Discordance between vibration controlled-transient elastography and ultrasound in cirrhosis assessment: prognostic implications for liver-related events in patients with chronic hepatitis B.

Accurate cirrhosis assessment is crucial to guide treatment and hepatocellular carcinoma (HCC) surveillance in chronic liver disease (CLD). Ultrasonography (USG) and Vibration-Controlled Transient Elastography (VCTE) are widely used non-invasive tools, but discordance results are frequently observed. We evaluated the risk of liver-related events (LREs) based on concordant and discordant USG and VCTE diagnoses of cirrhosis. This retrospective study included 532 chronic hepatitis B patients who underwent VCTE between 2014 and 2021. Key exclusion criteria were unreliable VCTE results and prior hepatic decompensation. Cirrhosis was defined as LSM ≥ 10 kPa by VCTE and by radiologists' reports for USG. LREs were defined as the development of HCC or cirrhosis complications. We found that discordant results occurred in 13% of patients. Overall, 17 LREs (3.2%) were observed during a 4.4-year of median follow-up period. Patients were stratified into four groups: concordant non-cirrhosis (Group 1, n = 418; 3 LREs [0.7%]); cirrhosis by USG only (Group 2, n = 21; 0 LREs); cirrhosis by VCTE only (Group 3, n = 50; 3 LREs [6.0%]); and concordant cirrhosis (Group 4, n = 43; 11 LREs [25.6%]).The 5-year LRE-free survivals were 99.3%, 100%, 94%, and 72.1% in groups 1, 2, 3, and 4, respectively. Multivariable Cox regression analysis showed significantly increased risk for LREs only in Group 3 (aHR 8.54, p = 0.01) and Group 4 (aHR 34.9, p < 0.001). In summary, discordant in cirrhosis diagnosis is a common clinical challenge. While concordant cirrhosis carries the highest LRE risk, VCTE serves a crucial complementary role in discordant cases by identifying high-risk patients potentially overlooked by USG alone.