Analysis Of Comprehensive Risk Factors Of Osteoporotic Fracture In Postmenopausal Women

Objective:To explore the independent risk factors of osteoporotic fracture in postmenopausal women(PMOF),and to compare the accuracy of independent risk factors in evaluating the risk of PMOF,so as to provide basis for the prevention and clinical treatment of PMOF.Methods:A total of 168 patients with postmenopausal osteoporosis(OP)treated in the Department of Orthopaedics of the first affiliated Hospital of Chengdu Medical College from October 2021 to December 2022 were divided into fracture group(n = 77)and nonfracture group(n = 91)according to the occurrence of osteoporotic fracture(OF).The following detection indexes were collected: bone mineral density(BMD)was measured by dual energy X-ray absorptiometry(DXA),and bone turnover markers(BTMs)were detected by electrochemiluminescence: type I collagen cross-linked C-terminal peptide(β-CTX),osteocalcin N-segment middle molecular fragment(N-MID),type I procollagen N-terminal peptide(PINP),25-hydroxyvitamin D [25(OH)D].Serum indexes such as calcium(Ca),phosphorus(P),estradiol(E2)and serum total bilirubin(TB)were measured,and general data such as age and body mass index(BMI)were collected.The observation indicators were obtained through the following two parts of experiments: 1)to determine the independent risk factors of PMOF: age,height,weight,BMI,lumbar BMD,femoral neck BMD,Wards triangle BMD,intertrochanteric BMD,total hip BMD,β-CTX,PINP,N-MID,Ca,P,25-(OH)D,E2,TB between fracture group and non-fracture group were compared by t-test or nonparametric test.Spearman correlation analysis was used to analyze the correlation between different variables.The correlation coefficient r > 0 showed positive correlation,r < 0 showed negative correlation.Binary logistic regression analysis was used to determine the risk factors of PMOF.Taking fracture as dependent variable,Ca,E2,25-(OH)D,β-CTX,PINP,lumbar BMD,femoral neck BMD,intertrochanteric BMD and total hip BMD as independent variables,Logistic regression analysis was established to analyze the risk factors of PMOF.2)determination of the accuracy of independent risk factors in the assessment of PMOF risk:a new evaluation model was established by combining different independent indexes:lumbar BMD,25-(OH)D,E2,Ca.The receiver operating characteristic(ROC)curve was used to analyze these models,and the area under the curve(AUC)was used to judge the accuracy of each index.Finally,De Long test was used to compare whether the AUC values of different variables were meaningful.Results:1)Analysis of independent risk factors of PMOF: lumbar BMD,femoral neck BMD,intertrochanteric BMD,total hip BMD,Ca,E2 and 25-(OH)D in the fracture group were lower than those in the non-fracture group,while β-CTX and PINP in the fracture group were significantly higher than those in the non-fracture group,but there was no significant difference in age,height,weight and BMI,Wards triangle BMD,N-MID,P,TB between the two groups.Spearman correlation analysis showed that β-CTX and PINP were positively correlated with PMOF,while other Ca,E2,25-(OH)D,lumbar BMD,femoral neck BMD,intertrochanteric BMD,total hip BMD were negatively correlated with PMOF.Adjusted Logistic regression analysis showed that Ca(adjusted OR 0.007 [95%CI0.000-0.154];P < 0.0l),E2(adjusted OR 0.963 [95%CI 0.937-0.989];P < 0.0l),25-(OH)D(adjusted OR 0.894 [95%CI 0.847-0.944];P < 0.01)and lumbar BMD(adjusted OR0.024 [95%CI 0.847-0.944]).These four indexes are still the risk factors of PMOF.2)Determination of the accuracy of independent risk factors in the assessment of PMOF risk: on this basis,a relevant model group containing different independent risk factors was established.Analyze these model groups by ROC curve.The AUC values and95% CI are as follows: model 1: 0.632(0.554-0.705),model 2: 0.737(0.663-0.802),model 3: 0.631(0.553-0.704),model 4: 0.644(0.567-0.716),model 5: 0.753(0.681-0.816),model 6: 0.660(0.583-0.731),model 7: 0.684(0.608-0.754).model 8: 0.769(0.698-0.831),model 9: 0.784(0.714-0.844),model 10: 0.703(0.627-0.771),model 11:0.782(0.711-0.841),model 12: 0.792(0.723-0.851),model 13: 0.728(0.655-0.794),model 14: 0.808(0.741-0.865),model 15: 0.819(0.752-0.874).De Long test showed that the AUC of model 9 was significantly higher than that of model 6,model 7 and model 10 in the model with two different independent risk factors.The AUC of model 6 was significantly lower than that of model 8.In the model with three different independent risk factors,the AUC of model 14 was significantly higher than that of model 13.There was no significant difference in AUC between model 14 and model 15.Finally,the AUC of the model containing 25-(OH)D,E2 and Ca was significantly higher than that of other model groups.Conclusion:1.25-(OH)D,lumbar BMD,E2 and Ca are independent risk factors of PMOF.Clinically,patients with abnormal results of the above four indexes should be informed of the risk of fracture,and early anti-osteoporosis and fracture prevention measures are recommended.2.The new model combined with 25-(OH)D,E2 and Ca is more accurate in evaluating the risk of PMOF.It is an effective tool for predicting PMOF and can be widely used as an effective model for clinical prediction of PMOF.