Advanced Imaging of Pancreatic Steatosis with Dixon MRI: Key Clinical Factors Revealed

Objective: In this cross-sectional study, Dixon MRI was used to assess pancreatic fat in individuals with prediabetes and recently diagnosed diabetes. The aim was to explore its correlation with various parameters and establish a cut-off value for predicting a fatty pancreas.

Methods: This study included 141 participants (67 males, 74 females; median age, 49.5 years). Of these, 55 patients were identified as having prediabetes, 55 as having recently diagnosed diabetes, and 31 as controls. All patients underwent chemical shift-encoded magnetic resonance imaging on a 1.5T scanner to quantify the amount of fat in the pancreas by assessing the proton density fat fraction (P-PDFF) in the head, body, and tail of the pancreas separately. Correlation and multivariable regression analyses were performed to investigate the relationship between P-PDFF with metabolic factors and anthropometric parameters such as age, weight, body mass index (BMI), waist circumference (WC), and laboratory results.

Results: The average P-PDFF across all subjects was 10.02%, with significantly higher P-PDFF values in the prediabetes group (10.64%) and the diabetes group (11.28%) compared to controls (6.69%) (P < .001). A cut-off value of ≥7.05% predicted pancreatic steatosis with 76.4% sensitivity and 67.7% specificity. P-PDFF was moderately correlated with WC (r = 0.502, P < .001). P-PDFF was also associated with glycemic factors (all P < .001) but was not directly linked to the development of diabetes.

Conclusion: Magnetic resonance imaging can quantitatively measure pancreatic steatosis, with WC being the key predictor of P-PDFF.

Cite this article as: Akhan BS, Yiğit H, Aral Y, Omma T, Karaca A, Koşar PN. Advanced imaging of pancreatic steatosis with dixon MRI: Key clinical factors revealed. Endocrinol Res Pract. 2025;29(3):163-170.