The associations of muscle and adipose tissue with bone mineral density (BMD) in perimenopausal women has been studied, but there are differences in site distribution, which lead to that the effect of muscle and adipose tissue on BMD and the site of action are still unclear.

To analyze the relationships of BMD in lumbar spine, femoral neck, total hip and adipose, muscle tissue in perimenopausal women, and provide theoretical basis for improving the quality of life of middle-aged and elderly and preventing and treatment of osteoporosis (OP) .

2 355 women aged 40-60 years who underwent medical examination in the Health Management Center, Affiliated Hospital of Guizhou Medical University from January 2018 to October 2021 were retrospectively selected and divided into pre-menopausal women (n=1 261) and post-menopausal women (n=1 094) according to whether they were menopausal or not. BMD of lumbar spine, femoral neck, and total hip were measured by Dual energy X-ray Absorptiometry (DXA) . Adipose and muscle tissue were measured by Bioelectrical Impedance Analysis (BIA) . To explore the association of adipose and muscle tissue with BMD of different site in perimenopausal women by Pearson correlation and multiple linear regression analysis. Receiver operating characteristic (ROC) curve was used to analyze the predictive value of appendicular lean mass (ALM) for postmenopausal OP.

Pearson correlation analysis showed that adipose mass, total body muscle mass, trunk muscle mass and ALM were positively correlated with BMD of lumbar spine, femoral neck and total hip in premenopausal, postmenopausal and overall women (P<0.05) .Multiple linear regression analysis showed that total body muscle mass and trunk muscle mass were the influencing factors of femoral neck BMD in overall female (β=-0.051, 0.099, P<0.05) , ALM was an influencing factor of BMD of lumbar spine, femoral neck and total hip in overall female (β=0.244, 0.199, 0.199, P<0.05) ; total body muscle mass was an influencing factor for BMD of femoral neck and total hip in postmenopausal women (β=-0.125, -0.075, P<0.05) . Trunk muscle mass was an influencing factor for BMD of femoral neck in postmenopausal women (β=0.150, P<0.05) . ALM was the influencing factor for BMD of lumbar spine, femoral neck and total hip in postmenopausal women (β=0.226, 0.216, 0.188, P<0.05) . ROC curve analysis showed that the AUC of ALM for predicting OP of total hip in postmenopausal women was 0.825〔95%CI (0.742, 0.908) 〕, with 81.82% sensitivity and 76.59% specificity when the cut-off value was 16.24 kg. The AUC of ALM for predicting OP of femoral neck in postmenopausal women was 0.760〔95%CI (0.692, 0.829) 〕, with 75.61% sensitivity and 61.89% specificity when the cut-off value was 16.68 kg. The AUC of ALM for predicting OP of lumbar spine in postmenopausal women was 0.641〔95%CI (0.603, 0.679) 〕, with 60.85% sensitivity and 60.17% specificity, when the cut-off value was 17.20 kg. The best site of ALM for predicting OP in postmenopausal women was the total hip (Z_{femoral neck}=-9.89, P<0.05) .

ALM is positively correlated with BMD of lumbar spine, femoral neck and total hip in postmenopausal women. ALM can predict the occurrence of OP at different body parts of postmenopausal women, of which the optimal site of prediction for OP is the total hip.