Subclinical Cushing S y n drome : A Review Lee F. Starker,
MD, PhD
a
, John W. Kunstman,
MD
a
, Tobias Carling,
MD, PhD
a,b,
*
KEYWORDS Subclinical Cushing syndrome Adrenal incidentalomas Dexamethasone-suppressive testing Corticotropin KEY POINTS Subclinical Cushing syndrome (SCS) may not be as clinically insignificant as previously postulated. Diagnosis of the syndrome is somewhat difficult, but possible. As most patients present with adrenal incidentalomas, a thorough biochemical workup by both a dedicated endocrinologist and endocrine surgeon is necessary for optimal outcome. Given the low rate of complications, minimally invasive adrenalectomy is recommended for patients with biochemically proven or suspected SCS who are appropriate surgical candidates.
INTRODUCTION
Subclinical Cushing syndrome (SCS) has recently become a topic of controversy and interest. The reasons for this are multifactorial. Adrenal masses known as incidentalomas, identified by radiographic imaging obtained for unrelated reasons, are detected more frequently, especially within our aging population; this is due to a combination of more refined diagnostic imaging technologies with higher resolution, and because cross-sectional imaging is much more widely used. This increase in the number of patients with adrenal masses undergoing biochemical evaluation has led to identification of patients with biochemical abnormalities without florid clinical symptoms. A unique situation has thus arisen for clinical intervention, as those patients without florid Cushing syndrome can be spared the development of significant physiologic derangements by early intervention and surgical resection. However, only a portion of patients with SCS progress to Cushing syndrome.
a Department of Surgery, Yale University School of Medicine, 333 Cedar Street, New Haven, CT, USA; b Yale Endocrine Neoplasia Laboratory, Yale University School of Medicine, 333 Cedar Street, New Haven, CT, USA * Corresponding author. Section of Endocrine Surgery, Department of Surgery, Yale University School of Medicine, 333 Cedar Street, FMB130, Box 208062, New Haven, CT 06520. E-mail address: [email protected]
Surg Clin N Am - (2014) -–http://dx.doi.org/10.1016/j.suc.2014.02.008 surgical.theclinics.com 0039-6109/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
2
Starker et al
Classic Cushing syndrome is characterized by the signs of hypercortisolism: moon face, buffalo hump, central obesity, easy bruising, proximal muscle wasting, deep purple striae, acne, hirsutism, and glucose intolerance. The syndrome is rare with an overall yearly incidence of 1 in 50,000.1 Primary adrenal lesions only account for roughly 15% of the cases of Cushing syndrome. About 50% of these lesions are adenomas, with the remaining 50% stemming from functional adrenal cortical carcinomas. Nevertheless, most studies report a prevalence of 5% to 24% for SCS in patients with an adrenal incidentaloma (AI). This broad range can likely be attributed to the differing results in the diagnostic criteria used over time.2–7 The concept of SCS was first identified and described in 1973 by Beierwaltes and colleagues8 and then again in 1981 by Charbonnel and colleagues.9 SCS was defined as an increase in the overall endogenous secretion of glucocorticoids without clinical manifestations of clinical Cushing syndrome. The overwhelming consensus is that the diagnosis can be made when there is a combination of 2 distinct findings: no overt clinical signs of Cushing syndrome and at least 2 distinct alterations in the hypothalamic-pituitary-adrenal (HPA) axis. Nonetheless, patients with SCS usually have an increased prevalence of obesity, hypertension, and type 2 diabetes. It is imperative to assess all patients with an AI for biochemical production before surgical resection, as the unopposed overproduction of glucocorticoids in this case can suppress the contralateral adrenal gland and lead to postoperative Addisonian crisis, which can be fatal.10 DEFINITION
