IVF versus surgery for infertility in women with deep and ovarian endometriosis: a partially randomized patient...

Endometriosis-associated infertility can be managed either surgically or through assisted reproductive technologies (ART). Given the recent substantial improvement in the accuracy of ultrasound diagnosis of ovarian endometriomas and deep infiltrating endometriosis, surgery is no longer required to detect these disease forms. Therefore, infertile women with ultrasound-diagnosed endometriosis face two alternatives; they can be either operated or straight embark in in vitro fertilization (IVF). In this study, we aimed to compare these two options, as no randomized clinical trial (RCT) has yet addressed this clinical scenario.

A multicenter, pragmatic, partially randomized patient preference trial (PRPT) was conducted at three Italian tertiary referral centers with expertise in both endometriosis surgery and IVF. The trial was designed to compare reproductive outcomes between IVF and surgery in infertile women with ultrasound-diagnosed ovarian and deep infiltrating endometriosis. Women with a history of previous surgery for endometriosis or prior IVF were excluded. Additional key eligibility criteria were age <40 years, regular menstrual cycles, endometriomas <4 cm, absence of hydrosalpinx, ureteral stenosis, or sub-occlusive bowel symptoms, and normal semen analysis. Women who had a strong preference and declined randomization could be enrolled in a parallel patient preference cohort (PPT). Women allocated to IVF were scheduled for up to three complete IVF cycles whereas those allocated to surgery attempted natural conception. The primary outcome was conception leading to live birth within 12 months of randomization. Analyses were performed according to the intention-to-treat principle. Considering a 20% absolute increase in live birth rate in the IVF group as clinically relevant, the preplanned sample size was 206 women (103 per group), at the usual level of α=5% and β=20%. The trial was registered on February 8, 2021 (NCT04743167).

Recruitment began in March 2021 and was discontinued in September 2024 in accordance with the Ethics Committees after the enrollment of 130 women (55 in the randomized cohort and 75 in the preference cohort), owing to investigators' increasing awareness of a markedly higher likelihood of success in women allocated to IVF. In the randomized cohort, the number of live births observed in the surgical and IVF groups was 7 out of 27 (26%) and 16 out of 28 (57%), respectively (p=0.019). The RR of live birth in women assigned to IVF was 2.20 (95% CI, 1.08-4.50). In the preference cohort, the number of live births was 7 out of 25 (28%) and 32 out of 50 (64%), respectively (p=0.003). The relative risks (RR) of live birth for women assigned to IVF in the randomized and preference cohorts were 2.20 (95% CI: 1.08-4.50) and 2.29 (95% CI: 1.18-4.43), respectively. When data from the two cohorts were combined, the RR remained consistent at 2.29 (95% CI: 1.41-3.70). Per protocol and survival analyses further supported these findings.

In this pragmatic PRPT, IVF was more effective than surgery for the treatment of infertility in women with ultrasound-diagnosed ovarian or deep infiltrating endometriosis.