Harmonization of serum cortisol methods for dynamic testing

Jacquelien Hillebrand; Annemieke Heijboer; Judith Bons; Herwaarden Antonius van

doi:10.1530/endoabs.110.P182

P182

Prev Next

Section Contents Cite

Endocrine Abstracts (2025) 110 P182 | DOI: 10.1530/endoabs.110.P182

ECEESPE2025 Poster Presentations Adrenal and Cardiovascular Endocrinology (169 abstracts)

Harmonization of serum cortisol methods for dynamic testing

Jacquelien Hillebrand ^1,2 , Annemieke Heijboer ^1,2,3 , Judith Bons ⁴ & Antonius van Herwaarden ⁵

Author affiliations

¹Amsterdam UMC, University of Amsterdam, Department of Laboratory Medicine, Endocrine Laboratory, Amsterdam, Netherlands, ²Amsterdam Gastroenterology, Endocrinology & Metabolism, Amsterdam, Netherlands, ³Amsterdam UMC, Vrije Universiteit, Department of Laboratory Medicine, Endocrine Laboratory, Amsterdam, Netherlands, ⁴Maastricht University Medical Center+, Department of Clinical Chemistry, Central Diagnostic Laboratory, Maastricht, Netherlands, ⁵Radboud University Medical Center, Department of Laboratory Medicine, Nijmegen, Netherlands

JOINT1711

Background: Dynamic tests are used in diagnosis and follow-up of endocrine disorders. Unfortunately, there is considerable variation in how these tests are performed, including the patient preparation and analytical methods used with different standardization. Despite these differences, the interpretation of dynamic tests is often based on fixed cut-off values. The (pre-)analytical variability means that dynamic test results are not similar between medical centers, leading to confusion and treatment delay. Dynamic testing for adrenal insufficiency or Cushing’s syndrome includes measuring serum cortisol. Our aim is to explore whether the different serum cortisol methods used can be harmonized to improve the interpretation of dynamic tests.

Methods: Blood was collected from 45 participants. Females (n=23) were not using oral contraceptives. Sera were aliquoted, frozen and sent to 25 Dutch laboratories. Serum cortisol was measured using in-house LC–MS/MS (n=5), Roche Cobas (n=4), Siemens Atellica/Centaur (n=6), Beckman Coulter DxI/Access (n=5) and Abbott Architect/Alinity (n=5) cortisol assays. One of the LC–MSMS methods was checked against the IFCC Cortisol Reference serum Panel and set as the expert method. Passing-Bablok regression analyses were performed and within method variation was determined. Maximum allowable imprecision was set at 8%.

Results: Cortisol measurements in laboratories using the same cortisol assay showed an imprecision of <8% except for the Beckman Coulter assay (31% of samples >8%). Regression analyses showed good agreement between the LC–MS/MS, Roche and Beckman Coulter and the expert method (slope 1.02, 0.99, 1.00 and intercept 1.1, 1.8, −7.0 nmol/L respectively), and slightly less for Abbott (slope 0.93, intercept 8.8 nmol/L). The Siemens assay showed significant deviation in slope (1.40) and intercept (−35 nmol/L).

Conclusions: This study shows that LC–MS/MS assays and Roche cortisol assays are well standardized and that harmonization is not necessary. Harmonization is possible for the Abbott assay whereas the Siemens assay requires re-standardization. The Beckman Coulter assay shows a high within–method variation that first needs improvement. Standardization or harmonization of serum cortisol assays along with aligning the preanalytical workup will pave the way for uniform interpretation of dynamic testing.