Peripheral quantitative computed tomography (pQCT) measures are associated with prior low trauma fracture in men.

UNLABELLED

Peripheral quantitative computed tomography (pQCT) assesses bone quantity and quality, complementary to current standard practice, and has potential to improve prediction of fracture risk. This study explored whether pQCT parameters were associated with prior fracture in men and found a number of parameters to be associated, particularly at the radius.

PURPOSE

Peripheral quantitative computed tomography (pQCT) provides information about bone structure and density complementary to dual x-ray absorptiometry. This study aimed to determine which pQCT parameters are associated with prior fracture.

METHODS

Participants were men (n = 508, age 33-96 years) from the Geelong Osteoporosis Study. Parameters at 4% (n = 469) and 66% (n = 436) of radial length, and 4% (n = 449) and 66% (n = 437) of tibial length were acquired (pQCT XCT 2000, Stratec Medizintechnik, Pforzheim, Germany), and mean standardised. Low trauma prior fractures in adulthood (≥ age 20 years) were radiologically confirmed when possible. Cross-sectional associations between pQCT and fracture were tested using logistic regression adjusting for confounders.

RESULTS

Prior low trauma fractures were identified for 106 participants. Fracture was negatively associated with parameters at the 4% radius site: bone mass (adjusted OR = 0.67; 95%CI = 0.52-0.86), total density (OR = 0.61; 95%CI = 0.47-0.78), trabecular density (OR = 0.62; 95%CI = 0.48-0.79) and cortical subdensity (OR = 0.61; 95%CI = 0.47-0.77). At the 66% radius site, fracture was associated with total density (OR = 0.69; 95%CI = 0.55-0.87) and cortical thickness (OR = 0.68; 95%CI = 0.54-0.86). Fracture was associated with the ratio of the cortical area at the 66% site to the total area at the 4% site (OR = 0.74; 95%CI = 0.58-0.94). Prior fracture was negatively associated with parameters at the 4% tibial site: total density (OR = 0.67; 95%CI = 0.52-0.86), trabecular density (OR = 0.64; 95%CI = 0.50-0.82) and cortical subdensity (OR = 0.72; 95%CI = 0.56-0.92). Fracture was negatively associated with cortical density at the 66% site (OR = 0.74; 95%CI = 0.58-0.94), and the ratio of the cortical area at the 66% site to the total area at the 4% site (OR = 0.65; 95%CI = 0.46-0.91), but were attenuated in adjusted models. No other associations were identified.

CONCLUSION

Prior fracture was associated with parameters at both the radius and tibia. This study highlights key pQCT parameters that may aid in the prediction of fracture risk.