Predictors of Successful Palliation of Compression Fractures with Vertebral Augmentation: Single-center Experience of 525 Cases
Received 16 September 2008; received in revised form 21 January 2009; accepted 26 January 2009.
Purpose
To determine the effectiveness of vertebral augmentation in relieving pain, differences in pain relief outcomes based on procedure type were investigated. Variables that potentially influence outcomes were identified.
Materials and Methods
A database of 525 cases (740 levels) treated for compression fractures with vertebroplasty, kyphoplasty, or S1-level sacroplasty was compiled. Average age was 75 years ± 12, and 72.4% of patients were female. Variables evaluated included age, sex, fracture etiology, procedure type, vertebral level treated, number of levels treated per procedure, and technical approach. Outcomes were assessed by a binary system of “responders” (ie, patients with improvement/resolution of pain) versus “non-responders” (ie, those with no change/worsening of pain) and with a four-level pain scale (1, pain resolution; 2, pain improvement; 3, no change; 4, worse pain) retrospectively applied from medical records. Univariate and multivariate analyses determined outcomes.
Results
Four-hundred and sixty-seven patients (89%) showed a response to treatment: 40% had pain resolution and 49% had pain improvement. Multivariate analysis showed that women and older patients had greater odds of being responders (odds ratios [ORs], 0.56 and 0.98, respectively; P = .016 and P = .048, respectively). Patients without cancer (OR, 1.60; P = .012) and women (OR, 2.05; P = .0002) were more likely to experience pain resolution. Increasing numbers of levels treated per case were associated with decreased odds of pain resolution (OR, 0.69; P = .0081). Sex and number of levels treated were independently predictive of pain scale outcomes (ORs, 2.0 and 0.71, respectively; P = .0003 and P = .015).
Conclusions
Vertebral augmentation procedures provide pain relief for a majority of patients regardless of underlying fracture etiology. There was no difference in pain outcomes among procedure types. Age and sex may be predictive of pain outcomes.
aDepartment of Internal Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
bDepartment of Radiology/Division of Neurointerventional Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
cDepartment of Radiation Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
Address correspondence to R.M.J., 9 Hawthorne Pl., Unit 6F, Boston, MA 02114
J.A.H. is a consultant for and minor shareholder in Cardinal Health (Dublin, Ohio) and Medtronic (Minneapolis, Minnesota). None of the other authors have identified a conflict of interest.
ABSTRACT