Cirrhosis and its disease-related complications are the 12th leading cause of overall mortality1 and the second leading cause of gastrointestinal death in the United States.2 The major disease-related complications from cirrhosis are related to portal hypertension (esophageal variceal hemorrhage, ascites, hepatic encephalopathy) and hepatocellular carcinoma.3 For the majority of affected patients, these clinical outcomes are directly related to the development of progressive hepatic fibrosis. Despite advances in the medical management of cirrhosis, there remains an opportunity to improve the detection rate of individuals with progressive fibrosis who might benefit from early intervention.
To date, liver biopsy has been the gold standard for detecting hepatic fibrosis. The majority of classification systems recognize 5 stages of fibrosis, graded as F0 (no fibrosis), F1 (portal fibrosis), F2 (periportal fibrosis), F3 (bridging fibrosis), and F4 (cirrhosis). Clinically significant fibrosis is generally defined by F2 involvement or greater. However, a number of studies have demonstrated excessive rates of sampling error (25%-40%) resulting in poor reproducibility. In addition, the extent of variation from observer interpretation by expert histopathologists may be as high as 20%.4 Given these limitations, a growing number of investigations have focused on validating noninvasive methods for detecting hepatic fibrosis.
A number of serum markers representing the process of hepatic fibrosis have been studied to date. Unfortunately, the majority of these studies are limited by retrospective study designs, low rates of liver biopsy reproducibility, and the inclusion of patients with a narrow spectrum of disease severity.5 Advances in cross-sectional imaging that focus on detecting static morphologic alterations of liver disease can be used to identify cases of established cirrhosis.6 However, the ability to detect early and intermediate stages of fibrosis with these techniques remains limited.7
The palpation and detection of tissue stiffness by physical examination has traditionally been associated with organic disease processes, including malignancy. The development of techniques to “palpate” solid-organ tissue by imaging has recently been identified.8 Regional imaging through wave propagation techniques has the potential to assess the mechanical properties of human tissue in health and disease.9 Measuring the velocity of wave propagation can be used to calculate “stiffness,” which may be highly associated with known endpoints of injury such as fibrosis. Recent work has focused on the technique of ultrasound-based elastography to predict the severity of hepatic fibrosis based on measurements of liver stiffness.10 To date, it appears that a correlation exists between mean liver stiffness (measured in kilopascals) and stage of fibrosis in patients undergoing liver biopsy.10–12 Similar findings are observed in patients with varying degrees of cirrhosis, as higher stiffness values correspond to the development of disease-related complications.13
In the current issue of Liver Transplantation, Carrion et al.14 examined the role of ultrasound-based transient elastography to detect advanced hepatic fibrosis in patients with recurrent hepatitis C infection after liver transplantation. In addition, the performance of hepatic venous pressure gradient (HVPG) measurement was done to assess the degree of portal hypertension in this cohort. From a total of 124 patients, the sensitivity and specificity of ultrasound-based elastography to detect stage 2–4 fibrosis from liver biopsy was 90% and 81%, respectively. Greater degrees of accuracy were observed for detecting stage 3 and stage 4 fibrosis, respectively. Notably, the correlation between liver stiffness and HVPG measurement was higher compared to the correlation between elastography and liver biopsy. In fact, 8 of 12 (67%) patients with significant portal hypertension and stage 0–1 fibrosis on liver biopsy had elastography stiffness values corresponding to stage 2–4 fibrosis. No individual patient with stiffness values equivalent to stage 0–2 fibrosis had more advanced disease on liver biopsy or portal hypertension by HVPG measurement.
