 1986). In a previous study (Davous et al. 1986), we found that a majority of elderly patients with senile dementia of the Alzheimer type (SDAT) were suppressors after dexamethasone administration. Nonsuppression was found in 50% of the most impaired patients, irrespective of their depression score. This result could suggest a cortisol hy-
From the Departments of Neurology (P.D.) and Nuclear Medicine (M.L.P.), Sainte Anne Hospital, Paris; the Departments of Chiabics (M.R.) and Pharmacy (C.A.), Charles Richet Hospital, Villier le Bel, and INSERM U 161 (Y.L.). Paris, France. Supported by Grant 83 C. 1026 from the Ministk de I’lndusttie et de la Recherche. Address reprint requests to Dr. P. Davous, Service de Neurologie, H6pital Sainte Anne, 1 Rue Cabanis, 75674 Paris Cedex 14, France. Received October 27, 1986; revised March 21, 1987.
0 1988 Society of Biological
Psychiatry
OO@-3223/88/$03.50
14
MOL. PSYCHIATRY l’x3x:?1.13 ?‘l
persecretion in dementia, but no signi~cant correlation could be established between plasma cortisol values and the severity of dementia as evaluated by the Mini Mental State Examination (Folstein et al. 197.5). As corticotropin-releasing factor (CRF) and adrenocorticotrophic hormone (ACTH) release are controlled by the hypothalamus and limbic areas (Carroll et al. 1976), in accordance with Janowsky’s theory (1972), it has been suggested that abnormal DST in SDAT might be related to an ~mpai~ent in the hypothalamic-pituita~-adrenal (HPA) axis rather than to cortical lesions (Raskind et al. 1982). It is also well known that cholinergic and noradrenergic systems, which are both impaired in SDAT, are involved in the regulation of ACTH secretion, the former being excitatory (Davis and Davis 1980: Risch et al. 1980), the latter inhibitory (Hillhouse et al. 1975, Sachar et at. 1980). Fu~he~orc, CRF-Iike immunoreactivity is normal in the hy~thalamus of SDAT patients (Bissette et al. 1985), although there is a reduction of choline acetyltransferase (Davies 1979) and noradrenaline (Yates et al. 1981) activities. If ACTH secretion is under cholinergic and noradrenergic control at the hypothalamic level, one should observe a modification of ACTH and cortisol secretion as a result of the imbalance between the two systems in Alzheimer’s disease. The purpose of our investigation was to determine whether or not cholinergic and adrenergic agonists can influence plasma cortisol concentrations and dexamethasone response in SDAT patients.
Methods The study group consisted of 28 SDAT patients (3 men, 25 women), 71-95 years old (mean age 84), who had been free of neuroleptic or antidepressant drugs for at least 3 months. All patients were hospitalized in the same geriatric department for at least 3 months. They were all able to walk and eat without assistance. The diagnosis of probable SDAT was established according to DSM-III (APA 1980) and NINCDS (McKhann et al. 1984) criteria. Patients with neurological signs or symptoms, previous stroke, or any history or evidence of arteriosclerosis were all excluded to rule out multiinfarct dementia. Dementia severity was evaluated by the Mini Mental State Examination (MMSE). The patients had scores from 0 to 19 (mean rt SD, 7.3 t. 5.3). On the basis of their dementia rating, they were subdivided in two subgroups with similar mean age and sex ratio. Seventeen patients had an MMSE score of O-9, and 11 patients had an MMSE score of 10-19. They were assigned, respectively, to group I (severe dementia) and group I1 (mild to moderate dementia). All but one of the patients included in this study had already been rated on a depressive symptom checklist and had been given a standard DST (Davous et al. 1986). They were free of other medical illnesses at the time of pharmacological investigation. As a control group, 14 elderly inpatients living under the same conditions (13 women, I man; mean age 85, range 75-93) were submitted to a standard DST. All of these patients had no previous history of depression, were normal on detailed neurological examination, and had not been exposed to any psychotropic drugs. They were free of cognitive impairment, as assessed by the MMSE: scores ranged from 22 to 27 (mean 25). Two patients had congestive heart failure, two had peripheral arteriopathy, and one had asthma. The others were free of medical illness and had been admitted for sociai reasons.
BIOL PSYCHIATRY 1988~23: 13-24
DST in Alzheimer’s Disease
15
Cholinergic Stimulation and DST was used as a cholinomimetic. Twenty-eight SDAT patients participated in the study 3 months after undergoing a standard DST. They acted as their own controls. The test was carried out over 4 days. Two milligrams of physostigmine was given orally 5 times a day from the second day at 7:00 AM to the fourth day at 2530 PM. One milligram dexamethasone was given on the third day at 11:OOPM. Blood samples for plasma cortisol determination were obtained at 4%) PM on the first day (baseline), third day (stimulation), and fourth day (suppression). Side effects (vomiting) occurred in 4 patients, who were then excluded from the study. Twenty-four patients completed the study without significant side effects. There were 14 patients in group I and 10 patients in group II. Physostigmine
Adrenergic Stimulation and DST Clonidine was used as an adrenomimetic. The study was carried out 4 months after the cholinergic stimulation. During that period, 4 patients had died of cardiac or pulmonary disease. Nineteen patients previously studied under cholinergic stimulation entered the study. They had been clinically stable, with no change of therapy in the intervening period. The test was carried out over 6 days. Clonidine was given orally, 0.15 mg twice a day at 8:00 AM and 8:00 PM, from the second day at 8:00 AM to the sixth day at 8:00 AM. Dexamethasone was given on the fifth day at 11:OOPM. Blood samples for plasma cortisol were obtained at 4:00 PM on the first day (baseline), fifth day (stimulation), and sixth day (suppression). Side effects (sedation and/or hypotension) were observed in 3 patients, who were then excluded from the study. Sixteen patients completed the study. There were 10 patients in group I and 6 patients in group II. In sum, 15 patients previously tested by a standard DST were studied consecutively by cholinergic and adrenergic stimulations and dexamethasone suppression. Cortisol was measured by radioimmunoassay (Davous et al. 1986). Dexamethasone nonsuppression was defined as any cortisol value greater than 5 p,g/dl.
Results Standard DST in SDAT Patients Compared with Controls of Similar Age Group (Table I) Seven of 27 SDAT and 5 of 14 elderly control patients were found to be nonsuppressors after dexamethasone administration. The mean postdexamethasone cortisol values were Table 1. Postdexamethasone Cortisol Values in Controls and SDAT Patients Divided into
SubgroupsAccording to Dementia Severity cortisol MMSE score
Controls
SDAT SDAT
groupI groupII
Postdexamethasone (mean f SEM)
@ange)
n
Age (mean 2 SD)
Nonsuppressors (n)
22-30 o-9 lo-19
14 16 11
85 -c 5.4 82.6 -c 3.1 86 k I
5 I 0
“p < 0.05 comparedwith group I.
(wW
5.2 2 1.1 6.5 2 1.4 2.3 + 0.4”
16
BIOL PSYCHIATRY 1988;23: 13-24
P. Davous et al.
Table 2. Baseline Cortisol Values in SDAT Patients According to Dementia Severity
MMSE
n SDAT group I SDAT group II
(mean +
1s
Cortisol level 4:oO (mean t REM)
score SEM)
(fig/W
4 t 0.9 12.4 ? I.1
10
PM
10.2
lr 0.8
13.1
t
1.7
not significantly different in SDAT and control patients. There was no correlation between co&o1 values and age or MMSE scores. Seven of 16 (43%) SDAT patients in group I were nonsuppressors. All of the patients in group I1 were DST suppressors. The most severly impaired patients (group I) had significantly higher postdexamethasone cortisol values than less affected patients (group II) (Mann-Whitney U-test, U = 45, p < 0.05).
Baseline Cortisol Levels in SDAT Patients Cortisol levels at 4:00 PM were determined twice before pharmacological stimulation in 18 demented patients. All of these subjects had plasma cortisol levels > 5 pg/dl in the two different samples. The mean cortisol values were, respectively, 11.2 k 0.9 (SEM) before administration of physostigmine and 9.5 2 0.8 (SEM) before administration of clonidine. These mean values are not significantly different (I = 1.40, df = 17). Nevertheless, the mean individual difference was 14% (range 0.7-66), and the individual values were not correlated. The cortisol values were not correlated with the MMSE scores. The most impaired patients tended to have lower values than patients less affected, and the difference approached significance (t = 1.66, df = 23, 0.1 < p < 0.2) (Table 2).
