Vertebroplasty — Sorting Things Out When Studies Conflict With Clinical Experience
By Dan Harvey
Vol. 11 No. 6 P. 24
Two recent, high-profile studies raised more questions about vertebroplasty than they were designed to answer. Their results contradicted clinical experience and surprised the medical community. The studies, conducted nearly simultaneously and a world apart—one at the Mayo Clinic in Rochester, Minn., and the other in Australia—produced similar results. The trials, reported in the August 9, 2009, issue of The New England Journal of Medicine (NEJM), found that vertebroplasty for pain relief performs no better than sham procedures using placebos and local anesthesia. The results were so surprising that the lead researcher of the Mayo Clinic trial even felt compelled to question his study’s data.
“I simply couldn’t believe what I was seeing,” says David F. Kallmes, MD, a Mayo Clinic interventional radiologist.
Even beyond the results, the studies were significant in one important aspect: They were the first randomized, double-blinded trials designed to compare vertebroplasty with sham procedures and placebos. Patients who participated in the studies experienced modest pain relief whether they underwent vertebroplasty or “dummy” injections.
When Kallmes learned that Australian researchers achieved similar results, he felt more confident about his own study’s conclusions, but the data seemed to defy experience.
“It’s one of those situations that just doesn’t seem to make sense and produces cognitive dissonance,” says R. Douglas Orr, MD, of the Cleveland Clinic’s Center for Spine Health in its Neurological Institute. “Previously, few had felt the need to question whether the procedure worked; results were that dramatic.”
Kallmes and his colleagues weren’t seeking to debunk the value of vertebroplasty. They sought to better comprehend the role of cement injections in altering pain and pain-related dysfunction related to at least three osteoporotic vertebral fractures. If anything, Kallmes believed the research would put to rest doubts about vertebroplasty’s efficacy and thought the study might raise some questions. “Obviously, that’s exactly what happened,” he says.
“It flies in the face of what most people experience,” says James F. Benenati, MD, president of the Society of Interventional Radiology (SIR) and medical director of the Peripheral Vascular Laboratory at Baptist Cardiac & Vascular Institute in Miami. “But I don’t think the studies truly represent what happens in clinical practice.”
The two studies, Benenati says, point to the fact that more validation is needed. “Much has been taken for granted. Consensus indicates efficacy. But we now live in a world that requires high-level evidence for FDA approval and reimbursement. As such, I credit the authors,” he says.
A minimally invasive procedure typically performed by interventional radiologists armed with imaging guidance, vertebroplasty involves injecting medical-grade bone cement into a vertebra with a compression fracture. The procedure has served many patients well, especially those for whom treatments such as analgesics, narcotics, conventional therapy, and surgery didn’t work. The injected cement stabilizes the collapsed vertebra; some liken the procedure to a dentist putting a filling in a tooth cavity. Vertebroplasty use has expanded rapidly in the past decade despite its expense and risks such as infection and additional spinal fractures. Many patients consider the expense and risks good trade-offs for achieving relief from incessant pain and restricted movement. Physicians often witness an almost immediate transformation after the treatment.
“A patient bedridden by pain and physical disability arises from their bed about four hours after the procedure,” says Orr.
But the NEJM studies contradict anecdotal and clinical experience. In a column published in the January issue of the Cleveland Clinic Journal of Medicine, Orr advanced the notion of cognitive dissonance related to the conflicting evidence. On the one side is clinical experience, where many physicians performing the procedure have witnessed dramatic results. “Most patients report significant improvement in pain immediately after the procedure,” wrote Orr. He also notes that such immediate relief is not common in spinal procedures. “In spine surgery, few procedures give such dramatic results with so few complications,” Orr says.
On the other side are the recent studies that, as Orr indicates, called into question “the efficacy of vertebroplasty for treating osteoporotic vertebral compression fractures and have led many of us to question many of our assumptions, not only about vertebroplasty, but also evidence-based medicine.”
In 2002, after securing funding from the National Institute of Arthritis and Musculoskeletal and Skin Diseases, Kallmes began the collaboration to compare vertebroplasty with a placebo. His study eventually included 131 randomly assigned patients. These patients suffered one to three painful osteoporotic vertebral compression fractures. For the study, they received either bone cement or a dummy injection.
Eventually, patients in both the treatment and the placebo groups reported that, on a scale of 1 to 10, their pain dropped by an average of about 3 points after 30 days, regardless of whether they received bone cement. Specifically, patients experienced similar improvements in pain and physical disability at one month postoperation.
As the researchers reported in their paper, the patients underwent the assigned intervention (68 vertebroplasties and 63 simulated procedures). In both groups, baseline characteristics were similar. At one month, there was no significant difference between the vertebroplasty group and the control group in either the RDQ score (difference, 0.7; 95% confidence interval [CI], –1.3 to 2.8; P = 0.49) or the pain rating (difference, 0.7; 95% CI, –0.3 to 1.7; P = 0.19). Following intervention, both groups experienced immediate improvement in disability and pain scores.
The researchers concluded that improvements “in pain and pain-related disability associated with osteoporotic compression fractures in patients treated with vertebroplasty were similar to the improvements in a control group.”
Meanwhile, on the other side of the world, Australian researchers headed by Rachelle Buchbinder, PhD, a rheumatologist and director of the department of clinical epidemiology at Cabrini Hospital in Malvern, worked with 78 patients with spine fractures who were treated at four Australian medical centers. Results revealed no substantial advantage with vertebroplasty up to six months after surgery compared with the control group.
