MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Sources of Innovation


Executives of emerging medical technology companies are rightfully proud of the innovations that have given rise to their companies, and are typically eager to describe how those good ideas have been developed into commercial products. One such narrative is presented in this issue's cover story, "Focused for Growth", in which Accuray president and CEO Euan S. Thomson takes readers through the early years of his company's efforts to develop an idea into a commercial product.

Over the years, MX has carried dozens of such stories. But in this issue, the magazine also takes a step further upstream to look at some of the elements that go into creating medtech innovation.

In "Location, Location—Innovation", authors Paris Kucharski and Scott Oldach identify the top U.S. regional sources of patented medtech innovations. And in "The Academic Connection", author Charles F. D'Agostino—with an able assist from a variety of academically affiliated organizations—describes the growing importance that universities are placing on their work with medtech manufacturers.

Together, these articles illustrate why regional development authorities of all sizes eagerly recruit medical device companies to establish facilities in their territories. And they also contribute to an understanding of some of the elements that go into making possible the innovations that lead to advanced medical technologies—and hence improvements in healthcare.

Creating such an understanding is especially important this year, when politicians and government agencies are considering many initiatives to reform medtech practices in areas such as product marketing, patent filing, postmarket safety, and business transparency. In some cases, such proposals threaten unintended consequences that could weaken medtech innovation.

The existence and success of medtech innovation should not be taken for granted. Supporting such innovation where it already exists, and nurturing its growth in promising new areas, requires carefully conceived and well-executed policies.

Copyright ©2008 MX

Too Much Blame, Not Enough Credit


It's a commonly heard claim: Doctors and patients insist on using the newest and best medical products. Such use has been blamed for everything from soaring healthcare costs to the impending bankruptcy of Medicare.

That's always been an oversimplified argument. Yes, there are devices that are no different from predecessor technologies other than in price. But in many cases, the newer and more expensive technologies make a real difference in terms of patient health. And that often saves money in the long run.

One study in the news makes the point in stark terms. In August, the online journal Circulation published a review by Duke University researchers of 34,000 heart-failure patients who are candidates for cardiac resynchronization therapy (CRT), which shocks hearts back into rhythm. It found that only in 12.4% of cases were CRT devices implanted—this despite earlier studies showing that CRT therapy, when used with drugs and accompanied by lifestyle changes, has tremendous benefits. It is applicable for 30–50% of heart-failure patients. In that patient population, it can reduce deaths by 36% and hospitalizations by 50%.

There are about 5 million people in the United States who suffer from heart failure. That means 1.5 million–2.5 million Americans could benefit from CRT, but only about 620,000 are receiving it. And, according to the Duke researchers, as many as 10% of those who are getting it may not need it. (The American Heart Association recommends CRT for those with a left ventricular ejection fraction of 35% or less.)

The numbers indicate that not enough people who could significantly benefit from the treatment are getting it. Clearly, we have a case where a device is being underused. You won't hear anything about it from those whose idea of healthcare reform is cutting back on new medical technology, though. Where is the outrage over as many as 2 million patients not getting proper care?

The biggest stumbling block, not surprisingly, is short-term cost. CRT devices are priced at about $33,000 apiece. Many hospitals are in a financial crunch, and there is pressure on Medicare and other insurers to cut reimbursements for new technology. So it is easy for hospital administrators to discourage their surgeons from using CRT. Another issue is that there are not enough doctors who have been trained how to implant CRT devices. And it appears that some doctors who could perform the procedure aren't aware of just how beneficial the treatment is.

But if the devices reduce hospitalizations by 50%, isn't the initial cost more than offset by reducing the need for patients to return to the hospital so often? And if they reduce deaths by 36%, isn't that a benefit that should be encouraged, regardless of cost?

Read more about underuse of technology at DeviceTalk - MD&DI's Blog on Medical Device Link.

“There is a fine balance between appropriate restraint and timely and efficient utilization of new technologies,” cardiologist Jonathan Piccini, a member of the Duke team, told Bloomberg News. “In the case of cardiac resynchronization therapy, there is an abundance of clinical trial data that shows it helps patients.”

This is why blanket reform proclamations are not the way to go. Some devices are overused, but some are underused, so reducing the use of new medical technology should not be the corner­stone of a reform policy. Instead, healthcare stakeholders and policymakers should look at clinical outcomes to determine which treatments help patients and which do nothing more than raise costs.

Erik Swain for The Editors

Patent Allows for Breakthrough Imaging of Implants


Glenbrook Technologies (Randolph, NJ) has received a U.S. patent for its x-ray microscope technology that displays real-time detailed motion studies of implanted devices. The patent covers the microscope's use in diagnostic and interventional radiology.

Using its MXRA camera that magnifies objects up to 20 times, the MicroFluor can examine and videorecord implants, such as stents and heart valves, while exposing patients and doctors to less radiation than standard fluoroscopic technology. The camera also has five times greater resolution than current fluoroscopes. For example, a 0.04-in. diameter stent can be displayed on video at almost 1 in. diameter, and the device's deployment can be recorded as a movie.

