Serum cortisol is still the gold standard and valid in clinical uses

cortisol free serum=addison's desease....;)

I think salivary cortisol can be used as good as serum and urine free cortisol

Http://www.clinchem.org/content/39/6/1353.full.pdf

Http://acb.sagepub.com/content/50/5/450.full.pdf+html

I like the overnight dexamethasone suppression test for screening. It is simple and cheap, and good suppression will rule out Cushing's. However, to make the diagnosis more specific, I like the salivary cortisol. The key to diagnosis of Cushing's is demonstrating the absence of diurnal variation. Serum is fine, but trying to collect blood samples at midnight is tough to do as an outpatient. Salivary cortisol is a good surrogate when drawing blood is not feasible.

The gold standard is Cortisol free urine (24-hour). Salivary cortisol in ambulatory pacientes ?salivary cortisol is influenced by some factors that are difficult to standardize in ambulatory patients - such as smoking - you should not smoke before testing because nicotine suppresses the enzyme that converts active cortisol in idle kortizon.Ne need to eat before one study also do not brush their teeth ...

Robert Steed is absolutely right, diurnal variation can be "easasily" excluded by saliva sampling, no matter that no real normalization is possible on saliva samples

The best screening test is the measurement of urinary cortisol in a 24-hour urine sample. This should be performed on three different collection samples in order to correct for incomplete urinary samples. If the urinary cortisol is elevated, an overnight dexamethasone suppression test may be indicated (J. Int Fed. Clin Chem. 6 (1994) 154-158.).

Sighart Golf

Clinical or basic science... I prefer several days of salivary cort with samples at wake-up, then every few hours to develop a response profile. Then you have more than just dichotomous "Cushings yes or no". Plasma good for binding globulins, CRH, etc. Urine good for metabolites. Really depends on what you want to know about the patient.

The diagnosis of Cushing's syndrome can be difficult. In cushing's syndrome elevated cortisol is present without the usual controlling mechanisms. Therefore the investigation of Cushing's Syndrome would involve assessing both. 1. assess whether there is elevated cortisol - a) urine free cortisol is good, b) random cortisol can be misleading, c) 9 a.m. and midnight cortisol (showing lack of diurnal variation) can be informative, d) salivary cortisol (9 am and midnight) looks promising but not used much in the UK yet. 2. Assess lack of usual control a) midnight and 9 am to assess lack of control, b) dexamethasone suppression test - 2 mg overnight.

In summary i agree with Walter Golf 24h urine cortisol to see whether there is "excess" cortisol, and overnight dexamethasone suppression to see whether cortisol responds to suppression.

The best test for screening is overnight dexamethasone suppression test (however the reference range is still controversial: 2? 3? or 5 mcg/dl?). Elevated midnight cortisol and increased free cortisol in 24h urine confirm diagnosis. On the other hand I have several dozens of patients with adrenal tumors with subclinical hypercortisolaemia in which surgery brought improvement of clinical status, despite normal urine free cortisol at the diagnosis. They all had lack of post-dexamethasone suppression (at present we use border on 3 mcg/dl, formerly 2 mcg/dl) , slightly elevated midnight cortisol and low (suppressed) morning levels of ACTH.

Presumably ENDOGENEOUS Cushing's syndrome:

All have limitations but depends on clinical context (clinical index of suspicion) and initial tests

CONFIRMATION OF DIAGNOSIS:

24h UFC -

Diffculty in accurate 24 h urine collections

False negatives in those with renal impairment.

Dexamethasone Suppression tests

Overnight (1mg or 2 mg) easy to perform or 48h LDDST (gold standard)

False positives: Drugs which increase Dex metabolism eg certain anti-convulsants; Non compliance; immunoassay interference; increased CBG eg OCP

Midnight Salivary Cortisol

If appropriately low excellent exclusion

Only MS methods really suitable

No EQA schemes

When obvious no diffculty but others may be required in equivocal cases

NB: New category of subclincal autonomous glucocorticoid hypersecretion (SAGH) - we find 24 h UFC usually high normal or borderline elevated and ACTH not always suppressed

I think that it depends on the type of your suspicious clinical judgment.

For example, when you want to diagnose subclinical cushing overnight DST is the best option. 

When you want to diagnose cushing in Renal failure, DST is the best option. 

In depression and obesity and pseudo cushing ,  you may need to other tests, such as CRH dexamethasone test or so on. 

Overally, overnight Dexamethasone Suppression test is easy test,  however,  maybe you need to help from other tests. 

Dr Alireza Arefzadeh. 

Www.drarefzadeh.com

An advantage of salivary cortisol is the ease of monitoring cortisol levels over an extended period and in 'at-home' normal everyday environments.  Interpretation and diagnosis of longitudinal 'profiles' is complex, and ideally includes detailed information about sleep, diet, activity, and psycho-social stress.  

Salivary Cortisol. CBG is absent from saliva, and the use of

salivary cortisol measurements offers a sensible alternative

in that the test does not require hospitalization. The diag-

nostic accuracy of a single midnight salivary cortisol level

has been established in several studies. The cutoff points

that define disease vary depending on the assay used.

In one study, a cortisol value greater than 2.0 ng/mL

(5.5 nmol/L) had a 100% sensitivity and a 96% specificity

for diagnosis of Cushing syndrome.

the sensitivity

and specificity of UFC is poor, and most centers

have replaced these assays with the more sensitive mea-

surement of urinary free cortisol. This is an integrated

measure of plasma free cortisol: as cortisol secretion

increases, the binding capacity of CBG is exceeded, result-

ing in a disproportionate rise in urinary free cortisol.

The overnight test is a useful outpatient screening

test  in which 1 mg of dexamethasone is given at

11 PM. A normal response is a plasma cortisol level of less

than 50 nmol/L (