Adoption and validation of the European Association for the Study of the Liver algorithm for the noninvasive diagnosis of advanced fibrosis in metabolic dysfunction-associated steatotic liver disease in low-resource South Asian settings

Abstract

Introduction Patients with significant liver fibrosis (SF) are likely to progress to advanced chronic liver disease (ACLD). Therefore, liver-directed therapy is indicated. The European Association for the Study of the Liver-2024 (EASL) recommends annual screening of patients with diabetes for SF/ACLD using the FIB-4 score, followed by vibration-controlled transient elastography (VCTE) in patients with FIB-4 ≥ 1.3 in ≤65-year-olds and ≥2.0 in > 65-year-olds. Because VCTE is not freely available in resource-limited settings, we revised the EASL algorithm to prioritise referrals for VCTE in such settings and validated it in an external cohort. Methods We conducted a cross-sectional study of adults with type 2 diabetes (T2DM) and ultrasonographic evidence of steatotic liver disease (SDL) attending three outpatient medical clinics in the Gampaha District, Sri Lanka. FIB-4 scores were calculated, and those with scores ≥1.3 underwent VCTE. SF was defined as liver stiffness measure (LSM) ≥ 8.0 kPa. Factors significantly associated with SF/ACLD were identified using multiple logistic regression (LR). We then developed a new criterion for VCTE referral and compared the number of referrals for VCTE when using the new criterion-based algorithm with the EASL algorithm. We validated the new criterion-based algorithm in an external cohort of 372 patients with MASLD. Results We studied 363 patients, and 128 had an FIB-4 score of≥1.3. Of them,121 underwent VCTE, and 76 had an LSM ≥ 8.0 kPa. On multivariable LR, VCTE-diagnosed SF/ACLD was independently associated with diabetes of ≥ 5 years duration (OR 3.8, p = 0.035), micro/macrovascular complications (OR 19.4, p = 0.016), and BMI of ≥ 25 kg/m2(OR 6.2, p = 0.003). We revised the VCTE referral criterion as “patients having EASL FIB-4 criterion plus one or more of the three other factors: diabetes of ≥ 5 years duration, presence of micro/macrovascular complications or BMI≥25 kg/m2,”. The number of VCTE referrals indicated using the EASL algorithm was 96, compared to 83 with the new criterion, resulting in a 13.5% reduction. In the external validation cohort, the new algorithm reduced the number requiring VCTE referral by 15.5%. Conclusions Adopting the new criterion for VCTE referral in patients with MASLD appears more cost-effective for detecting SF/ACLD in low-resource settings in South Asia.