Random-start ovarian stimulation in an oocyte donation programme: a large, single-centre, experience

Do live birth rates differ between recipients matched with donors using conventional ovarian stimulation compared with those using random-start protocols?

Retrospective analysis of 891 ovarian stimulations in egg donors (January-December 2018) and clinical outcomes in matched recipients (n = 935). Donors commenced ovarian stimulation on day 1-3 of the menstrual cycle (n = 223) or in the mid/late-follicular (n = 388) or luteal phase (n = 280) under a conventional antagonist protocol. Live birth rate of matched recipients was the main outcome.

Duration of stimulation and total gonadotrophin dose were comparable between conventional versus random-start groups. The number of collected eggs were similar (17.6 ± 8.8 versus 17.2 ± 8.5, P = 0.6, respectively). Sub-group analysis showed that stimulation length (10.2 ± 1.8 versus 9.8 ± 1.7 versus 10.4 ± 1.7, P < 0.001) and gonadotrophin consumption (2041.5 ± 645.3 versus 2003.2 ± 647.3 versus 2158.2 ± 685.7 IU, P = 0.01) differed significantly between the conventional, mid/late follicular and luteal phase groups, respectively. In matched recipients receiving fresh oocytes and undergoing fresh embryo transfer, the biochemical pregnancy (63.8% and 63.3%; P = 0.9), clinical pregnancy (54.6% and 56.1%; P = 0.8) and live birth rates (47.7% and 46.6%; P = 0.7) per embryo-transfer were similar between conventional versus random groups. Similar results were obtained in recipients receiving vitrified eggs. Euploidy rate was also comparable.

No notable variations were found in clinical outcomes using oocytes obtained from random-start protocols and those proceeding from conventional ovarian stimulation in oocyte donation treatments. Luteal-phase stimulation seems to require longer stimulation and higher FSH consumption. Random-start stimulation strategy does not impair the potential of the oocyte yield or clinical outcomes in oocyte donation cycles.