[Portal contrast medium-enhanced spiral computed tomography of the liver--the correlation of radiological and intraoperative findings and the evaluation of resectability].

PURPOSE: To evaluate the accuracy of spiral computed tomography during arterial portography (SCTAP) in the detection, localization, and resectability of liver tumors in a correlative study between radiology and intraoperative findings. METHOD AND MATERIALS: Retrospectively, SCTAP images of 168 consecutive patients before liver tumor resection were analyzed. The SCTAP studies (100 ml lopromid 300 by automated injector with a flow of 3 ml/s; slice thickness, table feed and reconstruction index 5 mm each; scan-delay 30 s; 120 kV; 250 mAs) were evaluated for the detection, localization, and resectability of focal liver lesions by three experienced radiologists in consensus and were correlated with histopathological and intraoperative findings where available (59/168). RESULTS: The sensitivity of SCTAP for the detection of liver tumors was 91% for all lesions and 84% for lesions < 1 cm. The specificity was only 19% due to a high rate of false-positive lesions (30%) and preselection effects. Typical pitfalls in false positive lesions were inhomogeneous liver perfusion near the portal vein, the falciform ligament, and the gallbladder (19/42). In 30% of the patients SCTAP correctly diagnosed inoperability, in 23% the intraoperative tumor expansion was larger than expected from SCTAP images, which would have changed operability. CONCLUSION: The SCTAP has a high sensitivity in the detection and localization of liver tumors and is a valuable method in the preoperative diagnostic procedure. The method is limited by many false-positive lesions often due to inhomogeneous liver perfusion and the insufficient evaluation of local tumor spread. Therefore, SCTAP should be replaced by MRI in the near future.