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.


Articles from 1998 In February

Continuous Improvement Shores up the Bottom Line

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


Kaizen's continuous improvement process views even a change that reduces manufacturing time by as little as half a second as a valuable contribution to a company's overall efforts.

It seems every year another management philosophy that will shorten product development or manufacturing time, enhance employee productivity, or contribute in some other way to fortifying a business's bottom line comes into fashion. Americans seem to approach work challenges looking for fast, dramatic results. The resurgence of reengineering—with its accompanying replacement of existing processes with entirely updated, seemingly more practical ones—demonstrates a desire for the huge reorganization that will deliver bigger, better whatever.

In contrast, the Japanese—whose management prowess has resulted in total quality management and statistical process control as well as an incredibly efficient and technologically advanced workforce—focus on processes and their individual elements. Improvements that cut manufacturing time by as little as half a second are viewed to be as valuable as those that save larger blocks of time.

In this month's Bottom Line, we revisit Japan's kaizen philosophy, which changes processes in a sustained series of incremental adjustments. Kaizen came to America in the 1980s at Toyota plants and has since expanded into other manufacturing operations, with dramatic results. Medical device companies have reaped rewards as well as awards. Both Critikon (Southington, CT) in 1994 and Perfecseal (Philadelphia) in 1997 have won the prestigious Shingo Prize for Excellence in Manufacturing, which recognizes companies in the United States, Mexico, and Canada that excel in productivity and process improvement, quality enhancement, and customer satisfaction. Critikon's use of kaizen generated a 50% jump in productivity, as measured by the units made per employee, and cut cycle time by 60% and material waste by 53%.

After adopting kaizen, Perfecseal saw its lead time fall by three-quarters. Further, on-time and complete shipments, which had only occurred 38% of the time, jumped to 95%. Productivity improved 50%, and setup time was cut by half.

As impressive as these numbers are, it's even more noteworthy that the steps that lead to them can be taken relatively quickly. Kaizen projects usually range from two to five days; larger goals are slated for completion in about 30. Further, all employees are deemed valuable contributors to problem solving. In addition to the morale boost this involvement produces, management benefits by having the shop floor point of view presented, because with kaizen, changes aren't sparked by a team brooding over a conference table. Instead, team members try new techniques on the floor until they find the best solution.

By adopting the kaizen philosophy and thus empowering employees, helping them upgrade their skills, and looking at all the small elements of existing systems in the process of making more-sweeping changes, organizations can expect to see gains in quality, cost, delivery time, safety, morale, and, ultimately, competitive position.

Tell me about the impact kaizen and other management philosophies have had on your company's bottom line. I'd like to hear from you!

Stacey L. Bell

Copyright ©1998 Medical Device & Diagnostic Industry

Lessons Learned, New Challenges: Life after FDA Reform

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


MDMA's former executive director examines the medical device industry's history, recent legislative victories, and the challenges it faces.

In late 1994, CEOs and top regulatory affairs officers at U.S. medical device companies and trade associations began meeting in attempts to chart a new direction for FDA—an agency that medical device manufacturers, by Congress's direction, must contend with every day. The task: change the unpredictable, subjective nature of this $1 billion—budget, 10,000-employee regulatory agency so it can "partner" with industry to deliver safe, efficacious medical technologies to patients more quickly. To say that people had differing views on how to accomplish this task is an enormous understatement. Three years later—after struggling, debating, lobbying, and warring with self-proclaimed consumer groups—the device industry succeeded: President Clinton signed the FDA Modernization Act of 1997 into law on November 21.

This is unquestionably the largest restructuring of FDA since the enactment of the Medical Device Amendments to the Federal Food, Drug, and Cosmetic Act in 1976 when Congress proclaimed that if a product was not a drug, a biologic, or a food, it was a medical device. Despite the vast differences between product lines (e.g., in vitro diagnostic test kits, IV catheters, linear accelerators, kidney dialysis machines, and automated external defibrillators), medical device manufacturers have been regulated under the same set of laws, for better or for worse, ever since. Now that the new law has been passed, the industry needs to examine how it is going to build on this recent success, revisit some of the lessons learned in the process, and identify what it is this industry seeks to accomplish in Washington in the years to come.

"It is not the writing of the laws—it is their execution" (Thomas Jefferson). Bruce Burlington, MD, and his staff at CDRH are to be commended for the improvements made in the management of the device program during the past two years. However, implementation of an entirely new law will be the ultimate test of their skills. Congress must continue to provide the constructive oversight that it has demonstrated in the past three years to ensure that the execution of the law is consistent with its intent. The national trade associations involved in the lobbying effort, namely the Medical Device Manufacturers Association (MDMA), Health Industry Manufacturers Association (HIMA), and National Electrical Manufacturers Association, also need to continue to provide leadership and direction.

