Originally Published MDDI June 2002

WASHINGTON WRAP-UP

A software-powered cleaning technology has caught the eye of FDA. Beyond possible savings for both OEMs and SUD reprocessors, the technology offers the prospect of lifting some of the agency's inspectional burdens.

James G. Dickinson

Another Device Classification Dispute

How clean is clean enough for a medical device, be it reprocessed or original equipment? And how do you validate that cleanliness to the satisfaction of both FDA and the end-user?

HyperFlo, a small Gilbert, AZ–based company, believes it has the answers to such timely questions. The firm has developed a new, fully programmable technology as a result of cleaning semiconductor components to exacting aerospace and military specifications.

HyperFlo's custom-built equipment, which is powered by software that allows for on-line validation, supports aqueous, semiaqueous, solvent, and cosolvent cleaning applications at temperatures from ambient to 250°F. Each machine features a single-chamber cleaning, rinsing, and vacuum-drying environment and employs pull-through rather than push-through kinetics for the cleansing fluids used.

As explained by the company's president, Charlotte Frederick, this new technology, which ranges in price from $65,000 to $180,000 depending on capacity, offers dollar savings to both OEMs and single-use device (SUD) reprocessors at a time when cost pressures are forcing everyone to be more efficient.

It also offers prospective savings to FDA, which has assumed new inspectional burdens with the growth of SUD reprocessing. If the agency could gain sufficient confidence in the automation and efficiency of HyperFlo's technology, FDA audits of sterilization validation could simply be a matter of reviewing software logs on-line—in theory, at least.

FDA also sees the technology's utility to the OEM and SUD reprocessor sectors of the industry. CDRH director of dental, infection control, and general hospital devices Timothy A. Ulatowski told me his division is interested in such a product.

"A product used in processing or reprocessing devices would be considered a manufacturing article and subject to evaluation under the quality system regulation," he said. "It could be subject to premarket review also, as part of the review of what is being processed or reprocessed, if it is an entirely new cleaning or sterilization technology. The manufacturing process—including the use of this article—would have to be validated."

So FDA wants to evaluate this technology, especially if it is to be used in the medical device industry to sterilize devices. Frederick isn't identifying any of HyperFlo's customers without their highly unlikely consent; she knows they don't like FDA inspections. Some major device OEMs have ordered from HyperFlo.

When the time comes for the first customers using this new technology to be the subject of an FDA inspection, HyperFlo's technology in that particular application will come under regulatory review. The results will likely be confidential—the company claiming a competitive (trade secret) advantage—so industry at large won't learn of the technology's regulatory attributes, only its economic ones.

Frederick thinks FDA should follow the military's lead and buy its own HyperFlo to test in myriad applications for the benefit of public health in general. The results would then be public.

One HyperFlo customer that did agree to be interviewed was Alliance Medical, a major SUD reprocessor who calculates that the new technology will save the company 75% of its present processing time, and increase its device reprocessing volume even more than that.

Alliance Medical confirmed these expectations by sterilizing GI biopsy forceps at the HyperFlo facility before placing an order. The company found the level of sterility to be as good as or better than its present, largely manual 99.99% standard. Before Alliance Medical can actually reprocess devices with HyperFlo on-site, it will have to obtain FDA approval for 510(k) addenda that it will submit for each device.

Another Device Classification Dispute

In another case of apparent FDA heavy-handedness in a device classification dispute, a small South Orange, NJ, device maker complained in April that its pending device reviews at FDA had been shut down by the abrupt imposition of FDA's Application Integrity Policy (AIP).

Although qualitatively dissimilar, the complaints of Endotec Inc., a hip and shoulder replacement systems maker, awaken echoes of two earlier device classification embarrassments for CDRH—controversies involving dental muscle monitor maker Myo-Tronics in 1997 and temporomandibular joint (TMJ) prosthesis maker TMJ Implants in 2000.

In those cases, FDA assertion of a new device classification against marketed products arbitrarily shut down the companies, aroused congressional indignation, and forced the agency to restore the embargoed products to the market after millions of dollars in sales had been lost.

In Endotec's case, only unmarketed products have been jeopardized (the company currently markets FDA-approved hip and shoulder replacement systems in the United States and has an investigational TMJ device); there are no allegations as in the Myo-Tronics and TMJ Implants cases of FDA employee bias against the sponsor.

Nevertheless, to Endotec, the business disruption felt the same, and just as unjustified. And in all three cases, a common theme is CDRH's alleged lack of technological sophistication about the embargoed devices.

Endotec's troubles began in September 1999 when the company submitted a 510(k) for its total-ankle-replacement system as substantially equivalent to devices currently marketed.

In response to the 510(k) notification, CDRH said it had determined that the device was not substantially equivalent because the Endotec device incorporates a flat superior bearing surface mating with a flat tibia component without any additional device-related constraints, and thus must be considered a nonconstrained ankle joint. Other devices, it said, are semiconstrained because of the presence of mechanical constraints built into them.

