Detection of Small Hypervascular Hepatocellular Carcinomas in Cirrhotic Patients: Comparison of Superparamagnetic Iron Oxide-Enhanced MR Imaging with Dual-Phase Spiral CT

OBJECTIVE: To compare the performance of superparamagnetic iron oxide (SPIO) -enhanced magnetic resonance (MR) imaging at 1.5T and dual-phase spiral computed tomography (CT) for the depiction of small hypervascular hepatocellular carcinomas (HCCs). MATERIALS AND METHODS: Forty-three patients with 70 small nodular HCCs (5-20 mm; mean, 13.7 mm) were examined. Diagnosis was based on the results of surgical biopsy in 22 patients and by the combined assessment of MR imaging, lipiodol CT, alpha feto-protein levels, and angiographic findings in 21. MR imaging consisted of respiratory-triggered turbo spin-echo T2-weighted imaging, T1-weighted fast low-angle shot, and T2* -weighted fast imaging with steady-state precession imaging before and after SPIO enhancement. CT imaging was performed with 5-mm collimation and 1: 1.4 pitch, and began 30 and 65 secs after the injection of 150 mL of contrast medium at a rate of 3 mL/sec. Two blinded observers reviewed all images independently on a segment-by-segment basis. Diagnostic accuracy was evaluated using receiver operating characteristics (ROC) analysis. RESULT: The mean areas (Az) under the ROC curves were 0.85 for SPIOenhanced MR imaging and 0.79 for dual-phase spiral CT (p .05). CONCLUSION: SPIO-enhanced MR imaging is more sensitive than dual-phase spiral CT for the depiction of small hypervascular hepatocellular carcinomas.

The Significance of Perfusion Defect at Myocardial Perfusion MR Imaging in a Cat Model of Acute Reperfused Myocardial Infarction

OBJECTIVE: To determine whether the size of a perfusion defect seen at myocardial perfusion MR imaging represents the extent of irreversibly damaged myocardium in acute reperfused myocardial infarction. MATERIALS AND METHODS: In nine cats, reperfused myocardial infarction was induced by occlusion of the left anterior descending coronary artery for 90 minutes and subsequent reperfusion for 90 minutes. At single-slice myocardial perfusion MR imaging at the midventricular level using a turbo-FLASH sequence, 60 short-axis images were sequentially obtained with every heart beat after bolus injection of gadomer-17. The size of the perfusion defect was measured and compared with both the corresponding unstained area seen at triphenyl tetrazolium chloride (TTC) staining and the hyperenhanced area seen at gadophrin-2-enhanced MR imaging performed in the same cat six hours after myocardial perfusion MR imaging. RESULTS: The sizes of perfusion defects seen at gadomer-17-enhanced perfusion MR imaging, unstained areas at TTC staining, and hyperenhanced areas at gadophrin-2-enhanced MR imaging were 20.4+/-4.3%, 29.0+/-9.7%, and 30.7+/-10.6% of the left ventricular myocardium, respectively. The perfusion defects seen at myocardial perfusion MR imaging were significantly smaller than the unstained areas at TTC staining and hyperenhanced areas at gadophrin-2-enhanced MR imaging (p < .01). The sizes of both the perfusion defect at myocardial perfusion MR imaging and the hyperenhanced area at gadophrin-2- enhanced MR imaging correlated well with the sizes of unstained areas at TTC staining (r = .64, p = .062 and r = .70, p = .035, respectively). CONCLUSION: In this cat model, the perfusion defect revealed by myocardial perfusion MR imaging underestimated the true size of acute reperfused myocardial infarction. The defect may represent a more severely damaged area of infarction and probably has prognostic significance.

Gadobenate Dimeglumine-enhanced MR of VX2 Carcinoma in Rabbit Liver: Usefulness of the Delayed Phase Imaging and Optimal Pulse Sequence

PURPOSE: To assess the diagnostic value of delayed imaging using gadobenate dimeglumine(MultiHance) and to determine the optimal pulse sequence for the detection of VX2 carcinoma lesions in the rabbit. MATERIALS AND METHODS: Twelve VX2 carcinomas implanted in the livers of eleven New Zealand rabbits were studied. All patients underwent an MR protocol consisting of precontrast T2-and T1-weighted sequences, followed by repetition of the T1-weighted sequence at 0 to 30 (arterial phase), 31-60 (portal phase), and 40 minutes (delayed phase) after the intravenous administration of 0.1 mmol/kg of gadobenate dimeglumine. The signal-to-noise ratio (SNR) of the liver and VX2 tumor, and the lesion-to-liver contrast-to-noise ratio (CNR) of precontrast and postcontrast MR images were quantitativlely analyzed, and two experienced radiologists evaluated image quality in terms of lesion conspicuity, artifact, mass delineation, and vascular anatomy. RESULTS: Liver SNR was significantly higher at delayed imaging than at precontrast, arterial, and portal imaging (p<0.05), while lesion SNR was significantly higher at delayed imaging than at precontrast imaging (p<0.05). Lesion CNR was higher at delayed imaging than at precontrast and portal phase imaging (p<0.05), but there was no difference between arterial and delayed imaging. The latter provided better mass delineation than precontrast, arterial and portal phase imaging (p<0.05). While in terms of lesion conspicuity and vascular anatomy, the delayed phase was better than the arterial phase (p<0.05) but similar to the precontrast and portal phase. During the delayed phase, the gradient-echo sequence showed better results than the spin-echo in terms of liver SNR, and lesion SNR and CNR (p<0.05). CONCLUSION: Because it provides better lesion conspicuity and mass delineation by improving liver SNR and lesion-to-liver CNR, the addition of the delayed phase to a dynamic MRI sequence after gadobenate dimeglumine administration facilitates lesion detection. For delayed-phase imaging, the gradient-echo sequence is superior to the spin-echo sequence.