Massachusets Healthcare Industry Transparency Bill Becomes Law


Despite vigorous objections from the medtech, pharmaceutical, and life sciences industries, in August Massachusetts Governor Deval Patrick (D–MA) signed into law the Act to Promote Cost Containment, Transparency, and Efficiency in the Delivery of Quality Health Care. A major provision of the law requires the Massachusetts Department of Public Health (DPH) to establish a pharmaceutical and medical device marketing code of conduct, and develop and impose compliance and reporting requirements on pharmaceutical and medical device companies that have employees involved in marketing or selling prescription drugs or medical devices in the state.


While companies are concerned about the costs and administrative burdens involved in compliance, they take particular issue with the public disclosure aspects of the law. Industry representatives believe this requirement threatens the integrity and security of proprietary information belonging to companies pursuing research and product development initiatives with partner firms, doctors, hospitals, or other organizations. In contrast to FDA disclosure laws and the Physician Payments Sunshine Act pending in Congress, the Massachusetts law requires public disclosure of any collaborative relationship or industry partnering soon after such actions are first initiated—risking exposure of product development plans to competitive firms.

The law will go into effect on January 1, 2009, and will be enforced by the Massachusetts attorney general. It provides for fines of up to $5000 for each transaction, occurrence, or event that violates the law.

In signing the bill into law, Governor Patrick said, it will "help ensure healthcare providers make choices about prescription drugs and medical devices for their patients based on therapeutic benefits and cost-effectiveness. I am confident the Department of Public Health, pursuant to its regulatory authority, will safeguard the confidentiality of companies' trade secrets and proprietary information and protect against roadblocks to medical research or the education of healthcare providers."


The Massachusetts Medical Device Industry Council (MassMedic; Boston) was an early critic of the legislation and aggressively pushed back on the bill as it was working its way through the state legislature. But like other industry leaders, MassMedic president Thomas J. Sommer is now taking Governor Patrick at his word that all stakeholders will have access to the Massachusetts Department of Public Health as it develops the final provisions of its code of conduct.

The complete text of the Act to Promote Cost Containment, Transparency, and Efficiency in the Delivery of Quality Health Care (MA Senate, no. 2863), which Governor Patrick signed into law in August, is available via

Copyright ©2008 MX

Omni Components Awarded for Performance

(Springfield, MA), a subsidiary of Orthofix, started working with Omni in 2002 when the company began providing engineering and manufacturing services to support Blackstone Medical’s cervical and thoracolumbar surgical products. Omni currently makes a line of critical implant products and tools with Blackstone Medical. Its precision engineering and machining services use computer numerical control Swiss machines, vertical milling, milling and turning machines, and electrical-discharge machining.

Two-Shots: Molding for Silicone and Thermoplastics


Silicone-thermoplastic two-shot molding has been used for more than 20 years in automotive and industrial applications, but has only recently been introduced
into the medical device market.

Two-shot silicone-thermoplastic molding is a method to create a silicone-and-thermoplastic part in one press and in one process. These parts are traditionally molded individually and assembled as one completed medical device component. The two-shot process eliminates secondary operations and assembly, which are often the main contributors to increasingly higher part costs. By reducing the chance of misalignments seen in traditional inserts or in overmolding processes, the two-shot process enables improved part performance and consistent part quality. Two-shot also allows manufacturers to eliminate a tool and the associated tool validation costs.

Most importantly, the technology provides design engineers with increased freedom in part design. Two-shot eliminates the need to design for assembly. It may also enable medical OEMs to bring high-quality, low-cost medical device components to the market faster than other molding processes.

Two-shot is not a new manufacturing technology. Silicone-thermoplastic two-shot molding has been used extensively for more than 20 years in automotive and industrial applications, but has only recently been introduced into the medical device market. The medical market was slow to adopt this process because until recently, there were no commercially available USP Class VI self-bonding grades of silicone.

See a Demonstration of Two-Shot Molding

Over a two-year period, silicone manufacturers developed commercially available self-bonding silicones for the medical market that could chemically bond to rigid thermoplastic polymers during the two-shot molding process and could maintain bond strength poststerilization. Different grades have been developed to bond to polycarbonate, polyester, polyamide (nylon), and polyetheretherketone (PEEK).

How It Works

Two-shot processing requires knowledge of silicone-thermoplastic chemistry and adhesion characteristics. The materials must maintain appropriate processing temperatures in order to adhere to one another. In the material selection process, it is critical to first determine requirements for sterilization, clarity, and biocompatibility.

A two-shot molding machine, such as the one seen here, uses principles of thermodynamics.

