Diagnostic strategies for the management of patients with clinically suspected deep-vein thrombosis

Abstract

Given the perceived inaccuracy of clinical diagnosis, patients with suspected deep vein thrombosis should have objective testing. Due to the inherent limitations of the reference method (contrast venography), several diagnostic strategies using noninvasive tests have been developed. These strategies share two components: anticoagulant therapy is initiated only in patients with an abnormal test, and serial testing is performed in patients with an initial normal test result. A thorough search of the literature was done to identify all studies that have evaluated the feasibility, accuracy, and safety of diagnostic strategies in patients with clinically suspected deep vein thrombosis. The safety of the individual diagnostic strategies was expressed as the total rate of venous thromboembolic complications. Feasibility was expressed as the mean number per patient of extra visits to the hospital and additional tests per patient. A total of 12 reports qualified for the analysis. The diagnostic strategies included venography, serial impedance plethysmography with and without 125I-fibrinogen leg scanning, serial ultrasound imaging with and without D-dimer determination, serial ultrasound imaging in combination with a clinical score, and a diagnostic work-up including ultrasound imaging, impedance plethysmography, D-dimer determination, and a clinical score. The observed venous thromboembolic complication rates varied between 0.4% and 2.6%. Feasibility was lowest for the initial serial impedance plethysmography strategy (mean number of extra hospital visits and mean number of additional tests, 4.1 per patient). Strategies that used the D-dimer test complimentary to ultrasound imaging or the combination of impedance plethysmography and a clinical score performed best (mean number of extra hospital visits and mean number of additional tests, approximately 0.3 per patient). All available noninvasive diagnostic strategies are as accurate and safe as contrast venography for the treatment of patients with clinically suspected deep vein thrombosis. The recently introduced simplified diagnostic strategies allow treatment decisions to be made on the day of presentation in most patients.