A C de Kat1, F J M Broekmans2, A C van Westing3, E Lentjes4, W M M Verschuren5, Y T van der Schouw3. 1. Department of Reproductive Medicine and Gynecology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands; Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. Electronic address: a.c.dekat@umcutrecht.nl. 2. Department of Reproductive Medicine and Gynecology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. 3. Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. 4. Department of Clinical Chemistry and Hematology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. 5. Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands; National Institute for Public Health and the Environment, Antonie van Leeuwenhoeklaan 9, 3721 MA Bilthoven, The Netherlands.
Abstract
OBJECTIVE: Anti-Müllerian hormone (AMH), a quantitative marker of ovarian reserve, is used for both clinical and research purposes in the field of reproductive medicine. Numerous AMH assays have been developed. Among other factors, the lack of large-scale comparisons of the various assays hinders the universal interpretation of AMH levels. Moreover, little is known of the practical performance of highly sensitive assays compared with conventional assays with regard to the very low AMH levels found in women nearing menopause. This study aimed to compare the measurements of the Gen II (Beckman Coulter) assay with those of the highly sensitive picoAMH (AnshLabs) assay. METHODS: This cross-sectional study included 1985 premenopausal women who completed the second visit of the population-based Doetinchem Cohort Study, with a mean age of 42±7years. AMH levels were measured with the Gen II and picoAMH assays. Passing-Bablok and Bland Altman analyses were performed and differences in the proportion of detectable samples were assessed. RESULTS: The results from the Gen II and picoAMH assays were highly correlated, with a Spearman correlation coefficient of 0.91. The Passing-Bablok regression formula was picoAMH=0.01+1.69*GenII, meaning that on average picoAMH levels were 69% higher than Gen II levels. Of the 670 samples with an undetectable AMH value with the Gen II assay, AMH could be detected in 78% with the picoAMH assay, at a median concentration [interquartile range] of 0.05 [0.01-0.14] ng/mL. CONCLUSION: These results indicate that, despite a high correlation, there is a large relative difference between results of the Gen II and picoAMH assays. The use of a highly sensitive AMH assay is likely to result in a large increase in the proportion of samples with detectable levels. This may enable research into women's health across the menopausal transition and research into the potential clinical benefits of distinguishing between women with very low ovarian reserve.
OBJECTIVE: Anti-Müllerian hormone (AMH), a quantitative marker of ovarian reserve, is used for both clinical and research purposes in the field of reproductive medicine. Numerous AMH assays have been developed. Among other factors, the lack of large-scale comparisons of the various assays hinders the universal interpretation of AMH levels. Moreover, little is known of the practical performance of highly sensitive assays compared with conventional assays with regard to the very low AMH levels found in women nearing menopause. This study aimed to compare the measurements of the Gen II (Beckman Coulter) assay with those of the highly sensitive picoAMH (AnshLabs) assay. METHODS: This cross-sectional study included 1985 premenopausal women who completed the second visit of the population-based Doetinchem Cohort Study, with a mean age of 42±7years. AMH levels were measured with the Gen II and picoAMH assays. Passing-Bablok and Bland Altman analyses were performed and differences in the proportion of detectable samples were assessed. RESULTS: The results from the Gen II and picoAMH assays were highly correlated, with a Spearman correlation coefficient of 0.91. The Passing-Bablok regression formula was picoAMH=0.01+1.69*GenII, meaning that on average picoAMH levels were 69% higher than Gen II levels. Of the 670 samples with an undetectable AMH value with the Gen II assay, AMH could be detected in 78% with the picoAMH assay, at a median concentration [interquartile range] of 0.05 [0.01-0.14] ng/mL. CONCLUSION: These results indicate that, despite a high correlation, there is a large relative difference between results of the Gen II and picoAMH assays. The use of a highly sensitive AMH assay is likely to result in a large increase in the proportion of samples with detectable levels. This may enable research into women's health across the menopausal transition and research into the potential clinical benefits of distinguishing between women with very low ovarian reserve.
Authors: Emily R Garnett; Purviben Jariwala; Kesha Rector; William E Gibbons; Paul W Zarutskie; Sridevi Devaraj Journal: Pract Lab Med Date: 2019-09-27