Assessment of the inter-platform reproducibility of ultrasound attenuation examination in nonalcoholic fatty liver disease

Abstract

Purpose: This study aimed to assess the inter-platform reproducibility of ultrasound attenuation examination in patients with nonalcoholic fatty liver disease (NAFLD).Methods: Between March 2021 and April 2021, patients with clinically suspected or known NAFLD were prospectively enrolled; each patient underwent ultrasound attenuation examinations with three different platforms (Attenuation Imaging [ATI], Canon Medical System; Tissue Attenuation Imaging [TAI], Samsung Medison; and Ultrasound-Guided Attenuation Parameter [UGAP], GE Healthcare) on the same day. The mean attenuation coefficient (AC) values of the three platforms were compared using repeated-measures analysis of variance with the Bonferroni correction. To evaluate inter-platform reproducibility, the AC values obtained for each platform were compared using Bland-Altman analysis with the calculation of 95% limits of agreement (LOA), intraclass correlation coefficients (ICCs), and coefficients of variation (CVs).Results: Forty-six patients (23 men; mean age±standard deviation, 52.3±12.4 years) were enrolled. The mean AC values showed significant differences among the three platforms (0.75±0.12, 0.80±0.11, and 0.74±0.09 dB/cm/MHz for ATI, TAI, and UGAP, respectively; P<0.001). For inter-platform reproducibility, the 95% LOAs were -0.22 to 0.11 dB/cm/MHz between ATI and TAI, -0.17 to 0.18 dB/cm/MHz between ATI and UGAP, and -0.08 to 0.20 dB/cm/MHz between TAI and UGAP, respectively. The pairwise ICCs were 0.790-0.797 in terms of absolute agreement among the three platforms; the CVs were 8.23%-9.47%.Conclusion: The AC values obtained from different ultrasound attenuation examination platforms showed significant differences, with significant inter-platform variability. Therefore, the AC values measured using different ultrasound attenuation examination techniques should not be used interchangeably for longitudinal follow-up of patients with NAFLD.

Interobserver Variability in Ultrasound‐Based Liver Fat Quantification

“…Lastly, it has been reported that measurements with attenuation imaging techniques have significant differences between ultrasound devices from different vendors. 22 Therefore, our results may not be generalizable for ultrasound devices from different manufacturers.…”

How many times should we repeat measurements of the ultrasound-guided attenuation parameter for evaluating hepatic steatosis?

“…However, there is no consensus on exactly how many measurements should be made for UGAP. In previous studies using UGAP, the numbers of measurements varied from 6 to 12 [ 24 , 34 – 36 ]. In contrast, studies have been conducted to arrive at a consensus on the number of repetitions required for modalities such as SWE [ 16 , 37 , 38 ].…”

“…Further studies comparing multiple ROIs and one ROI in the quality map using UGAP are warranted. Fourth, although there exist several ultrasound attenuation platforms, which should not be used interchangeably [ 36 ], only one platform (UGAP) was used to prove the hypothesis that a small number of repetitions would be enough to measure hepatic steatosis. Thus, further studies comparing many repetition protocols using several ultrasound attenuation platforms are needed.…”

“…Thus, further studies comparing many repetition protocols using several ultrasound attenuation platforms are needed. Finally, UGAP measurements were made by only one radiologist, although previous studies have demonstrated high intraobserver and interobserver reproducibility [ 35 , 36 ]. Additionally, the second set of six measurements might have been influenced by the first set of six measurements because all measurements were continuously performed by the same observer; however, this study focused on the preceding six measurements, which were not affected by the following six measurements.…”

Intra-individual comparison of two-dimensional shear wave elastography techniques using plane wave imaging and the multi-beam technique: are they interchangeable in measuring liver fibrosis?

“…Reliability using ICCs was classified as: ≥0.90, excellent; ≥0.75 to <0.90, good; ≥0.50 to <0.75, moderate; and <0.50, poor [17]. Pearson correlation coefficients were calculated to evaluate correlations between each 2D SWE technique and MR elastography and categorized as: 0-0.19, very weak; 0.2-0.39, weak; 0.40-0.59, moderate; 0.60-0.79, strong; and 0.80-1.0, very strong [36,37]. To evaluate whether inter-technique reproducibility is affected by the patient's BMI and causes of underlying chronic liver disease, Pearson correlation coefficient was calculated between the absolute between-technique differences in LS values and BMI; and Mann-Whitney U test was used to compare the absolute between-technique differences in LS values between patients with chronic hepatitis B and patients with other causes.…”