0013-7227/02/$15.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 87(5):1949 –1954
Copyright © 2002 by The Endocrine Society
Loss of Brain Volume in Endogenous Cushing’s
Syndrome and Its Reversibility after Correction
of Hypercortisolism
ISABELLE BOURDEAU, CÉLINE BARD, BERNARD NOËL, ISABELLE LECLERC,
MARIE-PIERRE CORDEAU, MANON BÉLAIR, JACQUES LESAGE, LUCIE LAFONTAINE,
ANDRÉ LACROIX
AND
Department of Medicine and Department of Diagnostic Radiology, Division of Endocrinology, Centre Hospitalier de
l’Université de Montréal, Montréal, Canada H2W 1T8
Chronic exposure to excess glucocorticoids results in cognitive and psychological impairment. A few studies have indicated that cerebral atrophy can be found in patients with
Cushing’s syndrome (CS), but its evolution after cure has not
been studied extensively. We report the presence of apparent
cerebral atrophy in CS and its reversibility after the correction of hypercortisolism. Thirty-eight patients with CS, including 21 with Cushing’s disease and 17 with adrenal CS were
studied. The control groups consisted of 18 patients with other
non-ACTH-secreting sellar tumors and 20 normal controls.
Twenty-two patients with CS were reevaluated after cure.
Subjective loss of brain volume was present in 86% of patients
C
USHING’S SYNDROME (CS) results from the chronic
exposure to supraphysiological levels of glucocorticoids and other steroids, which often produce neuropsychological and emotional modifications in affected patients (1–
3). Psychiatric disease is present in up to 66% of patients
suffering from CS and is expressed primarily by major atypical depression in 50 –54% of patients (4, 5). However, the
data on structural alterations of human brain in endogenous
CS are limited and not emphasized in endocrinology or radiology literature as being part of detectable manifestations
of this syndrome (6, 7). An older study using pneumoencephalography described the presence of diffuse cerebral
atrophy in patients with Cushing’s disease (CD) (8). More
recently, using magnetic resonance imaging (MRI), Starkman et al. (9) found decreased hippocampal formation volume in 27% of 12 patients with CS, in whom the values were
below the 95% confidence interval of a literature reference
normal population; however, the other areas of the brain
were not described. Recently, Simmons et al. (10), using a
subjective score in assessing computerized tomography
(CT)/MRI scans, also noted significant cerebral atrophy in 63
patients with CD, compared with a control group. The morphological changes observed in the brain imaging of CS
patients have been frequently referred as representing cerebral atrophy, but we prefer to use the term loss of brain
Abbreviations: ACS, Adrenal Cushing’s syndrome; CD, Cushing’s
disease; CS, Cushing’s syndrome; CT, computerized tomography; MRI,
magnetic resonance imaging; NST, no sellar tumors; OST, other (nonACTH-secreting) sellar tumors.
with Cushing’s disease and 100% of patients with adrenal CS.
The values for third ventricle diameter, bicaudate diameter,
and subjective evaluation were significantly increased in CS
groups in comparison with the control group (P < 0.001). Imaging reevaluated at 39.7 ⴞ 34.1 months after achieving
eucortisolism showed an improvement of the third ventricle
diameter (P ⴝ 0.001), bicaudate diameter (P < 0.0005), and
subjective evaluation (P ⴝ 0.05). We conclude that brain volume loss is highly prevalent in CS and is at least partially
reversible following correction of hypercortisolism. (J Clin
Endocrinol Metab 87: 1949 –1954, 2002)
volume in view of the lack of demonstration of true pathological cell atrophy or loss.
The potential reversibility of this premature loss in brain
volume in CS after correction of hypercortisolism was evaluated in only two reports, to date. Heinz et al. (11) described
a 9-yr-old boy with CD and apparent severe cerebral atrophy
who recuperated almost completely after cure. Starkman et
al. (12) reported a mean 3.2% ⫾ 2.5% (sd) increase of hippocampal formation volume following treatment of 22 patients with CD.
