Contents
Contributors
Search

The differentiation of true Cushing's syndrome from PCS can be challenging for the physician, since many of the latter patients have concomitant hypertension, obesity, muscle weakness, whereas anxiety and depression are frequent features of Cushing's syndrome. Furthermore, many biochemical tests commonly used for the diagnosis of Cushing's disease are not highly specific, resulting in a significant overlap of values: (i) 24-hr urinary free cortisol (UFC) levels can be mildly elevated in patients with PCS (7,8), (ii) the 1-mg overnight dexamethasone suppression test is highly sensitive but has relatively low specificity, and inadequate cortisol suppression suggestive of Cushing's syndrome has been documented in normal subjects (9), patients with obesity (10), major depression, alcoholism and other psychiatric disorders (11), or chronically ill patients (12), (iv) the low-dose 2-mg 2-day dexamethasone suppression test has a diagnostic accuracy of only 71% (13), and (v) early morning plasma total and free cortisol values significantly overlap in patients with PSC and in those with Cushing's disease (8).

In PCS, however, the hypothalamic-pituitary-adrenal axis, despite the CRH neuron hyperactivity, is otherwise normal and adequately, albeit not fully, restrained by the negative cortisol feedback (6). On the basis of this principle, the following tests can be used to distinguish between Cushing's disease and PCS:

(i) Dexamethasone suppression-ovine CRH test. The test consists of a 2-mg 2-day dexamethasone regime followed by intravenous ovine CRH 1µg/kg of body weight 2 hours after the last dexamethasone dose and a serum cortisol value measured 15 minutes after CRH administration. All patients with PCS suppress to less than 1.4 ug/dl (38 nmol/L), whereas all patients with Cushing's syndrome have serum cortisol values higher than these; thus, the test has close to 100% sensitivity, specificity, and diagnostic accuracy for PCS (see Table 1) (13).

(ii) A single midnight serum cortisol of less than 7.5 ug/dl (206 nmol/L) has a sensitivity of 96% and a specificity of 100% for ruling out the diagnosis of Cushing's disease (12). It should be taken into consideration that in this particular study, patients were hospitalized, the indwelling catheter was inserted 1 hr prior to sampling, and patients were instructed to fast and avoid physical activity 3 hrs prior to sampling, in order to avoid falsely elevated serum cortisol values. This cannot be used as a simple screening test, but it can be of help in patients who pose a diagnostic dilemma .

(iii) ovine CRH-stimulation test. Patients with PCS show a blunted ACTH response to ovine CRH, whereas about 80 %of patients with Cushing's disease have an exaggerated one (14,16).

(iv) Nighttime salivary cortisol can be a useful test for the diagnosis of Cushing's syndrome. A midnight value greater than 550 ng/dl (15.2 nmol/L) provides a specificity of 100% and a sensitivity of 93% for Cushing's syndrome. These values were determined in an inpatient setting with abstinence from food and physical activity for 3 hours before saliva collection (15). Further studies in an outpatient setting are necessary before extending the use of this criterion in the diagnosis of Cushing's syndrome.

Testing with newer peptides such as arginine vasopressin (AVP), lysine vasopressin (LVP), desmopressin (a synthetic vasopressin analogue) and hexarelin (a synthetic GHRH analogue) has been used for the differentiation between pituitary Cushing's disease and ectopic ACTH secretion. In most cases (but not all) patients with Cushing's disease demonstrate an exaggerated serum ACTH and cortisol response after the administration of the above peptides, whereas most patients with ectopic ACTH secretion do not (1). Not many studies though have focused on the use of these tests for the diagnosis of PCS. Coiro et al. showed that either desmopressin or hexarelin testing can differentiate between pituitary Cushing's disease and alcohol-induced PCS (16). Similarly, desmopressin was able to differentiate between women with pituitary Cushing's disease and depressive women (17). Both these studies did not include patients with ectopic ACTH production. Overnight pretreatment with 1 mg of dexamethasone before the morning administration of LVP was found to have a sensitivity of 96% and a specificity of 100% in correctly identifying patients with Cushing's syndrome, compared to control subjects (18).

Finally, it should be noted that PCS is not a progressive condition and it resolves with the treatment of or the improvement in the underlying condition. In ambiguous cases retesting of the patient after 3 or 6 months can be very helpful.