Cholinergic
Stimulation
and DST in SDAT (Table 3)
After the oral administration of physostigmine, most of the patients (20124) had higher plasma cortisol values (Figure I ). The mean f SEM value of plasmacortisol was 17.7 2 2.2. significantly different from the predrug values (paired t = 2.82, df = 23, p < 0.01). The range of relative cortisol variations was from - 72% to + 25 1%. Sixteen of 24 patients had levels 20% above their own baseline value, and 7 of them had levels 2
Table 3. Mean 2 Dexamethasone
Nonsuppressors Suppressors Total
SEM
Plasma Cortisol (p,g/dl) in SDAT Patients following Physostigmine n
Baseline
Postphysostigmine
DST
I1
13.07 + I 10.2 rt 1.3 I I 2 k 0.9 (A)
23.06 t 4” 13.3 + 2.2 (B) 17.7 5 2.2 (C)
17.9 k 4.2 2.5 k 0.3 9.5 * 2.5’
13 24
Patients have been separated according to DST results following phyrostigmine compared with A (r = 2.82) “p < 0.001 compared with B (I = 5 2 I) 3 Y4l. ‘p < 0.001 compared wth C (I y
"p < 0.01
and
administration.
BIOL PSYCHIATRY 1988;23:13-24
DST in Alzheimer’s Disease
17
. *
I
c Figure 1. individual plasma cot&of in SDAT patients fn = 24) following physostigmine stimulation (B) and dexamethasone suppression (C) compared with baseline values (A). Horizontal bar indicates the mean in each group.
standard deviations ahove the mean elevation of the group. Dexamethasone suppression was effective in 13 of the 24 patients. The suppression was highly significant in the whole group (paired t = -3.94, df = 23, p < 0.001). After physostigmine administration, the nonsuppressor patients had cortisol values significantly higher than the suppressors (Table 3) (t = 5.21, df = 22, p < 0.001). In fact, the cortisol level after DST was significantly correlated to the cortisol level following physostigmine adminis~tion (r = 0.626, p < 0.01). Baseline cortisol values were not significantly different between the two groups (t = 1.67, df = 22). There was no significant difference between the plasma cortisol values before the cholinergic stimulation and after the DST (t = 0.82, df = 23) in the population of SDAT patients taken as a whole. Studying the effect of cholinergic stimulation as a function of dementia severity, we did not find any significant correlation between MMSE scores and cortisol values measured after oral physostig~ne administration (r = -0.05). Comparison of the patients in groups I and II did not reveal any significant difference in absolute or relative values (Mann-Whitney U-test, U = 56, II = 43, respectively).
IX
BIOLPSYCHIATRk 1988:?3:13-24
Adrenergic
ct al
P. Ihous
Stimulation
and DST in SDAT (Table 4)
After the oral administration of clonidine, most of the patients (12/15) exhibited an elevation of their cortisol value. although it was smaller in magnitude than that observed after cholinergic stimulation. However, as a group. they showed a significant increase, with a mean value of 13.1 2 1.7 (paired t = 2.72, df = 14, p < 0.02) (Figure 2). Six patients showed marked stimulation, from 575%to 195%; 6 patients had a slight stimulation from 2.1% to 24%; and 3 patients had a suppression that ranged from - 4.8% to - 17.6%~ of baseline values. Dexamethasone suppression was effective in 11 of 16 patients. The suppression was highly significant in the whole group (paired t = -~6.75. df = 14. ~7 < 0.00 1). After clonidine administration, the nonsuppressor patients had plasma cortisoi values significantly higher than those of the suppressor patients (t = 4.45. df = 14. p < 0.001). The cortisol level after DST was significantly correlated with the plasma cortisol level after clonidine administration (r = 0.5 17, p < 0.05). Baseline cortisol values were not significantly different between the two groups (t = 1.20. df = 13). In contrast with the results obtained after cholinergic stimulation. there was a significant decrease in the cortisol concentrations of the whole group of patients after clonidine stimulation and dexamethasone suppression when compared with baseline values (paired I = -6.14,df = 14.~ < 0.001). There was no signiticant correlation between plasma cortisol values after clonidine administration and the MMSE scores (r = ~~0.12), nor any significant differences between the two groups according to the severity of dementia. whether considering absolute or relative variations (Mann-Whitney U-tests, U = 22. U = 23. respectively).
Comparative
Study
In the group of 15 demented patients who were consecutively studied with the standard DST and the DST under physostigmine and clonidine, 6 patients were suppressors on the 3 tests, irrespective of the pharmacological effect of drugs on cortisol secretion. In the whole group, the postdexamethasone cortisol values were not significantly different in basal conditions and after cholinergic or adrenergic stimulations (respectively, t = 1.94, t = 0.37, t = 1.28). Demented nonsuppressor patients identified in standard conditions (n = 4) maintained dexamethasone escape after pharmacological stimulation by physostigmine (3 cases), clonidine (3 cases), or both (2 cases). Comparing the individual responses to the cholinergic and adrenergic agonists, it appeared that the 7 patients who had the highest cortisol response to physostigmine were relatively insensitive to clonidine
Table 4. Mean + SEM Plasma Cortisol (kg/dl) in SDAT Patients following Clonidine and Dexamethasone Baseline
Postclonidine
10.4 + 2 8.4 k 0.7 9.2t-0.7(A)
20.4 2 3.4” 9.7 2 1 (B) 13.1k 1.7(C)
”
Nonsuppressors
5
Suppressors
I1
Total
16
Patients have been separated according to DST
“p <
0.02 compared with A (I =
“p< 0.001
compared with B (r = 4.45)
‘p < 0.001
compared wth
C (I =
results
following
clonidine
2.72). 6.75)
and with
A tr :
-6
141.
administration.
DST 7.7 -t I.1 1.6f 0.3 3.52 0.8'
BIOL PSYCHIATRY 1988:23:13-24
DST in Alzheimer’s Disease
CORTI
SOL
PG’DL
25
19
200
n
15 I
0
10.
A
T + n
5-
0 0
f
f __ __ __ __
mm
__
__
_-
-
I
A
R
C
Figure 2. Individual plasma cortisol in SDAT patients (n = 16) following clonidine stimulation (B) and dexamethasone suppression (C) compared with baseline values (A). Horizontal bar indicates the mean in each group.
and vice versa (Figure 3). Nevertheless, there was no significant inverse correlation between the cortisol variations induced by physostigmine and clonidine (r = - 0.293).
Discussion Basal Cortisol in SDAT The concentrations of plasma cortisol measured in baseline conditions at 4:00 PM were within the normal range, as defined in a similar age group with our method (range for controls 60-90 years old: 4-l 1.8). In our study, baseline plasma cortisol values were not correlated with the severity of dementia; the most impaired patients tended to have lower values than the mildly affected patients. Similar findings were reported by Pomara et al. (1984). Conversely, the early morning and midday levels tended to be higher in the most severe forms of dementia. Similar results were found by Davis et al. (1986). Thus, it can be hypothesized that SDAT patients have an impaired circadian rhythm of cortisol, which could be related to several parameters, such as the severity of dementia, locomotor activity, sleep disturbances etc. There is no evidence however that cortisol hypersecretion is a feature of SDAT and is related to the cholinergic deficit.
20
BIOL PSYCHIATR\r 198823:13-24
P. Lkivous et al.
2
1 Figure 3. Comparison of the percentage of cortisof variation from baseline (log-transfarmed) after physostigmine (A) and clonidine (B) in a group of 14 SDAT patients.