In this report, we studied a population of 38 patients with
endogenous CS, either with CD or adrenal CS (ACS), to
assess the presence and distribution of loss of brain volume,
using both objective (third ventricle diameter and bicaudate
diameter) and subjective measures on CT and/or MRI examinations. The possible relationship between urinary free
cortisol and tumor size with premature signs of decreased
brain volume was also examined. In addition, we evaluated
the evolution of brain volume after the correction of hypercortisolism in 22 patients.
Materials and Methods
Patients
In the first part of the study, clinical and radiological records of
patients who underwent CT and/or MRI of the pituitary gland during
the etiological investigation of CS were reviewed. Data were collected
retrospectively among clinical and radiological records from 1984 to
1996 and prospectively between 1997 and 1999. All patients and control
groups were younger than 65 yr of age, free of other chronic diseases,
previous cerebral trauma, or neurological diseases. Four groups were
included: patients with proven CS secondary to CD (21 patients) or
1949
1950
J Clin Endocrinol Metab, May 2002, 87(5):1949 –1954
Bourdeau et al. • Loss of Brain Volume in CS
primary adrenal causes (ACS, 17 patients) along with control groups
composed of patients with other (non-ACTH-secreting) sellar tumors
(OST, 18 patients) and normal controls with no sellar tumors (NST, 20
patients). The control groups were paired for age with the CD group. The
selection criteria were as follows: biochemically and surgically proven
ACTH-secreting pituitary corticotroph adenoma for CD and primary
cortisol secreting adenoma or bilateral macronodular hyperplasia for
ACS; pituitary adenoma of other etiology (OST) included nine prolactinomas, seven nonfunctioning adenomas, one GH-secreting tumor, and
one mixed GH- and PRL-secreting tumor. The NST group was part of
a group referred since 1983 for a possible increased sella turcica or an
asymmetric sellar floor on skull x-ray, which were shown to be normal
on CT investigation. Number, sex, and mean age for each group are
listed in Table 1.
In the second part of the study, we evaluated the evolution of the signs
of atrophy in 22 patients (14 with CD and 8 with ACS) in whom determinations were available following correction of hypercortisolism.
Five patients were studied retrospectively and 17 prospectively following the initial phase of the study. Mean age of patients included in the
follow-up study was 40.9 ⫾ 10.7 yr at the time of initial examination.
Cure was obtained surgically in 14 patients with CD and in 7 patients
with ACS. One patient with CS secondary to ACTH-independent macronodular adrenal hyperplasia and ectopic LH/human CG adrenal
receptors achieved normal urinary free cortisol levels following medical
therapy with long-acting leuprolide acetate (13). The time of correction
of hypercortisolism was defined as the date when postoperative hydrocortisone replacement was 30 mg or less per day, with normal free
urinary cortisol. The mean time between reaching normal cortisol replacement and imaging reevaluation was 39.7 ⫾ 34.1 months (mean ⫾
sd). The institutional review committee approved the study and patients
recruited prospectively provided informed consent.
Urinary-free cortisol levels measured in two to four baseline 24-h
collections were retrieved from patients’ charts. Urinary cortisol measurements were performed either by a commercial RIA kit (Quanticoat Kallestad Diagnostics, Chaska, MN) before 1997 and since then
by an immunofluorometric assay (Immuno I system, Bayer Corp.,
Tarrytown, NY).
Imaging techniques
All patients and control subjects were examined with conventional
coronal CT and/or coronal MRI examinations performed before pituitary/adrenal surgery or the beginning of medical therapy. CT and MRI
studies were done in the three hospital sites of Center Hospitalier de
l’Université de Montréal. For the purpose of consistency and validity of
the data, only coronal CT and MRI images perpendicular to the pituitary
gland were assessed. On MRI, measurements were done on routine
coronal spin-echo T1 images on a 1.5 and 1.0 TESLA MRI unit. Imaging
included sagittal and coronal T1 spin-echo sequence (350 – 600/18 –25/
2– 4, 3-mm-thick section/16 –20 cm FOV, 192 matrix) without and with
gadolinium. Only the more recent examinations included T2 spin-echo
sequences.
Measures
Three parameters were used to evaluate indices of apparent atrophy.
