[Focal liver lesions. A comparison between magnetic resonance under base conditions and after a superparamagentic contrast medium].

INTRODUCTION

Superparamagnetic iron oxide (Spio) is a negative contrast material which is phagocytosed by reticuloendothelial cells. It significantly decreases the signal intensity of the organs where it is taken up selectively, especially on T2-weighted images. According to previous reports, it improves the visualization of focal liver lesions with(out) a low content of Kupffer's cells. We investigated the yield of Spio-enhanced MRI in the detection and characterization of focal liver lesions.

MATERIALS AND METHODS

Eighteen patients underwent MRI before and after the administration of Spio particles. T1, PD, T2 SE and T2* GE images were acquired with a .5 T superconductive unit. The Spio dose was .075 mL/kg which was administered i.v. in 35-45 minutes' infusions. Images were acquired about 30-90 minutes after the end of administration. Scanning was repeated in one patient about 8 hours after the end of Spio administration.

RESULTS

Baseline MRI showed hepatocellular carcinoma (HCC) in 10 patients, metastases in 4, 1 dysplastic lesion, 1 single and 1 multiple hemangioma, 1 focal nodular hyperplasia (FNH) and 1 intrahepatic cholangiocarcinoma. No lesion was detected in one patient. Spio-enhanced MRI detected 7 more lesions that baseline MRI, with 11.47% sensitivity improvement. Moreover, Spio-enhanced MRI detected more lesions that baseline MRI in 5/18 patients (27.78%). PD/T2 SE images were the most sensitive ones, followed by T2* GE and finally by T1 SE. Nearly all HCC nodules (17/18), all metastatic lesions and the cholangiocarcinoma did not take up iron oxide and were relatively hyperintense to the liver. However, in one well-differentiated HCC, the nodule partially took up the contrast agent and became isointense, with high-signal borders on PD/T2 SE images. Both the dysplastic lesion and FNH took up iron oxide; the former was isointense to the liver, while the latter appeared slightly hyperintense, with a high-signal central scar. Finally, hemangiomas became hyperintense on Spio-enhanced T1 SE images and had slightly decreased signal intensity on Spio-enhanced T2 SE images.

CONCLUSIONS

In our experience iron oxide was an effective contrast material. Spio-enhanced MRI improved the detection and characterization of focal liver lesions relative to baseline MRI. Several liver-specific contrast agents are now available or in an advanced experimental stage. Therefore, future studies could be aimed at assessing the effectiveness of iron oxide versus other contrast agents and at establishing the role of Spio-enhanced MRI versus spiral CT and CT during arterial portography.