-1
R
LIST and SDAT In our sample, nonsuppression was found only in the most demented patients: all patients scoring 10 or more on the MMSE were suppressors. In the group of suppressor patients fn = 20), the plasma cortisol concen~tion after DST was lower than in the age-matched control group. This finding suggests either lower basal values in SDAT or a stronger suppression by dexamethasone in the group of less demented patients. Similar results have been reported by Pomara et al. (1984), who found that the percentage change in cortisol levels from the 8:OOAM baseline to the 8:00 AM postdexamethasone level correlated with dementia severity. From the literature, it appears that the demented patients often show dex~ethasone nonsupp~ssion (Raskind et al. 1982; Spar and Gemer 1982; Balldin et af. 1983; MacKeith 1984; Greenwald et al. 1986), but opposite results have been described (Cames et al. 1983; Castro et al. 1983; Charles et al. 1986) These discrepancies suggest methodological differences among the studies. The percentage of nonsuppressor patients observed in our study (Davous et al. 1986) is similar to the results obtained by Jenike and Albert (1984) and by Pomara et al. ( 1984). From these three studies, it appears that dexamethasone nonsuppression is correlated with dementia severity. However, the relationship between
DST in Alzheimer’s Disease
BlOL PSYCHIATRY 1988;23:13-24
21
dexamethasone suppression and mental impairment is questionable, as we did not find any significant correlation between the plasma cortisol levels after DST and the MMSE scores. In some studies, nonsuppression in SDAT patients has been related to age (Greenwald et al. 1986) and to depression (Cames et al. 1983), but none of these parameters were found to be significantly correlated with DST results by other investigators (Balldin et al. 1983; Jenike and Albert 1984; MacKeith 1984; Davous et al. 1986). Role of Physostigmine Physostigmine, a centrally active cholinesterase inhibitor, has been shown to increase cortisol secretion in normal subjects (Carroll et al. 1978; Risch et al. 1980, Doerr and Berger 1983), and experimental data in animals and humans indicate an action of the drug on the HPA axis (Krieger and Krieger 1967; Hillhouse et al. 1975; Risch et al. 1981). According to Davis and Davis (1980) and Risch et al. (1980), the physostigmineinduced cortisol secretion might be secondary to a stress reaction when the iv route is employed and side effects occur. Such an explanation seems unlikely when physostigmine is given orally, and if peripheral side effects are prevented or excluded by individual dosages. In fact, a recent paper by Risch et al. (1986) provided further evidence that physostigmine’s effects on plasma cortisol concentrations occur via central mechanisms. In the present study, despite probable interindividual differences in drug absorption, most of the patients exhibited an increase in cortisol secretion when given physostigmine orally. A central stimulation is likely, as patients with side effects were excluded from the study. Following stimulation by physostigmine, the DST showed that 55% of SDAT patients are suppressors and 45% avoid suppression. Nonsuppression was directly correlated with the cortisol secretion induced by physostigmine. If there is a relationship between DST nonsuppression and cholinergic activity, the lack of suppression by dexamethasone after physostigmine stimulation could provide another predictive index of therapeutic benefit from cholinomimetics. The absence of any significant correlation between the cortisol secretion changes and the severity of dementia is indicative of the absence of a direct relationship between the cholinergic deficit in SDAT and the HPA axis response to physostigmine. Mohs et al. (1985) found a linear relationship between percent increase in cortisol and mean percent change in symptoms induced by physostigmine. In a similar study, including 10 patients treated by 2 mg oral p~ysostigmine 5 times a day for 1 month, no significant correlation was observed between the relative changes in ADAS subscores (Rosen et al. 1984) and cortisol levels (r = - 0.140) (Davous et al. unpublished observations). In fact, if cortisol secretion were related to cholinergic activity, one should observe in SDAT a hypoactivity of the HPA axis when stimulated by physostigmine, with less response found in patients with more ace~lcholine depletion. Some alternative explanations can be put forth: (1) as suggested by Mountjoy et al. (1984), choline acetyltransferase (CAT) activity is not significantly reduced in SDAT patients compared with elderly controls; (2) there is an overactivity of the postsynaptic muscarinic receptors (Nordberg et al. 1983; Pomara and Stanley 1986); and (3) cortisol secretion is not significantly linked to the cholinergic system in SDAT. In order to understand the relationships between cortisol ~gulation and cholinergic mech~isms in SDAT, further studies are needed, addressing the following matters: intestinal absorption and central bioavailability of the cholinergic drugs, cholinergic status of the patient, basal function of the HPA axis and its reactivity to physiological and pharmacological parameters .
22
BIOL PSYCHIATRl
I’. I)avvus
et al.
IOXX:?3:13-24
Role of Clonidine Clonidine, a central alpha-adrenergic agonist, has been shown to bind to presynaptic as well as postsynaptic receptors (Kobinger 1978). Pharmacological studies suggest a rather complex mechanism of action related to other neurotransmitter systems, such as serotonin and acetylcholine (Komer and Head 1983). In fact, there is still much controversy on the site of action of clonidine at the synaptic level and throughout the different brain nuclei (Jhanwar-Uniyal et al. 1985). In normal subjects, the action of clonidine on cortisol secretion is not clearly demonstrated. Lanes et al. (1983), Siever et al. (1984), and Lechin et al. ( 1985) have reported a suppressive effect. Nakai et al. (1973) have found a stimulation of the HPA axis by alpha-adrenomimetics. La1 et al. (1975) and Doerr and Berger (1983) did not observe significant variations. Conversely, Siever et al. (1984) reported a significant reduction of cortisol secretion in depressed patients compared with controls. To our knowledge, there is no study in the literature on the effects of clonidine on cortisol secretion in SDAT. In the present study, clonidine induced a slight cottisol suppression in 3 cases and stimulation in 12 cases, 6 of them being highly stimulated. Some of the patients who were stimulated by clonidine escaped suppression by dexamethasone. DST nonsuppression could be predicted by the level of cortisol elevation induced by clonidine. The absence of any reliable index of noradrenergic activity in the patients studied, and our uncertainties about the effects of clonidine at the central level. make these results difficult to interpret. Furthermore, in contrast with CAT activity, the deficit in norepinephrine in SDAT has not been found to correlate with symptom severity (Mann et al. 1981; Perry et al. 1981; Francis et al. 1985). It is possible that the effects of clonidine on cortisol secretion, if they are directly mediated by the noradrenergic system, depend on individual norepinephrine release and receptor sensitivity. Another possible explanation is that the clonidine effect is mainly controlled by cholinergic mechanisms, as suggested by Delitala et al. (1983) and Zometzer (1985). According to this hypothesis. loss of normal locus coeruleus function might result in diminished cholinergic efficacy. Such an imbalance is compatible with the results observed in our study. In conclusion, there is no evidence from the present study that cortisol variations reflect abnormalities specific for central cholinergic or monoaminergic neurotransmission. but some results support the hypothesis of individual cholinergic-adrenergic imbalance in the HPA axis functional system in SDAT.
The authors thank A. Dickerson
for his help
in revision of the English text.
References American Psychiatric Association (1980): Diagnostk (ed 3). Washington, DC: APA.
and Statistical Manual of Mental Disorders
Balldin J, Gottfries CG, Karlsson 1, Lindstedt G, Langstrom G, Walinder J (1983): Dexamethasone Suppression Test and serum prolactin in dementia disorders. Br J Psychiatry 143:277-281. Bissette G, Reynolds GP, Kilts CD, Widerlov E, Nemeroff CB (1985): Corticotropin releasing factor-like immunoreactivity in senile dementia of the Alzheimer type. JAMA 254:3067-3069. Cames M, Smith JC, Kahn NH, Bauwens SF (1983): The Dexamethasone Suppression demented outpatients with and without depression. Psychiatp Res 9:337-344. Carroll BJ, Curtis GC. Mendel J (1976): Neuroendocrine Psychiatry 33: 1039-1044.
regulation
in depression.
