Diagnosing%20Cushing

1
Diagnosing Cushings Syndrome Not as Easy as it
Seems

• Theodore C. Friedman, M.D., Ph.D.
• Professor of Medicine-Charles Drew University
• Professor of Medicine-UCLA
• Magic Foundation
• Symposium on Cushings Syndrome
• February 22, 2009
• Las Vegas, NV

2
States of Glucocorticoid Excess

• ACTH-dependent States
• a. Pituitary Adenoma (Cushings Disease) 90-95
• b. Ectopic ACTH Syndrome
• ACTH-independent States
• a. Adrenal adenoma
• b. Adrenal carcinoma
• Exogenous Sources
• Glucocorticoid intake
• Psychiatric Conditions (Pseudo-Cushing Disorders)
• a. Depression
• b. Alcoholism
• Pregnancy

3
Pseudo-Cushing States

• High Cortisol Secretion Rate without Convincing
  Clinical Features of Cushing Syndrome

4
Eucortisolemic Cushing Syndrome

• Clinical Manifestations of Cushing Syndrome
  without evidence of increased cortisol levels
• Exogenous glucocorticoid administration
• Episodic (periodic) Cushing syndrome-common
• Recently cured Cushing syndrome

5
Need to Distinguish Early or Mild Cushings from
Other Diseases

• Cushings is considered rare, but may not be that
  rare.
• It is vastly under diagnosed.
• Other diseases that have some symptoms/signs in
  common with Cushings (PCOS or Metabolic
  Syndrome) are more common, but present
  differently from Cushings. The treatment is
  different for these other diseases
• Thus, a strategy needs to be developed to
  diagnose Cushings syndrome.

6
Is Cushings Syndrome Rare

• Probably under-diagnosed
• Catargi et al. JCEM 2003, 885808-200 consecutive
  overweight patients with type 2 diabetes, but no
  other stigmata of hypercortisolism. 4 (2)
  patietns were found to have Cushings syndrome
  and another 7 are being evaluated.
• Kadioglu et al. Endo Society 2004 86 P2-455- 100
  consecutive obese patients. Cushings syndrome
  was diagnosed in 11.
• Nishikawa et al. Endo Society 2004 86 P3-437-
  1020 patients with hypertension. 11 had Cushings
  syndrome and 10 had subclinical Cushings
  syndrome (2).
• These studies may have missed mild Cushings
  syndrome and may actually be low.
• Maybe Cushings syndrome is not so rare

7
Do all diseases progress from mild to severe?
Rapid onset
Linear
Delayed onset
8
Should Cushings be Diagnosed Early?

• Cushings Patients are miserable.
• Effective treatment (surgery) exists
• Lack of medicine for it, less pharmaceutical
  funding.
• Most doctors are not familiar with Cushings
  syndrome and may only be familiar with severe
  cases.

9
How to Diagnose Cushings Syndrome

• Careful history and physical
• Change in weight and body habitus
• Look at old pictures
• Not all patients have all signs and symptoms,
  especially early and periodic patients.
• Most published data compared severe Cushings
  with normals.
• Important to diagnose early before devastating
  sequelae develop.
• Initial diagnosis most difficult aspect of
  Cushings syndrome.
• Gestalt with as much information as possible
• Periodic Cushings common, so one positive test
  may be worth more than 10 negative tests
• Make the diagnosis before proceeding to the
  differential diagnosis??

10
IMPORTANT SYMPTOMS

• Wired at night
• Trouble sleeping-trouble falling asleep or
  frequent awakenings
• Severe fatigue-new onset
• Abrupt weight gain-without other cause such as
  decreased activity or depression
• Decreased ability to exercise
• Menstrual abnormalities
• Cognitive changes- brain fog
• Decreased Libido
• Symptoms of adrenal insufficiency-joint pains,
  cant get out of bed, nausea and vomiting
• Depression, anxiety, mood-swings

11
IMPORTANT SIGNS

• Central obesity
• Muscle atrophy
• Thin skin
• Buffalo hump
• Round, red face
• Bruising
• Extra hair growth
• Acne
• Loss of hair on head
• Stretch marks

12
Signs/Symptoms

• Most patients dont have all these signs/symptoms
• Many doctors may have only seen 1 case of
  Cushings and textbooks may show only severe
  cases.

