GE Healthcare. MR-Touch MR Elastography. pulse. imagination at work APPLICATION MONOGRAPH

Transcription

1 GE Healthcare MR-Touch MR Elastography imagination at work pulse S I G N A APPLICATION MONOGRAPH

2 Table of Contents Healthymagination Source MR Elastography Healthymagination Validation Oxford Analytica....3 MR-Touch Fact Sheet Healthymagination....5 Whitepaper Healthymagination white paper (traditional format) Healthymagination....7 SignaPulse white paper (SignaPulse format) signapulse supplement Product Launch/Technology Source MR-Touch teaser...product Brief A New Touch for MR Imaging SignaPulse Fall MR Elastography / MR-Touch story board rsna The Sound Diagnosis SignaPulse Spring Patient Management Impact Changing Patient Management with MR Elastography....SignaPulse Fall Changing Patient Management Patient Perspective.....SignaPulse Fall MRE Helps Guide Patient Management SignaPulse Fall Clinical cases Case Review C ompendium SignaP ulse Fall MRE Strengthens Practice Service Line SignaPulse Fall Evaluating Liver Fibrosis without Biopsy SignaPulse Fall Cost Effectiveness Preliminary Threshold Assessment ISPOR European C ongress Scenario Analysis... SignaP ulse Fall Non-invasive Liver Disease Evaluation. Current Trends in Non-invasive Evaluation of Liver Disease.... SignaPulse Spring Quantitative Fat Imaging for Evaluating Diffuse Liver Diseases... SignaPulse Spring Clinical Utility of Quantitative Fat Imaging SignaPulse Spring Non-invasive Evaluation of Liver Disease Using MR Elastography... SignaPulse Spring The facts and reimbursement figures that appear in articles in this monograph were true and accurate at the time of the article s original publication and reflect national averages Healthcare providers should always independently investigate and consider reimbursement rates and other factors in their own area 2

3 33 3

4 1 4

5 GE Healthcare Fact Sheet MR-Touch MR-Touch Fact Sheet What challenges does MR-Touch address and what solutions does it create? Chronic liver disease and cirrhosis are major public health problems worldwide. In 2004, these conditions were associated with nearly 40,000 deaths and a cost of at least $1.4 billion for medical services in the U.S. alone. 1,2 Liver biopsy is the current standard of care for detecting hepatic fibrosis, but its invasive nature limits its value to use as a screening tool for a large population. There are also limitations with the technique that include poor acceptance by patients, measurement errors, and cost. 3,4 Current noninvasive alternatives to liver biopsy are serum-based testing, 5 which is not reliable for detecting early disease, and transient ultrasound elastography, 6 which has technical limitations in patients with obesity and conditions such as ascites. MR-Touch uses the MR Elastography (MRE) technique to provide diagnostic information without the discomfort and risk of complications due to invasive procedures, enabling more frequent evaluation when closer monitoring is needed. What are the main benefits of the technology? By creating a vivid visual representation of liver tissue stiffness, MRE helps radiologists deliver a more confident diagnosis. Both comprehensive and noninvasive, the technique can appeal to patients and referring physicians, and can help expand the role of radiology into new areas. How does MRE work? MRE, a technique developed by Richard Ehman, MD, and colleagues at Mayo Clinic (Rochester, MN), uses low-frequency mechanical waves to probe the elastic properties of tissue. These mechanical waves are generated in the body through an external acoustic driver, which are then imaged using a special phase-contrast MR sequence. GE imagination at work 5

6 Fact Sheet MR-Touch Using a sophisticated mathematical algorithm, the mechanical wave data collected by the MR is then used to generate an elastogram, a diagnostic image that depicts tissue stiffness. How does MRE technology improve care and decrease costs? Improved quality: MRE is noninvasive and provides a color-coded visual representation of tissue stiffness overlaid on the anatomy. Mechanical properties of the liver tissue has been strongly correlated with the extent of fibrosis. 7 MRE technology also improves quality of care due to its attractiveness for use in early diagnosis. Diagnostic value: Sampling variability appears to be one of the major limitations of liver biopsy. 3 Even though small biopsy specimens may be sufficient for diagnostic purposes in certain situations, the possibility that sampling variability exists must be recognized, so that the absence of key findings does not rule out a suspected diagnosis. By showing information about liver stiffness over one or more cross sections of the entire liver, MR elastography provides a more comprehensive view than before available. Patient comfort: Pain and bleeding are the most common complications of liver biopsy, occurring in up to 84% of patients. 3 GE Healthcare Chalfont St. Giles Buckinghamshire UK MRE does not use contrast or ionizing radiation and provides a completely noninvasive test of liver tissue elasticity, thus resulting in greater patient comfort than invasive tests. Reduced cost: Given the novelty of the MRE technology, peer-reviewed academic or medical literature evaluating the potential cost-effectiveness of this noninvasive testing strategy in the diagnosis and management of liver fibrosis is currently limited. At this stage, there is no way to predict the willingness of payers to cover the procedure and the level of reimbursement. As a new technology, MRE is currently not reimbursed with its own CPT code*. If MRE is not reimbursed any more than a typical abdominal MRI scan, the reimbursement for a valid MRE scan would be similar to the 2010 national Medicare average payment rate for an abdominal MRI, i.e. $628 (CPT code 74183). Carlson et al. used data originally reported by Wong et al. and adjusted for inflation using the Consumer Price Index to arrive at an estimated cost of liver biopsy of $1,255. 8,9 MRE has high predictive value in distinguishing stiffness associated with normal liver tissue. 10 If one assumes that the cost of a liver biopsy is $1,255** and the cost of an MRE would be $628, then MRE would lower costs by at least 15% if it successfully avoids approximately two-thirds of unnecessary biopsies. ** Figures associated with US rates of reimbursement. Not globally applicable. ** When procedure-related morbidities are included, the true cost of a biopsy could be several times higher than the reported average cost About healthymagination GE s healthymagination is about better health for more people. We ve committed $6 billion to continuously develop innovations that help clinicians and healthcare providers deliver high-quality healthcare at lower cost to more people around the world. For more information about our healthymagination commitment, visit 1. Kim WR, Brown RS Jr, Terrault NA, et al. Burden of liver disease in the United States: summary of a workshop. Hepatology 2002; 36: Shaheen NJ, Hansen RA, Morgan DR, et al. The burden of gastrointestinal and liver diseases, Am J Gastroenterol. 2006;101: Rockey DC, Celdwell SH, Goodman ZD, et al. Liver Biopsy. American Association for the Study of Liver Disease (AASLD) Position Paper, Bravo AA, Sheth SG, Cehopra S. Liver biopsy. N Engl J Med. 2001; Feb 15;344(7): Smith JO, Sterling RK. Systematic review: non-invasive methods of fibrosis analysis in chronic hepatitis C. Aliment Pharmacol Ther. 2009; Sept 15;30(6): Stebbing J, Farouk L, Panos G, et al. A Meta-analysis of Transient Elastography for the Detection of Hepatic Fibrosis. J Clin Gastroenterol. 2009; Sept 9.;44(3): Yeh WC, Li PC, Jen YM, et al. Elastic modulus measurements of human liver and correlation with pathology. Ultrasound Med Biol 2002;28: Carlson JJ, Kowdley KY, Sullivan SD, Ramsey SD, Veenstra DL. An evaluation of the potential cost-effectiveness of non-invasive testing strategies in the diagnosis of significant liver fibrosis. J Gastroenterol Hepatol May;24(5): Wong J, Bennet W, Koff R, et al. Pretreatment evaluation of chronic hepatitis C: Risk, Benefits, and Costs. JAMA Dec 23-30; 280(24): Yin M, TalWalkar JA, Glaser KJ, et al. Assessment of Hepatic Fibrosis with Magnetic Resonance Elastography. Clin Gastroenterol and Hepatol. 2007;5(10): General Electric Company All rights reserved. General Electric Company reserves the right to make changes in specifications and features shown herein, or discontinue the product described at any time without notice or obligation. GE, GE Monogram, and imagination at work are trademarks of General Electric Company. GE Healthcare, a division of General Electric Company. GE imagination at work CRP EN-US

7 GE Healthcare White Paper healthymagination MRE Magnetic Resonance Elastography The promise of better outcomes and lower costs Vinod S. Palathinkara, PhD, Lloyd Estkowski, and David W. Lee, PhD GE Healthcare Background Chronic liver disease and cirrhosis are major public health problems worldwide. In 2004, these conditions were associated with nearly 40,000 deaths and a cost of at least $1.4 billion for medical services in the U.S. alone. 1,2 These figures are expected to increase due to aging, obesity, and end-stage liver disease caused by chronic hepatitis C. The major biological process responsible for clinical liver disease is progressive hepatic fibrosis. Liver biopsy is the current gold standard for detecting hepatic fibrosis. There are, however, limitations with the technique, which include poor acceptance by patients, measurement errors, and cost. 3,4 Current non-invasive alternatives to liver biopsy are limited to serum-based testing, 5 which is not reliable for detecting early disease, and transient ultrasound elastography, 6 which has technical limitations in patients with obesity and conditions such as ascites. 7,8 GE imagination at work 7

8 healthymagination MRE Magnetic resonance elastography (MRE) Magnetic resonance elastography (MRE), a technique developed by Richard Ehman, MD, and colleagues at Mayo Clinic (Rochester, MN), uses low-frequency mechanical waves to probe the elastic properties of tissue. These mechanical waves are generated in the body through an external acoustic driver, which are then imaged using a special phasecontrast MR sequence. Using a sophisticated mathematical algorithm, the mechanical wave data collected by the MR is then used to generate elastograms diagnostic images that depict a relative stiffness of tissues. MRE gives referring physicians a powerful new option for liver assessment. It is a new tool that provides diagnostic information without the discomfort and risk of complications due to invasive procedures, enabling more frequent evaluation when closer monitoring is needed. By creating a visual representation of liver tissue stiffness, MRE helps radiologists deliver a more confident diagnosis at a lower cost than previous techniques. Both comprehensive and non-invasive, MRE can appeal to patients and referring physicians and can help expand the role of radiology into new areas. More than anything else, MRE holds the promise of better outcomes at lower costs to the healthcare system. Patient management with MRE Multiple studies have shown that when added to a conventional MRI exam of the abdomen, MRE can provide additional information that clinicians need to improve the management of their patients with chronic liver disease. The long-term benefit is in using the information downstream to better utilize liver tests and procedures, and enhance the quality of patient care. MRE provides additional assessment of liver disease beyond routine lab and imaging tests, so that the patients can be more appropriately referred for further diagnosis options such as biopsy. Because liver biopsy is invasive, some patients with suspected liver disease may decline the procedure. As a result, some patients with significant liver disease are not properly identified as eligible candidates for appropriate treatment. MRE can enable referring physicians to assess more patients who may need liver biopsy and to identify patients who present tissue stiffness that is symptomatic of fibrosis. MRE could be a particularly useful tool for physicians to manage patients afflicted with hepatitis B and C, which can often lead to liver injury. Since MRE identifies tissue with elevated liver stiffness, and advanced fibrosis or cirrhosis leads to increased liver stiffness, patients with either type of liver disease can still be evaluated and monitored. MRE could be a better-tolerated, noninvasive method to risk-stratify patients who may have symptoms typical of fibrosis, such as elevated liver stiffness. MRE can also be used to evaluate the need for biopsies or to conduct that first biopsy in the future when evidence typical of hepatic fibrosis first presents on MRE