They included the third ventricle diameter measured at the level of the
foramen of Monroe (14, 15) and the bicaudate diameter measured in the
largest distance between the midportion of the head of the caudate nuclei
(16). Two blinded independent radiologists performed subjective estimation of the degree of apparent cerebral atrophy using a scale graded
from 0 –3: 0 (no atrophy), 1 (mild atrophy), 2 (moderate atrophy), and
3 (severe atrophy). Comments on the distribution of atrophy were also
recorded. Pituitary or adrenal tumor sizes were measured in their largest
diameter when they were radiologically visible.
Statistical analysis
A group-matched design was used for age, with mean age difference
within 2 yr between the OST and NST groups with the CD group. The
ACS group was collected later and was not matched for age with other
groups; they were on average 8 yr older than the other groups (Table 1).
There was no correlation between age and sex with the various measures
of brain volume loss. There was therefore no need to adjust the correlations for age and sex. CT and MRI were considered equivalent for
accuracy in evaluating cerebral volume loss. Comparison among the
four groups was performed using one-way ANOVA. When differences
were identified, the Tukey B procedure was performed as a multiplerange test. This approach allows comparing all pairs of means (three
pairs when three means are involved) and keeping the overall ␣ level
at 0.05, thus avoiding inflation of type one error (7). The relationship
among tumor sizes, 24-h urinary-free cortisol levels in CD and ACS, and
the degree of brain volume loss was studied using the Pearson product
moment correlation in CD and the Spearman product moment correlation in the ACS. The Spearman’s method is not parametric and better
adapted to the size of the adrenal tumors. Initial and last available
radiological evaluations of brain atrophy were compared using ANOVA
for repeated measures.
Results
Assessment of the loss of brain volume
The evaluation of brain volume loss in the various groups
of patients at the initial basal period is presented in Table 1.
Bicaudate diameter could not be determined in eight patients
and third ventricle diameter in three patients because CT
examinations contained magnified pictures that did not allow these determinations. There were no differences in the
various measures of indices of cerebral volume between the
OST and NST groups. Significant increases in third ventricle
diameter, bicaudate diameter, and subjective evaluation
were present both in the CD and ACS groups, compared with
the NST group. There were no differences in those parameters between the CD and ACS groups. The data for the CD
and ACS groups could thus be pooled in a combined CS
group and compared with the NST control group. The mean
third ventricle diameter value was 5.78 mm in the pooled CS
groups, which is 1.9 times wider than in NST (P ⫽ 0.001). The
average bicaudate diameter was 1.3 times larger in CS, compared with the NST (P ⫽ 0.001). The strongest association
was seen with the subjective assessment of cerebral atrophy
TABLE 1. Evaluation of apparent brain volume loss in the various groups of patients
CD (n ⫽ 21)
ACS (n ⫽ 17)
Pooled CS (CD ⫹ ACS; n ⫽ 38)
OST (n ⫽ 18)
NST (n ⫽ 20)
Sex
(M:F)
Mean age
(yr ⫾ SD)
Third ventricle diametera
(mm)
Bicaudate diameter
(mm)
Subjective evaluationb
8:13
1:16
9:29
4:14
3:17
37.3 ⫾ 11.1
45.4 ⫾ 12.0
41.3 ⫾ 12.0
35.4 ⫾ 8.4
37.3 ⫾ 12.8
5.61 ⫾ 2.33c (21)
6.00 ⫾ 1.83c (16)
5.78 ⫾ 2.10c (37)
3.68 ⫾ 1.24 (17)
3.08 ⫾ 1.13 (19)
15.81 ⫾ 4.88c (19)
18.30 ⫾ 3.51c (15)
16.90 ⫾ 4.45c (34)
12.82 ⫾ 3.80 (17)
12.74 ⫾ 2.47 (17)
1.69 ⫾ 1.03d (21)
1.94 ⫾ 0.83d (17)
1.80 ⫾ 0.94d (38)
0.17 ⫾ 0.38 (18)
0.10 ⫾ 0.31 (20)
Mean ⫾ SD; the number of subjects in each group is indicated in parentheses.
