Effect of threshold growth as a major feature on LI-RADS categorization.

Abstract

Liver Imaging Reporting and Data System (LI-RADS) uses major features (arterial phase hyperenhancement [APHE], "washout" [WO], "capsule," diameter, threshold growth [TG]) to codify probability of hepatocellular carcinoma for each observation. This study assessed the effect of removing TG as a major feature on LI-RADS categorization. In this HIPAA-compliant, IRB-approved study, all MR and CT clinical reports containing a standardized LI-RADS v2014 template between 4/15-1/17 were retrospectively reviewed for each LR-3, LR-4, and LR-5 reported observation. Two LI-RADS categories were then assigned: one using all LI-RADS major features and one after removing TG as a major feature. The two categories were compared descriptively. The study included 265 patients (172 [65%] male, mean age 63 [±10] years) with 489 observations (median diameter 14mm, IQR 10-20mm), of which 345 (71%) had APHE, 307 (63%) had WO, 86 (18%) had "capsule," and 72 (15%) had TG. Of 86 observations with TG, 47 (65%) were new observations ≥10mm, 14 (19%) had diameter increase ≥50% in ≤6months, and 11 (15%) had diameter increase ≥100% in >6months. Using all major features, 214/489 (44%) observations were LR-3, 129/489 (26%) were LR-4, and 146/489 (30%) were LR-5. After removing TG, 237/489 (48%) were LR-3, 119/489 (24%) were LR-4, and 133 (27%) were LR-5. Removing TG caused a category downgrade for 35/489 (7%, 95% CI 5-10) observations, including 13/146 (9%, 95% CI 3-14) LR-5 observations. 9% of LR-5 observations would be downgraded without TG.