Diagnostic Accuracy of the Hepatic Steatosis Index (HSI)

Characteristics of Diagnostic Performance

The Hepatic Steatosis Index (HSI) has a sensitivity of 93.1% for excluding steatosis when <30 and a specificity of 92.4% for diagnosing steatosis when >36, with an area under the curve (AUC) of 0.81, indicating good to very good diagnostic performance 1
An HSI <30 allows for the exclusion of hepatic steatosis with a sensitivity of 93.1%, making it excellent for ruling out the disease when the value is low 1
An HSI >36 allows for the diagnosis of hepatic steatosis with a specificity of 92.4%, indicating high confidence when the value is elevated 1

The HSI is calculated as: 8 × (ALT/AST) + BMI [kg/m²], with additional adjustments of +2 for type 2 diabetes and +2 for female sex 1
The HSI shows comparable performance to other validated non-invasive indices, such as the Fatty Liver Index (FLI) with an AUC of 0.85, and the NAFLD Liver Fat Score (NLFS) with an AUC of 0.86-0.87 1

The HSI has demonstrated effective diagnostic performance in both Western and Asian populations, suggesting its transcultural applicability 1
The HSI has a gray diagnostic zone between 30 and 36, where additional diagnostic methods are required 1
The presence of inflammation or hepatic fibrosis can affect the accuracy of the index, although to a lesser extent than conventional ultrasonography 1
The HSI does not discriminate between simple steatosis and steatohepatitis, nor does it evaluate the degree of fibrosis, so patients with high values require additional risk stratification 1, 2

The HSI is particularly useful as an initial screening tool in primary care to identify patients who require further evaluation with imaging methods or specialized hepatological assessment 1
For patients with HSI >36, evaluation of fibrosis risk using FIB-4 or other non-invasive methods is recommended to determine the need for referral to hepatology 2