Using Risk Analysis to Develop Coated Medical Devices

Posted: 5/1/2014

Link to Medical Device & Diagnostic Industry magazine -  Using Risk Analysis to Develop Coated Medical Devices



The popularity of coated devices is at an all-time high. However, there are challenging technical and regulatory obstacles when adding a coating to a device.

 

Coated-device test articles should be exposed to all the processing conditions, including sterilization.

The market introduction of drug-eluting cardiovascular stents has brought the use of coatings in medical devices to the forefront. The potential upside, both for increasing device efficacy and profits, is significant.1

Somewhat smaller gains in device performance can be achieved by applying a coating that does not contain a drug or biologic to a medical device. Applying a coating can improve the physical performance of the product. Such improvements include decreased frictional forces between the device and tissue or increased resistance to mechanical failure of a device surface.

Devices with a physical coating have clear development and regulatory pathways and relatively low regulatory risks. However, the technical obstacles of adding a coating can be challenging.

Sidebar:
Regulatory guidances

A later article will discuss the technical and regulatory issues associated with drug-eluting, or bioactive, coatings. This article outlines the regulatory requirements for developing coated medical devices.

The technical and regulatory issues for coated devices are presented here using risk analysis as the basis for device design. Some familiarity with design control requirements is necessary.2 Since coated devices are regulated by CDRH only, this article focuses exclusively on CDRH requirements.

Risk Analysis

To meet regulatory requirements for design control, the risks of the coated or surface-modified device must be analyzed. The most recent standard for performing such risk analyses is ISO 14971:2000. This standard has been expanded to include risk management as well as risk analysis and is recognized by FDA as a consensus standard.3,4

The risks associated with a coated device should include those for the uncoated device as well as any general patient hazards associated with the coating. These hazards include the following:

 

  • Unacceptable biological responses to the adhered coating.
  • Unacceptable response to portions of the coating that separate from the device, e.g., emboli of the coating released into the bloodstream, emboli that constitute wear particles shed from coated orthopedic devices, or molecular fragments of biodegradable coatings.
  • Unacceptable coating performance, where the coated device fails to perform as indicated in claims.


Standard methods for analyzing risk include failure modes and effects analysis (FMEA) and failure modes, effects, and criticality analysis (FMECA). These methods identify each component of the medical device, their potential modes of failure, the predicted risks posed by each failure mode, and the means to reduce the risks to acceptable levels.

The potential failure modes for the coating that could result in the patient hazards listed above can include

  • Unacceptable biological or toxicological interactions with the coating or leachable constituents of the coating.
  • Flaking or unintended removal of the coating from the device surface by physical, chemical, or biological means, or a combination thereof.
  • Inconsistent coating of the device surface.
  • Deterioration of the uncoated device function, material properties, or dimensions by the coating or coating process.

The risks posed to a patient by a coated device can be reduced to acceptable levels by using robust designs and materials. Reducing the risks involves creating specification requirements for the device and ensuring that they are consistently met. In some instances, a risk cannot be reduced to acceptable levels by modifying device design. Then a risk-benefit analysis performed by a qualified individual, usually a clinician, can determine if the benefits outweigh the risks. If so, the design can be advanced to the marketplace. If not, the device still may be marketable. However, its use might have to be limited, by labeling, to patient populations, end-use environments, or end-users where risks are at acceptable levels.

Design Input and Output

Sidebar:
Surface-Modified Devices and CDRH

According to CDRH, "design output should be expressed in terms that allow adequate assessment of conformance to design input requirements and should identify the characteristics of the design that are crucial to the safety and proper functioning of the device."2

The performance specifications for the coated device (design output) are derived from the design inputs. Design inputs include requirements of relevant standards, documented customer needs and expectations, and claims for the coating. Device performance requirements are also defined to mitigate patient risk by eliminating the cause or reducing the incidence of potential failure modes.

