NEW ORLEANS — The vast majority of spinal procedures involving bone morphogenetic protein (BMP) are off-label uses, researchers found.
BMP is widely used as a substitute for iliac crest bone graft and its use has been increasing at a rapid pace in recent years, investigators said at the American Academy of Orthopaedic Surgeons meeting here.
From 2003 to 2007, the use of BMP more than quadrupled from 23,900 to 103,194 procedures, according to Kevin Ong, PhD, managing engineer at Exponent, an engineering and scientific consulting firm based in Menlo Park, Calif.
Although the FDA has approved BMP only for anterior lumbar interbody fusion, only 16.6% of the procedures in the study involved the approved use, researchers said.
Although it is believed that BMP is safe and effective for a variety of these off-label uses, there are concerns about the protein. They include both its cost and medical issues such as potential heterotopic bone formation, transient bone resorption, soft tissue swelling, seroma formation, and radiculitis.
In particular, medical questions focus on BMP’s use in the anterior cervical spine. A study in the Journal of the American Medical Association last year found that BMP increases complications there. (See Spinal Fusion Agent Raises Costs, Complication Rates)
To examine the prevalence of the on- and off-label use of BMP in the U.S., Ong and colleagues examined data from the Nationwide Inpatient Sample from October 2002 through December 2007.
During that period there were 340,251 primary inpatient procedures involving BMP — and 92.8% were spine fusions.
The vast majority involved off-label uses, including 30% for posterior lumbar interbody fusion/transforaminal lumbar interbody fusion, 20.4% for posterolateral spine fusion, 13.6% for cervical fusions, and 3.9% for thoracolumbar fusions.
Ong said there were historical examples of agents used off-label that later became the standard of care for that indication, such as antibiotic bone cement and bone cement for vertebral compression fractures.
Although the possibility of complications is a concern in this case, one study presented at the meeting, by Steven Glassman, MD, of the Norton Leatherman Spine Center in Louisville, Ky., found recombinant human BMP-2 to be safe for posterolateral spine fusion.
He and his colleagues retrospectively reviewed the records of 1,037 patients who underwent the procedure from 2003 to 2006.
Overall, 18.3% had complications — 7.8% major and 10.2% minor.
In general, the complication rates, including pulmonary, cardiac, and renal problems, were consistent with previous reports of iliac crest bone graft, Glassman said.
Deep wound infection occurred in 2.12%, and hematoma with negative cultures occurred in 0.96%.
Psoas hematoma identified by CT scan occurred in 0.77% of patients.
Severe radicular symptoms were reported in just 0.68%, suggesting that the problems seen in the anterior cervical spine do not occur in posterolateral fusion.
“I think this really points out the fact that these bioactive technologies need to be evaluated carefully in both a site-specific and dose-specific manner,” Glassman said.
He said he could not explain why BMP might have different effects at different sites but suggested that some regions, such as the anterior cervical spine, don’t have enough space to cope with the inflammation that BMP causes.
Gary Friedlaender, MD, of Yale University, who served as a moderator of the session at which both studies were presented, said in an interview, “Whenever you use an agent off-label, you have to be very cautious and you have to have some good reason to believe it’s safe and effective.”
He said he did not know whether the off-label uses of BMP are, in fact, safe and effective, an issue that needs to be explored definitively in prospective clinical trials.
Even so, he added, “The preponderance of evidence is certainly that molecules like the BMPs and the other osteoinductive molecules have an enormously good safety record in general. I do think the tremendous off-label use [of BMP] underscores the intense need for molecules with these types of actions.”
Ong reported receiving research or institutional support from Kyphon and Stryker.
Glassman reported receiving royalties and research support from and serving as a paid consultant for Medtronic Sofamor Danek.
Friedlaender reported having past consulting agreements with Stryker Biotech and serving on the board of directors of BioMimetic Therapeutics.