Selecting a self-bonding silicone and thermoplastic material that is appropriate for the requirements is crucial. It is also important to choose a thermoplastic material with a high softening temperature that meets or exceeds 300˚F. Materials that offer high-heat stability allow for less differentiation in thermal dynamics of the silicone and thermoplastic molds. The higher the softening point of the thermoplastic material, the higher the temperature at which the silicone can be cured during processing. Higher temperatures permit faster curing and shorter cycle times.

Once a thermoplastic material is selected, the thermal data can be used in running rheology tests on the silicone. Rheology testing determines the temperature at which the silicone begins to cure, as well as the cure profile value for various temperatures. This testing allows the processing temperatures to be chosen before molding.

After the materials and temperature processing range are determined, the two-shot process begins. It is critical to produce thermoplastic parts first. The thermoplastic process needs to be defined and optimized before any silicone is injected into the tool. As long as the temperatures selected were based on accurate analytical testing, the two-shot process should be set up as a standard thermoplastic process, and then as a standard silicone process.

The Laws of Thermodynamics

Knowledge of thermodynamics is required for the two-shot silicone-thermoplastic molding process. In this process, the purpose is to solidify a thermoplastic melt and to try to heat and cure the liquid silicone rubber. A properly designed two-shot mold, based on thermodynamic principles, is the first step in being able to accurately process a two-shot part. The mold typically can be broken into four quadrants: three cold quadrants to cool the thermoplastic material and one hot quadrant for curing silicone.

The thermoplastic material is the first injection shot; the mold is then rotated 180˚ and the silicone is injected into the mold. The mold does not open until the silicone curing process inside the mold is complete. The final result is a completed silicone-thermoplastic medical device component.

The keys to successful processing are understanding the thermodynamics of the mold and choosing a material combination that is compatible at the required thermal conditions.


Designing medical device components that combine silicone and thermoplastic materials follows basic design considerations for the silicone and thermoplastic parts. Silicone can be used in applications that require high-temperature use, low compression set, and purity, but for which thermoplastic materials are not suitable. The silicone enables designing for applications for which thermoplastic materials cannot meet the specifications. Understanding and considering two-shot silicone-thermoplastic processing at the design stage results in a successful two-shot part.

At the early stages in the design process, expectations of part functionality should be determined and clearly defined. The part should be examined for characteristics that are critical to function and design, including locations of parting lines and gates.

After the basic design is complete and main characteristics are defined, the process of designing the part for two-shot molding begins. Typical thermoplastic and silicone design considerations can be used as a baseline for the two-shot silicone-thermoplastic molding process.

Choose a Thermoplastic Material that Can Withstand the Processing Temperature of Silicone. The silicone is cured in the high-heat mold. As a result of the two-shot silicone-thermoplastic molding process, the thermoplastic material must withstand a high mold temperature of around 300˚F to avoid distortion. Materials with high heat-distortion temperatures are recommended. These materials include (but are not limited to) polycarbonate, nylon, PEEK, and polyester.

Avoid Sharp Corners. For thermoplastics, sharp corners not only negatively affect the filling of the mold, but also affect the final properties of the part. Sharp corners in the material flow path can cause stresses in the material, creating uneven flow. Depending on the location, the uneven flow can lead to many defects such as nonfills, trapped air, and flow lines.

In silicones, sharp corners create tears in the silicone during demolding. Silicone flows more easily into a rounded corner than a sharp corner, which optimizes the flow path and helps prevent any possible flow defects.

There is one exception to this rule: A sharp corner may be acceptable in either a thermoplastic or a silicone part at the parting line. A sharp corner at this location is desirable because it provides a much better shutoff of material flow and it is easier to machine.

Figure 1. (click to enlarge) A two-shot silicone-thermoplastic cap is designed for a medical application.

Keep a Constant or Gradual Transition in the Wall Thickness. It is important to have uniform wall thickness in the thermoplastic. Uniformity helps mold filling and prevents warping and sink marks in the completed part. If a part design has thick sections in load-bearing areas, substitute by using uniformly thick ribs. Uniform wall thickness promotes more-uniform fills and faster cycle times, which ultimately result in a more consistent and reliable part. If thicker sections are necessary, employ gradual transitions. Figures 1 and 2 show the basic concept of a two-shot mold.

Figure 2. (click to enlarge) A side view of the cap shows how all silicone features are filled properly, with consistent wall sizes.

Unlike thermoplastics, silicone can have varying wall thicknesses. However, it is critical that the transition from a thin to a thick section is gradual. A gradual transition helps with mold filling. Keep in mind that the thicker the wall, the longer it will take to cure the silicone, which increases the cycle time and cost. Silicone can also be molded into thin membrane sections of 0.015 ± 0.0015 in. thickness.

Consider Material Gating Locations. Position the gate in an aesthetically pleasing location. But also think about the function and manufacturing of the part. Based on the critical areas that were previously defined, the gate needs to be located where it will not interfere with the functionality, such as a sealing surface or fitting location.