Performance specifications for coated devices include requirements for

  • Biocompatibility and toxicity of the coated device.
  • Stability of the chemical, physical, and biological properties of the coated device during processing, storage, shipping, handling, and in the biological environment, including adhesion of the coating to the device and durability of the coating in end-use environments.
  • Thickness and uniformity of the coating.
  • Physical performance of the coated device, or reliability.
  • Physical dimensions of the device.
  • Shelf-life requirements.
  • Physical performance of the coating to justify any claims made (reduced friction, reduced wear, improved hemocompatibility, etc.).

• Labeling, including instructions for use.

It is often difficult to define specifications in absolute terms with clearly defined acceptance criteria. When this is the case, the characteristics of one coating can be compared with those of a coating on a similar, marketed device intended for the same end-use in the same end-use environment. The specifications then can be defined in terms of performance relative to that demonstrated by the existing coated device.

Design Verification

Design verification tests demonstrate whether the coated device meets specifications. Test articles used in verification studies should be built using the same methods and materials that will be used to manufacture the finished devices. The test articles should also be exposed to all of the processing conditions that will be employed during device manufacture, including coating and sterilization.

If resterilization is to be allowed, exposure to the number of permitted sterilization cycles is required prior to testing. High-humidity environments can pose problems for hydrophilic coatings, and may require special packaging or handling requirements. And because the coating can be subjected to abrasive wear during shipping, the coated device must be exposed to simulated or real shipping conditions to ensure that it still meets specifications when it reaches the end-user.

The coating can degrade over time, so there is a need to ensure its performance after subjecting the coated device to real-time or accelerated-aging conditions. Results from the studies performed after accelerated aging can be used to support shelf-life claims in marketing applications, but they must be verified by also performing the same tests on a product that is subjected to real-time aging.5

Characterization of the conditioned device determines whether the coating is present in sufficient quantities or thickness in specified locations. The test also ensures that the device's composition is not unacceptably altered during processing, including sterilization, shipping, storage, and handling. The testing required to demonstrate that the coating meets its specifications includes standard bulk-chemical and physical analyses to identify, quantify, and characterize the presence of the coating. Analytical techniques for identifying surface composition include x-ray photoelectron spectroscopy, attenuated totally internal reflecting Fourier-
transform infrared analysis, secondary ion mass spectroscopy (SIMS), and time-of-flight SIMS, among others.

Atomic force microscopy (AFM) and ellipsometry can determine coating thickness, and AFM and profilometry can be used to assess surface roughness. The surface can also be visualized under high magnification using energy dispersive analysis by x-rays, scanning electron microscopy, or AFM to determine the coating's consistency. Visualization methods can also identify the presence of voids, cracks, or other irregularities. On a gross level, staining techniques can be used to ensure that the device has been uniformly coated.

 

Applying a coating to a device, like this coated coronary stent,
can improve the device's physical performance.

Other surface properties may need to be characterized and controlled depending on the end use of the device and the claimed performance of the coating. Such properties include surface charge, hydrophobicity or hydrophilicity, surface energy, and porosity.

Discussion of the surface analytical techniques that assess surface properties is beyond the scope of this article. Fortunately, there are university and industry experts and laboratories that specialize in surface analytical methods. These experts can be relied on to decipher the alphabet soup of surface-sensitive techniques. They can perform the necessary analyses to characterize the modified, or native, surface of a device.

Simulated-use testing is required to demonstrate that the coating or modified surface remains firmly attached to the bulk of the device. The device must be exposed to the stresses that would be applied during manufacture, packaging, sterilization, shipping, storage, implantation (or clinical manipulation), and use of the device. These stresses include simulation of the physiologic forces at the intended implant site or anatomical location.

When a method that is described in a standard is followed to test a coating's integrity, the simulations are mostly straightforward and do not require validation. For example, ISO 11070:1998 has a section for coated guidewires that requires the guidewire to be coiled several times around a cylinder whose radius is scaled to the diameter of the guidewire. The guidewire then is examined visually for the presence of cracks, splits, and dehesion.

