Diffusion-weighted imaging of hepatocellular carcinoma for predicting early recurrence and survival after hepatectomy.

PURPOSE: The effectiveness of imaging (including apparent diffusion coefficient [ADC] of diffusion-weighted magnetic resonance imaging [DWI]) and laboratory variables for predicting early tumor recurrence and overall survival after surgery in hepatocellular carcinoma (HCC) patients are analyzed. METHODS: The present study included 116 consecutive patients with HCC who underwent partial hepatectomy. Patients were classified into two groups: patients with and without early recurrence (<1 year). Preoperative imaging variables (tumor number, size, shape, capsule, ADC, and venous invasion) and laboratory variables were evaluated to predict early recurrence using univariate and multivariate analyses. Overall survival was calculated using the Kaplan-Meier method. RESULTS: Twenty patients (17 %) developed early recurrence after surgery. Multivariate logistic regression analysis showed that tumor ADC (p = 0.0002), aspartate aminotransferase (p = 0.0121), and serum prothrombin time activity percentage (p = 0.0082) were statistically significant for predicting early recurrence. The optimal ADC cutoff value for predicting early recurrence obtained from receiver operating characteristic analysis was ≤0.898 × 10(-3) mm(2)/s. In patients with ADC ≤0.898 × 10(-3) mm(2)/s, the 3- and 5-year survival rates (77 and 56 %, respectively) were significantly decreased compared with those in patients with ADC >0.898 × 10(-3) mm(2)/s (97 and 97 %, respectively; p = 0.0015). CONCLUSIONS: Low tumor ADC value by DWI was a risk factor for early postoperative HCC recurrence and was associated with lower patient survival rates.

Diffusion-weighted imaging in optic neuritis.

PURPOSE: To evaluate imaging characteristics of optic nerves by using magnetic resonance imaging, especially diffusion-weighted imaging (DWI) with apparent diffusion coefficient measurements in acute and chronic phases of optic neuritis (ON). MATERIALS AND METHODS: A retrospective study was conducted by using records of 14 patients with clinically suspected acute ON (15 nerves), 5 chronic ON (7 nerves), and 11 normal volunteers with no eye symptoms were used as controls. Magnetic resonance imaging was performed by a 1.5T scanner. Affected nerves were evaluated for sizes, signal characteristics on DWI and T2-weighted imaging (T2WI), contrast enhancement, and apparent diffusion coefficient values. Visually assessed characteristics were compared between the acute and chronic, whereas apparent diffusion coefficient values were assessed among acute ON, chronic ON, and the control groups by using the Fisher exact test and Mann-Whitney U test. RESULTS: There were significant differences in the diameter of the optic nerves, hyperintensity on DWI, and enhancement characteristics on post-enhanced images in acute and chronic phases of ON (P = .0001, P < .0001, and P = .0022, respectively), apparent diffusion coefficient values of the optic nerves in acute ON, chronic ON, and control groups also differed significantly from each other. CONCLUSION: In conclusion, DWI can add valuable information in assessment of damage to nerve and neuronal barriers and thus in predicting recovery in cases of ON.

Diffusion- and T2-weighted MR imaging of the liver: effect of intravenous administration of gadoxetic acid disodium.