Subjective evaluation: 0, no atrophy; 1, mild atrophy; 2, moderate atrophy; 3, severe atrophy.
c
P ⫽ 0.001 compared with NST.
d
P ⬍ 0.001 compared with NST.
a
b
Bourdeau et al. • Loss of Brain Volume in CS
J Clin Endocrinol Metab, May 2002, 87(5):1949 –1954 1951
in which the pooled CS obtained a mean ranking close to 2
(moderate atrophy), and the mean was near 0 (no atrophy)
in the NST control group (Table 2; P ⬍ 0.001).
Urinary-free cortisol levels were similarly elevated (normal ⬍330 nmol/d) in CD (1,376 ⫾ 774 nmol/d) and in the
ACS (1,127 ⫾ 609 nmol/d) groups. Urinary-free cortisol levels were not correlated with measures of apparent cerebral
atrophy except positively with the bicaudate diameter in the
CD group when one exceptionally elevated value (16,973
nmol/d) was discarded (r ⫽ 0.43, P ⫽ 0.04). In ACS, the
urinary-free cortisol was positively correlated only with the
subjective assessment of apparent cerebral atrophy (r ⫽ 0.41,
P ⫽ 0.04); there was, however, no correlation between urinary-free cortisol levels and brain volume loss when the
patients with ACS and CD were pooled in the same group.
A positive correlation was also found between pituitary tumor size and third ventricle diameter (r ⫽ 0.854, P ⬍ 0.001),
bicaudate diameter (r ⫽ 0.629, P ⫽ 0.002), and subjective
evaluation (r ⫽ 0.454, P ⫽ 0.02) in the CD group. In the
patients with ACS, there was no correlation between the size
of adrenal tumors and the extent of brain volume loss.
No correlation was found between age and any measures
of brain volume loss in this relatively young population. On
the basis of subjective evaluation, apparent cerebral atrophy
was present in 86% of patients with CD (14% mild, 43%
moderate, 29% severe) and 100% of ACS (35% mild, 35%
moderate, 30% severe), compared with 10% in controls (NST)
(P ⬍ 0.005). Subjective evaluation of loss of brain volume in
the groups with CD or ACS revealed a diffuse cortical and
subcortical apparent atrophy on the basis of sulcal and ventricle enlargement (Fig. 1).
Evolution of the loss in brain volume following correction
of hypercortisolism
There was a significant improvement in the estimations of
third ventricle, bicaudate diameter, and subjective evaluation in the group of 22 patients reevaluated after the correction of hypercortisolism (Table 2 and Fig. 1). The majority of
patients showed regression of the three measurements.
The effect of duration of correction of hypercortisolism on
this improvement was studied. In examining sequential estimations in individual patients in which repeated measures
were performed, there appeared to be a progressive improvement that has not reached a plateau yet in our patient
with the longest sequential follow-up period (Fig 2). The
subgroup of 11 patients with a follow-up from 7–24 months
(mean 15.5 ⫾ 4.8 months) was compared with the second
group of 11 patients with a follow-up of 25–123 months
TABLE 2. Evolution of the parameters of loss of brain volume
after correction of the hypercortisolism in 22 patients with
Cushing’s syndrome
Initial value
Following correction
of hypercortisolism
a
b
Third ventricle
diameter
(mm)
Bicaudate
diameter
(mm)
Subjective
evaluation
scale of atrophy
5.48 ⫾ 2.48a
3.95 ⫾ 1.96b
17.14 ⫾ 4.69
14.09 ⫾ 3.69c
1.70 ⫾ 1.09
1.23 ⫾ 1.05d
Mean ⫾ SD.
P ⫽ 0.001; c P ⬍ 0.0005; d P ⫽ 0.05 compared with initial values.
(mean 63.9 ⫾ 33.5 months); there were no differences in the
improvement of the two subgroup values for the three parameters of brain volume assessment.
Discussion
This study examined the effects of chronic excess of endogenous glucocorticoids in 38 patients with either pituitary
or adrenal etiologies of CS on the loss of brain volume. Using
modern neuroradiological diagnostic methods and both objective and standardized subjective measures of apparent
cerebral atrophy, our data show that loss of brain volume is
present in a very large proportion of patients either with
pituitary corticotroph adenomas or primary adrenal etiologies. The subjective assessment showed a diffuse pattern of
apparent cerebral atrophy in accordance with older pneumoencephalographic data (8).
