Safety of Excluding Suspected Deep Vein Thrombosis with a Single Whole-Leg Compression Ultrasound: Systematic Review and Meta-Analysis.

Abstract

Abstract 243 Background:In patients with suspected lower extremity deep vein thrombosis (DVT), compression ultrasonography (CUS) is typically used as the initial test to confirm or exclude DVT. Patients with a negative CUS and either a moderate-to-high probability and/or a positive D-dimer usually require repeat CUS testing after 5–7 days to exclude proximal propagation of distal DVT, reducing diagnostic efficiency. Whole-leg compression ultrasound may safely exclude proximal and distal DVT in a single evaluation. Purpose:To assess the safety of withholding anticoagulation in patients presenting with suspected lower extremity DVT following a single negative whole-leg CUS. Methods:MEDLINE, EMBASE, CINAHL, LILACS, Cochrane, Health Technology Assessments databases were searched from January 1970 to April 2009 without language restrictions. This search was supplemented by reviewing Google, Google Scholar, clinicaltrials.gov, meeting abstracts, conference proceedings, reference lists, and by contacting content experts. Study Selection and Data Extraction:Randomized controlled trials and prospective cohort studies of patients with suspected DVT with a negative whole-leg CUS, not treated with anticoagulation, and followed at least 90 days for occurrence of venous thromboembolism (VTE). Studies required objective confirmation of VTE events during follow-up. Two authors independently reviewed articles and extracted data. Results:Six studies were included totaling 4,229 patients with negative whole-leg CUS exams and not receiving anticoagulation. VTE or suspected VTE-related death occurred within the follow-up period in 24 (0.6%) patients. Of these 24 events, 9 (37.5%) were distal DVT, 7 (29.2%) were proximal DVT, 6 (25.0%) were non-fatal pulmonary embolism, and 2 (8.3%) were deaths, possibly related to VTE. Combined VTE event rate at 3 months was 0.46% (95% CI 0.22, 0.70). Limitations:Pretest probability assessment was not available for all analyzed patients. Conclusions:Withholding anticoagulation in patients with suspected DVT based on a single whole-leg CUS is associated with a low risk of VTE (0.46%) during 3 months of follow-up. This strategy represents a safe and efficient alternative to serial CUS testing in patients with suspected DVT. Disclosures:No relevant conflicts of interest to declare.