Reflux in the ovarian veins, with or without an obstructive venous outflow component, is reported to be the primary cause of pelvic venous insufficiency (PVI). The degree to which venous outflow obstruction plays a role in PVI is currently ill-defined.


We retrospectively reviewed the charts of 227 women with PVI who presented to the Center for Vascular Medicine from January 2012 to September 2015. Assessments and interventions consisted of an evaluation for other causes of chronic pelvic pain by a gynecologist; preintervention and postintervention visual analog scale (VAS) pain score; complete venous duplex ultrasound examination; and Clinical, Etiology, Anatomy, and Pathophysiology classification. All patients underwent diagnostic venography of their pelvic and left ovarian veins as well as intravascular ultrasound of their iliac veins. Patients were treated in one of six ways: ovarian vein embolization (OVE) alone (chemical ± coils), OVE with staged iliac vein stenting, OVE with simultaneous iliac vein stenting, iliac vein stenting alone, OVE with venoplasty, and venoplasty alone.


Of the 227 women treated, the average age and number of pregnancies was 46.4 ± 10.4 years and 3.36 ± 1.99, respectively. Treatment distribution was the following: OVE, n = 39; OVE with staged stenting, n = 94; OVE with simultaneous stenting, n = 33; stenting alone, n = 50; OVE with venoplasty, n = 8; and venoplasty alone, n = 3. Seven patients in the OVE and stenting groups (staged) and one patient in the OVE + venoplasty group required a second embolization of the left ovarian vein. Eighty percent (181/227) of patients demonstrated an iliac stenosis >50% by intravascular ultrasound. Average VAS scores for the entire cohort before and after intervention were 8.45 ± 1.11 and 1.86 ± 1.61 (P ≤ .001). In the staged group, only 9 of 94 patients reported a decrease in the VAS score with OVE alone. VAS score decreased from 8.6 ± 0.89 before OVE to 7.97 ± 2.10 after OVE. After the planned staged stenting, VAS score decreased to 1.33 ± 2.33 (P ≤ .001). Similarly, in the simultaneous group, preintervention scores were 8.63 ± 1.07 and decreased to 2.36 ± 2.67 after OVE + stenting (P ≤ .001).


The majority of patients in our series (80%) demonstrated a significant iliac vein stenosis. These observations indicate that the incidence of iliac vein outflow obstruction in PVI is greater than previously reported. In patients with combined ovarian vein reflux and iliac vein outflow obstruction, our data suggest that pelvic venous outflow lesions should be treated first and that ovarian vein reflux should be treated only if symptoms persist. In women with an outflow lesion, ovarian vein reflux, and a large pelvic reservoir, we recommend simultaneous treatment.