Evaluation of portal hypertension and varices by acoustic radiation force impulse imaging of the liver compared to transient elastography and AST to platelet ratio index

Abstract

Background: Acoustic radiation force impulse imaging (ARFI) is a new method of liver stiffness measurement (LSM). The aim was to compare ARFI, transient elastography (TE) and AST to platelet ratio index (APRI) for the noninvasive diagnosis of clinically significant portal hypertension (CSPH, hepatic venous pressure gradient; HVPG ≥ 10 mmHg) and esophageal varices (EV).

Materials and methods: LSM via ARFI and TE was performed in 88 consecutive patients with cirrhosis prior to HVPG measurement. The mean liver stiffness for ARFI was calculated out of 5 measurements for each lobe.

Results: LSM by TE and ARFI was not successful in 22 (25%) patients and 1 (1 %) patient, respectively, due to ascites or obesity. Both TE (r = 0.765; p < 0.001) and ARFI (r = 0.646; p < 0.001) correlated significantly with HVPG. At the optimal cut-off (16.8 kPa), TE (area under the curve, AUC 0.870) yielded a sensitivity and specificity of 89.7% and 75%, respectively, for predicting CSPH. At the optimal cut-off (2.58 m/s), the sensitivity and specificity for ARFI (AUC 0.855) were 71.4% and 87.5%, respectively. Using an APRI (AUC 0.838), the sensitivity and specificity were 69% and 87.5%, respectively. The AUC for the diagnosis of EV was 0.802 for TE (cut-off: 27.9 kPa), 0.743 for ARFI (cut-off: 2.74 m/s), and 0.805 for APRI (cut-off: 1.90).

Conclusion: ARFI shows a higher applicability particularly in obese and ascitic patients. All three investigated methods show a high diagnostic accuracy for CSPH. Notably, APRI performed not significantly different compared to ARFI for the diagnosis of CSPH.