13
The Diagnosis of Cushings Syndrome An
EndocrineSociety Clinical Practice GuidelineJ
Clin Endocrinol Metab. May 2008, 93(5)15261540

• Lynnette K. Nieman
• Beverly M. K. Biller
• James W. Findling
• John Newell-Price
• Martin O. Savage
• Paul M. Stewart
• Victor M. Montori

14
The Diagnosis of Cushings Syndrome An
EndocrineSociety Clinical Practice GuidelineJ
Clin Endocrinol Metab. May 2008, 93(5)15261540

• 1st line recommended tests
• UFC
• Low dose or overnight dexamethasone test
• Night-time salivary cortisols
• Testing for Cushings syndrome in patients with
  multiple and progressive features compatible with
  the syndrome
• Patients with an abnormal result see an
  endocrinologist and undergo a second test, either
  one of the above or, in some cases, a serum
  midnight cortisol or dexamethasone-CRH test.

15
The Diagnosis of Cushings Syndrome An
EndocrineSociety Clinical Practice GuidelineJ
Clin Endocrinol Metab. May 2008, 93(5)15261540

• Patients with 2 or more normal results should not
  undergo further evaluation.
• Recommend additional testing in patients with
  discordant results, normal responses suspected of
  cyclic hypercortisolism, or initially normal
  responses who accumulate additional features over
  time.

16
The Diagnosis of Cushings Syndrome An
EndocrineSociety Clinical Practice GuidelineJ
Clin Endocrinol Metab. May 2008, 93(5)15261540

• We recommend against any further testing for
  Cushing's syndrome in individuals with
  concordantly negative results on two different
  tests (except in patients suspected of having the
  very rare case of cyclical disease)
• Rarely patients have been described with episodic
  secretion of cortisol excess in a cyclical
  pattern with peaks occurring at intervals of
  several days to many months. Because the DST
  results may be normal in patients who are cycling
  out of hypercortisolism, these tests are not
  recommended for patients suspected of having
  cyclic disease. Instead, measurement of UFC or
  salivary cortisol may best demonstrate cyclicity.
  In patients for whom clinical suspicion is high
  but initial tests are normal, follow-up is
  recommended with repeat testing, if possible to
  coincide with clinical symptoms.

17
FIG. 1. Algorithm for testing patients suspected
of having Cushing's syndrome (CS)
18
Hypothesis

• Patients with full-blown Cushings syndrome
  started out with mild Cushings syndrome.
• It would be advantageous to diagnose these
  patients when they have mild disease before they
  are affected by hypercortisolemia.
• There are many case reports of patients with
  periodic Cushings syndrome.
• Some of these patients have hypercortisolism at
  regular intervals as documented by symptoms and
  laboratory measurements.
• Many patients report highs and lowseven if
  not regular.
• There has been no series examining the frequency
  of mild or periodic/episodic Cushings syndrome.
• Thus, we hypothesized that a high percentage of
  consecutive patients presenting with signs and
  symptoms of hypercortisolism have episodic and/or
  mild Cushings syndrome.

19
Episodic, Cyclical, Periodic

• Periodic and cyclical refer to changes in
  cortisol levels that occur on a regular
  predictable basis.
• Episodic refers to high cortisol levels that are
  random.
• Most of my patients are episodic.

20
WEB AGE

• MOST FOUND ME FROM THE INTERNET
• Cushings-help.com (I hosted several chats
  including Jan 2009)
• Most went to numerous other Endocrinologist,
  including Cushings specialists
• Told Your arms arent thin enough for Cushings
  or were dismissed with 1 normal test
• In most cases, patient suspected Cushings, in
  spite of doctor telling them its unlikely

