W1002 Usefulness of Real Time Virtual Sonography for Detection of Small Hepatic Nodules: A Prospective Study

Abstract

Purpose:Ultrasonography(US) has excellent spatial resolution, uses neither X-rays nor magnetic fields, is a portable machine that can be conveniently used at the bedside, and displays lesions in real time. The latter feature is convenient for curative percutaneous procedures such as radiofrequency ablation and percutaneous ethanol injection. However, US is not sensitive for the detection of small nodules in patients with advanced liver disease, and the ability to detect nodules varies depending on the experience and technical skill of the examiner. Real time virtual sonography(RVS) is the first system capable of demonstrating virtual multiplanar reconstructed images corresponding to ultrasonographic images in real time. The purpose of this study was to prospectively evaluate the diagnostic efficacy of RVS for detection of liver malignancy often overlooked by conventional US. Materials and Methods:Sixty-six patients with a presumptive diagnosis of hepatocellular carcinoma were assessed between January 2005 and December 2006 at Shimane University Hospital. All of the patients underwent CT angiography, either during arterial portography or hepatic arteriography. Patients with diffuse invasive hepatoma, 6 or more nodules, and nodules for which US-guided percutaneous therapy (>3 cm lesions) was not indicated were excluded. A final total of 59 patients with 140 nodules were included. After CT angiography, conventional US and RVS were performed. The number of nodules detected by these two methods as well as their characteristics (nodule diameter, location and echo pattern of the nodule, and the echo pattern of the liver parenchyma) were compared to determine which method was more sensitive and which characteristics had a stronger influence on sensitivity. Results:From a total of 140 nodules detected by CT angiography, 71 (50.7%) were detected by both conventional US and RVS, and 46 (32.9%) were detected by RVS but not by conventional US. Twenty-three nodules (16.4%) were not detected by either conventional US or RVS. Univariate analysis demonstrated that nodule size, location, and echo pattern of the nodule and the liver parenchyma, influenced the difference in detection ability between the two methods. Multivariate analysis suggested that the most important factor affecting the difference in detection ability was nodule size. Conclusion:The RVS system is useful for detection of non-hyperechoic nodules with 10 mm or smaller diameter located in the periphery of the liver.