Test in
Arch Grn
DST in Alzheimer’s Disease
BIOL PSYCHIATRY 1988:23:13-24
23
Carroll BJ, Greden JF, Rubin RT, Haskett R, Feinberg M, Schteingart D (1978): Nemotransmitter mechanism of neuroendocrine disturbance in depression. Acta Endocrinol (SUppl) 22@14. Castro P, Lemaire M, Toscano-Aquilar M, Herchuelz A (1983): Depression, dementia and the
Dexamethasone Suppression Test. Am J Psychiatry 140:1386. Charles GA, Lefevre A, Mire1 J, Rush AJ (1986): Suppression of cortisol following dexamethasone in demented patients. Psychiatry Res 17:173-182. Coppen A, Abou-Saleh M, Milln P, Metcalfe M, Harwood J, Bailey J (1983): Dexamethasone Suppression Test in depression and other psychiatric illness. Br J Psychiatry 142:498-X% Davies P (1979): Neurotransmitter-related enzymes in senile dementia of the Alzheimer type. Brain Res 171:319-327. Davis BM, Davis KL (1980): Cholinergic mechanisms and anterior pituitary hormone secretion. Biol Psychiatry 15303-310. Davis KL, Davis BM, Greenwald BS, Mohs RC, Mathe AA, Johns CA, Horvath TB (1986): Cortisol and Alzheimer’s disease. I. Basal studies. Am J Psychiatry 143:300-305. Davous P, Roudier M, Lamour Y, Susini de Luca H, Abramowitz C (1986): Dexamethasone Suppression Test in senile dementia of the Alzheimer type. Adv Behav Biol 29:355-359. Delitala G, Maioli M, Pacific0 A, Brianda S, Palermo M, Mannelli M (1983): Cholinergic receptor control mechanisms for L-dopa, apomotphine, and clonidine-induced growth hormone secretion in man. J Clin Endocrinol Metab 57: 1145-I 148. Doerr P, Berger M (1983): Physostigmine-induced escape from dexamethasone suppression in normal adults. Biol Psychiatry 18:261-268. Folstein MF, Folstein SE, McHugh PR (1975): Mini Mental State. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189-198. Francis PT, Palmer AM, Sims NR, Bowen DM, Davison AN, Esiri MM, Neary D, Snowden JS, Wilcock GK (1985): Neurochemical studies of early-onset Alzheimer’s disease. N Engl J Med 313:7-11. Greenwald BS, Mathe AA, Mohs RC, Levy MI, Johns CA, Davis KL (1986): Cortisol and Alzheimer’s disease, II: Dexamethasone suppression, dementia severity and affective symptoms. Am J Psychiatry 143sl42-446.
Hillhouse EW, Burden J, Jones MT (1975): The effect of various neurotransmitters on the release of corticotropin releasing hormone from the hypothalamus of the rat. Neuroendocrinology 17:1-11. Janowsky DS, El-Yousef MK, Davis JM, Sekerke HJ (1972): A cholinergic-adrenergic hypothesis of mania and depression. Luncet ii:632-635. Jenike MA, Albert MS (1984): The Dexamethasone Suppression Test in patients with presenile and senile dementia of the Alzheimer’s type. J Am Geriat Sot 32441444. Jhanwar-Uniyal M, Levin BE, Leibowitz SF (1985): Clonidine effects on catecholamine levels and turnover in discrete hypothalamic and extrahypothalamic areas. Brain Res 337:109116. Kobinger W (1978): Central alpha adrenergic systems as targets for hypotensive drugs. Rev Physiol Biochem Pharmacof 81:39-100.
Komer PI, Head GA (1983): Cardiovascular functions of central noradrenergic and serotoninergic neurons in conscious rabbits. Their contributions to the central action of clonidine: Chest (Suppl)2:335-338.
Krieger DT, Krieger HP (1967): Circadian pattern of plasma 17 hydroxycorticosteroid. Alteration by anticholinergic agents. Science 155:1421-1422. La1 S, Tolis G, Martin JB, Brown GM, Guyda H (1975): Effect of clonidine on growth hormone, prolactin, luteinizing hormone, follicle stimulating hormone and thyroid stimulating hormone in the serum of normal man. J Chin Endocrinol Metab 41:827-832. Lanes R, Herrera H, Palacios A, Moncada G (1983): Decreased secretion of cortisol and ACTH after oral clonidine administration in normal adults. Metabolism 32:568-570.
24
BIOL PSYCHIATRY 19X8:23: 13-24
P. Davous et al.
Lechin F, Van der Dijs B, Jakubowicz D, Camero R, Villa S, Lechin E, Gomez F (1985): Effects of clonidine on blood pressure, noradrenaline, cortisol, growth hormone and prolactin plasma levels in high and low intestinal tone subjects. Neuroendocrinology 40:253-26 1, MacKeith IG (1984): Clinical use of the DST in a psychogeriatric 145:389-393.
population.
Br ./ Ps~chiat~
McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984): Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of department of health and human services task force on Alzheimer’s disease. Neurology 34:939-944. Mann JJ, Stanley M, Neophytides A (1981): Central amine metabolism vivo relationship to cognitive deficit. Neurobiol Aging 2:57-60.
in Alzheimer’s
disease: In
Mohs RC, Davis BM, Johns CA, Mathe AA, Greenwald BS, Horvath TB, Davis KL (1985): Oral physostigmine treatment of patients with Alzheimer’s disease. Am J Psychiatry 142:28-33. Mountjoy CQ, Rossor MN, Iversen LL, Roth M (1984): Correlation of cortical cholinergic and GABA deficits with quantitative nemopathological findings in senile dementia. Brain 107:507-5 18. Nakai Y. lmura H, Yoshimi T. Matsukura S (1973): Adrenergic secretion in man. Acta Endocrinol 74:263-270.
control mechanism
Nordberg A, Larsson C, Adolfsson R, Lalfuzoff I. Winblad B (1983): Muscarinic pensation in hippocampus of Alzheimer patients. .I Neural Trunsm. 56: 13-19.
for ACTH
receptor com-
Perry EK, Tomlinson BE, Blessed G, Perry RH, Cross AJ, Crow ‘IT (1981): Noradrenergic cholinergic systems in senile dementia of Alzheimer type. Lancer ii: 149. Pomara N, Stanley M (1986): The cholinergic hypothesis of memory dysfunction disease revisited. Psychopharmacol Bull 22: 110-I 18.
and
in Alzheimer’s
Pomara N, Oxenkrug GF, McIntyre IM, Block R, Stanley M, Gershon S (1984): Does severity of dementia modulate response to dexamethasone in individuals with primary degenerative dementia? Biol Psychiatry 19: 1481-1487. Raskind M, Peskind E, Rivard MF, Veith R, Barnes R (1982): Dexamethasone Suppression Test and cortisol circadian rhythm in primary degenerative dementia. Am J Psychiatry 139: 1468-J 47 1. Risch SC, Cohen RM, Janowsky DS, Kalin NH, Murphy DL (1980): Plasma beta endorphin and cortisol elevation accompany the mood and behavioral effects of physostigmine in man. Science 209: 1545-1546. Risch SC, Kalin NH, Cohen RM, Weker JL, lnsel TR, Cohen ML, Murphy DL (1981): Muscarinic cholinergic influences on ACTH and beta-endotphin release mechanisms in human subjects. Peptides 2 (Suppl 1):95-97. Risch SC, Janowsky DS. Mott MA, Gillin JC, Kalir HH, Huey LY, Ziegler M, Kennedy B. Turken A (1986): Central and peripheral cholinesterase inhibition: Effects on anterior pituitary and sympathomimetic function. Psychoneuroendocrinology 11:22 I-230. Rosen WG, Mohs RC, Davis KL (1984): A new rating scale for Alzheimer’s Psychiatry 141:1356-1364.
disease.
Am J
Sachar EJ, Asnis G, Nathan S, Halbreich U, Tabrizi MA, HaJpern FS (1980): Dextroamphetamine and cortisol in depression. Morning plasma cortisol levels suppressed. Arch Gen P~chiut~ 37:755-757. Siever LJ, Uhde TW, Jimerson DC, Post RM, Lake CR, Murphy DL (1984): Plasma cortisol responses to clonidine in depressed patients and controls. Arch Gen Rpchiutn, 41:63-68. Spar JE, Gemer R ( 1982): Does the Dexamethasone depression? Am J Psychiatry 139:238-240.
Suppression
Test distinguish
Yates CM, Ritchie IM, Simpson J, Maloney AFJ, Gordon A (1981): Noradrenaline type dementia and Down syndrome. Lance? ii:3940. Zometzer SF (1985): Catecholamine NY Acad Sci 444~242-254.
system involvement
dementia from in Alzheimer-
in age-related memory dysfunction.
Ann
13
Pharmacological Modulation of Cortisol Secretion and Dexamethasone Suppression in Alzheimer’s Disease P. Davous, M. Roudier, M. L. Piketty, C. Abramowitz, and Y. Lamour
We have investigated the dexamethasone suppression of cortisol release in a group of 28 patients with senile dementia of the Alzheimer type (SDAT) after stimulation by physostigmine and clonidine, as compared with basal conditions. All patients but one had previously been evaluated with a depression symptom checklist and had submitted to a standard Dexamethasone Suppression Test (DST). SDAT patients showed normal baseline cortisol values measured at 4:00 PM. DST was reproducible, but nonsuppression did not appear to be a feature of the disease, nor of the dementia syndrome, although a majority of the most demented patients were found to be nonsuppressors. Physostigmine stimulated cortisol secretion in 20 of 24 cases, irrespective of the severity of dementia. Clonidine induced a secretion in 12 of 15 cases, but this was less than that observed after cholinergic stimulation, Physostigmine made cortisol release significantly less sensitive to the suppressive effect of dexamethasone than clonidine in SDAT. This double response should be tested as a possible predictor of a cholinergic therapeutic effect.
