Is color-flow duplex a good diagnostic test for detection of isolated calf vein thrombosis in high-risk patients?

Color-flow duplex scanning (CDS) is a good diagnostic test for lower extremity proximal deep vein thrombosis (DVT). This report aims to evaluate the diagnostic accuracy of CDS in detecting isolated calf DVT in two in-hospital populations. A total of 166 patients had routine DVT testing with both CDS and venography: 99 total joint arthroplasty patients and 67 symptomatic in-hospital patients. Isolated calf DVT was noted in 34% of arthroplasty patients and 12% of symptomatic in-hospital patients. Peroneal DVT was most common. The sensitivity, specificity, positive predictive value, and negative predictive value (with 95% confidence interval [CI]) of CDS in detecting isolated calf DVT in the symptomatic in-hospital group was 39% (16%-62%), 98% (94%-99%), 88% (65%-99%), and 81% (71%-91%), respectively. In the arthroplasty patients these values were 13% (3%-23%), 92% (85%-99%), 60% (30%-90%), and 55% (45%-65%), respectively. CDS has a low sensitivity in detecting isolated calf DVT among hospitalized patients and cannot be deemed an effective tool for identifying clots limited to only one or two tibial vessels.