The next FDA commissioner must take the same results-oriented approach Burlington has. Concepts like process improvement, reengineering, efficiency, and streamlining, often only mentioned following a management consultant's audit, should become part of the new FDA environment. With proper execution, this law will provide flexibility for each division in CDRH to further streamline the device approval process and, it is hoped, add consistency to the enforcement process.

The longer-term issues facing this industry are complex, including strengthening the relationship between industry and Congress, making sure that the next FDA commissioner is a strong manager by letting Congress know its needs, and identifying a leader on Capitol Hill to replace the retiring Senator Dan Coats (R—IN), who learned about our industry, established a continual and aggressive public relations front, and kept companies committed to Washington representation.


Since the passage of the 1976 amendments, most device companies have been on the offensive, adapting to the ever-changing marketplace. Through well-calculated technology acquisitions and concentration on international growth, the industry has become one of the leading high-technology sectors in the United States, boasting an impressive $5.1 billion export manufacturing surplus. Yet, although it is a perfect exemplification of the type of business that is fueling U.S. economic growth, the industry has been mysteriously naive in the ways of Washington policy making.

For example, from 1994 to 1996, the leaders of the three national medical device trade associations rarely spoke to one another while working on FDA reform legislation and waging weekly press release wars. The industry leaders would not have been successful in 1997 either, had the three groups not united to work as a team. We learned from the pharmaceutical companies that speaking as one is far more effective. The marketplace is where competition should exist, not on Capitol Hill.

Despite early problems, the industry has managed to score a number of impressive legislative victories since 1994, including the following:

  • Elimination of medical device user fees.
  • Passage of the FDA Export Reform and Enhancement Act of 1996.
  • Enactment of the FDA Modernization Act.

While these wins did not come easily, it will be far more difficult and much more expensive in the future if the industry does not maintain its present forward-thinking attitude and activity level.


It is time to educate the people in Washington who make decisions that affect our bottom line about the medical device industry—including congressmen, key senior staffers at HHS, FDA, HCFA, the U.S. Department of Commerce, the U.S. Trade Representative's office, and the domestic policy staff in the White House. Often, top policymakers only hear about medical devices when they're in the hospital or there is a product recall.

The level of ignorance among our policymakers is epitomized by some of their comments, such as, "Just how long does it take to get a 401(k) approved?" (a veteran member of the Senate Labor and Human Resources Committee in early 1997) and "Medical devices? . . . You mean like rubbers and stuff?" (a senior senator two days before a scheduled vote on FDA reform). Unfortunately, the primary congressional committees that create legislation that affects our industry and appropriate money to FDA every year have little comprehension of what medical device companies do. Lawmakers typically have little knowledge of the specialized needs of the medical device industry when enacting laws that protect intellectual property and control medical research, medical device approval times, marketing, and reimbursement. Moreover, without adequate congressional oversight, regulations promulgated by FDA often ignore the needs of the medical device industry.

The industry must decide how to best manage its Washington presence. Politicians must be aware that medical device manufacturers help save lives, introduce new technologies, provide high-paying jobs, and are true entrepreneurs. Manufacturers must also align themselves more closely with patients and physicians who use their technologies. Policymakers need to recognize the true value of this industry—not just in dollars and cents but in human value.

Further, the management of the leading companies needs to take at least two days each year to come to Washington to encourage the adoption of regulatory and legislative policies that benefit the industry and ultimately the patients. CEOs should not come to Washington only when they need something from Congress. Medical technology and innovation fairs could show off our technologies on Capitol Hill, and congressional hearings could examine the benefits of new, innovative technologies and how they positively affect patients.

By educating the young lawmakers who will one day be leaders and committee chairs, such as Congresspersons Joe Barton (R—TX) and Anna Eshoo (D—CA), the industry protects itself now and positions itself for future efforts. Education must be conducted by the industry as a whole, not by individual segments, e.g., orthopedic manufacturers versus IVD manufacturers.

When policymakers in Congress, the White House, and the executive branch contemplate issues critical to the bottom line of every medical device company, they need to understand that establishing uniform product liability laws, getting predictable FDA product approval times, assuring consistent reimbursement rates at HCFA, guaranteeing future access to biomaterials, preserving research and development tax credits, and the breaking down of trade barriers are all critical elements of this industry's future. The device industry needs to be on the offensive and participate in the debate when the federal government considers policies that affect its core businesses.


A strong manager is needed in the commissioner's office to run FDA. I believe this is possible only if an individual is brought in from outside the federal government. A background in science/medicine may be important, but identifiable success as a manager of a large, multidivisional organization should be the primary criterion. The next commissioner should audit FDA's budget, thoroughly review potential conflicts of interest at the deputy commissioner level, and hire a third-party auditor to analyze the work flow at each center. A CEO-level medical device advisory committee that is representative of the device industry (80% of companies have fewer than 50 employees) should be established to report industry issues directly to the commissioner. David Kessler's lack of interest in medical technology issues was not acceptable.