9 healthymagination MRE Figure 1 (all images below): Typical MR elastograms of normal volunteer and patients. Elastogram on volunteer patient is shown (right) and corresponding anatomic image (left). In the elastogram, relative stiffness is shown on a color scale, ranging from softest (purple) to hardest (red). For reference, a dashed outline has been superimposed on the elastogram to indicate the approximate location of the liver Note that the stiffness of normal liver tissue is very low and similar to that of adipose tissue. The spleen is usually considerably stiffer than other tissues, as shown by the corresponding red areas. A 61-year-old with elevated serum liver tests and nonalcoholic fatty liver disease. In this case of advanced liver fibrosis, the elastogram shows that the liver is much stiffer than subcutaneous tissues and overall stiffness of the liver. The heterogeneity of the stiffness of the liver is also increased (compared to volunteer in images shown above). A 61-year-old with hepatitis C, cirrhosis, and hepatocellular carcinoma.the oval outline in the anatomic image (left) shows the location of the hepatocellular carcinoma. The elastogram (right) shows a corresponding area of high stiffness in the right lobe of the liver (red arrow), as well as an area of very high stiffness in the left lobe of the liver (green arrow) that is consistent with advanced fibrosis. 3 Images courtesy of Dr. Richard Ehman, The Mayo Clinic 9

10 healthymagination MRE Clinical value of MRE Yin et al. evaluated the diagnostic performance of an optimized MRE protocol for assessing hepatic fibrosis among patients with diverse causes of chronic liver disease and in normal individuals. The summary of mean values and variance of liver stiffness from the 35 normal volunteers and 48 patients with chronic liver disease is shown in Figure 2. The mean liver stiffness value for normal individuals was 2.20 ± 0.31 kilopascal (kpa) (range, kpa). For the entire group of patients with varying degrees of chronic liver disease, the mean liver stiffness value was 5.80 ± 2.57 kpa (range, kpa). When assessed by stage of fibrosis, the mean liver stiffness value increased systematically with an excellent correlation between histologic fibrosis and shear stiffness obtained with MR elastography (R 2 = 0.94, P < 0.001) (see Figure 2). Further comparisons between the normal volunteers and the patient groups showed significantly higher mean liver stiffness values by fibrosis groups of F0 1 2, F3, and F4 compared with normal volunteers (P < ) (see Figure 3). Between the mild (F0 1 2) and severe (F3 4) fibrosis groups, the authors also found significant differences in mean liver stiffness measurements (P <.05) (see Figure 3). This study s results supported the hypothesis that MRE is effective for distinguishing normal, soft-liver tissue from stiff fibrotic-liver tissue with a very high negative predictive value. The severity of increased stiffness was shown to allow moderate to severe fibrosis to be distinguished non-invasively from mild fibrosis. It is important to assess the accuracy of MRE in relation to the accuracy of liver biopsy. A review of the available data on the accuracy of needle liver biopsy to define the stage of fibrosis reveals that significant sampling and interpretive error affects the diagnostic accuracy of liver biopsy. Needle liver biopsy evaluates only about 1/50,000 of the volume of the liver, so it may be Mean liver stiffness (kpa) (35) (14) (6) (5) (5) (18) Normal y =.1631 e x r 2 = Stage F0 Stage F1 Stage F2 Stage F3 Chronic liver disease Stage F4 Figure 2: Mean liver shear stiffness measurements for normal volunteers and patients. Mean liver stiffness increases with the increased fibrosis stage in patients. Shown is a summary of the mean shear stiffness measurements of the liver for the 35 normal volunteers and the 48 patients divided into the five different fibrosis stages, which are indicated as F0, F1... F4. Liver stiffness is significantly higher in patients than in the control group. The error bar for each group also illustrates the standard errors for each group. An exponential function fit well to the liver stiffness data with an r2 value of Chart from Yin et al., Gastroenterology and Hepatology,

11 healthymagination MRE affected by substantial sampling error. 10 Autopsy and laparoscopy studies that have evaluated the accuracy of liver biopsy for staging fibrosis and diagnosing cirrhosis have clearly shown that cirrhosis is missed on a single blind liver biopsy in 10% to 30% of cases. 11, 12, 13, 14, 15 The majority of this error is due to the under-staging of disease. Both the size of the biopsy and number of biopsies taken have a major effect on accuracy. Abdi et al. report that the correct diagnosis of cirrhosis with a single biopsy increased from 80% to 100% when three specimens were analyzed. 16 Similarly, in a study that evaluated the agreement between three biopsies taken at a single setting, Maharaj reported that cirrhosis was identified in all three biopsies in only 50% of the cases. 17 Rocky et al. suggest that sampling variability appears to be one of the major limitations of liver biopsy. 3 In a study of 124 patients with chronic HCV infection who underwent laparoscopy-guided left and right lobe liver biopsies, 33% of cases had discordant results by at least one histological stage. A smaller but substantial proportion of biopsies were discordant by at least two stages. Similarly, a single liver biopsy specimen may fail to distinguish steatohepatitis from simple steatosis and may mis-stage the disease by one, or less frequently, two stages if the specimen is much smaller than 2 cm. The authors caution that although even small biopsy specimens may be sufficient for diagnostic purposes in certain situations, the possibility that sampling variability exists must be recognized, so that the absence of key findings does not rule out a suspected diagnosis. By showing information about liver stiffness over one or more cross sections of the entire liver, MRE provides a more comprehensive view than before available. Liver stiffness (kpa) Liver patient Normal Stage 0 Stage 1 Stage 2 Stage 3 Stage 4 Normal P <.0001* P <.0001* Chronic liver disease Figure 3: Mean liver shear stiffness at different fibrosis stages. P <.0001* 5 Normal FO-1-2 F3-4 Kruskal Wallis Dunnett s Test a=0.5 Liver stiffness increases significantly with increased fibrosis extent as determined by liver biopsy examination. In the left diagram, significant differences (*) were observed in the liver stiffness between the normal control group and patient groups F0 1 2, F3, and F4. The P values all are less than The CI diamonds are shown for each group. In the right diagram, a significant difference also was observed between the mild fibrosis groups (F0 1 2) and the severe fibrosis groups (F3 4). The P value is less than.05. The center and the radius of the three circles indicate the mean and SD of the normal, F0 1 2, and F3 4 fibrosis groups. The data were analyzed with a Kruskal Wallis test followed by the Dunnett test. Chart from Yin et al., Gastroenterology and Hepatology,

12 healthymagination MRE Improved patient comfort and safety with MRE According to Rocky et al., pain is the most common complication of liver biopsy, occurring in up to 84% of patients. 3 The most important complication of a liver biopsy is bleeding. Severe bleeding requires hospitalization and increases the likelihood of transfusion or even radiological intervention or surgery. Such bleeding has been estimated to occur in between 1 in 2500 to 1 in 10,000 biopsies. Less severe bleeding, defined as that sufficient to cause pain or reduced blood pressure, but not requiring transfusion or intervention, occurs in approximately 1 in 500 biopsies. Mortality after liver biopsy is usually related to hemorrhage and is very uncommon. The most commonly quoted mortality rate is approximately 18, 19 1 in 10,000 to 1 in 12,000. MRE does not use contrast or ionizing radiation and provides a completely non-invasive test of liver tissue elasticity, thus resulting in high patient comfort. According to Ehman et al., the vibration used in MRE has amplitude that is very small (typically less than 0.1 mm) and does not cause discomfort to the patient. 20 Complications Risk Any pain 1:4 Significant pain 1:10 1:20 Bleeding 1:100 Bile leak 1:1,000 Death 1:10,000 1:12,000 Figure 4: Major complications of liver biopsy

13 healthymagination Health care system costs Carlson et al. use data originally reported by Wong et al. and adjusts for inflation using Consumer Price Index to arrive at an estimated cost of liver biopsy of $1,255,* but this estimate is based on cost rather than charge and does not include all expenses associated with the test. It also understates the true costs of a liver biopsy because it excludes procedure-related morbidities. 7, 21 Myers et al. use administrative databases from a large Canadian Health Region to identify percutaneous liver biopsies performed between 1994 and The study found that between 1994 and 2002, 3627 patients had 4275 liver biopsies. Thirty-two patients (0.75%) had significant biopsy related complications. Pain requiring admission (0.51%) and bleeding (0.35%) were most common. Six patients (0.14%) died; all had malignancies. The median direct cost of a hospitalization for complications was $4579 Canadian (range $1164-$29,641). As a new technology, MRE is currently not reimbursed as a standalone test with its own CPT code. Because the acquisition time for MRE is very short, the technique can be readily included in the protocol for an already-indicated abdominal MRI exam with little impact on the typical examination time of 30 to 45 minutes. If the entire cost of such an exam is attributed to the MRE procedure, then a conservative estimate of the cost of the MRE would be equivalent to the 2010 national Medicare average payment amount for abdominal MRI, i.e., $628 (CPT code 74183). At this stage, there is no way to predict the willingness of payers to cover an MRI examination conducted solely to perform MRE. To better quantify the costs associated with MRE and liver biopsy, a decisionanalytic model comparing diagnostic costs was constructed. 23 A targeted literature review was conducted and, in addition, a leading hepatologist and pathologist were consulted to identify the appropriate procedure codes associated with liver biopsy. The study assumed that MRE would be reimbursed using CPT-4 code (magnetic resonance [e.g., proton] imaging, abdomen; without contrast material[s]). All appropriate allowable charges were assigned to the identified procedure codes using the 2010 Medicare Physician Fee Schedule. Please note that all costs discussed here are US-based costs and are not globally applicable. Based on the model, the cost of a guided liver biopsy was $1,424 (ultrasound $164, surgical $881, pathology $347, laboratory $32) and the cost of an MRE (without contrast) was $946 (hospital setting) or $666 (non-hospital setting). Given the novelty of MRE technology, peer-reviewed academic/medical literature evaluating the potential cost-effectiveness of this non-invasive testing strategy in the diagnosis and management of liver fibrosis is not yet available. Nevertheless, scenario-based analysis of published comparisons of patients who had both biopsy and MRE is illustrative and insightful. *Figures associated with US rates of reimbursement. Not applicable globally. Significant complications were identified by reviewing medical records of patients hospitalized within seven days of a biopsy and those with a diagnostic code indicative of procedural complications. MRE is currently not reimbursed on its own CPT code

14 healthymagination MRE Scenario analysis Huwart et al. performed a blind comparison of MRE and liver biopsy for non-invasive staging of liver fibrosis and reported histopathologic staging of liver fibrosis according to the METAVIR scoring system as the reference. 24 The study analyzed 96 patients for whom both MRE and liver biopsy were performed (see Figure 5). It should be noted that the initial sample had 141 patients from whom liver biopsy specimens were collected, but only 127 liver biopsy specimens were suitable for fibrosis staging. This suggests that approximately 10% of the samples from a biopsy specimen may be unsuitable for staging. We consider three scenarios as a hypothetical example to illustrate the costs of performing MRE for evaluation of liver disease. In scenario 1, we assume that liver biopsy is 100% accurate and that the discrepancy in staging between MRE and liver biopsy is entirely due to the errors in MRE. From a cost perspective, this would mean that at some time, these patients would need some follow-up to get a definitive diagnosis. Since we are using biopsy as the reference standard, the cost for the follow-up would be assumed to be the cost of a biopsy. Biopsy (96) F0(22) F1-2(41) F3-4(33) Outcome Elevated liver enzymes Conforming outcome MRE (96) (24) Misclassification Follow-up Figure 5: Flow diagram of patients who underwent liver biopsy and MRE (Huwart et al.)

