[3-phase spiral CT of the liver, Value of non-contrast arterial and portal venous studies in the diagnosis of focal liver lesions].

PURPOSE: To assess the value of noncontrast, arterial and portal venous phase of triphasic helical CT in detecting and characterising focal liver lesions. MATERIAL AND METHODS: 120 patients with focal liver disease underwent triphasic helical CT examinations with a collimation of 6.5 mm at a table feed of 6.5 mm/s. The liver scans were obtained before the administration of 120 ml of non-ionic contrast material (flow 2 or 3 ml/s), at the arterial phase, and at the portal venous phase (20 s, respectively 60 s after injection). Patients were divided into four groups according to the underlying disease and enhancement pattern. The studies were evaluated retrospectively. RESULTS: A total of 269 lesions was seen. The noncontrast phase (NCP) revealed 86% of lesions, the arterial phase (AP) 95% and the portal venous phase (PVP) 91%. In the first group of hypovascular lesions (colorectal carcinoma) all lesions (73/73) were detected in the PVP. In the second group of hypervascular lesions (breast cancer, melanoma) the combination of AP and PVP revealed 73 of 74 lesions. In the third group of patients with unknown primary and detected lesions by sonography all 89 lesions were detected with the combination of AP and PVP. In the fourth group of patients with cirrhosis 3 of 33 lesions were detected exclusively during the AP and 3 other lesions exclusively during the NCP. To make a definitive diagnosis of focal liver lesions the value of the three phases was as follows: to characterise lesions the PVP was sufficient in 62%, the combination of PVP and AP in 27%, and the combination of all three phases in 11%. CONCLUSIONS: If hypovascular lesions are suspected examination during PVP is sufficient. In cases of hypervascular lesions and lesions of unknown primary AP and PVP should be combined. Unenhanced scans are of additional diagnostic value only in patients with liver cirrhosis.