Without Senator Coats, who is retiring this year, the medical device industry would not have any FDA reform legislation. Coats, Barton, and Eshoo have been the biggest Capitol Hill supporters of this industry for the past three years. Regardless of one's partisanship, these three individuals must be supported. It is unclear who is going to step up to the plate in Coats's absence after 1998. No clear favorite comes to mind other than perhaps Senator Bill Frist (R—TN). As a rising political star in the Republican party, Frist has already been tapped as its primary spokesperson for all health-care-related matters, from slowing the growth of Medicare and Medicaid to reimbursing doctors and debating cloning issues. If Frist can make the time, he is the candidate. One way or another, the void that will be created when Coats retires must be filled quickly.


The device industry has been given an opportunity to build upon its newly established foundation with key leaders in Congress. If, however, it becomes complacent with this recent victory, it will once again struggle to keep up with its counterparts in the pharmaceutical industry. We will never be able to compete head-on with pharmaceuticals in terms of media exposure, but with a cooperative campaign targeted to congressional leaders and committees, we can be effective. The leadership is going to have to come from within—CEOs willing to take the time to come to the nation's capital to keep the industry on the offensive. I hope you all choose to be a part of it.

Jeffrey J. Kimbell was executive director of MDMA from the fall of 1994 through January 1998. He is now CEO of Jeffrey J. Kimbell & Associates, based in Washington, DC.

Copyright ©1998 Medical Device & Diagnostic Industry

Device Industry Wins Big in FDA Reform

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


The FDA Modernization Act of 1997 makes many changes to the agency requested by the medical device industry.

A lthough they came to Capitol Hill nearly four years ago de manding release from FDA oversight, in November most members of the medical device industry felt satisfied with the compromises they had obtained. The FDA Modernization Act of 1997 addresses every cogent industry complaint made about FDA policy and practice since the reform movement began.

For example, no fewer than five rights previously granted to FDA by Congress were repealed:

  • FDA may no longer cite a company for claiming its product is FDA approved.
  • Device distributors no longer need to file medical device reports (MDRs), or register or list their products with FDA (but recordkeeping rules still apply).
  • Companies no longer need to annually certify under the MDR rule. User facilities also will be gradually let off the hook, except for their involvement in a sentinel system to be established by FDA. Under this system, FDA will require MDRs from a subset of user facilities that make up a representative profile of health-care providers.
  • Not all manufacturers of critical or implantable life-supporting devices will need to track their use by patients and practitioners. Instead, FDA must, under explicit statutory restrictions, specifically order that certain Class III devices be tracked.
  • Companies no longer need to submit postmarket surveillance plans for critical, life-supporting, or implantable devices, although FDA may require such plans on a case-by-case basis.

Although these changes may not directly benefit every device manufacturer, they illustrate the revisionist mood that swept Congress in response to industry arguments and FDA's own tacit recognition of its shortcomings.

However, removing old irritations is only a small part of the reform package President Clinton signed into law in November. The new law, which takes effect February 18, contains a number of innovations. The Medical Device Manufacturers Association and HIMA have drafted excellent analyses of these changes.

Under the new act, all Class I devices are exempt from 510(k) premarket notification unless they are intended for a use of substantial importance in preventing the impairment of health or unless they potentially pose an unreasonable risk of illness or injury. The Senate-House conference committee report specifically states that FDA cannot reclassify low-risk Class I devices to a higher category simply to avoid exempting them under this provision. In addition, the agency must publish, by January 20, a list of Class II devices that are exempt from 510(k) requirements. Other devices may be exempted later by FDA as a result of its own judgment or after a manufacturer-initiated petition process. FDA must process such petitions within 180 days or they will be deemed granted.

The new law also shortens 510(k) review times by requiring FDA to consider the extent to which postmarket controls may expedite the classification process, to seek only the least burdensome evidence of substantial equivalence, and to base substantial equivalence decisions on only those intended uses detailed in the proposed labeling. The Office of Device Evaluation may still require a label to state information about uses other than those intended by the manufacturer if there is a reasonable likelihood that the device will be used for such a purpose and that such use could cause harm. Decisions on 510(k)s still are held to the current 90-day deadline.

Reviews for 510(k) submissions also should see faster turnarounds because FDA is to expand its pilot program for independent reviews. All devices will be eligible for third-party review except for Class III devices, Class II devices that are intended to be permanently implantable or life-supporting, and devices requiring clinical data to support their 510(k). FDA has 30 days to act on an independent reviewer's recommendation.

In addition, FDA must provide review priority for devices that represent breakthrough technology, have no approved alternative, offer significant advantages over existing alternatives, or are in the best interest of patients.

