Despite Increased Miscarriage History With Uterine Anomalies, Hysteroscopic Metroplasty Resulted in Identical IVF Miscarriage, Live Birth, and Cumulative Live Birth Rates When Compared to Controls

INTRODUCTION: Septate uterus is a congenital anomaly affecting 1.4–2.3% of all women but can be in up to 55% of patients needing fertility treatment. All septate and T- or Y-shaped defects are quickly and safely repaired with hysteroscopy. However, the indication for hysteroscopic metroplasty remains controversial. OBJECTIVE: To assess miscarriage (MR), live birth (LBR), and cumulative live birth rates (CLBR) in patients undergoing IVF following hysteroscopic metroplasty for congenital anomalies compared to controls with normal baseline sonohysterogram (SIS). METHODS: This is a single-institution-surgeon, retrospective case-control study of all IVF cycles between 2016 and 2023 with at least one embryo transfer (ET). Cases included all patients who had a septate, T-, or Y-shaped uterine anomaly and underwent hysteroscopic metroplasty until there was no significant remnant septum (<0.5 cm). Septate anomalies were sub-categorized by extent of the septum from the fundus to the internal cervical os or evidence of side wall invagination resulting in a hypoplastic T- or Y-shaped cavity. A mild septum was defined as <20% extension, a moderate septum was 20–50%, and a severe or complete septum was >50% extension. Controls included all patients who had a normal uterus on initial SIS, including 3D imaging. The primary outcome was LBR. Secondary outcomes included MR, CLBR, and the number of hysteroscopic procedures required to correct the anomaly prior to ET. Data analysis was performed using R software. RESULTS: 112 patients (214 cycles) had a septate or T- or Y-shaped congenital anomaly, while 238 patients (428 cycles) had a normal uterus at initial evaluation and were assigned to the control group. Prior live birth was significantly lower in the anomaly group than the controls (0.24 vs 0.56, p=0.00007), but gravidity was not significantly different (1.08 vs 1.21, p=0.4). Prior miscarriage trended more often in the anomaly group (0.63 vs 0.48, p=0.16). After successful repair of the uterine anomaly, LBR (46.3% vs 48.4%, p=0.68), MR (10.8% vs 13.8%, p=0.55), and CLBR (72.3% vs 70.5%, p=0.83) were nearly identical between cases and controls. In cycles with elective euploid single-embryo transfers, there was no difference in LBR (45.6% vs 52.9%, p=0.56) and MR (9.7% vs 9.1%, p=1). LBR and MR were not significantly different among anomaly subgroups following complete repair (p=0.48 and p=0.09, respectively), and patients underwent a similar number of procedures to correct the anomaly (Table 1). CONCLUSIONS: Patients with uterine anomalies had lower parity and more miscarriages before repair when compared to controls without intrauterine defects. After hysteroscopic metroplasty, LBR, MR, and CLBR in these patients were nearly identical to normal controls. LBR and MR were not significantly different among anomaly classifications, although LBR trended lower with larger septum and T- or Y-shaped anomalies and MR trended higher in moderate and severe septa. Overall, our results strongly suggest a benefit of septum resection and metroplasty prior to IVF-ET.Table 1