In the absence of exogenous glucocorticoid use, patients with multiple and progressive features that are highly discriminatory for Cushing's syndrome should be tested for the endocrine disorder, according to new clinical practice guidelines developed by a task force of the Endocrine Society.
Testing is also recommended for patients with adrenal incidentaloma compatible with adenoma and children with combined reduced linear growth and increased weight, wrote lead author Dr. Lynnette K. Nieman of the National Institute of Child Health and Human Development and task force colleagues.
Although the signs and symptoms of full-blown Cushing's disease are clinically unmistakable, “the spectrum of clinical presentation is broad, and the diagnosis can be challenging in mild cases,” the authors wrote, noting that the guidelines were developed to help facilitate this process. Because the strength of the recommendations varies depending on the quality of evidence supporting them, “careful consideration of the patient's circumstances, values, and preferences is appropriate to determine the best course of action,” the authors wrote (J. Clin. Endocrinol. Metab. 2008 March 11 [doi:10.1210/jc.2008-0125]).
Pretest Probability
Before conducting any biochemical testing, physicians should obtain a thorough drug history to rule out iatrogenic Cushing's syndrome associated with excessive exogenous glucocorticoid exposure.
After exclusion of exogenous glucocorticoid use, diagnostic testing is recommended for individuals with multiple features considered discriminatory for the disease, including easy bruising, facial plethora, proximal myopathy, reddish-purple striae, thin skin in the young, and weight gain with decreasing growth velocity in children, the authors wrote.
In addition, because Cushing's syndrome is progressive, patients who demonstrate an accumulation of new features should be tested, as should patients who develop features of the condition, such as osteoporosis and hypertension, which are atypical in the general population for individuals their age. Finally, the presence of an incidentally found adrenal nodule contributes to the pretest probability of Cushing's syndrome. “Such patients usually do not present with overt clinical features of Cushing's syndrome, but biochemical hypercortisolism is present in a large fraction,” the authors wrote.
The authors recommend against widespread testing in any other patient group, including obese children, unless their statural growth has slowed.
Diagnostic Tests
With respect to initial diagnostic testing, the task force recommended any of the following: at least two urinary free cortisol (UFC) measurements, two late-night salivary cortisol measurements, a 1-mg overnight dexamethasone suppression test (DST), or the longer low-dose 2 day/2 mg per day DST.
The guidelines recommend against testing for any of the following because of their low diagnostic accuracy for Cushing's syndrome: random serum cortisol or plasma ACTH levels; urinary 17-ketosteroids; insulin tolerance test; loperamide test; and tests designed to determine the cause of Cushing's syndrome, such as pituitary and adrenal imaging and the 8-mg DST.
Patients with normal test results who have a high pretest probability for Cushing's syndrome based on clinical features, adrenal incidentaloma, or suspected cyclic hypercortisolism should be referred for further evaluation by an endocrinologist, while those with normal test results and a low pretest probability should be reevaluated in 6 months if symptoms persist. Patients with at least one abnormal test result should also be referred to an endocrinologist for further evaluation to confirm or exclude the diagnosis, the authors wrote.
Although the four recommended diagnostic tests have “acceptable” accuracy when the suggested cutoff points are used, “no test has optimally high specificity, so false positives may occur,” the authors noted. To minimize the possibility of a misdiagnosis, the guidelines recommend subsequent evaluation of abnormal initial test results using another one of the high-sensitivity tests. “We suggest the additional use of the dexamethasone-CRH test or the midnight serum cortisol test in specific situations,” the authors wrote. They recommended against the use of the desmopressin test, “except in research studies,” because its utility has not been validated.
With the exception of patients of having “the very rare case of cyclical disease,” further testing is not recommended for patients with negative results on two different tests, the authors noted. For patients with concordantly positive results from two different tests for whom there is no concern regarding possible non-Cushing's hypercortisolism, “we recommend tests to establish the cause of Cushing's syndrome,” they wrote. Finally, further evaluation and follow-up is suggested for those few patients with concordantly negative results who are suspected of having cyclical disease, and for those patients with discordant results in whom the pretest probability is high.
Special Populations
The guidelines also addressed diagnostic testing in special populations, recommending the use of UFC (and against the use of dexamethasone testing) in the initial evaluation of pregnant women, and suggesting the use of the 1-mg overnight dexamethasone suppression test rather than UFC for initial testing in patients with renal failure and in those suspected of having mild Cushing's syndrome. In patients who take antiepileptic drugs that are known to enhance dexamethasone clearance, dexamethasone testing was not recommended. Instead, the guidelines recommended measurements of nonsuppressed cortisol in blood, saliva, or urine. When cyclic Cushing's syndrome is suspected, “we suggest the use of UFC or midnight salivary cortisol tests rather than dexamethasone suppression tests,” the authors wrote.