Objective:

To investigate magnetic resonance imaging-proton density fat fraction (MRI-PDFF) distribution characteristics of non-alcoholic fatty liver disease (NAFLD) in overweight/obese children and to provide a reference for the development of the evaluation standard of liver fat quantification in children with NAFLD in the future.

Methods:

Enrolled 69 overweight/obese children who underwent 3.0 T magnetic resonance m-Dixon-quant sequence scanning of upper abdomen in Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from August 2022 to March 2023. The MRI-PDFF

laboratory biochemical index

and measurement data of human body index in all patients were recorded. Patients were divided into four groups (Grade 0、Grade 1、Grade 2、Grade 3) according to the weighted mean of MRI-PDFF (MRI-PDFF

wm

). Analyzed the variability of PDFF values between hepatic segments and their relationship with fatty liver grading. Explored the consistency between MRI-PDFF

I-Ⅷ

and MRI-PDFF

wm

in the diagnosis and grading of fatty liver. Compared the correlation between MRI-PDFF

I-Ⅷ

with alanine aminotransferase (ALT)

aspartate aminotransferase (AST)

and gamma-glutamyl transferas (GGT).

Results:

Among the 69 participants

58 patients (84.06%) were diagnosed with NAFLD. The MRI-PDFF of the NAFLD patients was (20.23±10.16)%. Hepatic segment

Ⅶ

had the highest MRI-PDFF (22.23±9.85)%. Hepatic segment Ⅴ had the h

ighest intraclass correlation coefficient (ICC=0.988

P

＜0.001). The range of MRI-PDFF in normal liver (2.59%) and mild fatty liver (7.15%) was lower than that in patients with moderate to severe fatty liver (9.71%-11.14%)

with statistically significant differences (

P

＜0.001). In the diagnosis of the normal liver and fatty liver (Kappa value was 1

P

＜0.001) and fatty liver grading (Kappa value was 0.880

P

＜0.001)

MRI-PDFF V showed the best consistency with MRI-PDFF

wm

. Liver MRI-PDFF

I-Ⅷ

had moderate and positive correlation with ALT (

r

=0.415-0.470

P

＜0.001)

and hepatic segment Ⅳb showed the highest correlation (

r

=0.470

P

＜0.001). Liver MRI-PDFF

I-Ⅷ

had slight and positive correlation with AST (

r

=0.252-0.337

P

＜0.005) and GGT (

r

=232-0.333

P

＜0.005).

Conclusion:

NAFLD has a high incidence in overweight/obese children

and the distribution of fat is usually uneven in the liver. The variability of hepatic fat distribution is increased with the degree of hepatic steatosis. Compared to other hepatic segments

MRI-PDFF

V

is more representative for the overall degree of liver steatosis and can be used as the preferred location for liver biopsy or other non-invasive quantitative evaluation. The liver fat content can reflect the damage degree of liver function to some extent.