In patients with pelvic venous disorders secondary to pelvic venous insufficiency (PVI), the optimal imaging modality is ill-defined. Transabdominal ultrasound (TAU) is widely used to identify the presence of iliac vein stenosis. The purpose of the present investigation is to determine the accuracy of TAU for determining the presence of an iliac vein area-reducing lesion compared with intravascular ultrasound (IVUS).


From January to December 2020, a retrospective review of prospectively collected data from 96 patients treated for symptomatic PVI at the Center for Vascular Medicine was performed. All patients had complete history and physical examination findings, demographics, CEAP (clinical, etiologic, anatomic, pathophysiologic), revised venous clinical severity score, and TAU, diagnostic venography, and IVUS measurements recorded in our electronic medical record system. All TAU measurements were performed by the same ultrasound technician with the patient in the supine position. Iliac vein diameters of the common femoral, external iliac, and common iliac veins and the inferior vena cava were obtained. Differences in body habitus were normalized by dividing the minimum diameter measurement of the stenotic vessel with that of the ipsilateral common femoral vein, subtracting this number from 1 and multiplying by 100 (stenosis = [1 – minimal diameter/common femoral diameter] × 100). The normalized stenoses were then compared with the IVUS-derived area reducing measurements. A receiver operating characteristic curve was created, and logistic regression analysis for the probability of predicting an area-reducing lesion of >50% and >60% with TAU was performed. The sensitivity, specificity, and positive and negative predictive values were calculated.


The average age of the entire cohort was 49.8 ± 13.5 years, with 69 women and 27 men. The CEAP distribution was as follows: C0, 5%; C1, 5%; C2, 10%; C3, 40%; C4a,b, 30%; C5, 7%; and C6, 3%. The average revised venous clinical severity score was 6.2 ± 2.6. The indications for intervention were leg symptoms alone in 43%, pelvic symptoms alone in 3%, and combined leg and pelvic symptoms in 54%. TAU identified a stenosis of ≥50% in 92 of the 96 patients (96%). For a ≥50% stenosis, a normalized diameter of ≤3 mm demonstrated a sensitivity, specificity, and positive and negative predictive value of 75%, 75%, 98%, and 12%, respectively. Logistic regression analysis indicated that TAU was significant in predicting the presence of a ≥60% area-reducing lesion (odds ratio, 1.03; 95% confidence interval, 1.01-1.05; P = .009). The area under the receiver operating characteristic curve (c-statistic) was 68.6%. The sensitivity, specificity, and positive and negative predictive values were 66.7%, 66.7%, 81.5%, and 47.6%, respectively, for a normalized diameter of ≥4 mm.


The ability of TAU to identify an iliac vein stenosis of ≥50% is 96%. The positive predictive value for TAU to identify a ≥60% iliac vein area-reducing lesion is high, with moderate sensitivity and specificity. For patients with symptoms consistent with pelvic venous disorders secondary to PVI, TAU is a good preintervention screening modality for properly trained vascular imaging specialists with findings that correlate well with IVUS measurements.