Precise 3D Localisation of a Cortical Thinning Defect Associated With Femoral Neck Fracture in Life
Osteoporosis - Assessment
Bone Biomechanics and Quality (Basic)
Oral Presentations, Presentation Number: 1127
Session: Concurrent Oral Session 22: Methods of Osteoporosis Assessment
Sunday, September 18, 2011 2:15 PM - 2:30 PM, San Diego Convention Center, Room 31ABC
* Kenneth Poole, University of Cambridge, UNITED KINGDOM, Graham Treece, University of Cambridge, UNITED KINGDOM, Paul Mayhew, University of Cambridge, UNITED KINGDOM, Martin Horak, Bulovka Hospital, Czech Republic, Jan Stepan, Charles University, CZECH REPUBLIC, Andrew H Gee, University of Cambridge, United Kingdom
Thinning of the proximal femoral cortex with ageing and osteoporosis is considered a cause of fracture and a modifiable target for bone active treatment. We applied a novel 3D cortical thickness mapping technique to ordinary clinical CT scans in order to discover whether femoral cortical bone in patients with an acute hip fracture was thinner than femoral cortical bone from community dwelling controls without fracture. The technique allowed us to pinpoint systematic differences and visualise the magnitude of the differences with a colour map displayed on an average femur model.
In a case-control study, we analysed Siemens CT scans from 75 women volunteers with acute fracture and 75 age-matched controls from the Prague Study of Hip Joint in Trauma. We classified the fracture location as femoral neck or trochanteric before creating cortical thickness maps of the unfractured contralateral hip. After non-rigid registration to an average femur shape and statistical parametric mapping, we were able to pinpoint statistically significant foci of cortical thinning associated with each fracture type.
The cortex was approximately 10% thinner in fracture cases than controls throughout the femur. More remarkably, there were several distinct foci of statistically significant (p<0.02) thinning of up to 30% which coincided anatomically with stereotypical fracture sites (femoral neck or trochanteric, fig. 1). The focal thinning of the head-neck junction associated with femoral neck fracture (fig. 1a) also coincided with the focus of peak tensile stress in Cristofolini's cadaveric and FE models of spontaneous femoral neck fracture (J Biomech 2007). Taken alongside the 3D density analyses of the whole hip of Li et al (Bone 2009), our results suggest that the DXA 'femoral neck' box ROI could eventually be replaced by more specific, fracture-critical ROIs in risk estimation, diagnostic and treatment strategies.
* Presenting Authors(s): Kenneth Poole, University of Cambridge, UNITED KINGDOM