Shear wave elastography with a new reliability indicator.

Non-invasive methods for liver stiffness assessment have been introduced over recent years. Of these, two main methods for estimating liver fibrosis using ultrasound elastography have become established in clinical practice: shear wave elastography and quasi-static or strain elastography. Shear waves are waves with a motion perpendicular (lateral) to the direction of the generating force. Shear waves travel relatively slowly (between 1 and 10 m/s). The stiffness of the liver tissue can be assessed based on shear wave velocity (the stiffness increases with the speed). The European Federation of Societies for Ultrasound in Medicine and Biology has published Guidelines and Recommendations that describe these technologies and provide recommendations for their clinical use. Most of the data available to date has been published using the Fibroscan (Echosens, France), point shear wave speed measurement using an acoustic radiation force impulse (Siemens, Germany) and 2D shear wave elastography using the Aixplorer (SuperSonic Imagine, France). More recently, also other manufacturers have introduced shear wave elastography technology into the market. A comparison of data obtained using different techniques for shear wave propagation and velocity measurement is of key interest for future studies, recommendations and guidelines. Here, we present a recently introduced shear wave elastography technology from Hitachi and discuss its reproducibility and comparability to the already established technologies.

Introduction

Many different non-invasive methods for liver stiffness (LS) assessment have been introduced over recent years. Of these, two main methods for estimating liver fibrosis using ultrasound elastography have become established in clinical practice: shear wave elastography (SWE) and quasi-static or strain elastography (SE)(. The European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) has published Guidelines and Recommendations that describe these technologies and provide recommendations for their clinical use(. In summary, shear waves are generated when a directional force is applied to a tissue, causing shear deformation. Shear waves are rapidly attenuated by tissues, and they travel at a much slower rate (between 1 and 10 m/s), and thus can be easily detected by longitudinal ultrasonic waves. They are not supported by liquids of low viscosity(. The stiffness of the liver tissue can be assessed based on shear wave velocity (the stiffness increases with the speed). A comparison of data obtained using different techniques for shear wave propagation and velocity measurement is of key interest for future studies, recommendations and guidelines. Most of the data available to date has been published using the Fibroscan (Echosens, France)(, point shear wave speed measurement using an acoustic radiation force impulse (ARFI) (Siemens, Germany)( and 2D SWE using the Aixplorer (SuperSonic Imagine, France)(. More recently, other manufacturers such as GE, Philips and Toshiba have introduced SWE technology into the market(. Here, we present a recently introduced SWE technology from Hitachi and discuss its reproducibility and comparability to already established technologies.

Shear wave measurement technology

Shear wave measurement (SWM) technology using the ARFI principle with a convex probe for the examination of the liver was recently introduced by Hitachi. It enables quantitative evaluation of shear wave velocity and adds a reliability indicator to each shear wave velocity measurement, providing a measured indication of the appropriateness of the measurement(. The aim of this report is to describe the principles of SWM and the contribution of the reliability indicator to the diagnostic performance.

Method

Shear Wave Measurement uses an ARFI (the push pulse) to generate shear waves in a small region of interest (ROI) in the liver, followed by longitudinal ultrasound tracking pulses that measure the speed of propagation of these shear waves. The SWM transmit/receive sequence is shown in Fig. 1. Push pulses are transmitted in one direction to generate shear waves; tracking pulses are alternately transmitted and received in two different directions to detect the speed of shear wave propagation. For each ‘push-track’ sequence, the shear wave velocity (Vs) is measured at multiple depths in the ROI and the sequence is automatically repeated, so the propagation speed is measured several times in a short period of time. This is followed by a probe cooling time, when the entire wave transmission is stopped. SWM conforms to standard regulations applicable to diagnostic ultrasound systems, related to power output measurements and probe surface heating.

Fig. 1

SWM transmit/receive sequence

Reliability Indicator, V N

It is common to display just one median V value for the data measured inside the ROI in point shear wave speed measurement methods. However, it can be difficult to assess whether a measurement is appropriate or not from the V value alone. Disturbances can result from body motion, respiratory movement of the patient, unsteady handling of the transducer by the examiner, etc. In addition, the movement of the microvasculature due to cardiac motion or vascular flow can make echo signals fluctuate over time in the parenchyma of the liver. Even when the standard deviation is displayed, it can be difficult to distinguish if the variation is due to tissue structure or measurement error. The SWM technique is designed to detect phase fluctuations caused by blood flow and micro-vibrations to overcome this problem, and as a result, the shear wave velocity (Vs) can be calculated with high accuracy.

