American Association for Clinical Chemistry
Better health through laboratory medicine
Clinical Laboratory Strategies: January 10, 2008

Screening for Mild Cushing’s Syndrome
New Study Highlights Diagnostic Challenges
By Deborah Levenson


With growing numbers of patients who have clinical symptoms that mimic those of Cushing’s syndrome—an extremely rare hormonal disorder caused by prolonged exposure of the body's tissues to high levels of the hormone cortisol—endocrinologists are increasingly called upon to exclude the condition in its mild form. An Italian research team recently published data that confirms that two screening tests commonly used in the U.S for Cushing’s syndrome offer acceptable sensitivity. This issue of Strategies examines their findings.

For endocrinologists, diagnosis of Cushing’s syndrome can present a challenge. Growing numbers of obese, depressed, diabetic, and hypertensive patients present with clinical symptoms—including weight gain, poor wound healing, immune suppression, and mood changes—that match those of Cushing’s. Yet the syndrome is extremely rare, occurring at a rate of only about 5 cases per million, according to the Cushing’s Support and Research Foundation. With such low incidence and because treatment can involve surgery, accurate screening and diagnosis is of the utmost importance.

Yet screening for the disease has been the topic of some controversy, in part because of concerns regarding specificity. Now, a recently published assessment of urinary free cortisol (UFC), 1-mg dexamethasone overnight suppression (OST), and serum cortisol at midnight in a large group of patients shows that OST and UFC—both of which are common in the U.S.—offer acceptable sensitivity. However, the research notes that specificity of these tests does vary considerably (Journal of Clinical Endocrinology and Metabolism 2007; 92: 11, 4123-4129).

The Findings

Researchers from the University of Milan and Division of Endocrinology and Metabolic Diseases, Ospedale San Luca, Istituto Auxologico Italiano in Milan, Italy retrospectively reviewed results from patients whose clinical symptoms suggested Cushing’s syndrome, but in whom the disorder was eventually excluded. The numbers of patients who underwent screening by UFC, OST, and serum cortisol at midnight were 3,461, 357, and 864, respectively.
UFC and OST at a 5 μg/dL cutoff had the highest specificities, 91% (95% CI, 90.2-92.1%) and 97% (95% CI, 96.3-98.5%), respectively. Results from midnight serum were less satisfying, yielding just 87% (95% CI, 84.3-91.1%) specificity with the 7.5 μg/dL cutoff, while the 1.8 μg/dL threshold resulted in an unacceptably high proportion of false positives at only 20% specificity (95% CI, 16.0-24.4%).

When dealing with a possible diagnosis of Cushing’s disease, lab results must be reliable, emphasized lead author Francesco Cavignini, MD, Professor of Endocrinology and Chief of Endocrinology and Metabolic Diseases at Instituto Auxalogico Italiano. “The main take home message of our research is that maximum effort should be made by the laboratory not to add inaccuracies to the already existing problems inherent to the clinician’s choice of the test and interpretation of the results,” he declared. “In patients with an abnormal result from any of these three screening tests, the two remaining tests should confirm/disprove the diagnosis.” That’s because the three tests offer information about different aspects of the disorder.

Cavignini also called for large scale studies to establish normative cutoffs for these tests –especially midnight cortisol—based on gender and age. “Obviously, a rare disease such as Cushing’s syndrome requires highly sensitive tests and this should favor the use of lower diagnostic thresholds,” he noted, explaining that lower cutoffs produce a marked decrease in specificity and increased risk of false positives.

No Midnight Cortisol Here

Lowering the diagnostic threshold for serum cortisol is a moot point in the U.S. because physicians don’t order it, noted Herschel Raff, PhD, Research Director and Clinical Supervisor in the Endocrine Research Laboratory at Aurora St. Luke’s Medical Center and Professor at Medical College of Wisconsin, both in Milwaukee. “You need to hospitalize the patient and draw blood while the patient is sleeping, and if he or she wakes up during that process, you have to wait until the next night,” he explained.

But Cavignini’s findings are valuable, especially because his study included large numbers of patients, Raff emphasized. “It’s the first really good specificity study. You need lots of samples and patients when studying specificity in a rare disease.” As for the message U.S. labs should take away from the paper, Raff said, “Both UFC and OST offer acceptable specificity.”

The study also makes another important point, according to Mary Lee Vance, MD, Professor of Internal Medicine and Meurosurgery and Associate Director of the General Clinical Research Center at the University of Virginia Health System in Charlottesville, VA. “The bottom line is that you should use the most precise method,” she noted. She maintained that labs should determine UFC by tandem mass spectrometry, noting that other UFC methods are often imprecise and can result in falsely high results.

While the Italian study confirms the specificity of OST, Vance does not order it. “The practical reality is that timing is everything with this test, and it rarely happens when it should. Another problem is that patients who are hypermetabolizers often aren’t suppressed the next morning and they get false-positive results. A third reason is that because dexamethasone is metabolized in the liver, certain medications can speed that process, resulting in false positives in the morning,” she explained.

While Vance prefers UFC, a recent paper by Raff and a colleague maintains that differences in methodology make interpretation of UCF data challenging, and issues with limited sensitivity and specificity compound the problem. (Journal of Clinical Endocrinology and Metabolism 2006; 91: 3746-3753). Biller maintained using mass spectrometry makes the test extremely reliable, pointing out that she always orders the test be done by this method.

Rise of the Salivary Cortisol Test

Another option that is gaining popularity as both a screening and diagnostic test is late-night salivary cortisol, according to Raff. “While the concept behind the test is similar to that behind midnight serum cortisol, you can send the patient home with a saliva sampling tube and he or she can get the sample at bed time.” One downside is that hypertension, advanced age, and certain psychiatric conditions can lead to false positives, but Raff’s paper notes that repeat testing is often normal. Raff’s institution uses late night salivary cortisol as a first-line test for Cushing’s, he said, adding that 11 studies have shown that sensitivity and specificity are both greater than 90%.

“Late night salivary cortisol is not perfect, but neither are any of these tests. The upshot is that diagnosis of mild Cushing’s remains a challenge,” Raff commented.


Update
Strategies February 22, 2007
Experts assembled by the CDC’s National Office of Public Health Genomics have issued a paper that recommends neither for nor against cytochrome P450 gene testing to guide treatment with selective serotonin inhibitor drugs for patients with depression. The report from the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working group echoes EGAPP’s preliminary recommendations. Read the report.


Was this issue of Strategies of interest to you? Please help us improve by rating this article.
Rate this page:       

Heads Up
Recent papers recommended by the Strategies editorial board