15 healthymagination MRE Huwart et al. report that when comparing results from MRE and biopsy, 24 of 96 (25%) were misclassified in their stage of fibrosis. Based on the reported sensitivity and specificity of MRE techniques, this misclassification is unusually high. Nevertheless, since biopsy is the reference standard, we assume that 25% of patients who underwent MRE would eventually require further evaluation, probably with a biopsy. We do not consider how one would identify the patients who are misclassified and ignore the impact on the outcome or the additional treatment costs due to misclassification. Our attempt here is to illustrate a methodology that can provide directional information on costs to explore potential for cost reduction, rather than to establish or estimate actual cost differentials. In scenario 1, we assume that biopsy is 100% accurate and all samples are good enough to make a diagnosis. We consider this as a worst-case scenario for MRE. The authors highlight the fact that liver biopsy is not an optimal reference examination and that they do not know if the reported discordant results between MRE and histopathology were caused by problems of inadequate biopsy sampling.* The authors also report that the two pathologists who reviewed the biopsy specimen were initially in agreement only on 81 of the 96 samples. Nevertheless, since biopsy is the reference standard, we have to assume that it provides clinically accepted basis for comparison. Elevated liver enzymes Biopsy (96) MRE (96) F0(22) F1-2(41) Outcome F3-4(33) Conforming outcome Scenario 1 Biopsy is 100% accurate; all samples are suitable. Costs for procedures: Biopsy = 96 x $1424 = $136,704 MRE = 96 x $ x $1424 = $123,100 Cost differential = $13,604, or 10% less than biopsy (24) Misclassification Follow-up Figure 6: Hypothetical examples of direct costs of MRE and liver biopsy for scenario 1. *In 26 of the 96 samples, the biopsy specimen was less than 25 mm

16 healthymagination MRE As reported by Huwart et al., 14 of 141 samples were unsuitable for diagnosis. Even though there are many studies that recommend a minimum sample length of 25 mm, the study reports that 23% of the samples were less than 25 mm in length. Thus the scenario that a few biopsy samples would be unsuitable is realistic. In scenario 2, we assume that to get 96 good samples, one would need to do 10% more samples (105.6 biopsies). This assumption does not imply that these patients would undergo an immediate repeat biopsy. The cost of this may result in an increased cost of diagnosis per person. Elevated liver enzymes Unsuitable sample (9.6) Biopsy (105.6) MRE (96) F0(22) F1-2(41) Outcome F3-4(33) Conforming outcome Scenario 2 Biopsy is 100% accurate, but 10% samples are unsuitable. Costs for procedures: Biopsy = 96 x $ x 1424 = $136,704 + $13,670 = $150,374 MRE = $123, x $1424 = $126,518 Cost differential = $23,856, or 16% less than biopsy (24) Misclassification Follow-up Figure 7: Hypothetical example of direct costs of MRE and liver biopsy for scenario

17 healthymagination MRE In scenario 3, we take into account biopsy leading to a misclassification. More than questioning the accuracy of the biopsy, this is reflective of the fact that biopsy is a sampling technique. Studies report biopsy mis-staging to be in the range of 10% to 33%. In this case, we assume 20% mis-staging. If 20% of biopsy samples are mis-staged, then the discordances between MRE and biopsy may decrease. However, for simplicity, we still assume that the discordance between MRE and biopsy would not change. Conclusion MRE is non-invasive and provides tissue stiffness information for the entire liver. It avoids the discomfort and risk of complications associated with other invasive procedure. In addition, elastograms that overlay tissue stiffness images over the anatomy avoid sampling error and provide richer information that could assist in diagnosis. Studies show that the technique has excellent sensitivity in depicting the elevated stiffness associated with hepatic fibrosis. Stiffness of normal liver tissue is very soft and comparable to that of subcutaneous fat. Studies have also shown that hepatic steatosis, a common condition, does not have a significant influence on liver stiffness and therefore does not confound the elastographic findings observed in fibrosis. In summary, the evidence supports the use of MRE as a triaging option for liver biopsy. The accuracy and the noninvasive nature of the technology offer the promise that MRE could improve outcomes, potentially at lower costs. Elevated liver enzymes Unsuitable sample (9.6) Biopsy (105.6) MRE (96) F0(22) Outcome F1-2(41) F3-4(33) (19.2) Follow-up Conforming outcome Scenario 3 Biopsy is only 80% accurate, but MRE still has 25% misclassifications. Costs for procedures: Biopsy = $136, x 96 x $1424 = $136,704 + $27,341 = $164,045 MRE = $123,100 Cost differential = $40,945, or 25% less than biopsy (24) Misclassification Follow-up Figure 8: Hypothetical examples of direct costs of MRE and liver biopsy for scenario

18 References 1 Kim, W.R., et al. Burden of liver disease in the United States: summary of a workshop. Hepatology 36, (2002). 2 Shaheen, N.J., et al. The burden of gastrointestinal and liver diseases. Am J Gastroenterol 101, (2006). 3 Rockey, D.C., et al. Liver biopsy. American Association for the Study of Liver Disease (AASLD), position paper, Bravo, A.A., et al. Liver biopsy. N Engl J Med 344(7), (2001). 5 Smith, J.O., Sterling, R.K. Systematic review: non-invasive methods of fibrosis analysis in chronic hepatitis C. Aliment Pharmacol Ther 30(6), (2009). 6 Stebbing, J., et al. A meta-analysis of transient elastography for the detection of hepatic fibrosis. J Clin Gastroenterol 44(3), (2010). 7 Carlson, J.J., et al. An evaluation of the potential cost-effectiveness of non-invasive testing strategies in the diagnosis of significant liver fibrosis. J Gastroenterol Hepatol 24(5), (2009). 8 Talwalkar, J.A. Elastography for detecting hepatic fibrosis: options and considerations. Gastroenterology 359(1), (2008). 9 Yin, M,. et al. Assessment of hepatic fibrosis with magnetic resonance elastography. Clin Gastroentero 5(10), (2007). 10 Afdhal, N.H., Nunes, D. Evaluation of liver fibrosis: a concise review. Am Gastroentero, 99(6), (2004). 11 Bruguera, M., et al. A comparison of the accuracy of peritoneoscopy and liver biopsy in the diagnosis of cirrhosis. Gut 15, (1974). 12 Poniachik, J., et al. The role of laparoscopy in the diagnosis of cirrhosis. Gastrointest Endosc 43, (1996). 13 Pagliaro, L., et al. Percutaneous blind biopsy versus laparoscopy with guided biopsy in diagnosis of cirrhosis. Dig Dis Sci, 28, (1983). 14 Olsson, R., et al. Sampling variability of percutaneous liver biopsy in primary sclerosing cholangitis. J Clin Pathol 48, (1995). 15 Angelucci, E., et al. Needle liver biopsy in thalassemia: analyses of the diagnostic accuracy and safety in 1184 consecutive biopsies. Br J Haematol 89, (1995). 16 Abdi, W., et al. Sampling variability on percutaneous liver biopsy. Arch Intern Med 15, (1979). 17 Maharaj, B., et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet 1, (1986). 18 Perrault, J., et al. Liver biopsy: complications in 1000 inpatients and outpatients. Gastroenterology 74, (1978). 19 McGill, D.B., et al. A 21-year experience with major hemorrhage after percutaneous liver biopsy. Gastroenterology 99, (1990). 20 Ehman, E.C., et al. Vibration safety limits for magnetic resonance elastography. Phys Med Biol 53(4), (2008). 21 Wong, J., et al. Pretreatment evaluation of chronic hepatitis C: risk, benefits, and costs. JAMA 280, (1998). 22 Myers, R.P., et al. Utilization rates, complications and costs of percutaneous liver biopsy: a population-based study including 4275 biopsies. Liver Int, (2008). 23 DeKoven, M. Cost comparison: liver biopsy versus abdominal MRI. Memo to GE Healthcare from IMS Health Incorporated. May 25, Huwart, L., et al. Magnetic resonance elastography for the non-invasive staging of liver fibrosis. Gastroenetrology, 135(1), 32-40, (2008)

19 2010 General Electric Company All rights reserved. General Electric Company reserves the right to make changes in specifications and features shown herein, or discontinue the product described at any time without notice or obligation. GE, GE Monogram, and imagination at work are trademarks of General Electric Company. GE Healthcare, a division of General Electric Company. White Paper healthymagination About GE Healthcare GE Healthcare provides transformational medical technologies and services that are shaping a new age of patient care. Our broad expertise in medical imaging and information technologies, medical diagnostics, patient monitoring systems, drug discovery, biopharmaceutical manufacturing technologies, performance improvement and performance solutions services helps our customers to deliver better care to more people around the world at a lower cost. In addition, we partner with healthcare leaders, striving to leverage the global policy change necessary to implement a successful shift to sustainable healthcare systems. Our healthymagination vision for the future invites the world to join us on our journey as we continuously develop innovations focused on reducing costs, increasing access, and improving quality around the world. Headquartered in the United Kingdom, GE Healthcare is a unit of General Electric Company (NYSE: GE). Worldwide, GE Healthcare employees are committed to serving healthcare professionals and their patients in more than 100 countries. For more information about GE Healthcare, visit our website at GE Healthcare 9900 Innovation Drive Wauwatosa, WI U.S.A. GE imagination at work

20 MAGNETIC RESONANCE ELASTOGRAPHY WHITE PAPER The promise of better outcomes and lower costs Vinod S. Palathinkara, PhD, Lloyd Estkowski, and David W. Lee, PhD GE Healthcare Background Chronic liver disease and cirrhosis are major public health problems worldwide. In 2004, these conditions were associated with nearly 40,000 deaths and a cost of at least $1.4 billion for medical services in the U.S. alone. 1,2 These figures are expected to increase due to aging, obesity, and end-stage liver disease caused by chronic hepatitis C. The major biological process responsible for clinical liver disease is progressive hepatic fibrosis. Liver biopsy is the current gold standard for detecting hepatic fibrosis. There are, however, limitations with the technique, which include poor acceptance by patients, measurement errors, and cost. 3,4 Current non-invasive alternatives to liver biopsy are limited to serum-based testing, 5 which is not reliable for detecting early disease, and transient ultrasound elastography, 6 which has technical limitations in patients with obesity and conditions such as ascites. 7,

21 WHITE P PER M NETIC RESON NCE EL O PHY Magnetic resonance elastography (MRE) Magnetic resonance elastography (MRE), a technique developed by Richard Ehman, MD, and colleagues at Mayo Clinic (Rochester, MN), uses low-frequency mechanical waves to probe the elastic properties of tissue. These mechanical waves are generated in the body through an external acoustic driver, which are then imaged using a special phase-contrast MR sequence. Using a sophisticated mathematical algorithm, the mechanical wave data collected by the MR is then used to generate elastograms diagnostic images that depict relative stiffness of tissues. MRE gives referring physicians a powerful new option for liver assessment. It is a new tool that provides diagnostic information without the discomfort and risk of complications due to invasive procedures, enabling more frequent evaluation when closer monitoring is needed. By creating a visual representation of liver tissue stiffness, MRE helps radiologists deliver a more confident diagnosis at a lower cost than previous techniques. Both comprehensive and non-invasive, MRE can appeal to patients and referring physicians and can help expand the role of radiology into new areas. More than anything else, MRE holds the promise of better outcomes at lower costs to the healthcare system