PURPOSE: We evaluated the effect of intravenous administration of gadoxetic acid disodium to hepatic lesions and liver parenchyma on T2-weighted (T2WI) and diffusion-weighted imaging (DWI). MATERIALS AND METHODS: One hundred and one consecutive patients with 259 hepatic lesions underwent T2WI and DWI (b-values of 500 and 1000 s/mm²) before and after gadoxetic acid administration. We compared the ratio of signal intensity (SIR) of the liver parenchyma and hepatic lesions, the ratio of contrast intensity of the lesion to the liver (CIR), the apparent diffusion coefficients (ADCs) of the liver and lesions, and lesion detectability between pre- and post-contrast images. RESULTS: SIRs, CIRs, and ADC of focal hepatic lesions were comparable on pre- and post-contrast images, and lesion detectability did not differ significantly between pre- and post-contrast T2WI and DWI. The SIRs of the liver parenchyma were significantly lower on post-contrast DWI (1.4±0.68 [b=500 s/mm²] and 1.71±0.67 [b=1000 s/mm²]) than pre-contrast images (1.89±0.68 [b=500 s/mm²] and 2.26±0.78 [b=1000 s/mm²]) (P<0.001). ADCs of the liver parenchyma were also significantly decreased on post-contrast DWI (0.77±0.32 mm²/s) than pre-contrast images (0.64±0.33 mm²/s) (P=0.001). CONCLUSION: T2WI and DWI after administration of gadoxetic acid are feasible and do not compromise the SIR, CIR, and ADC of focal hepatic lesions. However, the signal intensity of DWI and ADC value of the liver parenchyma were decreased on gadoxetic acid-enhanced hepatocyte phase images.

Magnetic resonance elastography as a predictor of insufficient liver enhancement on gadoxetic acid-enhanced hepatocyte-phase magnetic resonance imaging in patients with type C hepatitis and Child-Pugh class A disease.

OBJECTIVES: The aim of this study was to examine liver stiffness value measured by magnetic resonance elastography (MRE) and laboratory test results to find the best method for predicting insufficient liver enhancement on gadoxetic acid-enhanced hepatocyte-phase images. MATERIALS AND METHODS: The institutional ethics committee approved this retrospective study with waiver of informed consent. In total, 118 patients with Child-Pugh class A disease and type C hepatitis underwent MRE and gadoxetic acid-enhanced magnetic resonance imaging. During MRE examination, a pneumatic passive driver was used to obtain liver stiffness in kPa. Liver enhancement was assessed using liver-to-spleen contrast ratio (LSR), calculated using signal intensities of the liver and spleen on hepatocyte-phase magnetic resonance images obtained 20 minutes after contrast administration of gadoxetic acid. Insufficient liver enhancement was defined as an LSR lower than 1.5. The following laboratory test results were used as possible predictors of insufficient liver enhancement as well as liver stiffness measured by MRE: albumin, total bilirubin, aspartate aminotransferase, percentage prothrombin time, and platelet count. Correlation coefficients were calculated between LSR and these variables. Logistic analysis was performed to determine independent predictors of insufficient liver enhancement. RESULTS: All possible predictors investigated were significantly correlated with LSR. Logistic regression analysis revealed that MRE was the only variable to predict insufficient liver enhancement, with an odds ratio (95% confidence interval) of 2.03 (1.22-3.85) (P = 0.0138). A cutoff value of greater than 6.4 kPa yielded 95% specificity for predicting insufficient liver enhancement. CONCLUSION: Gadoxetic acid is not recommended in patients with liver stiffness greater than 6.4 kPa (consistent with severe fibrosis) because of insufficient liver enhancement on hepatocyte-phase images.

Cross-validation of MR elastography and ultrasound transient elastography in liver stiffness measurement: discrepancy in the results of cirrhotic liver.

PURPOSE: To evaluate individual differences in liver stiffness measurement using both MR elastography (MRE) and ultrasound transient elastography (UTE) in patients with chronic liver disease. MATERIALS AND METHODS: This study included 80 patients with chronic liver disease who underwent both UTE and MRE. MRE and UTE were performed using a pneumatic driver (60 Hz) and an ultrasound probe with a vibrator (50 Hz), respectively. Liver stiffness data measured using the two techniques (µ(UTE) and µ(MRE) ) were compared with respect to shear modulus. The patients were subdivided into four quartiles on the basis of average of the µ(UTE) and µ(MRE) values for each patient. RESULTS: The analysis of the 4 quartile groups revealed that µ(UTE) was significantly higher than µ(MRE) in the two most stiff liver groups: µ(UTE) versus µ(MRE) , 7.5 (1.2) versus 6.0 (0.72) kPa for the group with [µ(UTE) + µ(MRE) ]/2 of 5.6-8.0 kPa; 15.1(4.2) versus 6.7 (1.4) kPa for the group with >8.0 kPa. However, in the least stiff liver group (i.e., the group with [µ(UTE) + µ(MRE) ]/2 < 3.2 kPa), µ(UTE) was significantly lower than µ(MRE) . CONCLUSION: The shear modulus measured by UTE and MRE are not equivalent, especially in patients with stiff livers.

