FDA’s Handling of High-Risk Medical Devices Under the Microscope
By Bruce Patsner, M.D., J.D.
bpatsner@central.uh.edu
Recent scandals such as the debacle over safety issues surrounding the prescription antiinflammatory drug, Vioxx, as well as other medications, have strongly suggested that
FDA has recently enjoyed an uncomfortably-close relationship with the pharmaceutical
industry it must regulate.1 Another major public health safety issue which is just as
important and which, until recently has received relatively little media attention, concerns
the manner in which FDA approves and regulates medical devices, particularly those
high-risk devices posing the greatest potential safety risks to patients.2 Serious
controversy is now swirling around the relationship of FDA to the medical device
industry and whether the agency’s current regulatory framework is equipped to handle
the level of new medical device technologies entering the thirty-five year old medical
device approval process.3
Two recent developments strongly suggest that these concerns are going to be addressed.
First, The New York Times4 published a letter sent by nine scientists working at the
Center for Devices and Radiological Health (CDRH) at FDA to President Obama
charging that senior FDA officials had allegedly acted illegally by approving medical
devices for marketing which had been unanimously rejected as being unsafe by scientific
reviewers. Such a direct complaint to the White House is typically unusual by FDA
medical officers who will often stay within FDA’s hierarchy of authority or contact
pertinent officials in the Department of Health and Human Services. The second was the
publication by the U.S. Government Accountability Office on medical devices5 which
highlighted potential deficiencies in the manner in which FDA approves high-risk
medical devices for marketing in the U.S.6
The Current Regulatory Framework for Medical Device Marketing and Approval
The current regulatory framework for medical device approval was established in the
Medical Device Amendments of 1976 (MDA)7 to the Federal Food, Drug, and Cosmetic
Act of 1938 (FDCA).8 The MDA was an innovative piece of legislation resulting from
1
Howard Markel, Why America Needs a Strong FDA, 294 JAMA 2489 (2005).
Editorial, Substantial Complications, N.Y. TIMES, May 11, 2009, at A20.
3
See Edward M. Basile and Beverly H. Lorell, The Food and Drug Administration’s Regulation of Risk
Disclosure for Implantable Cardioverter Defibrillators: Has Technology Outpaced the Agency’s
Regulatory Framework?, 61 FOOD & DRUG L. J. 251 (2006).
4
Gardiner Harris, Rare F.D.A. Meeting to Discuss Complaints on Device Approval, N.Y. TIMES, (Apr. 21,
2009), available at http://www.nytimes.com/2009/04/22/health/policy/22fda.html (last visited May 31,
2009).
5
U.S. Gov’t Accountability Office, Medical Devices: FDA Should Take Steps to Ensure That High-Risk
Device Types Are Approved through the Most Stringent Premarket Review Process, (Jan. 15, 2009),
available at http://www.gao.gov/products/GAO-09-190 (last visited June 2, 2009).
6
Id.
7
21 U.S.C. § 360k; 21 C.F.R. § 860.3.
8
21 U.S.C. § 396.
2
the public health debacle surrounding women’s safety issues resulting from use of the
Dalkon Shield intrauterine contraceptive device. The 1976 MDA contained a three-class
classification system for medical devices.9 Class I devices pose the lowest risk to
consumers’ health, do not require FDA approval for marketing, and include devices such
as tongue depressors.10 Class II devices pose intermediate risk and often include special
controls including post-market surveillance and guidance documents.11 Class III devices
pose the greatest risk of death or complications and include most implantable surgical
devices such as cardiac pacemakers, coronary artery stents, automated external
defibrillators, and several types of implantable orthopedic devices for spine and hip
surgery.12
The 1976 MDA also contained two different routes for FDA to get new medical devices
to market.13 Class III devices already on the market prior to passage of the MDA were
exempt from the requirement to submit a marketing application and would serve as
“predicate devices” to which newer, but similar devices could be compared. New Class
III medical devices seeking FDA approval (after passage of the MDA on May 28, 1976),
could be approved, i.e. “cleared,” for marketing via an abbreviated process called a
510(k) clearance if they could demonstrate that the new device was substantially
equivalent to a pre-existing predicate medical device.14 If substantial equivalence could
be demonstrated, the manufacturer was not required to submit any new clinical trial data
on the device’s effectiveness and safety and could rely on old data.15 If no predicate
device existed, the manufacturer was required to submit a Pre-marketing Application
(PMA) requiring new safety and efficacy data.16 The PMA process is more expensive,
takes longer to complete, and is more complicated from both an administrative and
medical point of view.17
Issue #1: Lack of Original Data on Some High-Risk Class III Devices
In an era of limited agency resources and only incremental changes to new medical
devices, there were some unexpected consequences to the manufacturer’s option to select
the 510(k) clearance mechanism:18 (1) CDRH has favored it because it is faster and less
9
See supra note 7; see also U.S. Food & Drug Admin., Medical Devices: Overview of Device
Classification, available at http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/Overview/
default.htm (last visited June 2, 2009).
10
21 C.F.R. § 860.3; see also U.S. Food & Drug Admin., Medical Devices: Device Classification Panels,
available at http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/Overview/ClassifyYour
Device/ucm051530.htm (last visited June 2, 2009).
11
Id.
12
Id.
13
Id.
14
See 21 C.F.R. § 807.92(a)(3); see also U.S. Food & Drug Admin., Medical Devices: Premarket
Notification (510k), available at http http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance
/HowtoMarketYourDevice/PremarketSubmissions/PremarketNotification510k/default.htm (last visited
June 2, 2009).