22 MAGNETIC RESONANCE ELASTOGRAPHY WHITE PAPER How elastography works The image is captured in as little as 14 seconds, or one breath hold, in three steps: A simple, drum-like driver generates acoustic waves within the tissue of interest. A special MRI technique images minute displacements of the tissue that result from wave propagation. An advanced mathematical technique generates maps of tissue stiffness, known as elastograms. Patient management with MRE Multiple studies have shown that when added to a conventional MRI exam of the abdomen, MRE can provide additional information that clinicians need to improve the management of their patients with chronic liver disease. The long-term benefit is in using the information downstream to better utilize liver tests and procedures, and enhance the quality of patient care. MRE provides additional assessment of liver disease beyond routine lab and imaging tests, so that the patients can be more appropriately referred for further diagnosis options such as biopsy. Because liver biopsy is invasive, some patients with suspected liver disease may decline the procedure. As a result, some patients with significant liver disease are not properly identified as eligible candidates for appropriate treatment. MRE can enable referring physicians to assess more patients who may need liver biopsy and to identify patients who present tissue stiffness that is symptomatic of fibrosis. MRE could be a particularly useful tool for physicians to manage patients afflicted with hepatitis B and C, which can often lead to liver injury. Since MRE identifies tissue with elevated liver stiffness, and advanced fibrosis or cirrhosis leads to increased liver stiffness, patients with either type of liver disease can still be evaluated and monitored. MRE could be a better-tolerated, noninvasive method to riskstratify patients who may have symptoms typical of fibrosis, such as elevated liver stiffness. MRE can also be used to evaluate the need for biopsies or to conduct that first biopsy in the future when evidence typical of hepatic fibrosis first presents on MRE

23 WHITE PAPER MAGNETIC RESONANCE ELASTOGRAPHY Figure 1. (all images below): Typical MR elastograms of normal volunteer and patients. Elastogram on volunteer patient is shown (right) and corresponding anatomic image (left). In the elastogram, relative stiffness is shown on a color scale, ranging from softest (purple) to hardest (red). For reference, a dashed outline has been superimposed on the elastogram to indicate the approximate location of the liver Note that the stiffness of normal liver tissue is very low and similar to that of adipose tissue. The spleen is usually considerably stiffer than other tissues, as shown by the corresponding red areas. A 61-year-old with elevated serum liver tests and nonalcoholic fatty liver disease. In this case of advanced liver fibrosis, the elastogram shows that the liver is much stiffer than subcutaneous tissues and overall stiffness of the liver. The heterogeneity of the stiffness of the liver is also increased (compared to volunteer in images shown above). A 61-year-old with hepatitis C, cirrhosis, and hepatocellular carcinoma.the oval outline in the anatomic image (left) shows the location of the hepatocellular carcinoma. The elastogram (right) shows a corresponding area of high stiffness in the right lobe of the liver (red arrow), as well as an area of very high stiffness in the left lobe of the liver (green arrow) that is consistent with advanced fibrosis. Images courtesy of Dr. Richard Ehman, The Mayo Clinic 4 123

24 MAGNETIC RESONANCE ELASTOGRAPHY WHITE PAPER Clinical value of MRE Yin et al.. 9 Figure P Pa). For the.. Pa.. Pa). 2. P <.. - P <.. P <.. value.. I.. Needle.. The.. Abdi et al. report that the. of the cases. Mean liver stiffness (kpa) (35) (14) (6) (5) (5) (18) Normal y =.1631 e x r 2 = Stage F0 Stage F1 Stage F2 Stage F3 Chronic liver disease Stage F4 Figure F4. Liver the control group. The error bar for each group also illustrates the standard errors for each group..94. Chart from Yin et al., Gastroenterology and Hepatology,

25 WHITE PAPER MAGNETIC RESONANCE ELASTOGRAPHY Liver stiffness (kpa) Liver patient Normal Stage 0 Stage 1 Stage 2 Stage 3 Stage 4 Normal P <.0001* P <.0001* P <.0001* Chronic liver disease Normal FO-1-2 F3-4 Kruskal Wallis Dunnett s Test a=0.5 Liver stiffness increases significantly with increased fibrosis extent as determined by liver biopsy examination. In the left diagram, significant differences (*) were observed in the liver stiffness between the normal control group and patient groups F0 1 2, F3, and F4. The P values all are less than The CI diamonds are shown for each group. In the right diagram, a significant difference also was observed between the mild fibrosis groups (F0 1 2) and the severe fibrosis groups (F3 4). The P value is less than.05. The center and the radius of the three circles indicate the mean and SD of the normal, F0 1 2, and F3 4 fibrosis groups. The data were analyzed with a Kruskal Wallis test followed by the Dunnett test. Chart from Yin et al., Gastroenterology and Hepatology, Figure 3. Mean liver shear stiffness at different fibrosis stages. Rocky et al. suggest that sampling variability appears to be one of the major limitations of liver biopsy. 3 In a study of 124 patients with chronic HCV infection who underwent laparoscopy-guided left and right lobe liver biopsies, 33% of cases had discordant results by at least one histological stage. A smaller but substantial proportion of biopsies were discordant by at least two stages. Similarly, a single liver biopsy specimen may fail to distinguish steatohepatitis from simple steatosis and may mis-stage the disease by one, or less frequently, two stages if the specimen is much smaller than 2 cm. The authors caution that although even small biopsy specimens may be sufficient for diagnostic purposes in certain situations, the possibility that sampling variability exists must be recognized, so that the absence of key findings does not rule out a suspected diagnosis. By showing information about liver stiffness over one or more cross sections of the entire liver, MRE provides a more comprehensive view than before available. Improved patient comfort and safety with MRE According to Rocky et al., pain is the most common complication of liver biopsy, occurring in up to 84% of patients. 3 The most important complication of a liver biopsy is bleeding. Severe bleeding requires hospitalization and increases the likelihood of transfusion or even radiological intervention or surgery. Such bleeding has been estimated to occur in between 1 in 2500 to 1 in 10,000 biopsies. Less severe bleeding, defined as that sufficient to cause pain or reduced blood pressure, but not requiring transfusion or intervention, occurs in approximately 1 in 500 biopsies. Mortality after liver biopsy is usually related to hemorrhage and is very uncommon. The most commonly quoted mortality rate is approximately 1 in 10,000 to 1 in 12, ,19 Complications Risk Any pain 1:4 Significant pain 1:10 1:20 Bleeding 1:100 Bile leak 1:1,000 Death 1:10,000 1:12,000 Figure 4. Major complications of liver biopsy. 6 MRE does not use contrast or ionizing radiation and provides a completely non-invasive test of liver tissue elasticity, thus resulting in high patient comfort. According to Ehman et al., the vibration used in MRE has amplitude that is very small (typically less than 0.1 mm) and does not cause discomfort to the patient

26 MAGNETIC RESONANCE ELASTOGRAPHY WHITE PAPER Health care system costs Carlson et al. use data originally reported by Wong et al. and adjusts for inflation using Consumer Price Index to arrive at an estimated cost of liver biopsy of $1,255,* but this estimate is based on cost rather than charge and does not include all expenses associated with the test. It also understates the true costs of a liver biopsy because it excludes procedure-related morbidities. 7, 21 Myers et al. use administrative databases from a large Canadian Health Region to identify percutaneous liver biopsies performed between 1994 and The study found that between 1994 and 2002, 3627 patients had 4275 liver biopsies. Thirty-two patients (0.75%) had significant biopsy related complications. Pain requiring admission (0.51%) and bleeding (0.35%) were most common. Six patients (0.14%) died; all had malignancies. The median direct cost of a hospitalization for complications was $4579 Canadian (range $1164 $29,641). As a new technology, MRE is currently not reimbursed as a standalone test with its own CPT code. Because the acquisition time for MRE is very short, the technique can be readily included in the protocol for an already-indicated abdominal MRI exam with little impact on the typical examination time of 30 to 45 minutes. If the entire cost of such an exam is attributed to the MRE procedure, then a conservative estimate of the cost of the MRE would be equivalent to the 2010 national Medicare average payment amount for abdominal MRI, i.e., $628 (CPT code 74183). At this stage, there is no way to predict the willingness of payers to cover an MRI examination conducted solely to perform MRE. To better quantify the costs associated with MRE and liver biopsy, a decision-analytic model comparing diagnostic costs was constructed. 23 A targeted literature review was conducted and, in addition, a leading hepatologist and pathologist were consulted to identify the appropriate procedure codes associated with liver biopsy. The study assumed that MRE would be reimbursed using CPT-4 code (magnetic resonance [e.g., proton] imaging, abdomen; without contrast material[s]). All appropriate allowable charges were assigned to the identified procedure codes using the 2010 Medicare Physician Fee Schedule. Please note that all costs discussed here are US-based costs and are not globally applicable. Based on the model, the cost of a guided liver biopsy was $1,424 (ultrasound $164, surgical $881, pathology $347, laboratory $32) and the cost of an MRE (without contrast) was $946 (hospital setting) or $666 (non-hospital setting). Given the novelty of MRE technology, peer-reviewed academic/medical literature evaluating the potential cost-effectiveness of this non-invasive testing strategy in the diagnosis and management of liver fibrosis is not yet available. Nevertheless, scenario-based analysis of published comparisons of patients who had both biopsy and MRE is illustrative and insightful. *Figures associated with US rates of reimbursement. Not applicable globally. Significant complications were identified by reviewing medical records of patients hospitalized within seven days of a biopsy and those with a diagnostic code indicative of procedural complications. MRE is currently not reimbursed on its own CPT code

27 WHITE PAPER MAGNETIC RESONANCE ELASTOGRAPHY Biopsy (96) F0(22) F1-2(41) F3-4(33) Outcome Elevated liver enzymes Conforming outcome MRE (96) (24) Misclassification Follow-up Figure 5. Flow diagram of patients who underwent liver biopsy and MRE (Huwart et al.). Scenario analysis Huwart et al. performed a blind comparison of MRE and liver biopsy for non-invasive staging of liver fibrosis and reported histopathologic staging of liver fibrosis according to the METAVIR scoring system as the reference. 24 The study analyzed 96 patients for whom both MRE and liver biopsy were performed (see Figure 5). It should be noted that the initial sample had 141 patients from whom liver biopsy specimens were collected, but only 127 liver biopsy specimens were suitable for fibrosis staging. This suggests that approximately 10% of the samples from a biopsy specimen may be unsuitable for staging. We consider three scenarios as a hypothetical example to illustrate the costs of performing MRE for evaluation of liver disease. In scenario 1, we assume that liver biopsy is 100% accurate and that the discrepancy in staging between MRE and liver biopsy is entirely due to the errors in MRE. From a cost perspective, this would mean that at some time, these patients would need some follow-up to get a definitive diagnosis. Since we are using biopsy as the reference standard, the cost for the follow-up would be assumed to be the cost of a biopsy. Huwart et al. report that when comparing results from MRE and biopsy, 24 of 96 (25%) were misclassified in their stage of fibrosis. Based on the reported sensitivity and specificity of MRE techniques, this misclassification is unusually high. Nevertheless, since biopsy is the reference standard, we assume that 25% of patients who underwent MRE would eventually require further evaluation, probably with a biopsy. We do not consider how one would identify the patients who are misclassified and ignore the impact on the outcome or the additional treatment costs due to delay in misclassification. Our attempt here is to illustrate a methodology that can provide directional information on costs to explore potential for cost reduction, rather than to establish or estimate actual cost differentials

28 MAGNETIC RESONANCE ELASTOGRAPHY WHITE PAPER I.... N. Scenario 1. C C.... In scenario.... Scenario 2. C.. C Biopsy (96) F0(22) F1-2(41) F3-4(33) Outcome Unsuitable sample (9.6) Biopsy (105.6) F0(22) F1-2(41) F3-4(33) Outcome Elevated liver enzymes MRE (96) (24) Conforming outcome Misclassification Follow-up Figure 6.. Elevated liver enzymes MRE (96) Conforming outcome (24) Misclassification Follow-up Figure 7.. *I

