Association of intra-pancreatic fat deposition with bone marrow adiposity in young obese adults.

BACKGROUND: Ectopic fat accumulation is a key factor in impaired bone health in the obese population. Although hepatic fat fraction (HFF) has been established as an important marker of metabolic dysfunction, the effect of pancreatic fat fraction (PFF) on bone marrow adipose tissue (BMAT) in young obese adults remains unclear. The aim of this study was to investigate the independent association between PFF and bone marrow adiposity and to compare it with that of HFF. METHODS: This cross-sectional study included 155 young obese adults [mean age 35.01±6.25 years; mean body mass index (BMI) 31.9±4.3 kg/m2]. BMAT, PFF, and HFF were quantitatively measured using chemical shift-encoded (CSE) magnetic resonance imaging (MRI-PDFF sequence). Statistical analysis employed stratified multivariate linear regression and restricted cubic spline (RCS) models to assess the independent association between PFF and HFF with BMAT. Models were adjusted for age, gender, BMI, glycosylated hemoglobin (HbA1c), and triglycerides (TG). RESULTS: PFF exhibited only a weak correlation with HFF [Spearman correlation coefficient (rs) =0.180, P=0.037]. Consistent positive associations with BMAT were observed across all pancreatic sub-regions (head, body, and tail: rs =0.29-0.34, all P<0.001). In the fully adjusted model, PFF demonstrated a robust and independent positive association with BMAT (β=0.323, P=0.014). Although HFF remained statistically significant, its effect size was substantially smaller than that of PFF (β=0.206, P=0.016). RCS analysis revealed a continuous linear dose-response relationship between BMAT and whole pancreas (P for overall =0.006), head (P=0.013), body (P=0.009), and tail (P=0.015), yet the nonlinearity tests did not reach statistical significance (all P for nonlinear >0.05). Notably, the shape of this linear dose-response trajectory was consistent across the pancreatic head, body, and tail, suggesting a generally uniform pathological impact without an identifiable safety threshold. CONCLUSIONS: PFF is a more robust and independent predictor of BMAT than HFF in young adults with obesity. These findings imply the existence of a distinct pathogenic pathway linking the fatty pancreas to marrow degeneration. Consequently, the assessment of pancreatic fat content may serve as an independent associated factor for skeletal risk stratification in this population.