SCS presents in patients with a clinically nonfunctioning adrenal adenoma, likely identified for an unrelated reason on diagnostic imaging. Before the term SCS took hold, the term preclinical Cushing syndrome was used and is still used interchangeably. However, this terminology is misleading because the overwhelming majority of patients with SCS will not progress to clinically apparent Cushing syndrome, as the term implies. Therefore, the terminology of SCS has become the preferred and more accurate modality of description.1 To facilitate the definition of the disease, a National Institutes of Health (NIH) State of Science Conference was convened and a better, more easily identifiable term was chosen: subclinical autonomous glucocorticoid hypersecretion.11 The final consensus was that to concretely make the diagnosis, 2 distinct criteria must be fulfilled. First, the patient must not present with a clear Cushing syndrome phenotype, and none of the physical stigmata of the disease must be present. It is imperative that an experienced clinician examine the patient to be absolutely certain that there is no evidence of classic Cushing syndrome.7 Second, the patient must have presented with an adrenal mass that was identified in an incidental fashion, meaning that the radiographic imaging was obtained for another reason and this lesion was subsequently identified. The likely culprit of this cortisol hypersecretion is commonly a cortical adenoma, although adrenocortical carcinoma (ACC) must not be overlooked12,13; however, given the natural course of ACC one must question whether the terminology of SCS can truly apply.14 DIAGNOSIS Rationale
Diagnosis of hypercortisolism may be laborious, even in cases of overt disease. Because of the subtle nature of the clinical manifestations of SCS as already described, biochemical tests are crucial in establishing a diagnosis. Several additional
Subclinical Cushing Syndrome
factors increase the difficulty of biochemical testing in cases of SCS. When assessed on a population basis, cortisol secretion and activation of the HPA axis runs in a continuum from unmistakably hypoactive (ie, adrenal insufficiency) to normal to hyperactive. As a corollary, assigning arbitrary cutoff values for the numerous biochemical tests assessing cortisol secretion and the HPA axis risks neglecting disease when such cutoffs are set too stringently, and overtreatment when too lenient. This conundrum is especially true in patients with AI. Moreover, the circadian nature of cortisol secretion, which persists even in disease states, dampens the reproducibility of certain biochemical test results and increases the effort required by provider, patient, and laboratory in obtaining accurate data. This situation has a particularly notable effect in cases of SCS, where potentially small elevations in cortisol levels and the resulting perturbation of the HPA axis may fall below a given test’s threshold for detection. Furthermore, the myriad of tests available makes efficient and accurate diagnosis challenging even for experienced clinicians. A summary of described diagnostic criteria for SCS is given in Table 1, and individual tests are described here. Dexamethasone-Suppression Testing
Dexamethasone-suppression testing (DST) is the test most frequently used to evaluate mild elevations in cortisol secretion, and therefore plays a key role in diagnosis of SCS. Dexamethasone is an endogenous steroid, used because it does not Table 1 Proposed criteria for diagnosis of subclinical hypercortisolism (SCS)
Authors,Ref. Year
Criteria
Dexamethasone Dose, DST Cutoff
SCS Prevalence (%)
Reincke et al,60 1992
DST alone
1 mg, 3 mg/dL
12
Osella et al,29 1994
DST alone
1 mg, 5 mg/dL
16
Ambrosi et al,28 1995
DST plus 1 of CRH, CCR, ACTH, UFC
1 mg, 5 mg/dL
12
Tsagarakis et al,31 1998
Low-dose DST alone
2.5 mg/dL
25
Terzolo et al,33 1998
DST plus UFC
1 mg, 5 mg/dL
6
Mantero & Arnaldi,4 2000
2 of CRH, CCR, ACTH, UFC, DST
1 mg, 5 mg/dL
9.2
Rossi et al,47 2000
Low-dose DST plus 1 of CRH, CCR, ACTH, UFC
3.0 mg/dL
18.5
Valli et al,17 2001
Unilateral uptake on 131 I-norcholesterol scintigraphy
N/A
61.3
Emral et al,25 2003
DST and high-dose DST
3 mg, 3 mg/dL
5.7
Chiodini et al,20 2009
2 of ACTH, UFC, DST
1 mg, 3 mg/dL
29.6
Masserini et al,21 2009
2 of ACTH, UFC, DST
1 mg, 3 mg/dL
21.4
Eller-Vainicher et al,15 2010
3 of CCR, ACTH, UFC, DST
1 mg, 3 mg/dL
48.3
Di Dalmazi et al,66 2012
DST (5 mg/dL) or DST (1.8 mg/dL) plus UFC or ACTH
1 mg, 1.8 mg/dL or 5 mg/dL
21.3
Abbreviations: ACTH, adrenocorticotropic hormone; CCR, alteration in circadian cortisol rhythm (increased midnight serum or salivary cortisol level); CRH, blunted response to corticotropinreleasing hormone; DST, dexamethasone-suppression test; N/A, no data available; SCS, subclinical Cushing syndrome; UFC, elevated 24-hour urinary free cortisol.
