ECTS2013 Poster Presentations Osteoporosis: evaluation and imaging (31 abstracts)
1Center of Bone Diseases, Lausanne University Hospital, Lausanne, Switzerland; 2Lausanne University Hospital, Rheumatology, Lausanne, Switzerland; 3Lausanne University Hospital, Internal Medicine, Lausanne, Switzerland.
The aim of the study is to compare the performance of FRAX vs TBS adjusted FRAX using Leslie B et al.1 method to better identify women at high fracture risk. The OsteoLaus cohort (1500 women 5080 years living in Lausanne, CH) started in 2010. CRF for OP, FRAX, spine and hip BMD, VFA by DXA and TBS were recorded. Sensitivity and specificity in regard to vertebral fracture grade 2 and 3 has been calculated. Net reclassification improvement (NRI) had also been calculated. We included 911 women: mean age 65.2±7.9 year, BMI 25.7±4.4, mean spine BMD 0.931±0.163, TBS 1.289±0.100. As expected, correlation between BMD and site matched TBS is low (r2=0.16). Prevalence of VFx grade 2/3 and MOF are 7.5 and 15.0% respectively.
An incremental improvement in fracture identification was seen by using spine TBS in combination with FRAX. If validated in prospective cohorts, spine TBS may become clinically useful for enhancing fracture prediction from FRAX.
1. Leslie WD et al. Lumbar Spine TBS is a FRAX independent risk factor for fracture: the Manitoba BMD Cohort. ISCD Annual meeting 2013. Tampa, Florida Table 1.
Sensitivity (%) | Specificity (%) | |
Spine BMD (−2.5 T-score threshold) | 29.4 | 82.7 |
Spine TBS (−1.200 threshold) | 51.5 | 77.1 |
FRAX MOF (20% threshold) | 38.2 | 94.8 |
Spine TBS or FRAX MOF (20% thresholds) | 63.2 | 74.4 |
TBS adjusted FRAX All fracture (20% threshold) | 50.0 | 89.9 |
NRI for FRAX adjusted by TBS vs FRAX was +7.6% for VFx (P<0.001). |