| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Monday, May 1, 10:00 AM–12:00 PM
Abstracts 023–031
Moderators: Elmar M. Merkle, MD and Duyshant V. Sahani, MD
10:00 AM
Keynote Address: 1.5 T vs, 3.0 T of the Liver
Elmar M. Merkle, MD, Duke University Medical Center, Durham, NC
10:30 AM
023. SPIOs and Low Molecular Weight Gd Chelates Are Synergistic for Direct Visualization of Advanced Liver Fibrosis
Hughes F.; Chavez A.D.*; Bahl G.; Wolfson T.; Znamirowski R.; Hassanein T.; Bydder G.; Sirlin C.; Radiology, University of California, San Diego, San Diego, CA.
Address correspondence to A.D. Chavez (adchavez{at}ucsd.edu)
Objective: Super-paramagnetic iron oxides (SPIOs) and low-molecular weight gadolinium (Gd) chelates both increase the visibility of fibrosis on magnetic resonance (MR) images. SPIOs preferentially accumulate in and cause signal loss in non-fibrotic liver; Gd chelates preferentially accumulate in and cause signal enhancement in fibrosis. Because SPIOs and Gd chelates increase fibrosis visibility by different mechanisms, we hypothesized the two agents would be synergistic. The purpose of this study was to compare the visibility of advanced fibrosis on unenhanced, Gd-enhanced, SPIO-enhanced, and combined (SPIOs + Gd) contrast-enhanced MR images.
Materials and Methods: This was a retrospective study on 58 consecutive cirrhotics who had spoiled gradient echo (SGE) imaging of the liver at 1.5T with (a) no contrast agents, (b) Gd only, (b) SPIOs only, and (d) combined agents. For each type of contrast enhancement (none, Gd, SPIO, or combined), four breath-hold SGE sequences were analyzed: (a) 2D T1w (n = 58 patients); (b) 2D T2*w (n = 11); (c) 2D multi-echo (n = 5); and (d) 3D T1w (n = 32). MR images were evaluated qualitatively and, in 10 patients who underwent liver transplantation, compared to gross pathology. Also, multiple regions of interest were measured on fibrotic and non-fibrotic tissue. Contrast-to-noise (CNR) of fibrosis to background liver was calculated and compared with Bonferroni-corrected pairwise Wilcoxon tests. Tests were regressed on echo time for the multi-echo sequence.
Results: On all sequences, combined contrast-enhanced images showed fibrosis with high clarity as a meshwork of high-signal 1–3 mm thick reticulations surrounding 2–5 mm low-signal nodules, which corresponded to fibrosis and regenerative nodules on gross pathology. Fibrosis was less visible on SPIO- and Gd-enhanced images and was not visible, without direct comparison to enhanced images, on unenhanced images. CNR of combined contrast images exceeded CNR of SPIO-enhanced (p < 0.05 to p < 0.0001 for all comparisons) and Gd-enhanced (p < 0.0001) images, and CNR of SPIO-enhanced and Gd-enhanced images exceeded CNR of unenhanced images (p < 0.05 to p < 0.0001).
Conclusion: SPIOs and Gd chelates are synergistic in demonstrating advanced liver fibrosis on SGE images. The effect is robust and was confirmed on four different SGE sequences. In cirrhotics, combined contrast MR imaging directly visualizes fibrosis with high clarity.
024. Diagnosis of Cirrhosis by Helical CT: A Case-control Study with Feature Analysis and Assessment of Observer Agreement
Keedy A.W.1,2; Aslam R.1*; Westphalen A.1; Rybkin A.;C Chen M.1; Coakley F.V.1; 1. Department of Radiology, University of California San Francisco, San Francisco, CA; 2. School of Medicine, University of California Irvine, Irvine, CA.
Address correspondence to A.W. Keedy (akeedy{at}uci.edu)
Objective: To determine the accuracy of helical CT in the diagnosis of cirrhosis using a case-control study with multivariate feature analysis and assessment of observer agreement.