Introduction The Dexamethasone Suppression Test (DST) has now been widely studied in dementia, and conflicting results have been reported (Raskind et al. 1982; Spar and Gemer 1982; Balldin et al. 1983; Cames et al. 1983; Coppen et al. 1983; Jenike and Albert 1984; MacKeith 1984; Pomara et al. 1984; Charles et al. 1986; Davis et al. 1986; Greenwald et al. 1986). In a previous study (Davous et al. 1986), we found that a majority of elderly patients with senile dementia of the Alzheimer type (SDAT) were suppressors after dexamethasone administration. Nonsuppression was found in 50% of the most impaired patients, irrespective of their depression score. This result could suggest a cortisol hy-
From the Departments of Neurology (P.D.) and Nuclear Medicine (M.L.P.), Sainte Anne Hospital, Paris; the Departments of Chiabics (M.R.) and Pharmacy (C.A.), Charles Richet Hospital, Villier le Bel, and INSERM U 161 (Y.L.). Paris, France. Supported by Grant 83 C. 1026 from the Ministk de I’lndusttie et de la Recherche. Address reprint requests to Dr. P. Davous, Service de Neurologie, H6pital Sainte Anne, 1 Rue Cabanis, 75674 Paris Cedex 14, France. Received October 27, 1986; revised March 21, 1987.
0 1988 Society of Biological
Psychiatry
OO@-3223/88/$03.50
14
MOL. PSYCHIATRY l’x3x:?1.13 ?‘l
persecretion in dementia, but no signi~cant correlation could be established between plasma cortisol values and the severity of dementia as evaluated by the Mini Mental State Examination (Folstein et al. 197.5). As corticotropin-releasing factor (CRF) and adrenocorticotrophic hormone (ACTH) release are controlled by the hypothalamus and limbic areas (Carroll et al. 1976), in accordance with Janowsky’s theory (1972), it has been suggested that abnormal DST in SDAT might be related to an ~mpai~ent in the hypothalamic-pituita~-adrenal (HPA) axis rather than to cortical lesions (Raskind et al. 1982). It is also well known that cholinergic and noradrenergic systems, which are both impaired in SDAT, are involved in the regulation of ACTH secretion, the former being excitatory (Davis and Davis 1980: Risch et al. 1980), the latter inhibitory (Hillhouse et al. 1975, Sachar et at. 1980). Fu~he~orc, CRF-Iike immunoreactivity is normal in the hy~thalamus of SDAT patients (Bissette et al. 1985), although there is a reduction of choline acetyltransferase (Davies 1979) and noradrenaline (Yates et al. 1981) activities. If ACTH secretion is under cholinergic and noradrenergic control at the hypothalamic level, one should observe a modification of ACTH and cortisol secretion as a result of the imbalance between the two systems in Alzheimer’s disease. The purpose of our investigation was to determine whether or not cholinergic and adrenergic agonists can influence plasma cortisol concentrations and dexamethasone response in SDAT patients.
Methods The study group consisted of 28 SDAT patients (3 men, 25 women), 71-95 years old (mean age 84), who had been free of neuroleptic or antidepressant drugs for at least 3 months. All patients were hospitalized in the same geriatric department for at least 3 months. They were all able to walk and eat without assistance. The diagnosis of probable SDAT was established according to DSM-III (APA 1980) and NINCDS (McKhann et al. 1984) criteria. Patients with neurological signs or symptoms, previous stroke, or any history or evidence of arteriosclerosis were all excluded to rule out multiinfarct dementia. Dementia severity was evaluated by the Mini Mental State Examination (MMSE). The patients had scores from 0 to 19 (mean rt SD, 7.3 t. 5.3). On the basis of their dementia rating, they were subdivided in two subgroups with similar mean age and sex ratio. Seventeen patients had an MMSE score of O-9, and 11 patients had an MMSE score of 10-19. They were assigned, respectively, to group I (severe dementia) and group I1 (mild to moderate dementia). All but one of the patients included in this study had already been rated on a depressive symptom checklist and had been given a standard DST (Davous et al. 1986). They were free of other medical illnesses at the time of pharmacological investigation. As a control group, 14 elderly inpatients living under the same conditions (13 women, I man; mean age 85, range 75-93) were submitted to a standard DST. All of these patients had no previous history of depression, were normal on detailed neurological examination, and had not been exposed to any psychotropic drugs. They were free of cognitive impairment, as assessed by the MMSE: scores ranged from 22 to 27 (mean 25). Two patients had congestive heart failure, two had peripheral arteriopathy, and one had asthma. The others were free of medical illness and had been admitted for sociai reasons.
BIOL PSYCHIATRY 1988~23: 13-24
DST in Alzheimer’s Disease
15
Cholinergic Stimulation and DST was used as a cholinomimetic. Twenty-eight SDAT patients participated in the study 3 months after undergoing a standard DST. They acted as their own controls. The test was carried out over 4 days. Two milligrams of physostigmine was given orally 5 times a day from the second day at 7:00 AM to the fourth day at 2530 PM. One milligram dexamethasone was given on the third day at 11:OOPM. Blood samples for plasma cortisol determination were obtained at 4%) PM on the first day (baseline), third day (stimulation), and fourth day (suppression). Side effects (vomiting) occurred in 4 patients, who were then excluded from the study. Twenty-four patients completed the study without significant side effects. There were 14 patients in group I and 10 patients in group II. Physostigmine
Adrenergic Stimulation and DST Clonidine was used as an adrenomimetic. The study was carried out 4 months after the cholinergic stimulation. During that period, 4 patients had died of cardiac or pulmonary disease. Nineteen patients previously studied under cholinergic stimulation entered the study. They had been clinically stable, with no change of therapy in the intervening period. The test was carried out over 6 days. Clonidine was given orally, 0.15 mg twice a day at 8:00 AM and 8:00 PM, from the second day at 8:00 AM to the sixth day at 8:00 AM. Dexamethasone was given on the fifth day at 11:OOPM. Blood samples for plasma cortisol were obtained at 4:00 PM on the first day (baseline), fifth day (stimulation), and sixth day (suppression). Side effects (sedation and/or hypotension) were observed in 3 patients, who were then excluded from the study. Sixteen patients completed the study. There were 10 patients in group I and 6 patients in group II. In sum, 15 patients previously tested by a standard DST were studied consecutively by cholinergic and adrenergic stimulations and dexamethasone suppression. Cortisol was measured by radioimmunoassay (Davous et al. 1986). Dexamethasone nonsuppression was defined as any cortisol value greater than 5 p,g/dl.
Results Standard DST in SDAT Patients Compared with Controls of Similar Age Group (Table I) Seven of 27 SDAT and 5 of 14 elderly control patients were found to be nonsuppressors after dexamethasone administration. The mean postdexamethasone cortisol values were Table 1. Postdexamethasone Cortisol Values in Controls and SDAT Patients Divided into
SubgroupsAccording to Dementia Severity cortisol MMSE score
Controls
SDAT SDAT
groupI groupII
Postdexamethasone (mean f SEM)
@ange)
n
Age (mean 2 SD)
Nonsuppressors (n)
22-30 o-9 lo-19
14 16 11
85 -c 5.4 82.6 -c 3.1 86 k I
5 I 0
“p < 0.05 comparedwith group I.
(wW
5.2 2 1.1 6.5 2 1.4 2.3 + 0.4”
16
BIOL PSYCHIATRY 1988;23: 13-24
P. Davous et al.
Table 2. Baseline Cortisol Values in SDAT Patients According to Dementia Severity
MMSE
n SDAT group I SDAT group II
(mean +
1s
Cortisol level 4:oO (mean t REM)
score SEM)
(fig/W
4 t 0.9 12.4 ? I.1
10
PM
10.2
lr 0.8
13.1
t
1.7
not significantly different in SDAT and control patients. There was no correlation between co&o1 values and age or MMSE scores. Seven of 16 (43%) SDAT patients in group I were nonsuppressors. All of the patients in group I1 were DST suppressors. The most severly impaired patients (group I) had significantly higher postdexamethasone cortisol values than less affected patients (group II) (Mann-Whitney U-test, U = 45, p < 0.05).
