Mild morphometric vertebral fractures predict vertebral fractures but not non-vertebral fractures.

Abstract

In this meta-analysis of the control arms of four phase 3 trials, mild vertebral fractures were a significant risk factor for future vertebral fractures but not for non-vertebral fracture. A prior vertebral fracture is a risk factor for future fracture that is commonly used as an eligibility criterion for treatment and in the assessment of fracture probability. The aim of this study was to determine the prognostic significance of a morphometric fracture according to the severity of fracture. We examined the control (placebo) treated arms of four phase 3 trials. Vertebral fracture status was graded at baseline in 7,623 women, and fracture outcomes were documented over the subsequent 20,000 patient-years. Fracture outcomes were characterised as a further vertebral fracture, a non-vertebral fracture or a clinical fracture (non-vertebral plus clinical vertebral fracture). The relative risk of fracture was computed from the merged β coefficients of each trial weighted according to the variance. Mild vertebral fractures were a significant risk factor for vertebral fractures [risk ratio (RR) = 2.17; 95% CI = 1.70-2.76] but were not associated with an increased risk of non-vertebral fractures (RR = 1.08; 95% CI = 0.86-1.36). Moderate/severe vertebral fractures were associated with a high risk of vertebral fractures (RR = 4.23; 95% CI = 3.58-5.00) and a moderate though significant increase in non-vertebral fracture risk (RR = 1.64; 95% CI = 1.38-1.94). Prior moderate/severe morphometric vertebral fractures are a strong and significant risk factor for future fracture. The presence of a mild vertebral fracture is of no significant prognostic value for non-vertebral fractures. These findings should temper the use of morphometric fractures in the assessment of risk and the design of phase 3 studies.