Materials and Methods: We retrospectively identified 126 patients who underwent helical CT at our institution and who had a contemporaneous hepatic biopsy or resection demonstrating cirrhosis (n = 67) or normal liver (n = 59). Two experienced readers independently recorded the presence or absence of hepatic and extrahepatic findings of cirrhosis and portal hypertension and the overall likelihood of cirrhosis on a five-point scale from 1 (definitely absent) to 5 (definitely present and severe). Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic accuracy of CT findings, and observer agreement was assessed by kappa statistics for categorical variables and by correlation coefficients for continuous variables.
Results: The overall likelihood of cirrhosis was the most accurate and objective diagnostic feature, with an area under the ROC curve (Az) of 0.97 for reader 1 and 0.90 for reader 2 and a kappa value of 0.70. Individual findings that were accurate and objective were diaphragmatic surface nodularity (Az of 0.95 and 0.88 for reader 1 and 2, respectively, with a kappa value of 0.75), global or segmental volume loss (Az of 0.95 and 0.87 for reader 1 and 2, respectively, with a kappa value of 0.70), and superior diaphragmatic adenopathy (Az of 0.85 for both reader 1 and 2, with a kappa value of 0.78). Of note, visceral surface nodularity was accurate but less objective (Az of 0.97 and 0.87 for reader 1 and 2, respectively, with a kappa value of 0.59) while portal vein diameter was not significantly different between normal and cirrhotic patients as measured by either reader (p = 0.54 and 0.65).
Conclusion: Helical CT demonstrates high accuracy and interobserver agreement in the diagnosis of cirrhosis, suggesting CT may be an alternative diagnostic test in patients who have contra-indications to biopsy, or may be an adjunct to confirm or refute the diagnosis of cirrhosis in patients with equivocal biopsy findings.
025. Long-term Natural History of Focal Confluent Fibrosis (FCF) in the Cirrhotic Liver: A CT Serial Analysis
Brancatelli G.1,3*; Baron R.L.2; Federle M.P.3; Sparacia G.1; Pealer K.3; 1. Radiology, University of Palermo, Palermo, Italy; 3. Radiology, University of Pittsburgh, Pittsburgh, PA; 2. Radiology, University of Chicago, Chicago, IL.
Address correspondence to G. Brancatelli (gbranca{at}yahoo.com)
Objective: To assess the long-term natural history of FCF in the cirrhotic liver with computed tomography (CT).
Materials and Methods: Two radiologists retrospectively reviewed in consensus 118 liver CT scans (2–9 scans per patient; initial and final CT time range: 6–81 months) in 26 patients (19 men, 7 women; age range 32–68 years, mean age, 50 years). Helical CT scans were obtained before, at 30s and at 70s after injection of 125 mL of contrast medium at a rate of 4 or 5 mL/sec. Proof of cirrhosis was based on histology (n = 13) or on imaging findings (n = 13). Number, location, attenuation and presence of trapped vessels within FCF lesions were evaluated. Size variation was assessed with an arbitrary retraction index obtained by using a standard linear regression analysis with the mixed-effects model. The retraction index was defined as a variation of the ellipsoid volume of FCF lesions over time. A classification and regression tree analysis performed by using the Fisher exact test was utilized to build the most efficient classification algorithm combining retraction index and time interval.
Results: Each radiologist identified 41 FCF lesions. Twelve patients (46%) had a single lesion, 13 patients (50%) had two lesions and one patient (4%) had three lesions. Twenty-eight (68%) of 41 lesions were located either in segment IV, VII or VIII. Thirty-two (68%) of 41 lesions were hypoattenuating on NC images, 25 (62%) of 40 lesions were hypoattenuating on HAP and 20 (49%) of 41 lesions were isoattenuating on PVP. Seven (17%) of 41 lesions were or became hyperattenuating at follow-up on PVP. Trapped vessels were found in six (15%) of 41 lesions. The retraction index showed an overall significant increase over time (r = 0.423, p = < 0.0001). Classification and regression tree analysis demonstrated an increase of the retraction index of 1% in 40% of lesions at <=24 months, of 7% in 32% of lesions at 24–40 months, of 52% in 17% of lesions at 40–60 months and of 17% in 11% of lesions at > 60 months (F = 51,9, p = 0.0001).
