| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Thursday, May 10, 10:00 AM–12:00 PM
Abstracts 218-227
Moderator(s): Ihab Kamel and Claude Sirlin
10:00 AM
Keynote Address: Advances in Imaging of Diffuse Liver Disease
Claude Sirlin, University of California–San Diego, San Diego, CA
10:20 AM
218. Mapping Living Liver Donor Vascular Anatomy: Comparative Performance of Gd-BOPTA and 16 MDCT Angiography
Sindhwani V.*; Singh A.; Hertel M.; Samir A.; Sahani D. Massachusetts General Hospital, Boston, MA
Address correspondence to V. Sindhwani (VSINDHWANI{at}PARTNERS.ORG)
Objective: The purpose of this study was to compare the feasibility and diagnostic accuracy of MR-imaging protocol using Gd-BOPTA with CT angiography for assessment of liver vascular anatomy in pre-surgical evaluation of Liver Donors.
Materials and Methods: Prior to surgery, using a phased array torso surface coil, MRI imaging was performed on a 1.5-T scanner in 10 liver donors who also underwent 16MDCT angiography for assessment of hepatic vascular anatomy. The following sequences were obtained: axial and coronal T1 and T2 non-breath hold, T2 coronal breath-hold and SPGR axial in-phase and out-phase-dual echo. After intravenous administration of 40 cc of Multihance (gadobenate dimeglumine, Gd-BOPTA; Bracco, Milano, Italy), 3D coronal MRA was performed (2.5 mm thickness) in a breathhold to cover the arterial and portal venous phase. CT angiography with bolus triggering at 150 HU using a slice thickness of 1.25 mm for arterial phase and 2.5mm for portal venous phase was performed. On independent review of both image sets 2 radiologists who were kept blinded to the surgical findings evaluated the hepatic vascular anatomy and variants and also rated CT and MR image sets for image quality and diagnostic confidence on a 5-point scale 1 = poor, 3 = average, 5 = excellent). Surgery constituted the stand of reference.
Results: Depiction of hepatic arterial morphological characteristics by MRA correlated with surgical findings in all 10 patients. The examinations demonstrated abnormalities of portal and hepatic venous anatomy (5/10), replaced left hepatic artery in (2/10) and replaced RHA from SMA (2/10). The sensitivity and accuracy of detection of variants in hepatic vascular anatomy by MRA was 100%. The image quality and diagnostic confidence gradings by both readers for MRA and CTA averaged to 4.2 and 4.4 respectively.
Conclusion: Gd-BOPTA is highly accurate for detection of hepatic vascular variants in liver donors and compares well with 16MDCT. This may avoid the need of CT angiography for pre-operative evaluation in such patients.
219. Intraindividual Comparison of Axial and Coronal Reconstructed Images Using Primary CT Raw Data at 64-slice CT (MSCT) for the Detection of Hepatocellular Carcinoma (HCC) in Cirrhotic Patients
Marin D.*; Catalano C.; Guerrisi A.; Di Martino M.; Passariello R. Università degli Studi di Roma "La Sapienza", Rome, Italy
Address correspondence to D. Marin (danielemarin{at}hotmail.it)
Objective: To prospectively compare axial and coronal reconstructed images using primary CT raw data for the detection of hypervascular HCC at 64-slice CT (MSCT).
Materials and Methods: 70 consecutive patients with cirrhosis underwent multiphasic CT performed with a 64-slice scanner (Sensation 64, Siemens). 40 patients with hypervascular HCC were identified. CT imaging parameters were: collimation 0.6 x 64 mm, section thickness 3-mm, mAs, 250, and kVp, 120. All patients received nonionic contrast medium (400 mgI/mL; Iomeron 400, Bracco) at a rate of 5 mL/sec. A triple-phase protocol was started 18, 60, and 180 seconds after the trigger (threshold level set at an increase of 150 HU over the baseline CT number of the aorta). Coronal images were directly reconstructed by primary CT datasets. Each dataset (axial and coronal images) was separately interpreted by three observers using a commercially available viewer (LifeWeb realViewer, Ferrania). Two reading sessions were achieved with a time interval of 4 weeks to minimize memory bias. Sensitivity, positive predictive value, and area under the receiver operating characteristic curve (A(z)) were calculated. Timing for each interpretation was recorded.