21
Confirmed Cushings Patients

• 66 patients
• 61 females, 5 males
• 62 Caucasians, 2 Hispanic, 1 Black, 1 Pacific
  Islander
• Median age 38.5 years
• BMI was 35.9 8.5 kg/m2 (mean SD)
• Average weight gain was 67.7 40.2 pounds
• Patients were considered for Cushings syndrome
  if they had a rapid, unexplained weight gain and
  associated symptoms of hypercortisolism including
  adult-onset hirsutism and acne, menstrual
  irregularities and proximal muscle weakness.
• All subjects reported that their symptoms were
  more severe at certain times suggesting episodic
  hypercortisolism

22
Cushings excluded

• 54 subjects
• 52 females, two males
• All Caucasians
• Median age 36 years
• BMI was 32.9 8.0 kg/m2
• Average weight gain was 48.3 35 pounds
• Cushings syndrome was excluded by lack of
  progression of symptoms and lack of biochemical
  evidence.
• Many were diagnosed with other conditions,
  including growth hormone deficiency

23
Symptoms/ Signs
24
24-Hour Urinary Free Cortisol (UFC)

• Integration of plasma cortisol throughout the day
• Good assays (using HPLC or mass spectroscopy)
  have a normal range of 10-34 ?g, with higher
  levels for men.
• Normal range of many older assays is 20-100 ?g
  /day indicating some non-specificity or
  interference of the assay
• PseudoCushings patients may have normal values in
  newer assays, but elevated levels in older
  assays.
• Many Cushings patients have normal values in the
  new assay
• My data demonstrates that most Cushings patients
  are periodic, therefore patients need to collect
  multiple collections hopefully when they have
  high cortisol.
• May be normal if subject is high at night and low
  during the day.

25
UFC Cushing's UFCgt 34 micrograms/day(50/66) UFClt
34 micrograms/day(57/66)
200-470
180
160
140
120
micrograms/day
100
80
60
40
20
0
0
10
20
30
40
50
60
Cushing's Patients
26
UFC non-Cushing's One UFC gt 34 micrograms/day
(13/51) One UFC lt 34 micrograms/day (50/51)
90-145
80
70
60
50
micrograms/day
40
30
20
10
0
0
10
20
30
40
50
non-Cushing's Patients
27
10 hr urine Cortisol/Cr

• Corcuff, et al. Clinical Endocrinology 481998,
  503-508.
• Night-time (from 10 PM to 8 AM) UFC excretion
  (correct for g of creatinine)
• 16 nmol/umol was the cutoff
• Helpful in subjects with high night time cortisol
  excretion and low daytime cortisol excretion
• Correcting for US units 16 ug/g is a reasonable
  cut-off
• I need to tabulating our data, but this is a
  reasonable approach.

28
Urinary 17-OH Corticosteroids (17-OHS)

• One of the earliest tests
• Went out of favor about 10 years ago and has been
  (incorrectly) replaced by UFC.
• UFC is probably better for full-blown Cushings
  compared to obese and normal subjects.
• 17-OHS may be better for picking up mild cases.
• Can use the same collection for both, so its
  worthwhile to measure 17-OHS in addition to UFC.
• Can also express results per gram of creatinine
  to correct for obesity

29
17 OHS Cushing's One 17 OHS gt 6mg/day(52/63) One
17 OHS lt 6mg/day(53/63)
30-95
25
20
mg/day
15
10
5
0
0
10
20
30
40
50
60
Cushing's Patients
30
17 OHS non-Cushing's One 17 OHS gt 6mg/day
(15/50) One 17 OHS lt 6mg/day (48/50)
16-32
14
12
10
8
mg/day
6
4
2
0
0
10
20
30
40
50
non-Cushing's Patients
31
17 OHS/g Cr Cushing's One 17 OHS/g Cr gt 3.6
micrograms/g(45/61) One 17 OHS/g Cr lt 3.6
micrograms/g(45/61)
14-60
12
10
8
micrograms/day
6
4
2
0
0
10
20
30
40
50
60
Cushing's Patients
32
17 OHS/g Cr non-Cushing's One 17 OHS/g Cr gt 3.6
micrograms/g (15/50) One 17 OHS/g Cr lt 3.6
micrograms/g (46/50)
12.0
10.0
8.0
micrograms/day
6.0
4.0
2.0
0.0
0
10
20
30
40
50
non-Cushing's Patients
33
Diurnal Plasma Cortisol Test

