| Objective:To explore the relationship between body weight change and bone mass change and the relationship between s RANKL and OPG in RANKL/RANK/OPG pathway.Methods:A total of 332 persons aged 46 to 90 years old with an average age of 65.36±10.04 years were selected from the health examination Center of Subei People’s Hospital of Jiangsu province for two consecutive years from 2021 to 2022.There were164 males aged 48 to 90 years old with an average age of 67.12±10.90 years.There were 168 females,aged 46-82 years old,with an average age of 63.64±8.91years.Basic information of each subject was collected in detail,height and weight in2021 were measured at baseline,and body mass index(BMI)was calculated.Baseline levels of glycosylated hemoglobin(Hb A1c),blood calcium,creatinine,triglycerides(TG),total cholesterol(TC),high density lipoprotein(HDLC),low density lipoprotein(LDLC)were measured.Serum soluble nuclear factor-KB receptor activating factor ligand(s RANKL)and serum osteoprotectin(OPG)were measured at follow-up.Bone mineral density(BMD)of lumbar 1-4(L1-L4),lumbar 1-4(L1-4),femoral neck and total hip were measured by dual-energy X-ray absorptiometry(DXA)for each subject for two consecutive years.BMD was recorded in grams per square centimeter(g/cm~2)accurate to 3 decimal places.Changes in bone mineral density at each site(△=bone mineral density at follow-up site-bone mineral density at baseline site).According to the change rate of body Weight[(follow-up weight-baseline Weight)/baseline weight]×100%],they were divided into:weight loss≥3%group(weight-loss,W-L),weight stability group(weight-stability,W-S):loss or gain<3%and a weight gain of≥3%(weight-gain,W-G).According to the different data types,ANOVA test,nonparametric rank sum test or Chi-square test were used for comparison between groups,and correlation analysis was conducted.Linear regression was used to exclude confounding factors,and then Bonferroni post hoc test and Kruskal-Wallis H test(K-W H test)were used for pairwise comparison between groups.P<0.05 was considered statistically significant.Results:1.The baseline age of W-L group(n=72)was lower than that of W-S group(n=204)and W-G group(n=56)(63.78±11.14 VS 64.20±9.09 VS 71.64±10.19,P=0.034).The bone density of the first lumbar vertebra{-0.035(-0.064,0.035)VS-0.009(-0.042,0.023)VS 0.036(-0.012,0.049),P=0.042}decreased more;At follow-up,s RANKL(2642.56±435.47 VS 2882.13±787.66 VS 3250.30±538.47,P=0.009)was lower.Baseline BMI(25.40±2.89VS 24.84±2.81 VS 23.25±2.60,P=0.048)was higher,and the difference was statistically significant.There were no significant differences in baseline height,body weight,sex,Hb A1c,blood calcium,creatinine,TG,TC,HDLC,LDLC,△L2,△L3,△L4,△L1-4,△femoral neck,△total hip and OPG(P>0.05).2.The change rate of body weight was positively correlated with△L1(r=0.309,P=0.004)and s RANKL(r=0.352,P=0.001).It was negatively correlated with baseline BMI(r=-0.283,P=0.01)and OPG(r=-0.288,P=0.008)(P<0.05).3.With△L1 as the dependent variable and the change rate of body weight as the independent variable,the effect of the change rate of body weight on△L1 in each model was(β1=0.399,β2=0.42,β3=0.456,β4=0.452,β5=0.499)after gradually correcting confounding factors such as age,BMI,s RANKL and OPG,respectively.P<0.05.4.The stratified analysis was conducted with baseline BMI of 24kg/m~2 as the cut-off point,and the results showed that in the population with baseline BMI≤24 kg/m~2:Compared with W-G group,△L1 bone mineral density decreased more in W-L group{-0.0495(-0.0918,-0.016)VS 0.036(-0.01,0.056)},s RANKL(2303.67±605.40VS 3213.84±623.70),OPG(662.05±213.21 VS 1005.46±319.27)were lower,and the differences were statistically significant.There was no difference in△L1 bone mineral density,s RANKL,OPG between W-G group and W-L group in baseline BMI>24kg/m~2.Conclusion:1.The change rate of body weight was positively correlated with the change of first lumbar vertebra bone density and s RANKL,and negatively correlated with baseline BMI and OPG.2.When confounding factors were gradually included and simple linear regression and multiple linear regression were performed,the change rate of body weight as an independent variable still had an effect on the change of the first lumbar vertebra bone density.3.In the group with baseline BMI≤24 kg/m~2:Weight loss≥3%resulted in more reduction in first lumbar bone density and lower serum s RANKL and OPG levels. |
