Use of FibroScan–AST Score to Stratify High-Risk Nonalcoholic Steatohepatitis in US Veterans

Abstract

US Veterans have a higher prevalence of nonalcoholic steatohepatitis (NASH) compared with the general US population. 1 Recently, vibration-controlled transient elastography (VCTE, FibroScan) proved valuable in evaluating hepatic steatosis and fibrosis in adolescents, and monitoring for liver-related events and survival in patients with nonalcoholic fatty liver disease (NAFLD)-related compensated advanced liver disease. 2,3 Validated noninvasive assessments are required as a surrogate to liver biopsy in stratifying patients with NASH, 4 particularly those at the highest risk for disease progression and liver-related outcomes (ie, NAFLD Activity Score [NAS] 4 and stage 2 fibrosis [F2]). 5 The FibroScan-AST (FAST) score combines VCTE liver stiffness measurement (LSM) and controlled attenuation parameter (CAP) with aspartate aminotransferase (AST) that effectively risk stratifies patients with progressive NASH. 6 The utility of this test in the Veterans population is unknown. We examined the diagnostic accuracy of FAST score in the real world clinical setting to stratify subjects with low-and highrisk NASH. We performed a cross-sectional study from a prospectively evaluated cohort of suspected NAFLD in a tertiary care Veterans Affairs facility from January 1, 2017 to May 30, 2019. Of 448 Veterans, 249 were excluded (coexistent other liver diseases ¼ 170, biopsy and VCTE >12 months apart ¼ 58, VCTE failure ¼ 19, and FAST score unavailable ¼ 2). Thus, 199 US Veterans had AST, LSM by VCTE, CAP, and liver histology on an average within 3 months (0-12 months). A trained nurse did VCTE to measure LSM and CAP. Demographic, clinical, laboratory, and histologic data were collected. The NAS score and fibrosis stage, and NASH were defined using NASH Clinical Research Network 5 and Fatty Liver Inhibition of Progression criteria 4 , respectively. The FAST score was derived as recently

Validation of the FibroScan‐aspartate aminotransferase score by vibration‐controlled transient and B‐mode ultrasound elastography

“…Oeda et al reported that the cutoff values of 0.35 and 0.66 were chosen for the M probe, and 0.32 and 0.63 were chosen for the XL probe to obtain sensitivity and specificity values higher than 90%. 7 Puri et al 17 also reported the diagnostic performance of the FAST score using the cut-off values originally reported by Newsome et al 6 The FAST score had a Cstatistic of 0.75 (95% CI 0.69-0.81) and Youden index of 0.44 (95% CI 0.31-0.54). Using a FAST score of ≤0.35 as a cut-off to rule out NASH with NAS ≥4 and ≥F2 had a sensitivity and NPV of 1.00, and a score of ≥0.67 as a cut-off to rule in NASH with NAS ≥4 and ≥F2 had a specificity of 0.69, with a PPV of 0.26.…”

“…reported that the cut‐off values of 0.35 and 0.66 were chosen for the M probe, and 0.32 and 0.63 were chosen for the XL probe to obtain sensitivity and specificity values higher than 90% 7 . Puri et al 17 . also reported the diagnostic performance of the FAST score using the cut‐off values originally reported by Newsome et al 6 .…”

“…Two additional studies reported on the accuracy of the FAST score. 7,17 Oeda et al previously reported the diagnostic accuracy of the FAST score by VCTE for distinguishing NASH with NAS ≥4 and fibrosis stage ≥2; 82 and 84 patients were included in the verification and validation sets, respectively. 7 The FAST score was calculated for both M and XL probes.…”

“…Very similarly, Puri et al reported that 35.1% of cases did not meet either cut-off criterion. 17 This is a challenge for the FAST score, as this indicates that a biopsy is required in approximately 30% of cases.…”

The FibroScan-aspartate aminotransferase score can stratify the disease severity in a Japanese cohort with fatty liver diseases

“…In the subset using the FAST score (n = 107), 16.8%, 34.6%, and 48.6% of the patients were classified as rule in NASH, gray zone, and rule out NASH, respectively 12 . Puri et al investigated 199 United States veterans; among them, 29%, 36.5%, and 36.5% were classified as rule in NASH, gray zone, and rule out NASH, respectively 13 . A potential reason for the differences between the results of our study and those of other studies is the higher proportion of patients who underwent liver biopsy and had more progressive liver disease, thereby resulting in a referral bias.…”