3
4
Starker et al
cross-react with either immunologic or chemical assays for cortisol. In individuals with corticotropin (adrenocorticotropic hormone [ACTH])-dependent hypercortisolism (either overt or subclinical), dexamethasone administration suppresses the HPA axis and results in suppression of cortisol secretion. For SCS diagnosis, 1 mg overnight DST is the most commonly performed, owing to its simplicity of administration.4,15–17 In consensus statements, the NIH and the American Association of Clinical Endocrinologists/American Association of Endocrine Surgeons recommend the standard cutoff of 5 mg/dL as the upper limit of normal for serum cortisol following DST in patients with AI.11,18 This cutoff results in very few false-positive outcomes, as specificity can exceed 90%.12,15 However, most normal individuals have cortisol levels suppressed to below 1 mg/dL after overnight DST, and sensitivity for detection of SCS is poor at the 5-mg/dL cutoff. As such, some European groups have recommended that a lower cutoff value for DST be used when assessing for SCS.6 By lowering the serum cortisol level to a lower cutoff value to ascribe a positive DST result, sensitivity increases dramatically, from 20% to 30% at 5 mg/dL to 70% to 90% at 1.8 mg/dL (Table 2); this also results in an inverse decline in specificity. Accordingly a compromise cutoff of 3 mg/dL has been widely assessed and clinically validated, which attempts to minimize false-positive results while maintaining a modestly increased sensitivity in comparison with higher cutoff values.7,15,19–22 Several additional methods of DST have been evaluated for SCS diagnosis. The 2-day, 2-mg DST (sometimes referred to as low-dose DST) appears, for unclear reasons to be more accurate in patients with alcoholism, diabetes, or psychiatric disorders, but for most patients has no advantage.23,24 Higher-dose DST, using 3-mg or 8-mg dexamethasone doses, have been proposed to further increase specificity, but this has not been demonstrated in comparison testing and these doses do not appear to have a role in SCS diagnosis.24–26 Corticotropin (Adrenocorticotropic Hormone)
Hypercortisolism can be regarded as ACTH-independent when serum ACTH levels are suppressed to lower than 10 pg/mL, as in cases of most adrenal tumors causing Table 2 Sensitivity and specificity of 1 mg overnight dexamethasone-suppression test at diagnosing subclinical hypercortisolism using various cutoff values for morning serum cortisol measurement
Authors,Ref. Year
5 mg/dL Cutoff (Se/Sp)
3 mg/dL Cutoff (Se/Sp)
1.8 mg/dL Cutoff (Se/Sp)
No. of Patients
Criteria for SCS Diagnosis
Barzon et al,67 2001
44/100
—
75/72
83
Scintigraphy
Valli et al,17 2001
58/83
63/75
100/67a
31
Scintigraphy
Eller-Vainicher et al,68 2010
33.3/85.7
59/52.4
79.5/23.8
60
Postsurgical hypocortisolism
Morelli et al,45 2010
23.8/93.3
52.4/81.4
71.4/49.5
231
Clinical manifestationsb
Eller-Vainicher et al,15 2010
21.7/96.9
—
91.3/56.3c
55
Postresection improvementd
Abbreviations: SCS, subclinical Cushing syndrome; Se, sensitivity (%); Sp, specificity (%). a Study utilized 2.2 mg/dL cutoff. b Included presence of hypertension, type 2 diabetes mellitus, or vertebral fractures. c Study utilized 2.0 mg/dL cutoff. d Improvement in at least 2 of: cholesterol level, body weight, hypertension, serum glucose.
Subclinical Cushing Syndrome
overt hypercortisolism. Low ACTH levels are a frequent finding in AI, being present in more than 50% of cases in some studies.4,15,21,27 As such, the specificity of depressed ACTH levels alone in the context of diagnosing SCS is as low as 38% to 60% in some studies. Furthermore, as ACTH levels can be normal even in ACTHindependent hypercortisolism, ACTH is not regarded as a useful primary tool in SCS diagnosis without additional testing.26 Similarly, addition of CRH (corticotropinreleasing hormone) stimulation before ACTH testing does not increase sensitivity to screening for SCS.26,28,29 In patients with known SCS, ACTH testing can help to rule out occult ACTH-dependent hypercortisolism as the cause rather than AI. This scenario should be suspected in cases of AI where ACTH is not suppressed, and is especially useful in cases of bilateral or diffuse adrenal enlargement, either of which can occur in ACTH excess. To exclude the presence of pituitary or ectopic ACTHdependent hypercortisolism, ACTH should be less than 10 pg/mL. In uncertain cases (ie, ACTH >10 pg/mL but
MD, PhD
a
, John W. Kunstman,
MD
a
, Tobias Carling,
MD, PhD
a,b,
*
KEYWORDS Subclinical Cushing syndrome Adrenal incidentalomas Dexamethasone-suppressive testing Corticotropin KEY POINTS Subclinical Cushing syndrome (SCS) may not be as clinically insignificant as previously postulated. Diagnosis of the syndrome is somewhat difficult, but possible. As most patients present with adrenal incidentalomas, a thorough biochemical workup by both a dedicated endocrinologist and endocrine surgeon is necessary for optimal outcome. Given the low rate of complications, minimally invasive adrenalectomy is recommended for patients with biochemically proven or suspected SCS who are appropriate surgical candidates.
