[2026-07-23] A Simple Clinical Prediction Score for Significant Liver Fibrosis Using VCTE as the Reference Standard in Thai Patients With MASLD

Abstract

Background: Clinically significant liver fibrosis is the major determinant of adverse liver-related outcomes in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). Vibration-controlled transient elastography (VCTE) with liver stiffness measurement (LSM) is the standard tool for noninvasive assessment of hepatic fibrosis; however, its high cost and limited availability restrict its widespread use in Thai hospitals. A simple, accessible clinical tool may improve early detection and risk stratification.

Objectives: To develop and evaluate a simple clinical prediction score for identifying significant hepatic fibrosis (LSM-defined F2-F4) among Thai patients with MASLD obtained by VCTE as the reference standard.

Methods: This retrospective study analyzed 189 Thai MASLD patients who underwent VCTE at Maharat Nakhon Ratchasima Hospital, Thailand, between April 2023 and March 2024. Clinically significant hepatic fibrosis was defined as LSM stage F2-F4, while F0-F1 was considered no or mild fibrosis. Demographic, clinical, and laboratory parameters were evaluated, and independent predictors of steatosis were identified using multivariable logistic regression. The b coefficients were converted into weighted scores to create the hepatic fibrosis prediction score. Model performance was assessed by discrimination, calibration, and internal validation using 1000 bootstrap resamples.

Results: The mean (SD) controlled attenuation parameter was 257.5 (48.7) dB/m, and 96 patients (50.8%) had clinically significant hepatic fibrosis (F2-F4). Significant predictors included older age, body mass index, diabetes, and aspartate aminotransferase (AST) higher than 40 U/L. The final score incorporated these variables and demonstrated acceptable discrimination (area under the receiver operating characteristic curve [AUC], 0.743; 95% CI, 0.674-0.811) and an optimism-corrected AUC of 0.719 (95% CI, 0.651-0.786). Calibration was satisfactory (Hosmer-Lemeshow χ² = 6.66, P = .465). Patients classified as high risk (> 4.0 points) had an 89.5% prevalence of clinically significant fibrosis and a positive likelihood ratio of 8.42.

Conclusions: The hepatic fibrosis score is a simple, inexpensive, and internally validated clinical prediction tool for identifying patients at high risk of clinically significant hepatic fibrosis in Thai patients with MASLD. Although it is not intended to replace VCTE, the score may facilitate initial risk stratification and help prioritize patients for further fibrosis assessment, particularly in primary care and resource-limited settings. External validation is warranted before routine clinical implementation.