Cutoff values of midnight salivary cortisol for the diagnosis of overt hypercortisolism are highly influenced by methods

https://doi.org/10.1016/j.cca.2009.11.033Get rights and content

Abstract

Background

Midnight salivary cortisol (MSC) concentration has been considered as a sensitive marker of overt hypercortisolism. Because no studies on commercially available automated, non-isotopic MSC assays have been reported, we determined and compared the diagnostic performance of an automated electrochemiluminescent immunoassay (ECLIA, Elecsys E170) and an in-house radioimmunoassay (RIA) for MSC measurement.

Methods

The study involved 126 consecutive patients referred for evaluation because of symptoms of Cushing's syndrome, obesity, or the presence of incidentally discovered adrenal adenoma. Using detailed clinical, hormonal and radiological evaluation overt endogenous hypercortisolism was confirmed in 9 patients and was excluded in 117 patients.

Results

ROC analysis indicated that the best performance of MSC was obtained at cutoff value of 0.35 µg/dl using ECLIA (sensitivity, 100%; specificity, 88%) and 0.29 µg/dl (sensitivity, 100%; specificity, 71%) using RIA. When the findings were compared to those obtained from low dose dexamethasone test, both ECLIA and RIA of MSC showed a better diagnostic performance.

Conclusion

MSC measurement is useful for the diagnosis of overt hypercortisolism but the cutoff value is highly dependent on the method used. We recommend the use of automated ECLIA for MSC assay, and we propose further studies on other automated immunoassay analyzers potentially suitable for MSC measurements.

Introduction

The diagnosis of endogenous hypercortisolism has been considered as one of the most challenging problems in current endocrinological practice. According to recent guidelines, the diagnostic algorithm includes measurement of diurnal variation of plasma cortisol, determination of 24-hour urinary free cortisol, measurement of midnight salivary cortisol (MSC), and determination of plasma cortisol after a low dose (1 mg) dexamethasone suppression test (LDDST) [1].

Cortisol in saliva correlates well with the unbound fraction of plasma cortisol and is unaffected by changes in corticosteroid-binding globulin (CBG) levels [2]. Therefore, salivary cortisol measurement represents a convenient way to assess the concentrations of free, biologically active cortisol. Its characteristics such as non-invasive and painless sampling, and stability of samples at room temperature for weeks made it as an ideal diagnostic test. The first MSC assay was described by Umeda et al. in 1981 [3], and its diagnostic value was evaluated by Luthold et al. in 1985 [4]. Since then MSC measurement has become widely accepted and it is now recommended by the Endocrine Society Clinical Practice Guideline as a first line diagnostic test in the diagnosis of Cushing's syndrome [1].

In the majority of earlier studies MSC has been determined by radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA). Although the results obtained with different assays correlated well, the absolute values showed significant discrepancies and an assay-specific reference range was needed for the best analytical sensitivity and specificity of each assay. Because only a few studies [5], [6] on commercially available automated, non-isotopic MSC assays have been reported, we determined and compared MSC using an automated electrochemiluminescent immunoassay (ECLIA, Elecsys E170, Roche Diagnostics, Switzerland, Basel) and a traditional in-house RIA, and compared the results to those obtained with LDDST to evaluate the diagnostic performance of the two assays for the diagnosis of overt hypercortisolism.

Section snippets

Patients and salivary sample collection

The study involved 126 consecutive patients referred to the 2nd Department of Medicine, Semmelweis University for evaluation because of symptoms of Cushing's syndrome, obesity, or the presence of incidentally discovered adrenal adenoma. The study was approved by the local Ethical Committee. None of the patients was taking corticosteroids, oral contraceptives or other drugs interfering with the hypothalamic-pituitary-adrenal function. All patients underwent a detailed clinical, biochemical and

