Criteria for detecting significant chronic iliac venous obstructions with duplex ultrasound

The purpose of this study was to determine the sonographic criteria for diagnosis of iliac venous outflow obstruction by assessing the correlation of this method with intravascular ultrasound (IVUS) in patients with advanced chronic venous insufficiency (CVI).

The evaluation included 15 patients (30 limbs; age, 49.4 ± 10.7 years; 1 man) with initial CVI symptoms (Clinical class, Etiology, Anatomy, and Pathophysiology [CEAP] classification, CEAP1-2) in group I (GI) and 51 patients (102 limbs; age, 50.53 ± 14.5 years; 6 men) with advanced CVI symptoms (CEAP3-6) in group II (GII). Patients from both groups were matched by gender, age, and race. The Venous Clinical Severity Score was considered. All patients underwent structured interviews and duplex ultrasound (DU) examination, measuring the flow phasicity, the femoral volume flows and velocities, and the velocity and obstruction ratios in the iliac vein. The reflux multisegment score was analyzed in both groups. Three independent observers evaluated individuals in GI. GII patients were submitted to IVUS, in which the area of the impaired venous segments was obtained and compared with the DU results and then grouped into three categories: obstructions <50%, obstructions between 50% and 79%, and obstructions ≥80%.

The predominant clinical severity CEAP class was C1 in 24 of 30 limbs (80%) in GI and C3 in 54 of 102 limbs (52.9%) in GII. Reflux was severe (reflux multisegment score ≥3) in 3 of 30 limbs (10%) in GI and in 45 of 102 limbs (44.1%) in GII (P < .001). There was a moderately high agreement between DU and IVUS findings when they were grouped into three categories (κ = 0.598; P < .001) and high agreement when they were grouped into two categories (obstructions <50% and ≥50%; κ = 0.784; P < .001). The best cutoff points and their correlation with IVUS were 0.9 for the velocity index (r = -0.634; P < .001), 0.7 for the flow index (r = -0.623; P < .001), 0.5 for the obstruction ratio (r = 0.750; P < .001), and 2.5 for the velocity ratio (r = 0.790; P < .001). Absence of flow phasicity was observed in 62.5% of patients with obstructions ≥80%. An ultrasound algorithm was created using the measures and the described cutoff points with accuracy of 86.7% for detecting significant obstructions (≥50%) with high agreement (κ = 0.73; P < .001).

DU presented high agreement with IVUS for detection of obstructions ≥50%. The velocity ratio in obstructions ≥2.5 is the best criterion for detection of significant venous outflow obstructions in iliac veins.