This study investigated the relationship between lumbar vertebral marrow fat accumulation and bone metabolic profiles in postmenopausal women, and evaluated the ability of spectral CT–based fat density parameters to diagnose osteoporosis. This study adopted a cross-sectional design and enrolled 60 postmenopausal females aged between 50 and 79 years who underwent spectral CT and DXA examinations. Subjects were subdivided on the basis of age distribution and BMD levels. Spectral CT imaging was used to extract quantitative parameters reflecting vertebral fat content and HAP-associated density at the L3 segment. Differences and associations were analyzed using one-way ANOVA and correlation tests, while diagnostic value was assessed via ROC curves. With increasing age, bone marrow fat density (HAP fat) showed no significant difference across age groups, although correlation analysis indicated a positive association, while BMD decreased. Compared to the normal bone mass group, osteoporotic participants exhibited higher fat (HAP) values and lower HAP (water), HAP (iodine), Ca (fat), and Ca (water) density values. BMD was negatively correlated with fat (HAP) (r = −0.628, p < 0.001) and positively correlated with HAP (water) (r = 0.818, p < 0.001). Among all parameters, HAP (water) had the highest diagnostic accuracy for osteoporosis, with an AUC of 0.8676, sensitivity of 80.6%, and specificity of 79.3%. Bone marrow adiposity showed no significant differences across age groups but exhibited a weak positive correlation with age, and was negatively correlated with BMD in postmenopausal women. Spectral CT-derived fat and HAP density values, particularly HAP (water), are associated with postmenopausal osteoporosis and may serve as non-invasive imaging indicators for its detection.
Zhang et al. (Mon,) studied this question.