Cement Injections for Spinal Fractures Are Not Effective

About 750,000 older people each year in the USA suffer compression fractures of their vertebrae, usually related to advancing osteoporosis. The usual treatment – bed rest, pain medications, and a brace – are not very satisfactory. About 20 years ago, the injection of a cement-like substance into the fracture itself was introduced; this was called vertebroplasty, and it was intended to provide immediate stability and relief of pain. The number of such procedures done in the USA has doubled during the last 6 years. However, there have been no blinded or placebo-controlled randomized trials of the procedure done, so the effectiveness has not been proven. Now two such studies have been reported in the New England Journal of Medicine, with negative results.

The first study was done in Australia. Of 71 participants who completed the study, 35 received vertebroplasty (using polymethylmethacrylate injections) and 36 had a sham procedure. It was double-blind – i.e. neither the patients or those who were evaluating them knew which procedure they had had. Follow-up exams done at one week, and at 1, 3, and 6 months, with the level of pain at 3 months, using a 10-point scale, as the ‘primary outcome’.

Pain was reduced in both groups at each follow-up visit, as was pain at night pain, at rest, physical functioning, quality of life scores, and patient-perceived improvement. However, there were no beneficial effects of vertebroplasty over the sham procedure at any time up to 6 months.

The second study was done in several centers in the USA, UK, and Australia. The design was similar to the first study, but evaluation of results was done at one month, with patients being allowed to cross over to the other study group after one month if they wished.

There were 68 patients with vertebroplasties and 63 sham procedures. After one month there were no significant differences in disability scores or average pain intensity during the previous 24 hours, although both groups reported immediate improvements in pain and disability after their procedures. A small, non-significant increased rate of pain improvement at one month was reported in the vertebroplasty group; 64% reported this, compared 48% in the control group. After 3 months, more vertebroplasty patients had crossed over into the other group – 43% - than the other way around – 12%; this difference was statistically significant.

Injection of polymethylmethacrylate is not without its risks. Rarely, soft-tissue damage, nerve root pain and compression are the result of leakage, and, extremely rarely, pulmonary embolism has been reported. In view of these rare but possible risks and the absence of clear benefit to the patient, the use of vertebroplasty should most probably be abandoned. This represents a good example of President Obama’s desire for using evidence from well-designed clinical studies to evaluate continued use of popular but untested therapies.