However, ISO 11070 is not an FDA-recognized consensus standard, and as such, demonstration that a guidewire coating meets ISO 11070 requirements may not meet FDA requirements.6

If the device will be situated within the vasculature and be subjected to the cyclic pressures created by the beating heart, cyclic stresses must be applied to the device to simulate those present in the physiologic environment. To minimize the time required to perform the simulations, the stresses are typically cycled at rates in excess of those that would be normally experienced. For example, FDA published a guidance for cardiovascular stents recommending that stent designs be subjected to accelerated in vitro testing equivalent to approximately 10 years of real time.7 Such testing requires exposure to approximately 400 million cycles of physiologic pressures experienced in coronary arteries.

In many instances, standards or recommendations for simulated conditions of use do not exist or are not applicable to a device's configuration or end-use environment. In these instances, there is some room to be creative. For example, coated stylet and guidewire insertion forces can be estimated in a test where the device is inserted into raw beef brisket to simulate insertion into human muscle tissue. The force required for advancement and withdrawal is then measured. This method can determine if the guidewire's lubricious coating creates a reduced coefficient of friction. The adhesion of coatings onto screws used to fix orthopedic implants can be evaluated by advancing the screws into plastic-foam blocks. After the screws are removed, a visual inspection of the coating can locate cracks, delamination, or other defects.

The number of cycles applied during the tests (i.e., inserting/withdrawing and advancing/removing a screw) exceeds the number of labeled or anticipated insertions to provide a margin of safety. If the simulated-use methods are not identified in standards, it is important to validate their use-that is, correlate the simulated conditions with preclinical or clinical conditions. It is also important to discuss the test method with FDA before relying on it to produce results that support the efficacy of a coating.

As an alternative to simulated-use studies, the device can be exposed to anticipated worst-case conditions and then evaluated. If the anticipated worst-case conditions are defined clearly to represent conditions that are more stressful on the coated device than the clinical situation, the device's ability to meet its specifications after exposure to such conditions can verify the acceptable performance of the device.

Preclinical Studies

Sometimes it is difficult or impossible to evaluate the performance or safety of the device using in vitro test methods or under simulated end-use conditions. In these cases, preclinical animal studies may be necessary. It is important to document the safety profiles of all coating materials found by performing literature searches and summarizing in vitro and other preclinical test results before initiating the evaluation. The more documentation, the less data may be required from the animal study.
The studies must be performed to good laboratory practice (GLP) requirements. If specific GLP requirements are not met during the study, be prepared to justify the reasons for the failures.8

A pilot study is typically needed to refine surgical technique, animal handling requirements, device design, directions for use of the device, etc., before the pivotal trial begins. Using an animal species known to be a good model for the device to be evaluated can minimize regulatory concerns. Employing a surgical team that has experience with the required procedure can save time.

Whenever possible, the animal protocol, including case report forms, should be reviewed by FDA before the study begins. The selection of species, the surgical procedure, sample size, controls employed, statistical methods, and the study's primary and secondary endpoints all may need to be defended. If the coated devices will vary in dimensions, justify the use of the sizes of the devices selected for the study or evaluate the devices with the largest and smallest dimensions. If the device could be used repeatedly, or if it could be implanted up to a maximum period of time, an evaluation of the anticipated worst-case conditions should be included. Case report forms should include questions whose answers can be used to support claims for the device as well as device safety and efficacy.

Assessment of the safety of the coating could include evaluation of local, regional (downstream), and systemic effects. These assessments would require preparation and histopathological evaluation of tissues from a number of sites. Considerations for the response to the coating can include its effects on healing (i.e., is it delayed?) and fibrous capsule formation (i.e., does it affect the function of the device?). Other considerations include coagulation, complement activation, and the body's immunologic response to the coated device.

 

Biocompatibility and Toxicity Tests

The questions often asked of the product development team are: "What happens if the coating comes off? Will the patient be exposed to a toxic level of the coating?" The typical response is "Don't worry; it won't come off!" But it is difficult to show unequivocally that the coating will continue to adhere to the device under end-use conditions. 

It may be easier to address this question by first performing a literature search to determine the toxicity of all of the constituents of the coating. The search may reveal that the quantity of coating constituents that could be released is below reported toxic limits. If that is the case, a toxicity test can be performed in a suitable animal model to confirm the findings.