Midnight salivary and plasma cortisol and LDDST

As shown in Table 1, the MSC concentration determined by both methods were significantly higher in patients who proved to have Cushing's syndrome (ECLIA mean ± SE, 1.2 ± 0.41 µg/dl; RIA mean ± SD, 0.92 ± 0.38 µg/dl) compared to patients in whom Cushing's syndrome was excluded by detailed clinical, biochemical and hormonal testing (ECLIA mean ± SD, 0.2 ± 0.02 µg/dl; RIA mean ± SD, 0.28 ± 0.03 µg/dl). As expected, patients with Cushing's syndrome had significantly higher midnight plasma cortisol (23.34 ± 6.67 µg/dl)

Discussion

Salivary cortisol has been traditionally measured by RIA, or ELISA [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], and only a few studies used automated electrochemiluminescent assays or liquid chromatography tandem mass spectrometry (LC–MSMS) [20]. However, in routine clinical practice automated immunoassays are widely used, and we were unable to find a reference range of MSC for diagnosis of overt hypercortisolism using the automated ECLIA. In addition, we found only one

Acknowledgment

A.P. is a recipient of the Janos Bolyai Research Fellowship.

References (30)

  • RK Schindhelm et al.

    Salivary cortisol as an alternative for serum cortisol in the low-dose adrenocorticotropic hormone stimulation test?

    J Endocrinol Invest

    (2009 Jul. 28)
  • ...
  • H. Raff

    Utility of salivary cortisol measurements in Cushing's Syndrome and adrenal insufficiency

    J Clin Endocrin Metab

    (2009)
  • H. Raff et al.

    Late-night salivary cortisol as a screening test for Cushing's syndrome

    J Clin Endocrinol Metab

    (1998)
  • H. Raff et al.

    Comparison of two methods for measuring salivary cortisol

    Clin Chem

    (2002)
  • Cited by (45)

    • Cortisol: Analytical and clinical determinants

      2023, Advances in Clinical Chemistry
      Citation Excerpt :

      Other steroids may also significantly interfere in general and when performing LNSC testing specifically. Cross-reactivity impacts cut-offs and clinical performance of LNSC for CS screening [107] and may be further compounded by the presence of synthetic steroids. MS provides a highly specific cortisol measurement which eliminates potential cross-reactivity with cortisone and other steroids.

    • Clinical utility of an ultrasensitive late night salivary cortisol assay by tandem mass spectrometry

      2018, Steroids
      Citation Excerpt :

      LNSC test has been advocated as one of the first line screening tests in patients suspected of CS [4]. There have been reports of excellent sensitivity and specificity of LNSC for CS diagnosis, however in most of these reports RIA was used for LNSC [12–18]. LC-MS/MS based method for LNSC is considered method of choice due to high test specificity including lack of cross-reactivity to other steroid compounds, which is commonly encountered in immunoassays, and with much improved test sensitivity.

    • Late-night salivary Cortisol is unaltered in patients with polycystic ovarian syndrome (PCOS), irrespective of disease phenotype, and in obese women, irrespective of the presence of PCOS

      2017, Endocrine Practice
      Citation Excerpt :

      Late-night salivary cortisol (LNSC) measurement has recently gained widespread acceptance as a noninvasive, reliable, and easy-to-perform test in the diagnosis of Cushing syndrome (CS) (1,2). Despite similar diagnostic performance of different assays, proposed cutoff values of LNSC differ largely within studies depending on the patient and control group selection criteria, specific assay, and statistical methods (3–7). Racial, ethnic, and cross-country differences might also influence LNSC levels (8,9).

    • Subclinical Cushing Syndrome: A Review

      2014, Surgical Clinics of North America
      Citation Excerpt :

      Both salivary and serum assays have been evaluated in SCS; midnight serum cortisol (MSeC) measurement is more burdensome as it requires the patient to be present for phlebotomy, unlike midnight salivary cortisol (MSaC). Midnight measurements are not skewed by obesity or depressive disorders, although they are unreliable in patients with erratic sleep schedules or who are employed in night-shift work.42,43 MSaC has gained in popularity recently for diagnosis of overt hypercortisolism, despite challenges to standardization of laboratory practices and techniques.44

    • The Adrenal Cortex

      2013, The Immunoassay Handbook: Theory and Applications of Ligand Binding, ELISA and Related Techniques
    View all citing articles on Scopus
    View full text