The thermoplastic gate should be located at a thicker section to help eliminate sink marks and voids. When choosing a gate location, consider he material's flow path. If there is a point in the part where the flow will split and then rejoin, causing a knit line (or weld line), reexamine the position. Consider whether this knit line is at a point of high stress. Knit lines are weak spots in the part and will be the first point of failure if located in a high-stress area. If a knit line is unavoidable, properly locate the gate where the resultant weld line is in a non-load-bearing area.

One key advantage of two-shot silicone-thermoplastic molding is the ability to design the silicone layer to conceal the thermoplastic gate, providing a completed-part look. However, it is important to know the location of the silicone gate on the part. And a part may have multiple silicone locations. In these cases, the injection location needs to be situated where the silicone can flow.

Make Sure All Silicone Features on the Part Can Be Filled. A two-shot part may contain multiple two-shot features, such as soft-grip and sealing features, and membranes. To save on cost, depending on the part size, design the thermoplastic section to include runner segments to connect all silicone locations. For example, consider a cylindrical part that contains a seal on the top and on the bottom. Instead of gating the part on both sides and using two cold-runner drops, the thermoplastic cylinder can be designed with recessed channels to allow silicone to flow down from one silicone sealing feature to the other sealing feature.

Research Materials to Define Shrinkage. Shrinkage of a thermoplastic part can vary significantly, depending on the base thermoplastic and additives or fillers. Typical shrinkage of thermoplastics varies from 2 to 5%. The optimal shrinkage value to reference is the one supplied directly from the manufacturer.

In a two-shot design, the thermoplastic base part is typically used to create shutoff locations. If shrinkage is miscalculated, the silicone does not fill properly. Therefore, understanding thermoplastic shrinkage and considering that shrinkage in the design is critical.

Silicone varies significantly from thermoplastic. The liquid silicone is maintained at room temperature during plastication and is injected into a hot mold. The silicone expands during molding and shrinks as it cools. Typical silicone shrinkage is 2–3%. Factors such as mold temperature, cavity pressure, flow direction, and postcure affect the amount of shrinkage.

Design the Part to Optimize Bonding. It is important to optimize the part design to enable the strongest bond. Allow large areas of contact between the silicone and thermoplastic to create a significant bond. Also, where possible, include mechanical interlocks. Doing so will provide chemical and mechanical forces that function to bond the silicone and thermoplastic.


The two-shot technique is not necessarily the best option for every situation. Although the cost savings of two-shot molding can be significant, the lead time for tooling is longer, and tooling costs are higher compared with traditional molding.

The two main disadvantages of two-shot molding that are of interest to medical device OEMs are the longer lead times and higher tool costs. Justifying the initial investment is specific to each application. Production volume, lower piece price, and elimination of assembly are key economical justifications of two-shot molding.


Two-shot molding can provide significant benefits in part quality. Further, it provides a cost-efficient means of manufacturing medical device components comprised of adjoining silicone and thermoplastic parts. Two-shot molding eliminates costly secondary operations and assembly, the main contributors to increasingly higher part costs. It also eliminates the additional tooling and validation costs and improves part performance. Device OEMs can expect a process that offers consistent quality and allows freedom in component design.


The author would like to thank Mark Simon, Danny Ou, Adam Nadeau, and Chuck Klann for their contributions to the article.

Sarah Voss is the medical marketing product specialist at Saint-Gobain Performance Plastics (SGPPL; Portage, WI). She can be reached at

Copyright ©2008 Medical Device & Diagnostic Industry

In Memoriam


James M. Gibson Jr., an expert on medical device sterilization who served for a number of years on MD&DI's Editorial Advisory Board, passed away on August 1. He worked for Johnson & Johnson and Abbott Laboratories, among other companies, and most recently had his own consulting business, J.M. Gibson Associates, in the Tampa, FL, area.

Gibson often answered questions on ethylene oxide (EtO) sterilization for MD&DI readers.

The staff of MD&DI extends its deepest condolences to Gibson's wife, Jane, and the rest of his family.

M&A Moves


Ethicon Endo-Surgery (Cincinnati), a Johnson & Johnson company, has entered into a definitive agreement to acquire SurgRx (Redwood City, CA), a privately held developer of the advanced bipolar tissue sealing system used in the EnSeal family of devices. Terms of the agreement were not disclosed.

Zimmer Holdings Inc. (Warsaw, IN) is acquiring Abbott Spine (Austin, TX), a business unit of Abbott Laboratories (Abbott Park, IL), for approximately $360 million in cash. The boards of directors of both companies have approved the transaction, which is expected to close in the fourth quarter of 2008.

Royal Philips Electronics (Amsterdam) has reached an agreement to acquire Alpha X-Ray Technologies (Mumbai, India), a leading manufacturer of cardiovascular x-ray systems targeting the economy segment of the Indian market. Upon closing of the transaction in the fourth quarter of 2008, Alpha will become part of the cardiovascular x-ray business within Philips's healthcare sector. Financial details of the agreement were not disclosed.