This approach assumes anticipated worst-case conditions; for example, imagine that all of the coating is removed from the device in a catastrophic fashion. If the removed coating materials elicit an acceptable response when injected into a suitable animal model, it is possible to conclude that the coating material is suitable for the application or use in clinical trials. Although animal experiments may be fairly expensive, they often offer the best way to determine the toxicity of the coating. This is especially true if the coated device is to be situated in a body location where little toxicity information is available.

In general, biocompatibility issues are addressed by the recommendations of ISO 10993-1 for the duration and body contact of the coated device. FDA recognizes most but not all of the ISO 10993 series as consensus standards. The agency also suggests additional testing for some types of devices.
9

Design Validation

According to CDRH, "[Validation] testing should involve devices which are manufactured using the same methods and procedures expected to be used for ongoing production. While testing is always a part of validation, additional validation methods are often used in conjunction with testing, including analysis and inspection methods, compilation of relevant scientific literature, provision of historical evidence that similar designs and/or materials are clinically safe, and full clinical investigations or clinical trials."2

Some validation information is more readily available and less expensive to obtain than that gleaned from "full clinical investigations or clinical trials." This information should be documented in the coated device's design history file to the fullest extent possible. It may not be possible to completely validate a design based on existing clinical, scientific, and medical device reporting data for a currently marketed device. However, the insight gained from literature reviews of scientific and clinical publications, as well as searches of FDA's complaint databases, can direct, and sometimes minimize, clinical research efforts.

In cases where a clinical claim is to be made, clinical data will be necessary. However, 510(k) clearance or premarket approval (PMA) may be obtained for coated devices without making a clinical claim for the coating. Claims may be based on bench-test data or preclinical animal studies. The data and information supporting the claim must be documented for the end-user in the labeling and be acceptable to FDA.

Typically, clinical studies have not been performed on coated devices. This is because the claims for improvements in physical properties can be assessed using in vitro physical test methods. In addition, safety can be assessed using standard biocompatibility and toxicity test protocols.

Regulatory Submissions for Coated Devices

It is important to discuss an application for a new device coating with FDA. It is important to do so as soon as there is enough information available on the coating to enable a clear description of the technology and the test program that will be implemented. These discussions can be informal, but, particularly for a novel material or evaluation method, it may be better to request a presubmission meeting with FDA.10

If the coated device is not claimed to have improved clinical performance and a predicate device exists, a 510(k) submission most likely will be required to document the safety and efficacy of the device. A PMA application would typically be required only if there is no predicate for the coated device, or if the claim for the coating has not been allowed previously. As always, communication with FDA can help ensure that the correct application is submitted.

In most instances, the regulatory pathway is well defined. Reviewers usually are familiar with the issues of safety and efficacy and methods used to evaluate coated devices. It is important to adhere to the recommendations of the relevant guidance document and to design control requirements. Strict adherence ensures that the information provided to FDA on the safety and efficacy of the device will be accepted. 

It is interesting to entertain the possibility of submitting a special 510(k) for a coated device that is already cleared in its uncoated state. It can be argued that adding a coating to a device is a materials change, for which the submission of a special 510(k) is acceptable. However, the change can only be addressed in a special 510(k) if certain requirements are met. For example, the materials in the coating must have already been used in another legally marketed device cleared by the agency for the same intended use. According to CDRH,

A change . . . in formulation in a material or a change to a type of material that has been used in other legally marketed devices within the same classification regulation for the same intended use could be reviewed as a Special 510(k). This should be true for both non-contacting devices as well as implants and devices that contact body tissues or fluids.
11

Manufacturing Issues

The manufacturing process used to coat the medical device must be fully described in a PMA application. The description should outline all of the steps in the process. It should also identify all solvents and intermediates used. Special attention should be paid to those substances that must be reduced below specified limits in the finished device to ensure acceptable toxicological properties. The susceptibility of the process to contamination should be addressed. For example, how do the controls exercised over purchasing and inspecting raw materials and the manufacturing process, including packaging, ensure acceptable levels of contamination?

