Screening for Metabolic Dysfunction-Associated Steatotic Liver Disease-Related Advanced Fibrosis in Diabetology: A Prospective Multicenter Study.

Summary

In 654 patients with MASLD and type 2 diabetes and/or obesity, noninvasive algorithms using FIB-4 followed by VCTE (or 2D-SWE/ELF) performed well for triaging advanced fibrosis. FIB-4/VCTE showed excellent diagnostic performance for referral; FIB-4/ELF at 9.8 offered high NPV but lower PPV, supporting practical screening pathways in diabetology clinics.

Key Findings

• Among 654 patients, 17.6% had intermediate/high risk and 9.3% had high risk of advanced fibrosis.
• Diagnostic AUCs for high-risk AF: FIB-4 0.78, FibroTest 0.78, FibroMeter 0.74, ELF 0.82, SWE 0.84.
• FIB-4/VCTE algorithms showed strong performance for referral decisions; FIB-4/ELF (threshold 9.8) achieved NPV 88–89% with PPV 39–46%; FIB-4/2D-SWE achieved NPV 91% with PPV 58–62%.
• Age-adapted FIB-4 thresholds reduced NPV and PPV across algorithms.

Clinical Implications

Diabetology clinics can adopt a two-step screening algorithm—FIB-4 followed by VCTE where available, or ELF/2D-SWE—to efficiently triage patients for hepatology referral, leveraging high NPV to rule out advanced fibrosis.

Limitations

• Interim analysis; not all participants had biopsy, and composite reference may introduce heterogeneity.
• Conducted in specialized settings; real-world performance in primary care or resource-limited clinics warrants evaluation.

Future Directions

Validate algorithm performance and thresholds across diverse populations and care settings, assess outcomes of referral strategies, and integrate with diabetes care pathways and EHR-based prompts.