Predicting Clinically Significant Portal Hypertension in MASLD Using Routine Ultrasound and Clinical Data: A...

Area of Science:

• Hepatology
• Gastroenterology
• Noninvasive Diagnostics

Background:

• Metabolic dysfunction-associated steatotic liver disease (MASLD) can lead to clinically significant portal hypertension (CSPH).
• Accurate noninvasive identification of CSPH in MASLD patients is crucial but remains difficult.
• CSPH is a key factor in adverse outcomes for MASLD patients.

Purpose of the Study:

• To assess the utility of routine ultrasound and clinical variables for identifying CSPH in adults with MASLD.
• To evaluate the performance of noninvasive markers in predicting CSPH (defined as HVPG ≥ 10 mmHg).

Main Methods:

• Retrospective cohort study of 218 adults with MASLD who underwent hepatic venous pressure gradient (HVPG) measurement.
• Evaluation of demographic, laboratory, routine ultrasound, and liver stiffness data.
• Application of logistic regression and machine learning models for CSPH prediction.

Main Results:

• CSPH was present in 44 (20.2%) patients.
• Patients with CSPH exhibited less favorable noninvasive profiles, particularly in liver stiffness and spleen length.
• Machine learning models showed modest predictive performance (AUC up to 0.67), with portal vein calibre and liver stiffness being independently associated with CSPH.

Conclusions:

• Routinely available ultrasound and clinical data have limited ability to identify CSPH in MASLD.
• Current noninvasive markers are independently associated with CSPH but insufficient for standalone clinical triage.
• Further refinement of MASLD-specific noninvasive prediction strategies is needed before clinical implementation.