INTRODUCTION
Subclinical Cushing syndrome (SCS) has recently become a topic of controversy and interest. The reasons for this are multifactorial. Adrenal masses known as incidentalomas, identified by radiographic imaging obtained for unrelated reasons, are detected more frequently, especially within our aging population; this is due to a combination of more refined diagnostic imaging technologies with higher resolution, and because cross-sectional imaging is much more widely used. This increase in the number of patients with adrenal masses undergoing biochemical evaluation has led to identification of patients with biochemical abnormalities without florid clinical symptoms. A unique situation has thus arisen for clinical intervention, as those patients without florid Cushing syndrome can be spared the development of significant physiologic derangements by early intervention and surgical resection. However, only a portion of patients with SCS progress to Cushing syndrome.
a Department of Surgery, Yale University School of Medicine, 333 Cedar Street, New Haven, CT, USA; b Yale Endocrine Neoplasia Laboratory, Yale University School of Medicine, 333 Cedar Street, New Haven, CT, USA * Corresponding author. Section of Endocrine Surgery, Department of Surgery, Yale University School of Medicine, 333 Cedar Street, FMB130, Box 208062, New Haven, CT 06520. E-mail address: [email protected]
Surg Clin N Am - (2014) -–http://dx.doi.org/10.1016/j.suc.2014.02.008 surgical.theclinics.com 0039-6109/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
2
Starker et al
Classic Cushing syndrome is characterized by the signs of hypercortisolism: moon face, buffalo hump, central obesity, easy bruising, proximal muscle wasting, deep purple striae, acne, hirsutism, and glucose intolerance. The syndrome is rare with an overall yearly incidence of 1 in 50,000.1 Primary adrenal lesions only account for roughly 15% of the cases of Cushing syndrome. About 50% of these lesions are adenomas, with the remaining 50% stemming from functional adrenal cortical carcinomas. Nevertheless, most studies report a prevalence of 5% to 24% for SCS in patients with an adrenal incidentaloma (AI). This broad range can likely be attributed to the differing results in the diagnostic criteria used over time.2–7 The concept of SCS was first identified and described in 1973 by Beierwaltes and colleagues8 and then again in 1981 by Charbonnel and colleagues.9 SCS was defined as an increase in the overall endogenous secretion of glucocorticoids without clinical manifestations of clinical Cushing syndrome. The overwhelming consensus is that the diagnosis can be made when there is a combination of 2 distinct findings: no overt clinical signs of Cushing syndrome and at least 2 distinct alterations in the hypothalamic-pituitary-adrenal (HPA) axis. Nonetheless, patients with SCS usually have an increased prevalence of obesity, hypertension, and type 2 diabetes. It is imperative to assess all patients with an AI for biochemical production before surgical resection, as the unopposed overproduction of glucocorticoids in this case can suppress the contralateral adrenal gland and lead to postoperative Addisonian crisis, which can be fatal.10 DEFINITION
SCS presents in patients with a clinically nonfunctioning adrenal adenoma, likely identified for an unrelated reason on diagnostic imaging. Before the term SCS took hold, the term preclinical Cushing syndrome was used and is still used interchangeably. However, this terminology is misleading because the overwhelming majority of patients with SCS will not progress to clinically apparent Cushing syndrome, as the term implies. Therefore, the terminology of SCS has become the preferred and more accurate modality of description.1 To facilitate the definition of the disease, a National Institutes of Health (NIH) State of Science Conference was convened and a better, more easily identifiable term was chosen: subclinical autonomous glucocorticoid hypersecretion.11 The final consensus was that to concretely