Ten-year absolute fragility fracture risk and central DXA assessment: does hypovitaminosis D matter in menopausal women?

Mara Carsote; Valentin Radoi; Roxana Dusceac; Andreea Geleriu; Alexandra Mihai; Diana Paun; Rucsandra Danciulescu Miulescu; Catalina Poiana

doi:10.1530/boneabs.3.PP254

PP254

Prev Next

Section Contents Cite

Bone Abstracts (2014) 3 PP254 | DOI: 10.1530/boneabs.3.PP254

ECTS2014 Poster Presentations Osteoporosis: evaluation and imaging (43 abstracts)

Ten-year absolute fragility fracture risk and central DXA assessment: does hypovitaminosis D matter in menopausal women?

Mara Carsote ^1, , Valentin Radoi ² , Roxana Dusceac ¹ , Andreea Geleriu ¹ , Alexandra Mihai ¹ , Diana Paun ^1, , Rucsandra Danciulescu Miulescu ¹ & Catalina Poiana ^1,

Views

Author affiliations

¹I.Parhon, Bucharest, Romania; ²Davila UMPh, Bucharest, Romania.

Introduction: The FRAX model is a new method to assess fracture risk and this aspect is important in menopause. The DXA is still the golden standard but the relationship between these two instruments is complex.

Aim: We correlate the 10-year absolute fracture risk of hip (10-yH) and major fractures (10-yM) based on FRAX with or without bone mineral density (BMD) with lumbar DXA in menopause women, and we use a statistical model to find out whether the vitamin D status influences these correlations.

Materials and method: This is a cross-sectional study design. The inclusion criteria were: i) women in menopause for minimum 1 year; ii) central DXA scan (GE Lunar Prodigy) at Parhon National Institute of Endocrinology, Bucharest, Romania and iii) FRAX risk calculation (for Romania). The exclusion criteria were: i) anterior medication for osteoporosis or fragility fractures prevention and ii) active metabolic bone diseases. 25-Hydroxy vitamin D (25-OHD) was assessed (chemiluminescence; normal: 30–100 ng/ml).

The SPSS21 soft calculated linear regression, backward stepwise regression. Statistical significance (SS) was at P<0.05.

Results: 197 women (mean age of 57.7 years) were enrolled. The 25-OHD median was 15 ng/ml (5% had <30 ng/ml). Mean 10-yM was 4.5% and mean 10-yH was 1.12%. Mean lumbar BMD was 1.101±0.2 g/cm². The linear correlation coefficient was between lumbar BMD and 10-yM was: r=−0.25 respective, 10-yH r=−0.23 (P<0.005). Similar data were obtained when inputting neck BMD in FRAX calculation. The backward stepwise regression coefficients (adding 25-OHD) were for lumbar BMD and 10-yM r=0.308, r²=0.095, respective for 10-yH r=0.295, r²=0.087, and after adjusting for 25-OHF r=0.29, r²=0.089, respective r=0.285, r²=0.08 (P<0.05).