“Our two studies were similar but not quite the same,” says Kallmes. “The Australian study didn’t allow for crossover, which I feel is a better way to do it. Also, they had two-year follow-up, whereas we only had one year. Further, their study was smaller and required MRI.”
However, similarities in results can’t be easily dismissed. “While different, the studies were complementary,” adds Kallmes.
Data and Experience
Complicating the issue is the fact that healthcare professionals had no major problems (and only a few quibbles) with the studies’ methodologies. The issue is that they couldn’t reconcile the results with personal experience. As Orr wrote in his column, his first impulse was to sift through the paper for flaws that would invalidate the results. The problems he found involved the small number of patients, which perhaps diminished the statistical strength. Also, as a surgeon, Orr wondered whether there was a problem related to patient selection as interventional radiologists performed procedures. “However, even a surgeon’s ego cannot support this interpretation,” he wrote.
Ultimately, Orr realized such criticisms didn’t amount to “fatal flaws,” and the fact that the two studies (which he defines as well designed) reached the same conclusions enhanced their validity.
Benenati also carefully sifted through the data and concedes that he felt comfortable about what he read. “Still, I am not quite sure the studies truly represent what happens in most clinical practices,” he adds. He believes certain elements “sapped the strength” of the findings. “The one thing that bothered me in the studies is that the level of pain needed to require vertebroplasty was significantly lower than in the patients that we typically treat in clinical practice,” he explains. Essentially, patients with a high level of pain will not be impacted by placebo effect.
Kallmes welcomes the observations. “Questions raised weren’t critical but skeptical, and it’s important to carefully scrutinize any trial that produces results that appear out of line with anecdotal experience. One of the major questions involves patient population: Did we enroll the right patients and enough patients?” he says.
In his column, Orr suggests at least one answer: “One concern that does bear some scrutiny is that the trials were too small to identify subgroups that may benefit from the procedure,” he wrote. “In my experience, vertebral augmentation seems to have better results with certain types of fractures. [For instance, patients] with a mobile pseudarthrotic cleft pattern of fracture seem to do much better than those with the more common nonmobile fracture.”
Benenati agrees that patient selection is key. “You have to look at indications and disease process,” he says. “It is quite possible that certain patients would experience benefit more than others. From my own experience, these would include patients with acute vertebral compression fractures and those without associated masses or tumors. They tend to respond better.”
Patient Selection Studies
Two studies presented at this year’s SIR annual scientific meeting suggest patient selection is the key to optimal efficacy as well as collaboration between interventional radiologists and other medical experts.
One study performed in Italy involved more than 1,500 patients who were followed for more than seven years. Researchers studied 2,251 osteoporotic patients, including 1,811 women (average age of 65) who suffered from back pain due to vertebral collapse. Vertebroplasty was performed in 1,542 patients (1,302 women with an average age of 73) when medical treatment did not relieve the pain or improve their quality of life during a three-month period. After vertebroplasty, the patients continued receiving medical treatment from a rheumatologist. In 1,494 patients (96.9%), the average pretreatment pain score on the 11-point visual analog scale was 8.2 ± 1.8, dropping “significantly” to an average of 1.1 ± 1.6 after vertebroplasty treatment. A patient’s ability to perform everyday functions also improved.
In another study involving the treatment of nonosteoporotic compression fractures in patients with multiple myeloma, vertebroplasty resulted in reduced pain, disability, and medication usage. Researchers at the University of Arkansas for Medical Sciences treated 2,715 nonosteoporotic vertebral compression fractures in 792 patients with multiple myeloma (45% women with an average age of 63) over a six-year period. Of these fractures, 2,258 were treated with vertebroplasty and 457 with kyphoplasty, which involves the insertion of an inflatable balloon placed in the vertebra followed by bone cement application.
In this study, the average patient pain-intensity score (based on the 11-point visual analog scale) dropped significantly from 7 to 2.7. Also, 37% of patients reported a decrease in medication usage; 62% had no change; and 1% reported an increase in medication usage. Forty-eight percent of patients reported an improvement in their postprocedural activity level, and 83% would consider undergoing vertebroplasty or kyphoplasty again, if needed.
“These studies revealed that vertebroplasty had excellent benefits for patients with acute compression syndromes and pain,” says Benenati. “The only drawback is that the studies weren’t prospective, blinded, randomized trials. As a result, SIR is considering its own randomized trial. If this is conducted with careful patient selection, I am confident that vertebroplasty will be revealed to be a very valuable procedure.”
Kallmes hopes his study leads to even more carefully crafted prospective and randomized clinical trials, as these would only lead to improved patient care. “I guess that my dream come true would be that our study would convince the medical community that we don’t know everything we thought was true and that more studies would benefit patient care,” Kallmes says. “Before the two studies, physicians had no doubt that they were doing the right thing. But I now feel that indications to treat became too wide and very lax. We need to get a better handle on the specific patients that would best benefit from the procedure. The only way to do that is through conducting randomized, blinded, and prospective trials.”
Physicians intend to continue to perform the procedure. As Orr says, “While I prefer kyphoplasty over vertebroplasty, I don’t think there is a huge difference between them in terms of outcomes. I use kyphoplasty because I think it is marginally safer. But I will continue to do it, and I talk to patients about the trial results as well as my own experience. Most of my patients say that they just want their pain to go away, and they don’t care how and where the pain relief comes.”
— Dan Harvey is a freelance writer based in Wilmington, Del. He is a frequent contributor to Radiology Today.