Premarket approval (PMA) application review times will be expedited by creating a collaborative review process. In this process, device sponsors will now have a statutory right to meet with FDA within 100 days of its receipt of their submission to discuss any problems. The review will then proceed on a mutually agreed-upon schedule. FDA may also use any company's PMA clinical or preclinical test data, six years after approval, for the purpose of approving another company's device, determining if a product development protocol has been completed, establishing a performance standard or special control, or (re)classifying another device.

Among other compromises obtained, device sponsors may participate in device classification panels and access the nonconfidential data to the same degree as FDA. Also, under strictly limited circumstances, companies may disseminate peer-reviewed and certain other forms of information about FDA-unapproved uses of their legally marketed devices to health-care practitioners, pharmacy benefit managers, health insurance issuers, group health plans, or government agencies.

Finally, FDA guidances must be available to the public in written and—wherever feasible—electronic format during their development. Moreover, FDA employees must be trained in the legal status of guidances—especially that they are not enforceable.

These changes are just a few of the myriad reforms contained in the new law. This law also provides a solid foundation for FDA's recent, self-initiated probusiness reforms. For starters, it gives FDA its first statutory mission statement, directing the agency to improve employee training and requiring it to ensure that its GMP rules conform "to the extent practicable" with internationally recognized quality system standards for medical devices. The act also requires FDA to support the Office of the U.S. Trade Representative in harmonizing regulatory approaches and achieving mutual recognition agreements with other countries.

FDA's new, stated mission encompasses promoting public health through the efficient review of clinical research and taking timely, appropriate action on marketing applications; protecting public health by providing reasonable assurance of the safety and effectiveness of devices; reducing international regulatory burdens and harmonizing requirements; and carrying out its duties in consultation with experts in science, medicine, and public health, and in cooperation with consumers, users, and manufacturers.

Somewhere in the final stages of the act's travels through Congress, it lost the words "and Accountability" from its title. This is not to say that Congress won't hold FDA accountable for its performance—numerous "hammer" provisions require the agency to publish within set deadlines, and it must report its progress to Congress after one-year and five-year periods. In addition, both the General Accounting Office and the Congressional Budget Office will certainly be conducting formal, in-depth investigations of FDA's responses to the new law.

On top of all that, section 406 of the new law requires FDA to develop and publish, within one year, a detailed plan for meeting its statutory obligations. These obligations include establishing a mechanism to ensure compliance with all statutory deadlines by July 1, 1999, and eliminating all backlogs by January 1, 2000. FDA also must publish an annual report in the Federal Register, permitting opportunity for public comment, that describes how it is complying with the act. This report must include statistical information on its performance to date and identify any regulatory policy that has had a significant negative effect on the agency's performance.

Thus, in addition to its congressional watchdogs, FDA will have to contend with the scrutiny of its other diverse constituencies, especially those that have been the most critical in the past. FDA would be wise to keep its doors open to this input during the year so it can address comments in its annual report. The act contains so many reporting provisions now that it almost makes the words that were deleted from the bill's title, "The FDA Modernization and Accountability Act of 1997," sound like overkill.

Copyright ©1998 Medical Device & Diagnostic Industry

Issuing Product Certificates for CE Marking

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


Harald Rentschler, president of Medical Device Certification GmbH (Memmingen, Germany), explains how several notified bodies can issue company product certificates for CE marking.

Is it true that, at least in principle, a legal manufacturer (as defined in the European Medical Devices Directive) may choose notified body A to obtain CE marking for one product family (e.g., Class 2A nonactive devices) and notified body B for another product family (e.g., Class 2A active devices produced with a copacker EN 46002 certified for this class)?

If yes, which notified body will issue the product certificate for the active devices (in this case, according to MDD/annex V)? Both product families need to become part of the manufacturer's EN 46002—certified quality system, but is it possible to include both product families on the same quality system certificate?

There is no limit on the number of notified bodies a company may use if it manufactures different devices. However, a provision of Medical Devices Directive 93/42/EEC states that for any one device, only one application can be filed with a notified body. This requirement leads to the conclusion that it is not possible to cover all devices under one certificate if different product applications have been lodged with different notified bodies.

The certificate for the active devices can be issued by notified body B even though other devices are being certified by notified body A. Notified bodies may consider the results obtained by other notified bodies during their assessments to avoid reexamining certain general aspects.

With respect to a voluntary certification according to the EN 46000 series, all activities of the company can be covered by one certificate.

"Help Desk" solicits questions about the design, manufacture, regulation, and sale of medical products and refers them to appropriate experts in the field. A list of topics previously covered can be found in our Help Desk Archives. Send questions to Help Desk, MD&DI, 11444 W. Olympic Blvd., Ste. 900, Los Angeles, CA 90064, fax 310/445-4299, e-mail You can also use our on-line query form.

