Prevalent vertebral deformities predict mortality and hospitalization in older women with low bone mass. Fracture Intervention Trial Research Group

To determine the relationship between prevalent vertebral deformities and the risk of mortality and hospitalization in older women with low bone mass.

A prospective cohort study.

Eleven clinical centers in the United States.

A total of 6459 community-dwelling women with low bone mass aged 55 to 81 participated in the Fracture Intervention Trial (FIT), a multicenter clinical trial of alendronate that enrolled women into one of two study arms based solely on the presence or absence of existing radiographic vertebral deformities. There were 2027 women with at least one vertebral deformity enrolled in the vertebral fracture arm of FIT and followed prospectively for an average of 2.9 years, whereas 4432 women with no vertebral deformity were enrolled in the clinical fracture arm of FIT and followed prospectively for an average of 4.2 years.

Determination of prevalent vertebral deformities on baseline lateral thoracic and lumbar spine radiographs was made at the coordinating center using a combination of radiographic morphometry by digitization and semiquantitative radiologic interpretation. Deaths were confirmed by obtaining copies of original death certificates of all participants who died. Episodes of hospitalization were captured through adverse event reporting; hospitalizations resulting solely from adverse events containing the words "fracture" or "trauma" were excluded from the analyses.

During the follow-up period, 122 women died, and 1676 women were hospitalized on at least one occasion for reasons not related solely to fracture. Compared with women without prevalent vertebral deformities, those women with prevalent deformities had higher risks of mortality (age- and treatment assignment-adjusted relative risk 1.60, 95% confidence interval (CI), 1.10-2.32) and hospitalization (age- and treatment assignment-adjusted relative risk 1.18, 95% CI, 1.06-1.31). In addition, further adjustment for other factors, including smoking status, physical activity, hypertension, coronary heart disease, obstructive lung disease, any fracture since the age of 50, health status, total hip BMD, and body mass index did not alter the association between prevalent vertebral deformities and risk of mortality substantially (multivariate relative risk 1.49, 95% CI, 1.05-2.21). Adjustment for all these factors and diabetes also did not change the relationship between prevalent vertebral deformities and hospitalization (multivariate relative risk 1.14, 95% CI, 1.02-1.27). Rates of mortality and hospitalization increased with increasing number of prevalent vertebral deformities (tests for trend P < .01).

Prevalent vertebral deformities in older women with low bone mass are associated with increased risks of mortality and hospitalization. Only a portion of this increased risk was explained by other known predictors of these outcomes.