Power Doppler sonography assessment of tumor recurrence after chemoembolization therapy for hepatocellular carcinoma.

OBJECTIVE, This study was undertaken to determine the value of power Doppler sonography in the evaluation of recurrent hepatocellular carcinomas after transcatheter arterial chemoembolization therapy. SUBJECTS AND METHODS. Forty-five patients (age range, 45-81 years; mean age, 61 years) with hepatocellular carcinomas prospectively underwent power Doppler sonography, helical CT, and intraarterial digital subtraction angiography before and after transcatheter arterial chemoembolization therapy to evaluate for tumor recurrence. Three to 6 months after transcatheter arterial chemoembolization therapy, the Doppler signal; its location, and shape were evaluated. The results were compared with tumor vascularity as determined on helical CT and intraarterial digital subtraction angiography, which were used as the gold standards (n = 142). RESULTS. A sonographic signal was seen in 80 of 142 lesions with power Doppler sonography. Flow signal in lesions tended to be more difficult to detect in the left lobe (sensitivity, 74%) than in the right lobe (sensitivity, 93%). The location of the color signal was in the tumor's center, on its periphery, or both. No correlation between tumor recurrence and the location of a signal within a tumor was found. Power Doppler sonography had a sensitivity of 87%, a specificity of 85%, and an accuracy of 86% in revealing tumor recurrence after transcatheter arterial chemoembolization therapy. All lesions that showed linear signals in the center or at the periphery of the tumors proved to be recurring tumors (n = 20). In lesions with spotty signals, tumor recurrence was not revealed in nine of 60 lesions. CONCLUSION. Power Doppler sonography can be used for follow-up studies after transcatheter arterial chemoembolization therapy as a sensitive and cost-effective imaging technique. Although spotty signals were frequently seen, linear signals appeared to be specific for tumor recurrence.