Abstract Background: Nonalcoholic fatty liver disease is a broad spectrum of heterogeneous disorders, and clinically picked up cases represent just the tip of the iceberg. The earliest, most benign, most prevalent, and reversible stage fatty liver or “steatosis” is the most inconspicuous and hidden of all stages. Objectives: The objective of this study was to test the diagnostic accuracy of hepatic steatosis index (HSI), Framingham steatosis index (FSI), and lipid accumulation product (LAP) for mass screening by externally validating them against ultrasonographically (USG) detected hepatic steatosis among adult population of Southeast Rajasthan. Materials and Methods: The three noninvasive tools were compared against USG-detected hepatic steatosis in the asymptomatic adult population of Kota city. Cutoff values, sensitivity, specificity, positive predictive value, negative predictive value (NPV), and accuracy were determined using receiver operating curve (ROC) analysis. Results: The prevalence of hepatic steatosis in the present study was found to be 23.7% on USG. The cutoff value for HSI was 35.79, with a sensitivity, specificity, and accuracy of 74.70%, 81.27%, and 79.71%, respectively. The cutoff value for FSI was 1.788, with a sensitivity, specificity, and accuracy of 66.27%, 82.40%, and 78.57%, respectively. The cutoff value for LAP score was 24.219, with a sensitivity, specificity, and accuracy of 67.47%, 74.91%, and 73.14%, respectively. Conclusion: Although, the differences for the Area under ROC (AUROC) were not statistically significant for any pair, indicating that all three indices are suited for screening. However, FSI performed better than both HSI and LAP in terms of AUROC, specificity and overall accuracy.
Gupta et al. (Thu,) studied this question.