The strengths of this diagnostic accuracy study include (1) the use of consecutive patients, (2) the inclusion of a wide spectrum of disease severity, (3) the application of test (elastography) and reference (liver biopsy) standards in all patients, (4) interpreter blinding of results for all tests, and (5) the measurement of diagnostic test parameters including receiver operating characteristic curves. Limitations include (1) the incomplete application of all 3 tests to the entire study cohort, (2) the absence of a second observer for ultrasound-based elastography to determine intraobserver reliability, and (3) the failure to categorize all test results, including those with indeterminate findings based on limited biopsy size or incomplete elastography.15
Regardless of acquisition method and specimen length, the persistence of sampling variability associated with core needle liver biopsy remains a significant obstacle in considering this as a true reference standard.16 However, the use of liver biopsy will still be required as the reference standard for additional diagnostic accuracy studies unless alternate methods (such as the use of wedge biopsies, or partial or complete explanted liver) are used. A second issue concerns the actual probability for widespread application of ultrasound-based elastography. For patients with a body mass index ≥28 kg/m2 (with or without prior liver transplant), there is a 9-fold increase in the risk of technical failure of elastography, even in expert hands.17 Notably, the Barcelona group also observed a reduction in diagnostic performance among this patient subgroup. Unless refinements in technique can overcome this problem, a sizable proportion of patients in North America and other geographic areas where obesity and nonalcoholic fatty liver disease are prevalent will not benefit from this diagnostic technique.
This is the first published investigation to explore the association between image-based elastography and an established measure of hepatic physiologic function such as portal pressure. While issues surround the utility of HVPG measurement in clinical practice,18 the recognition of mild to moderate portal hypertension in patients with liver biopsy results describing only stage 1–2 fibrosis from recurrent hepatitis C provides additional support on the weakness of liver biopsy to predict outcome. These findings also lend additional support to observations regarding the role of HVPG in defining clinical outcomes from recurrent allograft hepatitis C.19 In this regard, the use of functional imaging may provide an opportunity to determine more accurately which patients should be given antiviral therapy for chronic hepatitis C (including those individuals with liver histology that underestimates the true extent of recurrent disease).20 Additional validation studies incorporating the use serial elastography and HVPG measurements are required to confirm these observations.
There is great enthusiasm for the incorporation of functional hepatic imaging techniques into the diagnostic armamentarium of clinical hepatology. In addition, these methods have the potential to fulfill the desired tenets of being noninvasive, reproducible, and accurate compared to the somewhat flawed yet reference standard of liver biopsy. Will these diagnostic tools be a reality in the near future? Perhaps. The measurement of elasticity alone may not be enough to accurately depict overall stiffness in vivo. A number of observations strongly suggest that tissue viscosity also plays an important role for determining organ stiffness. Viscosity may be influenced by blood flow and pressure, bile formation and flow, and hepatocyte integrity within the human liver. Currently, the ability to decipher material properties of biologic tissues is based on empiric observations with phantom experiments and mathematical models. Therefore, the measurement of tissue viscosity is often neglected and may affect the accuracy of elasticity assessment.21 Tissues that are heterogeneous in nature (such as a liver involved with chronic inflammation and fibrosis) will also have regional variations in stiffness that invalidate the assumption of elasticity in characterizing fibrosis. At the present time, ultrasound-based techniques for measuring stiffness are sensitive to movement along a single axis (i.e., 1-dimensional), which can limit accuracy if a specific organ moves in multiple dimensions following wave excitation and propagation. A technique known as vibro-acoustic elastography, which stimulates tissue motion through high-frequency sound waves, may overcome these limitations, as subtle degrees of motion at tissue boundaries can be assessed to measure both elasticity and viscosity.
Recently, the technique of magnetic resonance (MR) elastography has also been described. MR elastography synchronizes motion-sensitive imaging sequences with the application of acoustic waves in tissue media.8 Preliminary data in human subjects also support the feasibility and promise of MR elastography to quantify stiffness and predict stage of fibrosis in patients with chronic liver disease.22 The advantages of MR elastography over ultrasound-based methods include (1) a freely-oriented field of view, (2) no acoustical window requirement, (3) insensitivity to body habitus, (4) the ability to quantify steatosis, and (5) the ability to obtain a conventional MR examination at the same time.8 In addition, determining the effect of blood flow and hepatocyte integrity (in the setting of inflammation and steatosis) on liver stiffness could also be assessed with MR technology.
Ultimately, the diagnostic accuracy of image-based elastography for quantifying hepatic fibrosis in human subjects requires additional study. Patients with recurrent hepatitis C after liver transplantation are just one of several high-risk subgroups who could benefit from the use of functional imaging to more accurately predict disease progression. Additional issues that warrant study include the ability to predict complications of cirrhosis, the rapid assessment of treatment effects of novel agents designed to halt fibrosis progression, and the uncovering of important mechanistic information regarding the dynamic etiopathogenesis of fibrosis and hepatocarcinogenesis.