The Spine Journal 12 (2012) 900–901

Commentary

Commentary: Despite reports of catastrophic complications, why
recombinant human bone morphogenetic protein-2 should be available
for use in anterior cervical spine surgery
K. Daniel Riew, MDa, Eugene J. Carragee, MDb,*
a
Department of Orthopaedic Surgery, Washington University Medical Center, 660 S. Euclid Ave., St. Louis, MO 63110, USA
Department of Orthopedic Surgery, Stanford University School of Medicine, Stanford Medicine Outpatient Center, 450 Broadway St,
Redwood City, CA 94063, USA

b

Received 15 October 2012; accepted 15 October 2012

COMMENTARY ON: Woo EJ. Recombinant human bone morphogenetic protein-2: adverse
events reported to the Manufacturer and User Facility Device Experience database. Spine J
2012;12:894–9 (in this issue).

The Food and Drug Administration (FDA) first issued
their warnings about life-threatening complications associated with the usage of recombinant human bone morphogenetic protein (rhBMP) in the anterior cervical spine on July
1, 2008. Despite this, rhBMP continues to be used in the
cervical spine, and most disturbingly, catastrophic swelling,
respiratory distress, and wound complications continue to
be reported. In this issue of The Spine Journal, the FDA’s
Manufacturer and User Facility Device Experience database for rhBMP-2 is examined. Dr Woo [1] observes that
despite the 2008 public health warning, ‘‘one of the most
DOI of original article: 10.1016/j.spinee.2012.09.052.
FDA device/drug status: Not approved for this indication (rhBMP in
the anterior cervical spine).
Author disclosures: KDR: Royalties associated with the paper: Medtronic (F, Annual); Royalties: Biomet (C), MSD (C), Osprey (C); Stock
Ownership: Osprey (B), Expanding Orthopedics (F), Spineology (B), Spinal Kinetics (B), Amedica (B), Nexgen Spine (B), Vertiflex (B), Benvenue
Medical, Inc. (B), Paradigm Spine (B), PSD (B); Grants: Medtronic (IDE
participation, amount not disclosed, Paid directly to institution). EJC:
Stock Ownership: Intrinsic Spine (B), Bioassetts (B); Private Investments:
Simpirica (D); Research Support (Investigator Salary): NIH (C, Paid directly to institution/employer); Trips/Travel: The Spine Journal (A); Other
Office: NASS/The Spine Journal (E, Editor in Chief); Fellowship Support:
OREF (E, Paid directly to institution/employer), AO Foundation (E, Paid
directly to institution/employer).
The disclosure key can be found on the Table of Contents and at www.
TheSpineJournalOnline.com.
* Corresponding author. Department of Orthopedic Surgery, Stanford
University School of Medicine, Stanford Medicine Outpatient Center,
450 Broadway St, Redwood City, CA 94063, USA. Tel.: (650) 7217616; fax: (650) 721-3470.
E-mail address: carragee@stanford.edu (E.J. Carragee)
1529-9430/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.spinee.2012.10.010

striking (findings) is the continued reporting of these potentially life-threatening adverse events and the need for emergent intervention.’’
If this were not enough bad news, there is recent data
analysis from the FDA and others suggesting that high doses
of rhBMP-2 (40 mg) used in the AMPLIFY trial were associated with an incidence of new cancer events. By 30
months postoperatively, a five to sevenfold higher rate of
cancers was seen after rhBMP-2 exposure compared with
the controls. A potential mechanism for this observation is
that, although rhBMP-2 may not be carcinogenic in this
early time frame, the potent growth factor may be a promoter
of subclinical cancers. That is, rhBMP-2 exposure may affect already existing subclinical disease, making the cancer
cells grow faster or transform to greater invasiveness, becoming clinically apparent malignancies [2,3].
As both of these observed complications associated with
rhBMP-2 can be life-threatening, some have viewed these
findings as a call to remove rhBMP-2 from the market, or
perhaps ban the use specifically in the cervical spine. We
disagree. Although rhBMP can cause catastrophic complications, it is also the most powerful inducer of bone formation available commercially. We believe that judicious use
of appropriate doses, in special cases, should remain an option for surgeons and well-informed patients.
How can we justify such a position in light of the fact
that this is a potentially life-threatening product used for
a nonlife-threatening problem, especially when there are
viable alternatives? As clinicians, we are well aware that
virtually every therapeutic decision we make in medicine
involves an assessment of the risks versus the benefits.