• Normal individuals and patients with
  pseudo-Cushing states have a pronounced diurnal
  rhythm of cortisol with the highest values in the
  morning and lower values at night.
• Patients with Cushing syndrome lack their diurnal
  variation of cortisol.
• Papanicolaou et al. (JCEM, 1998, 831163-1167)
  compared morning and nighttime plasma cortisol in
  97 patients with proven Cushing syndrome and 31
  patients with pseudo-Cushing states.
• A midnight plasma cortisol greater than 7.5 ?g/dL
  makes Cushings syndrome likely.
• Patients taking oral estrogens (or birth control
  pills) will have an increase in their CBG and a
  falsely high serum cortisol level.
• Pretty good test, but hard to arrange.

34
Midnight plasma cortisol
Papanicolaou et al. (JCEM, 1998, 831163-1167)
35
Night Cortisol Cushing's Night cortisol gt 7.5
micrograms/dL (26/57) Night cortisol lt 7.5
micrograms/dL (31/57)
30
25
20
micrograms/dL
15
10
5
0
0
10
20
30
40
50
60
Cushing's Patients
36
Night Cortisol non-Cushing's night cortisol gt
7.5 micrograms/dL(11/44) night cortisol lt 7.5
micrograms/dL(33/44)
16
14
12
10
micrograms/dL
8
6
4
2
0
0
10
20
30
40
50
non-Cushing's Patients
37
Diurnal Salivary Cortisol Test

• Salivary cortisol levels reflect plasma cortisol
  levels.
• Midnight plasma cortisol measurement requires
  blood-drawing and may be difficult to obtain in
  an outpatient setting.
• Measured by a company in Wisconsin called
  ACL. Also Esoterix
• Uses a "Salivette" in which the patient chews on
  a cotton tube for 2-3 minutes. The samples are
  stable for a week at room temperature and
  salivary cortisol is independent of the rate of
  saliva production.

38
Diurnal Salivary Cortisol Test (2)

• 36/39 patients with Cushing syndrome had a
  salivary cortisol gt 3.6 nmol/L (0.13 ?g/dl).
• 38/39 normal volunteers had a value 3.6 nmol/l
  (mean 1.2 nmol/L) and 37/39 patients with
  rule/out Cushing syndrome had a value 3.6
  nmol/l (mean 1.6 nmol/L).

39
Salivary Cortisols Cushing's salivary cortisol gt
4.3 nmol/L(43/64) salivary cortisol lt 4.3
nmol/L(58/64)
20-80
18
16
14
12
nmol/L
10
8
6
4
2
0
0
10
20
30
40
50
60
Cushing's Patients
40
Salivary Cortisols non-Cushing's One salivary
cortisolgt 4.3 nmol/L(9/53) One salivary
cortisollt 4.3 nmol/L(53/53)
16-22
14
12
10
8
nmol/L
6
4
2
0
0
10
20
30
40
50
non-Cushing's Patients
41
Both UFC and Salivary Cortisol are unlikely to
pick-up mild Cushings

• Serum cortisol less than 20 mg/dl (lower in
  evening when CBG is lower) is mainly (but not
  exclusively) bound to CBG and therefore little
  free cortisol is present in the blood.
• This results in little increase in salivary
  cortisol or UFC.
• At serum cortisol concentrations exceeding this
  cut-off, then salivary cortisol and UFC will rise
  dramatically.

42
Salivary cortisol Conclusions

• Convenient for periodic patients as the patient
  can collect many samples easily
• Try to have the patient collect when high
  symptoms, but Im finding that multiple
  collections (up to 8) is probably the best
  approach
• No better or worse than UFC for picking up mild
  cases.

43
Overnight dexamethasone test

• Give 1 mg of dexamethasone at midnight- collect 8
  am plasma cortisol
• Cushings patients resistant to glucocorticoid
  feedback.
• Old cut-off 5 mg/dL, new cut-off 1.8, 2 or 3
  mg/dL. Value greater than that consistent with
  Cushings syndrome.
• Cortisol assay isnt that good at low values
• May get falsely high values if on oral estrogens.
• Only half of classic Cushings patients have the
  genetic defects leading to resistance to
  dexamethasone-probably lower in mild/episodic
  patients (Bilodeau et al. 2006 20 2871-2886
  Genes Dev.)
• Friedman, T.C. (2006) An Update on the Overnight
  Dexamethasone Suppression Test for the Diagnosis
  of Cushings Syndrome Limitations in Patients
  with Mild and/or Episodic Hypercortisolism.
  Experimental and Clinical Endocrinology and
  Diabetes 216 356-360.

