Utility of venous Ultrasound to Diagnose Deep venous Thrombosis when the Primary Indication is Pulmonary Embolus

Abstract

Introduction

With increasing frequency, physicians are requesting a duplex ultrasound (US) to detect lower extremity deep venous thrombosis (DVT) in the face of suspected pulmonary embolus (PE), even when leg symptoms are absent. The objective of this study was to determine the diagnostic accuracy and the prevalence of DVT detected by duplex ultrasonography in patients with suspected PE in the presence or absence of leg symptoms.

Methods

Medical records for the calendar years 2000 and 2001 were examined from inpatients referred for lower extremity venous US. Basic demographic information, comorbidities, and indications for the test were retrieved from the records. Subjects were categorized into groups on the basis of risk factors and/or symptoms of DVT. For positive studies, the location of the DVT was also recorded.

Results

Of the 740 patients, 143 patients were suspected of having PE, whereas the remaining 597 subjects were suspected of having a DVT alone. Among subjects with suspected PE, 2 (2%) of the 110 patients without leg symptoms/risk factors were positive for DVT, whereas 15 (45%) of the 33 patients with symptoms/risk factors were positive for DVT (p < 0.001). Among patients with questionable DVT, none of 24 asymptomatic patients were positive for DVT, whereas 65 (11%) of the 573 symptomatic patients were positive for DVT (p = 0.10).

Conclusions

Lower extremity duplex US to diagnose DVT in the absence of leg symptoms when PE is suspected is not an effective strategy. If US is contemplated as the initial examination in patients suspected of having a PE, it should be used in those patients who have suggestive symptoms in the limb or medical contraindications to contrast injection.