The objective measures, the bicaudate diameter and third
ventricle diameter, were both found to be significantly larger
in CS patients, compared with the normal control group
(NST); no such findings were present in the patients with
other sellar tumors (except for grade 1 subjective atrophy in
one case of acromegaly), excluding that a loss in brain volume is related to the mere presence of a pituitary lesion. In
this study, these objective measures are used as an indirect
sign of volume loss (hydrocephalus ex vacuo) in correlation
with the increased subarachnoid space. They could also be
related to an increased in cerebral spinal fluid volume in the
ventricles.
This study revealed that a similar degree of brain structure
modifications was present in the 17 primary adrenal CS
patients as in the 21 CD patient group; only two other ACS
patients had been studied previously (9). In a study of the
neuropsychiatric manifestations of patients with CS, a relationship was found between the neuropsychiatric disability
rating and ACTH levels (1), suggesting that ACTH levels
may play a role in cognitive function. Because we found no
difference in loss of brain volume between patients with
ACTH-dependent (CD) and ACTH-suppressed (ACS) patients, our results suggested that ACTH levels are not an
independent factor for cerebral structural alterations in endogenous CS.
The absence of consistent correlations between the 24-h
urinary-free cortisol values and the indices of brain volume
loss is not surprising, taking into account the lack of precise
data on the duration of the hypercortisolism, as well as the
variability of cortisol production in CS patients on a day-today basis (6). Difficulty to evaluate the onset of the disease
prevented us from determining the effect of duration of
exposure to excess cortisol production.
The mainly retrospective nature of this study posed some
methodological restrictions. The coronal studies were centered on the sella turcica but were not optimally standardized. They did not all include the landmarks necessary to
perform all the objective measurements. It did not allow us
to use planimetry (17, 18) or histogram curve analysis (19, 20)
methods. We have therefore included a set of subjective
assessments of the overall aspect of visible central nervous
system structures. The ACS group was collected later and
was not age matched with NST. However, because there was
1952
J Clin Endocrinol Metab, May 2002, 87(5):1949 –1954
Bourdeau et al. • Loss of Brain Volume in CS
FIG. 1. Coronal T1-weighted MRI. A, A 33-yr-old normal control woman without any sellar tumor. B, A 42-yr-old man with a surgically proven
ACTH-secreting pituitary macroadenoma shows brain volume loss characterized by increased subarachnoid space and ventricle enlargement.
C, A 33-yr-old woman with CS from a surgically proven adrenal adenoma also presents similar evidence of apparent brain volume. D, Control
MRI 12 months after correction of hypercortisolism in patient shown in C. Regression of brain volume loss is illustrated by the reduction of
ventricles and sulci dilatation.
no correlation between the age and different measurements
of brain volume loss among the groups, this does not modify
our conclusion.
Although limited in number, reported autopsy data revealed cerebral atrophy in CS (8, 21–23). Pharmacological
doses of glucocorticoids were also found to cause apparent
cerebral atrophy in 10% of a young population on CT scans
(24). In children treated for congenital adrenal hyperplasia
with slightly supraphysiological doses of glucocorticoids,
increased prevalence of temporal lobe atrophy was found on
MRI (25). Other studies demonstrated a possible link between the increased activity of the pituitary-adrenal axis and
cerebral atrophy in the context of alcoholism (26), depression
(27–29), and stress (30, 31).
Data on the evolution of the apparent cerebral atrophy
following correction of hypercortisolism have been very lim-
Bourdeau et al. • Loss of Brain Volume in CS
FIG. 2. Sequential determinations of objective indices of brain volume loss during 45 months of follow-up after correction of hypercortisolism in a woman with CD. Both parameters improve gradually and
have not reached a plateau yet.
ited, to date (11, 12). In our study of 22 CS patients reevaluated after correction of hypercortisolism, a mean improvement of 28% of the third ventricle diameter and 17.8% of the
bicaudate diameter was observed. The longer follow-up period and the differences in measures and areas of brain volume in this study, compared with that of Starkman et al. (12),
might explain the more important improvement noted here.