Although every effort is made to ensure the accuracy of this column, neither the experts nor the editors can guarantee the accuracy of the solutions offered. They also cannot ensure that the proposed answers will work in every situation.

Readers are also encouraged to send comments on the published questions and answers.

Copyright ©1998 Medical Device & Diagnostic Industry

Implants and Disposables Drive the Search for Stronger Glues

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


Biochemists are teaming up with marine biologists to examine an unlikely source of superadhesive technology.

Formulating the right adhesive for a particular medical device is a tricky affair, constrained only by the laws of chemistry and the expertise of the maker. Suppliers have literally thousands of possibilities at their fingertips, including epoxy resins, silicones, cyanoacrylates, and urethanes in a range of viscosities, strengths, and hardnesses. These adhesives, which must hold a USP Class VI rating to pass muster with FDA, might be cured thermally or by electron beam, visible light, or ultraviolet light. They might glue catheters or syringes, pacemakers or angioplasty balloons. Or they might bond the components inside electronic implantables.

Researchers hope to unravel the secret behind the byssus threads that help mussels adhere to underwater surfaces. Photo courtesy of J. Herbert Waite, PhD.

"If we look through all our products and can't find one that fits the particular need," says Paul Atkinson, senior engineer at Ablestik (Rancho Dominguez, CA), "we set loose our chemists to formulate one."

It is, essentially, R&D on the fly, with adhesives tailored to fit the specific product. Ablestik's specialty is putting together the electronics that run implantable medical devices. Chemists there alter the characteristics of adhesives by changing the type of filler or resin. More dramatic effects are obtained by changing the catalyst. "It is very much like baking a cake," notes Atkinson. "You change some of the components and it changes the flavor dramatically."

High-tech improvements are possible as well. Researchers at Tennessee's Oak Ridge Center for Composites Manufacturing Technology (CCMT) have devised a way to inexpensively alter epoxy resins so they can be cured very rapidly using an electron beam. "Historically, there have been few, if any, suitable epoxy materials out there that were E-beam curable," says Chris J. Janke, principal investigator for CCMT. "We have now found an effective way of modifying these materials." The technology could be used to create materials that equal or exceed heat-cured epoxies in performance, yet neither require long cure times, consume large quantities of energy, nor involve the expensive tooling needed for heat curing.


Still, the myriad choices of adhesives and the iterative enhancements in their formulations can obscure the fact that the current state of the art is sorely lacking. "No one is ever really happy with adhesives," says John M. Questel, president of Adhesive Consultants (Akron, OH), a testing laboratory that specializes in medical adhesives.

Several years ago, a major medical manufacturer sought advice from Questel and his crew regarding aortal balloons, which were still in the development phase. The company needed to attach a Mylar balloon to a polyurethane tube. The balloon had to withstand pressure of about 250 psi in order to clear an obstructed artery. But pumping the balloon up caused it to blow off its mount. "If it pops off in an artery, it sooner or later gets to the brain and kills the patient," Questel says. His team was eventually able to find the right structural adhesive but only after exhaustive research.

An electron micrograph shows the dark granular filaments that hold the byssus of the zebra mussel to the substratum. Photo courtesy of Thomas Bonner, PhD.

Questel comments that balloon pressure ratings are determined largely by the strength of the adhesive. Such limitations are common in the medical device industry, where device specifications are matched to the capabilities of adhesives. Yet adhesives are still favored over other bonding methods. "Many disposable medical devices are now using adhesives for assembly purposes because they are the most economical way to do it," says Questel. "Even though the adhesive is about $1000 a gallon, it is very economical because one drop is all you need to assemble a catheter."

Ideally, new adhesives would offer the advantages of the current breed and greater strength at the same time. That may be possible, but chemists may have to dispense with the known laws of chemistry and embrace the still unknown laws of nature.


"It is downright embarrassing that some sailors make their livings chipping barnacles off ship bottoms — and those bonds are made to pretty dirty surfaces under salty, biochemically active water," says Robert Baier, PhD, professor and director of the Industry/University Center for Biosurfaces at the State University of New York (SUNY) Buffalo. "You would think after all these centuries of chipping them away, we would have learned how to copy them to at least hold two parts of a pacemaker together."

Baier is among a clutch of biochemists at institutions across the United States who believe future successes in medical adhesives will spring from knowledge gained from barnacles, mussels, and algae. These unassuming creatures have the remarkable ability to stick themselves in the midst of rushing water on the propellers and hulls of ocean-going ships, at the mouths of huge water intake pipes, and along the water line of supertankers, "where the detachment forces are most ferocious," as Baier notes.

The glues that hold these very different animals in place are virtually identical, he says. Nature has devised a method for producing this ultimate superglue and conserved it across species. If chemists could make a synthetic version, it might be the ideal bonding material for medical devices designed to be placed inside the human body, which is essentially a bag of water. The problem is that attempts at reverse engineering have failed to reveal what nature has evolved over the past several millions of years. "We have not come to understand the complexity of what occurs in natural systems," Baier says.

