Iliac Vein Stenting With or Without Ovarian Vein Embolisation for May-Thurner Syndrome-Associated Pelvic Venous Disorder in Women: A Received-Treatment Retrospective Cohort Study.

Abstract

Objective: To compare clinical outcomes after iliac vein stenting (IVS) alone, simultaneous IVS plus ovarian vein embolisation (OVE), and staged IVS followed by OVE in women with May-Thurner syndrome-associated pelvic venous disorder.

Methods: This single-centre retrospective cohort study enrolled 184 women (aged 18-65 years) with imaging-confirmed MTS and haemodynamically significant PeVD at Fujian Maternity and Child Health Hospital (January 2021-May 2025). The primary estimand was treatment received: IVS alone without subsequent OVE (Group A, n=71), simultaneous IVS plus OVE (Group B, n=54), or staged IVS then OVE (Group C, n=59, including 45 planned staged cases and 14 staged-after-IVS reassessment cases). The primary outcome was >=50% reduction in VAS pain score at 12 months. Confounding was addressed using multivariable regression, propensity score matching (PSM), stabilised inverse probability weighting (IPTW), overlap weighting, augmented inverse probability weighting (AIPW), and repeated-measures models. OVE timing analyses were exploratory. Exact-date and time-updated OVE sensitivity analyses were performed, and available follow-up was extended through August 31, 2026.

Results: At 12 months, response rates were 54.9% for IVS alone, 74.1% for simultaneous IVS+OVE, and 69.5% for staged IVS+OVE (overall P=0.059). Receipt of OVE was associated with higher adjusted odds of response compared with IVS alone (adjusted OR 2.64, 95% CI 1.21-5.73; P=0.014), supported by IPTW (OR 3.12, bootstrap 95% CI 1.09-6.87). PSM (OR 2.60, 95% CI 0.93-7.29), overlap-weighted (OR 2.42, 95% CI 0.66-8.84), and AIPW estimates were directionally consistent but imprecise. Planned staged and staged-after-IVS reassessment cases had 12-month response rates of 73.3% and 57.1%, respectively. Recorded 30-day events were mostly Grade I; one simultaneous IVS+OVE patient had Grade IIIa coil migration requiring snare retrieval, with no sequelae. Median follow-up was 37.3 months (IQR, 23.7-54.2; maximum, 66.5), with 96.6% stent patency among 29 patients assessed at 60 months.

Conclusions: In this received-treatment analysis, adjunctive OVE was associated with higher 12-month symptom-response rates compared with IVS alone. No statistically significant difference was observed between simultaneous and staged OVE, but this exploratory timing comparison should not be interpreted as equivalence. Because residual confounding remains and OVE adds procedure-specific events, these findings support selective consideration of OVE for patients with substantial ovarian vein reflux rather than routine incorporation for all patients.