Factors Associated with Disagreement of Fibrosis Stages between 2D-Shear Wave Elastography and Transient Elastography in Chronic Hepatitis B.

Abstract

The agreement of elastography techniques in chronic Hepatitis B (CHB) needs evaluation. We aimed to evaluate, in CHB, the agreement between transient elastography (TE) and two-dimensional shear wave elastography (2D-SWE), analyzing the factors related to the disagreement of measures. CHB patients underwent liver stiffness measures with both TE and 2D-SWE on the same day. For concordance analysis, we defined liver fibrosis as F0/1 vs. F ≥ 2, F0/1-F2 vs. F ≥ 3 and F0/1-F2-F3 vs. F4 for both methods. Logistic regression analysis was used to identify the variables independently associated with the disagreement between methods. A total of 150 patients were enrolled. Liver fibrosis categorization according to TE was: F0-F1 = 73 (50.4%), F ≥ 2 = 40 (27.6%), F ≥ 3 = 21 (14.5%) and F4 = 11 (7.6%), and according to 2D-SWE was: F0/F1 = 113 (77.9%), F ≥ 2 = 32 (22.1%), F≥ 3 = 25 (17.2%) and F4 = 11 (7.6%). It was observed that 20.0% of the sample had steatosis (CAP≥ 275 dB/m). TE and SD-SWE estimated equal fibrosis stages in 79.3% of cases. Spearman's correlation coefficient was 0.71 (p < 0.01). Kappa values for F ≥ 2, F ≥ 3 and F = 4 were: 0.78, p < 0.001; 0.73, p < 0.001; and 0.64, p < 0.001, respectively. Diabetes mellitus (DM) (OR 5.04; 95%CI: 1.89-13.3; p < 0.001) and antiviral treatment (OR 6.79; 95%CI: 2.33-19.83; p < 0.001) were independently associated with discordance between both methods. In CHB, there is strong correlation and good agreement between TE and 2D-SWE in identifying fibrosis stages. Diabetes mellitus and antiviral therapy may impact the agreement of stiffness measures obtained with these elastographic methods.