Non-invasive tools are increasingly used to stratify fibrosis in chronic hepatitis B (CHB). Liver stiffness measurement (LSM) by transient elastography (FibroScan) is widely adopted, while blood-based scores such as FIB-4 and APRI are simple and inexpensive. We aimed to evaluate the diagnostic performance of FIB-4 and APRI for the detection of advanced fibrosis in CHB, using FibroScan as the reference.

We retrospectively included CHB patients who underwent FibroScan examination at F.Hached University Hospital (Sousse, Tunisia). Demographic, clinical and laboratory data were collected. Fibrosis stages were classified as F0–F1, F2, F3 and F4 based on LSM; advanced fibrosis was defined as F3–F4. FIB-4 and APRI scores were calculated from routine blood tests. Receiver operating characteristic (ROC) curves were constructed to assess the ability of FIB-4 and APRI to detect advanced fibrosis. Area under the ROC curve (AUC), optimal cut-offs , and diagnostic performance at classical thresholds were reported.

Eighty CHB patients were analyzed; advanced fibrosis (F3–F4) was present in 22/80 (27.5%) and absent in 58/80 (72.5%). Mean FIB-4 and APRI values were significantly higher in patients with advanced fibrosis compared with those without (FIB-4: 4.86 vs 1.88, APRI: 1.50 vs 0.46).

For the detection of advanced fibrosis, AUC was 0.91 for FIB-4 and 0.94 for APRI. The optimal cut-off for FIB-4 was 2.57, yielding a sensitivity of 100% and a specificity of 85.5%. The optimal APRI cut-off was 0.65, also providing a sensitivity of 100% and a specificity of 85.5%.

At classical thresholds, FIB-4 ≥1.45 showed 100% sensitivity and 60.0% specificity (PPV 50.0%, NPV 100%) for advanced fibrosis, whereas FIB-4 ≥3.25 provided 59.1% sensitivity and 85.5% specificity (PPV 61.9%, NPV 83.9%). An APRI ≥1.0 yielded 59.1% sensitivity, 90.9% specificity (PPV 72.2%, NPV 84.7%).

Both FIB-4 and APRI demonstrated excellent diagnostic performance for the detection of advanced fibrosis, with AUCs above 0.90 and high accuracy at optimized cut-offs. Classical rule-in thresholds (FIB-4 ≥3.25 and APRI ≥1.0) provided good specificity, whereas lower cut-offs offered very high sensitivity and NPV. These findings support the use of FIB-4 and APRI as simple triage tools to identify CHB patients who warrant confirmatory assessment by transient elastography.