Results: Mean interpretation times were 6 minutes (2–10 minutes) and 5.75 (2.50–9 minutes) for axial and coronal images, respectively (no significant difference). Mean sensitivity and positive predictive value for HCC were 78% and 73% on axial images, 75% and 77% on coronal images, respectively. No significant difference in sensitivity and positive predictive value among the images was detected. The mean A(z) values were 0.80 and 0.79 for axial and coronal images, respectively (no significant difference). However, with regard to subphrenic lesions the mean sensitivity of coronal (77%) images was statistically significantly superior to that of axial (65%) images (p < 0.05).
Conclusion: The analysis of axial and coronal images provides similar results in terms of interpretation time and detection of HCC lesions. Coronal images increase conspicuity of subdiaphragmatic lesions.
220. Imaging Accuracy of Hepatocellular Carcinoma (HCC) Staging in Cirrhotic Patients Undergoing Liver Transplantation
Furlan A.2*; Almusa O.2; Marsh J. W.1; Federle M. P.2 1. University of Pittsburgh, Starzl Transplant Institute, Pittsburgh, PA; 2. UPMC, Department of Radiology, Division of Abdominal Imaging, Pittsburgh, PA
Address correspondence to A. Furlan (furlana{at}upmc.edu)
Objective: To assess the accuracy of CT and MRI for pre-transplant evaluation and staging of HCC in patients with cirrhosis with respect to the Milan and UCSF criteria, for predicting acceptable outcomes after liver transplantation.
Materials and Methods: 1029 consecutive cirrhotic patients (646 M, 383 F, mean age 54 years) who underwent liver transplantation at UPMC in the last 5 years were evaluated. The last imaging report before surgery (970 CT; 59 MRI) and the pathologic report of the explanted liver were reviewed retrospectively. Note was made of the number and dimensions of lesions reported as HCC along with signs of macro-vascular invasion. Patients were then classified according to the Milan (solitary nodule <5 cm or 2 or 3 nodules, none >3 cm, no vascular invasion) and UCSF (solitary nodule <6.5 cm or 2 or 3 nodules, none >4.5 cm and total diameter <8 cm, no vascular invasion) criteria.
Results: Concordance between the radiological and pathologic staging was demonstrated in 993 (96.50%) and 1003 (97.47%) patients according to the Milan and UCSF criteria, respectively. Underestimation and overestimation by imaging was found in 28 (2.72%) and 8 (0.78%) patients respectively in relation to the Milan criteria and in 19 (1.85%) and 7 (0.68%) patients in relation to the UCSF criteria.
Conclusion: CT and MRI are highly accurate at staging HCC with respect to both the Milan and UCSF criteria that have been correlated with outcomes after liver transplantation.
222. T Staging of Hepatocellular Carcinoma: Accuracy of Multi-phase Helical CT in Adult Transplant Patients
Jeon T.*; Lee W.; Kim S.; Park M.; Park K.; Lim H. K. Samsung Medical Center, Seoul, South Korea
Address correspondence to T. Jeon (hathor97.jeon{at}samsung.com)
Objective: To evaluate the accuracy of multi-phase helical CT for T staging of hepatocellular carcinoma in adult transplant patients by correlating pathologic T staging.
Materials and Methods: For a 10-year period, we collected 53 adult transplant patients with pathologically proven hepatocellular carcinoma and liver cirrhosis who had preoperative three-phase helical CT. Their CT images were retrospectively analyzed by consensus of three radiologists without knowledge of pathologic findings. T staging of HCC was made according to the recently revised AJCC/UICC system (2002) as follows: T1 (single tumor without vascular invasion), T2 (single tumor with vascular invasion or multiple tumors, none >5 cm), T3 (multiple tumors, any >5 cm or tumor involving a major branch of the portal or hepatic vein) and T4 (tumor[s] with direct invasion of adjacent organs or perforation of visceral peritoneum). CT staging was compared with the pathologic staging, and its diagnostic accuracy was determined.
Results: At pathology, 31 patients had single tumor and 22 patients had multiple tumors with a total number of 83 HCCs. The patient number of T1–T4 was 30, 15, 7 and 1, respectively. The size of hepatocellular carcinomas was 0.4-6.5 cm (mean, 2.7 cm). At CT, 61 of 83 hepatocellular carcinomas were correctly diagnosed (sensitivity, 74%), and 30 of 53 patients were correctly staged (sensitivity, 57%). The sensitivity of each stage was 57% (17/30), 53% (8/15), 71% (5/7) and 0% (0/1) in T1–T4, respectively. Eighteen patients (34%) were understaged mostly due to missed lesion (n = 17, tumor size; 0.4–1.2 cm [mean, 1.0 cm]). Five patients (9%) were overstaged mostly due to arterial enhancing pseudolesion (n = 4). The sensitivity of early T stage (T1+T2) and advanced T stage (T3+T4) was 78% and 63%, respectively.
