Advances in MR Elastography in Liver

Introduction:

As a public health problem, chronic liver disease (CLD) is a major cause of morbidity and mortality in the United States. It affects approximately 1.8% of the population and leads to 13 deaths per 100,000 persons[1]. A variety of etiologies can cause CLDs, including nonalcoholic fatty liver disease (NAFLD), chronic hepatitis virus infection, alcohol abuse, primary sclerosing cholangitis, primary biliary cirrhosis, and autoimmune hepatitis[2]. Cellular death and inflammation caused by CLD are central elements in hepatic fibrogenesis, which will progress to cirrhosis associated with the development of life-threatening complications of portal hypertension (PHTN), hepatic decompensation, and hepatocellular carcinoma [3]. There is compelling evidence indicate that with the removal of the underlying etiologies, liver fibrosis may regress or stabilize[4]. If treated early, the hepatic parenchyma may even return to almost normal[5]. Therefore, early detection of fibrosis and accurate diagnosis of etiology are essential for monitoring treatment efficacy, disease progression, and for establishing prognosis in patients with CLDs.

Liver biopsy is the gold standard for assessing hepatic inflammation, cellular injury, and fibrosis. However, it is an invasive procedure associated with complications of pain and bleeding, which results in frequent refusal by patients for serial measurements. Other major disadvantages include sampling errors and substantial inter- and intra-observer variation that limits the suitability of biopsy for providing dynamic information for assessing treatment efficacy and disease progression[6 7]. Serum laboratory tests, such as platelet count, aspartate aminotransferase-to-platelet radio index, or APRI[8], FIB-4[9], are useful and easy to perform but have limited accuracy in diagnosing early stage of inflammation and fibrosis[10]. Ultrasound-based elastography technologies (e.g., transient elastography, shear wave elastography, etc.) are inexpensive non-invasive methods for assessing liver stiffness. However, they evaluate only a small portion of the liver with a single parameter, which may yield substantial sampling error and incomplete information for disease assessment. Another limitation of ultrasound examinations is that they may fail in patients with obesity, ascites, or narrow intercostal space [11].

MR elastography (MRE) uses a modified phase-contrast imaging sequence to detect propagating shear waves within the liver. It enables the evaluation of a large portion of the liver and provides multiple mechanical properties that are associated with different pathophysiologic states. With significant advances made in MR technology, MRE has been demonstrated to be a highly accurate non-invasive diagnostic tool in detecting and monitoring various CLDs[12–14].

This comprehensive review summarizes current knowledge of the technical advances and innovations of hepatic MRE development (Table 1), and the clinical applications in various hepatic diseases.

Table.1.

Summary of advances in MRE of liver and their utility

Advances/Innovations	Utility
Spin-echo MRE (SE-MRE)	Less sensitive to liver iron overload Less sensitive to motion artifact More rapid data acquisition
3D-MRE	Larger coverage volume More accurate measurements by reducing errors due to oblique wave propagation and edge artifacts
Multifrequency MRE	Provides multiple mechanical parameters that can evaluate different pathologic processes
Flexible driver	Improve patient comfort
Free-breathing MRE	Improve tolerance, particularly pediatric subjects and sedated patients
Automated liver stiffness estimation	Reduce inter-observer variation in liver stiffness measurements

Technology development

Image quality enhancement

Liver MRE is a well-accepted substitute for biopsy in screening and follow-up. To further improve its technical reliability and discover new imaging biomarkers, investigators have made numerous improvements in imaging sequence, active/passive actuator, and elastographic inversion algorithms [15–18].

Nowadays, the most widely available commercial MRE technique is gradient-recalled echo MRE (GRE-MRE)[19]. It has been well-validated in several large cohorts of clinical studies[20–23]. However, the conventional GRE-MRE technique can have technical failures due to susceptibility artifacts (e.g., iron overloaded liver) and insufficient shear wave penetration/encoding in the deeper structure of the liver (e.g., obesity)[24 25]. Iron-overloaded liver and obesity are not unusual in patients with CLDs[26]. Thus, researchers have developed a dedicated spin-echo based echo-planar (SE-EPI) MRE sequence. It has been demonstrated to be intrinsically insensitive to T2* susceptibility[27 28] and allow rapid acquisition with mitigated motion artifacts[29] (Figure 1). SE-EPI MRE is considered to have a significantly higher technical success rate than GRE-MRE [24 25 30–32].

