Objective To explore the clinical application value of MRI fat quantification parameter ［vertebral bone marrow fat fraction （FF）］ combined with 25?hydroxyvitamin D ［25（OH）D］ in predicting fracture risk in patients with osteoporosis. Methods A total of 90 patients with osteoporosis who were admitted to the hospital from January 2023 to April 2024 were selected as the subjects. Among them， 50 patients with osteoporotic vertebral compression fractures were included in the fracture group， and 40 patients without fractures were included in the control group. All patients underwent the iterative decomposition of water and fat with echo asymmetry and least?squares estimation （IDEAL?IQ） method of MRI to measure the FF of each vertebra among L_1?5 and the average FF of L_1?5. Serum 25 （OH） D level was detected by electrochemiluminescence method. FF and serum 25（OH）D levels of the two groups were compared. The correlation of FF， 25（OH）D and bone mineral density （BMD） was analyzed. Multivariate logistic regression analysis was conducted to screen the risk factors for fracture in patients with osteoporosis. Receiver operating characteristic （ROC） curves were used to evaluate the predictive value of FF， 25（OH）D， and their combination for fracture in patients with osteoporosis. Results Patients in the fracture group were older than those in the control group. BMD and serum 25（OH）D level were lower than those of the control group （P < 0.05）. The FF of L₂ and average FF of L_1?5 in the fracture group were higher than those in the control group （P < 0.05）. Correlation analysis results showed that the FF of L₂ and the average FF of L_1?5 were negatively correlated with BMD （P < 0.05）， while serum 25 （OH） D level was positively correlated with BMD （P < 0.05）. Multivariate logistic regression analysis showed that age and FF of L₂ were independent risk factors for fracture in patients with osteoporosis， while BMD and 25（OH）D were protective factors （P < 0.05）. ROC curves indicated that the AUC values of FF of L2 and 25（OH）D for predicting fracture were 0.714 （95% CI： 0.606 ~ 0.822） and 0.774 （95% CI： 0.672 ~ 0.876）. The AUC of joint prediction was 0.923 （95% CI： 0.867 ~ 0.978）， which was significantly larger than that of separate prediction （P < 0.05）. Conclusions FF of L2 and serum 25（OH）D are related to fracture in patients with osteoporosis. Age and BMD are factors influencing the occurrence of fracture in patients with osteoporosis. FF of L2 and 25（OH）D have certain predictive value for fracture risk in patients with osteoporosis， and combined detection of the two can improve predictive efficiency.