Mass-forming autoimmune pancreatitis and pancreatic carcinoma: differential diagnosis on the basis of computed tomography and magnetic resonance cholangiopancreatography, and diffusion-weighted imaging findings.

PURPOSE: To differentiate mass-forming autoimmune pancreatitis (AIP) from pancreatic carcinoma by means of analysis of both computed tomography (CT) and magnetic resonance imaging (MRI) findings. MATERIALS AND METHODS: Ten patients with mass-forming AIP diagnosed by revised clinical criteria of Japan Pancreas Society and 70 patients with pathologically proven pancreatic carcinoma were enrolled in this retrospective study. Two radiologists independently evaluated the CT and MR imaging findings. The sensitivity, specificity, and odds ratio of significant imaging findings and combinations of findings were calculated. RESULTS: Seven findings were more frequently observed in AIP patients: (i) early homogeneous good enhancement, (ii) delayed homogeneous good enhancement, (iii) hypoattenuating capsule-like rim, (iv) absence of distal pancreatic atrophy, (v5) duct penetrating sign, (vi) main pancreatic duct (MPD) upstream dilatation ≤ 4 mm, and (vii) an apparent diffusion coefficient (ADC) ≤ 0.88 × 10(-3) mm(2) /s. When the findings of delayed homogeneous enhancement and ADC ≤ 0.88 × 10(-3) mm(2) /s were both used in diagnosis of mass-forming AIP, a sensitivity of 100% and a specificity of 100% were achieved. When 4 of any of the 7 findings were used in the diagnosis of AIP, a sensitivity of 100% and a specificity of 98% were achieved. CONCLUSION: Analysis of a combination of CT and MR imaging findings allows for highly accurate differentiation between mass-forming AIP and pancreatic carcinoma.

Effects of gadoxetic acid on liver elasticity measurement by using magnetic resonance elastography.

The purpose of this study was to evaluate the effect of gadoxetic acid (Gd-EOB-DTPA) on measurements of liver stiffness by using magnetic resonance elastography (MRE). In this study, 104 consecutive patients (mean age, 67.7±9.4 years) underwent MRE using a 1.5-T MR scanner equipped with a cylindrical passive driver that was placed across the right chest wall for delivering vibrations. Axial gradient-echo images, which were automatically converted to elastograms that represented stiffness (kPa), were acquired using a continuous sinusoidal vibration of 60 Hz. Two raters independently placed a region of interest on the right lobe of the liver on the elastograms obtained before and after Gd-EOB-DTPA was administered. Liver stiffness was measured using these two elastograms and compared using a paired t test and correlation analysis. No significant difference was observed in liver stiffness before and after Gd-EOB-DTPA was administered (Rater 1, P=.1200; Rater 2, P=.3585). The correlation coefficients were 0.986 (Rater 1) and 0.984 (Rater 2), indicating excellent correlation between the stiffness values before and after Gd-EOB-DTPA was administered. Liver stiffness measured by MRE did not differ before and after Gd-EOB-DTPA was administered.

Imaging study of early hepatocellular carcinoma: usefulness of gadoxetic acid-enhanced MR imaging.

