A Simple Conventional Ultrasound-Based Risk Score for Predicting Progression to Cirrhosis in Patients With Compensated Early-Stage Liver Fibrosis

Background: Patients with compensated early-stage liver fibrosis (F1–F2) are at risk of progressing to cirrhosis, yet routine surveillance with liver biopsy or transient elastography is limited by invasiveness, cost, or accessibility. Simple, non-invasive tools are therefore needed for risk stratification. This study aims to develop a simple risk scoring system based on conventional ultrasound features and evaluate its ability to predict progression to cirrhosis within 3 years in patients with compensated early-stage liver fibrosis (F1–F2).Methods: This retrospective study included 502 patients with biopsy-proven compensated early-stage liver fibrosis. Four ultrasound features—liver surface, liver parenchymal echogenicity, liver margin morphology, and splenic volume (calculated as length × thickness × width × 0.52)—were semiquantitatively scored (0–2 points each) at baseline. The total score (0–8) was used to classify patients into low-risk (≤4) and high-risk (>4) groups. The primary endpoint was progression to compensated cirrhosis within 3 years. A Cox proportional hazards model was used to assess the association between the risk score and cirrhosis progression. Model discrimination was evaluated using the C-index and time-dependent ROC curves. Progression-free survival curves were plotted using the Kaplan–Meier method and compared by the log-rank test.Results: Among 502 patients, 374 (74.5%) were classified as low-risk (Risk Score ≤4) and 128 (25.5%) as high-risk (Risk Score >4). The 3-year cumulative cirrhosis progression rates were 8.6% and 35.2%, respectively (log-rank p < 0.0001). The risk score was an independent predictor (hazard ratio (HR) = 1.43, 95% confidence interval (CI) 1.29–1.59, p < 0.001), with a C-index of 0.716 and a 36-month time-dependent area under the receiver operating characteristic curve (AUC) of 0.734. Bootstrap internal validation yielded an optimism-corrected C-index of 0.716. The model showed good calibration (Brier score = 0.118). Decision curve analysis demonstrated positive net benefit of the score across threshold probabilities of 0.05–0.35. The score showed higher C-index (0.716 vs. 0.650, 0.631, and 0.660) and 36-month AUC (0.734 vs. 0.679, 0.672, and 0.675) compared to FIB-4, APRI, and LSM. Adding the score to liver stiffness measurement (LSM) significantly improved reclassification (net reclassification improvement (NRI) = 0.184, p = 0.007; integrated discrimination improvement (IDI) = 0.048, p = 0.025). Albumin was also an independent protective factor (HR = 0.93, 95% CI 0.87–0.99, p = 0.024) in the multivariable sensitivity model.Conclusion: This simple risk score (0–8) based on conventional ultrasound features (liver surface, parenchymal echogenicity, liver margin, and splenic volume) is an independent predictor of progression to cirrhosis within 3 years in patients with compensated early-stage liver fibrosis. This simple scoring system may provide a convenient and cost-effective non-invasive tool for risk stratification in clinical practice; however, these findings require validation in prospective, multicenter cohorts.