29 WHITE PAPER MAGNETIC RESONANCE ELASTOGRAPHY 10 In scenario 3, we take into account biopsy leading to a misclassification. More than questioning the accuracy of the biopsy, this is reflective of the fact that biopsy is a sampling technique. Studies report biopsy mis-staging to be in the range of 10% to 33%. In this case, we assume 20% mis-staging. If 20% of biopsy samples are mis-staged, then the discordances between MRE and biopsy may decrease. However, for simplicity, we still assume that the discordance between MRE and biopsy would not change. Scenario 3 Biopsy is only 80% accurate, but MRE still has 25% misclassifications. Costs for procedures: Biopsy = $136, x 96 x $1424 Unsuitable sample (9.6) Biopsy (105.6) Elevated liver enzymes = $136,704 + $27,341 = $164,045 MRE = $123,100 Cost differential = $40,945, or 25% less than biopsy MRE (96) (24) F0(22) F1-2(41) F3-4(33) Outcome (19.2) Conforming outcome Misclassification Follow-up Follow-up Figure 8. Hypothetical examples of direct costs of MRE and liver biopsy for scenario 3. Conclusion MRE is non-invasive and provides tissue stiffness information for the entire liver. It avoids the discomfort and risk of complications associated with other invasive procedure. In addition, elastograms that overlay tissue stiffness images over the anatomy avoid sampling error and provide richer information that could assist in diagnosis. Studies show that the technique has excellent sensitivity in depicting the elevated stiffness associated with hepatic fibrosis. Stiffness of normal liver tissue is very soft and comparable to that of subcutaneous fat. Studies have also shown that hepatic steatosis, a common condition, does not have a significant influence on liver stiffness and therefore does not confound the elastographic findings observed in fibrosis. In summary, the evidence supports the use of MRE as a triaging option for liver biopsy. The accuracy and the noninvasive nature of the technology offer the promise that MRE could improve outcomes, potentially at lower costs. References 1 Kim, W.R., et al. Burden of liver disease in the United States: summary of a workshop. Hepatology 36, (2002). 2 Shaheen, N.J., et al. The burden of gastrointestinal and liver diseases. Am J Gastroenterol 101, (2006). 3 Rockey, D.C., et al. Liver biopsy. American Association for the Study of Liver Disease (AASLD), position paper, Bravo, A.A., et al. Liver biopsy. N Engl J Med 344(7), (2001). 5 Smith, J.O., Sterling, R.K. Systematic review: non-invasive methods of fibrosis analysis in chronic hepatitis C. Aliment Pharmacol Ther 30(6), (2009). 6 Stebbing, J., et al. A meta-analysis of transient elastography for the detection of hepatic fibrosis. J Clin Gastroenterol 44(3), (2010). 7 Carlson, J.J., et al. An evaluation of the potential cost-effectiveness of non-invasive testing strategies in the diagnosis of significant liver fibrosis. J Gastroenterol Hepatol 24(5), (2009). 8 Talwalkar, J.A. Elastography for detecting hepatic fibrosis: options and considerations. Gastroenterology 359(1), (2008). 9 Yin, M,. et al. Assessment of hepatic fibrosis with magnetic resonance elastography. Clin Gastroentero 5(10), (2007). 10 Afdhal, N.H., Nunes, D. Evaluation of liver fibrosis: a concise review. Am Gastroentero, 99(6), (2004). 11 Bruguera, M., et al. A comparison of the accuracy of peritoneoscopy and liver biopsy in the diagnosis of cirrhosis. Gut 15, (1974). 12 Poniachik, J., et al. The role of laparoscopy in the diagnosis of cirrhosis. Gastrointest Endosc 43, (1996). 13 Pagliaro, L., et al. Percutaneous blind biopsy versus laparoscopy with guided biopsy in diagnosis of cirrhosis. Dig Dis Sci, 28, (1983). 14 Olsson, R., et al. Sampling variability of percutaneous liver biopsy in primary sclerosing cholangitis. J Clin Pathol 48, (1995). 15 Angelucci, E., et al. Needle liver biopsy in thalassemia: analyses of the diagnostic accuracy and safety in 1184 consecutive biopsies. Br J Haematol 89, (1995). 16 Abdi, W., et al. Sampling variability on percutaneous liver biopsy. Arch Intern Med 15, (1979). 17 Maharaj, B., et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet 1, (1986). 18 Perrault, J., et al. Liver biopsy: complications in 1000 inpatients and outpatients. Gastroenterology 74, (1978). 19 McGill, D.B., et al. A 21-year experience with major hemorrhage after percutaneous liver biopsy. Gastroenterology 99, (1990). 20 Ehman, E.C., et al. Vibration safety limits for magnetic resonance elastography. Phys Med Biol 53(4), (2008). 21 Wong, J., et al. Pretreatment evaluation of chronic hepatitis C: risk, benefits, and costs. JAMA 280, (1998). 22 Myers, R.P., et al. Utilization rates, complications and costs of percutaneous liver biopsy: a population-based study including 4275 biopsies. Liver Int, (2008). 23 DeKoven, M. Cost comparison: liver biopsy versus abdominal MRI. Memo to GE Healthcare from IMS Health Incorporated. May 25, Huwart, L., et al. Magnetic resonance elastography for the non-invasive staging of liver fibrosis. Gastroenetrology, 135(1), 32-40, (2008). 129

30 GE Healthcare Sound diagnosis has a new look. MR Elastography a picture of confidence Today, an innovative technique known as MR elastography can capture a compelling visual image of the liver, using sound waves to detect the stiffness of tissue that can indicate liver disease. How elastography works The image is captured in as little as 14 seconds, or one breath hold, in three steps: A simple, drum-like driver generates acoustic waves within the tissue of interest. A special MRI technique images minute displacements of the tissue that result from wave propagation. An advanced mathematical technique generates maps of tissue stiffness, known as elastograms. 1 30

31 MR-Touch MR-Touch, a visual palpation device from GE Healthcare, is the first commercial product for MR elastography. hard soft Conventional MR liver image of a normal liver MR-Touch elastogram of a normal liver Conventional MR liver image with liver disease MR-Touch elastogram with liver disease For Referring Physicians: A sound alternative MR-Touch gives referring physicians a powerful new option for liver assessment. Its non-invasive examination will help to minimize discomfort and reduce the potential risk of infection, while enabling more frequent evaluation when closer monitoring is needed. For Radiologists: A sound diagnosis By creating a profound visual representation of liver tissue stiffness, MR-Touch helps radiologists deliver a more confident diagnosis. At the same time, it helps expand the role of radiology into new areas. For Administrators: A sound investment MR-Touch enables diagnostic procedures at a lower cost than previous techniques. Both comprehensive and non-invasive, the technique can appeal to patients and referring physicians, helping to drive additional procedures and referrals. It sounds like the future. But it is here today. While MR elastography is an innovative technology, an investment in MR-Touch can bring immediate clinical value to organizations dedicated to clinical excellence. For patients, clinicians and administrators, MR-Touch represents a sound choice today no matter how you look at it. GE Healthcare 3000 North Grandview Waukesha, WI USA 2011 GE Medical Systems, doing business as GE Healthcare. All rights reserved. MR EN-US DOC

32 GE Healthcare The best ideas are yours. You know what you need better than anyone. Guided by your voice, we re making it easier for you to find the right technology and care for your patients. That s why we ve introduced three platforms to help you find the solution you need: Optima, Brivo and Discovery. So keep sharing your ideas and we ll keep listening. Healthcare Re-imagined. Continue the conversation at Discovery Breakthrough technology. Advanced applications. Competitive edge. Brivo Essential functionality. Intuitive design. Operational ease. Optima Optimized productivity. Streamlined workflow. Versatile General Electric Company

33 ISSUE SPOTLI ELASTO P A N Imaging. O MR elastography what is it? I C N I. I I SignaPUL 133

34 MR ELASTOGRAPHY ISSUE SPOTLIGHT Discussion With the advent of MRI. Yet that was just the beginning. Over OL powerful new diagnostic applications. MRE provides a different type of contrast tissue stiffness. Initial exploration of this new capability has focused on diseases that are already known to cause local changes in tissue stiffness.. The addition of MRE to a standard MRI. C. 3 At Mayo C. Dr.. I.. MRE await investigation. Wave I Soft Hard Figure 1: MR elastography is used here to characterize the relative stiffness in soft tissue. Top row: C two different individuals. Center row: Mechanical waves are generated in the. Bottom row: showing the stiffness of tissue. The patient disease. liver stiffness appearance. I C N. Wave I Hard I C N 134 Figure 2: Left: C. Center: Mechanical waves are. Right:. Soft Lloyd Estkowski, MR manager for Body Applications at GE Healthcare, contributed to this article. References:. C P. Vibration safety. P....J. L P.J.. Magnetic resonance elastography by direct visualization of propagating acoustic strain waves MR Elastography inspires new wave of. Diagnostic I. 4.. MR elastography of liver... 49

35 135

36 N By Vinod S. Palathinkara, PhD, Lloyd Estkowski, and David W. Lee, PhD.. I. By.... C. I... alone. C infection

37 TEC NICAL INNO ION NETIC ONANCE ELASTO P L.. C 6. C N.. Using a stiffness. I O... Clinical value of MRE P.. Mean Liver Stiffness (kpa) (35) Normal (14) (6) (5) (5) (18) Stage F0. Stage F1 Stage F2 Stage F3 Stage F4 Chronic Liver Disease Figure 1... I.. N. 9. C SignaPUL 137

38 MAGNETIC ONANCE EL OGRAPHY TECHNICAL INNOVATION I.. Both the size of the biopsy and number of biopsies taken have a major effect on accuracy. Abdi et al. report that the. of the cases. Rockey et al. be one of the major limitations of liver biopsy. In a study C.... mation about liver stiffness over one or more cross sections. Patient comfort of MRE. patients. 3 The most important complication of liver biopsy.. L transfusion or intervention. Mortality after liver biopsy is. 138 Complications Risk Death Bile leak Any pain Figure 2. Major complications of liver biopsy.. the vibration has amplitude in abdominal tissue that. not cause discomfort. Health care system costs limited. of liver diseases. Carlson et al.. Consumer Price Index to this estimate understates the true costs of a liver biopsy because it excludes procedure-related morbidities. Myers et al. Canadian... complications. The median direct cost of a hospitalization C. CP. Because the acquisition time is conventional MRI examination protocol adds very little to. If MRE is not reimbursed any more than a typical abdominal MRI. N. 65

