Baveno VI and Expanded Baveno VI criteria succesfully predicts the absence of high‐risk gastro‐oesophageal varices in a Chilean cohort

BACKGROUND: Baveno VI and expanded Baveno VI criteria have been recommended to circumvent the need for endoscopy screening in patients with a very low probability of varices needing treatment (VNT). AIM: To validate these criteria in a Latin American population. METHODS: The ability of Baveno VI criteria (liver stiffness measurement (LSM) 150 x 103/muL) and expanded Baveno VI criteria (LSM 110 x 103/muL) to exclude the presence of VNT was tested in a prospectively recruited cohort of patients with Child-Pugh A liver cirrhosis and with no previous variceal haemorrhage who attended the liver clinics of three major hospitals in Chile. RESULTS: Three hundred patients were included. The median (IQR) age was 61 [18-86] years, median MELD was 8.0 (6-17), median LSM was 17.2 (10.2-77) kPa and median platelet count was 137 (23-464) x 10(3) /muL. The main aetiology was non-alcoholic fatty liver disease (67.3%). VNT were present in 18% of patients. The Baveno VI criteria had a sensitivity of 98.1% and a specificity of 38.2%, potentially sparing 31.3% of upper endoscopies with a very low risk of missing VNT (1.1%). The expanded Baveno VI criteria had a sensitivity of 90.7% and a specificity of 61%, potentially sparing 51.3% of upper endoscopies with a risk of missing VNT of 3.6%. Both criteria were independently associated with the absence of VNT. CONCLUSION: We validated the Baveno VI and expanded Baveno VI criteria in Chilean population, potentially sparing 31.3% and 51.3% of endoscopies, respectively, with a very low risk of missing VNT. Fondecyt 1191183.