Copyright ©2008 MX

Assessing the Biological Safety of Polymers


Illustration by iSTOCKPHOTO

One of the greatest challenges in medical device chemical characterization is performing adequate assessment of biological or toxicological risks from extractables that can compromise patient safety. EN/ISO 10993-17 has clearly stated that risk assessments are a part of material biocompatibility and that they are necessary for the assurance of biological safety.1 Toxicological hazard of the chemical constituents of the materials from which a medical device is made should be considered in biological safety. Therefore, the first step for a biological safety assessment is to characterize the chemicals of the materials. Toxicological hazards can be identified by understanding the toxicity of materials or extracted chemicals.

A systematic analysis of biological risks can be found in the general principles set out in clause 3 of ISO 10993-1.2 Do not be tempted to use the matrix in ISO 10993-1 as a checklist to perform a standard set of tests. Instead, use the principles to develop an appropriate scientific evaluation program based on the specifics of the device.

The results of all tests should be interpreted in the context of the overall risk assessment to determine whether a specific outcome indicates acceptable risk. Such a collaborative approach emphasizes the need for an overall scientifically valid risk assessment. The manufacturer, the analytical chemists, and the toxicological risk assessor must have input and each must be a significant contributor to the assessment process. A common question asked is: Does assessment of the risk that may be posed by a particular material or combination of materials require chemical characterization? This article addresses chemical characterization and defines the role of toxicological risk assessment for medical devices.


Risk assessment has only recently been integrated into international standards and endorsed as an integral part of chemical characterization and biocompatibility studies for medical devices. The suitability of a medical device for a particular use involves balancing any identified risks with the clinical benefit to the patient associated with its use.

EN/ISO 10993-17 states that “among the risks to be considered are those arising from exposure to leachable substances arising from medical devices.” The standard provides a method for calculating maximum tolerable levels that may be used by “other standards-developing organizations, government agencies, and regulatory bodies. Manufacturers and processors may use the allowable limits derived to optimize processes and aid in the choice of materials in order to protect patient health.”

Risk assessment, as explained in EN/ISO 10993-17, is really a tool that has evolved to enable decision making. Manufacturers and processors may use derived allowable limits to aid in choosing the most appropriate material for a particular medical device application.

Toxicological risk assessments have a long history with strong ties to Europe (BS 5736 series standards), FDA, Environmental Protection Agency (EPA), and the Occupational Safety and Health Administration (OSHA). Now ISO 10993 standards for medical devices prescribe the use of toxicological risk assessments for biological studies, including materials characterization and degradation studies. To be effective, the risk assessment must be well organized, documented, and evidence-based for use in support of decision making with respect to product or material safety.

Glossary of Terms

Toxicological Hazard: A property of the chemical constituents of the materials from which a medical device is made and chemical composition should be considered in relation to hazard identification.

Risk Characterization: The final stage in the risk assessment process and involves predicting the frequency and severity of effects in exposed populations.

Chemical Category: A group of chemicals whose physico­chemical and toxicological properties are likely to be similar or follow a regular pattern as a result of structural similarity.

The aim of the assessment should be to identify any biological hazards inherent in the materials used in the medical device and to estimate the risks resulting from hazards for the intended use. The goal is to develop a process that ultimately protects public health and establishes the safety of medical devices. The objective is supported by EN/ISO 10993-17 in subclause 4.3 of the general principles for establishing allowable limits, which states that “the safety of medical devices requires an absence of unacceptable health risk.”

A medical device manufacturer is responsible for ensuring its devices' biological safety and for documenting the assessment of toxicological risks and establishing the effectiveness of the analysis. Evidence must be provided that an appropriate toxicological risk assessment has been carried out so that the OEM can ensure that public health is not endangered.

EN/ISO 10993-17 also adds that “where risks associated with exposure to particular leachable substances are unacceptable, this part of ISO 10993 can be used to qualify alternative materials or processes.” This is another example of the way risk assessment can be used as a mechanism for critical decision processes.

Additional information from biocompatibility tests or on prior use of the materials may be used to provide a basis for further assessment of risks. Acceptable results from appropriate biological tests, such as those listed in the EN/ISO 10993 series of standards, may give a degree of assurance that the risk of adverse reactions occurring during clinical use is low. These tests differ from classical toxicity tests in that they typically attempt to mimic the conditions of clinical exposure to medical devices.

Standardized toxicological tests are amenable to the generation and comparison of data from a wide range of test materials within or across chemical platforms. Because standardized protocols must be broadly applicable for the study of a variety of different materials, they cannot realistically be expected at the same time to address highly focused mechanistic toxicological issues associated with only one or a few chemical compounds.3 This point of view is also expressed in an updated UK competent authority guidance note from 5 EC Medical Devices Directive, published January 2006.4 The Guidance on the Biological Safety Assessment states that

These tests, commonly termed biocompatibility tests, differ from basic toxicity tests in that they typically attempt to mimic the conditions of clinical exposure to medical devices and thus provide an indication of the probability of adverse effects arising during use. They tend, as a result, to be less sensitive than basic toxicity tests and are thus a less discriminating indicator of risk. Biocompatibility test data should therefore be used to complement an assessment based on materials characterization, rather than as a replacement for it.

