From basic ovarian physiology to ovarian stimulation

From this follows the central clinical paradox that occupies much of the lecture: a young woman may turn out to be a poor responder, while an older patient may retain an unexpectedly high reserve. The chronological and biological age of the ovary diverge, and relying on year of birth alone leads either to an insufficient or to an excessive gonadotropin dose and a higher risk of complications.

To avoid this error, the speaker argues for a multifactorial approach to ovarian reserve assessment. This means interpreting anti-Mullerian hormone and antral follicle count together as complementary markers, framed by clinical context — the history of previous cycles, prior gynecological interventions, body mass index, and comorbidities. No single parameter is a sufficient basis for a decision on its own.

At the physiological level, the lecture recalls the two-cell, two-gonadotropin model of steroidogenesis, in which LH drives androgen production in theca cells while FSH ensures their aromatization to estrogens in granulosa cells. Understanding this mechanism explains why, in a subset of patients, adequate follicular growth requires not only FSH but also sufficient LH activity, and it sets the basis for differentiated prescribing.

The practical conclusion Fatemi arrives at is to abandon the one-size-fits-all protocol in favor of a personalized stimulation strategy. A correct baseline assessment of reserve, a realistic prediction of the oocyte yield, and a starting dose matched to the individual patient phenotype improve the efficiency of the cycle and reduce the likelihood of both poor response and hyperstimulation.

The lecture opens the morning session of the Summer School of Reproductologist 2023 and serves as the physiological foundation for the talks that follow on receptor pharmacogenomics, LH deficiency, and the management of poor-prognosis patients. It is addressed to reproductive specialists who seek to build a stimulation protocol on an understanding of mechanisms rather than on fixed templates.