Never change a winning team—liver stiffness measurement and platelet count as key surrogates for clinically significant portal hypertension

Carvedilol is recommended in patients with compensated advanced chronic liver disease (cACLD) and clinically significant portal hypertension (CSPH) to prevent hepatic decompensation (1). However, as hepatic venous pressure gradient (HVPG) measurement, the minimally invasive gold standard for the diagnosis of CSPH, is not broadly available, decision rules based on non-invasive tests (NITs) are needed (2). As such, the Baveno VII criteria can be used to rule in [liver stiffness measurement (LSM) ≥25 kPa] or rule out CSPH (LSM ≤15 kPa & platelet count ≥150 G/L) (3) and have been shown to predict first hepatic decompensation (4). However, up to ~50% of patients remain in the diagnostic grey zone, and additional tools such as the ratio between von Willebrand Factor and platelet count [i.e., von Willebrand Factor antigen/thrombocyte ratio (VITRO) score] or spleen stiffness measurement (SSM) are needed (3,5-7). At the same time, LSM and platelet count can be combined in the so-called ANTICIPATE model and its derivatives to estimate the probability of CSPH, aiding the decision on whether or not to start carvedilol (e.g., if CSPH-probability is ≥60%) (8-10). Although evidence from randomized controlled trials (RCTs) supporting the concept that NIT-guided treatment with a non-selective beta-blocker (NSBB) improves outcomes is lacking, several studies have shown that subjects with a high non-invasively estimated probability of CSPH are at considerable risk of clinical complications (8,9), with a discrimination that is comparable to HVPG (9).

Led by the Liver Health Consortium in China (CHESS) group, a recent study extended our knowledge on non-invasive risk stratification for CSPH in patients with cACLD (11). Specifically, the authors derived a new model based on LSM and platelet count, weighted according to their regression coefficients as obtained from a meta-analysis. They subsequently demonstrated that the model could distinguish between the presence or absence of CSPH more accurately than the ANTICIPATE model and predict hepatic decompensation in several retrospective cohorts. We would like to bring up the following considerations:

First and foremost, the authors argue for a new model, as the current simplified decision rules (i.e., assuming a ≥60% probability of CSPH when LSM is between 15–20 kPa and platelet count <110 G/L, or when LSM is 20–25 kPa and platelet count <150 G/L) may have limited granularity for different combinations of these two NITs. However, we want to stress that the ANTICIPATE models do indeed provide granular probabilities at all different combinations, and can therefore not only be used to estimate the current probability of CSPH, but potentially also to monitor changes over time (for the formula, see respective publications).

Second, the current study validates the combined use of LSM and platelet count as key parameters to judge the presence of CSPH. However, it is unclear to what extent the new model improves current clinical practice, as its accuracy to predict hepatic decompensation was identical to the ANTICIPATE model, and subsequent risk groups were not compared to the ANTICIPATE model or other approaches using additional NITs (3,6,7). Also, calibration and clinical utility are as important as discrimination for clinical prediction models, of which the latter two aspects have not yet been evaluated, and thus, preclude clinical application at this point (12). To base decisions on models predicting the probability of CSPH, the estimated probability must be close to the observed (i.e., real) probability, as reflected by adequate calibration. Otherwise, this may result in significant over- or undertreatment. Here, studies from the Baveno Cooperation have shown that while the ANTICIPATE model provides adequate discrimination in several settings, including etiological cure and contemporary cohorts with predominant steatotic liver disease (SLD) patients, calibration may be improved by updated models and coefficients in the latter settings (7,13,14).

In line, while the new model reduced the grey zone (from 50% to 23%), 42% of the patients (vs. 19% for ANTICIPATE) were allocated to the high-risk group (increasing to 55–57% in cohorts evaluating hepatic decompensation) (11). Would all of these patients be subjected to NSBB treatment? Decision curve analyses and validation in contemporary cohorts focusing on patients with SLD and effectively treated for chronic hepatitis B/C will help to understand the additional benefits derived from this model.

Third, what this study also clearly demonstrates is the large diagnostic grey zone of the current Baveno VII criteria. This significantly impairs the reliance on NITs and leaves around half of the patients in need of additional tools (6,10). Here, several approaches have been discussed (2,3): First, an additional test can be introduced, such as the VITRO score or SSM. Specifically, a VITRO score <1.5 or an SSM <21–25 kPa (using the 50 or 100 Hz probe of vibration-controlled transient elastography) rules out CSPH. At the same time, a VITRO score ≥2.5 or SSM >50–55 kPa rules in CSPH, and these patients should be treated with carvedilol. Importantly, the decompensation risk in the remaining grey zone is negligible (6), supporting a watch-and-wait strategy with a reevaluation of NIT after another year. Second, the CSPH probability based on the ANTICIPATE models can be calculated: If the CSPH probability is ≥60% [recently studied for the combinations of LSM ≥20 kPa and platelet count <150 G/L (10)], CSPH can be assumed (CSPH rule-in), and these patients may be treated with carvedilol. In the remaining patients, NSBB indication for other reasons should be evaluated; moreover, endoscopy or HVPG measurement may be performed to identify CSPH or varices and justify the use of NSBB (10).