Process validation is the major task that must be completed before marketing a coated device. The validation demonstrates that the coating and coated medical device can be produced to consistently meet predetermined specifications. Process validation does not officially need to be completed before submitting a PMA application. However, it is important to document that the coating process is consistent and reproducible before initiating verification and validation studies. The documentation will help ensure that the information provided to the agency is representative of the device as it will be delivered to the end-user. If the process or materials change significantly, a new application will be necessary, resulting in costly delays of product introduction to the marketplace.

Conclusion


The process for developing a coated device is the same as the process for developing an uncoated device. However, additional testing is required for the coated device. The testing includes a repeat of the tests performed on the uncoated device to demonstrate that the coating doesn't degrade device performance to an unacceptable degree. Also included is testing to demonstrate that the coating performs as expected and evaluations of the coating to assure that its chemical, physical, and biological properties are stable after exposure to all processing, shipping, handling, storage, and end-use conditions. Typically, surface-sensitive techniques characterize surface properties of the coating, including its surface chemical composition, thickness, roughness, and uniformity.

The regulatory pathway for most coated devices is the submission of a 510(k) in accordance with the clearly defined recommendations of guidance documents that were written in the 1990s. So, for devices with physical coatings, product development and regulatory risks are low, and the improved physical properties of the device can yield a noticeable effect on device performance. Further, familiarity with the coating process and the techniques to assess the suitability of the coating forms a basis for the development of coatings containing bioactive substances.

 

 

 References

1. Erik Swain, "Coatings: The Next Generation," Medical Device & Diagnostic Industry 25, no. 7 (2004): 70-77.
2. Design Control Guidance for Medical Device Manufacturers, [on-line] (Rockville, MD: FDA: CDRH, 1997); available from Internet: www.fda.gov/cdrh/comp/designgd.html.
3. Harvey Rudolph, "Do We Need Medical Device Risk Management Certification?" Medical Device & Diagnostic Industry 24, no. 11 (2003): 44-49.
4. Mike W Schmidt, "The Use and Misuse of FMEA in Risk Analysis," Medical Device & Diagnostic Industry 25, no. 3 (2004): 56-61.
5. Shelf Life of Medical Devices, [on-line] (Rockville, MD: FDA: CDRH, 1991); available from Internet:
http://www.fda.gov/cdrh/ode/%20415.pdf.
6. Recognition and Use of Consensus Standards; Final Guidance for Industry and FDA Staff, [on-line] (Rockville, MD: FDA: CDRH, 2001); available from Internet:
www.fda.gov/cdrh/ost/guidance/ 321.html.
7. Draft Guidance for the Submission of Research and Marketing Applications for Interventional Cardiology Devices: PTCA Catheters, Atherectomy Catheters, Lasers, Intravascular Stents, [on-line] (Rockville, MD: FDA: CDRH, 1995); available from Internet:
www.fda.gov/cdrh/ode/846.pdf.
8. Preclinical Studies and Good Laboratory Practice, [on-line] (Rockville, MD: FDA: CDRH, 2003); available from Internet:
www.fda.gov/cdrh/devadvice/ide/related.shtml.
9. Memorandum G95-1: "Required Biocompatibility Training and Toxicology Profiles for Evaluation of Medical Devices," [on-line] (Rockville, MD: FDA: CDRH, 1995); available from Internet:
www.fda.gov/cdrh/g951.html.
10. Early Collaboration Meetings Under the FDA Modernization Act (FDAMA); Final Guidance for Industry and for CDRH Staff, [on-line] (Rockville, MD: FDA: CDRH, 2001); available from Internet:
www.fda.gov/cdrh/ode/guidance/310.html.
11. The New 510(k) Paradigm-Alternate Approaches to Demonstrating Substantial Equivalence in Premarket Notifications-Final Guidance, [on-line] (Rockville, MD: FDA: CDRH, 1998); available from Internet:
www.fda.gov/cdrh/ode/parad510.html.

Phil Triolo, PhD, RAC, is president of Phil Triolo and Associates LC, an organization that assists companies in developing new medical devices and combination products while meeting regulatory requirements.

Copyright ©2005 Medical Device & Diagnostic Industry