DIAGNOSTIC ACCURACY OF ULTRASOUND FOR GRADING METABOLIC DYSFUNCTION-ASSOCIATED STEATOTIC LIVER DISEASE AND ITS ASSOCIATION WITH BODY MASS INDEX: A SYSTEMATIC REVIEW
Keywords:
Non-alcoholic fatty liver disease; ultrasonography; hepatic steatosis; diagnostic accuracy; body mass index; sensitivity and specificityAbstract
Background: NAFLD (also known as MASLD) is a chronic disease of the liver that is associated with obesity and metabolic syndrome. Ultrasound is frequently used to identify and stage these types of liver diseases using non-invasive methods.
Objective: To evaluate the accuracy of ultrasound in detecting different grades of NAFLD based on BMI.
Methods: All known publications evaluating ultrasound imaging for the evaluation and classification of nonalcoholic fatty liver disease (NAFLD) were systematically reviewed to determine diagnostic accuracy (sensitivity and specificity) and NAFLD incidence/prevalence in relation to BMI.
Results: A total of 32 studies were evaluated in this systematic review. Sensitivity for identifying liver fat via ultrasound was between 68% and 94%, while specificity was between 70% and 95%. The incidence/prevalence of NAFLD in the reviewed studies, regardless of population characteristics, was between 25% and 72%, with higher incidence/prevalence rates observed in populations with higher BMI. In addition, there is an established correlation between BMI and the ultrasound classification of fatty infiltration (increased BMI is associated with more severe classifications of fatty infiltration). Also, quantitative ultrasound methods appear to outperform conventional ultrasound methods for diagnosing moderate to severe degrees of fatty infiltration.
Conclusion: Ultrasound is a dependable imaging technique that is commonly used to classify NAFLD, with strong diagnostic capabilities specifically among patients with elevated BMI or weight. A lack of standardization in the grading of NAFLD, as well as poor sensitivity within the diagnosis of mild NAFLD, highlight the need for complete standardization amongst diagnostic protocols.


