AUTHOR=Zhang Jing , Li Qiyuan , Wang Yao , Sun Li , Zhang Qingyuan , Gao Chuanping 

TITLE=The q-Dixon sequence for MRI predicts osteoporotic vertebral compression fractures

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2025.1591044

DOI=10.3389/fendo.2025.1591044

ISSN=1664-2392

ABSTRACT=ObjectivesTo evaluate whether q-Dixon sequence-based fat fraction (FF) values of the lumbar spine can predict osteoporotic vertebral compression fracture (OVCF) risk in older adult(s) osteoporosis patients.Materials & methodsThirty OVCF patients and 15 osteoporosis patients were enrolled. Areas of interest (ROIs) were manually drawn using the post-processing workstation, and FF values of the patient’s L1–L4 vertebrae (except the fractured vertebrae) were measured. The Pearson correlation test was used to analyze the correlation between the average lumbar spine FF value and bone mineral density (BMD). The receiver operating characteristic curve (ROC) and area under the curve (AUC) were used to evaluate the prediction efficiency of the FF values and to determine the best cut-off value for prediction.ResultsThe average lumbar spine FF value and the FF values of the L1–L4 vertebrae in the fracture group were significantly higher than those in the non-fracture group, and there was no significant difference in BMD between the two groups. ROC analysis showed that the AUC of the average lumbar spine FF value was 0.822, with sensitivity = 73.3%, specificity = 86.7%, and cut-off value = 57.27%. Among the L1–L4 vertebrae, the FF value of L2 vertebrae had the highest AUC of 0.870, and the cutoff value was 56.62%.ConclusionThe FF values of the lumbar spine measured by the q-Dixon sequence can help predict OVCF risk and provide complementary information to BMD measurements. The FF value of L2 vertebrae has the best prediction efficiency. However, these findings should be interpreted with caution given the relatively small sample size (n=45) and manual ROI segmentation method used in this study