Conclusion: The diagnostic accuracy of multi-phase helical CT is not satisfactorily high in T staging of hepatocellular carcinoma in adult transplant patients with hepatocellular carcinoma and liver cirrhosis, due to limited detectability of small lesion and pseudolesion. However, it may be helpful in distinguishing the early T stage from advanced T stage of hepatocellular carcinoma.
223. Percutaneous Radiofrequency Ablation of Peri-intestinal Hepatic Tumors: Initial Experience with Hydrodisplacement Technique
Anaya C.*; Gomez A.; Raman S.; Lu D. UCLA Medical Center, Los Angeles, CA
Address correspondence to C. Anaya (canaya{at}mednet.ucla.edu)
Objective: Radiofrequency ablation (RFA) is a minimally invasive therapeutic procedure proven to be effective and safe in the treatment of hepatic malignancies. One of its limitations continue to be the treatment of subcapsular hepatic tumors which are in close proximity to hollow viscous. We report our initial experience in 5 patients using the hydrodisplacement technique as a protective measure to separate liver form adjacent bowel.
Materials and Methods: Five patients diagnosed with subcapsular, peri-intestinal hepatic tumors underwent percutaneous radiofrequency ablation with hydrodisplacement to protect and displace adjacent bowel or stomach (Two patients diagnosed with primary hepatocellular carcinoma and three diagnosed with metastatic colon carcinoma). In this technique, a 22-gauge needle is first positioned under ultrasound guidance and 200–400 ml of D5W infused to create a fluid interface between the tumor and adjacent bowel. The procedure is done under continuous sonographic monitoring to ensure tissue separation and follow-up needle location. In one case tissue displacement was not achieved by D5W alone, and required the placement of a fluid filled balloon catheter to aid in the separation of the two structures. All patients had overnight admission for clinical follow-up. Post ablation contrast CT was performed in all cases within 24 hours.
Results: Peritoneal D5W infusion alone was adequate for the separation of tumor and bowel in four patients. One patient required a balloon catheter to be used for successful tissue separation. All patients recovered satisfactorily with no signs of bowel injury upon clinical follow-up. Post ablation contrast CT where performed within 24 hours of the procedure and demonstrated adequate perfusion ablation defects with no radiographic signs of bowel injury.
Conclusion: Hydrodisplacement is an effective tool designed to increase the safety of hepatic RFA when treating lesions that carry a higher risk of abdominal organ injury. By creating a fluid interface through image guided peritoneal fluid infusion, the two tissues of interest can be separated. When fluid interface is suboptimal the technique can be complemented by a balloon interposition technique.
224. Characterization of Cirrhotic Liver Reticulations with Unenhanced and Contrast-Enhanced MR (CEMR) Imaging: An In Vivo and Ex Vivo Study with Pathology Correlation
Bahl G.2*; Motta G.1; Chavez A.1; Collins J.1; Wolfson T.1; Hassanein T.1; Gamst A.1; Bydder G.1; Behling C.1; Sirlin C.1 1. University of California, San Diego, San Diego, CA; 2. Wayne State University/Detroit Medical Center, Detroit, MI
Address correspondence to G. Bahl (gautam.bahl{at}gmail.com)
Objective: MR imaging depicts parenchymal reticulations in cirrhotic livers. Prior studies have postulated that reticulations represent liver fibrosis. Since cirrhotic livers develop a spectrum of nodules and perfusion alterations, reticulations conceivably represent a fibrosis surrogate rather than fibrosis per se. This study aimed to confirm that liver reticulations on MR represent fibrosis and characterize the MR signature of fibrosis.
Materials and Methods: Part 1 of this study prospectively assessed 40 consecutive, freshly resected postsurgical and postmortem human liver specimens (n = 32 cirrhotic livers, 8 nonfibrotic livers). Specimens in saline were scanned at 1.5 T using high-resolution (voxel size 0.01–0.5 mm3) spin echo sequences, placed in 2 mM gadolinium (Gd) for 48–72 hours, re-scanned, sectioned, photographed, and submitted for histology. Tissue relaxation values were measured. Part 2 retrospectively assessed 20 consecutive clinical patients who underwent contrast-enhanced MR (CEMR) imaging (voxel size 2–20 mm3) within six months before liver transplantation. Explanted livers were sectioned into 5–10 mm slices in the in vivo imaging plane and photographed. Ex-vivo MR images (Part 1) and pre-transplant MR images (Part 2) were compared to pathology photographs and trichrome-stained histology slides.
