Hepatic Venous Pressure Gradient, Noninvasive Tests, and Prognosis Across the Subtypes of Advanced Steatotic Liver...

Steatotic liver disease encompasses metabolic dysfunction-associated steatotic liver disease, alcohol-related liver disease, and a mixed type (metabolic dysfunction-associated and alcohol-related liver disease). We investigated the course of compensated advanced chronic liver disease across steatotic liver disease subtypes, the impact of clinically significant portal hypertension, and the role of noninvasive tests.

This study included patients with steatotic liver disease and compensated advanced chronic liver disease from 17 centers undergoing hepatic venous pressure gradient and noninvasive test assessment through December 2023. Primary outcomes were first hepatic decompensation and liver-related mortality.

Among 696 patients (alcohol-related liver disease, 22%; metabolic dysfunction-associated and alcohol-related liver disease, 20%; metabolic dysfunction-associated steatotic liver disease, 58%), 94% had Child-Pugh A, and 60% had clinically significant portal hypertension. Over a median follow-up of 3.8 years, 170 patients decompensated, and 137 died (58% liver-related). The 3-year cumulative decompensation incidence was highest in alcohol-related liver disease (26%), followed by metabolic dysfunction-associated and alcohol-related liver disease (21%) and metabolic dysfunction-associated steatotic liver disease (15%); yet differences were no longer significant after adjusting for age, sex, Model for End-stage Liver Disease, albumin, and hepatic venous pressure gradient. Similarly, liver-related mortality did not differ by steatotic liver disease subtype in adjusted models. Clinically significant portal hypertension independently predicted decompensation across all subtypes. The ANTICIPATE-NASH model showed high and comparable diagnostic accuracy for clinically significant portal hypertension across alcohol-related liver disease, metabolic dysfunction-associated and alcohol-related liver disease, and metabolic dysfunction-associated steatotic liver disease (area under the receiver operating characteristic curve = 0.831-0.894). An ANTICIPATE-NASH probability ≥60% for ruling-in clinically significant portal hypertension outperformed the original ANTICIPATE model in all subtypes and independently predicted decompensation with similar effect sizes (adjusted subdistribution hazard ratio, 2.57-2.71).

Higher crude decompensation rates in alcohol-related liver disease and metabolic dysfunction-associated and alcohol-related liver disease reflect more severe liver disease at presentation, yet outcomes are comparable across the steatotic liver disease spectrum after accounting for disease severity. ANTICIPATE-NASH accurately identifies clinically significant portal hypertension across steatotic liver disease subtypes, and both invasive and noninvasive measures of portal hypertension predict decompensation and liver-related death.