Pitfalls in the use of color-flow duplex ultrasound for screening of suspected arterial injuries in penetrated extremities.

We compared color-flow duplex ultrasonographic scanning (CFD) to arteriography in 67 patients who sustained 75 penetrating injuries to the extremities without obvious arterial injury. There were 72 negative and three (4.0%) positive CFD studies and no failed attempts. With arteriography as the "gold standard," CFD had a specificity of 99% and a sensitivity of 50%, positive and negative predictive values of 66% and 97%, and an accuracy of 96%. Small pseudoaneurysms were missed in an axillary and an aberrant radial artery, and a genicular artery pseudoaneurysm was misread as originating from the popliteal artery. Cautious interpretation of negative studies appears warranted, particularly in the axilla and in bifurcated arteries. Extremity arteries should be completely imaged to rule out aberrant anatomy. Questionable studies should be confirmed arteriographically. With these caveats, color-flow duplex scanning may be useful for screening extremities with penetrating injuries thought to harbor occult arterial injuries.