Different ovarian response by age in an anti-Mullerian hormone-matched group undergoing in vitro fertilization

Recently, serum anti-Mullerian hormone (AMH) has been used as a good marker of ovarian response during in vitro fertilization (IVF). However, in the clinical setting, we felt that ovarian response was clearly different by age with the same AMH level. Then in this study we evaluated the relationship between serum AMH, age and parameters related to ovarian response and compared these parameters in regard to age within serum AMH-matched group. The relationship of these parameters were evaluated retrospectively in patients undergoing their first IVF cycle under a GnRH agonist flare up protocol (n = 456) between October 2008 and October 2010 in our clinic. To understand the relations between variables described above, principal component analysis (PCA) was performed. PCA revealed patients' age was at the different dimension from serum AMH and other variables. Therefore at first we segregated all patients into Low, Normal and High responder groups by their serum AMH using cut-off value of receiver operator characteristics curve analysis. Secondary, we divided each responder group into four subgroups according to patients' age. The high aged subgroups required a significantly higher dose of gonadotropin and a longer duration of stimulation; however, they had significantly lower peak E2 and a smaller number of total oocytes as well as M2 oocytes compared to the low aged subgroups. The influence of aging on the ovarian response was clearly seen in all groups; the ovarian response tended to decrease as patients' age increased with the same AMH level. Therefore serum AMH in combination with age is a better indicator than AMH alone.