A randomized trial of diagnostic strategies after normal proximal vein ultrasonography for suspected deep venous thrombosis: D-dimer testing compared with repeated ultrasonography

Background: With suspected deep venous thrombosis and normal results on proximal vein ultrasonography, a negative (D)-climer result may exclude thrombosis and a positive (D)-dimer result may be an indication for venography. Objective: To evaluate and compare the safety of 2 diagnostic strategies for deep venous thrombosis. Design: Randomized, multicenter trial. Setting: Four university hospitals. Patients: 810 outpatients with suspected deep venous thrombosis and negative results on proximal vein ultrasonography. Interventions: Erythrocyte agglutination (D)-dimer testing followed by no further testing if the result was negative and venography if the result was positive (experimental) or ultrasonography repeated after 1 week in all patients (control). Measurements: Symptomatic deep venous thrombosis diagnosed initially and symptomatic venous thromboembolism during 6 months of follow-up. Results: Nineteen of 408 patients (4.7%) in the (D)-dimer group and 3 of 402 patients (0.7%) in the repeated ultrasonography group initially received a diagnosis of deep venous thrombosis (P < 0.001). During follow-up of patients without a diagnosis of deep venous thrombosis on initial testing, 8 patients (2.1% (95% Cl, 0.9% to 4.0%]) in the (D)-dimer group and 5 patients (1.3% [Cl, 0.4% to 2.9%1) in the repeated ultrasonography group developed symptomatic venous thromboembolism (difference, 0.8 percentage point [CI, -1.1 to 2.9 percentage points]; P > 0.2). Venous thromboembolism occurred in 1.0% (Cl, 0.2% to 2.8%) of those with a negative (D)-dimer result. Limitations: seventy patients (8.6%) deviated from the diagnostic protocols, and 9 patients (1.1 %) had inadequate follow-up. Conclusion: In outpatients with suspected deep venous thrombosis who initially had normal results on ultrasonography of the proximal veins, a strategy based on (D)-dimer testing followed by no further testing if the result was negative and venography if the result was positive had acceptable safety and did not differ from the safety of a strategy based on withholding anticoagulant therapy and routinely repeating ultrasonography after 1 week.