PURPOSE: To describe imaging findings of early hepatocellular carcinoma (HCC) at gadoxetic acid-enhanced magnetic resonance (MR) imaging, dynamic contrast material-enhanced computed tomography (CT), CT during arterial portography (CTAP), and CT during hepatic arteriography (CTHA) and to compare the diagnostic performance of each modality for small (≤ 2 cm) HCC. MATERIALS AND METHODS: The institute ethics committee deemed study approval unnecessary. One hundred eight resected small lesions in 64 patients were diagnosed as a dysplastic nodule (DN) (n = 12), progressed HCC (n = 66), or early HCC (n = 30). All but two patients underwent all imaging examinations. The imaging characteristics of the lesions with each modality were determined. To evaluate the diagnostic performance of the modalities, two radiologists graded the presence of HCC with use of a five-point confidence scale. The area under the receiver operating characteristic curve (A(z)), sensitivity, and specificity of each modality were compared. RESULTS: The imaging features that are statistically significant for differentiating an early HCC from a DN include fat-containing lesions at dual-echo T1-weighted MR imaging (seen in 16 of the 30 early HCCs and none of the DNs), low attenuation at unenhanced CT (seen in 13 of the 30 early HCCs and none of the DNs), low attenuation at CTAP (seen in 11 of the 30 early HCCs and none of the DNs), and low signal intensity at hepatocyte phase gadoxetic acid-enhanced MR imaging (seen in 29 of the 30 early HCCs and none of the DNs). The diagnostic performance of gadoxetic acid-enhanced MR imaging (A(z), 0.98 and 0.99) was significantly greater than that of contrast-enhanced CT (A(z), 0.87) and CTHA-CTAP (A(z), 0.85 and 0.86) owing to its significantly higher sensitivity (P < .001). CONCLUSION: Gadoxetic acid-enhanced MR imaging is the most useful imaging technique for evaluating small HCC, including early HCC.

Diagnosis of colorectal hepatic metastases: comparison of contrast-enhanced CT, contrast-enhanced US, superparamagnetic iron oxide-enhanced MRI, and gadoxetic acid-enhanced MRI.

PURPOSE: To compare the diagnostic accuracy of contrast-enhanced computed tomography (CE-CT), contrast-enhanced ultrasonography (CE-US), superparamagnetic iron oxide-enhanced magnetic resonance imaging (SPIO-MRI), and gadoxetic acid-enhanced MRI (Gd-EOB-MRI) in the evaluation of colorectal hepatic metastases. MATERIALS AND METHODS: In all, 111 patients with colorectal cancers were enrolled in this study. Of the 112 metastases identified in 46 patients, 31 in 18 patients were confirmed histologically and the remaining 81 in 28 patients were confirmed by follow-up imaging. CE-CT, CE-US, SPIO-MRI, and Gd-EOB-MRI were evaluated. Mean (of three readers, except for CE-US) area under the receiver operating characteristic curve (A(z) ), sensitivities, and positive predictive values (PPV) were calculated. Each value was compared to the others by variance z-test or chi-square test with Bonferroni correction. RESULTS: For all lesions, mean A(z) and sensitivity of Gd-EOB-MRI (0.992, 95% [56/59]) were significantly greater than those of CE-CT (0.847, 63% [71/112]) and CE-US (0.844, 73% [77/106]). For lesions ≤1 cm, mean A(z) and sensitivity of Gd-EOB-MRI (0.999, 92% [22/24]) were significantly greater than those of CE-CT (0.685, 26% [13/50]) and CE-US (0.7, 41% [18/44]). Mean A(z) (95% CI) of SPIO-MRI for all lesions (0.966 [0.929-0.987]) and lesions ≤ 1 cm (0.961 [0.911-0.988]) were significantly greater than those of CE-CT and CE-US. Mean sensitivity of SPIO-MRI for lesions ≤1 cm (63%, 26/41) was significantly greater than that of CE-CT. CONCLUSION: Gd-EOB-MRI and SPIO-MRI were more accurate than CE-CT and CE-US for evaluation of liver metastasis in patients with colorectal carcinoma.

Staging liver fibrosis by using liver-enhancement ratio of gadoxetic acid-enhanced MR imaging: comparison with aspartate aminotransferase-to-platelet ratio index.

