Abstract
Background/Objectives: While gradient-echo (GRE)-based chemical shift-encoded magnetic resonance imaging (CSE-MRI) offers precise method for measuring adiposity in bone marrow, its limitation lies in the need for additional imaging. On the other hand, spin-echo (SE)-based CSE-MRI can seamlessly integrate into conventional protocols. Recently, a novel technique called the vertebral bone quality (VBQ) score has been introduced. The objective of this study was to investigate the association between fat fraction (FF) measured by GRE-based CSE-MRI (FF(GRE)) and FF measured by SE-based CSE-MRI (FF(SE)) or the VBQ score. Methods: A retrospective study with 344 patients assessed the correlation between FF and the VBQ score and each measurement's correlation with age using Pearson's correlation (r). Concordance between FF(GRE) and FF(SE) was assessed using Lin's concordance correlation coefficient (ρ(c)). Vertebral lesions (n = 41) were categorized as benign and malignant, and the Mann-Whitney U test was used for comparison. Results: FF(GRE) demonstrated strong positive correlations with FF(SE) and the VBQ score (r = 0.861 and 0.708, respectively). However, the concordance between FF(GRE) and FF(SE) was poor (ρ(c) = 0.295). All measurements moderately correlated with age (FF(GRE), r = 0.583; FF(SE), r = 0.477; VBQ score, r = 0.468). FF was significantly higher in benign lesions (FF(GRE), p = 0.004; FF(SE), p = 0.007), while the VBQ score did not show statistically significant differences between the two groups (p = 0.089). Conclusions: FF(GRE) exhibited a high correlation with the VBQ score. FF(SE) showed a strong correlation with FF(GRE), but replacing FF(GRE) with FF(SE) may be challenging. Both FF and the VBQ score moderately correlated with age. FF demonstrated statistically significant differences between benign and malignant lesions, while the VBQ score did not provide a distinguishable separation.