Академический Документы
Профессиональный Документы
Культура Документы
Introduction
Cushing syndrome is a clinical entity that reflects
chronic and excessive exposure of the body to gluco
corticoids. It includes numerous clinical features that
reflect the widespread distribution of glucocorticoid
receptors in multiple target organs.1,2 In most instances,
Cushing syndrome develops secondary to the use of
exogenous glucocorticoids, so-called iatrogenic Cushing
syndrome. Spontaneous or endogenous Cushing syn
drome is far less frequent and results from an excessive
production of cortisol by the adrenal cortex. From a
pathophysiological perspective, three conditions can
lead to endogenous Cushing syndrome.2,3 In the major
ity of cases, the excessive production of cortisol is driven
by an inappropriate secretion of adrenocorticotropic
hormone (ACTH) by a pituitary corticotroph adenoma,
so-called Cushing disease. Rarely, ACTH can be pro
duced by a nonpituitary (ectopic) tumor. This entity
is called the ectopic ACTH syndrome. Finally, cortisol
can be autonomously oversecreted by unilateral or, more
rarely, by bilateral adrenal tumors.
Department of
Endocrinology
(A.Tabarin), Clinical
Epidemiology Unit
(USMR) & CIC-EC7,
INSERM, (P. Perez),
Universit Bordeaux 2,
Centre Hospitalier
Universitaire de
Bordeaux, 146Rue Lo
Saignat, F33076
Bordeaux CEDEX,
France.
Correspondence to:
A. Tabarin
antoine.tabarin@
chu-bordeaux.fr
REVIEWS
Key points
Common clinical experience and a few epidemiological studies suggest that
endogenous Cushing syndrome is a rare condition
Systematic screening studies performed in specific populations that exhibit
nonspecific symptoms of Cushing syndrome (mainly patients with diabetes
mellitus) reveal a prevalence of occult Cushing syndrome that reaches 3.3%
A screening strategy is justified when its efficacy is supported by evidence
and if benefits outweigh potential drawbacks
Examination of the indications for and against screening for occult Cushing
syndrome reveals that the cons exceed the pros
Additional studies are needed before we can acknowledge, from an
evidence-based perspective, the usefulness of systematic screening for occult
Cushing syndrome
Cushing disease
Pseudo-Cushing syndrome
Glucose intolerance
Hirsutism
Acne
Hypertension
Myopathy
Purple striae
*
*
Easy bruising
Visceral obesity
25
50
Prevalence of symptoms (%)
75
100
REVIEWS
prevalence of occult Cushing syndrome in patients with
diabetes mellitus reached 9.4% and was 4.8-fold more
frequent than in the control group. Similarly, Leibowitz
etal.26 performed a screening study using the 1mg DST in
90 overweight individuals with poorly controlled diabetes
mellitus (HbA1c >8%) and identified 3.3% of patients with
occult Cushing syndrome. Another screening study per
formed in 100 patients with newly diagnosed type1 and
type2 diabetes mellitus who were not selected for their
clinical characteristics, such as poor metabolic control
or obesity, identified one case (1%) of occult Cushing
syndrome.27 Two methodologically acceptable negative
studies have also been published. A screening strategy per
formed in 154 elderly men with type2 diabetes mellitus
using the late-night salivary cortisol test did not identify
any case of occult Cushing syndrome.28 Moreover, Mullan
etal.29 did not find any case of occult Cushing syndrome
among 201 consecutive patients with diabetes mellitus
who had at least two of the following criteria: HbA1c >7%,
overweight (BMI >25kg/m2) and hypertension.29 Other
studies with questionable methodologies have yielded
divergent inconclusive results.30,31
Several arguments may account for the discrepancies
between positive and negative series. Sample variability
is important to consider. If the true prevalence of occult
Cushing syndrome is around 1%, the probability of iden
tifying at least one case in a cohort of 200 patients is not
more than 87%. The characteristics of the investigated
population also influence the prevalence of Cushing syn
drome. Intuitively, the presence of this condition is more
probable in patients with central obesity, hypertension
and long-standing, uncontrolled diabetes mellitus24,26 than
in patients with mild impairment of glycemic control29 or
newly diagnosed diabetes mellitus.27 Screening exclusively
or predominantly men for a disease that occurs mainly in
women is also a shortcoming of some studies.27,28 With
drawal of patients from the study who had a positive
screening test but did not undergo confirmation tests
may also hamper the identification of patients with occult
Cushing syndrome.28,29 Obviously, the type and sensitivity
of the screening investigation is of paramount importance
and will be discussed later.
100%
(200 patients)
1 Screening step
1 mg DST
Impaired 1 mg DST
(cortisol >60 nmol/l)
26%
Drop-out: 5 patients
2 Confirmation step
ACTH, cortisol,
circadian rhythm, UFC
Midnight cortisol, 4 mg DST
24%
Normal endocrine
evaluation
30 patients
(15% false-positive rate)
Drop-out: 3 patients
3 Imaging step
8%
Adrenal tumor:
4.0% (8 patients)
Cushing disease:
1.5% (3 patients)
No visible tumor:
1.5% (3 patients)
Figure2 | Algorithm used to screen for occult Cushing syndrome in obese patients
with type2 diabetes mellitus. Abbreviations: ACTH, adrenocorticotropic hormone;
DST, dexamethasone suppression test; UFC, urinary free cortisol. Adapted with
permission from The Endocrine Society Catargi, B. etal. J. Clin. Endocrinol.
