| Background Osteoarthritis(OA)is the most common form of arthritis.With the increased aging process in the population,the prevalence and incidence of osteoarthritis(OA)present a markedly increased trend.Knee OA is one of the most commonly affected joints and the leading cause of chronic pain and disability in the elderly.OA confers a huge burden on both individuals and health economies.The clinical diagnosis of knee OA is mainly dependent on X-ray,which is not sensitive to detect early abnormal joint structural changes of OA.With the extensive application of magnetic resonance imaging(MRI),early joint structure changes in OA are gradually been recognized.Osteoporosis is an another age-related skeletal degenerative disease,which are affected by many factors that is shared by OA.The impact of bone density on OA is still controversial despite decades of research about OA and OP.Objective The aims of this study were to examine related factors of systemic bone mineral density,and the associations between systemic bone mineral density and knee structural changes in paients with knee OA.To further understand the role of systemic bone mineral density in knee OA and provide a scientific basis for the prevention and treatment of knee OA.Methods 206 OA patients who met the criteria of symptomatic knee OA and agreed to participate in the follow-up study were enrolled from Department of Rheumatology and Immunology in the First Affiliated Hospital of Anhui Medical University.This study was conducted from January 2012 to November 2013.14 subjects were excluded from this study due to the lack of relevant examinations,leaving 192 subjects in the current study.Questionnaires were filled by patients.Heights and weights were measured and body mass index(BMI)was calculated.The dual-energy X-ray absorptiometry(DXA)was used to measure total body,hip and spine BMD(L1-L4),trunk and leg lean mass and fat mass.Cartilage defects were assessed on MRI T1-weighted fat suppression fast spin echo,using the software program Osiri X.Bone marrow lesions(BMLs)were evaluated on T2-weighted fat suppression fast spin echo.Results 1.Univariable linear regression analyses showed that age,female sex,menopause period,pregnancy,trunk and leg lean mass had associations with total T score,all P <0.05.Multivariable linear regression analyses showed that age(?:-0.056,95%CI:-0.072,0.041),female sex(?:-0.760,95%CI: 1.312,3.650)and menopause period(?:-0.075,-0.112,-0.039)were significantly and negatively associated with total T score(P<0.05).There were significant and positive associations between pregnancy(?:1.163,95%CI: 0.152,2.174),trunk lean mass(?:0.061,95%CI: 0.014,0.108),leg lean mass(?:0.141,95%CI: 0.077,0.205)and total T score(P<0.05).2.Univariable linear regression analyses showed that age,female sex,menopause period,pregnancy,trunk and leg lean mass had associations with total body BMD,all P<0.05.Multivariable linear regression analyses showed that age(?:-0.007,95%CI:-0.009,-0.005),female sex(?:-0.104,95%CI:-0.151,-0.058)and menopause period(?:-0.010,95%CI:-0.014,-0.005)were significantly and negatively associated with total body BMD(P<0.05).There were significant and positive associations between pregnancy(?: 0.150,95%CI: 0.019,0.281),trunk lean mass(?: 0.007,95%CI: 0.001,0.013),leg lean mass(?:0.017,95%CI: 0.009,0.025)and BMD at total body(P<0.05).3.Univariable linear regression analyses showed that age,female sex,menopause period,pregnancy,trunk and leg lean mass had associations with hip BMD,all P<0.05.Multivariable linear regression analyses showed that age(?:-0.006,95%CI: 0.967,1.367),female sex(?:-0.060,95%CI:-0.119,0.000)and menopause period(?:-0.006,95%CI:-0.012,0.000)were significantly and negatively associated with hip BMD(P<0.05);There were significant and positive associations between pregnancy(?: 0.225,95%CI: 0.043,0.408),leg lean mass(?:0.018,95%CI: 0.008,0.028)and BMD at hip(P<0.05).4.Univariable linear regression analyses showed that age,female sex,menopause period,pregnancy,trunk and leg lean mass have associations with spine BMD,all P<0.05.Multivariable linear regression analyses showed that age(?:-0.005,95%CI:-0.007,-0.002),female sex(?:-0.094,95%CI:-0.157,-0.030)and menopause period(?:-0.008,95%CI:-0.014,-0.002)were significantly and negatively associated with spine BMD P<0.05.There were significant and positive associations between BMI(?: 0.010,95%CI: 0.004,0.016),pregnancy(?: 0.324,95%CI: 0.142,0.505),trunk lean mass(?:0.009,95%CI: 0.001,0.016),leg lean mass(?: 0.024,95%CI: 0.013,0.035)and BMD at total body,hip and spine(P<0.05).5.Relationship between hip BMD and joint structural changes: total hip BMD was associated with cartilage defects at medial tibial and lateral femoral sites in unadjusted analyses,all P<0.05.After adjustment for multiple confounders,total hip BMD was significantly and negatively associated with cartilage defects at medial tibial(OR:0.61,CI:0.40,0.91)and lateral femoral(OR:0.59,CI:0.40,0.85)sites,all P <0.05;Total hip BMD was associated with BMLs at lateral tibial sites in unadjusted analyses,P<0.05. After adjustment for multiple confounders,hip BMD was significantly and negatively associated with BMLs at medial tibial(OR:0.62,CI:0.39,0.98),lateral tibial(OR:0.62,CI:0.33,0.99)sites,all P <0.05.6.Relationship between spine BMD and joint structural changes: spine BMD was associated with medial tibial cartilage defects before adjustment,P<0.05.After adjustment for multiple confounders,the association between hip BMD and medial tibial cartilage defects became of borderline significance,but remained negative significant for cartilage defects at lateral tibial sites(OR:0.58,CI:0.38,0.89),P<0.05.In univariable analyses,spine BMD was associated with lateral tibial BMLs,P<0.05;however,this association became non-significant after adjustment for confounders.7.Relationship between total body BMD and joint structural changes: total body BMD was associated with lateral femoral cartilage defects before adjustment,P<0.05.After adjustment for multiple confounders,total body BMD was significantly and negatively associated with cartilage defects at lateral femoral(OR: 0.70,CI:0.49,0.99),P<0.05.In univariable analyses,total body BMD was associated with medial tibial and lateral tibial BMLs,P<0.05;however,these associations became non-significant after adjustment for confounders.Conclusions1.While age,female sex and menopause may be risk factors for BMD,BMI,pregnancy and lean mass may be protective on BMD in patients with knee OA.2.BMD particularly at the hip was negatively associated with knee cartilage defects and BMLs,suggesting high level of systemic BMD may play a protective role against joint structural changes in knee OA.These findings needed to be confirmed by longitudinal studies. |
PDF Full Text Request