Conclusion: Changes in the liver from FCF occurs substantially and rapidly in some patients and more slowly in others. Follow-up imaging may be beneficial in characterizing these lesions.
026. Combined Contrast-Enhanced MR Imaging Permits Accurate Non-Invasive Staging of Liver Fibrosis in Non-Alcoholic Fatty Liver Disease
Sirlin C.B.*; Chavez A.D.; Wolfson T.; Znamirowski R.; Collins J.; Hassanein T.; Bydder G.; Radiology, University of California, San Diego, San Diego, CA.
Address correspondence to C.B. Sirlin (csirlin{at}ucsd.edu)
Objective: Non-alcoholic fatty liver disease (NAFLD) is a major source of morbidity, mortality, and health-care costs, affects children and adults, and is increasing in prevalence. Diagnosis and follow up of NAFLD is a major challenge. Biopsy, the current gold standard, samples only a fraction of the liver and has risks. We previously showed that combined contrast-enhanced (CCE) MR imaging, in which super-paramagnetic iron oxides and a low molecular weight gadolinium chelate, are given in combination, permits accurate staging of liver fibrosis in patients with viral hepatitis and alcoholic liver disease. The purpose of this study was to assess non-invasive staging of fibrosis using CCE MR imaging in NAFLD.
Materials and Methods: This retrospective, cross-sectional study reviewed 26 consecutive patients with clinically and histologically confirmed NAFLD who underwent, over a 9-month period, CCE MR imaging at 1.5T. One blinded radiologist reviewed CCE spoiled gradient recalled echo images (TR/TE/FA = 140ms/4.6ms/70°) twice, several weeks apart, in a different random order. MR interpretations were independent and there was no training in between. Fibrosis was scored on MR qualitatively using a 5-point ordinal scale designed to parallel the 5-point modified Brunt scale used for histological staging of fibrosis in NAFLD. In a separate session from the MR interpretations, a single hepatopathologist determined the histological fibrosis stage for each patient. Spearman rank correlations were calculated between MR interpretations 1 and 2 and between each MR interpretation and histological fibrosis stage. Diagnostic performance was assessed.
Results: On pathology, patients had modified Brunt scores of 0 (normal, n = 5 patients), 1 (mild fibrosis, n = 3), 2 (moderate, n = 2), 3 (severe, n = 1), and 4 (cirrhosis, n = 15). CCE MR images showed fibrosis as high-signal reticulations. Reticulations were thicker, denser, and more clearly visible in patients with more advanced fibrosis. Spearman rho between interpretation 1 and interpretation 2 was .985. Spearman rho between binarized interpretations (0,1,2 vs. 3,4) was 100%. Spearman rho between MR and pathology was .979 and .964 for interpretations 1 and 2, respectively. Sensitivity for modified Brunt scores of 3 or 4 (severe fibrosis or cirrhosis) was 94% (15/16) for both interpretations. Specificity was 100% (10/10).
Conclusion: CCE imaging permits accurate non-invasive staging of liver fibrosis in NAFLD using pathology as the reference. Intra-observer agreement is excellent.
027. Evaluation of Hepatic Steatosis by Opposed-phase MR Imaging: Effect of Liver Iron Deposition in Patients with Cirrhosis
Westphalen A.C.1*; Qayyum A.1; Lee J.1; Lamba A.1; Yeh B.M.; Lu Y.; Coakley F.V.; Departments of Radiology and Pathology, University of California San Francisco, San Francisco, CA.
Address correspondence to A.C. Westphalen (antonio.westphalen{at}radiology.ucsf.edu)
Objective: We undertook this study to determine the effect of liver iron on the evaluation of hepatic steatosis by opposed-phase MR imaging in patients with cirrhosis.
