The number of oocytes retrieved during IVF: a balance between efficacy and safety.

STUDY QUESTION

What is the relationship between the number of oocytes collected in fresh IVF treatments and the likelihood of cumulative delivery rate (fresh and frozen) per oocyte aspiration, severe ovarian hyperstimulation syndrome (OHSS) and thromboembolic events?

SUMMARY ANSWER

Cumulative delivery rate per aspiration increases up to 20 oocytes retrieved and then evens out while the incidence of severe OHSS increases more rapidly from around 18 oocytes and thromboembolic events, although rare, occurs in particular if 15 or more oocytes are retrieved.

WHAT IS KNOWN ALREADY?: Previous studies have shown that the number of oocytes retrieved for IVF is a positive predictor of live birth in fresh cycles. Few studies have investigated cumulative live birth rates and OHSS in relation to the number of aspirated oocytes.

STUDY DESIGN, SIZE, DURATION: Retrospective population-based registry study including 39 387 women undergoing 77 956 fresh IVF cycles in the period 2007-2013 and 36 270 consecutive transfers of frozen/thawed embryos in the period 2007-2014.

PARTICIPANTS/MATERIALS, SETTING, METHODS: Data from The Swedish National Quality Registry of Assisted Reproduction (Q-IVF) including all IVF cycles with oocyte retrieval performed in public or private infertility clinics during the study period, was cross-linked to the National Patient Register regarding diagnostic codes (ICD 10) for severe (OHSS) and thromboembolic events. Oocyte donation cycles were excluded.

MAIN RESULTS AND THE ROLE OF CHANCE

Live birth delivery rate in fresh cycles increased up to 11 oocytes retrieved and then evened out, where the live birth rate was 30.3% for a 34-year-old woman. The cumulative delivery rate per aspiration, including fresh transfer and all subsequent transfers of frozen-thawed embryos (FET cycles) per oocyte retrieval, increased up to approximately 20 oocytes where it reached 45.8%. The adjusted odds ratio (AOR) for live birth by the number of oocytes was 1.064 (95% CI: 1.061; 1.067). The incidence of severe OHSS increased significantly by the number of oocytes, particularly if more than 18 oocytes were retrieved. The AOR for OHSS by the number of oocytes was 1.122 (95% CI: 1.08; 1.137). Thromboembolic events were rare, a total of 16 events in 14 patients were observed, and occurred in particular if 15 or more oocytes were retrieved.

LIMITATIONS, REASONS FOR CAUTION: All FET cycles might not be included. Some embryos cryopreserved between 2010 and 2013 might still result in additional births until 2018. Furthermore the gonadotrophin dose was not included in the Q-IVF Registry in the study period, thus adjustment for dose was not possible.

WIDER IMPLICATIONS OF THE FINDINGS

The results suggest a shift at approximately 18-20 oocytes where the cumulative delivery rate per aspiration levels off and, at the same time, the incidence of severe OHSS increases more rapidly. Thromboembolic events, although rare, should also be taken into consideration at stimulation regimes for IVF. Evaluating data taking both efficacy and the most serious safety aspects into account, is a new approach and of crucial importance both for patients undergoing IVF and their physicians.

STUDY FUNDING/COMPETING INTEREST: Financial support was received through an agreement relating to research and the education of doctors (ALFGBG-70 940) and grant from the Hjalmar Svensson Research Foundation. None of the authors declares any conflict of interest.