Baier has been able to glean, however, that the system appears to be similar, at least generally, to a two-part epoxy, comprising a thin coat of biofilm—the slime that naturally grows on submerged surfaces—and a secretion from the animal. The foot of the mussel, for example, releases an exudate that is basically a complex phenolic compound that interacts with the biofilm on the underwater surface to form an adhesive.

There are two major barriers to synthesizing this exudate. One is finding out how to add the requisite second hydroxyl group to the benzene ring, which together form the phenol—an amino acid called tyrosine. "Nature somehow makes the extra hydroxyl group on that tyrosine outside the cell," Baier says. "We don't know how that is being done." The second barrier is learning how to cross-link the compound to provide strength against shearing.

Baier and his SUNY colleagues may be near conquering the first barrier—the oxidation of the tyrosine to form a synthetic compound similar to the mussel exudate. They have programmed bacteria to synthesize a major portion of this compound. But there is a problem. "Even if we could make a quart jar of the mussel adhesive cement in the lab, the chance of getting it to cross-link and harden like an epoxy cement is very low right now because we haven't yet discovered how the cross-linking takes place in nature," Baier says. The answer appears to involve enzymes that catalyze this molecule into a sticky form.


An understanding of this transformation might come from transmission electron micrographs being taken by Thomas Bonner, PhD, professor of biology at SUNY Brockport. Bonner's micrographs show the adhesive as granules of glue resting beside granules of the enzyme, a type of catechol oxidase. Evidence suggests that these granules open up and merge. An understanding of how this process occurs may come from a detailed analysis of the extraordinarily complex structure of these granules, as well as from studies of material called byssus secreted by the foot of the mussel.

"Electron micrography allows us to see the way the byssus is organized at a high level of magnification," Bonner says. "Then we get to look at the interface between the byssus and the substrate to understand the roles they play."

The byssus, explains Bonner, is outside the living cells that compose the animal's foot, and is the mussel's equivalent of fingernails. Both the byssus and human fingernails are composed of protein. But unlike fingernails, the byssus forms tethers, called byssus threads. The adhesive connects the byssus to the underwater surface, and the byssus tethers the mussel to the surface.

Studies at SUNY Brockport indicate that after the granules become part of the byssus and participate in the attachment process—the actual gluing—they are transformed into a structure composed of very densely packed filaments. Still evolving is an understanding of the role played by the biofilm that coats the surface being attached. This slime is made of living organisms, primarily bacteria, some algae, and perhaps fungi, along with decomposing organic material.

An electron micrograph shows the dark interactive surface of a natural animal glue. Photo courtesy of T. Bonner.

"There is preliminary evidence that the byssus is initially released in a semi-liquid or gel state, which then flows into the interstices of the biofilm and solidifies, trapping parts of the biofilm in the byssus itself," Bonner explains. "The reason we think this is that we find a fair number of bacteria trapped in the bottom of the bys-sus where it interacts with the surface."

Bonner is studying the zebra mussel, best known for its propensity to clog water intakes for power plants and water treatment plants along the Great Lakes. Marine animals such as the New England blue mussel also secrete a byssus and threads, but in these animals the threads are elastic and serve to absorb the force of the water as it rushes past. These threads elongate and then recoil to their original position when the force is removed. This natural "shock absorber" enhances the animals' ability to adhere to the surface and may provide a model for developers of medical adhesives.


The connector between the two surfaces might be a monolayer—rather than a long tether—composed of some attachment factor, suggests J. Herbert Waite, PhD, a professor of marine biology and biochemistry and joint professor of chemistry and biochemistry at the University of Delaware College of Marine Studies (Newark). "Here you would not necessarily have to worry about curing the adhesives; you could exploit the fact that the adhesive can stick opportunistically to any hard surface."

Waite is dissecting natural adhesives into their tiniest components, while search-ing for the underlying rules that will make sense of the data being uncovered in his laboratory. Unfortunately, performing reverse engineering is not a viable option. "Doing so assumes that these organisms subscribe to our rules," Waite explains. "Since nature has shown time and time again that it is capable of defining its own rules, the approach is fraught with uncertainty."

It is abundantly clear that the chemical functional groups on the macromolecules created by these animals are not present in any of the current generation of manmade adhesives, Waite says. That in it-self is a great motivator to push on with the research.

One of the great mysteries driving Waite involves the cross-linking that occurs within the adhesive, specifically the way this protein folds in upon itself. Current chemical understanding dictates that protein folding occurs in the middle of a molecule, which would draw the material together, minimizing its interface with the surface and, consequently, reducing adherence. But this natural superglue does the opposite. It spreads out, maximizing surface-to-volume ratio. "It certainly is not doing what other proteins are doing," Waite says.

