Is vertebroplasty really no more than an expensive placebo?


Ever since the simultaneous publication in the New England Journal of Medicine of studies from Mayo Clinic in Rochester, USA, and Monash University, Malvern, Australia, which questioned the effectiveness of vertebroplasty, specialists across the spine world have been hotly debating whether pain-relief obtained from vertebroplasty is merely the result of a “placebo” effect. Many clinicians are startled by the results of the studies, but remain positive about the procedure.

The Mayo Clinic’s randomised, controlled trial (RCT) compared vertebroplasty to a sham procedure in 131 patients in 11 centres in the USA, UK, and Australia. The second RCT, from Monash University, compared vertebroplasty to a sham operation in 78 patients at four centres in Australia.

Both studies found that relief of pain from vertebral compression fractures, as well as improvement in pain-related dysfunction were similar in patients treated with vertebroplasty and those treated with a sham procedure. “We found no beneficial effect of vertebroplasty over a sham procedure at one week, one, three, or six months among patients with painful osteoporotic vertebral fractures,” wrote Buchbinder et al.

While these results have left some specialists praising the quality of these studies and calling for more such rigorous clinical trials, particularly placebo-controlled, randomised trials for other interventions, others remain vigorous defenders of their own experiences with vertebroplasty.

“We are not saying that vertebroplasty does not work, because it somehow does,” says David F Kallmes, lead author of the study from Mayo Clinic. “But both sets of patients experienced significant improvements in pain and function a month following the procedure, whether they received cement injections or not. Improvements may be the result of local anaesthesia, sedation, patient expectations, or other factors.”

Rachelle Buchbinder, lead author of the study from Australia, told The Back Letter publication that personal anecdotal impressions can be very misleading. “The problem of misleading personal experience is why we need rigorous, placebo controlled trials,” she says.

Clinicians have greeted the results of both trials with mixed reactions

Gunnar Andersson, chair, spinal deformities, Rush University Medical Center, Chicago, USA and one of the editors-in-chief of Spinal News International, a sister publication of Interventional News, says, “Vertebroplasty is a helpful procedure for patients with painful vertebral fractures who do not respond to non-operative treatment. I personally prefer the kyphoplasty procedure which I think has the potential of providing added safety and a more consistent filling of the vertebral body with bone cement.”

Has he been influenced in any way by the studies in the NEJM? “My primary conclusion based on these studies, is that vertebroplasty should not be routinely performed in patients with chronic back pain and vertebral fractures. It is first necessary to determine whether or not there are other causes of pain. A second conclusion is that it becomes important to involve the patient in the decision before performing the vertebroplasty. The patient needs to know that the results are not uniformly good. Thus, it becomes important to have informed patient choice. Thirdly, the results have strengthened my preference for kyphoplasty,” he says.

Jean B Martin, diagnostic and interventional radiologist, Geneva University Hospital, Switzerland and also on the Spinal News International editorial board, was trained by Hervé Deramond. He says, “Since 1998, we have been performing around 600 vertebroplasty procedures per year. From the beginning, the promising results of vertebroplasty were there for all to see; I have seen patients who were obliged to take bed rest, due to pain, for more than three months, dancing in the room, hours after the procedure. I have never received so many thanks, and expressions of gratitude, since I started. It is a wonderful procedure for the patient when clinically adapted.

“These results published in NEJM do not change my position. Most of my colleagues involved in the clinical management of these patients have read the paper and are perplexed, but due to 15 years of practice and the evident clinical results seen in their own patients, they, too, have not changed their position and still have confidence in the technique. My opinion will not change, because I have seen the clinical effect with my own eyes,” he says.

Gerrit Bonacker, chief physician of the Praxisklinik Mittelhessen, Germany, agrees. When asked whether these studies in the NEJM had influenced him in any way, he says: “No. I am convinced that the treatment of vertebral fractures by vertebroplasty and kyphoplasty is superior to conservative treatment. But it is very important to have the right indication, that is, a painful and new vertebral body fracture. In the two trials published in the NEJM, this was not considered enough.”

