The Impact of the Biological Variability or Assay Performance on AMH Measurements: A Prospective Cohort Study With AMH Tested on Three Analytical Assay-Platforms

• Main Authors: Leif Bungum, Julia Tagevi, Ligita Jokubkiene, Mona Bungum, Aleksander Giwercman, Nick Macklon, Claus Yding Andersen, Tobias Wirenfeldt Klausen, Niels Tørring, Ajay Kumar, Sven Olaf Skouby
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2018-10-01
• Series: Frontiers in Endocrinology
• Subjects: anti mullerian hormone; antral follicle count; biomarkers; infertility counseling; female infertility
• Online Access: https://www.frontiersin.org/article/10.3389/fendo.2018.00603/full

This study examined longitudinal, age-related and intra-individual variation in Anti-Müllerian Hormone (AMH) in regular menstruating women and correlated the hormonal levels to the antral follicle count (AFC). The impact of variations on an algorithm for calculation of follitropin-dose for ovarian stimulation were also tested. The study was carried out at a fertility clinic of a tertiary university hospital and had a prospective trial design. Twenty-six healthy women not receiving infertility treatment aged 22 to 50 years participated. Blood sampling for hormonal analysis was done every fifth day throughout three consecutive menstrual cycles, AFC was determined with 3-dimentional ultrasound and AMH measured by different assays from Beckman Coulter, Roche and Ansh Labs. Outcome measures were maximum and minimum difference in absolute and relative terms for each study subject during the test-period, coefficient of variation (Cv) for AMH for each cycle and cycle-day and correlation between AMH and AFC. The impact from variable AMH levels on an algorithm calculating follitrophin-delta dose in ovarian stimulation was explored. A significant longitudinal age-independent variation in AMH-levels and coefficient of variation in cycles and cycle days was found. A strong correlation between AMH-levels and AFC was confirmed and a case of significant divergence between assays was seen. Variations in AMH had a significant impact on an algorithm calculated dosage of gonadotrophins in ovarian stimulation. The finding of a substantial longitudinal variation in AMH question one recording being sufficient in quantifying gonadotrophins for ovarian stimulation, decision making and prognostication related to infertility treatment and counseling. Occasionally, commercial assays may fail to recognize specific AMH cleavage-products.