Through reams of correspondence since that initial determination, Endotec and CDRH have continued to disagree about the nature of the Endotec device and how it should be classified. According to Endotec regulatory affairs executive Jared Pappas, the Endotec device uses anatomical constraints within the ankle and thus should be considered semiconstrained, even though the constraints are not mechanical. "It's better to keep as much of the anatomical structure as possible," he says.

The definitions that CDRH relies on for a Class III ankle prosthesis, he says, came from an orthopedics panel 20 years ago and no longer apply, given the evolution of technology. "That panel 20 years ago developed the definition based on one scientific paper and one device that had bad results," Pappas says. "Over the years it has been determined that the mechanical semiconstrained devices don't work. It's ironic that Endotec today could get 510(k) clearance for a device that doesn't work and start marketing it tomorrow, but we can't get approval for one that does work—and is being successfully used in several other countries today—because it doesn't fit an outdated definition."

At one point in the exchange of letters, CDRH suggested that Endotec consider applying for a reclassification of the device. The company submitted such a petition in August 2001. The following month, FDA began a 21-day inspection at the company's facility and conducted other inspections at two clinical sites.

Endotec says it responded in detail to the investigators' written observations and heard no more until it received a February 14, 2002, letter from CDRH director David Feigal invoking the AIP. The letter said the inspections had revealed a "systemwide failure to ensure the integrity of data," and cited a number of inspection observations in support of that finding. Feigal also concluded that data submitted about the study were unreliable based on sponsor records and deficient device accountability.

Pappas says his sense of injustice is profound, especially because CDRH officials have told the company "there is absolutely no fraud or wrongful acts involved in anything [Endotec has] done." And, he says, CDRH's own AIP requirements say that normally a company is warned and has a chance to respond before the restriction is issued. "But we never had that opportunity," Pappas says.

Pappas also says that while Endotec is not accusing CDRH of anything, the company finds it odd that the AIP was invoked within weeks of the end of the time period CDRH had for making a determination on the company's reclassification request. Now, as a result of the AIP, CDRH need not complete its review of the petition. He says he also finds it strange that the AIP came shortly after Endotec started applying political pressure to get action on the reclassification petition, about which it had not heard anything for nearly six months.

Asked to comment on Endotec's complaints, Feigal said—just as he did at an equivalent point in the TMJ Implants' dispute—that the center would not make any comment on this case because it is pending.

Pappas says the company has been in contact with FDA several times since the AIP was invoked, including an appeal to CDRH ombudsman Les Weinstein, who did not respond. Endotec is putting together corrective actions, he said, and bringing in an independent auditor to go over its records. Endotec has also asked the FDA commissioner's office to intervene and reverse application of the AIP.

In a March 12 letter to the commissioner's office, Pappas said, "Since there have been no dishonest or subversive acts, or allegations of such acts, the application of the AIP in this case is clearly not justified." He also disputed many of the statements made in Feigal's AIP letter. And he said FDA's actions violated common law and agency procedure.

"If we had been given the opportunity to defend our case, then perhaps this extreme action could have been avoided," Pappas wrote. "Five months have passed since the audit and our response to the FDA form 483. During this period corrective actions were taken to improve our clinical trial procedures and their implementation. FDA should have included these facts in their deliberations of whether to apply the AIP. It did not do so. . . . We also should have received notification that our response to the FDA form 483 was inadequate and that the audit's observations were, in fact, charges. This information would have given us the opportunity to respond to the audit more accurately had we known that it would lead to punitive action."

Endotec wanted CDRH to send the reclassification petition to an independent panel for assessment, and was confident that a panel would rule in its favor.

"We believe we have an extremely strong case if we can get the matter away from FDA and to an independent panel," Pappas says, adding that company officials don't understand why CDRH continues to cling to the old definition and requirements. "If it's to save face," he says, "what's wrong with saying that a decision was made in 1982 based on the best information available then, but now it has to be changed because we've learned more?"

Pappas says that since the reclassification petition does not rely on any of the clinical trial data that CDRH is questioning (using instead peer-reviewed journal articles to make its points), the company believes that review of the petition should proceed despite the AIP.

On April 8 of this year, Endotec had what Pappas says was a "very positive" meeting with CDRH director of application integrity Carl De Marco, who indicated that on the basis of the company's corrective action plan and ongoing outside audit, the AIP could be lifted in a short time.

Endotec also is continuing to work through its members of Congress to bring pressure on FDA and is asking potential patients who want to be able to use the device to contact their congressional representatives as well. "People who need this device are in a lot of pain, and we could help them," Pappas says. "We're hurting people by allowing this classification [dispute] to continue."

Copyright ©2002 Medical Device & Diagnostic Industry