Vertebroplasty

Abstract

Painful osteoporotic vertebral compression fractures are a common source of morbidity in the elderly, and are becoming even more prevalent with the aging population. Both vertebroplasty and kyphoplasty are common interventions to manage pain and shorten recovery time, but limited evidence exists to support their use. Recently, we highlighted a multicenter randomized controlled trial by Wardlaw et al. (Lancet 373: 1016–1024, 2009) suggesting that kyphoplasty resulted in earlier pain and functional improvements at one month. However, we noted that the study was non-blinded and industry funded, and also that general usage of this procedure may not be indicated as the symptoms are often self-limited. The first two blinded, randomized, placebo-controlled trials of vertebroplasty have now been reported in the New England Journal of Medicine (NEJM 361;6:557–579), and they do not support a beneficial effect of vertebroplasty as compared with a sham procedure in patients with painful osteoporotic vertebral fractures. Buchbinder et al. examined pain, physical functioning, quality of life, and perceived improvement at 1 week and at 1, 3, and 6 months in 78 participants. Vertebroplasty did not result in a significant advantage in any measured outcome at any time point. These findings are at odds with most, but not all previous reports, and are corroborated by the subsequent reported study. Kallmes et al. also investigated vertebroplasty versus a simulated procedure in 131 patients. Their outcome measures also included disability and pain at 3, 14, 30, and 90 days. Unlike the previous report, patients were permitted to cross over to the other study group after 1 month. Again, improvements in pain and disability were similar in both groups regardless of treatment. Although a trend toward a higher rate of clinically meaningful improvement in pain was noted in the vertebroplasty group at 1 month, this was not statistically significant. Interestingly it was noted that significantly more patients in the control group crossed over to the vertebroplasty group at 3 months, but a difference in pain outcomes was not seen to explain this phenomenon. This raises the possibility that a subgroup of patients exists that may be benefited by vertebroplasty. In his commentary, Weinstein (NEJM 361;6:619–621) also noted a possible bias since more patients than predicted were able to correctly guess which treatment they had received, there was a potential benefit in understanding the treatment effect in those who guessed their treatment accurately. Overall these results are somewhat surprising, given that they contradict numerous earlier reports. It had even been suggested that performing a randomized placebo-controlled trial of vertebroplasty versus simulated procedure was unnecessary given the many positive studies in the literature. Consistent across all studies has been the favorable natural history of this condition. In both of the reports discussed above, there was a significant reduction in overall pain and disability after treatment, irrespective of treatment group. This may contribute towards difficulty interpreting prior non-placebo controlled trials. As well, there is a significant placebo effect which may be enhanced by the usage of an invasive therapy such as vertebroplasty, and it has been suggested as well that the injection of local anesthetic during both treatment groups may in itself constitute a treatment modality, but this is unlikely to be a lasting effect. In addition, these studies did not examine specifically the population of patients who had failed extensive conservative treatment measures, which is more likely a primary indication for treatment by neurosurgeons. Both studies investigated patients with a mean pain duration of 12 to 20 weeks, and the Kallmes study excluded patients with more than one year of symptoms. For this population cement augmentation may offer the only reasonable possibility for pain relief. Thus, although these studies do not support the general use of vertebroplasty for painful osteoporotic vertebral compression fractures, it is possible that subgroups exist for which this would be a useful treatment, and further studies are needed to clarify this point. ANN M. PARR, M.D. MICHAEL Y. WANG, M.D.