Baseline Cortisol Levels in SDAT Patients Cortisol levels at 4:00 PM were determined twice before pharmacological stimulation in 18 demented patients. All of these subjects had plasma cortisol levels > 5 pg/dl in the two different samples. The mean cortisol values were, respectively, 11.2 k 0.9 (SEM) before administration of physostigmine and 9.5 2 0.8 (SEM) before administration of clonidine. These mean values are not significantly different (I = 1.40, df = 17). Nevertheless, the mean individual difference was 14% (range 0.7-66), and the individual values were not correlated. The cortisol values were not correlated with the MMSE scores. The most impaired patients tended to have lower values than patients less affected, and the difference approached significance (t = 1.66, df = 23, 0.1 < p < 0.2) (Table 2).
Cholinergic
Stimulation
and DST in SDAT (Table 3)
After the oral administration of physostigmine, most of the patients (20124) had higher plasma cortisol values (Figure I ). The mean f SEM value of plasmacortisol was 17.7 2 2.2. significantly different from the predrug values (paired t = 2.82, df = 23, p < 0.01). The range of relative cortisol variations was from - 72% to + 25 1%. Sixteen of 24 patients had levels 20% above their own baseline value, and 7 of them had levels 2
Table 3. Mean 2 Dexamethasone
Nonsuppressors Suppressors Total
SEM
Plasma Cortisol (p,g/dl) in SDAT Patients following Physostigmine n
Baseline
Postphysostigmine
DST
I1
13.07 + I 10.2 rt 1.3 I I 2 k 0.9 (A)
23.06 t 4” 13.3 + 2.2 (B) 17.7 5 2.2 (C)
17.9 k 4.2 2.5 k 0.3 9.5 * 2.5’
13 24
Patients have been separated according to DST results following phyrostigmine compared with A (r = 2.82) “p < 0.001 compared with B (I = 5 2 I) 3 Y4l. ‘p < 0.001 compared wth C (I y
"p < 0.01
and
administration.
BIOL PSYCHIATRY 1988;23:13-24
DST in Alzheimer’s Disease
17
. *
I
c Figure 1. individual plasma cot&of in SDAT patients fn = 24) following physostigmine stimulation (B) and dexamethasone suppression (C) compared with baseline values (A). Horizontal bar indicates the mean in each group.
standard deviations ahove the mean elevation of the group. Dexamethasone suppression was effective in 13 of the 24 patients. The suppression was highly significant in the whole group (paired t = -3.94, df = 23, p < 0.001). After physostigmine administration, the nonsuppressor patients had cortisol values significantly higher than the suppressors (Table 3) (t = 5.21, df = 22, p < 0.001). In fact, the cortisol level after DST was significantly correlated to the cortisol level following physostigmine adminis~tion (r = 0.626, p < 0.01). Baseline cortisol values were not significantly different between the two groups (t = 1.67, df = 22). There was no significant difference between the plasma cortisol values before the cholinergic stimulation and after the DST (t = 0.82, df = 23) in the population of SDAT patients taken as a whole. Studying the effect of cholinergic stimulation as a function of dementia severity, we did not find any significant correlation between MMSE scores and cortisol values measured after oral physostig~ne administration (r = -0.05). Comparison of the patients in groups I and II did not reveal any significant difference in absolute or relative values (Mann-Whitney U-test, U = 56, II = 43, respectively).
IX
BIOLPSYCHIATRk 1988:?3:13-24
Adrenergic
ct al
P. Ihous
Stimulation
and DST in SDAT (Table 4)
After the oral administration of clonidine, most of the patients (12/15) exhibited an elevation of their cortisol value. although it was smaller in magnitude than that observed after cholinergic stimulation. However, as a group. they showed a significant increase, with a mean value of 13.1 2 1.7 (paired t = 2.72, df = 14, p < 0.02) (Figure 2). Six patients showed marked stimulation, from 575%to 195%; 6 patients had a slight stimulation from 2.1% to 24%; and 3 patients had a suppression that ranged from - 4.8% to - 17.6%~ of baseline values. Dexamethasone suppression was effective in 11 of 16 patients. The suppression was highly significant in the whole group (paired t = -~6.75. df = 14. ~7 < 0.00 1). After clonidine administration, the nonsuppressor patients had plasma cortisoi values significantly higher than those of the suppressor patients (t = 4.45. df = 14. p < 0.001). The cortisol level after DST was significantly correlated with the plasma cortisol level after clonidine administration (r = 0.5 17, p < 0.05). Baseline cortisol values were not significantly different between the two groups (t = 1.20. df = 13). In contrast with the results obtained after cholinergic stimulation. there was a significant decrease in the cortisol concentrations of the whole group of patients after clonidine stimulation and dexamethasone suppression when compared with baseline values (paired I = -6.14,df = 14.~ < 0.001). There was no signiticant correlation between plasma cortisol values after clonidine administration and the MMSE scores (r = ~~0.12), nor any significant differences between the two groups according to the severity of dementia. whether considering absolute or relative variations (Mann-Whitney U-tests, U = 22. U = 23. respectively).
Comparative
Study
In the group of 15 demented patients who were consecutively studied with the standard DST and the DST under physostigmine and clonidine, 6 patients were suppressors on the 3 tests, irrespective of the pharmacological effect of drugs on cortisol secretion. In the whole group, the postdexamethasone cortisol values were not significantly different in basal conditions and after cholinergic or adrenergic stimulations (respectively, t = 1.94, t = 0.37, t = 1.28). Demented nonsuppressor patients identified in standard conditions (n = 4) maintained dexamethasone escape after pharmacological stimulation by physostigmine (3 cases), clonidine (3 cases), or both (2 cases). Comparing the individual responses to the cholinergic and adrenergic agonists, it appeared that the 7 patients who had the highest cortisol response to physostigmine were relatively insensitive to clonidine
Table 4. Mean + SEM Plasma Cortisol (kg/dl) in SDAT Patients following Clonidine and Dexamethasone Baseline
Postclonidine
10.4 + 2 8.4 k 0.7 9.2t-0.7(A)
20.4 2 3.4” 9.7 2 1 (B) 13.1k 1.7(C)
”
Nonsuppressors
5
Suppressors
I1
Total
16
Patients have been separated according to DST
“p <
0.02 compared with A (I =
“p< 0.001
compared with B (r = 4.45)
‘p < 0.001
compared wth
C (I =
results
following
clonidine
2.72). 6.75)
and with
A tr :
-6
141.
administration.
DST 7.7 -t I.1 1.6f 0.3 3.52 0.8'
BIOL PSYCHIATRY 1988:23:13-24
DST in Alzheimer’s Disease
CORTI
SOL
PG’DL
25
19
200
n
15 I
0
10.
A
T + n
5-
0 0
f
f __ __ __ __
mm
__
__
_-
-
I
A
R
C
Figure 2. Individual plasma cortisol in SDAT patients (n = 16) following clonidine stimulation (B) and dexamethasone suppression (C) compared with baseline values (A). Horizontal bar indicates the mean in each group.
and vice versa (Figure 3). Nevertheless, there was no significant inverse correlation between the cortisol variations induced by physostigmine and clonidine (r = - 0.293).
Discussion Basal Cortisol in SDAT The concentrations of plasma cortisol measured in baseline conditions at 4:00 PM were within the normal range, as defined in a similar age group with our method (range for controls 60-90 years old: 4-l 1.8). In our study, baseline plasma cortisol values were not correlated with the severity of dementia; the most impaired patients tended to have lower values than the mildly affected patients. Similar findings were reported by Pomara et al. (1984). Conversely, the early morning and midday levels tended to be higher in the most severe forms of dementia. Similar results were found by Davis et al. (1986). Thus, it can be hypothesized that SDAT patients have an impaired circadian rhythm of cortisol, which could be related to several parameters, such as the severity of dementia, locomotor activity, sleep disturbances etc. There is no evidence however that cortisol hypersecretion is a feature of SDAT and is related to the cholinergic deficit.
20
BIOL PSYCHIATR\r 198823:13-24
P. Lkivous et al.
2
1 Figure 3. Comparison of the percentage of cortisof variation from baseline (log-transfarmed) after physostigmine (A) and clonidine (B) in a group of 14 SDAT patients.
