Agreement and accuracy of shear-wave techniques (point shear-wave elastography and 2D-shear-wave elastography) using transient elastography as reference

Abstract

Objective: We aimed to evaluate the agreement/accuracy of point shear-wave elastography (p-SWE) and 2D-shear-wave elastography (2D-SWE) for liver fibrosis staging using transient elastography (TE) as the reference.

Methods: This retrospective study analyzed data from people with chronic liver diseases submitted to TE, p-SWE, and 2D-SWE. Liver fibrosis stages were defined using the TE's 'rule of five': normal (<5 kPa); suggestive of compensated-advanced chronic liver disease (cACLD) (10-15 kPa); highly suggestive of cACLD (15-20 kPa); suggestive of clinically significant portal hypertension (>20 kPa). Agreement and accuracy of p-SWE and 2D-SWE were assessed. Optimal cutoffs for p-SWE and 2D-SWE were identified using the point nearest to the upper left corner of the ROC curves.

Results: A total of 289 participants were included. The correlation between TE and 2D-SWE (rho = 0.59; P < 0.001) or p-SWE (rho = 0.69; P < 0.001) was satisfactory. The AUROCs (95% CI) of 2D-SWE and p-SWE for TE ≥ 5 kPa; TE ≥ 10 kPa; TE ≥ 15 kPa and TE ≥ 20 kPa were 0.757 (0.685-0.829) and 0.741 (0.676-0.806); 0.819 (0.770-0.868) and 0.870 (0.825-0.915); 0.848 (0.803-0.893) and 0.952 (0.927-0.978); 0.851 (0.806-0.896) and 0.951 (0.920-0.982), respectively. AUROCs of 2D-SWE were significantly lower compared with p-SWE for detecting cACLD. Optimal thresholds of 2D-SWE and p-SWE for TE ≥ 15 kPa were 8.82 kPa (sensitivity = 86% and specificity = 79%) and 8.86 kPa (sensitivity = 90% and specificity = 92%), respectively.

Conclusion: LSM by p-SWE and 2D-SWE techniques were correlated with TE. LSM by p-SWE seems to be more accurate than 2D-SWE to identify patients with more advanced fibrosis.