Performance of three clinical scores for steatosis and steatohepatitis and their interaction with metabolic syndrome in obese individuals

Abstract

BACKGROUND

Non-invasive clinical scores are widely used to detect hepatic steatosis and steatohepatitis, but their accuracy in individuals with obesity is limited. Most of these tools were developed for non-obese populations and do not account for metabolic dysfunction-associated steatotic liver disease (MASLD) spectrum. Moreover, the potential modifying effect of metabolic syndrome (MetS) on the diagnostic performance of these scores remains unclear. Given the global burden of obesity and MASLD, there is a pressing need to refine diagnostic strategies for early detection. We hypothesized that diagnostic performance may vary by MetS status and can be improved with adjusted thresholds.

AIM

To evaluate and optimize three clinical scores for steatosis and metabolic dysfunction-associated steatohepatitis (MASH), including assessment by MetS status.

METHODS

This cross-sectional study included 95 individuals undergoing bariatric surgery at a hospital in Brazil. Clinical scores [non-alcoholic fatty liver disease liver fat score (NLFS), hepatic steatosis index (HSI), and fatty liver index (FLI)] were calculated from preoperative data. Liver biopsy was used as the reference standard to assess steatosis and MASH. Diagnostic accuracy was evaluated using the area under the receiver operating characteristic curve, and optimal cut-offs were determined by Youden’s index. Logistic regression with interaction terms assessed whether MetS modified the diagnostic performance of each score across histological outcomes.

RESULTS

Sixty-six individuals (69.5%) had steatosis, and fifteen (15.8%) had moderate-to-severe steatosis and MASH. The area under the receiver operating characteristic curves for any steatosis was 0.676 (NLFS), 0.540 (HSI), and 0.468 (FLI); for moderate-to-severe steatosis, 0.671 (NLFS), 0.659 (HSI), and 0.700 (FLI); and for MASH, 0.671 (NLFS), 0.625 (HSI), and 0.639 (FLI). Standard cut-offs performed poorly; optimized thresholds improved both sensitivity and specificity. NLFS outperformed FLI for any steatosis (P = 0.021). No significant interactions were found between MetS and any score (all P > 0.05), indicating that diagnostic accuracy did not significantly differ by MetS status.

CONCLUSION

NLFS, HSI, and FLI show limited accuracy in obese individuals. Adjusting thresholds improves performance. Diagnostic utility remains consistent regardless of MetS, supporting their use across the MASLD spectrum.

Keywords: Metabolically-dysfunction-associated steatotic liver disease; Liver steatosis; Steatohepatitis; Obesity; Clinical scores

Core Tip: This study evaluated and optimized the diagnostic performance of three clinical scores - non-alcoholic fatty liver disease liver fat score, Hepatic Steatosis Index, and Fatty Liver Index - for detecting steatosis and steatohepatitis in obese individuals with metabolic dysfunction-associated steatotic liver disease. Using liver biopsy as a reference, we showed that standard thresholds performed poorly, but accuracy improved with optimized cut-offs. Metabolic syndrome status did not affect score performance significantly, suggesting that these tools can be applied across the metabolic dysfunction-associated steatotic liver disease metabolic spectrum. Our findings provide practical insights into improving non-invasive diagnosis in high-risk populations, where early detection is essential but challenging.