Related Experiment Video
Updated: Apr 10, 2026
Author Spotlight: Advancing Hepatic Fibrosis Diagnosis Using Magnetic Resonance Elastography and AI
Published on: July 21, 2023
Prognostication of Liver Disease Patients via the WFUMB 'Rule-of-4' Algorithm Using SSI-2D-SWE-Based Liver Stiffness
Christian Sebesta1,2,3, Georg Kramer1,2,3,4, Nina Dominik1,2,3
1Division of Gastroenterology and Hepatology, Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria.
Background:
The World Federation for Ultrasound in Medicine and Biology (WFUMB) guidance update 2024 proposed a 'rule-of-4' using ARFI liver stiffness measurement (LSM) to stratify the risk of decompensation events. This rule identifies advanced chronic liver disease (ACLD) at a threshold of ≥ 13 kPa and indicates a high probability of clinically significant portal hypertension (CSPH) above 21 kPa.
Methods:
We prospectively enrolled 187 patients undergoing same-day SSI-2D-LSM (Aixplorer SuperSonic Imagine) and hepatic venous pressure gradient (HVPG) measurement. Patients were stratified into three LSM groups < 13 kPa, 13-21 kPa and > 21 kPa and followed for decompensation events for 1 year (Y1).
Results:
The cohort comprised 31 (16.6%) patients with LSM < 13 kPa (non-ACLD); 33 (17.6%) LSM 13-21 kPa (ACLD) and 123 (65.8%) LSM > 21 kPa (high CSPH probability). The corresponding median HVPG values were 5 [IQR: 3-7] vs. 10 [7-14] vs. 17 [13-21] mmHg, respectively, and the corresponding CSPH prevalence was 6.5%, 54.5% and 91.9%. In ROC analysis to predict CSPH, SSI-2D-LSM demonstrated high accuracy (AUC: 0.83). Validating the new WFUMB extension of the 'rule-of-4', the recommended > 21 kPa threshold identified CSPH with a specificity of 80.0% and a PPV of 83.1%. In competing risk analysis, the Y1-decompensation rate was 0%, 10% and 24.7% of patients in the < 13 kPa, 13-21 kPa and > 21 kPa groups, respectively. For prediction of Y1-decompensation risk in the compensated ACLD (cACLD) subgroup, HVPG demonstrated the highest predictive accuracy (AUC: 0.92), while SSI-2D-LSM (AUC: 0.75) performed similarly to vibration-controlled transient elastography (VCTE: AUC: 0.77; AUC-comparison: p = 0.837).
Conclusions:
The WFUMB 'rule-of-4' for ARFI-LSM allows for accurate and point-of-care non-invasive risk stratification of patients with liver disease. Specifically, the short-term risk of decompensation starts at SSI-2D-LSM ≥ 13 kPa and becomes considerable at SSI-2D-LSM > 21 kPa (indicating high CSPH probability). These findings support the broader implementation of ARFI-LSM for risk assessment in clinical routine.