Figure 1.

Representative images of MRE in a patient with increased iron deposition in the liver. Magnitude, wave, and elastograms with overlayed confidence map (>0.95 confidence level in the checkboard) of GRE-MRE (top row) and SE-EPI-MRE (bottom row) are shown from left to right. The confidence level is calculated based on goodness of fit, as well as signal/phase to noise ratios in both magnitude and wave images.

SE-EPI MRE allows rapid data acquisition of the x, y, and z components of the vector tissue motion over a large volume of the liver in a reasonable time. It allows a 3D vector-based inversion algorithm for data processing (Figure 2). The tissue stiffness estimation based on this 3D MRE method is more robust and accurate than 2D scalar MRE because it requires fewer assumptions about the polarization and propagation direction of the waves and thus can handle more complex shear wave motion in organs with complicated shapes such as the liver, spleen, and pancreas [33–36]. It has been demonstrated that the 3D MRE has higher diagnostic accuracy than 2D MRE in diagnosing NAFLD advanced fibrosis [37].

Figure 2.

Scan coverage of 4-slice 2D scalar GRE-MRE and 32-slice 3D vector EPI-MRE in the liver.

MRE research is motivated by clinical implementations to provide high sensitivity of elasticity, viscosity, and poroelastic properties to structural variations of biological tissues at multiple scales(15). The frequency dispersion of parameters measured by MRE has been explored to characterize the dynamic responses of structure elements in biological tissues[38–40]. Some investigations demonstrated the feasibility of low-frequency MRE, which is potentially more sensitive to the fluid phase of the tissue[41 42].

Multiple parameters calculation

Compared to 2D-MRE, 3D-MRE allows a more comprehensive analysis of the steady-state dynamic shear wave propagation in the entire liver. Thus, it enables calculation of multiple MR parameters that are sensitive to viscoelastic and compressible alterations of liver tissue in the progression of CLDs.

It has been demonstrated that the MRE-assessed liver stiffness has a static component that is mainly determined by extracellular matrix composites and liver structure (e.g., hepatic fibrosis, necrosis, loss of hepatocytes, regeneration, etc.), and a dynamic component that is affected by intrahepatic hemodynamic changes (e.g., perfusion, congestion, and inflammation)[43 44].

Advanced elastography methods explore multiple mechanical quantities include the model-free properties and model-based viscoelastic parameters [45–54] (Figure3). Among them, liver viscosity was found to be correlated with fibrosis, but not to steatosis or disease activity (inflammation) [55]. The dispersions of shear wave velocity and attenuation were found to be associated with the degree of steatosis [56]. The damping ratio and the loss modulus were found to increase significantly at the early onset of liver injury or necroinflammation [57]. The volumetric strain was found to be a promising biomarker in predicting portal hypertension in a preclinical study[58]. Another potential application is slip interface imaging [59], which can be used to characterize boundary conditions of the focal lesions to predict interface adhesiveness, which may provide promises in determining the invasiveness and malignancy of the tumor.

Figure 3.

Multiple mechanical properties derived from 3D vector MRE.

Improve patient experience

Nowadays, in most published studies, a rigid plastic pneumatic driver was used in hepatic MRE scans, which may cause discomfort for patients. To improve the patient experience during the scan, a flexible and soft pneumatic driver has been developed recently[60]. Compared to the rigid driver, the flexible and soft driver conforms better to the anterior chest wall, and is closer to the liver, which enables the propagation of more uniform shear waves and potential improvement in liver stiffness estimation accuracy[66] (Figure 4). There are studies proved that the repeatability and reproducibility of the flexible driver are as good as the rigid one[16].

Figure 4.

Images showing different driver designs and corresponding MRE images obtained in the same subject.

Courtesy of J. Chen, PhD, Rochester, MN.

