Clinical evaluation of suspected deep vein thrombosis guides the decision to anticoagulate prophylactically but does not impact the decision to perform after hours duplex venous scanning or increase its yield

Abstract

Objective: The utility of after-hours duplex venous scanning (DVS) for suspected deep vein thrombosis (DVT) in emergency department (ED) patients has been debated. Availability of safe prophylactic low molecular weight heparin, cost containment efforts, and retention of scarce sonographers have to be balanced against 24/7 demand for services. We determined the incidence of DVT in DVS ordered after-hours, correlation between Wells' score and prophylactic anticoagulation as well as urgently performed DVS, and complications of delaying DVS until regular hours.

Methods: Records of all ED encounters between July 1, 2009 and June 30, 2010 associated with a DVS ordered after-hours were reviewed. The decisions to prophylactically anticoagulate and whether to perform DVS urgently or delayed until regular hours were at the discretion of the ED physician and a vascular surgeon. DVS findings, number of urgent and delayed studies, Wells' scores, D-dimers, and outcomes were recorded.

Results: DVT was found in 12% (22) of 181 DVS ordered after-hours. DVT was found in 19% of 42 DVS done urgently and in 10% of 139 DVS delayed an average 10 hours 17 minutes (P = NS). All patients had Wells' scores and 43 had D-dimers. Furthermore, 76% of patients with a Wells' score ≥3 had prophylactic anticoagulation whereas only 39% of patients with a Wells' score <3 had prophylactic anticoagulation (P = .0001). In contrast, 36% of patients with a Wells' score ≥3 had urgent DVS and 20% of patients with a Wells' score <3 had urgent DVS (P = NS). Prophylactic anticoagulation was given to 86% of patients eventually found to have DVT vs 40% of patients eventually found to have no DVT (P < .0001). There were no pulmonary emboli or bleeding complications.

Conclusions: The incidence of DVT in ED patients who had urgent after-hours DVS was no different than in those whose DVS was delayed until regular hours. High pretest probability can be achieved with clinical evaluation prior to DVS, and this guided the decision to prophylactically anticoagulate but did not impact the decision to perform urgent DVS. Most patients eventually found to have DVT did receive prophylactic anticoagulation, and delay of DVS did not result in complications. We believe that most patients in whom there is high clinical suspicion for DVT can safely get prophylactic anticoagulation and delayed DVS. Patients in whom there is low clinical suspicion should not get urgent DVS.