Incident Vertebral Fractures in Patients not Undergoing Vertebroplasty
Received 28 April 2008; received in revised form 4 February 2010; accepted 11 February 2010. published online 26 April 2010.
Purpose
To determine whether vertebroplasty increases the risk of new (ie, incident) vertebral fractures by comparing the rates of incident fractures among a group of patients with painful vertebral fractures who did not undergo vertebroplasty with a group of patients who did.
Materials and Methods
A retrospective chart review was performed to identify new-onset fractures after initial vertebroplasty evaluation in two groups, including patients who underwent vertebroplasty within 1 week of initial evaluation (group 1) and those who did not (group 2). Group 2 was further limited to patients with acute or subacute prevalent fractures (group 2A) after exclusion of those with exclusively chronic prevalent fractures. Survival analyses were performed to compare time to diagnosis and frequency of incident fractures in these three groups of patients.
Results
Group 1 included 269 patients, group 2 included 107 patients, and group 2A included 82 patients. Compared with group 2, incident fractures in group 1 occurred significantly earlier (log-rank statistic, 0.01) and more frequently (hazard ratio, 2.9; 95% CI, 1.2–8.4).
Conclusions
Among patients with acute or subacute vertebral fractures presenting for consideration of vertebroplasty, those who undergo the procedure experience more and earlier incident fractures than those who do not, but the observed differences failed to reach statistical significance. Further work is needed in this area to better define relative risks of incident fracture between treated and nontreated patients.
Department of Radiology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905
Address correspondence to D.F.K.
D.F.K. is a paid consultant for Bone Support (Chanhassen, Minnesota) and has received funding for research from Arthrocare (Sunnyvale, California), Cardinal (Dublin, Ohio), and Stryker (Kalamazoo, Michigan). Neither of the other authors has identified a conflict of interest.
ABSTRACT