Another technical concern related with patient experience is that the conventional hepatic MRE should be performed using expiration breath holds to avoid respiratory motion artifacts in the images [17, 18]. However, some patients have difficulty performing adequate end-expiration breath holds (e.g., pediatric and sedated patients). To eliminate the need for breath holds, investigators developed a non-gated, free-breathing, single-shot, multi-slice 2D EPI-MRE technique with a view-sharing-based reconstruction strategy[61], which can generate elastograms every 0.8 seconds and accomplish 100 time points within 1.5min. This implementation of free-breathing MRE has comparable repeatability and provides accurate averaged liver stiffness measurement compared with conventional breath-held MRE[62]. Additionally, this non-gate free-breathing MRE is capable of using the respiratory cycle to measure liver stiffness and other third-order mechanical parameters that may be helpful in disease diagnosis[63] (Figure 5). Another group demonstrated the feasibility of a respiratory-triggered (RT) SE-EPI MRE, which also yields comparable results of breath-held MRE[15]. Free-breathing MRE technique will be very beneficial for pediatric and sedated patients and will improve the comfort and patient experience for the general population as well.

Figure 5.

A. Images of free-breathing MRE in a healthy volunteer acquired at different time points during the scan. B. Graph plot showing excellent correlation between liver stiffness measured from free-breathing MRE and with breath hold, and illustrating excellent agreement between the two methods

Minimize inter-observer variation

MRE was introduced in 2007 for the clinical application of measuring liver stiffness, and are widely available on many MRI vendors, such as GE, Siemens, and Philips, with standardized hardware (Resoundant, Inc.) and inversion software (MMDI). To further improve inter-observer reproducibility and remove the need for manual analysis in MRE, a fully automated segmentation algorithm has been developed for calculating liver stiffness. This automated method is highly consistent with the measurements manually performed by expert readers in both 2D MRE and 3D MRE[64 65].

The repeatability coefficient of 2D MRE has been claimed as 19% by Quantitative Imaging Biomarkers Alliance (QIBA), which means a measured change in hepatic stiffness of 19% or larger indicates that a true change in stiffness has occurred with 95% confidence[66]. In a pilot repeatability study of 9 healthy volunteers and 6 patients, free-breathing MRE has a comparable RC value of 21% compared with that of breath-held MRE of 20%. In the same study cohort, 3D MRE provides a superior RC value of 10%[62].

Clinical applications

NAFLD

With the rising prevalence of obesity, NAFLD has become the leading cause of CLDs in the Western world[67]. Approximately 20–25% of NAFLD patients develop nonalcoholic steatohepatitis (NASH), leading to faster fibrosis progression to end-stage liver disease and hepatocellular carcinoma, which are established risk factors of liver-related death[68]. In preclinical studies, the liver stiffness derived from MRE has been proved to be sensitive to anti-fibrotic treatment, which supports the use of MRE as a non-invasive method to evaluate treatment efficacy longitudinally[69]. The multi-parametric 3D MRE combined with MRI-assessed proton density fat fraction shows high accuracy in predicting the NAFLD activity score (NAS) and NASH diagnosis in both preclinical models and clinical patients with NAFLD [43 44].

Even though MRE has been demonstrated to be highly accurate in diagnosing advanced fibrosis in patients with NAFLD[37 70 71], there are still diagnostic challenges in the detection of NASH[72]. In recent studies, the damping ratio and loss modulus have been shown to differentiate early onset of inflammation from fibrosis, even before the development of histologically detectable inflammatory cellular invasion[73]. A streamlined imaging protocol for NASH clinic has been recently established for virtual NAS prediction. This abbreviated imaging protocol takes as little as 5 minutes magnet time with the combination of multi-frequency 3D MRE and multi-echo Dixon imaging. It offers the advantage to predict not only NAS (the most commonly used surrogate end-point in NASH trials) but also separate estimations of the three components of NAS (steatosis, inflammation/ballooning, and fibrosis), which are individually targeted in certain experimental monotherapies[43] (Figure 6). This virtual NAS also reflects the histologic changes of NASH resolution in patients after bariatric surgery[74]. It is conceivable that this streamlined liver imaging protocol can be appropriately counseled about the risk of NAFLD disease progression and advised to implement therapeutic interventions.

