Endocrine Abstracts

Introduction: The 250 μg cosyntropin stimulation test (CST) is used to diagnose non-classic congenital adrenal hyperplasia (NCCAH). The current recommendation to perform CST is when follicular morning 17-hydroxyprogesterone (17OHP) is higher than 6 nmol/l, and CST is considered positive for NCCAH diagnosis when the 60-minutes post-CST 17OHP serum level is above 30 nmol/l. These cut offs are mainly derived from radioimmunoassay (RIA) data. Recently, a validated enzyme-linked immunosorbent assay (ELISA) has widely replaced RIA in the measurement of serum 17OHP. This study aimed to determine the RIA and ELISA-based 17OHP cut offs at which CST should be performed.

Material and methods: We conducted a retrospective cohort study at Maccabi Healthcare Services, an Israeli Health Maintenance Organization (HMO). Data was retrieved from adult females (≥16 years) with suspected NCCAH, referred for CST during 2001–2020. Clinical indications that led to NCCAH testing were hirsutism, irregular menses, acne, alopecia, or infertility. NCCAH was defined as post-CST 17OHP serum level >30 nmol/l. Serum 17OHP levels were assayed by direct RIA (Wizard gamma counter, Perkin-Elmer) from January 2000 through March 2015, and by ELISA (IBL International, Tecan) from April 2015 to December 2020. We allocated the individuals into two groups according to the assay method used. For each group, a ROC curve was generated and optimal pre-CST 17OHP threshold with the highest sensitivity and specificity determined.

Results: Cosyntropin testing was performed in 2409 female subjects (1564 in the RIA and 845 in the ELISA groups). The mean (±SD) age was 24.1±7 years. NCCAH was diagnosed in 74 (4.7%) of the RIA group and 63 (7.5%) of the ELISA group. The mean (±SD) pre- and post-CST 17OHP levels were lower in the RIA group as compared to the ELISA group (4.1±6.4 vs. 5.9±9.0 and 9.9±15.3 vs. 12.3±17.3, respectively, P<0.0001). Using ROC analysis, the optimal pre-CST 17OHP cut off values predicting NCCAH were 6.05 nmol/l in the RIA group (sensitivity 93.2%, specificity 91.7%) and 8.16 nmol/l in the ELISA group (sensitivity 93.7%, specificity 92.3%). When the guideline-recommended pre-CST 17OHP cut off value of 6 nmol/l was used in the ELISA group, sensitivity was 95.2%, and specificity decreased to 84%.

Conclusions: Our study showed a significant number of unnecessary cosyntropin tests. The optimal RIA-based pre-CST 17OHP cut off was comparable with that recommended in the guidelines. However, the results suggest adopting a higher 17OHP cut off when using ELISA. Further studies that incorporate genetic data are needed.