OBJECTIVE: To compare the diagnostic ability of gadoxetic acid-enhanced hepatocyte-phase MR images with aspartate aminotransferase-to-platelet ratio index (APRI) to predict liver fibrosis stage. MATERIALS AND METHODS: Our study included 100 patients who underwent gadoxetic acid-enhanced MRI and either liver biopsy or liver surgery. Liver fibrosis stage was histologically determined according to the METAVIR system: F0 (n=16), F1 (n=17), F2 (n=10), F3 (n=21) and F4 (n=36). Four measures were used as imaging-based fibrosis markers: liver-spleen contrast ratio, liver-enhancement ratio, corrected liver-enhancement ratio and spleen index. APRI represented a blood test-based fibrosis marker. The diagnostic ability of those fibrosis markers were compared through receiver-operating characteristic analysis. RESULTS: The area under the curve (AUC) for APRI prediction of severe fibrosis (≥F3 and F4) was significantly greater than that of corrected liver-enhancement ratio. However, corrected liver-enhancement ratio had a greater AUC for prediction of mild fibrosis (≥F1) than APRI, although the difference was insignificant. CONCLUSION: Corrected liver-enhancement ratio with gadoxetic acid-enhanced MRI is correlated to the stage of liver fibrosis. APRI, however, has greater reliability for predicting severe fibrosis and cirrhosis than does the imaging-based fibrosis marker tested in this study.

Detection of pancreatic carcinoma and liver metastases with gadoxetic acid-enhanced MR imaging: comparison with contrast-enhanced multi-detector row CT.

PURPOSE: To intraindividually compare gadoxetic acid-enhanced magnetic resonance (MR) imaging with contrast material-enhanced multi-detector row computed tomography (CT) in detection of pancreatic carcinoma and liver metastases. MATERIALS AND METHODS: The ethics committee approved this retrospective study with waiver of informed consent. This study included 100 patients (53 men, 47 women; mean age, 67.8 years) consisting of 54 patients with pathologically confirmed pancreatic carcinoma (mean size, 33 mm) and 46 without a pancreatic lesion. Sixty-two liver metastases (mean size, 10 mm) in 15 patients with pancreatic carcinoma were diagnosed at pathologic examination or multimodality assessment. Three readers blinded to the final diagnosis interpreted all MR (precontrast T1- and T2-weighted and gadoxetic acid-enhanced dynamic and hepatocyte phase MR images) and tetraphasic dynamic contrast-enhanced CT images and graded the presence (or absence) of pancreatic carcinoma and liver metastasis on patient-by-patient and lesion-by-lesion bases. Receiver operating characteristic analysis, McNemar test, and Fisher test were performed to compare the diagnostic performance of CT and MR imaging. RESULTS: No significant differences were observed between CT and MR images in depiction of pancreatic carcinoma. However, MR imaging had greater sensitivity in depicting liver metastasis than did CT for two of the three readers in the MR imaging-versus-CT analysis (85% vs 69%, P = .046) and for all three readers in the lesion-by-lesion analysis (92%-94% vs 74%-76%, P = .030-.044). CONCLUSION: Gadoxetic acid-enhanced MR imaging was equivalent to dynamic contrast-enhanced CT in depicting pancreatic carcinoma and had better sensitivity for depicting liver metastases, suggesting the usefulness of gadoxetic acid-enhanced MR imaging for evaluation of patients with pancreatic carcinoma.

Outcome of hypovascular hepatic nodules revealing no gadoxetic acid uptake in patients with chronic liver disease.

