[Evaluation of the efficacy of the Baveno VI criteria in diagnosing clinically significant portal hypertension under different etiologic control statuses].

Objective: To evaluate the efficacy of the Baveno VI criteria in the diagnosis of clinically significant portal hypertension (CSPH) under two conditions: non-etiological control and complete etiological control. Methods: A cross-sectional study was conducted. Clinical data were collected from 232 cases with compensated advanced chronic liver disease who underwent hepatic venous pressure gradient measurement and had liver stiffness measurement (LSM) ≥10 kPa at Nanfang Hospital, Southern Medical University, between July 2019 and December 2024. Patients were categorized into the etiological control group (78 cases) and the non-etiological control group (154 cases) according to the definition of etiological control. Patients meeting the criteria for etiological control or near-control, with alanine transaminase≤upper limit of normal (ULN) were further defined into the near-etiological control group (110 cases). Intergroup comparisons of normally distributed and skewed quantitative data were performed using the independent sample t-test and the Mann-Whitney U test, respectively. Categorical variables were compared using χ2 tests. The diagnostic performance of LSM in predicting CSPH under controlled and uncontrolled conditions was evaluated based on sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Results: The PPVs of LSM≥25 kPa for diagnosing CSPH in the overall population, the subgroups of the etiological-control, non-control, and nearly control were 82.8%, 93.3%, 79.6%, and 92.0%, respectively, with specificities of 89.3%, 96.3%, 86.8%, and 95.3%. The PPVs of LSM≥25 kPa for diagnosing CSPH were 90.0% and 76.9%, respectively, with the specificities of 93.8% and 85.0% in the non-control group with ALT≤ULN and ALT>ULN. LSM≥25 kPa demonstrated PPVs of 90.0% and 76.9% for diagnosing CSPH in populations with ALT≤ULN and ALT>ULN, respectively, with specificities of 93.8% and 85.0% in the non-etiological control group. The sensitivity and NPV for excluding CSPH using LSM≤15 kPa combined with platelet count (PLT)≥150×109/L in the etiological-control group, the non-control group, and the near-control group were 100% in the overall population. The combined model of LSM≥25 kPa and LSM≤15 kPa+PLT ≥150×109/L for diagnosing and excluding CSPH yielded gray zone rates of 60.3%, 67.9%, 56.5%, and 64.6% in the overall population and subgroups, respectively. The combined model of LSM≥25 kPa and LSM≤15 kPa+PLT≥110×109/L yielded gray zone rates of 50.8%, 59.0%, 46.8%, and 58.2% in the overall population and subgroups, respectively. Conclusion: The Baveno VI criteria reliably and effectively identify CSPH in patients with cACLD with well-controlled etiologies but remain to be validated in the non-controlled etiology group.