Purpose To investigate the clinical utility and longitudinal change of acoustic radiation force impulse (ARFI) ultrasonographic (US) imaging in patients with chronic hepatitis B. Materials and Methods A retrospective cohort study of patients with chronic hepatitis B who underwent serial ARFI examinations in a tertiary referral center in Taiwan between 2012 and 2016 was conducted. The clinical information and noninvasive liver stiffness measurement tests (ARFI, Fibrosis-4 index [FIB-4], and FibroScan) were collected. Serial ARFI measurements were compared in patients without antiviral therapy (nontreatment group) and during antiviral therapy (treatment group). The linear mixed-effects model with random coefficients was used to compare longitudinal repeated measurements. Results A total of 559 patients undergoing serial ARFI examinations were included. The ARFI value correlated with FIB-4 (correlation coefficient = 0.55, P < .001) and FibroScan (correlation coefficient = 0.69, P < .001) results. There were 314 patients with five or more ARFI measurements in more than 3 years of follow-up. The ARFI value remained unchanged in the nontreatment group (n = 189, from 1.11 to 1.11 m/sec, time trend P = .911). However, the ARFI value declined significantly in the treatment group (n = 125, from 1.63 to 1.37 m/sec, time trend P < .001), both in patients with cirrhosis (n = 51, from 2.15 to 1.75 m/sec, time trend P < .001) and in those without (n = 74, from 1.27 to 1.11 m/sec, time trend P < .001). Conclusion ARFI US imaging is an important clinical noninvasive test for liver stiffness measurement and can be used for serial measurements in the management of chronic hepatitis B. Antiviral therapy significantly reduces liver stiffness during longitudinal follow-up.

Background and Aims: Scarce data exist on the effect of nucleos(t)ide analogue (NA) discontinuation on hepatocellular carcinoma (HCC) risk in HBeAg-negative chronic hepatitis B (CHBe−). Therefore, we assessed whether HCC risk is increased in noncirrhotic CHBe− patients who discontinue compared to those remaining on NAs. Methods: This cohort study included 650 consecutive non-cirrhotic Caucasian or Asian patients with CHBe− without a history of HCC who discontinued NAs after a median of 5 or 3 years (cases, n = 325; Caucasians: 143, Asians: 182) or remained on NA therapy beyond 5 or 3 years respectively (controls, n = 325; Caucasians: 223, Asians: 102). Propensity score (PS) 1:1 matching was applied to adjust for patients' origin, age and sex.Results: During a median follow-up of 44 months, HCC developed in 7/325 cases and 9/325 controls or 7/245 PS-matched cases and 7/245 PS-matched controls with 5year cumulative HCC incidence of 5.1% and 4.9% respectively (log-rank, P = .836). No difference in 5-year HCC risk was observed between cases and controls of Caucasian (3.0% vs 4.8%; log-rank, P = .510) or Asian origin (1.3% vs 2.2%; log-rank, P = .873). In