Proton density fat-fraction is an accurate biomarker of hepatic steatosis in adolescent girls and young women.

Abstract

To compare complex quantitative magnetic resonance imaging (MRI) with MR spectroscopy (MRS) for quantification of hepatic steatosis (HS) and determine clinically significant MRI-based thresholds of HS in female youths. This prospective, cross-sectional study was conducted in 132 healthy females (11-22 years, mean 13.3 ± 2). Proton density fat-fraction (PDFF) was measured using complex quantitative MRI and MRS. Body mass index (BMI), fasting labs [glucose, insulin, alanine aminotransferase (ALT), and other metabolic markers] were obtained. Outcomes were measured using regression analysis, Spearman-rank correlation, and receiver operator characteristics (ROC) analysis. HS was defined as MRI-PDFF >5.6%. HS was detected by MRI-PDFF in 15% of all subjects. Linear regression demonstrated excellent correlation and agreement [r(2) = 0.96, slope = 0.97 (95 %CI: 0.94-1.00), intercept = 0.78% (95 %CI: 0.58-0.98%)] between MRI-PDFF and MRS-PDFF. MRI-PDFF had a sensitivity of 100% (95 %CI: 0.79-1.00), specificity of 96.6% (95 %CI: 0.91-0.99), and a kappa index of 87% (95 %CI: 0.75-0.99) for identifying HS. In overweight subjects with HS, MRI-PDFF correlated with ALT (r = 0.84, p < 0.0001) and insulin (r = 0.833, p < 0.001), but not with BMI or WC. ROC analysis ascertained an optimal MRI-PDFF threshold of 3.5% for predicting metabolic syndrome (sensitivity = 76 %, specificity = 83 %). Complex quantitative MRI demonstrates strong correlation and agreement with MRS to quantify hepatic triglyceride content in adolescent girls and young women. A low PDFF threshold is predictive of metabolic syndrome in this population. • Confounder-corrected quantitative MRI (ccqMRI) effectively measures hepatic triglyceride content in adolescent girls. • MRS and ccqMRI strongly correlate in liver proton density fat-fraction (PDFF) detection. • A PDFF threshold of 3.5% may be predictive of paediatric metabolic syndrome.