39 TECHNICAL INNOVATION MAGNETIC RESONANCE ELASTOGRAPHY MRE is effective for distinguishing stiffness... with very high negative-predictive value. scan, the reimbursement for a valid MRE scan would be similar to the 2010 national Medicare average payment rate for an abdominal MRI, i.e. $628 (CPT code 74183). At this stage, there is no way to predict the willingness of payers to cover the procedure and the level of reimbursement. MRE has the potential to significantly reduce cost as a triage for liver biopsy. The information MRE provides could be used to assess if and when a patient should undergo liver biopsy. If one assumes that the cost of a liver biopsy is $1,255 and the cost of an MRE would be $628, then MRE would lower costs by at least 15% if it successfully avoids approximately two-thirds of unnecessary biopsies (Figure 3). Yin et al. showed that MRE has high predictive value in distinguishing stiffness associated with normal liver tissue. 8 Even though liver biopsy is accurate in identifying fibrosis, due to its basis as a sampling technique, absence of evidence of fibrosis from biopsy does not rule out fibrosis. MRE may also be a useful tool that helps guide clinicians in localizing the area of biopsy. If MRE achieves a 65% success rate in triaging, it would reduce the total costs by approximately 15%. With a negative-predictive value 8 of 97% (95% CI, 83.8%-99.8%), the threshold of 65% for MRE is far less than the combined true positive, false positive, and false negative rates for MRE. Conclusion MRE is non-invasive and provides tissue stiffness information for the entire liver and avoids the discomfort and risk of complications associated with other invasive procedures. In addition, elastograms avoid sampling errors and provide 1. Kim WR, Brown RS Jr, Terrault NA, et al. Burden of liver disease in the United States: summary of a workshop. Hepatology 2002; 36: Shaheen NJ, Hansen RA, Morgan DR, et al. The burden of gastrointestinal and liver diseases, Am J Gastroenterol. 2006;101: Rockey DC, Celdwell SH, Goodman ZD, et al. Liver Biopsy. American Association for the Study of Liver Disease (AASLD) Position Paper, Bravo AA, Sheth SG, Cehopra S. Liver biopsy. N Engl J Med. 2001; Feb 15;344(7): Smith JO, Sterling RK. Systematic review: non-invasive methods of fibrosis analysis in chronic hepatitis C. Aliment Pharmacol Ther. 2009; Sept 15;30(6): Stebbing J, Farouk L, Panos G, et al. A Meta-analysis of Transient Elastography for the Detection of Hepatic Fibrosis. J Clin Gastroenterol Sept 9.;44(3): Cerlson JJ, Kowdley KY, Sullivan SD, Ramsey SD, Veenstra DL. An evaluation of the potential cost-effectiveness of non-invasive testing strategies in the diagnosis of significant liver fibrosis. J Gastroenterol Hepatol May;24(5): Yin M, TalWalkar JA, Glaser KJ, et al. Assessment of Hepatic Fibrosis with Magnetic Resonance Elastography. Clin Gastroenterol and Hepatol. 2007;5(10): Afdhal NH, Nunes D: Evaluation of Liver Fibrosis: A Concise Review. Am J Gastroenterol Jun;99(6): Bruguera M, Bordas JM, Mas P, et al. A comparison of the accuracy of peritoneoscopy and liver biopsy in the diagnosis of cirrhosis. Gut 1974; 15: SignaPUL Liver Biopsy: $1,255 Total Cost 100 Patients > MRE: $628 65% 35% Liver Biopsy: $1,255 Total Cost 15% Figure 3. Comparison of direct costs when MRE is used in triaging for biopsy richer information that could assist in diagnosis. Studies show that the technique has excellent sensitivity in differentiating stiffness associated with normal liver tissue and fibrotic tissue. Stiffness of normal liver tissue is comparable to that of subcutaneous fat; studies have also not reported any influence of steatosis on tissue stiffness. In summary, the evidence supports the use of MRE as a triaging option for liver biopsy. The accuracy, lower costs, and the noninvasive nature of the technology offer the promise that MRE could improve outcomes at lower costs. 11. Poniachik J, Bernstein DE, Reddy KR, et al. The role of laparoscopy in the diagnosis of cirrhosis. Gastrointest Endosc. 1996;43: Pagliaro L, Rinaldi F, Ceraxi A, et al. Percutaneous blind biopsy versus laparoscopy with guided biopsy in diagnosis of cirrhosis. Dig Dis Sci. 1983;28: Olsson R, Hagerstrand I, Broome U, et al. Sampling variability of percutaneous liver biopsy in primary sclerosing cholangitis. J Clin Pathol. 1995;48: Angelucci E, Baronciani D, Lucarelli G, et al. Needle liver biopsy in thalassemia: Analyses of the diagnostic accuracy and safety in 1184 consecutive biopsies. Br J Haematol. 1995;89: Abdi W, Millan JC, Mezey E. Sampling variability on percutaneous liver biopsy. Arch Intern Med. 1979;15: Maharaj B, Maharaj RJ, Leary WP, et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet. 1986:1: Rockey DC, Celdwell SH, Goodman ZD, Nelson RC, Smith AD. Liver Biopsy. American Association for the Study of Liver Disease (AASLD) Position Paper, Ehman EC, Rossman PJ, Kruse SA, et al. Vibration safety limits for magnetic resonance elastography. Phys Med Biol. 2008;53(4): Wong J, Bennet W, Koff R, et al. Pretreatment evaluation of chronic hepatitis C: Risk, Benefits, and Costs. JAMA Dec 23-30; 280(24): Myers RP, Fong A, Shaheen AAM. Utilization rates, complications and costs of percutaneous liver biopsy: a population-based study including 4275 biopsies. Liver Int May;28(5): Talwalker JA. Elastography for detecting hepatic fibrosis: options and considerations. Gastroenterology Jul; 35(1):

40 GE Healthcare Wide bore means compromise, right? Wrong. Some have said that the benefits of a 70cm bore justified making trade-offs in image quality. We disagree. So we ve developed an MR system featuring patient comfort, productivity, and exquisite image quality. Now, you can do wide-bore MR the right way the Optima MR450w General Electric Electric Company Company

41 CLINIC L LUE LI I IN Changing P C... Liver I). 1. Clinic N. C. The O.. 18 SignaPUL 141

42 LIVER IMAGING CLINICAL VAL Dr. Richard L. Ehman I to probe. I C. I I C MRE. 3. O C C.. I... Principal I NI. I O N.. C and N NI C National I I NIH.. P I. Figure 1.. I.. N

43 CLINIC L LUE LI I ING Dr. Jayant A. Talwalkar. Talwalkar, MD, M.P.H. C C I Clinic in N. Dr. I I C L Mayo Clinic in Clinic. L C P C L. Dr. Principal I NI. Figure 2.. I.. Figure 3. C... References: 1. L. I C L.. 2. C P.. P Clin. 4. L C.... C L. P P I C..... L. P... C Conclusion I. I. 20 SignaPUL 143

44 LIVER IMAGING CLINICAL VALUE The Picture That Paints a Thousand Words But Left a P. business and real estate.. P Cirrhosis (PBC. Three.. Marshall N Center in Boston. no other doctor had done: He encouraged. Advocating for change PBC support. advocate for the PBC Liver Foundation, helping strongly believed they deserved... PBC at the Cleveland C. -. While the chances of survival PBC. I - PBC at only 47 years of age... torch. - I. PBC,. MRE revealed the extent and distribution. Hope on the horizon CNN Clinic trials I). and supple tissue at different rates. not on a color scale. nity to access MRE.. Mayo C.. He even. I

45 CLINIC L LUE LIVER I IN Getting the whole picture. PBC. I I. PBC PBC.. PBC. I. I.... C. I PBC. I. PBC..... P C C..... O. 22 SignaPUL 145

46 LIVER IMAGING CLINICAL VAL MRE in Clinical Practice: Case Review C. P.. I. grows stronger... Sites contributing MRE cases Sites currently using MRE Sites pending MRE implementation San Diego, CA San Diego, CA CA I L Rochester, MN I Seoul, Korea Pittsburgh, PA C C P Singapore New Delhi, I Hong Kong 146 St. Paul, MN 9

47 CLINIC L LUE LIVER I IN Dr. Russell N. Low Contributor: Russell N. Low, MD, Medical Director Sharp and Children s MRI Center, San Diego, CA Case 1:.. a Figure 1... b a Figure 2.. b 10 SignaPUL a Figure 3.. C. MRE Finding: C.. b 147

48 LIVER IMAGING CLINICAL VALUE Contributor: Susanne Bonekamp, DVM, PhD, Research Associate in the Division of Clinical MRI, Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD Case 1: C I. P. Dr. Susanne Bonekamp Figure 4a. MRI. Figure 4b. MRI. MRE Finding:.. Case 2: C I. Patient. Figure 5a. MRI. Figure 5b. MRI. MRE Finding: METAVIR I

49 CLINIC L LUE LI I IN Dr. Utaroh Motosugi Contributor: Utaroh Motosugi, MD, PhD, Assistant Professor, Department of Radiology, University of Yamanashi, Yamanashi, Japan Case 1:. LP I. LP): 712 IU/L I L. L.. Figure 7. MRE Finding:.. Case 2: HC. I CC). Figure 6a. I. Figure 6b.. Figure 6c.. MRE Finding: CC. 12 SignaPUL 149

50 LIVER IMAGING CLINICAL VALUE Contributor: Jeong Ming Lee, MD, Associate Professor, Seoul National University Hospital, Seoul, Korea Case 1: A 14-year-old with glycogen storage disease. Conventional MRI is not able to show definite changes of fibrosis. Dr. Jeong Ming Lee Figure 8. Figure 9. MRE Finding: fibrosis. Pathology reveals F3 fibrosis. Case 2: CC) with central necrosis. Figure 10. Figure 11. MRE Finding:. P III HCC

51 CLINIC L LUE LI I IN Dr. Scott Reeder Contributor: Scott Reeder, MD, PhD, Associate Professor, Section Chief of MRI, Department of Radiology, University of Wisconsin-Madison, Madison, WI Case 1: L a BMI MRE Finding:.. I. Figure 12.. Case 2:. MRE Finding:. C. O.. a b c Figure SignaPUL 151

52 LI IMAGING CLINICAL LUE Contributors: Sabine F. Bensamoun, MD, PhD, Researcher CNRS in the Biomechanics and Bioengineering Laboratory at the University of Technology of Compiègne, Compiègne, France and Fabrice Charleux, MD, MRI radiologist, Department of Radiology, Polyclinic St. Côme, Compiègne, France Dr. Sabine F. Bensamoun Case 1:.. MRE Finding:.. Dr. Fabrice Charleux Figure 14. Acknowledgement C P P

53 CLINIC L LUE LI I IN C P P L. P P.. (SP C C C. I P. P. C SP P P. P.. C I... SP P.. (MN I. I N I HCC P. N I.... C CP N I SignaPUL 153

54 LIVER IMAGING CLINICAL VALUE Contributors: Peter B. Wold, MD, Body Imaging Section, St. Paul Radiology, and Coleman Smith, MD, Hepatology Section Head, Minnesota Gastroenterology, P.A., St. Paul, MN Case 1: L CV anti-body is referred for hepatology consult. HC. P C spontaneous clearance of HCV infection. MRE Finding:. Diffuse hepatic steatosis also present at out-of-phase gradient MRI. Clinical data and MRI suggest nonalcoholic steatohepatitis (N. L disease. P. N. Dr. Peter B. Wold Dr. Coleman Smith Case 1. Case 2: CV and stage I-II diagnosed at biopsy four years prior presents for hepatology consultation. I. MRE Finding:I-II. L hypertension at MRI. Case

55 CLINIC L LUE LI I ING MRE in Clinical P L By Sudhakar K Venkatesh, MD, FRCR, Department of Diagnostic Imaging and Seng Gee Lim, MD, FRACP, Department of Gastroenterology and Hepatology, National University Health System Introduction C.. L.. L. C... I... I. 4 SignaPUL L. 5 I... (MRE). I... Clinical Study. L.. L.. 155

