OBJECTIVETo evaluate the sensitivity and specificity of hepatic ultrasonography (US) for the diagnosis of hepatic steatosis in obese children, using ¹H magnetic resonance spectroscopy (¹H MRS) as the reference standard.

METHODSA total of 162 obese children with age of 10.5 ± 2.2 years and BMI of 28 ± 4 were enrolled in this study. They accepted hepatic US and (1)H MRS examinations. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of US were calculated for the overall presence of hepatic steatosis by comparison with ¹H MRS results.

RESULTSUsing quantitative criteria of liver fat content (LFC) >5% determined by (1)H MRS, 95 children(58.6%)were diagnosed as having hepatic steatosis. The sensitivity and specificity of US in diagnosing steatosis were 91.6% (87/95) and 50.7% (34/67) respectively, with PPV of 72.5% (87/120), and NPV of 81.0% (34/42). Considerable overlap in LFC measured by ¹H MRS was observed between different grades from US findings: absent (LFC interquartile range: 1.3%-3.9%), mild (2.4%-10.7%), moderate (7.1%-20.2%) and severe (7.6%-28.8%) steatosis.

CONCLUSIONSThe US can yield a high sensitivity and low specificity in the diagnosis of hepatic steatosis in obese children, suggesting it can be used as a screening tool for hepatic steatosis. To improve diagnostics, ¹H MRS is needed to determine LFC.