Comparison of vertebroplasty and balloon kyphoplasty for treatment of vertebral compression fractures: a meta-analysis of the literature

BACKGROUND CONTEXT: Previous investigators have reported on benefits and risks associated with vertebroplasty and kyphoplasty, but there are limited comparison data available. Additionally, much of the data is from retrospective studies and case series. PURPOSE: The purpose of this study is to review the literature and perform a meta-analysis of pain relief and risk of complications associated with vertebroplasty versus kyphoplasty. STUDY DESIGN: A meta-analysis of the literature on effectiveness of pain control and risk of complications after vertebroplasty versus balloon kyphoplasty. Outcomes measures include visual analog scale and complications. METHODS: A comprehensive review of the literature was performed. All studies providing information on pain relief and complications were included. Preoperative, postoperative, and change in visual analog scale (VAS) scores were tabulated. Data were analyzed to identify if a significant improvement in the VAS score occurred. Changes in the VAS scores were compared for vertebroplasty and kyphoplasty to determine if there was a significant difference. RESULTS: A total of 1,036 abstracts were identified. Of these, 168 studies met the inclusion criteria. Mean pre- and postoperative VAS scores for vertebroplasty were 8.36 and 2.68, respectively, with a mean change of 5.68 (p<.001). The mean pre- and postoperative VAS scores for kyphoplasty were 8.06 and 3.46, respectively, with a mean change of 4.60 (p<.001). There was statistically greater improvement found with vertebroplasty versus kyphoplasty (p<.001). The risk of new fracture was 17.9% with vertebroplasty versus 14.1% with kyphoplasty (p<.01). The risk of cement leak was 19.7% with vertebroplasty versus 7.0% with kyphoplasty (p<.001). CONCLUSIONS: Both vertebroplasty and kyphoplasty provided significant improvement in VAS pain scores. Vertebroplasty had a significantly greater improvement in pain scores but also had statistically greater risk of cement leakage and new fracture. (C) 2008 Elsevier Inc. All rights reserved.