Despite a significant improvement following correction of
hypercortisolism, it must be stressed that the values did not
return to the normal control group values (Table 2); longer
follow-up will be necessary to determine whether complete
correction will be possible. The partial reversibility of anatomical modifications of brain volume is in agreement with
the recovery of the brain choline level (marker of degradation
products of cell membrane) in the frontal and thalamic areas
in CS following correction of hypercortisolism (32).
The pathogenesis of loss in brain volume induced by
chronic glucocorticoid excess is probably multifactorial (2,
33). Glucocorticoids can occupy both MRs and GRs. MRs are
usually protected from glucocorticoid exposure by the catabolic effects of 11␤-HSD2, which converts cortisol into the
inactive cortisone molecule; however, 11␤-HSD2 is not expressed in hippocampus or other limbic structures allowing
MR activation by glucocorticoids (2). High-affinity MRs,
mainly localized in hippocampal regions, are heavily occupied by basal levels of adrenal steroids during the diurnal
cycle; elevation of glucocorticoid concentrations in stress and
CS increases the occupation of the lower-affinity GRs, which
are more largely distributed in several regions of the brain
(2). The occupation of MRs in hippocampus cells in culture
by physiological concentrations of cortisol is essential to
maintain neuronal cell survival and function; the occupation
of GRs by supraphysiological doses of glucocorticoids initially leads to decreased cell excitability and a reversible
phase of atrophy of apical dendrites of CA3 pyramidal neurons in culture (33, 34). If the exposure to excess glucocorticoids persists, neuronal cell death can occur.
Glucocorticoids have been shown to increase the synaptic
accumulation of glutamate, possibly through an effect on its
removal by glial cells; the enhanced stimulation of N-methyld-aspartate receptors will increase intracellular cytosolic
J Clin Endocrinol Metab, May 2002, 87(5):1949 –1954 1953
Ca2⫹ in postsynaptic neurons, which activates several processes leading to increased susceptibility to injury (cell
endangerment) and cell death (2, 33, 34). Whereas chronic
absence of glucocorticoids leads to an apoptotic-like hippocampal cell death (2), the cell endangerment following excess
glucocorticoids does not involve an apoptotic mechanism
(35). The mechanisms implicated are not fully understood yet
but may also involve effects of glucocorticoids on several
growth factors and receptors including brain-derived neurotropic factor, nerve growth factor, basal fibroblast growth
factor, and TGFs (2). A reduced glucose metabolism is observed on positron emission tomography scan of patients
receiving high-dose glucocorticoids (36); this may be secondary to the inhibitory action of glucocorticoids on glucose
transporter synthesis and translocation in neurons and glial
cells (33, 37). It must be mentioned that the studies of the
effects of glucocorticoids have been conducted essentially in
hippocampal cells in culture and it is unknown whether the
same mechanisms apply to neurons in other brain regions.
However, the partial reversibility of cerebral atrophy after
correction of hypercortisolism would certainly indicate that
the decrease in brain volume is not only secondary to neuronal cell death. Dexamethasone is very potent in treating
cerebral edema (38), and the loss of brain volume in CS could
be secondary, in part, to a loss in water; this decrease in water
content of brain tissue has been assessed in vivo in animal
models (39).
We conclude that loss of brain volume is not limited to
hippocampal volume formation and is very prevalent in
patients with CS from pituitary or adrenal etiologies. The loss
of brain volume is at least partially reversible after the correction of hypercortisolism. Signs of cerebral atrophy should
be evaluated in patients suffering of CS. Further studies will
be necessary to determine whether complete reversal of brain
volume loss is possible and whether these observations correlate with neuropsychological improvement (1, 40).
Acknowledgments
We thank Marc Dumont, M.Ps., for statistical analysis; Louis Bélair,
M.D., for contribution to this work; and Victoria Baranga for secretarial
assistance.
Received June 14, 2001. Accepted January 22, 2002.
Address all correspondence and requests for reprints to: Céline Bard,
M.D., Department of Diagnostic Radiology, Hôtel-Dieu du CHUM, 3840
St-Urbain Street, Montréal, Québec, Canada H2W 1T8. E-mail:
[email protected].