Metab. 88, 58085813 (2003).
Inclusion criteria
Suspected Cushing
syndrome (%)
Proven Cushing
syndrome* (%)
Leibowitz
etal.26
90
3.3
3.3
Catargi
etal.24
200
7.0
2.0
Chiodini
etal.33
294
>30years, in-patients
9.4
1.0
Liu etal.28
154
0.0
0.0
Reimondo
etal.27
99
Newly diagnosed DM
1.0
1.0
Taniguchi
etal.59
77
In-patients, T2DM
7.8
2.6
Mullan
etal.29
201
0.0
0.0
*Patients with definitive diagnosis of Cushing syndrome obtained after surgical excision of a pituitary
corticotroph adenoma or cortisol-secreting adrenal adenoma. Abbreviations: DM, diabetes mellitus;
n,number of patients screened; T2DM, type2 diabetes mellitus.
REVIEWS
Box 1 | Indication criteria for screening
Burden of target disease justifies action
Natural history of the disease is adequately understood
Preclinical or early symptomatic phase is long enough
Diagnostic performance of screening test is good
Accepted treatment or useful intervention for patients
with the disease is available
Early diagnosis improves clinical outcomes
Screening does more good than harm
Screening is acceptable by the screened population
and the health-care system
Based on Sackett etal.54 and WHO criteria for screening.55
Adrenal incidentalomas
This Review will not address the peculiar situation of
adrenal tumors incidentally discovered during abdomi
nal imaging. Depending on the referral processes, diag
nostic tests and criteria, biochemical hypercortisolism,
usually of mild intensity, may be identified in a variable
percentage (220%) of patients without clinical features
of Cushing syndrome.4752 Although many issues remain
unsolved concerning the definition and consequences
of this entity, several academic societies advocate the
indication of biochemical screening for occult Cushing
syndrome in this situation.10,51,53
REVIEWS
work load and costs. Given that type2 diabetes mellitus
has been the most extensively and rigorously studied
situation, we will examine the question of screening for
occult Cushing syndrome in patients with this disorder.
REVIEWS
and patients with overt Cushing syndrome.50,73 However,
divergent observations on the evolution of the metabolic
condition in the absence of specific intervention on the
adrenal tumors have been published.76,77 Consequently,
the long-term consequences of SCSAs are not precisely
known. The considerable heterogeneity across studies on
the biochemical definition of SCSAs may account for these
discrepancies.52 Be that as it may, whether or not SCSAs
markedly affects patients health is still a matter of debate.
Thus, to date, the possible effect of occult Cushing syn
drome on long-term morbidity in patients with diabetes
mellitus is not evidence-based.
REVIEWS
700
600
500
Cortisol (nmol/l)
400
300
200
100
0
10
11
12
HbA1c (%)
13
14
15
16
Drawbacks of screening
Altogether, the influence of pathological context, disease
spectrum and low prevalence of occult Cushing syndrome
REVIEWS
0.1
99
0.2
0.5
95
1,000
500
200
90
80
100
50
5
10
20
70
20
10
60
40
2
1
30
50
20
30
0.5
40
50
0.2
0.1
60
0.05
70
0.02
80
0.01
0.005
90
0.002
0.001
95
10
5
0.5
0.2
99
0.1
Pre-test
probability (%)
Likelihood
ratio
Post-test
probability (%)
REVIEWS
conclusions have also been reported.77 In addition, dia
betes mellitus, obesity, hypertension and dyslipidemia may
persist in a number of patients despite biochemical cure
of overt Cushing syndrome.23,103 A prospective, random
ized study that compares the benefits of surgery to that of
the currently available pharmacological tools for diabetes
mellitus is, therefore, necessary to conclude on the benefit
of specific treatment of hypercortisolism in patients with
occult and mild Cushing syndrome.
Acceptability of screening
Whether the screening process is acceptable to the
screened population and to the health-care system is a
matter of debate. The convenience of the 1mg DST or latenight salivary cortisol sampling that can be performed in
ambulatory patients94 is well-acknowledged and would
probably be accepted by patients with diabetes mellitus
who are familiar with repeated, multiple biochemical
investigations for their metabolic disease. This hypothesis
remains to be demonstrated for subsequent confirma
tory steps in patients with positive screening results and
who are not followed up in referral clinics or hospitals.
From this perspective, it should be noted that 1030%
of patients who initially gave informed consent during
the diabetes mellitus, obesity and osteoporosis screening
studies refused further endocrine evaluation after a posi
tive initial investigation.24,25,28,29,33,42 The acceptability of
the workload for the medical system is also questionable,
given the epidemic increase in the prevalence of type2
diabetes mellitus and that 525% of patients with diabetes
mellitus screen positive.24,25,2729
Last, but not least, the cost of a strategy with numer
ous uncertainties as mentioned above has to be assessed.
