AUTHOR=Dong Bingtian , Chen Yuping , Lyu Guorong , Chen Yongjian , Qin Ran 

TITLE=Quantitative Assessment of Portal Hypertension by Two-Dimensional Shear Wave Elastography in Rat Models of Nonalcoholic Fatty Liver Disease: Comparison With Four Composite Scores

JOURNAL=Frontiers in Medicine

VOLUME=Volume 9 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2022.844558

DOI=10.3389/fmed.2022.844558

ISSN=2296-858X

ABSTRACT=Background: Measurement of hepatic venous pressure gradient is the gold standard to assess portal hypertension (PH) but is invasive with potential complications. We aimed to assess the performance of liver and spleen stiffness measurement (LSM and SSM, respectively) using two-dimensional shear wave elastography (2D-SWE), and composite scores including liver stiffness-spleen diameter to platelet ratio score (LSPS), platelet count (PLT)/spleen diameter ratio (PSR), aspartate aminotransferase (AST)/alanine aminotransferase ratio (AAR), and AST-to-PLT ratio index (APRI) for diagnosing PH in nonalcoholic fatty liver disease (NAFLD) rat models.
Methods: Animal models with PH in NAFLD were established in 65 rats, which then underwent 2D-SWE measurements. Morphological and biological parameters were collected to calculate the four composite scores. Correlations of noninvasive methods with portal venous pressure were evaluated using the Spearman correlation analysis. The area under the receiver operating characteristic curve (AUC) was used to assess the performance of noninvasive methods in predicting PH.
Results: LSM and SSM were significantly associated with portal venous pressure (r = 0.636 and 0.602, respectively; all P < 0.001). AUCs of LSM and SSM in the diagnosis of PH were 0.906 (95% confidence interval [CI]: 0.841‒0.970) and 0.870 (95% CI: 0.776‒0.964), respectively, which were significantly higher than those in composite scores. The AUCs for LSPS, PSR, AAR, and APRI were 0.793, 0.520, 0.668, and 0.533, respectively, for diagnosing PH. The AUCs of the combined models of LSM and SSM, LSM and PLT, SSM and PLT, and LSM, SSM and PLT were 0.923, 0.913, 0.872, and 0.923, respectively. The four combined models showed no statistical differences compared to LSM and SSM in evaluating PH (all P > 0.05).
Conclusions: LSM and SSM by 2D-SWE can be used as promising noninvasive parameters to diagnose PH in NAFLD with higher accuracy than that of composite scores. The combined models, as compared to LSM and SSM, did not significantly improve the performance in diagnosing PH.