Toxicological hazard is a property of the chemical constituents of the materials from which a medical device is made, and chemical composition should be considered in relation to hazard identification. If significant risks arising from hazardous residues are identified by chemical characterization, their acceptance should be assessed in line with established toxicological principles. Biocompatibility tests identified in EN/ISO 10993 may be used to provide further assessment of risk.

Components of Risk Assessment

EN/ISO 10993-17 is an ambitious, much needed guidance document that defines and documents consistent practices for evaluation of the risk factors for specific leachable substances. The probability that an adverse effect will arise from exposure to a chemical depends on its inherent toxicity, but also on the amount to which a subject is exposed and the route of that exposure.

EN/ISO 10993-17 provides a systematic method for assessing complex solutions or extracts. The standard uses four basic steps that are commonly used in the risk assessment process. These steps, defined by the National Academy of Sciences (NAS), are as follows:5

  • Hazard identification.
  • Dose-response assessment.
  • Exposure assessment.
  • Risk characterization.

These four steps, when accurately defined and evaluated, result in a statistically derived probability that an adverse effect will occur at a defined exposure level. Risk characterization is the process in which the dose-response assessment and exposure assessments are integrated to predict risk to specific populations. Risk characterization is the final stage in the risk assessment process and involves predicting the frequency and severity of effects in exposed populations.

To establish a tolerable intake (TI) for a specific leachable substance, modifying factors are applied to the data for noncancer endpoints so that an appropriate intake value can be established. For example, the modifying factor is derived as the product of various component uncertainty factors.

One example of a commonly used uncertainty factor is the factor used in extrapolating the effects of animal studies to humans. If only limited long-term exposure studies were available, a higher uncertainty factor leading to a lower acceptable exposure in the human population would be employed. It is noted in the standard that when this factor is combined with other uncertainty factors, modifying factors may be expected to differ by two orders of magnitude. Uncertainty factors and ultimately the modifying factors are derived on a case-by-case basis. They are highly dependent on the quality of the toxicological database.

Figure 1. (click to enlarge) This graph demonstrates the quantitative relationship between the level of exposure and the intensity or occurrence of a resulting adverse health effect. Dose determines the biological response.

An important step in any estimation of chemical toxicity is generating a dose-response curve, a graphic representation of the quantitative relationship between the level of exposure and the intensity or occurrence of a resulting adverse health effect. Figure 1 shows a typical dose-response relationship. A dose or concentration of a chemical substance that does not produce any adverse effect—i.e., no observed adverse effect level (NOAEL)—is identified, usually from toxicological studies involving animals, but sometimes from epidemiological studies of human populations. A modifying factor is applied to the NOAEL to derive a tolerable daily intake (TDI). TDI is the intake or concentration to which it is believed a person can be exposed daily over a lifetime without deleterious effect.

Manufacturing, assembling, packaging, and sterilization of medical devices tend to result in a multiplicity of process chemicals that can potentially migrate into surrounding tissues and body fluids. Many of these chemicals are complex mixtures, often with poorly defined toxicological profiles. The profiles can become increasingly important because moving from a chemical with well-established risks to a chemical we know less about can make it difficult to define the hazard, so a higher risk will be assigned.

Extracts and Mixtures

Risk assessment of extracts or mixtures remains a complex problem. It is now recognized that significant data gaps exist in the area of mixtures toxicology, and these can complicate accurate risk assessments.6

Figure 2. (click to enlarge) Potential extractables from polymeric biomaterials could migrate into the surrounding environment.

It is difficult to judge the leaching risks associated with one pure substance; it is even more difficult if a solution or extract is a complex mixture of different compounds. Most analytical chemists are acutely aware that leachable residue is likely to be a blend of different chemicals (see Figure 2). The resulting biological effect of combined exposure to several agents can be characterized as additive, supraadditive (synergistic), or infraadditive (antagonistic). Another type of interaction, potentiation (a special form of synergism), may be observed. In cases of potentiation, one of two agents exerts no effect upon exposure. But when exposure to both agents together occurs, the effect of the active agent is increased.

The assumption is that compounds with similar metabolic pathways or even with similar structures will have an additive effective. Sometimes, a small change in chemical structure produces sharply different toxicological effects. In addition, there is the possibility that mixtures will have a synergistic effect (i.e., far greater than additive, so that the risk to humans is magnified). Or the effect could be antagonistic, for which the various residues cancel each other out.