This work was supported by a grant from the Association des Radiologistes du Québec, Fonds de la Recherche en Santé du Québec (FRSQ
980702). This work was presented in part at the 81st Annual Meeting of
The Endocrine Society, San Diego, California, 1999.
References
1. Starkman MN, Schteingart DE 1981 Neuropsychiatric manifestations of patients with Cushing’s syndrome. Relationship to cortisol and adrenocorticotropic hormone levels. Arch Intern Med 141:215–219
2. De Kloet ER, Vreugdenhil E, Oitzl MS, Joels M 1998 Brain corticosteroid
receptor balance in health and disease. Endocr Rev 19:269 –301
3. Forget H, Lacroix A, Somma M, Cohen H 2000 Cognitive decline in patients
with Cushing’s syndrome. J Int Neuropsychol Soc 6:20 –29
4. Sonino N, Fava GA, Raffi AR, Boscaro M, Fallo F 1998 Clinical correlates of
major depression in Cushing’s disease. Psychopathology 31:302–306
5. Dorn LD, Burgess ES, Dubbert B, Simpson SE, Friedmann T, Kling M, Gold
PW, Chrousos GP 1995 Psychopathology in patients with endogenous Cush-
1954
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
J Clin Endocrinol Metab, May 2002, 87(5):1949 –1954
ing’s syndrome: ‘atypical’ or melancholic features. Clin Endocrinol (Oxf) 43:
433– 442
Nieman LK 2001 Cushing’s syndrome. In: De Groot LJ, Jameson JLE, eds.
Endocrinology. Philadelphia: WB Saunders Co.; 1691–171
Osborn AG 1994 Diagnostic neuroradiology. St. Louis: Mosby Year Book Inc.;
651
Momose KJ, Kjellberg RN, Kliman B 1971 High incidence of cortical atrophy
of the cerebral and cerebellar hemispheres in Cushing’s disease. Radiology
99:341–348
Starkman MN, Gebarski SS, Berent S, Schteingart DE 1992 Hippocampal
formation volume, memory dysfunction, and cortisol levels in patients with
Cushing’s syndrome. Biol Psychiatry 32:756 –765
Simmons NE, Do HM, Lipper MH, Laws Jr ER 2000 Cerebral atrophy in
Cushing’s disease. Surg Neurol 53:72–76
Heinz ER, Martinez J, Haenggeli A 1977 Reversibility of cerebral atrophy in
anorexia nervosa and Cushing’s syndrome. J Comput Assist Tomogr 1:415– 418
Starkman MN, Giordani B, Gebarski SS, Berent S, Schork MA, Schteingart
DE 1999 Decrease in cortisol reverses human hippocampal atrophy following
treatment of Cushing’s disease. Biol Psychiatry 46:1595–1602
Lacroix A, Hamet P, Boutin JM 1999 Leuprolide acetate therapy in luteinizing
hormone-dependent Cushing’s syndrome. N Engl J Med 341:1577–1581
Hawrylyshyn PA, Tasker RR, Organ LW 1976 Third ventricular width and the
thalamocapsular border. Appl Neurophysiol 39:34 – 42
Dyke CG, Davidoff LM 1940 The pneumo-encephalographic appearance of
hemangioblastoma of the cerebellum. Am J Roentgenol Radiat Ther 44:1– 8
Barr AN, Heinze WJ, Dobben GD, Valvassori GE, Sugar O 1978 Bicaudate
index in computerized tomography of Huntington disease and cerebral atrophy. Neurology 28:1196 –1200
Winer BJ 1971 Statistical principles in experimental designs. New York:
McGraw Hill
Synek V, Reuben JR 1976 The ventricular-brain ratio using planimetric measurement of EMI scans. Br J Radiol 49:233–237
Schwartz M, Creasy H, Grady CL, DeLeo J, Frederickson HA, Cutler NR,
Rapoport SI 1985 Computed tomographic analysis of brain morphometrics in
30 healthy men. Ann Neurol 17:147–157
Huber VG 1982 Measurements of the intracranial CSF spaces by computer
tomography. Fortschr Röngenstr 137:42– 47
Trethowan WH, Cobb S 1952 Neuropsychiatric aspects of Cushing’s syndrome. Arch Neurol Psychiatry 67:283–309