However, Martin also acknowledges the importance of the studies. “Studies like this provoke reconsideration and rethinking of our medical attitude, which can only be helpful for patients. The most important issue here though, is the simple truth: We should not be treating images, but patients.”

Andersson finds that “This study eliminates some of the ‘hype’ surrounding the procedure. All patients with vertebral fractures should not have vertebroplasty and the results should not be presented to the patient as uniformly successful. Further, careful evaluation of the patient should occur before vertebroplasty is considered.”

Stephan Becker, Vienna, Austria, adds another perspective, “Both studies have revealed something unexpected, which is why the medical community is looking closely for bias, fault in the design or anything else. I congratulate the teams as both studies met the thresholds to be published in the NEJM, which shows their quality. But I seriously think that the studies have come a little too late, because opinion is now set and everybody has experience with the technique.”

Further studies needed

Experts are suggesting that one of the major limitations of the studies is the inclusion and exclusion criteria. While some speculate that the trials should have been directed towards individuals with more acute fractures, others say that the studies should have homed in on those patients with more severe pain.

“Although the inclusion criteria called for painful vertebral fractures, a large number of patients in both studies were found not to have vertebral fractures when screened for participation. This indicates to me that it was not the vertebral fracture that was the primary reason for referral, but rather the chronic back pain which could have a number of other sources,” says Andersson.

Another factor to consider, points out Martin, is that “This study presents a small group of patients treated by vertebroplasty versus sham, with no significant difference in follow-up. Only a small difference is observed at one month with a small drop of morphine treatment in the vertebroplasty group, and at three months with small difference in pain scores. No clinical data are notified and patients are analysed by precise and general questionnaires.”

Martin also told Spinal News International that it is common, after a vertebroplasty procedure, that the patient may complain that there is still pain. “It is up to the medical physician to evaluate that pain, find its origin and treat it. In one study, patient evaluation is performed at one week, but one of the usual transitory pains are muscle and periostal injury which can be really painful and last about 5–10 days, depending on the type and needle size,” he said.

Spine specialists also draw attention to the fact that the second evaluation in the study is at one month. At that time vertebroplasty is usually effective. In this study the difference observed in the two populations is not significant, but this is probably due to the fact that the questions are back-pain-specific, but not fracture-specific, they say.

The third evaluation at three months showed equivalent results, because many of the benign fractures, not needing vertebroplasty in the first place, will consolidate in six weeks by themselves.

“There are also other points to consider while evaluating the study,” said Martin. “Such as the exceptionally high complication rate described. Specifically, infection in the vertebroplasty group can be a confounding factor as such infections can be painful for months. Also rib fractures and second vertebral fractures can be painful for a long period of time, which are curiously higher in the sham group,” he said.

The need of the hour, Andersson feels, “Is determining which patients benefit versus which patients do not benefit from vertebroplasty. I also think we need to compare vertebroplasty to other alternatives such as kyphoplasty.”

Martin agrees: “We need a larger study with more precise questionnaires, focused on vertebral fractures and not generally back pain. A clinical evaluation has to include specific tests associated with fracture pain and load bearing. Other techniques, which alleviate arthritic and disc pain should be included in both groups, depending on clinical signs during the follow-up. There should be fewer exclusion criteria. In addition, kyphoplasty, with similar needle calibre to vertebroplasty, should be included especially when the fractures are recent according to clinical standards and the angle of the kyphosis or lordosis is largely modified.”

Bonacker says, “We need more studies that compare vertebroplasty to kyphoplasty. The Fracture Reduction Evaluation (FREE) study shows that the kyphoplasty is superior to non-surgical care. So why not compare the two methods with each other? I am looking forward to the results of two major studies, currently underway in the USA.”

Note: The opinions in this article are solely of the contributors and do not necessarily reflect the opinion of Interventional News.