Results: No reticulations were visible in liver specimens without fibrosis. In all ex vivo cirrhotic liver specimens and all cirrhotic patients, reticulations on MR corresponded to fibrosis as confirmed by pathology. As measured ex vivo, fibrosis had similar proton density as nonfibrotic liver tissue, but longer T1 (782 vs 710 msec) and longer T2 (106 vs 79 msec). Consequently, fibrosis was low signal on T1W images, high signal on T2W or T2*W, and invisible on proton density. After placement of specimens in Gd or after injection of patients with Gd, fibrosis enhanced, had shorter T1 than liver (310 vs. 412), and was high signal on T1W images. Fibrosis visibility was higher on CEMR than unenhanced images.
Conclusion: Reticulations on MR represent liver fibrosis. Fibrosis has longer T1 and T2 than liver, but similar proton density. Gd increases fibrosis visibility. In sum, liver fibrosis has a characteristic MR imaging signature, which permits its direct visualization.
225. Noninvasive Predictors of Steatosis: Comparison of MRI Evaluation with Body Mass Index
Bahl M.*; Qayyum A.; Westphalen A. C.; Noworolski S.; Chu B.; Coakley F. V.; Merriman R.; Tien P. University of California, San Francisco, San Francisco, CA
Address correspondence to M. Bahl (manisha.bahl{at}ucsf.edu)
Objective: To compare the relative utility of MRI with body mass index for the noninvasive evaluation of hepatic steatosis in patients with diffuse fatty liver disease.
Materials and Methods: This study was part of an ongoing prospective study funded in part by the NIDDK. Committee on Human Research approval and patient consent were obtained, and the study was in compliance with the Health Insurance Portability and Accountability Act. Fifty-two patients (15 men and 37 women) with non-alcoholic fatty liver disease (n = 29) and chronic hepatitis C (n = 23) underwent contemporaneous MRI and liver biopsy. The mean liver and spleen signal intensity was measured using 12 and 3 regions of interest for the liver and spleen, respectively. Regions of interest were placed at anatomically matched levels on dual-phase gradient echo and T2-weighted fast spin echo images with and without fat suppression. Liver fat was determined by the relative liver signal intensity loss on out-of-phase and T2-weighted fat suppressed images. Average visceral fat area was measured at the L2/3, L3/4, and L4/5 intervertebral disk levels on axial T1-weighted images using a free-form region-of-interest and manual thresholding to select intra-abdominal fat. Spearman correlation coefficients were used to compare histopathological liver fat grade with relative liver signal intensity loss, average visceral fat area, and body mass index.
Results: The median liver fat grade was 1 (range, 0–3). The mean body mass index was 29 (range, 17.3–46.8). The mean liver signal intensity loss on opposed-phase imaging was 18.8% (range, -8.40-71.43%). The mean liver signal intensity loss on T2-weighted imaging was 22.5% (range, -27.14-65.71%). The mean visceral fat area was 106.4 cm2 (range, 11.44–353.10 cm2). Histopathologic liver fat correlated well with relative liver signal intensity loss on opposed-phase imaging (r = 0.78), fat suppressed T2-weighted imaging (r = 0.75) and average visceral fat area measured on MR images (r = 0.77), (p < 0.01). Only a moderate correlation was observed between histopathologic liver fat and body mass index (0.53).
Conclusion: Histopathologic liver fat is strongly correlated with signal loss on opposed phase and fat saturated MRI and with visceral fat area measured at MRI, but is only weakly correlated with body mass index in patients with diffuse liver disease.
226. Diffusion Imaging in HIV/HCV-Related Liver Disease: Distribution of ADC Values in the Right and Left Lobes of the Liver
Dymond M.*; Qayyum A.; Westphalen A.; Lu Y.; Merriman R.; Tien P.; Noworolski S.; Vigneron D. University of California San Francisco, San Francisco, CA
Address correspondence to M. Dymond (melissa.dymond{at}radiology.ucsf.edu)
Objective: To determine homogeneity of diffusion imaging and correlation with histopathology in early diffuse liver disease.