Cope O, Raker JW 1955 Cushing’s disease. The surgical experience in the care
of 46 cases. N Engl J Med 253:119 –129
Soffer LJ, Iannaccone A, Gabrilove JL 1961 Cushing’s syndrome. A study of
fifty patients. Am J Med 30:129 –146
Bentson J, Reza M, Winter J, Wilson G 1978 Steroids and apparent cerebral
atrophy on computed tomography scans. J Comput Assist Tomogr 2:16 –23
Nass R, Heier L, Moshang T, Oberfield S, George A, New MI, Speiser PW
1997 Magnetic resonance imaging in the congenital adrenal hyperplasia pop-
Bourdeau et al. • Loss of Brain Volume in CS
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
ulation: increased frequency of white-matter abnormalities and temporal lobe
atrophy. J Child Neurol 12:181–186
Marchesi C, De Risio C, Campanini G, Piazza P, Grassi M, Chiodera P,
Vescovi PP 1994 Cerebral atrophy and plasma cortisol levels in alcoholics after
short or a long period of abstinence. Prog Neuropsychopharmacol Biol Psychiatry 18:519 –535
Rothschild AJ, Benes F, Hebben N, Woods B, Luciana M, Bakanas E, Samson
JA 1989 Relationships between brain CT scan findings and cortisol in psychotic
and nonpsychotic depressed patients. Biol Psychiatry 26:565–575
Sheline YL, Wang PW, Mokhtar HG, Csernansky JG, Vannier MW 1996
Hippocampal atrophy in recurrent major depression. Proc Natl Acad Sci USA
98:3908 –3913
Bremner JD, Narayan M, Anderson ER, Staib LH, Miller HL, Charney DS
2000 Hippocampal volume reduction in major depression. Am J Psychiatry
157:115–118
Lupien SJ, de Leon M, de Santi S, Convit A, Tarshish C, Nair NP, Thakur
M, McEwen BS, Hauger RL, Meaney MJ 1998 Cortisol levels during human
aging predict hippocampal atrophy and memory deficits. Nat Neurosci
1:69 –73
Bremner JD, Randall P, Scott TM, Bronen RA, Sebyl JP, Southwick SM 1995
MRI-based measurement of hippocampal volume in patients with combatrelated post-traumatic stress disorder. Am J Psychiatry 151:973–981
Khiat A, Bard C, Lacroix A, Boulanger Y 2000 Recovery of the brain choline
level in treated Cushing’s patients as monitored by proton magnetic resonance
spectroscopy. Brain Res 862:301–307
McEwen BS, Sapolsky RM 1995 Stress and cognitive function. Curr Opin
Neurobiol 5:205–216
Sapolsky RM 1994 The physiological relevance of glucocorticoid endangerment of the hippocampus. Ann N Y Acad Sci 746:294 –304
Masters JN, Finch CE, Sapolsky RM 1989 Glucocorticoid endangerment of
hippocampal neurons does not involve deoxyribonucleic acid cleavage. Endocrinology 124:3083–3088
Fulham MJ, Brunetti A, Aloj L, Raman R, Dwyer AJ, Di Chiro G 1995
Decreased cerebral glucose metabolism in patients with brain tumors: an effect
of corticosteroids. J Neurosurg 83:657– 664
Horner HC, Packan DR, Sapolsky RM 1990 Glucocorticoids inhibit glucose
transport in cultured hippocampal neurons and glia. Neuroendocrinology
52:57– 64
Criscuolo GR 1993 The genesis of peritumoral vasogenic brain edema and
tumor cysts: a hypothetical role for tumor-derived vascular permeability factor. Yale J Biol Med 66:277–314
Tjuvajev J, Uehara H, Desai R, Beattie B, Matei C, Zhou Y, Kreek MJ,
Koutcher J, Blasberg R 1996 Corticotropin-releasing factor decreases vasogenic brain edema. Cancer Res 56:1352–1360
Dorn LD, Burgess ES, Friedman TC, Dubbert B, Gold PW, Chrousos GP 1997
The longitudinal course of psychopathology in Cushing’s syndrome after
correction of hypercortisolism. J Clin Endocrinol Metab 82:912–919