In the treatment of infertility in women with ovarian or deep infiltrating endometriosis, in-vitro fertilization (IVF) is associated with better chances of achieving a live birth compared with endometriosis surgery, according to a pragmatic, partially randomized patient preference trial.
The primary outcome of a live birth within 12 months of randomization occurred more frequently in the IVF group than in the surgery group both in the randomized cohort (57 percent vs 26 percent; p=0.019) and the preference cohort (64 percent vs 28 percent; p=0.003). [Am J Obstet Gynecol 2026;doi:10.1016/j.ajog.2026.10.024]
Compared with endometriosis surgery, IVF was associated with a more than twofold higher likelihood of achieving a live birth in the randomized cohort (relative risk [RR], 2.20, 95 percent confidence interval [CI], 1.08–4.50) and the preference cohort (RR, 2.29, 95 percent CI, 1.18–4.43). Combined data from the two cohorts yielded a consistent result (IVF vs surgery: RR, 2.29, 95 percent CI, 1.41–3.70).
Results of the per-protocol and survival analyses further supported the findings.
First author Dr Laura Benaglia from the Policlinico of Milan, Milan, Italy, and colleagues cautioned that the marked advantage of IVF over surgery observed in the analyses “should not be interpreted as a justification for systematically favouring IVF in all women with endometriosis-related infertility.”
The findings “do not provide a universal answer” but contribute to “more evidence-based and informed counselling,” Benaglia and colleagues continued.
The authors advised clinicians to inform their patients about the risks of each intervention—the potential damage to ovarian reserve and the higher risk of major complications in cases of deep infiltrating disease with endometrial surgery, as well as possibility of pelvic infections, progression of deep peritoneal lesions, and spontaneous hemoperitoneum during pregnancy with IVF.
“Treatment decisions should therefore be individualized and shared, taking into account prior surgical history, pain symptoms, response to medical therapy, age, ovarian reserve, semen analysis, economical constraints, and patient preferences,” Benaglia and colleagues said.
The trial included 130 infertile women with ultrasound-diagnosed ovarian and deep infiltrating endometriosis. These participants were randomly assigned to undergo IVF or endometriosis surgery. Those who had a strong preference and declined randomization were enrolled in a parallel patient preference cohort. Participants in the surgery group who were dissatisfied with their predicted prognosis based on the Endometriosis Fertility Index score were allowed to shift to IVF earlier, and those allocated to IVF could be operated on if complications occurred, if symptoms significantly worsened, or if they requested for it.
A total of 55 women were included in the randomized cohort (28 in the IVF group and 27 in the surgery group), and 75 were included in the preference cohort (50 in the IVF group and 25 in the surgery group). Women in the IVF group received up to three complete IVF cycles, whereas those in the surgery group attempted natural conception.
The intention-to-treat population included 78 participants in the IVF group and 52 in the surgery group overall. The median age of the participants was 34 years, with between 15 and 18 months of infertility duration. The median BMI was 21–22 kg/m2.