The Clinical Study On Predicting Fragile Fracture Risk Of Patients With Type 2 Diabetes Mellitus Using FRAX Tool And Hip Structure Analysis

BackgroundThe incidence of type 2 diabetes mellitus(T2DM)is increasing year by year,and it has become a frequent disease among the middle-aged and elderly people.With the aging of the global population,osteoporosis(OP)has gradually became a more and more common disease,and they often coexist.In recent years,most studies believe that the risk of osteoporotic fracture in patients with T2DM has increased significantly,especially the middle-aged and elderly patients with T2DM,due to multiple factors such as diabetic complpications,the use of hypoglycemic drugs,menopause,falls,tobacco,alcohol and adverse living habits,has become a high-risk group of osteoporotic fracture.At present,bone mineral density(BMD)measured by dual-energy X-ray absorptiometry(DXA)is mainly used to evaluate bone mass and diagnose osteoporosis.And use the Fracture Risk Assessment Tool(FRAX)to predict the fracture risk of patients in the next ten years.However,many studies have confirmed that FRAX underestimates the actual fracture risk of type 2 diabetes mellitus patients.More valuable methods are urgently needed to improve the effectiveness of FRAX in assessing type 2 diabetes mellitus patients.In addition,Imany fragile fractures in diabetic patients occur in people with bone mineral density T>-2.5,suggesting that factors other than BMD affect bone strength and fracture risk.Changes in hip geometric parameters may partly explain the high fracture risk beyond BMD in patients with T2DM.ObjectiveTo explore the methods and value of improving fracture risk assessment in middle-aged and elderly patients with T2DM,and to discuss the correlation between hip geometry with bone mineral density and fracture risk of middle-aged and elderly patients with T2DM.Objects and methods367 middle-aged and elderly patients with T2DM were recruited retrospectively,including 202 postmenopausal women and 165 men over 50 years old.General clinical data,fracture risk factors and laboratory biochemical indicators of all subjects were collected.Using FRAX to predict the fragile fracture risk in the next 10 years.Then to calculate the probability of major osteoporotic fracture(PMOF)and hip fracture(PHF)in ten years by increasing the age by 10 years and T value of femoral neck minus 0.5 standard deviation(SD)respectively.OSTA index of all subjects was calculated.Dual-energy X-ray absorptiometry(DXA)and its own hip structure analysis software(HSA)were used to determine the bone density values of the lumbar spine,femoral neck and the hip geometric parameters of all subjects.According to the different methods of correcting FRAX scores of diabetic patients,all subjects were divided into three groups:group A:uncorrected FRAX score;group B:FRAX score was corrected by age plus 10 and group C:FRAX score was corrected by femoral neck T value minus 0.5 standard deviation;According to PHF≥3%as the treatment cut-off point,the patients were divided into two groups:the same treatment recommendations group and the different treatment recommendations group.According to gender,they were divided into postmenopausal female patients with T2DM and male patients over 50 years old with type 2 diabetes mellitus.Then according to age and WHO diagnostic criteria for osteoporosis separately,male and female patients were divided into three subgroups:under 60 years old,60-70 years old and over 70 years old,and normal bone mass,osteopenia and osteoporosis.All subjects were divided into non-fracture group and fracture group according to whether they had fracture in the past.All data were processed by SPSS 25.0 and GraphPad Prism 7.0 statistical software.The measurement data conforming to normal distribution are expressed by mean±standard deviation,while the measurement data not conforming to normal distribution are expressed by median and quartile spacing.Enumeration data are expressed by the number of cases and percentage(%).The data were analyzed by statistical methods such as t-test,nonparametric test,variance analysis,Kruskal-Wallis test,Pearson correlation analysis,Spearman correlation analysis and multiple regression analysis.Result1.Comparison between group B and group A:PHF of group B was higher than group A.There was no statistical difference in the ratio of patients recommended for treatment between the two groups.2.Comparison between group C and group A:The PMOF and PHF in group C were 0.6%and 0.4%higher than those in group A respectively.The ratio of patients recommended for treatment of group C was higher than that of group A.The age,the rate of patients with previous fracture and the rate of patients with alcohol intake of the same treatment recommendations group are lower than the different treatment recommendations group,while the weight,BMI and T value of femoral neck are higher than the different treatment recommendations group.3.Comparison among the three subgroups(<60 years old group,60-70 years old group and<70 years old group):3.1 Postmenopausal women with T2DM:PMOF,PHF,NN-BR,osteoporosis detection rate increased gradually among the three subgroups,while OSTA index,FN-BMD,NN-CSA,NN-CT decreased gradually,and NN-Z,IT-CSA and NSA of<70 years old group were lower than those of<60 years old group.3.2 Male patients