3-D DISTRIBUTION OF BONE PARAMETERS ACROSS THE DISTAL FEMUR FROM WEIGHT BEARING CT FOLLOWS PATTERNS ACCORDING TO JOINT SPACE NARROWING PHENOTYPE: A MULTICENTER OSTEOARTHRITIS STUDY INVESTIGATION

Abstract

Weight bearing CT (WBCT) technology has been increasingly used in imaging of knee osteoarthritis in recent years because of its ability to assess the weight bearing joint space. However, relatively little is known regarding the extent to which this cone beam technology can provide useful information on peri-articular bone beyond the subchondral regions. To explore relationships between WBCT-derived 3-D cortical bone thickness and trabecular bone attenuation at the distal femur with concurrent joint space narrowing phenotypes. WBCT imaging was obtained ancillary to the Multicenter Osteoarthritis Study at the 144-month visit. After semi-automatic segmentation, femoral cortical thickness (fCT) and trabecular attenuation (fTA) maps were created for each of 663 available distal femurs using cortical bone mapping. A template was registered to each individual distal femur and a 3-D statistical shape model created, followed by transfer of fCT and fTA distribution maps onto the template. A statistical parametric mapping (SPM) general linear model adjusted for age, sex, BMI, and the first 5 shape modes (controlling for effects of systematic misregistration) was used to test the dependence of fCT and fTA in turn on the experimental variables of concurrent radiographic medial and lateral OARSI joint space narrowing (JSN) grade. One knee from each participant was selected for inclusion, taking the side with worse compartmental baseline JSN grade or randomly if equal. 16 knees were excluded due to unachievable registration, with a final study set of a single knee from 386 individuals. 219 were female, mean ± SD age was 63.6 ± 9.6 yrs, mass 82.3 ± 17.7 kg, height 169 ± 9 cm, and BMI 28.5 ± 5.0 kg/m 2 . SPM significance ROIs according to baseline medial (left, figs 1 & 2) and lateral (right, figs 3 & 4) JSN phenotypes are shown for fCT (top row, figs 1 & 3) and fTA (bottom row, figs 2 & 4). SPM revealed significantly greater fCT by up to ∼0.1mm and fTA by up to ∼40 attenuation units (AU) for each increment in baseline JSN grade along the outer margin of the respective compartments (unmasked blue zones in each figure). Significantly greater fCT by up to ∼0.1mm for each increment in medial JSN grade also spread across the anterior aspect of the medial femoral condyle suggesting a wider regional relationship. 3-D distribution of fCT and fTA follow distinct patterns according to OARSI JSN phenotype, relating to recognised distributions of marginal osteophytosis and subchondral sclerosis. These results indicate that cone beam WBCT can quantify cardinal bony structural features of osteoarthritis and that their distribution is related to specific osteoarthritis OARSI JSN phenotypes. National Institutes of Health, University of Kansas (R01AR071648), University of Iowa. (U01AG18832), University of California-San Francisco (U01AG19069), Boston University (U01AG018820). NS is a consultant for Integra BioLife, Trice Medical and Pacira Biosciences. The authors would like to thank participants and staff of the MOST study. CORRESPONDENCE ADDRESS: tom@turmezei.com