The explanation will depend on coming up with new rules to explain the behavior of this protein. Waite is confident that the rules can be determined without rewriting current principles of organic chemistry. The research now being conducted will eventually produce effective models to explain the actions of these macromolecules. "This will allow people to make effective adhesive molecules, as well as the enzymes needed to synthesize them," Waite adds.

Producing these natural superglues in bulk might be a great help to the medical device industry. Researchers in this field might learn enough from the natural processes to build a "peptide maker" that could produce quantities of these glues on demand. Among the leading candidates are yeast and virally infected insect cell lines. The infected insect cells intrigue Waite the most because they have a lot of the enzymes that appear to be needed to do the processing that gives the adhesive proteins their requisite "stickiness."

Exactly when such molecular production lines will be available, however, is difficult to predict. Waite believes science may be able to imitate the adhesion of marine invertebrates within 10 years. "But that may not be the direction we want to go," he says. "When you characterize a very complex strategy, you have hurdles that you overcome along the way. The proteins and partial strategies that come from these can frequently have very useful spin-offs that can be applied without knowing the whole answer."

Copyright ©1998 Medical Device & Diagnostic Industry

Empowering Your Employees

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


Kaizen gives employees the power to make rapid changes and keeps companies competitive.

When Perfecseal (Philadelphia) adopted the Japanese management philosophy kaizen, or continuous improvement, in 1994, the company saw dramatic results. With the help of TBM Consulting Group (Durham, NC), Perfecseal set up a week-long kaizen event. The long-term rewards of kaizen included improvements in Perfecseal's productivity. "Our 49-day lead time dropped to 12 days," says John Martis, operations vice president. "Our on-time and complete external shipments were about 38%; now they're 95%. We had a 50% improvement in productivity measured in sales per team per day, and a 50% reduction in setup time."

In addition, "employee response was very favorable," Martis says. "People who have been involved in kaizen events not only voice their opinions but actually have a hand in making changes. If it is a week-long event, at the end of the week they can say, 'We had this idea and we got it done.'"

In 1996 at Acuson (Mountain View, CA), inventory accuracy was fluctuating between 85 and 95% despite a cycle-counting system. A kaizen team assigned to the problem in mid-1997 discovered that the problem was not with the employees doing the counting but with the outdated and poorly documented procedures for tracking inventory. The kaizen team rewrote the procedures in an easy-to-understand format. By October 1997, inventory accuracy reached 99%.

"That's an example of a problem in which the real root cause would never have been identified without talking to the people who actually do the counting and the inventory transactions," Pat McMahon, continuous improvement manager for Acuson, says. "The kaizen team felt great because, for the first time, they were given credit for knowing something about the process. I think they felt complimented that management gave them such a vote of confidence."

Companies across the United States are using kaizen to reduce manufacturing and delivery time and increase employee satisfaction.


"Kaizen is a process for improving the elements of the manufacturing system," says Jeff Madsen, director of continuous quality improvement at Johnson & Johnson Medical­Vascular Access (Southington, CT). "During kaizen, employees look at how you move material, how many people you use, what the net output of a system is, how much inventory you have, how you schedule, how you maintain and attain equipment reliability--all the elements that go together to make up the total production system of turning materials into finished products."

The central concept of kaizen, which was popularized in the United States by Toyota in the 1980s, is to give employees the power to solve problems and the resources and opportunities they need to do so, without spending a lot of time in meetings getting approvals. Kaizen emphasizes action. Teams don't sit in rooms thinking about problems or designing plans; they are out on the floor trying new techniques until they find pragmatic ways to improve performance. A team's time frame for establishing change can be as little as two days or as long as a week. Sometimes teams are given a longer period--usually 30 days--for large-scale projects.


Although kaizen itself is a system of rapid change, kaizen training can take several days. Tom Moran, director of business development for TBM Consulting Group, describes a first-time kaizen session as a "five-day, one-night blitz." The first day is devoted to team training. The second day employees gather data about a problem and brainstorm solutions (the second night may be spent physically changing the work environment if it seems necessary after the brainstorming session). On the third and fourth days, employees test the solution, make adjustments, and document each change as it is made. On the fifth day, the kaizen team presents its solution.

As a company becomes more familiar with the kaizen process, training time can be reduced. "When we first started, we probably spent a day in training before we let our teams begin work," says Madsen. "Now we spend about half a day."

Kaizen training generally consists of three phases. The first phase stresses the need to improve operations to remain competitive. "In every kaizen, there's a concrete result: eliminating steps in the process, eliminating unnecessary work, or reducing the staffing level," says Madsen. "The solution might involve using less manufacturing floor space for your equipment and people. There are many, many small ways in which you reduce waste or become more effective as a result of a kaizen team's work."

