Biomedical Engineering Reference
In-Depth Information
experts that there is reasonable assurance of the
safety and effectiveness of a device under its condi-
tions of use.”
Based on this definition, reviewers at the FDA
d
individually, and collectively within a branch or
division
d
evaluate the valid scientific evidence in
each application. In the rest of this section and the
next one, we discuss a few general concepts related to
safety and effectiveness, risk analysis, and scientific
methods that FDA reviewers often take into consid-
eration when reviewing a new device application.
These are based on our own experience, and do not
reflect official guidance or policy from the agency.
Device manufacturers should always consult relevant
guidance documents and other official materials
when deciding what type of testing results or other
evidence is needed for individual device applications,
as these will vary greatly depending on the device
type and clinical area.
From an engineering perspective, an FDA
reviewer should follow the same thought process as
the design engineers responsible for the device.
When designing any new object
d
not just a medical
device
d
a design engineer generally has two sets of
constraints in mind. First, there are performance
specifications, or things that he or she
wants
the
device to do. When designing a spinal fusion cage,
for example, he or she wants the cage to fit into the
intervertebral disc space and support load while
a bony fusion develops.
Efficacy
or
effectiveness
relate to how well the device is able to meet these
specifications (in ideal or less-than-ideal circum-
stances). If the device is not effective in achieving its
intended purpose, the patient may be putting them-
selves at risk (e.g., by having an unnecessary surgery,
or by receiving treatment with an ineffective device
when they could have been treated with an effective
one). Therefore, an important part of any submission
to the FDA is clearly communicating what the
intended use of the device is, that is, who and what it
is for, and what it is expected to accomplish. FDA
engineers and scientists want to know whether the
device will meet its performance specifications and
be
effective
.
Second, the design engineer considers the poten-
tial failure modes, or the things that he or she
does
not want
the device to do. If it is an implantable
device, for example, it should not break, wear out,
harm the tissues around it, or cause an adverse bio-
logical reaction.
Safety
relates to how well the device
prevents these things from happening. So, in addition
to evaluating effectiveness, FDA engineers and
scientists also want to know that the device is
unlikely to fail, and is therefore
safe
. FDA reviewers
spend the majority of their time evaluating safety
rather than effectiveness, because the effectiveness of
many devices depends greatly upon interactions with
the body's environment, which can only be observed
during a human clinical trial. Before a manufacturer
can conduct a clinical trial on an unapproved device,
however, they should first demonstrate to the FDA
that potential risks associated with use of the device
have been mitigated or minimized, based on labora-
tory, bench, or animal testing results.
The FDA reviewer begins with the risk analysis
that the manufacturer has developed and presented
for their device and considers many of the same
questions the device designer started with. What
are the possible failure modes of this device? How
might the device harm the patient? Or inversely, how
might the patient's body damage the device?
Generally the list of failure modes is limited to things
that could go wrong
if the device is used correctly
,
because devices are required to be packaged with
adequate labeling that includes directions for use.
The FDA often considers whether those directions
are clear, or if they are confusing or misleading, to
ensure that a well-intentioned person is able to use
the device correctly and without difficulty.
Each category of potential failure mode can typi-
cally be divided into even more specific ones. A
device could break in many different ways and at
different locations, due to compressive forces, tensile
forces, shear forces, or torque. A device with multiple
components could wear out at different interfaces.
The device could harm the body's tissues in a number
of different ways, either physically (by cutting or
abrading tissue) or chemically, depending on the
biocompatibility of the materials. Consequences of
each failure mode are also considered because some
types of failure may be more serious or harmful to the
patient than others. Identifying how or under what
circumstances each failure mode could occur helps to
determine what kind of tests or analyses are needed
to guard against those failures, and the reviewer
judges whether appropriate tests or analyses have
been performed in each case.
If the manufacturer has performed tests to address
all the important failure modes, then the reviewer
assesses the technical details of the testing reports.
The FDA rarely requires that any test or analysis be
performed according to one prescribed method, but