56 LIVER IMAGING CLINICAL VALUE Method I.5T MR scanner.. of the liver. OI). Results. O. L representative of significant fibrosis. Using MRE, 55% of the patients had liver. Management L N. N. NA NA and liver function tests. elevated liver stiffness and high viral DN.. I viral DN NA levels.. Dr. Sudhakar K. Venkatesh Sudhakar K. Venkatesh, MD, FRCR, is Assistant Professor of Diagnostic Radiology, National. He graduated residency training in Radiodiagnosis at the in Radiodiagnosis at SGPGIMS, L India. N College of Radiologists (FRCR), L. I he joined the National University Hospital (NUH). Dr. the Mayo Clinic (Rochester, MN). Dr. Venkatesh s research interests are in liver diseases, specifically I. Dr. Venkatesh is the principal investigator of a research project of liver fibrosis and differentiation of focal liver lesions. Case One L. Liver. The patient refused percutaneous liver biopsy. MRE Finding of fibrosis.. 156

57 CLINIC L LUE LI I ING C Dr. Seng Gee Lim Seng Gee L CP P N. N L I... L a C L. N.. L.. L. About the facility N N N N. N C O O O P O. I N L L... MRE Finding OI.. C. MRE Finding... N I C C P C... L. C. L C... L IGO L IGO..... C C. C 5. P P. C... N.. C. SignaPUL Summary N.. L. L. 157

58 HEALTH ECONOMICS BEYOND THE SCAN MRE Cost Effectiveness: Preliminary Threshold Assessment By David W. Lee, PhD, Vinod Palathinkara, PhD, GE Healthcare and Mitch DeKoven, IMS Health Introduction Liver biopsy is the most specific test to assess the nature and severity of liver diseases. 1 While often considered the reference standard, liver biopsy can yield false-negative results in nearly one-third of cases 2,3 and is characterized by a morbidity rate of 3% and a mortality rate of 0.03%. 2 As such, noninvasive methods are being developed as a means for detecting liver fibrosis. Magnetic Resonance Elastography (MRE) is a promising noninvasive technique for evaluating tissue stiffness and liver tissue characterization. 4 This study estimated the accuracy needed for MRE to be cost-neutral, as compared to liver biopsy, from a US Medicare perspective. Methods We conducted a targeted literature review to identify the full range of services that accompany both a liver biopsy and MRE. A leading hepatologist and pathologist were consulted in order to identify the appropriate procedure codes associated with both modalities. We assumed that MRE would be reimbursed using CPT-4 code (Magnetic resonance [eg, proton] imaging, abdomen; without contrast material[s]). 5 Payment amounts were assigned to the identified procedure codes using the Medicare 2010 Physician Fee Schedule, the Medicare 2010 Hospital Outpatient Prospective Payment System, 6 and the Medicare 2010 Clinical Laboratory Fee Schedule. Payment rates were subsequently stratified into both payer and patient responsibility, based upon patient out-of-pocket responsibility. The payment amounts were incorporated into a decisionanalytic model that compared the costs for patients with suspected liver fibrosis via two scenarios: 1) biopsy or 2) MRE followed by biopsy when the MRE test was positive. We assumed MRE negative predictive values (NPV) of 0.8, 0.9 and 0.95 and patients with a false-negative MRE would ultimately receive a biopsy. Results The cost of an ultrasound-guided liver biopsy in the hospital setting, from the US Medicare payer perspective, is $1, (ultrasound: $120.96; surgical: $730.97; pathology: $271.10; laboratory: $25.54; evaluation and management: $ (Table 1). 158 Table 1: Liver biopsy costs* Liver Biopsy Costs Professional Fee Facility Fee Total Ultrasound $23.67 $97.29 $ Surgical (i.e., biopsy) $79.38 $ $ Pathology $ $ $ Laboratory $25.54 $25.54 Physician Evaluation and Management $ $ $ Total $ $ $ *Cost to Medicare program (80% of allowable charge) excluding cost sharing 59

59 ON SC N L ECONO ICS Table 2: MRE Costs* Hospital Setting Non-Hospital Setting Cost Category Professional Fee Facility Fee Total Professional Fee Only I.89. $ $ P.59. $ Total $ $ $ $ *C... NP.... I NP.... MRE Test Negative Rate Conclusions.. C. Editor s Note: This study was presented as a poster at the ISPOR 13th Annual European Congress held in Prague, Czech Republic, on November 6 to 9, Negative Predictive Value (NPV) N Figure 1:.. 60 SignaPUL References. C. L. N... P P. C... P.... N P L. L P C P. P. 6. O P P 159

60 MRE Cost Effectiveness: A Scenario Analysis By Vinod S. Palathinkara, PhD, and David W. Lee, PhD, GE Healthcare Given the novelty of MRE, peer-reviewed academic/medical literature on direct evaluation of potential cost-effectiveness of the technique in the diagnosis and management of liver fibrosis is not yet available. Nevertheless, indirect comparisons based on studies that compare stiffness measured using MRE and biopsy results can provide insights. A scenario analysis based on a peer reviewed literature comparing observations of a patient who had both biopsy and MRE is illustrated here. MRE is currently not reimbursed as a stand-alone test with its own CPT code. Because the acquisition time for MRE is very short, the technique can be readily included in the protocol for an already indicated abdominal MRI exam with little impact of the typical examination time of 30 to 45 minutes. If the entire cost of such an exam is attributed to the MRE procedure, then a conservative estimate of the cost of MRE would be equivalent to the 2010 national Medicare average payment amount for abdominal MRI, i.e. $628 (CPT code 74183). At this stage, there is no way to predict the willingness of payers to cover an MRI examination conducted solely to perform MRE. To better quantify the costs associated with MRE and liver biopsy, a decision-analytic model comparing diagnostic costs was constructed. 1 A targeted literature review was conducted and a leading hepatologist and pathologist were consulted to identify the appropriate procedure codes associated with liver biopsy. The study assumed that MRE would be reimbursed* using CPT-4 code (magnetic resonance [e.g., proton] imaging, abdomen; without contrast material[s]). All appropriate allowable charges were assigned to the identified procedure codes using the 2010 Medicare Physician Fee Schedule. All costs discussed here are US-based and are not globally applicable. Based on the model, the cost of a guided liver biopsy is $1,424 (ultrasound $164, Figure 1. Flow diagram of patients who underwent liver biopsy and MRE (Huwart et al.). Biopsy (96) F0(22) F1-2(41) F3-4(33) Outcome Elevated liver enzymes Conforming outcome *MRE is currently not reimbursed on its own CPT code. MRE (96) 160 (24) Misclassification Follow-up 61

61 BEYOND SC N L ECONO ICS Scenario 1. F0(22) C Biopsy (96) F1-2(41) Outcome F3-4(33) C Elevated liver enzymes Conforming outcome MRE (96) (24) Misclassification Follow-up Figure 2... Scenario Analysis I. 2. I... I... Scenario 2. C.. C Unsuitable sample (9.6) Biopsy (105.6) Elevated liver enzymes MRE (96) F0(22) F1-2(41) Outcome F3-4(33) Conforming outcome (24) Misclassification Follow-up Figure SignaPUL 161

62 HEALTH ECONOMICS BEYOND THE SCAN Scenario 3 Biopsy is only 80% accurate, but MRE still has 25% misclassifications. Costs for procedures: Biopsy: $136, x 96 x $1,424 = $136,704 + $27,341 = $164,045 MRE: $123,100 Cost differential = $40,945, or 25% less than biopsy Unsuitable sample (9.6) Biopsy (105.6) Elevated liver enzymes MRE (96) F0(22) Outcome F1-2(41) F3-4(33) (19.2) Follow-up Conforming outcome (24) Misclassification Follow-up Figure 4. Hypothetical examples of direct costs of MRE and liver biopsy for scenario 3. Huwart et al, report that in comparing results from MRE and biopsy, 24 of 96 (25%) were misclassified in their stage of fibrosis. Based on the reported sensitivity and specificity of MRE techniques, this misclassification is unusually high. Nevertheless, since biopsy is the reference standard, we assume that 25% of patients who underwent MRE would eventually require further evaluation and likely receive a biopsy. We do not consider how one would identify the misclassified patients and ignore the impact on the outcome or the additional treatment costs due to delay in misclassification. Our attempt here is to illustrate a methodology that can provide directional information on costs to explore potential for cost reduction rather than to establish or estimate actual cost differentials. In scenario 1, we assume that biopsy is 100% accurate and all samples are sufficient for diagnosis. This is also considered a worst-case scenario for MRE. The study s authors highlight the fact that liver biopsy is not an optimal reference examination and that they do not know if the reported discordant results between MRE and histopathology were caused by problems of inadequate biopsy sampling*. The authors also report that the two pathologists who reviewed the biopsy specimen were initially in agreement for 81 of the 96 samples. Nevertheless, since biopsy is the reference standard, we assume it provides a clinically accepted basis for comparison. As reported by Huwart et al, 14 of 141 samples were unsuitable for diagnosis. While many studies recommend a minimum sample length of 25mm, the study reports that 23% of the samples were less than 25mm in length. Thus the scenario that a few biopsy samples would be unsuitable is realistic. 162 In scenario 2, we assume that to obtain 96 good samples, one would need to do 10% more samples (105.6 biopsies). This assumption does not imply that these patients would undergo an immediate repeat biopsy. The cost of this may result in an increased cost of diagnosis per person. In scenario 3, we take into account a biopsy leading to a misclassification. More than questioning the accuracy of the biopsy, this is reflective of the fact that biopsy is a sampling technique. Studies report biopsy mis-staging to be in the range of 10 to 33%. If we assume 20% of biopsy samples are mis-staged, then the discordances between MRE and biopsy may decrease. However, for simplicity, we still assume that the discordance between MRE and biopsy will not change. Summary We have illustrated a scenario-based analysis to compare potential cost differentials of MRE with other techniques that provide similar information. No recommendation on the likelihood of any scenario is being made. As the three scenarios show, the cost differentials for MRE (over biopsy) can range from 10% to 25% lower than direct biopsy costs. The key assumptions that determine the actual cost savings are accuracy of biopsy, sample suitability of biopsy, and accuracy of MRE. *In 26 of the 96 samples, the biopsy specimen was less than 25mm. References 1. DeKoven, M, Cost comparison: Liver Biopsy versus Abdominal MRI, Memo to GE Healthcare from IMS Health Incorporated; May 25, Huwart, L, Sempoux, C, Vicaut, E, Salameh, N, Annet, L, Danse, E, Peeters, F, Ter Beek, LC, Rahier J, Sinkus, R, Horsmans, Y, and Van beers, BE, Magnetic Resonance Elastography for the non-invasive staging of liver fibrosis, Gastroenetrology, 135(1): 32-40, July

63 GE Healthcare Outrun obsolescence. MR technology advances quickly. Bridging the technological gaps between new purchases is a must for maintaining your system s value and your high level of patient care. Only GE Healthcare backs every MR system with the legendary Continuum our promise that you can upgrade your system to the latest MR technology without replacing your magnet. So you maintain your imaging edge and a strong return on your investment. You can t always see what s down the road. But with our MR Continuum, you re assured the next corner will bring something extraordinary. MR Re-imagined. To learn more, visit continuum.html 2009 General Electric Company GE Medical Systems, a General Electric company, going to market as GE Healthcare. 163

64 name of author info about clinic, etc. Evaluation of Liver Diseases 164 Liver disease includes Hepatitis B, Hepatitis C, non-alcoholic fatty liver disease (NAFLD), non-alcoholic steatohepatitis of NAFLD. Hepatitis C. 3 evaluation of liver diseases, GE Healthcare recently hosted - 69

