Interobserver agreement of complete compression ultrasound for clinically suspected deep vein thrombosis.

The interobserver variability of compression ultrasound of proximal and distal veins in clinically suspected deep vein thrombosis was assessed. One hundred one symptomatic legs of all patients referred for clinically suspected deep vein thrombosis on 21 consecutive workdays were examined by two investigators independently according to a standardized protocol of complete compression ultrasound (CCUS) with 28 predefined venous segments between groin and ankle. Incompressible vein segments were defined as thrombotic. Cohen's kappa coefficient was used to calculate interobserver variability regarding diagnosis of deep vein thrombosis. Kappa for entire lower extremity was 0.94 (95% CI, 0.87-1). Kappa for proximal veins was 1; for calf veins 0.9 (95% CI, 0.79-1). For the posterior tibial veins and peroneal veins, kappa was 0.84 (95% CI, 0.66-1) and 0.77 (95% CI, 0.59-0.94), respectively. The results show that almost complete interobserver agreement can be achieved in compression ultrasound of both proximal and distal deep veins conducted according to a standardized examination protocol in clinically suspected deep vein thrombosis.