Agreement between fragility fracture risk assessment algorithms as applied to adults with chronic spinal cord injury.

STUDY DESIGN: Cross-sectional.

OBJECTIVES: The objective of the study was to determine and report agreement in fracture risk stratification of adults with spinal cord injury (SCI) using (1) Canadian Association of Radiologists and Osteoporosis Canada (CAROC) and Canadian Fracture Risk Assessment (FRAX) tools with and without areal bone mineral density (aBMD) and (2) SCI-specific fracture thresholds.

SETTING: Tertiary rehabilitation center, Ontario, Canada.

METHODS: Community-dwelling adults with chronic SCI (n=90, C2-T12, AIS A-D) consented to participation. Femoral neck aBMD values determined 10-year fracture risk (CAROC and FRAX). Knee-region aBMD and distal tibia volumetric BMD values were compared to SCI-specific fracture thresholds. Agreements between CAROC and FRAX risk stratifications, and between fracture threshold risk stratification, were assessed using prevalence- and bias-adjusted Kappa statistics (PABAK).

RESULTS: CAROC and FRAX assessment tools showed moderate agreement for post-menopausal women (PABAK=0.56, 95% confidence interval (CI): 0.27, 0.84) and men aged ⩾50 years (PABAK=0.51, 95% CI: 0.34, 0.67), with poor agreement for young men and pre-menopausal women (PABAK⩽0). Excellent agreement was evident between FRAX with and without aBMD in young adults and in those with motor incomplete injury (PABAK=0.86-0.92). In other subgroups, agreement ranged from moderate to substantial (PABAK=0.41-0.73). SCI-specific fracture thresholds (Eser versus Garland) showed poor agreement (PABAK⩽0).

CONCLUSION: Fracture risk estimates among individuals with SCI vary substantially with the risk assessment tool. Use of SCI-specific risk factors to identify patients with high fracture risk is recommended until a validated SCI-specific tool for predicting fracture risk is developed.