The BISTIM study: a randomized controlled trial comparing dual ovarian stimulation (duostim) with two conventional ovarian stimulations in poor ovarian responders undergoing IVF

• Published in: Human reproduction (Oxford) 2023-05, Vol.38 (5), p.927-937
• Main Authors: Massin, N, Abdennebi, I, Porcu-Buisson, G, Chevalier, N, Descat, E, Piétin-Vialle, C, Goro, S, Brussieux, M, Pinto, M, Pasquier, M, Bry-Gauillard, H
• Format: Article
• Language: English
• Subjects: COVID-19; Female; Fertilization in Vitro - methods; Humans; Original; Ovary; Ovulation Induction - methods; Pregnancy; Pregnancy Rate
• ISSN: 0268-1161; 1460-2350
• DOI: 10.1093/humrep/dead038

Abstract STUDY QUESTION Is the total number of oocytes retrieved with dual ovarian stimulation in the same cycle (duostim) higher than with two consecutive antagonist cycles in poor responders? SUMMARY ANSWER Based on the number of total and mature oocytes retrieved in women with poor ovarian response (POR), there is no benefit of duostim versus two consecutive antagonist cycles. WHAT IS KNOWN ALREADY Recent studies have shown the ability to obtain oocytes with equivalent quality from the follicular and the luteal phase, and a higher number of oocytes within one cycle when using duostim. If during follicular stimulation smaller follicles are sensitized and recruited, this may increase the number of follicles selected in the consecutive luteal phase stimulation, as shown in non-randomized controlled trials (RCT). This could be particularly relevant for women with POR. STUDY DESIGN, SIZE, DURATION This is a multicentre, open-labelled RCT, performed in four IVF centres from September 2018 to March 2021. The primary outcome was the number of oocytes retrieved over the two cycles. The primary objective was to demonstrate in women with POR that two ovarian stimulations within the same cycle (first in the follicular phase, followed by a second in the luteal phase) led to the retrieval of 1.5 (2) more oocytes than the cumulative number of oocytes from two consecutive conventional stimulations with an antagonist protocol. In a superiority hypothesis, with power 0.8 alpha-risk 0.05 and a 35% cancellation rate, 44 patients were needed in each group. Patients were randomized by computer allocation. PARTICIPANTS/MATERIALS, SETTING, METHODS Eighty-eight women with POR, defined using adjusted Bologna criteria (antral follicle count ≤5 and/or anti-Müllerian hormone ≤1.2 ng/ml) were randomized, 44 in the duostim group and 44 in the conventional (control) group. HMG 300 IU/day with flexible antagonist protocol was used for ovarian stimulation, except in luteal phase stimulation of the duostim group. In the duostim group, oocytes were pooled and inseminated after the second retrieval, with a freeze-all protocol. Fresh transfers were performed in the control group, frozen embryo transfers were performed in both control and duostim groups in natural cycles. Data underwent intention-to-treat and per-protocol analyses. MAIN RESULTS AND THE ROLE OF CHANCE There was no difference between the groups regarding demographics, ovarian reserve markers, and stimulation parameters. The me