PURPOSE: To elucidate the natural history of hypovascular nodules that appear hypointense on hepatocyte-phase gadoxetic acid-enhanced MR images by focusing on hypervascularization over time. MATERIALS AND METHODS: In this study, 135 hypovascular nodules revealing no gadoxetic acid uptake in 53 patients were examined. All nodules were retrospectively examined using serial follow-up computed tomography (CT) and MRI examinations until hypervascularity was observed on arterial-phase dynamic CT or gadoxetic acid-enhanced MR images, or on CT during hepatic arteriography. Logistic regression analysis was used to investigate the association between hypervascularization and MR findings including a presence of fat assessed by a signal drop on opposed-phase T1-weighted images. RESULTS: Of the 135 nodules, 16 underwent hypervascularization. The size of the nodules and the presence of fat in the nodules were independent indicators of hypervascularization. The 1-year cumulative risk of hypervascularization was 15.6%. This risk was significantly increased in the case of nodules >10 mm (37.6%, P < 0.01) and fat-containing nodules (26.5%, P < 0.01). CONCLUSION: Hypovascular nodules that appear hypointense on hepatocyte-phase gadoxetic acid-enhanced MR images may progress to conventional hypervascular hepatocellular carcinoma. Nodules more than 10 mm in diameter and containing fat are at high risk for developing hypervascularization.

Distinguishing hepatic metastasis from hemangioma using gadoxetic acid-enhanced magnetic resonance imaging.

PURPOSE: To retrospectively determine the findings of gadoxetic acid-enhanced magnetic resonance imaging (MRI) to distinguish hemangioma and metastasis of the liver. MATERIALS AND METHODS: The University's ethics committee approved this retrospective study. We assessed 45 patients without chronic liver disease who underwent gadoxetic acid-enhanced MRI. Diagnosis of 58 metastases and 47 hemangiomas was confirmed using histopathology or multimodality evaluation. Two radiologists independently assessed the following MRI findings of metastasis and hemangioma: their appearance on T2-weighted images (T2WI) and dynamic contrast enhancement patterns after gadoxetic acid administration. The metastasis and hemangioma findings were compared using Fisher exact test. The lesion-to-liver signal intensity ratio on hepatocyte phase was compared using the Mann-Whitney U test. Multivariate analysis was performed to identify independent imaging findings distinguishing the 2 diseases. Receiver operating characteristic analysis was used to estimate the diagnostic ability of gadoxetic acid-enhanced MRI to distinguish metastasis from hemangioma. RESULTS: The lesion-to-liver signal intensity ratio was comparable in both diseases. Peripheral-dot enhancement, ring-like, geographic, and moderate late-phase enhancement, rapid contrast filling, and bright signal on T2WI could differentiate between the 2 diseases. In multivariate analysis, bright signal on T2WI (94%-98% in hemangioma and 13%-25% in metastasis) and ring-like enhancement (4% in hemangioma and 58%-60% in metastasis) were the independent findings suggesting hemangioma and metastasis, respectively. The areas under the receiver operating characteristic curves to distinguish metastasis from hemangioma were 0.95 and 0.98 for Reader 1 and 2, respectively. CONCLUSION: Reliable findings to distinguish hepatic metastasis from hemangioma on gadoxetic acid-enhanced MRI were ring-like enhancement on arterial-phase images and bright signal on T2WI.

Double-dose gadoxetic Acid-enhanced magnetic resonance imaging in patients with chronic liver disease.

OBJECTIVES: determine the effect of double-dose gadoxetic-acid (Gd-EOB-DTPA) on lesion-liver contrast ratio in arterial- and hepatocyte-phase images and arterial-phase image quality in patients with chronic liver disease. MATERIALS AND METHODS: the ethics committee at our institute approved this study. This study included 28 patients (13 with Child-Pugh class A and 15 with class B) with 54 hepatocellular carcinomas. All patients received the standard Gd-EOB-DTPA dose (0.025 mmol/kg bodyweight) and double dose (0.050 mmol/kg bodyweight). The lesion-liver contrast ratio was evaluated in arterial- and hepatocyte-phase images. The artifacts in arterial-phase images were evaluated with a 4-point scale. Wilcoxon signed-rank test were used for comparisons. RESULTS: the hepatocyte-phase lesion-liver contrast ratio after the double dose was significantly higher than that after the standard dose in patients with Child-Pugh class B disease(standard dose vs. double dose; 0.20 ± 0.16 vs. 0.25 ± 0.17; P < 0.0001); however, the ratio after both the standard and double doses was equivalent in patients with Child-Pugh class A disease (0.35 ± 0.18 vs. 0.35 ± 0.14; P = 0.3038). The double dose significantly increased the arterial-phase lesion-liver contrast ratio (0.34 ± 0.19 vs. 0.58 ± 0.33; P < 0.0001). The artifacts in the arterial-phase images were more prominent after the standard dose (2.7 vs. 2.4 for reader 1, 2.8 vs. 2.4 for reader 2; P = 0.0195 and 0.0010). CONCLUSIONS: administration of double dose of Gd-EOB-DTPA provided better arterial enhancement of hepatocellular carcinomas in patients with chronic liver disease, and also improved the lesion-liver contrast in hepatocyte-phase images in patients with Child-Pugh class B disease.

