Pitfalls in lower extremity venous duplex scanning.

Results of scans performed on 1074 patients over an 18-month period were evaluated to define the limitations of lower extremity venous duplex scanning. Eighty-four patients had confirmatory phlebography performed within 24 hours of their venous duplex scanning. In 71 patients scans were considered diagnostic (sensitivity 91%, specificity 95%). Eighteen studies (13 equivocal, 5 misinterpretations) were scrutinized to determine the limitations of venous duplex scanning compared to phlebography and are the focus of this analysis. Seven patients had phlebograms documenting only infrapopliteal thrombus, seven had normal phlebographic findings, and four had findings consistent with chronic thrombosis. In the seven patients with infrapopliteal thrombus, four had normal imaging outcomes but abnormal Doppler flow patterns, whereas three had both normal imaging results and flow patterns. All four patients with chronic thrombosis had identifiable thrombus and abnormal flow patterns by venous duplex scanning, but in each case thrombus age was indeterminate. Of the seven patients with normal phlebographic results, five had incompressible segments of the superficial femoral vein on imaging, one had abnormal Doppler flow without visualized thrombus and without apparent reason, one had venous duplex scanning visualized thrombus with a normal outcome on phlebography. These data suggest that the diagnostic yield of lower extremity venous duplex scanning may be improved by (1) meticulous infrapopliteal vein examination, (2) better estimation of the age of the thrombotic process, and (3) recognizing segmental incompressibility of the superficial femoral vein within the adductor canal as a normal finding especially in the absence of abnormal Doppler flow or imaged thrombus.