Medical Device Daily Washington Editor
GAITHERSBURG — Tuesday's orthopedic and rehabilitative devices advisory panel meeting of the FDA, to review the Bryan artificial cervical disk made by Medtronic Sofamor Danek (Memphis, Tennessee), commenced just about the time that the agency announced that it had given PMA approval to the company's Prestige artificial cervical disc (Medical Device Daily, July 18, 2007).
Then, by day's end, the company had a second reason to celebrate as the panel voted 7-1 to recommend conditional approval of its Bryan artificial cervical disc device.
Despite solid clinical data for the Bryan, the agency had sent the PMA application to the advisory panel due mainly to a pair of fairly novel design elements of the device.
One of these is that the Bryan employs no screws to affix it to the spine, relying instead on a pocket milled into the vertebrae to serve as a source of fixation. This is abetted further by a beaded porous coating on the upper and lower surfaces of the disc to develop traction between the disc and the adjacent vertebrae.
The second point of novelty — which served as a flashpoint with the sole panelist who voted against approvability — was the use of polyurethane in the interior of the disc. While the polyurethanes used in the Bryan are generally well characterized, their use in this specific application is without precedent.
Subjects in the control arm received the Atlantis cervical plate, also made by Sofamor Danek, for stabilization of fusion done with "commercially available allograft," according to FDA documents.
Stephen White, VP of R&D for Sofamor Danek, said that the Bryan's porous titanium shell is also found in some hip replacement devices. The polyurethane sheath that wraps around the device's middle was incorporated to facilitate a one-piece insertion into the spinal column and to retain saline lubricant. The Bryan allows 11 degrees of flexion, extension, and left and right bending, and White said that the domed shape of the Bryan "maximizes the stability of this device."
White also said of the polyurethane nucleus that "this and similar materials are currently 510(k)-cleared in two devices."
FDA indicated that the final bench-top test, used to assess lifetime durability, subjected one of the devices to 40 million cycles, or the equivalent of 295 years of in vivo use. That test resulted only in a "hole in the rim of the nucleus" of the device.
FDA documents indicated that another bench test that examined the wear caused by the interface of the titanium shell and the polyurethane interior from 10 million cycles at 130 newtons. No large pieces of polyurethane broke off, and the test generated less than 15 milligrams of debris, with more than 90% smaller than one micron. Tests in bovine solution yielded "comparable results."
Ann Ferriter, a mechanical engineer who served as the lead reviewer for the agency on the application, said that this application involved the first use of this combination of materials and employed "a novel method of affixing" the device to the bone.
"We commend the sponsor for enrolling roughly equal numbers of men and women in their trial," Ferriter said, adding that the overall success rate was more than 80% for the Bryan and about 70% for the controls.
She said that the Bryan had "roughly the same adverse event rate" as the controls.
FDA documents stated that the company performed worst-case testing "at FDA's request," but panelist John Kirkpatrick, MD, a professor of orthopedics at the University of Florida College of Medicine (Gainesville), asked why wear testing was restricted to what he described as "the neutral zone." He said that this is "the area of the stress-strain curved that sees very little stress" in normal daily activities.
A total of 117 patients were enrolled but never received treatment, a source of some concern for the panel.
However, eight enrollees could not follow through due to insurance issues, and 32 who were randomized to the control arm dropped out of the trial due to "dissatisfaction with the results of randomization." Eight were lost to failing inclusion/exclusion criteria, and 18 left the study because of improvement in their symptoms without either surgery.
Steve Kurtz, PhD, president of Exponent (Menlo Park, California), an engineering consultancy, said that his firm had looked at retrieved Bryan discs — 15,000 of which implanted outside the U.S. — and said that he saw "no evidence of damage from oxidation" or other forces out to 6.1 years in the polyurethane center of the disc.
Paul Anderson, MD, an associate professor of orthopedic surgery at the University of Wisconsin School of Medicine and Public Health (Madison), told the panel that he had explanted some of the units from animal models and that, even with microscopes and Fourier transformation infrared spectroscopy, the polyurethane components of those discs "are virtually identical to controls," which were bagged and sealed after manufacture.
However, one of the panelists, Sanjiv Naidu, MD, an orthopedic surgeon with the Penn State Fredericksen Outpatient Center , termed as "bunk" the assertion that there was no degradation in serious quantities.
And he complained about the lack of data for a friction coefficient between the polyurethane core and the titanium shell, saying that he was "not thrilled with the material data presented to the panel."
None of the other panelists backed his position, however.
Despite Naidu's reservations, the other panelists voted to recommend approval of the device with a number of conditions.
One was that the company would conduct a trial "expeditiously" to examine reports of kidney problems seen in some rodent models. The kidney problems were consistent with lab-acquired microbials, but the lab did not undertake an effort to make a differential diagnosis.
Another condition was that the post-approval study will examine kyphosis (spinal curvature) and heterotopic ossification.
Kathryn Simpson, PhD, medical officer of Sofamor Danek, emphasized that the trial demonstrated superiority of the cervical disc compared to surgical fusion, but the panel voted to recommend to the FDA not to allow the firm to make such a claim in its product literature.