LIVERSTAT for the diagnosis of compensated advanced chronic liver disease in patients with type 2 diabetes

Document Type

Article

Publication Date

3-1-2026

Abstract

Introduction: LIVERSTAT is a non-invasive artificial intelligence-based test that provides risk stratification for metabolic dysfunction-associated steatotic liver disease. We aimed to study the performance of LIVERSTAT compared with the Fibrosis-4 Index (FIB-4) as a stand-alone test and as a first-line test to identify patients for liver stiffness measurement (LSM) for the diagnosis of compensated advanced chronic liver disease (cACLD) in patients with type 2 diabetes (T2D). Materials and Methods: This is a cross-sectional study of patients with T2D who underwent transient elastography. cACLD was defined as LSM ≥10 kPa. As a stand-alone test, LIVERSTAT Class D and increased FIB-4 ≥1.3 (≥2.0 if age ≥65 years old) were considered as having cACLD. As a first-line test, LIVERSTAT Class D and increased FIB-4 were considered as requiring LSM. Results: We analysed data for 221 patients (mean age 61 years, 41% male, cACLD 26%). The area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, negative predictive value and misclassification rate for LIVERSTAT were 0.66, 32%, 88%, 47%, 79% and 27%, respectively. The corresponding values for FIB-4 were 0.61, 39%, 81%, 41%, 79%, and 30%, respectively. When using LIVERSTAT as a first-line test, the proportion of patients requiring LSM was 17% (38/221), while the proportion of false negatives was 19% (34/183). The corresponding values for FIB-4 were 24% (54/221) and 19% (31/167), respectively. Conclusion: LIVERSTAT has similar accuracy as FIB-4 when used as a stand-alone test or as a first-line test to identify patients for LSM for the diagnosis of cACLD in patients with T2D.

Keywords

cACLD, fibrosis, Fibrosis-4 Index, MASLD, Metabolic dysfunction-associated steatotic liver disease

Publication Title

Medical Journal of Malaysia

ISSN

0300-5283

DOI

pubmed.ncbi.nlm.nih.gov/41914579/

Funders

University of Malaya Special Research Fund (BK006-2016, BKS067-2017)

Volume

81

Issue

2

First Page

209

Last Page

215

Publisher

Malaysian Medical Association

Share

COinS