15
Id.
16
Id.
17
Id.
18
Bruce Patsner, Riegel v. Medtronic, Inc, Revisiting Pre-emption for Medical Devices, J. LAW, MED. &
labor-intensive; (2) sponsors have favored it because it is faster and they only need
demonstrate that a new device is substantially equivalent to a pre-existing one; and (3)
the 510(k) clearances have become more complicated as the newer devices began to
resemble preexisting predicate devices less and less, and the predicate devices to which
they were being compared were no longer even the “gold standard” for that particular
class of devices.19
Congress acknowledged this potential problem in 1990 when it enacted the Safe Medical
Devices Act.20 This amendment to the FDCA mandated that all currently and newly
marketed Class III medical devices that could not be reclassified as Class I or II devices
would ultimately undergo the more rigorous PMA process, regardless of whether they
had “cleared” for marketing via the 510(k) mechanism.21 The 1990 Act was a major
effort to correct inappropriate over-reliance on 510(k) clearance for Class III medical
device marketing. Unfortunately, despite this Congressional mandate, for almost two
decades FDA has failed to “call” for PMAs for 27 types of Class III medical devices
which have been allowed to be marketed via 510(k) clearance.22 The end result was that
some potentially dangerous devices were marketed without submission of any original
safety or efficacy data to FDA for review. It is only with the publication of the GAO
Report and some adverse media attention directed at CDRH23 that FDA announced it has
begun to review the safety and effectiveness of older medical devices. But, there is a
second issue which was raised by the controversy over the failure to call for PMAs for
existing Class III devices.
Issue #2: Should the Current Medical Device Approval Process Be Overhauled?
The GAO Report’s criticism of FDA’s continued reliance on the 1976 MDA review and
approval process for high-risk (e.g. class III) medical devices was more than just a formal
reminder that the medical devices division of FDA has not lived up to the administrative
burden placed on it by the Medical Device Safety Act of 1990. More pointedly, the GAO
noted that the high-risk Class III medical devices should essentially always be approved
for marketing via the PMA process, even if an abbreviated marketing mechanism (i.e.,
510(k) clearance) were available. This may be true from a public health point of view. It
also might be preferable from a liability point of view as well, given the fact that
manufacturers are entitled to federal preemption protection against tort claims in civil
suits in state courts only if their product was approved for marketing by FDA via a PMA
as a result of the recent Riegel v. Medtronic, Inc.,24 decision.
ETHICS (forthcoming, Summer 2009) (on file with author).
19
Bruce Patsner, The Problem of Direct-to-Consumer Advertising (DTCA) of Restricted, Implantable
Medical Devices: The Current Regulatory Approach Must be Changed, 64 FOOD & DRUG L. J. 1 (2009).
20
The Safe Medical Devices Act of 1990, Pub. L. No. 101-629 (1990).
21
Id.
22
Rebecca Voelker, More Safety Data Sought for High-Risk Devices, 301 JAMA 1867 (2009).
23
Medical News Today, FDA to Review Safety, Effectiveness of Some Older Medical Devices, (Apr. 13,
2009), available at http://www.medicalnewstoday.com/printerfriendlynews.php?newsid=145794 (last
accessed June 2, 2009).
24
552 U.S. ----, ----, 128 S.Ct. 999, 1017, 169 L.Ed.2d 892 (2008).
These two facts notwithstanding, the only way for all Class III medical devices to be
required to reach U.S. markets only via a PMA would be for Congress to amend the
FDCA. This would be reasonable only if sufficient data exists to demonstrate that the
bifurcated marketing mechanisms originally set forth in the 1976 MDA are no longer a
viable regulatory framework for high-risk Class III medical devices. What is not clear is
whether the adverse publicity CDRH is encountering regarding the appropriate level of
safety and efficacy data for high-risk medical devices in the media25 and peer-review
medical literature26 is the result of: (1) a medical device approval matrix that has now
become outmoded; (2) a simple lack of sufficient medical and administrative personnel to
reverse the tendency to employ 510(k) clearance instead of PMA whenever remotely
feasible; (3) a political problem in which otherwise competent negative medical reviews
about safety problems for high-risk devices are simply being ignored or over-ridden by
superior FDA officials who have decided to make industry-favorable approval decisions
despite safety concerns; or some combination of these. What is clear is that overall
manner FDA is handling safety issues for prescription medical devices is going to come
under the same level of Congressional and public scrutiny, and skepticism, that its
handling of prescription drug safety issues has. Although that may not bode well for
some in industry and FDA, it may be a welcome signal to the new FDA Commissioner27
and a real benefit to consumers in this country.
Health Law Perspectives (June 2009), available at:
http://www.law.uh.edu/healthlaw/perspectives/homepage.asp
25
Alicia Mundy, Congress Moves to Beef Up FDA as Crises Mount, WALL ST. J., July 30, 2008, at A4.
Gregory D. Curfman, Stephen Morrissey, and Jeffrey M. Drazen, The Medical Device Safety Act of 2009,
NEW. ENG. J. MED. 1550 (2009).
27
Gardiner Harris, Obama’s Choice to Lead F.D.A. Doesn’t Shy Away From Difficult Situations, N.Y.
TIMES, (May 6, 2009), available at http://www.nytimes.com/2009/05/07/health/policy/07fda.html (last
accessed June 2, 2009).
26