Normal range of hepatic fat fraction on dual- and triple-echo fat quantification MR in children

To evaluate hepatic fat fraction on dual- and triple-echo gradient-recalled echo MRI sequences in healthy children. We retrospectively reviewed the records of children in a medical check-up clinic from May 2012 to November 2013. We excluded children with abnormal laboratory findings or those who were overweight. Hepatic fat fraction was measured on dual- and triple-echo sequences using 3T MRI. We compared fat fractions using the Wilcoxon signed rank test and the Bland-Altman 95% limits of agreement. The correlation between fat fractions and clinical and laboratory findings was evaluated using Spearman's correlation test, and the cut-off values of fat fractions for diagnosing fatty liver were obtained from reference intervals. In 54 children (M:F = 26:28; 5-15 years; mean 9 years), the dual fat fraction (0.1-8.0%; median 1.6%) was not different from the triple fat fraction (0.4-6.5%; median 2.7%) (p = 0.010). The dual- and triple-echo fat fractions showed good agreement using a Bland-Altman plot (-0.6 ± 2.8%). Eight children (14.8%) on dual-echo sequences and six (11.1%) on triple-echo sequences had greater than 5% fat fraction. From these children, six out of eight children on dual-echo sequences and four out of six children on triple-echo sequences had a 5-6% hepatic fat fraction. When using a cut-off value of a 6% fat fraction derived from a reference interval, only 3.7% of children were diagnosed with fatty liver. There was no significant correlation between clinical and laboratory findings with dual and triple-echo fat fractions. Dual fat fraction was not different from triple fat fraction. We suggest a cut-off value of a 6% fat fraction is more appropriate for diagnosing fatty liver on both dual- and triple-echo sequences in children.