 K.D. Riew and E.J. Carragee / The Spine Journal 12 (2012) 900–901

For example, although acetaminophen is widely perceived
to be a safe and an effective medication, according to one
article, ‘‘Acetaminophen overdose is the leading cause for
calls to Poison Control Centers (O100,000/y) and accounts
for more than 56,000 emergency room visits, 2,600 hospitalizations, and an estimated 458 deaths due to acute liver
failure each year’’ [4]. Yet, the FDA, appropriately, has
no plans to consider banning the use of acetaminophen.
To be clear, the overwhelming majority of cervical fusion cases do not require rhBMP. However, like any other
potentially dangerous drug with some beneficial effect,
we believe rhBMP-2 can have its uses. For example, if a patient with multiple comorbidities associated with an increased risk for pseudarthrosis requires a multilevel
cervical arthrodesis, one can do an anteroposterior operation to increase the rate of solid fusion. Alternatively, one
could accept the risk of pseudarthrosis, knowing that a reoperation may be necessary for some patients. But what if that
patient has comorbidities that increase the risk of circumferential surgery or reoperation? In such cases, judicious
use of rhBMP, at an appropriate dose, may in fact be the
less dangerous alternative.
Bone morphogenetic protein is a naturally occurring
substance that the body normally makes to heal bones, albeit at infinitesimally small levels. Trace amounts of it exist
in the commercial demineralized bone matrix products. At
such small doses, bone morphogenetic protein in demineralized bone matrix is not known to produce deleterious reactions. Unfortunately, it also does not induce bony healing
at a reproducibly high rate. Therefore, it is obvious that
minute doses do not appear to cause problems, whereas
large doses result in a high incidence of complications. In
theory, at least, there may exist an ideal dose at which bone
healing is reproducibly enhanced, but the incidence and severity of adverse reactions are minimized.
For the past decade, the first author (KDR) has been using
rhBMP in the cervical spine in patients with special needs
for fusion augmentation. Initially, when high doses (1–3
mg/level) were used, retropharyngeal edema and fluid was
common, although easily treated with steroids. Reoperations
were rare and there were no deaths in my experience. More
recently, we have most commonly used 0.2 to 0.4 mg per
level. In addition, we place a small dose (20–40 mg) of
a high-potency particulate steroid (eg, methylprednisolone)
in the wound before closure. These patients anecdotally have
less dysphagia than those in whom we do not use rhBMP because we do not routinely use steroids in those individuals.
Although the fusion rate is not 100%, we have noted a much
higher fusion rate for these high-risk patients than historical
controls. We are in the process of determining if even lower
doses can still be effective. We inform each patient about the
FDA and American Academy of Orthopaedic Surgeons
warnings (available at www.aaos.org). In addition, the risk

901

of cancer is disclosed. Only if the patient is well informed
and willing to sign a special consent, do we use the product.
In conclusion, we strongly believe that rhBMP-2 use
should not be banned in the cervical spine. Rather, it should
be used prudently in the small subset of patients for whom
the risks of a failed surgery clearly outweigh the risks of
very low-dose rhBMP-2 fusion augmentation. Surgeons
and patients should understand that like many potent drugs,
rhBMP is dangerous and only clinicians who are well informed about its potential catastrophic complications
should use it. This caveat is particularly important in the
anterior cervical spine.
We believe that patients can be appropriately informed
of the risks involved and make reasonable decisions. Furthermore, patients must be well informed so that they can
be alert to the signs of progressive retropharyngeal edema
and how this effect may be delayed in presentation. Special
care must be taken: the potential for life-threatening complications must be weighed against its possible benefits;
the lowest possible doses should be used; and prudent patient care during the postoperative period, especially in
the week or so after discharge, is essential.
Acknowledgments
The first author (KDR) receives several hundred thousand dollars per year in royalties from Medtronic, the
maker of rhBMP-2, for a posterior cervical instrumentation
system. He does not receive direct financial benefit from the
sale of rhBMP-2 or any products associated with its FDA
approved usage. The second author (EJC) has no financial
relationship with Medtronic or any other company
manufacturing osteobiologic products to promote fusion.
Both authors acknowledge that in their surgical practices
if a cervical spine patient requires circumferential surgery,
the fees are thousands of dollars more than for a one-sided
operation. If the fusion fails and they perform a reoperation,
the fees are in the tens of thousands of dollars for the complex revision operation.
References
[1] Woo EJ. Recombinant human bone morphogenetic protein-2: adverse
events reported to the Manufacturer and User Facility Device Experience database. Spine J 2012;12:894–9.
[2] Even J, Eskander M, Kang J. Bone morphogenetic protein in spine surgery: current and future uses. J Am Acad Orthop Surg 2012;20:547–52.
[3] Food and Drug Adminstration. Food and Drug Administration Executive Summary for P050036 Medtronic’s AMPLIFYTM rhBMP-2
Matrix Orthopaedic and Rehabilitation Devices Advisory Panel.
2010; Available at: http://www.fda.gov/downloads/advisorycommit
tees/committeesmeetingmaterials/medicaldevices/medicaldevicesadvis
orycommittee/orthopaedicandrehabilitationdevicespanel/ucm220079.
pdf. Accessed October 10, 2012.
[4] Lee WM. Acetaminophen and the U.S. Acute Liver Failure Study
Group: lowering the risks of hepatic failure. Hepatology 2004;40:6–9.