44
Overnight dexamethasone test
0800 h cortisol (?g/dL)
Patient
45
Overnight dexamethasone test

• Conclusion test useless for excluding Cushings
  syndrome.
• If someone has a high value after dexamethasone,
  may help with the diagnosis of Cushings
  syndrome, but those patients usually are severe
  and can be diagnosed anyway
• If patient suppresses to overnight dexamethasone,
  adrenal adenoma or ectopic is unlikely.
• I am now doing a prospective study using 0.25 mg
  of overnight dexamethasone, 1 mg of dexamethasone
  and the 2 mg/2 day dexamethasone test.
• All my patients suppress on the 2 mg/2 day test
• 0.25 mg may be helpful, but so far a lot of
  overlap between Cushings and Cushings excluded.

46
Dexamethasone-CRH test

• Patients with pseudo-Cushings states show a
  diminished response to exogenous CRH and a
  greater inhibition of cortisol production by
  glucocorticoids than patients with Cushings
  syndrome.
• Yanovski et al. (JAMA 1993, 2692232-2238)
  studied 39 patients with surgery confirmed
  Cushings syndrome and 19 patients with
  pseudo-Cushing states. Both groups of patients
  had UFC between 90-360 ug/day (nl 20-100 ug/day).
• Dexamethasone (0.5 mg) is given every 6 hours
  for 8 doses, starting at noon. The last dose is
  given at 6 A.M, 2 hours before the CRH test.
  Ovine CRH (1 mg/kg) is then given at 8 A.M.
  Plasma samples were analyzed for cortisol and
  ACTH at 4 basal time points (-15, -10, -5 and 0)
  and at 5, 15, 30, 45 and 60 minutes after oCRH.

47
Dexamethasone-CRH test

• Using a cutoff of 1.4 mg/dL, a plasma cortisol
  drawn 15 minutes after oCRH administration
  (following dexamethasone suppression) was able to
  completely separate patients with pseudo-Cushing
  states from those with Cushing syndrome. This
  was much better than just performing a oCRH test
  or dexamethasone test alone.
• Subsequently, many articles have shown the test
  is not full-proof
• Timing is crucial.
• Has not been tested in mild or periodic patients.
• The dex-CRH test is expensive and time consuming.
  I found that most of my patients with mild
  Cushings syndrome had low cortisol values
  following the test.

48
Pituitary MRI

• In literature approximately 50 of patients with
  Cushing disease have a visible tumor on MRI
  (older, non-dynamic, lower power MRIs).
• 10 of normal volunteers have MRIs consistent
  with a pituitary adenoma (Hall et al. Ann.
  Intern. Med., 1994, 120817-820).
• Now 3 Tesla doing dynamic MRIs can pick up small
  tumors are done.
• Patients without Cushings syndrome or with
  adrenal/ectopic Cushings can have a pituitary
  incidentaloma.
• Friedman, T.C., Zuckerbraun, E., Lee, M.L.,
  Kabil, M.S., Shahinian, H.K. (2007) Dynamic
  Pituitary MRI Has High Sensitivity and
  Specificity for the Diagnosis of Mild Cushings
  Syndrome and Should be Part of the Initial
  Workup. Hormone and Metabolic Research
  39451-456.

49
Pituitary MRI
23 of 24 patients had had a MRI consistent with a
pituitary lesion

12
10
8
Tumor size (mm)
6
4
2
0
Pt
50
Pituitary MRI-Cushings Syndrome-excluded
10

9
8
7
Tumor size (mm)
6
5
4
3
2
1
0
Pt
51
Pituitary MRI

• 23 of 24 patients had had a MRI consistent with a
  pituitary lesion (21 with a microadenoma, two
  with pituit