In summary, the paper by Liu et al. (11) confirms that models based on LSM and platelet count provide accurate estimates of the probability of CSPH and can guide the decision on treatment with carvedilol. Nevertheless, additional NITs (VITRO, SSM) or a more granular estimation of CSPH probability still seem to be required to further reduce the grey zone. Whether the model by Liu et al. (11) will be added to the armamentarium of NIT for PH that are ready for clinical application remains to be determined by the emergence of (external) validation. At the same time, it becomes increasingly evident that NITology needs to shed ballast with Baveno VIII in 2026, given the increasing number of tools that may overcomplicate the diagnosis of CSPH.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, HepatoBiliary Surgery and Nutrition. The article has undergone external peer review.

Peer Review File: Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-463/prf

Funding: The work was supported by the Clinical Research Group MOTION, Medical University of Vienna, Vienna, Austria—a project funded by the Clinical Research Groups Programme of the Ludwig Boltzmann Gesellschaft (grant No. LBG_KFG_22_32) with funds from the Fonds Zukunft Österreich (to G.S., M.J., and M.M.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-2025-463/coif). G.S. received travel support from Amgen. M.J. has received grant support from Gilead Sciences Inc. and has served as a speaker and/or consultant for Gilead Sciences Inc. and Echosens. M.M. received grant support from Echosens, served as a consultant and/or advisory board member and/or speaker for AbbVie, AstraZeneca, Echosens, Eli Lilly, Gilead, Ipsen, Takeda, and W. L. Gore & Associates, and received travel support from AbbVie and Gilead. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Mandorfer M, Aigner E, Cejna M, et al. Austrian consensus on the diagnosis and management of portal hypertension in advanced chronic liver disease (Billroth IV). Wien Klin Wochenschr 2023;135:493-523. [Crossref] [PubMed]
2. Mandorfer M, Abraldes JG, Berzigotti A. Non-invasive assessment of portal hypertension: Liver stiffness and beyond. JHEP Rep 2025;7:101300. [Crossref] [PubMed]
3. Semmler G, Jachs M, Mandorfer M. Non-invasive tests-based risk stratification: Baveno VII and beyond. Clin Mol Hepatol 2023;29:105-9. [Crossref] [PubMed]
4. Song BG, Goh MJ, Kang W, et al. Validation of non-invasive diagnosis of CSPH in patients with compensated advanced chronic liver disease according to Baveno VII. Liver Int 2023;43:1966-74. [Crossref] [PubMed]
5. Semmler G, Hartl L, Mendoza YP, et al. Simple blood tests to diagnose compensated advanced chronic liver disease and stratify the risk of clinically significant portal hypertension. Hepatology 2024;80:887-900. [Crossref] [PubMed]
6. Jachs M, Hartl L, Simbrunner B, et al. The Sequential Application of Baveno VII Criteria and VITRO Score Improves Diagnosis of Clinically Significant Portal Hypertension. Clin Gastroenterol Hepatol 2023;21:1854-63.e10. [Crossref] [PubMed]
7. Jachs M, Odriozola A, Turon F, et al. Spleen stiffness measurement by vibration-controlled transient elastography at 100 Hz for non-invasive predicted diagnosis of clinically significant portal hypertension in patients with compensated advanced chronic liver disease: a modelling study. Lancet Gastroenterol Hepatol 2024;9:1111-20. [Crossref] [PubMed]
8. Pons M, Rivera-Esteban J, Ma MM, et al. Point-of-Care Noninvasive Prediction of Liver-Related Events in Patients With Nonalcoholic Fatty Liver Disease. Clin Gastroenterol Hepatol 2024;22:1637-45.e9. [Crossref] [PubMed]
9. Jachs M, Hartl L, Simbrunner B, et al. Prognostic performance of non-invasive tests for portal hypertension is comparable to that of hepatic venous pressure gradient. J Hepatol 2024;80:744-52. [Crossref] [PubMed]
10. Dajti E, Villanueva C, Berzigotti A, et al. Exploring algorithms to select candidates for non-selective beta-blockers in cirrhosis: A post hoc analysis of the PREDESCI trial. J Hepatol 2025;82:490-8. [Crossref] [PubMed]
11. Liu C, You H, Zeng QL, et al. Carvedilol to prevent hepatic decompensation of cirrhosis in patients with clinically significant portal hypertension stratified by new non-invasive model (CHESS2306). Clin Mol Hepatol 2025;31:105-18. [Crossref] [PubMed]
12. Strandberg R, Jepsen P, Hagström H. Developing and validating clinical prediction models in hepatology - An overview for clinicians. J Hepatol 2024;81:149-62. [Crossref] [PubMed]
13. Semmler G, Lens S, Meyer EL, et al. Non-invasive tests for clinically significant portal hypertension after HCV cure. J Hepatol 2022;77:1573-85. [Crossref] [PubMed]
14. Pons M, Augustin S, Scheiner B, et al. Noninvasive Diagnosis of Portal Hypertension in Patients With Compensated Advanced Chronic Liver Disease. Am J Gastroenterol 2021;116:723-32. [Crossref] [PubMed]