Diagnosis of colorectal hepatic metastases: Contrast-enhanced ultrasonography versus contrast-enhanced computed tomography versus superparamagnetic iron oxide-enhanced magnetic resonance imaging with diffusion-weighted imaging.

PURPOSE: To compare the diagnostic accuracy of contrast-enhanced ultrasonography (CE-US), contrast-enhanced CT (CE-CT), and superparamagnetic iron oxide-enhanced MRI (SPIO-MRI) with diffusion-weighted imaging (DWI) in the evaluation of colorectal hepatic metastases. MATERIALS AND METHODS: Thirty-six patients with colorectal cancers were prospectively enrolled and retrospectively evaluated. Of the 86 metastases identified, 16 were confirmed histologically and the remaining 70 were confirmed by follow-up imaging. CE-CT and SPIO-MRI + DWI were independently evaluated by two readers, whereas CE-US was evaluated by consensus reading of two different readers. Area under receiver operating characteristic curve (A(z)), sensitivities, and positive predictive values (PPVs) were calculated and compared. RESULTS: For both readers, SPIO-MRI+DWI had significantly greater A(z) (0.879 and 0.904) and sensitivity (78% and 87%) for all lesions compared with CE-CT (0.779 and 0.793; 59% and 59%) and CE-US (0.811; 69%), and significantly greater A(z) (0.783 and 0.837) and sensitivity (56% and 73%) for lesions ≤1 cm compared with CE-CT (0.562 and 0.601; 20% and 22%) and CE-US (0.66; 37%). For lesions >1 cm, there was no significant difference in A(z), sensitivity and PPV between all the image sets. CONCLUSION: SPIO-MRI with DWI was the most reliable modality for evaluation of liver metastases particularly for lesions ≤1 cm.

Magnetic resonance elastography of the liver: preliminary results and estimation of inter-rater reliability.

PURPOSE: The aim of this study was to estimate inter-rater reliability and validate magnetic resonance elastography (MRE) as a tool to measure liver elasticity. MATERIALS AND METHODS: The study protocol was approved by the institutional review board at our institution. In all, 10 normal volunteers and 110 patients, who provided written informed consent, were enrolled. The pathological fibrosis score was applied as a standard reference of liver fibrosis in 21 patients. MRE was performed with a 1.5-T magnetic resonance imaging scanner with a cylindrical passive longitudinal shear wave driver placed over the right chest wall to deliver vibrations. A gradient-echo MRE sequence was used to acquire axial wave images, which were automatically converted to elastograms representing elasticity (in kilopascals, or kPa). The region of interest was placed in the right lobe of the liver on elastograms by two raters independently. To evaluate interrater reliability, the intraclass correlation coefficient was calculated. The elasticity measurements correlated with the pathological fibrosis score (F1-F4) in 21 patients. RESULTS: The intraclass correlation coefficient was almost perfect (0.993) between the elasticities measured by the two raters. The mean elasticity value for patients with F4 was 5.7 kPa; F3, 4.4 kPa; F2, 3.1 kPa; F1, 2.2 kPa; and F0, 2.1 kPa. CONCLUSION: MRE is a reliable tool to measure liver elasticity.

