The value of serum dehydroepiandrosterone sulfate in differential diagnosis of Cushing's syndrome

Abstract

Objective: To evaluate the changes and diagnostic value of serum dehydroepiandrosterone sulfate (DHEAS) in Cushing's syndrome (CS) with different etiologies. Methods: The study retrospectively recruited patients diagnosed as CS in Drum Tower Hospital affiliated to Nanjing University Medical School between January 2012 and June 2019, including 36 patients (8 males, 28 females, with an average age of 44 years) with Cushing disease (CD) and 64 patients (6 males, 58 females, with an average age of 39 years) with adrenal CS (ACS). Meanwhile, 97 patients diagnosed as nonfunctional adrenal adenoma (NFA) were also included as controls. Clinical characteristics, laboratory data, adrenocorticotropic hormone (ACTH), serum DHEAS level and sex-and age-adjusted DHEAS ratio of the three groups were collected. The sensitivity and specificity of DHEAS and its ratio in differential etiology diagnosis of CS were compared using receiver operating characteristic (ROC) curve analysis. Results: Compared to NFA group, ACS patients had lower DHEAS levels [0.39 (0.39, 0.63) μmol/L vs 2.96 (1.92, 4.60) μmol/L, P<0.01] and lower DHEAS ratio [0.58 (0.27, 0.98) vs 3.95 (3.08, 6.83), P<0.01]. DHEAS [6.49 (4.32, 11.63) μmol/L] and DHEAS ratio [9.17 (4.49, 15.41)] in CD patients were significantly higher compared to those in NFA and ACS patients (all P<0.01). There were 53 ACS patients (82.8%) with suppressed ACTH level (<2.2 pmol/L) and 11 patients (17.2%) with normal/high ACTH level (≥2.2 pmol/L). The level of 24 hour urine free cortisol in normal/high ACTH level group was lower than the suppressed ACTH group [(1 299±511) nmol/24 h vs (1 972±876) nmol/24 h, P=0.04]. No significant differences were found in the DHEAS and DHEAS ratio between the two groups. ROC analysis showed that the area under the curve of serum DHEAS and DHEAS ratio in diagnosing ACS from CD was 0.997 and 0.990, respectively. The optimal cut-off values for DHEAS and its ratio were 2.06 μmol/L and 2.10, respectively. The diagnostic sensitivity and specificity of DHEAS were 97.5% and 100%, and those of DHEAS ratio were 95.0% and 100%, respectively. Conclusion: There are significant differences in serum DHEAS level and DHEAS ratio between ACS and CD patients, which might be used as indicators for the identification of the two main CS etiologies, especially in the identification of ACS patients without plasma ACTH suppression from CD patients.