Epidemiologic data in the French population acknowl
edge a prevalence of type2 diabetes mellitus of around
1.5 million in individuals older than 50years. Thus, in
France, the cost of the first ambulatory screening proce
dure in patients with diabetes mellitus would be around
76 million. The cost of complementary evaluations
including biochemical and imaging studies in 300,000
individuals who had a positive screening test would be
286 million. If we assume that 2% of patients have occult
Cushing syndrome and will undergo surgery, the total
cost will be at least 445 million without including pos
sible adverse events of surgery. The cost of treating true
positive patients would be around 181 million but that
of endocrine and diagnostic imaging investigations for
false positives would amount to at least 264 million.
Such sums have to be contrasted with the number of
scientific uncertainties surrounding screening that we
have discussed above.
Conclusions
In conclusion, systematic screening studies in selected
populations deserve credit for showing that the true
prevalence of Cushing syndrome might be higher than
previously thought. However, occult Cushing syndrome
may potentially be mild in most patients, which raises
important issues with regard to the difficulty to identify
and confirm Cushing syndrome. In addition, we lack the
Cons
Performance of
diagnostic tests
Outcomes and beneficial
impact of treatment?
Acceptability of screening
by patients and health
care system?
Severity?
Pros
Prevalence from
screening studies
REVIEWS
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
www.nature.com/nrendo
2011 Macmillan Publishers Limited. All rights reserved
REVIEWS
55. Wilson, J. & Jungner, G. Principles and practice
of screening for disease. Public Health Papers
No. 34. (WHO, Geneva, 1968).
56. Adriaanse, M.C. & Snoek, F.J. The psychological
impact of screening for type 2 diabetes.
Diabetes Metab. Res. Rev. 22, 2025 (2006).
57. Barton, M.B. et al. Decreasing womens
anxieties after abnormal mammograms:
acontrolled trial. J. Natl Cancer Inst. 96,
529538 (2004).
58. Brett, J., Austoker, J. & Ong, G. Do women who
undergo further investigation for breast
screening suffer adverse psychological
consequences? A multi-centre follow-up study
comparing different breast creening result
groups five months after their last breast
screening appointment. J. Public Health Med. 20,
396403 (1998).
59. Taniguchi, T., Hamasaki, A. & Okamoto, M.
Subclinical hypercortisolism in hospitalized
patients with type 2 diabetes mellitus. Endocr. J.
55, 429432 (2008).
60. Gao, S. Combining binomial data using the
logistic normal model. J. Stat. Comput. Simul. 74,
293306 (2004).
61. Clayton, R.N. Mortality in Cushings disease.
Neuroendocrinology 92 (Suppl. 1), 7176
(2010).
62. Hammer, G.D. et al. Transsphenoidal
microsurgery for Cushings disease: initial
outcome and long-term results. J. Clin.
Endocrinol. Metab. 89, 63486357 (2004).
63. Storr, H.L., Chan, L.F., Grossman, A.B. &
Savage, M.O. Paediatric Cushings syndrome:
epidemiology, investigation and therapeutic
advances. Trends Endocrinol. Metab. 18,
167174 (2007).
64. Pecori Giraldi, F., Moro, M. & Cavagnini, F.
Gender-related differences in the presentation
and course of Cushings disease. J. Clin.
Endocrinol. Metab. 88, 15541558 (2003).
65. Wajngot, A., Giacca, A., Grill, V., Vranic, M. &
Efendic, S. The diabetogenic effects of
glucocorticoids are more pronounced in lowthan in high-insulin responders. Proc. Natl Acad.
Sci. USA 89, 60356039 (1992).
66. van Rossum, E.F. & Lamberts, S.W.
Polymorphisms in the glucocorticoid receptor
gene and their associations with metabolic
parameters and body composition. Recent Prog.
Horm. Res. 59, 333357 (2004).
67. Devenport, L., Knehans, A., Sundstrom, A. &
Thomas, T. Corticosterones dual metabolic
actions. Life Sci. 45, 13891396 (1989).
68. Sonino, N., Boscaro, M., Fallo, F. & Fava, G.A.
Aclinical index for rating severity in Cushings
syndrome. Psychother. Psychosom. 69, 216220
(2000).
69. Sarlis, N.J., Chanock, S.J. & Nieman, L.K.
Cortisolemic indices predict severe infections in
Cushings syndrome due to ectopic production of
adrenocorticotropin. J. Clin. Endocrinol. Metab.
85, 4247 (2000).
70. Sonino, N., Fava, G.A., Raffi, A.R., Boscaro, M.
& Fallo, F. Clinical correlates of major depression
in Cushings disease. Psychopathology 31,
302306 (1998).
71. Strack, A.M., Horsley, C.J., Sebastian, R.J.,
Akana, S.F. & Dallman, M.F. Glucocorticoids
and insulin: complex interaction on brown
adipose tissue. Am. J. Physiol. 268,
R1209R1216 (1995).
72. Webb, S.M. et al. Evaluation of health-related
quality of life in patients with Cushings
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.