<< An Evidence-Based Approach To Emergency Ultrasound

Ultrasound For Deep Venous Thrombosis

TOC Will Appear Here

Ultrasound For Deep Venous Thrombosis

Ultrasound For Deep Venous Thrombosis

Evaluation of a painful, swollen leg is a common work-up in the ED. The differential diagnosis usually includes deep venous thrombosis (DVT), cellulitis, venous stasis, or some combination of these. Ultrasound is the test of choice to identify or exonerate DVT, and with the increasing availability of ultrasound machines, more emergency clinicians are becoming facile in performing and interpreting this examination at the bedside. (See Figures 8 and 9.)

Review Of The Literature

Many studies have examined the ability of emergency physicians to accurately diagnose DVT on BUS. A 1997 study looked at whether emergency physicians could accurately diagnose DVT using color-flow Doppler after a modest training program and found a sensitivity of 100% and specificity of 75%.64 A recent review by Burnside et al summarized these findings.65 They performed a systematic review of emergency physician-performed ultrasound for the diagnosis and exclusion of DVT. They included original prospective studies of emergency physician-performed ultrasound in symptomatic patients and compared the results with a second venous ultrasound performed by radiology or a vascular lab. Of 1162 publications reviewed, 6 studies met inclusion criteria and relevance screening.66-71 When these 6 studies were combined, 132 cases of DVT were detected in 936 patients. Pooled sensitivities and specificities were 95% (95% CI, 0.87-0.99) and 96% (95% CI, 0.87-0.99), respectively.65 Despite some limitations, the overall pooled sensitivity and specificity are very high, suggesting that properly trained emergency clinicians can diagnose DVT at the bedside using ultrasound with an accuracy approaching that of formal studies.65

Shortly after the Burnside review, Kline et al published a study evaluating the sensitivity and specificity in detecting DVT on ultrasound by a heterogeneous group of providers.72 This was a prospective, single-center study conducted at an urban academic ED in which faculty, residents, and midlevel providers underwent a training course for 3-point compression BUS. The overall sensitivity and specificity were 70% and 89%, respectively.

Time And Cost Savings

In general, it is reasonable to postulate that ultrasonography performed by an emergency clinician at the bedside would almost invariably take less time than obtaining a comprehensive study in radiology or the vascular lab. Studies examining the time and potential costs saved by emergency clinician-performed ultrasound to detect DVT are limited.

In 2004, a prospective, single-blind, observational study by Theodoro et al studied the time savings associated with emergency physician -performed ultrasound studies versus studies performed by radiologists.68 A convenience sample of 156 patients was enrolled. A statistically significant difference (P < 0.0001) existed between the mean time from triage to ED disposition (95 min) and the mean time from triage to radiology disposition (220 min), with excellent agreement between the 2 studies. A similar trend was described, but not explicitly studied, by Blaivas et al.71 In this study, the median scan time for emergency physician- performed ultrasound was 3 minutes, 28 seconds. The longest time required to complete an emergency physician-performed ultrasound study was 18 minutes, 20 seconds, whereas the vascular lab performed a complete duplex ultrasound examination within 8 hours. Although scant, the literature supports what is already intuitive—that BUS is considerably faster than studies performed by radiology or in the vascular lab.

Many radiology departments and vascular labs perform full-length lower-extremity ultrasound and use sonographic techniques including color-flow Doppler, augmentation, and respiratory variation of veins to assess for DVT. Compression alone has also been shown to be a powerful detector of DVT. Patent veins should compress completely when downward pressure is applied with an ultrasound probe. Clots may appear hyperechoic, but since some are anechoic, failure to visualize a clot does not rule out its presence. Regardless of its echogenicity, when a clot is present, the vein does not fully compress.

A study by Biondetti et al compared vein compressibility alone to a gold standard of venography.73 They found that vein compressibility had a sensitivity of 87% and specificity of 100% for detecting DVT. Six of the 7 false-negative results were from isolated distal DVT. The sensitivity of ultrasound for detecting proximal DVT was 98%.