When specific toxicological information is unavailable for a particular chemical, a modeling method customary to predictive toxicology can be used to ascertain potential hazards. OEMs use both qualitative and quantitative modeling methods relating chemical structure to biological activity. These methods, called structure-activity relationship (SAR) analyses, have been applied to the prediction and characterization of chemical toxicity.7

SARs are basic to toxicological investigations and are used in risk assessment. The fundamental premise is that the structure of a chemical implicitly determines its physical and chemical properties and reactivities. The properties, in interaction with a biological system, determine its biological and toxicological properties.6

SARs are accepted and provide great benefit; however, there are limitations. There is no single SAR test capable of predicting a particular activity or property for all compounds. A provision for the use of SAR to reduce testing needs is included under EPA's high production volume program.8 Under this program, a chemical category is defined as “a group of chemicals whose physicochemical and toxicological properties are likely to be similar or follow a regular pattern as a result of structural similarity.” The similarities should be based on a common functional group, common precursors, or breakdown products (resulting in structurally similar chemicals). The goal of developing a chemical category is to use interpolation and extrapolation to assess chemicals rather than to conduct additional testing with animals. The specific concern centers on animal welfare and the goal is to minimize the use of animals in the testing of chemicals.

The staff of the Office of Science and Engineering Laboratories (OSEL) at FDA has been responsible for conducting risk assessments on compounds released from medical device materials.9 OSEL staff were involved with the development of EN/ISO 10993-17, which established allowable limits using health-based risk assessment. The standard describes a process for determining an allowable or TI for a single chemical entity. However, it does not provide a good way to evaluate the toxicological significance or biological effect of a solution or extract containing a complex mixture of a number of different chemicals. Synergistic or antagonistic effects are not accurately determined or predicted.

The method outlined in EN/ISO 10993-17 (Method for the Establishment of Allowable Limits for Leachable Substances) was used by CDRH to establish TI for diethylhexyl phthalate (DEHP) released from PVC medical devices. “The safety assessment approach used by FDA/CDRH to derive the TI values is essentially identical to the method used by other regulatory agencies and advisory bodies to establish health protective exposure levels for DEHP (and other compounds).”10 The process is used to ascertain the safety of DEHP, set the precedence for this approach, and evaluate the safety or risk of exposure to extracted chemicals.

This process works well when dealing with a single chemical entity. However, as pointed out previously, antagonistic and synergistic effects are not accurately determined or predicted when multiple chemicals have been extracted. For this reason, biocompatibility tests listed in EN/ISO 10993–series standards should be used to complement a risk assessment process as described in EN/ISO 10993-17.

A second international standard, ISO 14971, “Medical Devices—Application of Risk Management to Medical Devices,” gives guidance with respect to evaluation of toxicological hazards.11 Annex I, “Guidance on Risk Analysis Procedure for Biological Hazards,” also provides suggestions for toxicological hazards caused by chemical constituents with the potential for biological harm. According to the standard, three major factors can be used to estimate toxicological risks, as follows:

  • The chemical nature of the materials.
  • Prior use of the materials.
  • Biological safety data.

The amount of data required and the extent of the investigation depend on the intended use or purpose and on the nature and duration of patient contact. Therefore, material intended for manufacturing an implantable device requires a more extensive investigation than a surface device contacting intact skin.

Collectively, knowing the material's composition, including additives and processing aids, prior use of the materials in a predicate device or similar device, and biological safety tests should provide predictive evidence of any toxicological hazard to patients. Although EN/ISO 10993-17 can be used to establish allowable limits for individual chemicals, biological safety tests when used to complement the risk assessment can give another measure of assurance.

In practice it is not possible to carry out complete chemical characterization of a complex mixture obtained from extracts of device materials. Therefore, the integration of chemical and biological information is critical to any assessment of the toxicity of complex mixtures. EN/ISO 10993-17 deals with establishing allowable limits for each individual chemical. ISO 14971 relies on biological safety data as one of the factors to evaluate toxicological hazards. In combination, appropriate biological and chemical tests provide a way to deal with some of the weaknesses of assessments of complex chemical substances.

Biological safety data provide another level of predictive evidence that none of the extracted chemical substances are potentially harmful to patients. In vitro tests, such as cytotoxicity and hemolysis, provide predictive evidence that extracted substances singularly and collectively are not toxic to mammalian cells.

In vitro tests enable a large number of combinations of chemicals to be assayed using a single test article. They are very useful in studies of acute toxicity and also biotransformation products of extractables. The sample extract or mixture is treated as a whole and tested as-is. Supporting data derived from in vitro and in vivo biological tests can help risk assessors make meaningful predictions as to likely human response. Cell studies can help identify the mechanism by which a substance has produced an effect in the animal bioassay. These tests have the ability to predict any unexpected potentiation or synergistic effects not accounted for by EN/ISO 10993-17 that may result in toxicity.


A comprehensive chemical characterization program that integrates the evaluation of extractables, device material stability, and toxicological risk assessment provides predictive evidence of safety and effectiveness of the device and all its constituents. It is important to give consideration to any potential biological or chemical interactions between the biological environment and the device.

