Searchable abstracts of presentations at key conferences on calcified tissues
Bone Abstracts (2016) 5 OC3.6 | DOI: 10.1530/boneabs.5.OC3.6

ECTS2016 Oral Communications Clinical trials, FGF-23 and focal osteoporosis (6 abstracts)

Focal osteoporosis associated with hip fracture involves both trabecular and cortical bone; a 3D cortical bone mapping study of cases and controls using clinical CT

Ken Poole 1 , Linda Skingle 1 , Andrew Gee 1 , Thomas Turmezei 1 , Fjola Johannesdottir 1 , Karen Blesic 1 , Collette Rose 1 , Madhavi Vindlacheruvu 1 , Simon Donnell 5 , Jan Vaculik 2 , Pavel Dungl 2 , Martin Horak 3 , Jan Stepan 2 , Jonathan Reeve 4 & Graham Treece 1


1University of Cambridge and Addenbrooke’s Hospital, Cambridge, UK; 2Charles University, Prague, Czech Republic; 3Homolka Hospital, Prague, Czech Republic; 4University of Oxford, Oxford, UK; 5University of East Anglia, Norwich, UK.


Focal cortical thinning and loss of trabecular structure in the proximal femur is associated with hip fracture. We analysed clinical computed tomography (CT) scans in cases and controls to explore their contributions to hip fracture in women.

We used cortical bone mapping (CBM) and statistical parametric mapping (SPM) after combining women with hip fracture from FEMCO and Prague Hip Joint in Trauma studies (n=138, 52 Trochanteric and 86 Femoral Neck) and comparing their bone parameters with 121 healthy age-matched female controls from previous Cambridge and Prague studies. Our aim was to determine how well focal measures of Cortical (Mass Surface Density, CMSD) and Trabecular bone (Endocortical trabecular density, ECTD), compared with areal BMD (aBMD, DXA-like from CT) in discriminating hip fractures from controls using ROC analysis. We aligned participant selection and CT scanning criteria across centres, and also modelled study site within the general linear model. An average single measure of 3D CMSD or ECTD was taken for each patient from previously determined bone mapping ROIs. We tested the ability of age, height and average measures of either aBMD, CMSD, ECTD or combined CMSD+ECTD to correctly discriminate fractures. The corresponding AUC values and 95% CI’s were calculated (Table 1).

Femoral neck and trochanteric fracture patients had different patterns of focal osteoporosis. As reflected in ROC analysis and odds ratios, the ability to effectively discriminate hip fractures from controls was dependent upon including both trabecular and cortical measurements. AUC for discrimination of hip fracture type was 0.794 for areal BMD, and 0.822 for 3D measures. These 3D CBM measures warrant testing in a prospective female cohort.

Table 1
aBMD AUC (95% CI)CMSD AUC (95% CI)ECTD AUC (95% CI)CMSD+ECTD AUC (95% CI)
All hip Fx0.79 (0.73–0.85)0.73 (0.67–0.79)0.83 (0.77–0.88)0.82 (0.76–0.87)
Neck Fx0.77 (0.70–0.83)0.79 (0.72–0.84)0.81 (0.73–0.85)0.84 (0.77–0.88)
Troch Fx0.74 (0.66–0.80)0.70 (0.60–0.77)0.80 (0.73–0.85)0.82 (0.75–0.87)

Volume 5

43rd Annual European Calcified Tissue Society Congress

Rome, Italy
14 May 2016 - 17 May 2016

European Calcified Tissue Society 

Browse other volumes

Article tools

My recent searches

No recent searches.