Hemangioma and Hepatocellular Carcinoma: Distinction with Superparamagnetic Iron Oxide-Enhanced MR Imaging

PURPOSE: To compare liver hemangioma with hepatocellular carcinoma (HCC), as seen on superparamagneticiron oxide (SPIO)-enhanced MR images. MATERIALS AND METHODS: The study involved 30 patients with 51 focal hepatic mass lesions (31 hemangiomas, 20 HCCs). Breath-hold T1-weighted fast low angle shot (FLASH) and respiratory-triggered T2-weighted turbo-spin echo (TSE) images were obtained at 1.5 T before and after intravenous administration of SPIO particles. For quantitative analysis, percentage signal intensity change (PSIC) and contrast-to-noise ratio (CNR) of the lesions were calculated for T1-weighted FLASH and T2-weighted TSE before and after intravenous administration of SPIO particles. In addition, lesion conspicuity and imaging artifacts were analyzed qualitatively. RESULTS: After SPIO administration, percentage signal intensity increase on T1-weighted FLASH images was 73.0 +/-22.1% for hemangiomas and 21.8 +/-12.6% for HCCs, the difference being significant (p < 0.05). Taking a signal increase of 40% on postcontrast T1-weighted FLASH as the cut-off value, sensitivity and specificity for hemangiomas were 96.8% and 100%, respectively. In addition, the percentages of signal intensity loss on T2-weighted TSE images for hemangiomas and HCCs were 35.5 +/-17.2% and 0.2 +/-10.5%, respectively (p < 0.05). A comparison of lesion to liver CNR before and after SPIO infusion showed readings-for hemangiomas and HCCs, respectively - of 15.4 +/-6.0 and 4.7 +/-4.4 on T1-weighted FLASH images, and -2.6 +/-0.7 and 2.7 +/-4.4 on T2-weighted TSE images (p < 0.05). Qualitative analysis indicated that the conspicuity of HCCs was noticeably greater on postcontrast T2-weighted TSE images than on precontrast images (p < 0.05). CONCLUSION: The positive enhancement seen on T1-weighted FLASH images and the negative enhancement on T2 weighted TSE observed in liver hemangiomas after the administration of SPIO particles are valuable diagnostic features that can help characterize hemangiomas and differentiate them from HCCs.

Detectability of Hepatocellular Carcinoma: Comparison of Gd-DT PA-Enhanced and SPIO-Enhanced MR Imaging

PURPOSE: To compare the detectability of hepatocellular carcinoma (HCC) using superparamagnetic iron oxide (SPIO)-enhanced T2-weighted turbo spin echo (TSE), SPIO-enhanced T2*-weighted FISP, and dynamic Gd-DTPA-enhanced fast low-angle shot (FLASH) MR images. MATERIALS AND METHODS: In order to assess their hepatic lesions, 25 patients (20 men and 5 women) with HCC were enrolled in an MR study in which both gadolinium and SPIO were used. Since the lesions were most conspicuous during the phase of dynamic arterial dominant phase of dynamic gadolinium-enhanced imaging, this was the phase used for analysis. Images were analyzed qualitatively and quantitatively, and to compare the diagnostic value of gadolinium-enhanced imaging with that of SPIO-enhanced imaging for the detection of HCCs, a receiver-operated characteristic curve was obtained. RESULTS: Qualitative analysis revealed a significantly higher percentage of signal loss and a higher liver-lesion contrast-to-noise ratio on SPIO-enhanced FISP imaging than on SPIO-enhanced T2-weighted TSE imaging (p<0.05). It also showed that the lesions were most clearly visible on SPIO-enhanced FISP imaging(and signifi-cantlyso), followed by SPIO-enhanced T2-weighted TSE imaging, and dynamic gadolinium-enhanced imag-ing. Imaging artifacts were more prominent on SPIO-enhanced T2-weighted TSE than on SPIO-enhanced PISF imaging or dynamic gadolinium-enhanced imaging(p<0.05). According to ROC analysis, SPIO-enhanced T2-weighted turbo spin echo(TSE) or SPIO-enhanced FISP imaging achieved higher accuracy than did dynamic gadolinium-enhanced FLASH imaging(p<0.05). CONCLUSION: For the detection of hepatocellular carcinomas, SPIO-enhanced MR imaging is better than gadolinium-enhanced FLASH imaging.