with T2DM:PMOF and PHF increased gradually and OSTA index decreased gradually among the three subgroups.IT-CSA,IT-CSMI and IT-Z of>70 years old group were lower than<60 years old group.The detection rate of osteoporosis in>70 years old group was higher than that of<60 years old group.4.Comparison among three subgroups(normal bone mass group,osteopenia group and osteoporosis group):4.1 Postmenopausal women with T2DM:age,course of disease,menopausal years,PMOF,PHF,NN-BR,IT-BR,FS-BR,fracture detection rate increased gradually,while weight,BMI,OSTA index,lumbar BMD,FN-BMD,NN-CSA,NN-CSMIL NN-Z,NN-CT,IT-CSA,IT-CSMI,IT-Z,IT-CT,FS-CSA.FS-CSMI,FS-Z and FS-CT decreased gradually.F-CP in normal bone mass group was higher than that in osteoporosis group.4.2 Male patients with T2DM:age,course of disease,PMOF,PHF,NN-BR,IT-BR,FS-BR,fracture detection rate were gradually increased,weight,BMI,OSTA index,lumbar BMD,FN-BMD,NN-CSA,NN-CSMIL NN-Z,NN-CT,IT-CSA,IT-CSMI,IT-Z,IT-CT,FS-CSA,FS-CSMI FS-Z,FS-CT decreased gradually.F-CP level in normal bone mass group was higher than that in osteoporosis group.5.Comparison of parameters between non-fracture group and fracture group:Compared with the non-fracture group,PMOF,PHF and BR in the fracture group were higher than those in the non-fracture group,while FN-BMD,OSTA index,FN-CSA,FN-CSMI,FN-Z and FN-CT were lower than those in the non-fracture group.6.The ROC results interpretation of relevant parameters to identify fragilefracture:By ROC curve analysis,the area under the curve was:Five-GPH combined with FRAX score>FRAX score>Five-GPH>FN-BMD>OSTA index.7.Correlation analysis between hip geometric mechanical parameters and various factors:Correlation analysis showed that in postmenopausal women with T2DM,FN-CSA was positively correlated with weight,BMI,OSTA index,F-CP,lumbar BMD,and FN-BMD,and negatively correlated with age,disease course,menopause years,PMOF,PHF,and AKP levels;FN-CSMI was positively correlated with weight,BMI,OSTA index,lumbar BMD,and FN-BMD,and negatively correlated with age,disease course,menopause years,PMOF,and PHF.FN-Z was positively correlated with weight,BMI,OSTA index,UA,lumbar BMD,FN-BMD,and negatively correlated with age,disease course,menopause years,PMOF,and PHF.FN-CT was positively correlated with weight,BMI,OSTA index,F-CP,lumbar BMD,and FN-BMD,and negatively correlated with age,disease course,menopause years,PMOF,PHF,and AKP levels.FN-BR was positively correlated with age,disease course,menopause years,PMOF,PHF,AKP level,and negatively correlated with weight,BMI,OSTA index,F-CP,lumbar BMD,and FN-BMD.NSA was positively correlated with weight,OSTA index,TG,L2-BMD,lumbar spine total BMD,and negatively correlated with age and menopause years.In male patients over 50 years old,FN-CSA was positively correlated with weight,BMI,OSTA index,TG,UA,lumbar BMD,and FN-BMD,but negatively correlated with age,disease course,PMOF,PHF,and HDL-C levels.FN-CSMI was positively correlated with weight,height,BMI,OSTA index,lumbar BMD and FN-BMD,but negatively correlated with PMOF and PHF.FN-Z was positively correlated with weight,height,BMI,OSTA index,lumbar BMD,and FN-BMD,but negatively correlated with age,disease course,PMOF,and PHF.FN-CT was positively correlated with weight,BMI,OSTA index,TG,UA,lumbar BMD,FN-BMD,and was negatively correlated with age,disease course,PMOF,PHF,HDL-C,and AKP levels.FN-BR was positively correlated with age,disease course,PMOF,PHF,HDL-C level,and negatively correlated with body weight,BMI,OSTA index,HbAlc,UA,lumbar BMD,and FN-BMD.NSA is positively correlated with TG and UA and negatively correlated with HDL-C.8.Multiple linear regression analysis of hip geometric and mechanical parameters,FRAX score and bone density:After adjusting for age,weight,BMI,duration of diabetes,duration of menopause,and HbAlc,in postmenopausal women with T2DM,CSA,CSMI,Z,FN-CT,BR were not correlated with PMOF,PHF,and lumbar BMD.But CSA,CSMI,Z,and FN-CT were positively correlated with FN-BMD,while BR was negatively correlated with FN-BMD.NSA had no correlation with PMOF,PHF,lumbar and femoral BMD.In male T2DM patients over 50 years old,CSA,CSMI,Z,and FN-CT were not correlated with PMOF,PHF,and lumbar BMD,but positively correlated with FN-BMD,and BR was positively correlated with PHF and negatively correlated with FN-BMD.NSA was not correlated with PMOF,PHF,L1-BMD,L4-BMD,FN-BMD,but positively correlated with L2-L3 BMD and lumbar spine total BMD.Conclusion1.Compared with conventional FRAX in predicting fracture risk of middle-aged and elderly patients with T2DM,applying T value of femoral neck minus 0.5SD method to correct FRAX score can improve fracture risk prediction value,so as to carry out clinical intervention timely and reduce the fracture risk of patients.2.The lower the bone mass of middle-aged and elderly patients with T2DM,the thinner the femoral neck cortex,the greater the pressure,the smaller the bone stiffness,the smaller the bending resistance,the higher the brittleness,the more unstable the hip geometric structure.The hip geometric structure can reflect the changes of femoral neck bone strength in middle-aged and elderly patients with T2DM to some extent.3.Geometric and mechanical parameters of hip combined with FRAX score can greatly improve the accuracy of fracture risk prediction in middle-aged and elderly patients with T2DM.