65 ISSUE SPOTLIGHT LIVER DISEASE -'! '! Figure 1. 4_XST\X^[^Vh ^U 7T_PcXcXb 1 _atep[t]rt ^U 71b0V RPaaXTa bcpct 4R^]^\XRP[[h 7T_PcXcXb 2 Xb cwt d]sta[hx]v RPdbT U^a ^]T cwxas P a^gx\pct[h ^U P[[ [XeTa cap]b_[p]cb Pc P R^bc ^U!' DB3 _Ta _a^rtsdat CWT [XUTcX\T R^bc ^U 7T_PcXcXb 2 fxcw^dc cap]b_[p]c Xb TbcX\PcTS Pc _Ta X]SXeXSdP[ 4 Assessing liver disease background 0[P]X]XT cap]bp\x]pbt 0;C P]S Pb_PacPcT cap]bp\x]pbt 0BC PaT bt]bxcxet X]SXRPc^ab ^U [XeTa SP\PVT ^a X]Ydah Ua^\ SXUUTaT]c ch_tb ^U SXbTPbT htc cwtbt T]ih\T [TeT[b PaT ]^c P _atrxbt X]SXRPc^a ^U [XeTa SP\PVT ^a _a^v]^bxb 1^cW ctbcb X]SXRPcT ^][h cwt _atbt]rt ^U [XeTa SXbTPbT ]^c bcpvx]v P]S RP]]^c QT dbts c^ VdXST catpc\t]c 0RR^aSX]V c^ 3a CP[fP[ZTa ]^a\p[ 0;C ep[dtb S^ ]^c P[fPhb X]SXRPcT \X[S [XeTa SXbTPbT CWaTT btad\ ctbcb WT[_ TeP[dPcT ^etap[[ [XeTa Ud]RcX^] BTad\ P[Qd\X] \TPbdaTb W^f ft[[ cwt [XeTa Xb _a^sdrx]v cwt _a^ctx] P[Qd\X]* btad\ QX[XadQX] STcTa\X]Tb W^f ft[[ cwt [XeTa R[TPab P Q[^^S fpbct _a^sdrc QX[XadQX]* P]S _a^cwa^\qx] cx\t 8=A \TPbdaTb cwt R[^ccX]V cx\t ^U _[Pb\P Pb P] X]RaTPbT X] _a^cwa^\qx] cx\t \Ph X]SXRPcT [XeTa SP\PVT 0b fxcw 0;C 0BC cwtbt ctbcb PaT P[b^ ]^c R^]R[dbXeT U^a bcpvx]v [XeTa SXbTPbT 70 BXV]P?D;B4 ~ B_aX]V! 2daaT]c[h cwt V^[S bcp]spas U^a R^] a\x]v P R[X]XRP[ SXPV]^bXb P]S PbbTbbX]V cwt btetaxch ^U ]TRa^X] P\\PcX^] Xb [XeTa QX^_bh 1X^_bh Xb P[b^ dbts c^ TeP[dPcT _^bbxq[t R^]R^\XcP]c SXbTPbT _a^rtbbtb P]S PbbTbb Qa^bXb P]S cwtap_tdcxr X]cTaeT]cX^] CWT _ax\pah V^P[ ^U QX^_bh Xb c^ SXUUTaT]cXPcT bcpct! Ua^\ bcpvt! " P]S c^ SXPV]^bT RXaaW^bXb } Tg_[PX]b 9^W] A ;PZT <3 _a^utbb^a ^U <TSXRX]T P]S BdaVTah D]XeTabXch ^U <X]]Tb^cP <TSXRP[ BRW^^[ X] ^asta c^ STcTa\X]T fwtcwta cwtat Xb P] X]SXRPcX^] U^a catpc\t]c } 7T PSSb cwpc \PZX]V P SXPV]^bXb ^U RXaaW^bXb RP] P[b^ WPeT X\_^acP]c X\_[XRPcX^]b bdrw Pb bratt]x]v U^a 722 P]S S^bX]V P]cXeXaP[b 3a ;PZT Tg_[PX]b cwpc P[cW^dVW QX^_bh Xb cwt V^[S bcp]spas Xc Xb R^bc[h ap]vtb Ua^\ $ c^! DB3 P]S _^btb axbzb c^ _PcXT]cb ]P\T[h \^STaPcT _PX] X]! P]S W^b_XcP[XiPcX^] X] c^ " 8c WPb P \^acp[xch apct ^U fwxrw X]RaTPbTb c^ # X] _PcXT]cb fxcw RP]RTa CWT _a^rtsdat Xb P[b^ Pbb^RXPcTS fxcw _^ct]cxp[ bp\_[x]v Taa^ab P]S bdqytrcxet WXbc^[^Vh VaPSX]V 5TfTa cwp] _^acp[ caprcb \Ph QT [X]ZTS fxcw bp\_[x]v Taa^ab ^]T bcdsh U^d]S Qa^bXb bcpvt SXbR^aSP]RT X] "" ^U RPbTb $?PcW^[^Vh bzx[[ btcb P[b^ epah fxst[h 0]^cWTa bcdsh U^d]S! ^U Qa^bXb P]S RXaaW^bXb PaT d]stabcpvts fxcw cwt bxit ^U QX^_bh QTX]V P [XZT[h Rd[_aXc 6 0b P atbd[c btr^]s ^_X]X^]b PaT ^UcT] at`dxats 1 65

66 Figure 2. A real problem for liver biopsy is the ability to detect small the complications and potentially unreliable biopsy results has led to increased interest in non-invasive methods for preference will further accelerate the decline in the Non-invasive options for chronic liver disease management accurate prediction of the full spectrum of disease

67 ISSUE SPOTLIGHT LIVER DISEASE Serum fibrosis markers Ideal non-invasive test for monitoring chronic liver disease 7 Transient and ultrasound elastography

68 Liver fibrosis Normal Biopsy stage F1 Biopsy stage F2 Biopsy stage F3 Biopsy stage F4 Figure 3. (Images courtesy of Mayo Clinic, Rochester, MN) Hard Performance of MR Elastography in difficult situations Figure 4. Figure 5. Hard Hard MR Elastography

69 ISSUE SPOTLIGHT LIVER DISEASE 9 Conclusion Photo L to R. Row 1: Row 2: Row 3: Row 4: Row 5: Row 6: References:

70 By Kenji Asano, ME 1, PSD/Application Architect; Ersin Bayram, PhD 1, Body MR Applications Development Manager; Huanzhou Yu, PhD 1, Scientist; and Scott B. Reeder, MD, PhD 2, Section Chief of MRI and Cardiovascular Imaging 1 GE Healthcare, Waukesha, WI, 2 University of Wisconsin, Madison, WI Healthy volunteer Fatty liver 170 Figure 1. Fatty liver images are courtesy of Dr. Claude Sirlin, University of California, San Diego, CA. 75

71 ISSUE SPOTLIGHT LIVER DISEASE A promising MR-based technique IDEAL IQ provides volumetric whole-liver coverage in a single breath-hold and generates estimated T2* and triglyceride fat fraction maps in a non-invasive manner. It is intended for breath-held abdominal imaging to evaluate diffuse liver diseases such as hepatic steatosis of the liver and corrects for challenging confounding factors such as T2* decay. The technique is designed for water-triglyceride fat separation with simultaneous T2* correction and estimation based on the IDEAL technique. Six gradient echoes are typically collected using the 3D Fast SPGR sequence in one or two repetitions. The IDEAL IQ reconstruction produces water and triglyceride fat images, and relative triglyceride fat fraction and R2* maps from the six echo source data. T2* correction and estimation T2* decay causes signal dephasing and T2* in liver can be shorter than the normal range in cases of iron overload. It has been assumed that T2* decay results in negligible signal loss among the echoes in conventional Dixon methods, but it can be significant on the time scale of Dixon echo shifts leading to substantial errors in hepatic fat estimates (Figure 2). IDEAL IQ uses a novel construct of a complex field map to include the effects of T2* into the signal model (Figure 3). By acquiring six echoes and estimating the complex field map using an iterative least square method, it is possible to achieve simultaneous water-triglyceride fat decomposition and T2* estimation in a single breath-hold. 8 Signal 10% triglyceride fat T2* = Infinity OP=W-F IP=W+F time (ms) 10% triglyceride fat Figure 2. Signal level plots as a function of time using typical in-phase (IP) and out-of-phase (OP) imaging with and without T2* effects. IP and OP images would have been sufficient to calculate the triglyceride fat fraction if there were no T2* effects. However, with the T2* effects, the signal s decay will result in under-estimation of triglyceride fat fraction. Signal 10% triglyceride fat T2* = 10 ms OP=W-F Multi-peak fat spectrum modeling IP=W+F time (ms) 0% triglyceride fat Conventional chemical shift-based water-fat separation methods use a relatively simple signal model that assumes both water and fat have a single resonance frequency. However, it is well known that fat has multiple spectral peaks, at least six of which can be resolved at clinical field strengths (Figure 4). This inaccuracy in the signal model results in two undesired effects. First, water and fat are incompletely separated. Second, methods designed to S i (t) i = 1, 2 num echoes IDEAL recon Complex fieldmap W F ˆ j R2*/2 ˆ real imaginary triglyceride fat fraction map { R2* Figure 3. Source echoes are decomposed into T2*-corrected water, T2*-corrected triglyceride fat, and complex field map. The water and fat images are used to calculate a triglyceride fat fraction map. The real part of the complex field map corresponds to B 0 field inhomogeneity, while the imaginary part contains information on R2* (=1/T2*) SignaPULSE Spring 2011

72 Figure 4. Dr. Scott B. Reeder Frequency (Hz, at 3T) s(t) = ( w + e t e t where is is the complex 9 Water Fat R2* map Fitting residue Multi-peak IDEAL IQ Single-peak IDEAL IQ R2* (1/s) Figure

73 ISSUE SPOTLIGHT LIVER DISEASE Initial guess S i Huanzhou Yu, PhD, Scientist Source signals Complex based T2* IDEAL Step 1: Resolve ambiguity Water, fat Magnitude based reconstruction Step 2: Fine tuning Water, fat Weighted combination W, F, triglyceride fat fraction Kenji Asano, ME, PSD/ Application Architect Figure 6. Magnitude-based reconstruction Ersin Bayram, PhD, Body MR Applications Development Manager References

74 Clinical Utility of By Scott B. Reeder, MD, PhD, Division Chief of MRI, University of Wisconsin, Madison, WI Case 1 IDEAL IQ used in conjunction with MR-Touch may play an important role in the non-invasive assessment of diffuse liver disease. Figure 1. Case 2 Elastogram Triglyceride fat fraction map R2* map Figure

75 Case 3 Dr. Scott B. Reeder T2 with fat-sat Out-of-phase Coronal T2 SSFSE In-phase Elastogram Triglyceride fat fraction map R2* map Figure 3. Discussion

76 Non-invasive Evaluation By Russell N. Low, MD, Sharp and Children s MRI Center and Tarek Hassanein, MD, Southern California Liver Centers Discovery MR450 Patient history The MR-Touch application utilizes an active driver (above) to generate shear waves that are propagated to the abdominal cavity via the passive driver. Figure 1. LAVA

77 Dr. Russell N. Low MR technique MR finding Histopathology: Discussion MR-Touch parameters TR 50 ms TE 24.6 ms Bandwidth khz Flip angle 30 degrees Matrix size 256 x 64 Slice thickness 10 mm Gap 1 mm Dr. Tarek Hassanein

78 GE Healthcare At GE, we are committed to helping increase access to healthcare while improving its quality and lowering its cost. Just like physicians everywhere. So by investing in new innovations, we are empowering the world s healthcare professionals to do what they do best: caring for patients around the world. Every day, doctors are bringing better health to more people and GE Healthcare technologies are behind them General Electric Company