Another variable in scanning technique is performing 2-point compression (femoral vein at the groin only and popliteal vein) versus full-length lower-extremity ultrasound. The obvious advantage of the 2-point compression technique is the shorter time required to perform the examination. But how does it compare in terms of diagnostic accuracy?

A recent study by Bernardi et al prospectively randomized 2098 patients with suspected DVT to serial 2-point ultrasound in combination with a D-dimer test versus whole-leg color-flow Doppler ultrasound.74 One group of patients was randomized to full-length lower-extremity ultrasound, including evaluation of the calf veins. The other group was examined with 2-point ultrasound (common femoral vein and popliteal vein compression only) and a D-dimer test. Those with a negative ultrasound and positive D-dimer underwent repeat ultrasound at 1 week’s time. Those with a negative 2-point ultrasound and negative D-dimer and those with a negative repeat ultrasound at 1 week were not treated with an anticoagulant and were not further investigated. The incidence of proximal DVT on 1-week testing in the group with a positive D-dimer test was 5.5%. The incidence of confirmed symptomatic DVT on 3-month follow-up was 1.2% versus 0.9% in the whole-leg group versus 2-point ultrasound group, respectively. This study supports the idea that 2-point compression in combination with a D-dimer test is equivalent to whole-leg color-flow Doppler ultrasound in the management of symptomatic patients with suspected DVT. The authors also note that the initial prevalence of proximal DVT was similar in both groups (22.1% vs 20.2%), suggesting that proximal DVT always involves the common femoral vein, the popliteal vein, or both. They suggest that superficial and deep femoral veins are usually not worth investigating.

Evaluation Of Calf Veins

The evaluation of distal (calf) veins for DVT is also a controversial topic. The natural history of distal DVT is that in the majority of cases, the thrombi are resorbed and do not propagate, while a small number (0% to 29%)75 do propagate and extend to involve the popliteal vein, thus becoming proximal DVT. Proximal DVTs carry a high risk of embolization, and as such, the use of anticoagulation in patients with isolated distal DVTs is controversial. The goal, then, is to identify patients who have distal DVTs that then develop into proximal DVTs. (See Figures 8 and 9) One common practice pattern is to refer ED patients for a repeat ultrasound in 1 week’s time.

A study by Kearon et al studied outpatients with suspected DVT and negative results on proximal vein ultrasound.76 Patients received a repeat ultrasound in 1 week and then were followed for 6 months. Three of 402 patients (0.7%) were diagnosed with DVT on the 1-week ultrasound. Of the other 399, 5 patients (1.3%) were found to have DVT within the 6-month follow-up period. This study supports the concept that with a negative 1-week ultrasound, the rate of symptomatic DVT during the next 6 months is very low. Care must be taken by the emergency clinician to communicate with the patient’s primary care provider, as many providers seem unaware of the need for a repeat ultrasound.77

Finally, when using ultrasound to evaluate a patient for DVT, is it useful to scan the contralateral leg? Pennell et al sought to answer this question in a 2008 study of 239 patients with unilateral symptoms and a DVT on ultrasound; 47 patients (19.7%) were found to have thrombosis in the (asymptomatic) contralateral leg. Patients with active malignancy carried a 38% incidence of an asymptomatic contralateral clot, so the authors recommend routine scanning of both legs in this population.78


Bedside ultrasound to detect DVT offers excellent sensitivity and specificity when performed by skilled emergency clinicians. Its use in the ED is likely to decrease patients’ length of stay. The 2-point compression technique is sufficient for diagnosing proximal DVT. Evaluation of calf veins for distal DVT is not usually necessary as long as a follow-up ultrasound can be obtained within a week.

Get Quick-Read Evidence-Based Updates
Enter your email to get free evidence-based content delivered to your inbox once per month.
Please provide a valid email address.