943 A Step-Wise Approach to Non-Invasive Diagnosis of Advanced Fibrotic Disease in NAFLD

Abstract

INTRODUCTION: The accurate diagnosis of advanced fibrosis using non-invasive methods remains a challenge in patients with nonalcoholic fatty liver disease (NAFLD). Recently, improved accuracy has been suggested with a new approach which uses a two-step model to diagnose advance fibrosis; first classifying patients based on NAFLD Fibrosis Score (NFS) and then, based on the Liver Stiffness Measurements (LSM) as measured by the Fibroscan®. Our study examines the usefulness of this approach in reducing the number of liver biopsies needed in NAFLD patients. METHODS: Patients were included if they had Fibroscan® and liver biopsy done within 6 months of each other during 2017 and 2018 at a large tertiary center. The NFS, LSM and Fibrosis (F) score on biopsy were assessed. LSM cut offs of 10 and 15 kPa were derived from the Baveno IV consensus. First, patients were divided into three groups based on NFS score: low risk (< -1.455), intermediate (-1.455-0.675) and high risk (>0.675). A Fibroscan® was then performed in each group. Patients with an NFS >-1.455, LSM > 10kPA and/or a combination of both were thought to qualify for a liver biopsy. We then compared the accuracy of using NFS alone, LSM alone or a Step-Wise Approach (SWA) to the biopsy result. Descriptive statistics analyzed the data. RESULTS: 98 patients met our inclusion criteria and underwent analysis. By being classified as intermediate or high risk using NFS alone, LSM alone, or SWA; 64, 64, and 48 patients qualified for liver biopsies, respectively. For low risk categories, using NFS, LSM and SWA, we classified 34, 34 and 50 patients respectively as having no significant fibrosis, 3,0,3 of them were misclassified respectively. The sensitivity, specificity, positive predictive value and negative predictive value for SWA were 88%, 64.4%, 45.8%, and 94%. SWA had the highest diagnostic accuracy. CONCLUSION: SWA has a high specificity and negative predictive value and accurately prevented 16 biopsies. We propose that SWA is beneficial in clinical practice to decrease the number of liver biopsies needed for NAFLD patients as shown in Figure 1. Furthermore, our results show that patients with NFS < -1.455 gain little benefit from subsequent LSM to rule out advanced fibrosis. Conversely, patients with high risk NFS and high risk LSM may have the biopsy deferred and be treated as compensated cirrhosis unless indicated otherwise.