-1
R
LIST and SDAT In our sample, nonsuppression was found only in the most demented patients: all patients scoring 10 or more on the MMSE were suppressors. In the group of suppressor patients fn = 20), the plasma cortisol concen~tion after DST was lower than in the age-matched control group. This finding suggests either lower basal values in SDAT or a stronger suppression by dexamethasone in the group of less demented patients. Similar results have been reported by Pomara et al. (1984), who found that the percentage change in cortisol levels from the 8:OOAM baseline to the 8:00 AM postdexamethasone level correlated with dementia severity. From the literature, it appears that the demented patients often show dex~ethasone nonsupp~ssion (Raskind et al. 1982; Spar and Gemer 1982; Balldin et af. 1983; MacKeith 1984; Greenwald et al. 1986), but opposite results have been described (Cames et al. 1983; Castro et al. 1983; Charles et al. 1986) These discrepancies suggest methodological differences among the studies. The percentage of nonsuppressor patients observed in our study (Davous et al. 1986) is similar to the results obtained by Jenike and Albert (1984) and by Pomara et al. ( 1984). From these three studies, it appears that dexamethasone nonsuppression is correlated with dementia severity. However, the relationship between
DST in Alzheimer’s Disease
BlOL PSYCHIATRY 1988;23:13-24
21
dexamethasone suppression and mental impairment is questionable, as we did not find any significant correlation between the plasma cortisol levels after DST and the MMSE scores. In some studies, nonsuppression in SDAT patients has been related to age (Greenwald et al. 1986) and to depression (Cames et al. 1983), but none of these parameters were found to be significantly correlated with DST results by other investigators (Balldin et al. 1983; Jenike and Albert 1984; MacKeith 1984; Davous et al. 1986). Role of Physostigmine Physostigmine, a centrally active cholinesterase inhibitor, has been shown to increase cortisol secretion in normal subjects (Carroll et al. 1978; Risch et al. 1980, Doerr and Berger 1983), and experimental data in animals and humans indicate an action of the drug on the HPA axis (Krieger and Krieger 1967; Hillhouse et al. 1975; Risch et al. 1981). According to Davis and Davis (1980) and Risch et al. (1980), the physostigmineinduced cortisol secretion might be secondary to a stress reaction when the iv route is employed and side effects occur. Such an explanation seems unlikely when physostigmine is given orally, and if peripheral side effects are prevented or excluded by individual dosages. In fact, a recent paper by Risch et al. (1986) provided further evidence that physostigmine’s effects on plasma cortisol concentrations occur via central mechanisms. In the present study, despite probable interindividual differences in drug absorption, most of the patients exhibited an increase in cortisol secretion when given physostigmine orally. A central stimulation is likely, as patients with side effects were excluded from the study. Following stimulation by physostigmine, the DST showed that 55% of SDAT patients are suppressors and 45% avoid suppression. Nonsuppression was directly correlated with the cortisol secretion induced by physostigmine. If there is a relationship between DST nonsuppression and cholinergic activity, the lack of suppression by dexamethasone after physostigmine stimulation could provide another predictive index of therapeutic benefit from cholinomimetics. The absence of any significant correlation between the cortisol secretion changes and the severity of dementia is indicative of the absence of a direct relationship between the cholinergic deficit in SDAT and the HPA axis response to physostigmine. Mohs et al. (1985) found a linear relationship between percent increase in cortisol and mean percent change in symptoms induced by physostigmine. In a similar study, including 10 patients treated by 2 mg oral p~ysostigmine 5 times a day for 1 month, no significant correlation was observed between the relative changes in ADAS subscores (Rosen et al. 1984) and cortisol levels (r = - 0.140) (Davous et al. unpublished observations). In fact, if cortisol secretion were related to cholinergic activity, one should observe in SDAT a hypoactivity of the HPA axis when stimulated by physostigmine, with less response found in patients with more ace~lcholine depletion. Some alternative explanations can be put forth: (1) as suggested by Mountjoy et al. (1984), choline acetyltransferase (CAT) activity is not significantly reduced in SDAT patients compared with elderly controls; (2) there is an overactivity of the postsynaptic muscarinic receptors (Nordberg et al. 1983; Pomara and Stanley 1986); and (3) cortisol secretion is not significantly linked to the cholinergic system in SDAT. In order to understand the relationships between cortisol ~gulation and cholinergic mech~isms in SDAT, further studies are needed, addressing the following matters: intestinal absorption and central bioavailability of the cholinergic drugs, cholinergic status of the patient, basal function of the HPA axis and its reactivity to physiological and pharmacological parameters .
22
BIOL PSYCHIATRl
I’. I)avvus
et al.
IOXX:?3:13-24
Role of Clonidine Clonidine, a central alpha-adrenergic agonist, has been shown to bind to presynaptic as well as postsynaptic receptors (Kobinger 1978). Pharmacological studies suggest a rather complex mechanism of action related to other neurotransmitter systems, such as serotonin and acetylcholine (Komer and Head 1983). In fact, there is still much controversy on the site of action of clonidine at the synaptic level and throughout the different brain nuclei (Jhanwar-Uniyal et al. 1985). In normal subjects, the action of clonidine on cortisol secretion is not clearly demonstrated. Lanes et al. (1983), Siever et al. (1984), and Lechin et al. ( 1985) have reported a suppressive effect. Nakai et al. (1973) have found a stimulation of the HPA axis by alpha-adrenomimetics. La1 et al. (1975) and Doerr and Berger (1983) did not observe significant variations. Conversely, Siever et al. (1984) reported a significant reduction of cortisol secretion in depressed patients compared with controls. To our knowledge, there is no study in the literature on the effects of clonidine on cortisol secretion in SDAT. In the present study, clonidine induced a slight cottisol suppression in 3 cases and stimulation in 12 cases, 6 of them being highly stimulated. Some of the patients who were stimulated by clonidine escaped suppression by dexamethasone. DST nonsuppression could be predicted by the level of cortisol elevation induced by clonidine. The absence of any reliable index of noradrenergic activity in the patients studied, and our uncertainties about the effects of clonidine at the central level. make these results difficult to interpret. Furthermore, in contrast with CAT activity, the deficit in norepinephrine in SDAT has not been found to correlate with symptom severity (Mann et al. 1981; Perry et al. 1981; Francis et al. 1985). It is possible that the effects of clonidine on cortisol secretion, if they are directly mediated by the noradrenergic system, depend on individual norepinephrine release and receptor sensitivity. Another possible explanation is that the clonidine effect is mainly controlled by cholinergic mechanisms, as suggested by Delitala et al. (1983) and Zometzer (1985). According to this hypothesis. loss of normal locus coeruleus function might result in diminished cholinergic efficacy. Such an imbalance is compatible with the results observed in our study. In conclusion, there is no evidence from the present study that cortisol variations reflect abnormalities specific for central cholinergic or monoaminergic neurotransmission. but some results support the hypothesis of individual cholinergic-adrenergic imbalance in the HPA axis functional system in SDAT.
The authors thank A. Dickerson
for his help
in revision of the English text.
References American Psychiatric Association (1980): Diagnostk (ed 3). Washington, DC: APA.
and Statistical Manual of Mental Disorders
Balldin J, Gottfries CG, Karlsson 1, Lindstedt G, Langstrom G, Walinder J (1983): Dexamethasone Suppression Test and serum prolactin in dementia disorders. Br J Psychiatry 143:277-281. Bissette G, Reynolds GP, Kilts CD, Widerlov E, Nemeroff CB (1985): Corticotropin releasing factor-like immunoreactivity in senile dementia of the Alzheimer type. JAMA 254:3067-3069. Cames M, Smith JC, Kahn NH, Bauwens SF (1983): The Dexamethasone Suppression demented outpatients with and without depression. Psychiatp Res 9:337-344. Carroll BJ, Curtis GC. Mendel J (1976): Neuroendocrine Psychiatry 33: 1039-1044.
regulation
in depression.