Figure 6.

Examples of imaging analyses and predicted probabilities of NASH and NAS with 68% confidence intervals from (A, B) two clinical patients. The prediction model is composed of 3 imaging parameters, including shear stiffness, damping ratio, and fat fraction.

Chronic viral hepatitis

In patients with viral hepatitis, MRE has been demonstrated to be more accurate than aspartate aminotransferase-to-platelet ratio index (APRI) and many other non-invasive biomarkers in detecting significant fibrosis[75]. Additionally, MRE provides a cutoff value of 2.8kPa as a threshold for initiating antiviral therapy in patients with hepatitis C virus (HCV) infection [76].

Portal hypertension

The MRE-assessed liver stiffness has previously performed well in the detection of clinically significant PHTN[77 78], and in the estimation of the presence of esophageal varices[79]. One group found that the MRE-assessed liver stiffness was significantly higher in patients with cirrhotic PHTN, when compared with non-cirrhotic PHTN[80]. Recent work showed that the ratio between the spleen and liver stiffness can distinguish cirrhotic PHTN and non-cirrhotic PHTN[81]. Moreover, there was a preclinical study showing that a prediction model with multiple parameters derived from MRE has the potential to monitor PHTN progression[58].

Hepatocellular carcinoma

As the most common primary hepatic malignancy, the prognosis of patients with HCC is related to the aggressiveness and recurrence of the tumor[82]. One group found that each 1-kPa increase in tumor stiffness was associated with a 16.3% increase in the risk for tumor recurrence[83], which demonstrated that MRE has high diagnostic accuracy for predicting HCC development and stratifying the risk of HCC development.

The MRE-derived slip interface imaging may be related to the microvascular invasion of HCC, which can be used to stage malignancy and predict prognosis[84] (Figure 7).

Figure 7.

A 57-year-old female with a complete slip interface and no microvascular invasion at pathology. There is a stiffer HCC in left lobe compared with background liver tissue. The HCC-liver interface is clearly delineated in the octahedral shear strain (OSS) map, which indicates this HCC was well-incapsulated. The risk of microvascular invasion may be lower compared with those patients without clear slip interface shown in between HCC and liver.

Courtesy of J. Wang, MD, Guangzhou, Guangdong, People’s Republic of China.

Summary

Hepatic MRE has been demonstrated to be the most accurate non-invasive technique in diagnosing fibrosis and cirrhosis with liver stiffness measurement. With recent technology developments, multiparametric MRE can provide more promising parameters for evaluating pathogenic changes during disease progression of CLD, with substantially improved patient experience via more rapid, comfortable, and reliable imaging.

Synopsis.

Magnetic Resonance Elastography (MRE) has been well-accepted as the most accurate non-invasive technique in diagnosing fibrosis and cirrhosis in patients with chronic liver disease (CLD). To further improve the technical reliability and discover new MRE biomarkers, investigators made innovative progress in imaging sequence design, active/passive actuator design, and elastographic inversion algorithms. The accuracy of hepatic MRE in distinguishing the severity of disease has been validated in different studies of patients with CLDs, including nonalcoholic fatty liver disease, chronic hepatitis virus infection, portal hypertension, and hepatocellular carcinoma. The advanced hepatic MRE has been established as a reliable, comfortable, and inexpensive alternative to liver biopsy for disease diagnosing, progression monitoring, and clinical decision making in patients with CLD.

This comprehensive review will summarize current knowledge of the technical advances and innovations of hepatic MRE development, and the clinical applications in various hepatic diseases.

Key Points.

• MR elastography (MRE) has been well-accepted as the most accurate non-invasive assessment of hepatic fibrosis and cirrhosis in patients with chronic liver disease (CLD).

• Investigators have made numerous improvements in imaging sequence, active/passive actuator, and elastographic inversion algorithms to enhance the technical reliability and discover new MRE biomarkers.

• Current updates about applications of hepatic MRE in disease diagnosing, progression monitoring, and clinical decision making for patients with CLD.