A Combined Model of Spleen Stiffness and Baveno VII Criteria for Clinically Significant Portal Hypertension in Compensated Advanced Chronic Liver Disease: An International Multicenter Study

ABSTRACT Aims The renewing Baveno VII consensus proposed criteria for the diagnosis of clinically significant portal hypertension (CSPH) in patients with compensated advanced chronic liver disease (cACLD). The performance of a combined model of spleen stiffness measurement (SSM) by spleen‐dedicated 100 Hz (SSM@100 Hz) or conventional 50 Hz (SSM@50 Hz) and Baveno VII criteria to rule‐in or rule‐out CSPH had not been well validated. This study aims to compare the performance of the combined model with Baveno VII criteria alone to rule‐in and rule‐out CSPH in cACLD. Methods This international multicenter study included cACLD patients who underwent paired liver stiffness measurement (LSM), SSM@100 Hz or SSM@50 Hz, platelet count (PLT), and hepatic venous pressure gradient (HVPG). CSPH was defined as HVPG ≥ 10 mmHg. Patients with SSM@100 Hz were prospectively recruited from China between August 2021 and March 2022, while a global cohort of patients with SSM@50 Hz from Croatia, Japan, and Singapore was retrospectively included between December 2014 and June 2022. The diagnostic performance of different models was assessed using sensitivity, specificity, positive predictive value, and negative predictive value. Results A total of 206 patients with cACLD were recruited from seven university centers and 110 patients were included in the final analysis (54 from the SSM@100 Hz cohort and 56 from the SSM@50 Hz cohort). The success rate of SSM@100 Hz was significantly higher than that of SSM@50 Hz (103/105 [98.1%] vs. 86/101 [85.1%]; p < 0.001). While the combined model (SSM > 50 kPa or LSM ≥ 25 kPa) and Baveno VII criteria (LSM ≥ 25 kPa) had a positive predictive value and specificity > 90% to rule‐in CSPH, the combined model correctly ruled‐in more cACLD patients with CSPH compared to Baveno VII criteria alone (35/110 [31.8%] vs. 22/110 [20.0%]; p < 0.001). Furthermore, the combined model (SSM < 21 kPa or [LSM ≤ 15 kPa and PLT ≥ 150 × 10 9 /L]) and Baveno VII criteria (LSM ≤ 15 kPa and PLT ≥ 150 × 10 9 /L) had a sensitivity and negative predictive value > 90% to rule‐out CSPH. Compared to the Baveno VII criteria alone, the combined model correctly ruled‐out more patients without CSPH, although there was no statistical difference (39/110 [35.5%] vs. 34/110 [30.9%]; p = 0.063). The findings remained broadly similar when subgroup analyses were performed in the SSM@100 Hz cohort and the SSM@50 Hz cohort. Notably, the combined model reduced patients in the gray zone compared to Baveno VII criteria alone (36/110 [32.7%] vs. 54/110 [49.1%]; p < 0.001). Conclusions Whether using SSM@100Hz or SSM@50Hz, the combined model of SSM and Baveno VII criteria was superior to Baveno VII criteria alone to rule‐in and rule‐out CSPH in cACLD patients, which may guide therapeutic decisions by minimizing cACLD patients in the gray zone. Trial Registration ClinicalTrials.gov; No.NCT05251272.