Materials and Methods: Committee on Human Research approval and patient consent were obtained, and the study was in compliance with the Health Insurance Portability and Accountability Act. Thirty-one women with a mean age of 49 (range, 33–60) with human immunodeficiency syndrome and hepatitis C (HIV/HCV) related liver disease underwent MRI including single-shot fast spin echo diffusion tensor imaging of the liver. Twenty-one of the women underwent contemporaneous liver biopsy. The mean apparent diffusion coefficient (ADC) was calculated from 3 regions of interest measuring 3.2 cc (2 in the right and 1 in the left lobe) placed at the mid-level of the liver in the axial plane. The intraclass correlation coefficient for ADC was calculated to determine homogeneity of diffusion. Wilcoxon rank sum test was used to assess the relationship between mean ADC and histopathology for the 21 subjects with liver biopsy.
Results: The mean ADC value was 1.53 (range, 0.84–2.09), 1.72 (range 1.31–2.22) and 2.00 (range 1.34–2.48) in the right posterior segment, right anterior segment and left lateral segment of the liver, respectively. The correlation coefficients between regions of interests were 0.75 between right posterior and anterior segments, 0.63 between right posterior segment and left lobe, and 0.35 between right anterior segment and left lobe. The intraclass correlation of ADC between all 3 segments of the liver was 33%. For the 21 subjects with liver biopsy 5 subjects had grade 0 and 16 subjects had grade 1 steatosis. All subjects had early stage fibrosis (median 1, range 0–3) and variable inflammation (median 5, range 2–7). ADC was significantly correlated with steatosis (p < 0.01), and demonstrated a trend toward correlation with early fibrosis (p = 0.09).
Conclusion: In this preliminary study, hepatic apparent diffusion coefficients varied mildly in HIV/HCV liver disease, however the average apparent diffusion coefficient correlated significantly with steatosis but not fibrosis in these patients.
227. Role of 1H-MR Spectroscopy in the Evaluation of Hepatic Steatosis and Steatohepatitis in HIV-infected Subjects
Holalkere N.*; Sahani D. V.; Hadigan C. M.; Hahn P. F.; Mueller P. R. Massachusetts General Hospital, Boston, MA
Address correspondence to N. Holalkere (nholalkere{at}partners.org)
Objective: To evaluate HIV-infected adults for the presence of hepatic steatosis and steatohepatitis on proton MR spectroscopy (MRS) and to correlate the MRS results with clinical markers.
Materials and Methods: A total of 33 HIV-infected subjects (24M:9F) without cirrhosis or alcohol abuse within 3 years were enrolled prospectively and evaluated with single voxel proton MRS on a 1.5-T system (GE). Spectra were obtained using point-resolved spectroscopy (PRESS) sequence (TR/TE = 3,000/30 msec) and absolute metabolite concentrations were quantified by LC Model algorithm. Clinical markers such as liver function test, lipid profile, CD4 T cell count, HBV/HCV serology, homeostatic model for assessment of insulin resistance (HOMA-IR), body mass index (BMI), visceral abdominal fat area (VAT) measured by CT, body fat on bioelectrical impedance analysis and on dual X-ray absorptiometry (DXA) were obtained within a week of liver MRS. These parameters were correlated with total hepatic triglycerides (TL) (sum of peaks at 0.9, 1.3 and 2 ppm) and ratios of choline (Cho) to TL and glycogen complex (Glyu) to TL on MRS using Spearman correlation.
Results: Elevated TL (>0.045 mM) were present in 13 (40%), Cho/TL (>0.04) in 13 (40%), Glyu/TL (>0.25) in 1 (3%) and both Cho and Glyu/TL in 1 (3%) subjects. Subjects with elevated TL had a higher mean HOMA-IR, BMI, DXA and VAT compared to subjects without elevated TL (p < 0.05). A strong positive correlation between TL and VAT (r = 0.77), TL and HOMA-IR (r = 0.71) and negative correlation between Cho/TL and HOMA-IR (r = -0.5314) and Glyu/TL and HOMA-IR (r = -0.4968) was observed (p < 0.05). There was no correlation of TL, Cho/TL or Glyu/TL with aminotransferases or CD4 T cell counts or positive HCV/HBV serology. In a multivariate regression model, VAT was the strongest independent predictor of TL (p < 0.001).
Conclusion: Prevalence of hepatic steatosis is high in HIV-infected adults on MRS and strongly correlates with insulin resistance and visceral adipose tissue. Steatohepatitis as evident by elevated choline to lipid ratio on MRS is also common but lacks correlation with clinical markers of steatohepatitis. MRS has the potential to diagnose and monitor hepatic steatosis and steatohepatitis noninvasively and may possibly replace invasive biopsies.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
