Determinants of fracture risk among individuals with spinal cord injury: a case control study
MetadataShow full item record
Background/Objectives: Low areal bone mineral density (aBMD) of the hip and knee region has been associated with fracture risk in individuals with SCI; however the contribution of bone micro-architecture to fracture risk has not been evaluated. The primary objective of this study was to determine whether a relationship exists between indices of bone strength (aBMD at the distal femur and proximal tibia; trabecular vBMD; average hole size, HA; cortical thickness, CTh; buckling ratio, BR; cross-sectional moment of inertia, CSMI; and polar moment of inertia, PMI) and potential fracture risk factors (gender, age, bisphosphonate use, time post-injury, fractures, and completeness of injury). The secondary objectives were to 1) determine whether indices of bone strength can discriminate between SCI patients with and without fragility fractures; 2) determine if these indices of bone strength correlate with the number of fractures sustained; and 3) determine the proportion of individuals with SCI who have a trabecular vBMD at the ultra-distal tibia that is below 72mg/cm3. Materials and Methods: A nested case-control study was performed. Forty seven men (n=33) and women (n=14) with chronic SCI (C2-T12 AIS A-D) with a duration of paralysis of at least two years were included in this study. Subjects with SCI were questioned about the cause, location, and time of the lower extremity fragility fractures. Fracture presence was verified by x-rays. aBMD of the distal femur and proximal tibia were determined using dual energy x-ray absorptiometry (DXA). Trabecular volumetric bone mineral density (vBMD) and HA were measured at 4% of the tibia length, and CTh, BR, CSMI, and PMI were measured at 66% of the tibia length of individuals with chronic SCI using peripheral quantitative computed tomography (pQCT). Linear and multiple regression models were used to determine significant correlates (age, gender, completeness of injury, duration of injury, bisphosphonate use, and fractures) of indices of bone strength, while logistic regression was used to assess the relationship between indices of bone strength and fragility fractures. To assess the relationship between multiple fragility fractures and indices of bone strength, a poisson regression analysis was performed. Results: Risk factors found to be related to the indices of bone strength include gender, completeness of injury, duration of injury, bisphosponate use, and prior fractures. An increase in HA (OR=1.081, 95% CI=1.001-1.166, p=0.0470), a decrease in aBMD in the distal femur (OR=0.988, 95% CI=0.978-0.998, p=0.0226), and a decrease in CSMI (OR=0.098, 95% CI=0.012-0.838), p=0.0338) were associated with fractures. Fractures were not associated with aBMD at the proximal tibia, trabecular vBMD, CTh, or BR. The poisson regression model predicting the number of fragility fractures sustained among individuals with chronic SCI from aBMD, vBMD, HA, CTh, CSMI, PMI, and BR were each statistically significant. Finally, only 7.7% of our population had a trabecular vBMD fracture threshold of less than or equal to 72mg/cm3. We found a trabecular vBMD fracture breaking point of approximately 126mg/cm3 and 115mg/cm3 at the ultra distal tibia in individuals with complete and incomplete SCI, respectively. Conclusion: Specific bone strength measures, specifically aBMD at the distal femur, HA, and CSMI are associated with fracture risk and may improve our ability to identify individuals with SCI at high risk of fracture. Larger population based studies are needed to determine the most appropriate risk factors that contribute to bone loss and understand the role and importance of these and other indices of bone strength on skeletal fragility in individuals with SCI.