Test in
Arch Grn
DST in Alzheimer’s Disease
BIOL PSYCHIATRY 1988:23:13-24
23
Carroll BJ, Greden JF, Rubin RT, Haskett R, Feinberg M, Schteingart D (1978): Nemotransmitter mechanism of neuroendocrine disturbance in depression. Acta Endocrinol (SUppl) 22@14. Castro P, Lemaire M, Toscano-Aquilar M, Herchuelz A (1983): Depression, dementia and the
Dexamethasone Suppression Test. Am J Psychiatry 140:1386. Charles GA, Lefevre A, Mire1 J, Rush AJ (1986): Suppression of cortisol following dexamethasone in demented patients. Psychiatry Res 17:173-182. Coppen A, Abou-Saleh M, Milln P, Metcalfe M, Harwood J, Bailey J (1983): Dexamethasone Suppression Test in depression and other psychiatric illness. Br J Psychiatry 142:498-X% Davies P (1979): Neurotransmitter-related enzymes in senile dementia of the Alzheimer type. Brain Res 171:319-327. Davis BM, Davis KL (1980): Cholinergic mechanisms and anterior pituitary hormone secretion. Biol Psychiatry 15303-310. Davis KL, Davis BM, Greenwald BS, Mohs RC, Mathe AA, Johns CA, Horvath TB (1986): Cortisol and Alzheimer’s disease. I. Basal studies. Am J Psychiatry 143:300-305. Davous P, Roudier M, Lamour Y, Susini de Luca H, Abramowitz C (1986): Dexamethasone Suppression Test in senile dementia of the Alzheimer type. Adv Behav Biol 29:355-359. Delitala G, Maioli M, Pacific0 A, Brianda S, Palermo M, Mannelli M (1983): Cholinergic receptor control mechanisms for L-dopa, apomotphine, and clonidine-induced growth hormone secretion in man. J Clin Endocrinol Metab 57: 1145-I 148. Doerr P, Berger M (1983): Physostigmine-induced escape from dexamethasone suppression in normal adults. Biol Psychiatry 18:261-268. Folstein MF, Folstein SE, McHugh PR (1975): Mini Mental State. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189-198. Francis PT, Palmer AM, Sims NR, Bowen DM, Davison AN, Esiri MM, Neary D, Snowden JS, Wilcock GK (1985): Neurochemical studies of early-onset Alzheimer’s disease. N Engl J Med 313:7-11. Greenwald BS, Mathe AA, Mohs RC, Levy MI, Johns CA, Davis KL (1986): Cortisol and Alzheimer’s disease, II: Dexamethasone suppression, dementia severity and affective symptoms. Am J Psychiatry 143sl42-446.
Hillhouse EW, Burden J, Jones MT (1975): The effect of various neurotransmitters on the release of corticotropin releasing hormone from the hypothalamus of the rat. Neuroendocrinology 17:1-11. Janowsky DS, El-Yousef MK, Davis JM, Sekerke HJ (1972): A cholinergic-adrenergic hypothesis of mania and depression. Luncet ii:632-635. Jenike MA, Albert MS (1984): The Dexamethasone Suppression Test in patients with presenile and senile dementia of the Alzheimer’s type. J Am Geriat Sot 32441444. Jhanwar-Uniyal M, Levin BE, Leibowitz SF (1985): Clonidine effects on catecholamine levels and turnover in discrete hypothalamic and extrahypothalamic areas. Brain Res 337:109116. Kobinger W (1978): Central alpha adrenergic systems as targets for hypotensive drugs. Rev Physiol Biochem Pharmacof 81:39-100.
Komer PI, Head GA (1983): Cardiovascular functions of central noradrenergic and serotoninergic neurons in conscious rabbits. Their contributions to the central action of clonidine: Chest (Suppl)2:335-338.
Krieger DT, Krieger HP (1967): Circadian pattern of plasma 17 hydroxycorticosteroid. Alteration by anticholinergic agents. Science 155:1421-1422. La1 S, Tolis G, Martin JB, Brown GM, Guyda H (1975): Effect of clonidine on growth hormone, prolactin, luteinizing hormone, follicle stimulating hormone and thyroid stimulating hormone in the serum of normal man. J Chin Endocrinol Metab 41:827-832. Lanes R, Herrera H, Palacios A, Moncada G (1983): Decreased secretion of cortisol and ACTH after oral clonidine administration in normal adults. Metabolism 32:568-570.
24
BIOL PSYCHIATRY 19X8:23: 13-24
P. Davous et al.
Lechin F, Van der Dijs B, Jakubowicz D, Camero R, Villa S, Lechin E, Gomez F (1985): Effects of clonidine on blood pressure, noradrenaline, cortisol, growth hormone and prolactin plasma levels in high and low intestinal tone subjects. Neuroendocrinology 40:253-26 1, MacKeith IG (1984): Clinical use of the DST in a psychogeriatric 145:389-393.
population.
Br ./ Ps~chiat~
McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984): Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of department of health and human services task force on Alzheimer’s disease. Neurology 34:939-944. Mann JJ, Stanley M, Neophytides A (1981): Central amine metabolism vivo relationship to cognitive deficit. Neurobiol Aging 2:57-60.
in Alzheimer’s
disease: In
Mohs RC, Davis BM, Johns CA, Mathe AA, Greenwald BS, Horvath TB, Davis KL (1985): Oral physostigmine treatment of patients with Alzheimer’s disease. Am J Psychiatry 142:28-33. Mountjoy CQ, Rossor MN, Iversen LL, Roth M (1984): Correlation of cortical cholinergic and GABA deficits with quantitative nemopathological findings in senile dementia. Brain 107:507-5 18. Nakai Y. lmura H, Yoshimi T. Matsukura S (1973): Adrenergic secretion in man. Acta Endocrinol 74:263-270.
control mechanism
Nordberg A, Larsson C, Adolfsson R, Lalfuzoff I. Winblad B (1983): Muscarinic pensation in hippocampus of Alzheimer patients. .I Neural Trunsm. 56: 13-19.
for ACTH
receptor com-
Perry EK, Tomlinson BE, Blessed G, Perry RH, Cross AJ, Crow ‘IT (1981): Noradrenergic cholinergic systems in senile dementia of Alzheimer type. Lancer ii: 149. Pomara N, Stanley M (1986): The cholinergic hypothesis of memory dysfunction disease revisited. Psychopharmacol Bull 22: 110-I 18.
and
in Alzheimer’s
Pomara N, Oxenkrug GF, McIntyre IM, Block R, Stanley M, Gershon S (1984): Does severity of dementia modulate response to dexamethasone in individuals with primary degenerative dementia? Biol Psychiatry 19: 1481-1487. Raskind M, Peskind E, Rivard MF, Veith R, Barnes R (1982): Dexamethasone Suppression Test and cortisol circadian rhythm in primary degenerative dementia. Am J Psychiatry 139: 1468-J 47 1. Risch SC, Cohen RM, Janowsky DS, Kalin NH, Murphy DL (1980): Plasma beta endorphin and cortisol elevation accompany the mood and behavioral effects of physostigmine in man. Science 209: 1545-1546. Risch SC, Kalin NH, Cohen RM, Weker JL, lnsel TR, Cohen ML, Murphy DL (1981): Muscarinic cholinergic influences on ACTH and beta-endotphin release mechanisms in human subjects. Peptides 2 (Suppl 1):95-97. Risch SC, Janowsky DS. Mott MA, Gillin JC, Kalir HH, Huey LY, Ziegler M, Kennedy B. Turken A (1986): Central and peripheral cholinesterase inhibition: Effects on anterior pituitary and sympathomimetic function. Psychoneuroendocrinology 11:22 I-230. Rosen WG, Mohs RC, Davis KL (1984): A new rating scale for Alzheimer’s Psychiatry 141:1356-1364.
disease.
Am J
Sachar EJ, Asnis G, Nathan S, Halbreich U, Tabrizi MA, HaJpern FS (1980): Dextroamphetamine and cortisol in depression. Morning plasma cortisol levels suppressed. Arch Gen P~chiut~ 37:755-757. Siever LJ, Uhde TW, Jimerson DC, Post RM, Lake CR, Murphy DL (1984): Plasma cortisol responses to clonidine in depressed patients and controls. Arch Gen Rpchiutn, 41:63-68. Spar JE, Gemer R ( 1982): Does the Dexamethasone depression? Am J Psychiatry 139:238-240.
Suppression
Test distinguish
Yates CM, Ritchie IM, Simpson J, Maloney AFJ, Gordon A (1981): Noradrenaline type dementia and Down syndrome. Lance? ii:3940. Zometzer SF (1985): Catecholamine NY Acad Sci 444~242-254.
system involvement
dementia from in Alzheimer-
in age-related memory dysfunction.
Ann