Materials and Methods: This was a retrospective, single institution study. Committee on Human Research approval was obtained and compliance to HIPAA regulations was observed. Patient consent was not required. We identified 38 patients with cirrhosis who underwent abdominal MR imaging and who had contemporaneous liver biopsy. The etiology of cirrhosis in the study group was hepatitis C (n = 22), hepatitis B (n = 6), cryptogenic (n = 5), primary sclerosing cholangitis (n = 2), alcoholic cirrhosis (n = 1), non-alcoholic steatohepatitis (n = 1), and auto-immune cirrhosis (n = 1). None of the patients had hemochromatosis. Two radiologists independently quantified steatosis by the relative loss of signal intensity on out-of-phase compared to in-phase T1-weighted gradient-echo images. A single hepatopathologist independently reviewed the histopathologic slides for all patients without knowledge of radiological data. Fat content was determined as the overall impression of the percentage fat-containinghepatocytes on hematoxylin and eosin-stained preparation. Iron deposition was identified on Pearl's Prussian-blue stain and was graded based on the ease of observation of iron deposits and required magnification. Generalized linear models, including a mixed random effects model, were used to determine the effect of iron deposition on Spearman's correlation coefficient for relative signal loss versus histopathologic fat percentage.
Results: The mean histopathologic liver fat percentage was 3.4% (range, 0–25). Liver iron deposition was found in 25 of 38 patients. For both readers, relative signal loss on opposed-phase imaging was closely and significantly correlated (p < 0.05) with histopathologic liver fat percentage in patients without iron deposition (r = 0.7, r = 0.6), but no such correlation was found in patients with iron deposition (r = 0.1, and r = –0.31, p > 0.05).
Conclusion: Iron deposition is common in patients with cirrhosis, and evaluation of hepatic steatosis by opposed-phase MR imaging is inaccurate in the presence of such deposits; appropriate caution should be exercised in assessing hepatic steatosis by opposed phase MR imaging in patients with suspected end-stage liver disease.
028. Can Diffusion-Weighted MR Imaging Reliably Assess Anti Angiogenic Response in Hepatocellular Carcinoma?
Holalkere N.*; Sahani D.; Zhu A.; Mueller P.R.; Abdominal Imaging and Interventions, Massachusetts General Hospital, Boston, MA.
Address correspondence to N. Holalkere (nholalkere{at}partners.org)
Objective: To study if Diffusion Weighted Imaging (DWI) can serve as a non-invasive surrogate to measure early changes of anti-angiogenic therapy in patients with locally advanced HCC using perfusion CT as a correlate.
Materials and Methods: In this phase II clinical trial, 8 consecutive patients (M:F 5:3, mean age 58 yrs) with advanced HCC, underwent DWI MR imaging using a phased array coil on a 1.5-T MR (Signa, Siemens) before and 10 to 12 days after a single dose of Bevacizumab (Avastin). DWI was performed using a non-breath hold EPSE sequence (B value 500; 5,500/95) with a slice thickness of 6 mm, gap of 1.8 mm and 128 x 128 matrix. Apparent diffusion co-efficient (ADC) maps were generated and ADC values were calculated by placing a ROI in the tumor and background liver. All target lesions 2 cm or larger were included for evaluation and changes in ADC values were compared using paired student T test. A perfusion CT (CTp) was concurrently performed and tumor blood flow (BF), blood volume (BV), Mean transit time (MTT) and permeability surface (PS) was calculated using body perfusion 2 software (GE). Comparison was made between the ADC and CTp parameters.
Results: A total of 18 lesions (range 1 to 4 lesions per patient) with size range of 4.5 to 11.8 cm were studied. Cirrhosis was present in two patients and one patient demonstrated portal vein thrombosis. The mean ADC values before treatment and after treatment for HCC were as follows 1.3 x 10–3 mm2/sec (range 0.9 to 1.5 x 10–3 mm2/sec) and 1.4 x 10–3 mm2/sec (range 1.2 to 1.5 x 10–3 mm2/sec) (p = 0.5). Similarly, the background liver demonstrated a mean ADC value of 1.4 and 1.5 x 10–3 mm2/sec before and after treatment (p = 0.3). There was no significant difference (p = 0.3) in ADC of tumor and liver parenchyma. However, significant reduction in the tumor BF (p = 0.002), BV (p = 0.01) and PS (p = 0.002) was observed on CTp.
