Combined analysis from two large studies verified that the noninvasive tools Fibrosis-4 index, aspartate aminotransferase-to-platelet ratio, and NAFLD fibrosis score were able to detect advanced fibrosis and fibrosis progression in patients with nonalcoholic fatty liver disease.
“Although major international consortia are currently engaged in efforts to qualify noninvasive measures of disease assessment in NAFLD, they are still years away from fruition,” Mohammad S. Siddiqui, MD, from the Virginia Commonwealth University, and colleagues wrote. “There is therefore an unmet need to further refine existing clinical tools to assess disease stage and its progression.”
While several previous studies have established the general relationship between these noninvasive tools and fibrosis stage, Siddiqui and colleagues analyzed their diagnostic performance with a retrospective analysis that included the full spectrum of individuals with NAFLD including varying age, BMI, race and type 2 diabetes.
The study comprised 1,904 patients with NAFLD and fibrosis, of whom 58% had nonalcoholic steatohepatitis, 20% had stage 3 fibrosis, and 8% had stage 4 fibrosis.
With a specificity fixed at 90%, the positive predictive values (PPV) for detecting any fibrosis was higher than 90% in FIB-4 (92%), APRI (92%), and NFS (91%), although the negative predictive values (NPV) were low.
For detecting the presence of advanced fibrosis at a specificity of 90%, FIB-4 had a 65% PPV and 81% NPV with a cut-off of 1.95 (95% CI, 0.78-0.82), APRI had a 61% PPV and 79% NPV with a cut-off of 1.53 (95% CI, 0.74-0.79), and NFS had a 63% PPV and 90% NPV (95% CI, 0.76-0.8).
FIB-4 (C-statistic = 0.81), APRI (C-statistic = 0.82), and NFS (C-statistic = 0.8) were also able to detect the progression to advanced fibrosis among patients. Multivariate analysis showed that 1-unit change in model score correlated with change in fibrosis by 0.26 (95% CI, 0.15-0.37) using FIB-4, by 0.33 (95% CI, 0.2-0.45) using APRI, and by 0.19 (95% CI, 0.07-0.31) using NFS.
“Our findings confirm published literature showing limited sensitivity of these models,” Siddiqui and colleagues wrote. “However, the NPV for these models is high and thus may be used clinically to exclude the presence of moderate and advanced fibrosis to identify patients who can be safely spared confirmatory yet invasive testing.” – by Talitha Bennett
Disclosure: The authors report no relevant financial disclosures.