make the diagnosis, 2 distinct criteria must be fulfilled. First, the patient must not present with a clear Cushing syndrome phenotype, and none of the physical stigmata of the disease must be present. It is imperative that an experienced clinician examine the patient to be absolutely certain that there is no evidence of classic Cushing syndrome.7 Second, the patient must have presented with an adrenal mass that was identified in an incidental fashion, meaning that the radiographic imaging was obtained for another reason and this lesion was subsequently identified. The likely culprit of this cortisol hypersecretion is commonly a cortical adenoma, although adrenocortical carcinoma (ACC) must not be overlooked12,13; however, given the natural course of ACC one must question whether the terminology of SCS can truly apply.14 DIAGNOSIS Rationale
Diagnosis of hypercortisolism may be laborious, even in cases of overt disease. Because of the subtle nature of the clinical manifestations of SCS as already described, biochemical tests are crucial in establishing a diagnosis. Several additional
Subclinical Cushing Syndrome
factors increase the difficulty of biochemical testing in cases of SCS. When assessed on a population basis, cortisol secretion and activation of the HPA axis runs in a continuum from unmistakably hypoactive (ie, adrenal insufficiency) to normal to hyperactive. As a corollary, assigning arbitrary cutoff values for the numerous biochemical tests assessing cortisol secretion and the HPA axis risks neglecting disease when such cutoffs are set too stringently, and overtreatment when too lenient. This conundrum is especially true in patients with AI. Moreover, the circadian nature of cortisol secretion, which persists even in disease states, dampens the reproducibility of certain biochemical test results and increases the effort required by provider, patient, and laboratory in obtaining accurate data. This situation has a particularly notable effect in cases of SCS, where potentially small elevations in cortisol levels and the resulting perturbation of the HPA axis may fall below a given test’s threshold for detection. Furthermore, the myriad of tests available makes efficient and accurate diagnosis challenging even for experienced clinicians. A summary of described diagnostic criteria for SCS is given in Table 1, and individual tests are described here. Dexamethasone-Suppression Testing
Dexamethasone-suppression testing (DST) is the test most frequently used to evaluate mild elevations in cortisol secretion, and therefore plays a key role in diagnosis of SCS. Dexamethasone is an endogenous steroid, used because it does not Table 1 Proposed criteria for diagnosis of subclinical hypercortisolism (SCS)
Authors,Ref. Year
Criteria
Dexamethasone Dose, DST Cutoff
SCS Prevalence (%)
Reincke et al,60 1992
DST alone
1 mg, 3 mg/dL
12
Osella et al,29 1994
DST alone
1 mg, 5 mg/dL
16
Ambrosi et al,28 1995
DST plus 1 of CRH, CCR, ACTH, UFC
1 mg, 5 mg/dL
12
Tsagarakis et al,31 1998
Low-dose DST alone
2.5 mg/dL
25
Terzolo et al,33 1998
DST plus UFC
1 mg, 5 mg/dL
6
Mantero & Arnaldi,4 2000
2 of CRH, CCR, ACTH, UFC, DST
1 mg, 5 mg/dL
9.2
Rossi et al,47 2000
Low-dose DST plus 1 of CRH, CCR, ACTH, UFC
3.0 mg/dL
18.5
Valli et al,17 2001
Unilateral uptake on 131 I-norcholesterol scintigraphy
N/A
61.3
Emral et al,25 2003
DST and high-dose DST
3 mg, 3 mg/dL
5.7
Chiodini et al,20 2009
2 of ACTH, UFC, DST
1 mg, 3 mg/dL
29.6
Masserini et al,21 2009
2 of ACTH, UFC, DST
1 mg, 3 mg/dL
21.4
Eller-Vainicher et al,15 2010
3 of CCR, ACTH, UFC, DST
1 mg, 3 mg/dL
48.3
Di Dalmazi et al,66 2012
DST (5 mg/dL) or DST (1.8 mg/dL) plus UFC or ACTH
1 mg, 1.8 mg/dL or 5 mg/dL
21.3
Abbreviations: ACTH, adrenocorticotropic hormone; CCR, alteration in circadian cortisol rhythm (increased midnight serum or salivary cortisol level); CRH, blunted response to corticotropinreleasing hormone; DST, dexamethasone-suppression test; N/A, no data available; SCS, subclinical Cushing syndrome; UFC, elevated 24-hour urinary free cortisol.
