Predicting response to non-selective beta-blockers with liver-spleen stiffness and heart rate in patients with liver cirrhosis and high-risk varices

Abstract

Introduction: Non-selective beta-blockers (NSBB) are used for primary prophylaxis in patients with liver cirrhosis and high-risk varices (HRVs). Assessing therapeutic response is challenging due to the invasive nature of hepatic venous pressure gradient (HVPG) measurement. This study aims to define a noninvasive machine-learning based approach to determine response to NSBB in patients with liver cirrhosis and HRVs.

Methods: We conducted a prospective study on a cohort of cirrhotic patients with documented HRVs receiving NSBB treatment. Patients were followed-up with clinical and elastography appointments at 3, 6, and 12 months after NSBB treatment initiation. NSBB response was defined as stationary or downstaging variceal grading at the 12-month esophagogastroduodenoscopy (EGD). In contrast, non-response was defined as upstaging variceal grading at the 12-month EGD or at least one variceal hemorrhage episode during the 12-month follow-up. We chose cut-off values for univariate and multivariate model with 100% specificity.

Results: According to least absolute shrinkage and selection operator (LASSO) regression, spleen stiffness (SS) and liver stiffness (LS) percentual decrease, along with changes in heart rate (HR) at 3 months were the most significant predictors of NSBB response. A decrease > 11.5% in SS, > 16.8% in LS, and > 25.3% in HR was associated with better prediction of clinical response to NSBB. SS percentual decrease showed the highest accuracy (86.4%) with high sensitivity (78.8%) when compared to LS and HR. The multivariate model incorporating SS, LS, and HR showed the highest discrimination and calibration metrics (AUROC = 0.96), with the optimal cut-off of 0.90 (sensitivity 94.2%, specificity 100%, PPV 95.7%, NPV 100%, accuracy 97.5%).

Keywords: Elastography; Hepatic venous pressure gradient (HVPG); High-risk varices; Liver cirrhosis; Liver stiffness (LS); Machine learning; Non-selective beta-blockers (NSBB); Primary prophylaxis; Spleen stiffness (SS); Variceal hemorrhage.

Fig. 1

Description of follow-up timeline from enrollment. Eligible patients underwent baseline clinical assessment, laboratory tests, and elastography measurements liver stiffness > 20 kPa and platelet count < 150.000 × 10⁹ cells/L underwent EGD for EVs screening within 10 days. Patients with HRVs were prescribed NSBBs, and re-evaluated at 3/6 months, and 1-year post-therapy initiation. a Flowchart reporting patients who completed follow-up in the derivation cohort, while b reports patients who completed follow-up in the validation cohort

Fig. 2

a Presents a 3D visualization of a logistic regression model, depicting the relationship between the predicted outcome and three input variables, offering a comprehensive view of the model’s decision boundary. b Displays the calibration plot comparing the expected versus observed risk for the model, evaluating how well the predicted probabilities align with actual outcomes

Fig. 3

After initial evaluation and detection of HRVs, patients should be treated with NSBB if they present no contraindications. According to our findings, before NSBB therapy initiation LS, SS, and HR should be registered and repeated after 3 months to determine NSBB response non-invasively. Responses can be either evaluated by considering the multivariate model (1) or by considering each variable singularly (2)