Compression Ultrasound for DVT Diagnosis

To the Editor:

I read with interest the article by Drs Wuhl and Graham on the diagnosis and treatment of pulmonary embolism (February 2005). When discussing the diagnosis of deep vein thrombosis (DVT), the authors comment on page 18 that "a single negative IPG or Doppler study is insufficient to rule out DVT." To date, there have been 3 large, well-conducted trials investigating the safety and feasibility of a single, comprehensive (ie, imaging the entire leg) compression ultrasound (CUS) to rule out DVT.^1-3 All 3 studies demonstrated that in centers where high-quality ultrasound equipment is available and the operators well trained, a single CUS that visualizes the entire leg is not inferior to repeated, limited CUS. The 3-month failure rates (ie, the percentage of patients who tested negative for DVT but developed a thromboembolic event) in these studies were 0.3% to 0.8%?rates that are at least as good as the 3-month failure rates reported with other strategies. Because 2 of these studies did not enroll a large number of patients with a high pretest probability of DVT,^2,3 the use of alternative strategies in these patients is recommended. Thus, in outpatients with a low or intermediate pretest probability of DVT, a single comprehensive CUS using high-quality equipment and experienced operators is a safe strategy for DVT diagnosis.

Matthew T. Rondina, MD University of Utah Health Sciences Center, Salt Lake City

Ann Intern Med.

1. Stevens SM, Elliott CG, Chan KJ, et al. Withholding anticoagulation after a negative result on duplex ultrasonography for suspected symptomatic deep venous thrombosis. 2004;140:985-991.

Thromb Haemost.

2. Elias A, Mallard L, Elias M, et al. A single complete ultrasound investigation of the venous network for the diagnostic management of patients with a clinically suspected first episode of deep venous thrombosis of the lower limbs. 2003;89:221-227.

Thromb Haemost.

3. Schellong SM, Schwarz T, Halbritter K, et al. Complete compression ultrasonography of the leg veins as a single test for the diagnosis of deep vein thrombosis. 2003;89:228-234.

The Authors Reply:

We appreciate Dr Rondina's comments about the reliability of CUS to diagnose DVT. It is encouraging that the false-negative rate of CUS is lower in the studies he cites than has been recognized (30%) for many years; however, we do not think that these data are sufficient to change the standard of care.

Our comments were limited to the specific clinical situation in which clinical suspicion for pulmonary embolism is high and the ventilation/perfusion scan results are equivocal. The 2 studies that included insufficient numbers of patients with a high pretest probability of DVT are therefore inapplicable. Also, the unusually low false-negative rates are limited to "centers where high-quality ultrasound equipment is available and the operators well-trained."

We all look forward to the day in which we can safely rely on 1 good sonographic study to exclude the likelihood of DVT, but we do not think that day has arrived.

Mark G. Graham, MD, FACP

Jefferson Medical College, Philadelphia