Comparable Diagnostic Accuracy but Discordant Thresholds Between FibroTouch and FibroScan: A Biopsy‐Based Comparative Study

Aims Accurate noninvasive fibrosis staging is essential in chronic liver disease (CLD). Although both FibroTouch and FibroScan are widely used, differences in technical design may affect measurement consistency. Whether their diagnostic thresholds for liver fibrosis are interchangeable remains unclear. The study aims to compare their diagnostic performance and threshold agreement in a biopsy‐confirmed CLD cohort. Methods This single‐center, cross‐sectional study included 339 patients with biopsy‐confirmed CLD from February 2014 to March 2019, all of whom underwent biochemical testing, FibroTouch, and FibroScan. Diagnostic performance was evaluated using the area under the receiver operating characteristic curves (AUROCs). Univariable and multivariable linear models were used to identify factors associated with inter‐device liver stiffness measurement (LSM) differences. Misclassification was calculated using manufacturer‐recommended thresholds. Subgroup analysis was conducted among patients with metabolic dysfunction–associated fatty liver disease (MAFLD). Results Both devices showed excellent diagnostic performance for detecting fibrosis ≥ F2, ≥ F3, and F4, with AUROCs exceeding 0.900, outperforming the fibrosis‐4 index (FIB‐4) and aspartate aminotransferase‐to‐platelet ratio index (APRI) in the entire cohort (FIB‐4 AUROCs: 0.775, 0.801, and 0.815; APRI AUROCs: 0.771, 0.754, and 0.749; all p < 0.001). FibroTouch required higher LSM thresholds than FibroScan at ≥ F2 (9.8 vs. 8.5 kPa) and ≥ F3 (13.1 vs 11.5 kPa). Cirrhosis ( β : −4.144, p = 0.001), Grade 4 hepatic inflammation ( β : −6.257, p = 0.001), and elevated alkaline phosphatase ( β : −0.004, p = 0.007) were independently associated with inter‐device discordance. Misclassification rates were 36.9% (125/339) for FibroTouch and 32.2% (109/339) for FibroScan, primarily due to overestimation. In the MAFLD subgroup ( n = 113), diagnostic accuracy remained high (AUROCs > 0.910), but threshold gaps widened further. Conclusion FibroTouch and FibroScan provide comparable diagnostic accuracy for fibrosis staging in CLD. However, they are not interchangeable, and the use of device‐specific thresholds is mandatory to reduce misclassification, especially in patients with MAFLD.