Distinguishing hypervascular pseudolesions of the liver from hypervascular hepatocellular carcinomas with gadoxetic acid-enhanced MR imaging.

PURPOSE: To retrospectively determine findings at gadoxetic acid-enhanced magnetic resonance (MR) imaging in hypervascular pseudolesions that were observed at computed tomography (CT) during hepatic arteriography, with special focus on distinguishing these pseudolesions from hypervascular hepatocellular carcinomas (HCCs). MATERIALS AND METHODS: The institute ethics committee deemed study approval unnecessary. The study population comprised 80 patients (55 men, 25 women) with chronic liver disease who underwent CT during hepatic arteriography and arterial portography, gadoxetic acid-enhanced MR imaging, and follow-up dynamic contrast material-enhanced CT. The diagnosis of 104 pseudolesions and 123 HCCs was confirmed by means of histopathologic or multimodality evaluation. Two radiologists assessed the MR imaging findings of HCCs and pseudolesions in consensus, including the signal intensities (SIs) of the lesions on T2-weighted, diffusion-weighted (DW), and contrast-enhanced hepatocyte-phase images. The findings of nodular pseudolesions and HCCs were compared with the Fisher exact test. Additionally, the hepatocyte-phase SI ratio (ratio of lesion SI to liver SI) for HCCs and pseudolesions was compared by means of the Mann-Whitney U test. RESULTS: There were 62 wedge-shaped, 32 nodular, and 10 linear pseudolesions. On gadoxetic acid-enhanced hepatocyte-phase MR images, 15% of pseudolesions (16 of 104) were hypointense compared with surrounding liver tissue. The mean hepatocyte-phase SI ratio of HCCs (0.65 +/- 0.14 [standard deviation]) was significantly lower (P < .01) than that of the nodular pseudolesions (0.95 +/- 0.11). The optimal cutoff value of hepatocyte-phase SI ratio for distinguishing between HCC and nodular pseudolesion was 0.84. No nodular pseudolesions were visible on DW images. CONCLUSION: Gadoxetic acid-enhanced hepatocyte-phase MR imaging and DW imaging could be used to distinguish hypervascular pseudolesions from hypervascular HCCs; a hepatocyte-phase SI ratio below 0.84 and visibility on DW images were findings specific for HCCs rather than pseudolesions.

High-b-value diffusion-weighted MR imaging of hepatocellular lesions: estimation of grade of malignancy of hepatocellular carcinoma.

PURPOSE: To evaluate the effectiveness of diffusion-weighted magnetic resonance imaging (DWI) in estimating the grade of malignancy of hepatocellular carcinoma. MATERIALS AND METHODS: Dynamic contrast-enhanced computed tomography (CE-CT) and DWI (b value, 1000 s/mm(2)) were performed on 73 patients. Using DW images, the lesions were classified as "visible" or "invisible." The apparent diffusion coefficient (ADC) of the lesions was measured. Furthermore, the lesions were classified as hypervascular or iso-hypovascular using arterial phase CE-CT images. The image findings for each lesion type were compared. RESULTS: The 73 patients had 98 hepatocellular lesions, of which 12 were histologically diagnosed as dysplastic nodules; 39, well-differentiated HCCs; 33, moderately differentiated HCCs; and 14, poorly differentiated HCCs. The mean ADC values of moderately poorly-differentiated HCCs were significantly lower than well-differentiated HCCs and dysplastic nodules (P < 0.01). On DW images, >90% of moderately (30/33) and poorly differentiated HCCs (13/14) were visible, while 51% of well-differentiated HCCs (20/39) and all dysplastic nodules were invisible. Of 22 iso-hypovascular lesions, 4 were visible on DW images and were poorly differentiated HCCs, whereas 18 were invisible and were dysplastic nodules (12/18) or well-differentiated HCCs (6/18). CONCLUSION: A combination of hypovascularity and visibility on DW images can help distinguish poorly differentiated HCCs from low-grade hepatocellular lesions (dysplastic nodules and well-differentiated HCCs).