BMC Medical Imaging
Fatty pancreas disorder on CT: associations with mesenteric fat attenuation, irregular contour, and reduced main pancreatic duct diameter
By Myriam Boumaiza, Caroline Prouvot, Sylvain Grange and 7 others.
Abstract
To identify clinical and radiological factors associated with fatty pancreas disorder (FPD) on computed tomography (CT), with a particular emphasis on body composition parameters. In this retrospective single-center study, 389 adult patients who underwent ERCP between September 2014 and June 2023 and had a prior CT scan were included. CT-defined FPD was defined as unenhanced pancreatic attenuation ≤ + 15 HU, approximating the lowest quartile of the distribution. CT-derived measurements included visceral fat area (VFA), subcutaneous fat area, skeletal muscle area, skeletal muscle density (SMD), main pancreatic duct (MPD) diameter, and liver and spleen attenuation. Factors associated with FPD were identified using univariable and multivariable logistic regression. The definition was tested in sensitivity analyses using a pancreas-to-spleen attenuation ratio, an absolute threshold of ≤ 30 HU, and pancreatic attenuation modelled as a continuous variable. Interobserver reproducibility of pancreatic contour classification was assessed in a random subsample of 50 patients. FPD was present in 106 patients (27.2%). Patients with FPD were significantly older (81.1 ± 9.1 vs. 71.5 ± 17.4 years, p < 0.001) and had higher VFA (242.1 ± 116.3 vs. 200.2 ± 117.0 cm², p = 0.005). Multivariate analysis identified older age (OR = 1.06; 95% CI: 1.03–1.09; p < 0.001), absence of smooth pancreatic contour (OR = 0.40; 95% CI: 0.20–0.80; p = 0.010), smaller MPD diameter (OR = 0.65; 95% CI: 0.50–0.85; p = 0.002) and lower mesenteric fat attenuation (OR = 0.96; 95% CI: 0.94–0.99; p = 0.017) as independent predictors of FPD. Pancreatic attenuation showed negative correlations with age (r = − 0.34) and VFA (r = − 0.25), and positive correlations with psoas SMD (r = + 0.26) and MPD diameter (r = + 0.26). FPD was independently associated with advanced age, irregular pancreatic contour, reduced MPD diameter, and lower mesenteric fat attenuation. Pancreatic contour remained robust across alternative FPD definitions, whereas the association with MPD diameter was definition-dependent. Question: Which clinical and CT imaging factors are independently associated with fatty pancreas disorder (FPD) on routine CT? Findings: Fatty pancreas disorder was independently associated with older age, loss of a smooth pancreatic contour, a smaller main pancreatic duct diameter, and lower mesenteric fat attenuation on CT. Clinical relevance statement: Fatty pancreas disorder is associated with a distinct metabolic and pancreatic morphological phenotype that can be recognized on routine CT.










