Monica Pons1, Salvador Augustin, Bernhard Scheiner, Maeva Guillaume, Matteo Rosselli, Susana G Rodrigues, Horia Stefanescu, Mang M Ma, Mattias Mandorfer, Mayka Mergeay-Fabre, Bogdan Procopet, Philipp Schwabl, Arnulf Ferlitsch, Georg Semmler, Annalisa Berzigotti, Emmanuel Tsochatzis, Christophe Bureau, Thomas Reiberger, Jaime Bosch, Juan G Abraldes, Joan Genescà. 1. 1Liver Unit, Department of Internal Medicine, Hospital Universitari Vall d'Hebron, Vall d'Hebron Institut de Recerca (VHIR), Faculty of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain; 2Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Instituto de Salud Carlos III, Madrid, Spain; 3Division of Gastroenterology and Hepatology, Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria; 4Vienna Hepatic Hemodynamic Laboratory, Medical University of Vienna, Vienna, Austria; 5Service d'Hépatologie CHU Toulouse Rangueil, Institut Cardiomet et Université Paul Sabatier, Toulouse, France; 6UCL Institute for Liver and Digestive Health and Sheila Sherlock Liver Centre, Royal Free Hospital and UCL, London, United Kingdom; 7Swiss Liver Center, UVCM, Inselspital, Department of Biomedical Research, University of Bern, Bern, Switzerland; 8Liver Unit, Regional Institute of Gastroenterology and Hepatology "Octavian Fodor," University of Medicine and Pharmacy "Iuliu Hatieganu," Cluj-Napoca, Romania; 9Division of Gastroenterology (Liver Unit), University of Alberta, Edmonton, Canada; 10Division of Internal Medicine II, Krankenhaus Barmherzige Brüder, Vienna, Austria; 11Liver Unit, Hospital Clinic, IDIBAPS, University of Barcelona, Barcelona, Spain.
Abstract
INTRODUCTION: We aimed to explore the prevalence of portal hypertension in the most common etiologies of patients with compensated advanced chronic liver disease (cACLD) and develop classification rules, based on liver stiffness measurement (LSM), that could be readily used to diagnose or exclude clinically significant portal hypertension (CSPH) in clinical practice. METHODS: This is an international cohort study including patients with paired LSM/hepatic venous pressure gradient (HVPG), LSM ≥10 kPa, and no previous decompensation. Portal hypertension was defined by an HVPG >5 mm Hg. A positive predictive value ≥90% was considered to validate LSM cutoffs for CSPH (HVPG ≥10 mm Hg), whereas a negative predictive value ≥90% ruled out CSPH. RESULTS: A total of 836 patients with hepatitis C (n = 358), nonalcoholic steatohepatitis (NASH, n = 248), alcohol use (n = 203), and hepatitis B (n = 27) were evaluated. Portal hypertension prevalence was >90% in all cACLD etiologies, except for patients with NASH (60.9%), being even lower in obese patients with NASH (53.3%); these lower prevalences of portal hypertension in patients with NASH were maintained across different strata of LSM values. LSM ≥25 kPa was the best cutoff to rule in CSPH in alcoholic liver disease, chronic hepatitis B, chronic hepatitis C, and nonobese patients with NASH, whereas in obese NASH patients, the positive predictive value was only 62.8%. A new model for patients with NASH (ANTICIPATE-NASH model) to predict CSPH considering body mass index, LSM, and platelet count was developed, and a nomogram was constructed. LSM ≤15 kPa plus platelets ≥150 × 10/L ruled out CSPH in most etiologies. DISCUSSION: Patients with cACLD of NASH etiology, especially obese patients with NASH, present lower prevalences of portal hypertension compared with other cACLD etiologies. LSM ≥25 kPa is sufficient to rule in CSPH in most etiologies, including nonobese patients with NASH, but not in obese patients with NASH.
INTRODUCTION: We aimed to explore the prevalence of portal hypertension in the most common etiologies of patients with compensated advanced chronic liver disease (cACLD) and develop classification rules, based on liver stiffness measurement (LSM), that could be readily used to diagnose or exclude clinically significant portal hypertension (CSPH) in clinical practice. METHODS: This is an international cohort study including patients with paired LSM/hepatic venous pressure gradient (HVPG), LSM ≥10 kPa, and no previous decompensation. Portal hypertension was defined by an HVPG >5 mm Hg. A positive predictive value ≥90% was considered to validate LSM cutoffs for CSPH (HVPG ≥10 mm Hg), whereas a negative predictive value ≥90% ruled out CSPH. RESULTS: A total of 836 patients with hepatitis C (n = 358), nonalcoholic steatohepatitis (NASH, n = 248), alcohol use (n = 203), and hepatitis B (n = 27) were evaluated. Portal hypertension prevalence was >90% in all cACLD etiologies, except for patients with NASH (60.9%), being even lower in obese patients with NASH (53.3%); these lower prevalences of portal hypertension in patients with NASH were maintained across different strata of LSM values. LSM ≥25 kPa was the best cutoff to rule in CSPH in alcoholic liver disease, chronic hepatitis B, chronic hepatitis C, and nonobese patients with NASH, whereas in obese NASH patients, the positive predictive value was only 62.8%. A new model for patients with NASH (ANTICIPATE-NASH model) to predict CSPH considering body mass index, LSM, and platelet count was developed, and a nomogram was constructed. LSM ≤15 kPa plus platelets ≥150 × 10/L ruled out CSPH in most etiologies. DISCUSSION: Patients with cACLD of NASH etiology, especially obese patients with NASH, present lower prevalences of portal hypertension compared with other cACLD etiologies. LSM ≥25 kPa is sufficient to rule in CSPH in most etiologies, including nonobese patients with NASH, but not in obese patients with NASH.