Published online Jun 7, 2023. doi: 10.3748/wjg.v29.i21.3318
Peer-review started: February 27, 2023
First decision: March 14, 2023
Revised: March 23, 2023
Accepted: May 6, 2023
Article in press: May 6, 2023
Published online: June 7, 2023
Processing time: 94 Days and 5.3 Hours
Chronic liver disease is a growing problem worldwide. Accurate assessment of liver fibrosis is important for treatment prioritization, surveillance, and determination of prognosis. Liver biopsy is still considered as the gold standard for staging liver fibrosis. However, liver biopsy is an expensive and invasive diagnostic tool. Its main complications are bleeding and pain, which limit its clinical application. Recently, the application of two-dimensional shear wave elastography (2-D SWE) in the diagnosis of non-invasive assessment of liver fibrosis has developed rapidly. However, the presence of artifacts leads to inaccurate liver stiffness (LS) measurements.
Although 2-D SWE artifacts of the liver are common in clinical practice, they are poorly recognized, and there is even no clear definition. To the best of our knowledge, only a few review articles have been published. Knowledge of the artifacts is essential to improve operation technology to obtain high-quality images. It is very important to obtain accurate measurements in an attempt to optimize its performance and application value.
We aim to investigate the presence and influence of artifacts in 2-D SWE of liver.
In this study, we performed 2-D SWE examination in patients with chronic liver disease by a novice and an expert. The elastogram was divided into four locations: top-left, top-right, bottom-left, and bottom-right. The occurrence frequency of artifacts in different locations was compared. The effect of artifacts on the LS measurements was evaluated by comparing the elastogram with the most artifacts (EMA) and the elastogram with the least artifacts (ELA).
Each operator had 720 elastography images were included for analysis. The percentage of elastograms with artifacts and the area of artifacts in the novice were significantly higher than those in the expert (both P < 0.001). Comparing the occurrence frequency of artifacts in all locations of the two operators, it was found that both operators had the highest frequency of bottom-left, followed by top-left and bottom-right, and top-right had the lowest frequency. This study showed that the LS values and standard deviation values of the EMAs were higher than those of the ELAs. Both operators had lower stability index values and intraclass correlation coefficient values for EMAs than ELAs.
Artifacts are common when using 2-D SWE to measure LS, especially for the novice. Our results showed artifacts were more likely to occur in the bottom-left corner of the elastogram. Artifacts may lead to the overestimation of LS and reduce the repeatability and reliability of LS measurements.
In this study, we only analyzed a small sample of data from two operators of one device. Therefore, a larger sample study involving more operators and devices needs to be conducted in future studies.