Vertebroplasty proves no better for pain than placebo in two groundbreaking trials

Interventional radiologists are coming to grips with the implications of two groundbreaking clinical trials indicating that vertebroplasty relieves pain from osteoporotic vertebral fractures no better than a sham version of the procedure.

Results from Mayo Clinic in Rochester, MN, and Cabrini Medical Center in Malvern, Australia, were published simultaneously Aug. 5 in the New England Journal of Medicine.

Unlike past trials that examined only treated patients, the Mayo and Cabrini studies compared results from patients who underwent vertebroplasty with patients who were randomly selected to receive sham interventions.

Their findings suggest that the clinical benefit of vertebroplasty may stem from a placebo effect rather than the therapeutic impact of cement injected to stabilize fractured vertebral segments.

In the Mayo trial, 131 patients with painful osteoporotic spinal fractures received either vertebroplasty with a standard administration of polymethylmethacrylate cement or a sham procedure without cement injection. Dr. David F. Kallmes, a professor of radiology, and colleagues at Mayo Clinic and 10 collaborating facilities in the U.S., U.K., and Australia found that both groups experienced an immediate improvement in disability and pain following intervention. No significant differences were observed between the two groups at one and three months after the procedures.

First author Rachelle Buchbinder, Ph.D., and colleagues in Australia found that 38 patients who underwent vertebroplasty for osteoporotic fractures experienced about the same level of pain relief and quality-of-life improvement as 40 patients who received a sham procedure.

The results explain why vertebroplasty is effective, a question that previous trials did not answer, Kallmes said in an interview. The new studies led researchers to reject the assumption that injected cement is the source of pain relief. A lidocaine anesthetic drug used for both sets of patients in both studies may confer a benefit, but placebo effect is the most plausible reason why patients felt less pain after the procedures, he said.

Interventional radiologists initially challenged these findings, mainly on methodological grounds. Dr. J. Kevin McGraw, section head of vascular and interventional radiology at Riverside Methodist Hospital in Columbia, OH, expressed concern that the Australian study was susceptible to patient selection bias because more than 30% of patients who met its inclusion criteria decided not to participate.

McGraw was involved with the early stage of the Mayo study as well. He recalled that many patients refused to volunteer because they feared being selected for a sham procedure that would deny them pain treatment. Such experience explains why Mayo and Australian researchers needed five and four years, respectively, to recruit volunteers for their trials. Both efforts ultimately involved small numbers of patients, he said.

"We need larger studies and larger patient numbers," he said.

Dr. Allan L. Brook, director of interventional neuroradiology at Montefiore Medical Center in New York City, repeated McGraw's concerns about patient selection and trial population size. From his perspective, the level of pain experienced by patients in the two studies was also substantially less severe than pain levels described by his patients before treatment.

The two small-scale trials should not be considered statistically powerful enough to set a new standard for vertebroplasty's efficacy, he said.

"It is hard to believe 78 patients in one study and 131 in another can match the positive experience of 100,000 patients and millions worldwide who are successfully treated with vertebroplasty augmentation,' he said.

However, Kallmes said the new findings do not contradict previously published trials. The pain relief experiences of patients who received vertebroplasty in the Mayo and Australian studies are similar to findings from many cohort studies in the medical literature.

"Everybody is right," he said. "It is not that people are not getting better with vertebroplasty-they are," he said. "The problem is that we saw the same benefit in the placebo group."

The findings suggest to Kallmes that vertebroplasty and kyphoplasty, a similar interventional procedure, are overprescribed. He agreed with his critics that larger trials are needed.

In the meantime, Kallmes has adjusted his practice to account for the new findings. He advises patients that the benefits of vertebroplasty do not necessarily relate to injected cement and that the underlying mechanism for the procedure's success is not as well understood as previously thought.

The results offer a reason to encourage patients to continue medical therapy for their pain as long as possible, he said. For patients in severe pain and those who are given high doses of narcotics for their pain, augmentation with vertebroplasty or kyphoplasty should be strongly considered.

An estimated 180,000 vertebroplasty and kyphoplasty interventional procedures are performed annually in the U.S. The Society of Interventional Radiology issued a position paper in 2007 attesting to their safety and efficacy. It recommended reimbursement.