644: change in serial liver stiffness measurement by magnetic resonance elastography and outcomes in non-alcoholic fatty liver disease

The impact of disease progression in NAFLD on liver outcomes remains poorly understood. We aimed to investigate NAFLD progression using longitudinal liver stiffness measurements (LSM) by serial magnetic resonance elastography (MRE) and the association with liver outcomes.All adult patients with NAFLD who underwent at least two serial MREs for clinical evaluation at Mayo Clinic, Rochester, between 2007 and 2019 were identified from the institutional database. Progression and regression were defined based on LSM change of 19% above or below 19% of initial LSM, respectively, based on Quantitative Imaging Biomarker Alliance consensus. The association between change in LSM and liver-related outcomes occurring after the last MRE was examined using time-to-event analysis. A total of 128 participants underwent serial MREs (53% female, median age 59 years). The median time between paired MREs was 3.4 (range 1-10.7) years. NAFLD progression (LSM = +0.61 kPa/year) was identified in 17 patients (13.3%). NAFLD regression (-0.40 kPa/year) occurred in 35 patients (27.3%). Stable LSM was noted in 76 participants (59.4%). In NAFLD without cirrhosis at baseline (n = 75), cirrhosis development occurred in 14% of LSM progressors and 2.9% of non-progressors (p = 0.059) over a median 2.7 years of follow-up from the last MRE. Among those with compensated cirrhosis at baseline MRE (n = 29), decompensation or death occurred in 100% of LSM progressors and 19% of non-progressors (p < 0.001) over a median 2.5 years of follow-up after the last MRE.Noninvasive monitoring of LSM by conventional MRE is a promising method of longitudinal NAFLD monitoring and risk estimation of liver-related outcomes in NAFLD.