Article Text

Download PDFPDF
A4.1 Lumbar bone mineral density and long-term cardiovascular risk – a cross-sectional study
  1. Claudiu Popescu1,2,
  2. Alexandra Koşevoi-Tichie1,
  3. Denisa Predeţeanu1,2
  1. 1Internal Medicine and Rheumatology Department, “Sfanta Maria” Clinical Hospital, Bucharest
  2. 2“Carol Davila” University of Medicine and Pharmacy, Bucharest


Background and Objectives Osteoporosis and cardiovascular morbidity are strongly associated with age and menopause. The study aimed to evaluate the association of cardiovascular risk (CVR), estimated using SCORE charts, with BMD measured using lumbar dual X-ray absorptiometry (DXA).

Materials and Methods We included female in-patients, who underwent a lumbar DXA measurement between January-June 2013, which revealed a T-score ≤ -2.5. The correlations of normally distributed data were assessed using Pearson coefficients and t tests respectively. For non-normally distributed data we used Spearman coefficients and Mann Whitney tests respectively. Differences between qualitative data were evaluated using χ2 or Fisher tests where appropriate (SPSS v. 17). Statistical significance was stated if p < 0.05. The predictive value was assessed using multiple linear regression models. Sensitivity and specificity were studied using ROC analysis. All patients gave informed consent. The study was approved by the local ethics committee.

Results A total of 101 female patients fulfilled the inclusion criteria. All lumbar spine DXA measurements correlated significantly and negatively with the SCORE estimate (for example Z-score: r = - 0.468; p < 0.001). When partial correlations were calculated controlling for age, none of the above correlations were significant (p > 0.2). Instead, regardless of age, these bone variables correlated significantly and positively with height (for example LBM: r = 0.350; p < 0.001), weight (for example BMD: r = 0.347; p < 0.001) and with residence (in the sense that urban dwellers had significantly higher lumbar bone indices than rural patients; p < 0.001). A regression model including height, weight and residence predicted the lumbar T-score (F = 12.6; p < 0.001), explaining 28.7% of its variation (R2 = 0.287). The ROC area for lumbar bone mass and fragility fractures at the time of the DXA measurement was 0.795 (95% CI: 0.668 – 0.923; p < 0.001). The fracture threshold was 29.7 g, with 71.4% sensitivity and 82.9% specificity.

Conclusions Osteoporosis and cardiovascular morbidity are independent variables which evolve separately in the context of ageing. A prospective study using whole body bone indices and better cardiovascular (surrogate) end-points would strengthen this conclusion. Romanian urban female patients exhibit better lumbar DXA bone indices compared to rural women.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.