Conclusion: In our initial experience, DWI was ineffective to assess treatment response following a single dose of anti angiogenic therapy for advanced HCC despite significant response on perfusion CT.
029. PET/CT Findings in the Liver after Radiotherapy
Iyer R.B.; Balachandran A.*; Macapinlac H.A.; Munden R.F.; Diagnostic Radiology, University of Texas M.D. Anderson Cancer Center, Houston, TX.
Address correspondence to A. Balachandran (abalachandran{at}di.mdacc.tmc.edu)
Objective: To compare changes in Fluorodeoxyglucose (FDG) uptake on PET and attenuation changes on CT in irradiated liver after radiotherapy for distal esophageal cancer.
Materials and Methods: Twenty six patients with esophageal carcinoma who underwent pre-operative chemo-radiation were included in the study. All patients had baseline PET/CT and follow-up PET/CT approximately 6 weeks after completion of 50.4 Gy of radiation therapy to the distal esophageal tumor. None of the patients had documented liver metastases. Retrospective review of the PET/CT images pre and post radiation was performed by 2 radiologists experienced in PET/CT. The maximum standard uptake value (max SUV) was measured using a region of interest cursor over the left and right lobes of the liver as well as the primary tumor before and after radiotherapy. Changes in attenuation of the right and left lobe of the liver was also documented on CT before and after radiotherapy.
Results: In two patients FDG uptake in the irradiated left lobe of the liver adjacent to the irradiated esophageal tumor increased by greater than 50 % (54% and 133% respectively) with no significant change in FDG uptake in the non-irradiated right lobe of the liver and no evidence of liver mass on CT. One patient had an approximately 50% decrease in max SUV of the liver in both the right and left lobes after therapy. One patient had a 60% increase in FDG uptake in the right lobe without corresponding lesion on CT and with no significant change in the left lobe. In the remaining 22 patients, no significant difference in FDG uptake between the right and left liver was seen post irradiation. Atrophy and characteristic changes in attenuation of the irradiated liver on CT was seen in fifteen patients.
Conclusion: Hepatic injury is a known sequela after therapeutic irradiation. Changes in FDG uptake in the irradiated left lobe of the liver was seen infrequently and was variable (3/26 patients) while CT changes of atrophy and decreased attenuation were more frequently seen (15/26 patients) at 6 weeks after completion of therapy. This suggests that FDG-PET is a relatively insensitive measure of radiation change in the liver at this time point post therapy as compared to CT. However if, increased FDG uptake is seen in the liver, it is important to correlate with CT findings and distinguish radiation injury from metastatic tumor during pre-operative staging of patients with potentially operable esophageal cancer.
030. Orthotopic Adult Liver Transplantation: What Happens After Hepatic Artery Thrombosis?
Horrow M.M.1*; Blumenthal B.M.1; Manzarbeitia C.2; Reich D.J.2; 1. Radiology, Albert Einstein Medical Center, Philadelphia, PA; 2. Liver Transplantation, Albert Einstein Medical Center, Philadelphia, PA.
Address correspondence to M.M. Horrow (horrowm{at}einstein.edu)
Objective: To determine the time course, types of imaging and clinical outcomes of adult patients with hepatic artery thrombosis (HAT) after orthotopic liver transplantation.
Materials and Methods: Retrospective review of all patients with liver transplantations during the past 10 years.