3
4
Starker et al
cross-react with either immunologic or chemical assays for cortisol. In individuals with corticotropin (adrenocorticotropic hormone [ACTH])-dependent hypercortisolism (either overt or subclinical), dexamethasone administration suppresses the HPA axis and results in suppression of cortisol secretion. For SCS diagnosis, 1 mg overnight DST is the most commonly performed, owing to its simplicity of administration.4,15–17 In consensus statements, the NIH and the American Association of Clinical Endocrinologists/American Association of Endocrine Surgeons recommend the standard cutoff of 5 mg/dL as the upper limit of normal for serum cortisol following DST in patients with AI.11,18 This cutoff results in very few false-positive outcomes, as specificity can exceed 90%.12,15 However, most normal individuals have cortisol levels suppressed to below 1 mg/dL after overnight DST, and sensitivity for detection of SCS is poor at the 5-mg/dL cutoff. As such, some European groups have recommended that a lower cutoff value for DST be used when assessing for SCS.6 By lowering the serum cortisol level to a lower cutoff value to ascribe a positive DST result, sensitivity increases dramatically, from 20% to 30% at 5 mg/dL to 70% to 90% at 1.8 mg/dL (Table 2); this also results in an inverse decline in specificity. Accordingly a compromise cutoff of 3 mg/dL has been widely assessed and clinically validated, which attempts to minimize false-positive results while maintaining a modestly increased sensitivity in comparison with higher cutoff values.7,15,19–22 Several additional methods of DST have been evaluated for SCS diagnosis. The 2-day, 2-mg DST (sometimes referred to as low-dose DST) appears, for unclear reasons to be more accurate in patients with alcoholism, diabetes, or psychiatric disorders, but for most patients has no advantage.23,24 Higher-dose DST, using 3-mg or 8-mg dexamethasone doses, have been proposed to further increase specificity, but this has not been demonstrated in comparison testing and these doses do not appear to have a role in SCS diagnosis.24–26 Corticotropin (Adrenocorticotropic Hormone)
Hypercortisolism can be regarded as ACTH-independent when serum ACTH levels are suppressed to lower than 10 pg/mL, as in cases of most adrenal tumors causing Table 2 Sensitivity and specificity of 1 mg overnight dexamethasone-suppression test at diagnosing subclinical hypercortisolism using various cutoff values for morning serum cortisol measurement
Authors,Ref. Year
5 mg/dL Cutoff (Se/Sp)
3 mg/dL Cutoff (Se/Sp)
1.8 mg/dL Cutoff (Se/Sp)
No. of Patients
Criteria for SCS Diagnosis
Barzon et al,67 2001
44/100
—
75/72
83
Scintigraphy
Valli et al,17 2001
58/83
63/75
100/67a
31
Scintigraphy
Eller-Vainicher et al,68 2010
33.3/85.7
59/52.4
79.5/23.8
60
Postsurgical hypocortisolism
Morelli et al,45 2010
23.8/93.3
52.4/81.4
71.4/49.5
231
Clinical manifestationsb
Eller-Vainicher et al,15 2010
21.7/96.9
—
91.3/56.3c
55
Postresection improvementd
Abbreviations: SCS, subclinical Cushing syndrome; Se, sensitivity (%); Sp, specificity (%). a Study utilized 2.2 mg/dL cutoff. b Included presence of hypertension, type 2 diabetes mellitus, or vertebral fractures. c Study utilized 2.0 mg/dL cutoff. d Improvement in at least 2 of: cholesterol level, body weight, hypertension, serum glucose.
Subclinical Cushing Syndrome
overt hypercortisolism. Low ACTH levels are a frequent finding in AI, being present in more than 50% of cases in some studies.4,15,21,27 As such, the specificity of depressed ACTH levels alone in the context of diagnosing SCS is as low as 38% to 60% in some studies. Furthermore, as ACTH levels can be normal even in ACTHindependent hypercortisolism, ACTH is not regarded as a useful primary tool in SCS diagnosis without additional testing.26 Similarly, addition of CRH (corticotropinreleasing hormone) stimulation before ACTH testing does not increase sensitivity to screening for SCS.26,28,29 In patients with known SCS, ACTH testing can help to rule out occult ACTH-dependent hypercortisolism as the cause rather than AI. This scenario should be suspected in cases of AI where ACTH is not suppressed, and is especially useful in cases of bilateral or diffuse adrenal enlargement, either of which can occur in ACTH excess. To exclude the presence of pituitary or ectopic ACTHdependent hypercortisolism, ACTH should be less than 10 pg/mL. In uncertain cases (ie, ACTH >10 pg/mL but
