Baveno VII Endoscopy-Sparing Criteria in MASLD- Predominant Compensated Cirrhosis: A Prospective Validation from a Gulf Arab Population (Kuwait)

Abstract Background & Aims: The Baveno VII non-invasive criteria can safely defer screening endoscopy in compensated advanced chronic liver disease, but their performance in metabolic dysfunction-associated steatotic liver disease (MASLD), particularly with obesity, and in Gulf Arab populations is incompletely characterised. We validated the Baveno VII endoscopy-sparing criteria in a Kuwaiti, MASLD-predominant compensated cirrhosis cohort, with pre-specified disaggregation of MASLD by obesity status and exploratory validation of the ANTICIPATE-NASH model against clinical CSPH surrogates. Methods: Consecutive Kuwaiti adults with compensated cirrhosis underwent concurrent vibration-controlled transient elastography, platelet count, and reference-standard upper gastrointestinal endoscopy at Al Adan Hospital between January 2022 and April 2026. Baveno VI, Expanded Baveno VI, Baveno VII, and ANTICIPATE-NASH criteria were applied, with high-risk varices (HRV) as the primary endpoint. Aetiologies were classified per the 2023 multi-society MASLD nomenclature. Pre-specified subgroups compared MASLD with and without obesity (BMI ≥30 kg/m²). Wilson confidence intervals, DeLong's method, multivariable logistic regression, and Hosmer–Lemeshow calibration were applied. Reporting followed STARD 2015, STROBE, and TRIPOD. Results: Of 451 patients screened, 380 were analysed (mean age 53.4 years; 60.0% male; mean BMI 33.1 kg/m²; MASLD 66.3%; HRV prevalence 20.0%). The Baveno VII rule-out criterion (LSM <20 kPa AND platelets >150 × 10⁹/L) was satisfied by 37.6%, with miss rate 1.4% (95% CI 0.2–5.0%), NPV 98.6%, and safe performance across all subgroups. Expanded Baveno VI exceeded the 5% safety threshold in MASLD with obesity (7.1%, 95% CI 3.3–14.6%). ANTICIPATE-NASH ≥75% improved positive predictive value for CSPH by 16.8 points over LSM ≥25 kPa alone in MASLD with obesity (84.1% versus 67.3%; p<0.001; H-L p=0.31). On multivariable analysis, LSM, platelet count, albumin, and MELD were independent predictors of HRV; BMI, type 2 diabetes, and MASLD aetiology were non-significant after adjustment for LSM. Conclusions: The Baveno VII rule-out criterion is safe in this cohort. Expanded Baveno VI should not be applied in MASLD with obesity. ANTICIPATE-NASH generalises to a Gulf Arab population without recalibration and reduces overtreatment when used as the rule-in tool. These findings support a three-tier non-invasive endoscopy-sparing framework for Gulf Arab compensated cirrhosis.