Results: Of 485 liver transplantations, 14 developed 1° HAT (2.9 %) and 7, 2° HAT after other hepatic artery complications. Immediate 1° HAT (within 1 week of transplant) occurred in 5 cases, all diagnosed by ultrasound (US), confirmed by angiography and treated by thrombectomy with 2/5 alive after 1 year. Delayed HAT (< 10 months post transplant) occurred in 16 cases, 9, 1° and 7, 2°. Delayed 1° HAT was diagnosed in 6 by US and angiography, 2 during operations for other reasons and 1 with a false positive US with parvus tardus flow and a positive MRA and angiogram. Delayed 2° HAT occurred in 7 cases, 4 after treatment of a pseudoaneurysm and 3 after hepatic artery stenosis, all diagnosed with US and angiography. In the 16 cases of delayed HAT, 6 required intervention: 4 retransplantations (1 died), 1 thrombectomy, 1 thrombolysis. Outcomes in remaining 10 patients were 1 death due to sepsis at 4 months, 2 late deaths at 27 and 33 months due to biliary complications and 7 living patients with HAT (mean 47.7 months, range 10–79). US imaging in these 10 patients showed development of hepatic arterial collateral flow in 7, 2 confirmed at angiography. This collateral flow was often parvus tardus in nature and not found in the normal common hepatic artery location at the porta hepatis. Clinical followup in these 10 cases showed biliary complications and/or sepsis in 7. 15 of all 21 patients who developed HAT (72%) are currently alive.
Conclusion: US is the screening study of choice for HAT, with confirmation by angiography. The majority of our patients with HAT are alive. Thrombectomy is treatment for immediate HAT. Retransplantation is required in some patients with delayed HAT. Non-surgical candidates can live with delayed HAT, most developing arterial collaterals, but often with biliary complications. Arterial collaterals in patients with HAT may be a cause of false positive US.
031. Tumor Volume and Tumor Volume Doubling Time of Small Hepatocellular Carcinomas Evaluated with Contrast-enhanced 3D Gradient-echo Sequence (VIBE): Comparison of 2D vs. 3D Measurements; Preliminary Results
Chandarana H.*; Lee V.S.; Hecht E.; Taouli B.; Radiology, NYU Medical Center, New York, NY.
Address correspondence to H. Chandarana (chandh02{at}med.nyu.edu)
Objective: To compare tumor volume (TV) and tumor volume doubling time (TVDT) of untreated small hepatocellular carcinomas (HCCs) < 3 cm evaluated with contrast-enhanced 3D gradient-echo sequence (VIBE), using two dimensions (2D) and planimetric measurement (3D).
Materials and Methods: We identified retrospectively 10 untreated small HCCs in 8 cirrhotic patients (7 men, 1 woman, mean age 53 y) in our institution. All patients underwent serial MRIs (2–4 MRIs) with a minimum of 2.5 months interval time between two studies. Patients were scanned with 1.5 T scanners using volumetric interpolated breath-hold examination (VIBE) sequence before and after dynamic Gadolinium injection using a timing run on the aorta. Mean reconstructed slice thickness was 2.2 mm (range, 1.9–3.1 mm). Explant correlation was available in all patients, with all HCCs histologically confirmed. Twenty MRIs were evaluated by two observers in consensus. HCCs were identified on the best post-contrast phase. TV was calculated using 2D (max. and minimum dimensions according to WHO criteria) and planimetric 3D measurements (by calculating the surface of tumor x slice thickness). TVDT was calculated using the Schwartz formula [TVDT = (t x log 2)/log (TV1/TV0); where TV1 is follow-up TV, TV0 baseline TV, t the interval time between two MRIs]. Correlation between TV and TVDT obtained with the 2 datasets was evaluated.
Results: Mean maximum size of HCCs was 1.6 ± 0.4 cm (range, 1.0–2.2 cm) at baseline, and 1.9 ± 0.5 cm (range, 1.1–2.5 cm) at follow-up. Mean delay between MRIs was 115 days (range, 76–200 d). Patients underwent liver transplantation with a mean delay of 71 days (range, 31–143 d) after the latest follow-up MRI. Mean baseline and follow-up TV were 1.6–3.3 (2D) and 1.9–3.5 cm3 (3D planimetric). Mean-median and range of TVDT were 315–263 days (75–807 d) for 2D data; and 273–240 days (91–654 d) for 3D data. TV and TVDT from the 2 datasets were significantly correlated (for TV, r = 0.91; p < 10–3; for TVDT, r = 0.66; p < 0.01).
Conclusion: These preliminary results show that TV and TVDT of small HCCs could be correctly estimated using 2D measurements only. However, a larger number of patients should be evaluated.
* Will present paper
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
