US Markers in the Noninvasive Assessment of Hepatic Steatosis in Metabolic Dysfunction-associated Steatotic Liver Disease.

Abstract

Background Attenuation coefficient (AC) is an established US marker for assessing patients with liver steatosis. However, improvements in diagnostic performance are warranted. Purpose To evaluate the diagnostic performance of AC, normalized local variance (NLV), hepatorenal ratio, and their combinations for liver histopathologic steatosis in participants with metabolic dysfunction-associated steatotic liver disease (MASLD). Materials and Methods This secondary analysis of the prospective Innovative Liver Elasticity, Attenuation, and Dispersion Ultrasound Study (iLEAD) (June 2019 to March 2023) included participants with biopsy-proven MASLD who underwent multiparametric liver US; AC, NLV, and hepatorenal ratio were measured. The diagnostic performance of each imaging marker and their combinations for determining steatosis grade were assessed using the area under the receiver operating characteristic curve (AUC) with the DeLong test, and reclassification improvement was also assessed using the category-free net reclassification improvement (NRI). Results A total of 102 participants (mean age, 52 years ± 16.1 [SD]; 51 male) were evaluated. AC showed the highest diagnostic performance for determining steatosis grade across all grades (AUC for grade ≥S1, 0.93 [95% CI: 0.87, 0.99]), compared with NLV (0.88 [95% CI: 0.77, 0.99]; P = .47) and hepatorenal ratio (0.85 [95% CI: 0.71, 0.98]; P = .30). Four models were constructed for identifying steatosis grades: model 1, AC only; model 2, AC plus NLV; model 3, AC plus hepatorenal ratio; model 4, AC plus NLV plus hepatorenal ratio. Models 2-4 showed no evidence of a difference in performance from model 1 for identifying grade S1 or higher (AUC, 0.93-0.98; P = .16, .15, and .07, respectively), but an improvement in risk reclassification was observed for model 4 (NRI, 1.18 [95% CI: 0.54, 1.77]; P < .001) compared with model 1 (reference). Performance was similar across models in identifying steatosis grade S2 or greater (AUC, 0.89-0.90; all P ≥ .47) and for grade S3 (AUC, 0.89-0.90; all P ≥ .72), but an improvement in risk reclassification was observed for model 3 compared with model 1 (≥S2 steatosis: NRI, 0.58 [95% CI: 0.20, 0.94] [P = .002]; S3 steatosis: NRI, 0.71 [95% CI: 0.26, 1.14] [P = .002]). Conclusion Liver US-derived AC performed better than NLV and hepatorenal ratio in determining steatosis grade in participants with MASLD; when these markers were combined, reclassification of steatosis grade improved. Clinical trial registration no. NCT04012242 © RSNA, 2026 Supplemental material is available for this article.