The Relationship Between Iron Levels And Bone Mineral Density

Osteoporosis(OP)caused by progressive decrease in bone mineral density and destruction of bone tissue microstructure has already become an important public health problem all over the world.The incidence of osteoporosis is generally higher in the elderly and postmenopausal women,and the risk is higher in women than in men.According to the previous studies,the prevalence of OP in the global range is increasing yearly,and there are 50%of the white women over 50 years old have osteoporosis,and about 20%of the patients lead to osteoporotic fracture will die in a variety of complications within one year after suffer from the fracture.Dual energy X-ray Absorptiometry(DXA)is commonly used in clinical in order to measure the bone mineral density(BMD)and diagnose osteoporosis.The risk of osteoporosis over the age of 50 will be lower in those who have a higher BMD in the younger age.Some scholars have suggested that iron may promote the loss of bone mass;it is also shown that the level of serum ferritin in patients with hemochromatosis or other hemopathy is high,and this type of patient is often combined with osteoporosis,so that a portion of the osteoporosis is considered to be caused by iron accumulation.The mechanism of the effect of iron level on bone mineral density is not clear now,but it is known that iron overload and iron deficiency both affect bone metabolism.Iron levels in humans can be measured by iron intake,circulating iron(serum iron)and storing iron(serum ferritin).Objective:The purpose of our study was to analyze the influencing factors and correlation between iron and bone mineral density of neck and lumbar spine,and to verify the effect of high iron level on bone mineral density of human body.By comparing the correlation between different iron levels and bone mineral density,the iron index with the best correlation with bone mineral density was determined.At the same time,the relationship between iron and bone density was investigated whether the relationship between iron and bone mineral density was consistent in different sites.Methods:Using the data of National Health Nutrition Examination Survey(NHANES)from 2005 to 2010,analyze the factors that may affect the bone mineral density and the association between the iron,serum iron,serum ferritin and the bone mineral density in the young women by using the IBM SPSS 24.0 software.Generalized Linear Models(GLMs)and Generalized Additive Models(GAMs)were used to testing the linear or non-linear relationship between iron intake,serum iron,serum ferritin and the bone mineral density.Logistic Regression models were used to comparing the R~2 among iron intake,serum iron and serum ferritin.Akaike information criterion(AIC)and Bayesian information criterion(BIC)were calculated to tradeoff the complexity of the estimation model and the goodness of fitting data which in order to prevent the complexity of the model from being too high due to the high precision.Results:1.A total of 31034 people were surveyed in the NHANES from 2005 to 2010.We excluded the individuals with missing or invalid measurements of iron intake,serum iron,serum ferritin and bone mineral density in femoral neck and lumbar spine,had 4000 women individuals.the average age of them is 27.70 years old.Furthuremore,we had 1480 women individuals with the average age of 24.91,after excluding the individuals with missing or invalid measurements of iron intake,serum iron,serum ferritin and bone mineral density in arms and legs and head.The levels of bone mineral density and iron measurements were all within the normal reference range in the two parts of ours study.2.In the adjusting GLM models,the individuals in aim one showed that serum iron and serum ferritin were negatively associated with bone mineral density in femoral neck(serum iron:?=-0.032,P=0.024;serum ferritin:?=-0.071,P<0.001)and lumbar spine(serum iron:?=-0.030,P=0.034;serum ferritin:?=-0.084,P<0.001).And iron intake was only negatively associated with femoral neck bone mineral density(?=-0.027,P=0.048).The individuals in aim two explained that serum ferritin was only negatively related to femoral neck(?=-0.066,P=0.012)and lumbar spine(?=-0.056,P=0.026)bone mineral density.What's more,iron intake was not associated with both femoral neck and lumbar spine BMD.There was no correlation between three iron indexes and bone mineral density of limbs and head.3.We performed model fitting in the individuals,which have a large sample of individuals,in aim one.We compared the fitting degree in serum iron and serum ferritin,which had correlation with BMD in the generalized linear regression analysis.The results showed that serum ferritin was better in femoral neck BMD(R~2=0.254,AIC=-5955.9,BIC=-5949.7)and lumbar spine BMD(R~2=0.233,AIC=-5809.3,BIC=-5803.0)than serum iron(femoral neck:R~2=0.251,AIC=-5935.4,BIC=-5929.1;Lumbar spine:R~2=0.227,AIC=-5769.8,BIC=-5763.5).It is the best model to reflect BMD.4.In the generalized Additive Model,we also use the individuals in aim one to see whether there has nonlinear relationship between iron intake,serum iron,serum ferritin and BMD in femoral neck and lumbar spine clearly.Conclusion:1.The bone mineral density of the subjects was affected by age,race and BMI,and there were significant differences in bone density among different parts.The bone mineral density of the head was the highest,and the bone density of the upper limb was the smallest.2.Serum iron and serum ferritin were negatively correlated with bone mineral density in femoral neck and lumbar spine,but there was no correlation between iron intake and bone mineral density.3.Serum ferritin is the best iron index among the three iron indexes,which has the best correlation with bone mineral density of femoral neck and lumbar spine.4.The negative correlation between iron and bone mineral density was affected by the site,but the bone mineral density of limbs and head was not affected by iron levels.