The reliability indicator is calculated for each set of V values acquired from multiple Push-Track sequences and multiple V measurements at different depths inside the ROI for each measurement sequence. Values are rejected from the V set using three defined criteria, and the ratio of the remaining V values (Vs after rejection /total number of Vs) is defined as Vs efficacy rate, and displayed as a percentage (the V N value). Median V and interquartile range (IQR) are calculated from the histogram of each V set after rejection (Fig. 2). The three rejection conditions are illustrated in Fig. 3.

Fig. 2

SWM measurement. The region of interest and the histogram are displayed. The display item [unit] and description are shown as well

Fig. 3

Schematic diagram of rejection conditions. Negative Vs (1): Vs takes a negative value when the peak of track 2 is detected at an earlier time than the peak of track 1 due to, for example, disturbed shear waves. Shear waves in this case are not correctly detected and thus the value is rejected. Vs is outside of a defined range (2): A certain range of Vs values is defined depending on the organs and tissues being examined. If Vs values are beyond that range, the value is regarded as a detection error of the shear wave speed and rejected. Phase fluctuation detected at a particular depth (3): As described earlier, phase fluctuations due to blood vessels and blood flow are different from shear waves and rejected as a detection error

Reliability of SWM

SWM accuracy was evaluated using seven phantoms of different stiffness (with range of shear wave propagation velocities between 1.19 and 3.79 m/secs). The measured SWM V values were compared with stiffness values obtained with the mechanical tester INSTRON Model 2519 (Illinois Tool Works Inc., USA). For each phantom, 10 different probe positions were selected and at each measuring position, average Vs, coefficient of variation (%CV; standard deviation / average) showing measurement reproducibility, and accuracy level (ΔV %) showing the difference compared with the mechanical tester were acquired. The measurements were taken at depths of 20, 40 and 60 mm. The measurement re-producibility (%CV) was between 3 and 16% and measurement accuracy level (ΔV ) within ±15%. No significant difference in V values were observed depending on the depth. Furthermore, V N remained over 80% for the phantoms up to 3 m/sec of stiffness, indicating that the measurements are reproducible. On the other hand, with the stiffest phantom, when the shear wave speed was on the border of the defined range, the V N dropped to between 25 to 60%, showing that the measurement reliability decreased. The above analysis and a clinical study by Yada et al.( have demonstrated that the measurement reliability can be evaluated quantitatively in a clinical setting using VN.

Initial clinical experience

Our first and preliminary results show that data can be reliably measured in 92% of subjects, with failures mainly related to obesity. So far, all healthy subjects showed values below 7.1 kPa (Fig. 4) and patients with liver cirrhosis had values above 13.8 kPa. This corresponded well with other equipment used (S2000 [Siemens], Logiq E9 [GE] and Aixplorer [Supersonic]). Patients with intermediate fibrosis had values between 5.1–14.4 kPa. Lower values were obtained in patients with a lower degree of fibrosis, however, a significant overlap was observed as the same is also seen using other equipment, including Fibroscan(. Future comparative studies are needed to confirm the results.

Fig. 4

A healthy subject. Series of images showing the reliability of data (A–E)

Discussion

Two main elastography methods using ultrasound elastography have become established in clinical practice: quasi-static or strain elastography (SE)( for the examination of the liver(, pancreas(, lymph nodes(, anorectum(, thyroid(, and other organs and SWE mainly for non-invasive liver stiffness assessment(. Ultrasound-based SWE techniques are available for quantitative elastography measurements in clinical practice, including vibration-controlled transient elastography (VCTE), shear-wave point quantification and 2D (3D) SWE techniques(. Different techniques have been thoroughly evaluated and the methods are described in detail in the EFSUMB guidelines on elastography(. Here, we were first to describe that shear wave measurement technology enables quantitative evaluation of shear wave velocity and adds a reliability indicator to each shear wave velocity measurement.

Conclusion

The recently introduced SWE technology from Hitachi allows quantitative evaluation of shear wave velocity and adds a reliability indicator to each shear wave velocity measurement. Future studies should investigate the comparability of the new SWE technology from Hitachi to the already established SWE technologies. Open questions include the comparability and reproducibility between the different manufacturers with respect to normal SWE values and the cut off values for liver cirrhosis as well as the different stages of fibrosis.