Xiang Hongjin, Ling Wenwu, Ma Lin, Yang Lulu, Lin Tang, Luo Yan
Department of Ultrasound, West China Hospital of Sichuan University, Chengdu, China.
Quant Imaging Med Surg. 2022 May;12(5):2855-2865. doi: 10.21037/qims-21-863.
Measurements of liver stiffness obtained with 2-dimensional shear wave elastography (2D-SWE) have been widely used to clinically assess liver fibrosis. However, differences between different 2D-SWE systems can lead to confusion when interpreting measurements. This study investigated the variability between a recently released sound touch elastography (STE) system and a supersonic shear imaging (SSI) system and assessed the degree of intersystem discrepancy using the different liver stiffness value (LSV) thresholds recommended by the Society of Radiologists in Ultrasound (SRU) for assessing liver fibrosis.
A total of 4,152 patients who had undergone STE and SSI on the same day were enrolled in this retrospective study. First, intrasystem agreement for STE and SSI was assessed. Then, intraclass correlation coefficients (ICCs) and the Bland-Altman method were used to assess intersystem variability for all cases, classified according to the thresholds recommended by the SRU. The effects of age, gender, and body-mass index (BMI) were evaluated using multivariate linear regression analysis and attributive intervals were computed for STE and SSI at each of the different thresholds.
The ICCs for STE and SSI intrasystem agreement were 0.94 [95% confidence interval (CI): 0.937-0.943; P<0.001] and 0.984 (95% CI: 0.984-0.985; P<0.001), respectively. The 95% limit of agreement (LOA) for all cases ranged from -6.96 to 7.44 kPa. The 95% LOA increased as the threshold values rose, and intersystem variability was obvious, even at the smallest threshold (the 95% LOA at values ≤5 kPa was -0.85 to 2.08 kPa, while that at values >17 kPa was -20.81 to 14.71 kPa). The adjusted R for age, gender, and BMI was only 0.018 (all P value <0.05).
There was clear variability between STE and SSI, in contrast with some previous studies with small sample sizes, and consistent with others. Intersystem variability increased with the elevation of the LSV thresholds recommended by the SRU. Gender and BMI had little effect on intersystem variability. Future research could compare STE and SSI in different liver diseases, assessing the feasibility of the SRU-recommended thresholds in proven pathologies and evaluating the test-retest repeatability.
二维剪切波弹性成像（2D-SWE）所测得的肝脏硬度已被广泛用于临床评估肝纤维化。然而，不同的2D-SWE系统之间的差异可能会在解释测量结果时导致混淆。本研究调查了最近发布的声触诊组织量化（STE）系统和超声剪切波弹性成像（SSI）系统之间的变异性，并使用美国放射学会超声学会（SRU）推荐的不同肝脏硬度值（LSV）阈值来评估系统间差异程度，以评估肝纤维化。
本回顾性研究共纳入4152例同日接受STE和SSI检查的患者。首先，评估STE和SSI的系统内一致性。然后，使用组内相关系数（ICC）和Bland-Altman方法评估所有病例的系统间变异性，并根据SRU推荐的阈值进行分类。使用多元线性回归分析评估年龄、性别和体重指数（BMI）的影响，并计算STE和SSI在每个不同阈值下的归因区间。
STE和SSI系统内一致性的ICC分别为0.94[95%置信区间（CI）：0.937-0.943；P<0.001]和0.984（95%CI：0.984-0.985；P<0.001）。所有病例的95%一致性界限（LOA）范围为-6.96至7.44kPa。随着阈值升高，95%LOA增大，即使在最小阈值时系统间变异性也很明显（值≤5kPa时95%LOA为-0.85至2.08kPa，而值>17kPa时为-20.81至14.71kPa）。年龄、性别和BMI的调整R值仅为0.018（所有P值<0.05）。
与一些之前小样本量的研究不同，STE和SSI之间存在明显变异性，这与其他研究一致。系统间变异性随着SRU推荐的LSV阈值升高而增加。性别和BMI对系统间变异性影响较小。未来的研究可以比较STE和SSI在不同肝脏疾病中的情况，评估SRU推荐阈值在已证实病理中的可行性，并评估重测重复性。
