Can we rely upon venous ultrasonography to detect DVT?
Melkon Hacobian, MD
Provided by NATF
Posted on ClotCare January, 2008*
Reference: Schellong SM, Beyer J, Kakkar AK, Halbritter K, Eriksson BI, Turpie AG, Misselwitz F, Kälebo P. Ultrasound Screening for Asymptomatic Deep Vein Thrombosis After Major Orthopaedic Surgery: the VENUS Study. J Thromb Haemost. 2007 Jul;5(7):1431-7.
Invasive contrast venography is the diagnostic standard for assessing the presence or absence of deep vein thrombosis in clinical studies. The limitations of this procedure including inter-observer variable interpretation, radiation burden, exposure to contrast agents, and the cumbersome and occasionally painful nature of the procedure itself. These limitations have spurred increased interest in finding reliable noninvasive alternatives to this valuable diagnostic test.
Venous ultrasound as a noninvasive, easily repeatable technique without any contraindications has become the most widely acceptable and available test for the detection of DVT. But failure to demonstrate sufficient accuracy compared with invasive venography and logistical and technical difficulties with central adjudication have been limitations of this test. Using complete compression ultrasound (CCUS) by standardizing the procedure and extending it to the entire venous system of the leg addresses these limitations and has been shown to result in excellent inter-observer agreement for the detection of DVT in proximal and distal veins. On the other hand, the high degree of standardization of the protocol and the possibility of generation a video document for shipment and central review may overcome the difficulties of previous less standardized ultrasound protocols to detect DVT in asymptomatic patients.
The VENUS study was designed to validate centrally adjudicated CCUS against venography in a large population of patients after elective hip or knee replacement surgery. This was a sub study of two dose finding trials comparing oral rivaroxaban (a novel direct factor Xa inhibitor) with subcutaneous enoxaparin administered postoperatively. The mandatory venography was performed according to a standardized procedure 7 ± 2 days after the surgery, and the CCUS was performed within 24 hours of venography by the sonographers blinded to the venograms results. Central committees, who were also blinded to the results, adjudicated these studies independently.
This was the largest study to compare these two diagnostic techniques in a patient population frequently included in anticoagulant drug trials. Although the inter-observer agreement for centrally adjudicated venography was excellent, it was poor for ultrasound. This could be the result of the difference in the levels of adjudication experience of those reading venograms and sonograms.
This study demonstrated a high specificity for CCUS consistent with previous ultrasound validation studies using venography as the gold standard, since the DVT diagnosis had to be confirmed by the agreement of all three senior readers in the adjudication process.
On the other hand, the sensitivity of centrally adjudicated CCUS was found to be poor, between 20% and 30%, falling in the lowest range of rates achieved in studies reported previously. Contributing factors were thought to be the documentation and examination protocol itself and the performance of the sonographers.
In conclusion, centrally adjudicated CCUS is not a reliable test to detect asymptomatic DVT in patients early after hip and knee replacement. Ultrasound will not be able to replace venography as an endpoint measure of the actual rate of DVT, because ultrasound drastically underestimates the DVT rate in this setting. This finding might be problematic in proof-of-concept phase II as well as in confirmatory phase III drug trials. Future studies will be required to evaluate if better results can be obtained with a modified and/or more extensive ultrasound examination and documentation protocol or in different patient populations in whom the physical nature of a thrombus may be different.
Dr. Hacobian completed his internship and residency in Intenral Medicine at UMass Memorial Medical Center. In July of 2008, Dr. Hacobian will begin his Cardiology fellowship at the Maine Medical Center in Portland, ME. As a Venous Thromboembolism Research Group fellow, he is currently involved with investigator initiated and multicenter trials at Brigham and Women's Hospital His current research interests include Warfarin Pharmacogenomics and novel anticoagulant treatment strategies for venous Thromboembolism.
This posting originally appeared on the website of the North American Thrombosis Forum (NATF) and has been provided on ClotCare with NATF's permission. See www.natfonline.org for more information about NATF.