The integration of chemical and biological information is critical to the assessment of toxicity of complex mixtures or device extracts. The guidance provided by EN/ISO 10993-17 and ISO 14971 has made it clear that together, biological safety tests, knowledge of the material's composition (including additives and processing aids, and prior use of the materials) in a predicate device or similar device should provide predictive evidence of any potential toxicological hazard to patients.

More from the Authors: Read an article from Albert and Hoffmann on ISO 10993-18 online.

The value of risk assessments has long been recognized by international organizations and now EN/ISO 10993. In the risk assessment process, a decision must be made related to risk versus benefit. Once risk assessment has been completed, the focus turns to risk management. Part 17 of EN/ISO 10993 states that “manufacturers and processors may use the allowable limits derived to optimize processes and aid in the choice of materials in order to protect patient health.” Decisions should be made using the results of risk assessment, biological safety testing, and safe clinical use of predicate devices as described in ISO 14971. When coupled or linked to biological safety testing, a successful biological safety assessment becomes a highly useful decision-making tool.

Dave Albert is a senior scientist at NAMSA's Ohio laboratory. Reach him at Amy Hoffmann serves as a technical specialist supporting NAMSA's Ohio chemistry department. She can be contacted at


1. ISO 10993-17:2002, “Biological Evaluation of Medical Devices—Part 17: Establishment of Allowable Limits for Leachable Substances” (Geneva: International Organization for Standardization, 2002).

2. ISO 10993-1:2003, “Biological Evaluation of Medical Devices—Part 1: Evaluation and Testing” (Geneva: International Organization for Standardization, 2003).

3. “Risk Assessment in the Federal Government—Managing the Process,” National Research Council (Washington, DC: National Academy Press, 1983).

4. Guidance Note from 5 EC Medical Devices Directive—Guidance on the Biological Safety Assessment (Bootle Merseyside UK: UK Competent Authority, January 2006): 1–10.

5. VL Reynolds, “Applications of Emerging Technologies in Toxicology and Safety Assessment,” International Journal of Toxicology 42 (2005): 135–137.

6. DE Albert, B Kanegsberg, and E Kanegsberg, “Toxicological Risk Assessment for Medical Devices—What Is It?” Controlled Environment 8, no. 10 (2005): 32–33.

7. AP Worth, “The Tiered Approach to Toxicity Assessment Based on the Integrated Use of Alternative (Non-Animal) Tests” in Predicting Chemical Toxicity and Fate, ed. MTD Cronin, and DJ Livingstone (Washington, DC: CRC Press, 2004): 391–412.

8. JD McKinney et al., “The Practice of Structure Activity Relationships (SAR) in Toxicology,” Toxicological Sciences 56 (2000): 8–17.

9. “Chemical Hazard Data Availability Study: What Do We Really Know about the Safety of High Production Volume Chemicals?” (Environmental Protection Agency, Office of Pollution Prevention and Toxics, 1998).

10. “Safety Assessment of Di (2-ethylhexyl)phthalate (DEHP) Released from PVC Medical Devices” (Rockville, MD: FDA, 2004): 1–60.

11. ISO 14971:2000(E), “Medical Devices—Application of Risk Management to Medical Devices, Annex C: Guidance on Risk Analysis Procedures for Toxicological Hazards” (Geneva: International Organization for Standardization, 2000).

Copyright ©2008 Medical Device & Diagnostic Industry

Accuray by the Numbers


(click to enlarge)
Share price for Accuray Inc. (NASDAQ: ARAY; Sunnyvale, CA) versus the S&P 500, since the company went public in February 2007. Although promising clinical results and product orders bode well for Accuray, the company's share price has endured a downward slide since its initial public offering, when it briefly neared the $30 level.

In its first full year as a publicly held company, Accuray Inc. (Sunnyvale, CA) has maintained a strong balance sheet and solid revenue growth. The company was profitable in all four quarters, and ended its fiscal year in June 2008 with $159 million in cash and investments, and no debt. Accuray reported a backlog of $647 million, with approximately $359 million in sales contracts for the CyberKnife robotic radiosurgery system and another $288 million associated with services and other recurring revenue. Additionally, Accuray's revenue for fiscal 2008 reflected an annual growth rate of 50%, increasing to $210.4 million from $140.5 million the previous year.

Accuray estimates that the worldwide market opportunity for CyberKnife systems is approximately 7500 units. Currently, there are more than 140 CyberKnife systems installed around the world, with 90 in the Americas, 20 in Japan, 18 in the rest of Asia, and 12 in Europe. In June, Accuray received regulatory approval in Japan to market the CyberKnife system for extracranial radiosurgery. As a result, the company sees great potential for growth in this key market. Previously, the CyberKnife system was only approved for intracranial treatments, and despite that limitation, Japan was the largest market for the system outside of the United States.

Copyright ©2008 MX