The second phase of the training process addresses kaizen philosophy and methodology, showing how they fit into a plant's production cycle or processes. "We gather data on the current situation through time observation, counting inventory, measuring square footage of the floor space, and creating what we call a spaghetti diagram, which follows the steps of the process all over the factory and sometimes outside," says Moran. "Then we brainstorm, maybe come up with a new layout for the equipment or process or with a method to reduce the inventory that allows flow to continue but reduces the amount of inventory that's in process." Kaizen treats any point in the manufacturing process where goals are not being met or there is room for improvement as a potential problem to be solved.

The third phase of kaizen teaches teams analytical data collection and problem-solving techniques, including training in working as a team, creating agendas, keeping minutes, handling action items, and focusing on the team mission. Teams are taught skills such as brainstorming; making checklists, diagrams, and surveys; and creating histograms, Pareto charts, run charts, and statistical process control charts.


Although building teams and teaching them problem-solving skills is important to kaizen, getting input from every employee is what makes kaizen work. Kaizen philosophy asserts that every employee, regardless of education or experience, has valuable suggestions to contribute. For example, new employees are frequently most aware of processes that are difficult, confusing, or poorly documented. Diverse life experiences can also shed new light on a problem. Respecting every employee's potential contribution to the improvement process is kaizen's greatest strength, says McMahon. As an example, he points to Silicon Valley's large immigrant population. "Some of those people have gone through life experiences that I hope never to encounter, and they know problem-solving. They have all kinds of clever and innovative ways of dealing with difficulties."

Customer comments about products or services that need improvement are another place to start looking for areas to improve. This also is a good time to look outside the medical device industry for process improvement ideas and skills. Often, whole industries develop bad practices because they compare themselves only to each other. Many companies find that they learn the most not from studying their competitors but from seeing how companies outside their industry do things. For example, John Martis says Perfecseal benchmarked itself against Prince Corp. (Holland, MI), an automotive interior manufacturing facility, and Johnson Controls (South Bend, IN), a manufacturer of thermostat controllers. Pat MacMahon notes that Acuson has learned its statistical processes from National Semiconductor (Santa Clara, CA) and survey techniques from The Gallup Organization (Princeton, NJ).


Like any process, kaizen has its drawbacks, primarily caused by the speed of implementation and the top-down corporate effort and commitment needed for this management technique.

Keeping up with new procedures can be difficult when changes are made so quickly. "We are an ISO-certified facility," says Martis, "so we have to make sure that our standard operating procedures accurately reflect what we're doing. You have to be certain that you follow up on the paperwork side of kaizen." In addition, he notes, if too many kaizen problem-solving projects are put on a 30-day schedule instead of completed during a short-term kaizen event, they can become difficult to manage. "As we went through the kaizen process, we learned to focus on getting things done during events and not letting a lot of things languish afterward."

Effort, attention, and managerial support are necessary for kaizen to work. "Kaizen is probably more difficult than it appears," Madsen says. "People who are beginning to do these things for the first time may underestimate the amount of effort it takes." For instance, team formation and training take time, which can temporarily disrupt the production process. Productivity may also take a dip while kaizen-driven changes are being made. Short-term slowing may discourage a company from pursuing longer-term goals.

Kaizen also can fail if it is given only lip service, says Sheila Kessler of Competitive Edge (Fountain Valley, CA), a consulting firm whose clients have included Beckman Instruments, Inc. (Fullerton, CA) and Johnson & Johnson (New Brunswick, NJ). The company must support a culture open to employee-driven, rather than manager-driven, change. "In our culture, our executives and our managers are supposed to know the answers, but with kaizen our executives and our managers need to know the right questions," Kessler says.


When it works, kaizen can be beneficial to the company and rewarding for employees. More than a simple series of process improvement steps, kaizen affects the entire company culture: it encourages open communication, continual change, teamwork, and taking personal responsibility for the day-to-day procedures one uses on the job.

By striving to improve all elements of the production system, medical device companies can streamline and improve their manufacturing processes. In turn, employees can apply firsthand experience to improve their working conditions and, as members of kaizen teams, become active agents of change within their company.

Kim Campbell Thornton is a writer based in Lake Forest, CA.

Illustration by Ken Coffelt

Copyright ©1998 Medical Device & Diagnostic Industry

Design Considerations for EMI Gaskets

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


Selecting the proper EMI gasket is a complex task that requires careful examination of all criteria.

T he medical device industry clearly understands how to use EMI gaskets to address the design issues associated with electromagnetic compatibility of electronic products and to meet regulatory requirements. However, what is not universally grasped are the specific design considerations necessary for selecting the proper gasket for a particular product. This article describes different types of EMI gaskets and their applications. Moreover, it examines the specific issues that determine the appropriate gasket choice.


EMI gaskets can be grouped according to their material composition.

Wire Mesh. Wire mesh gaskets are composed of tin-plated, copper-clad steel or Monel wire that has been knitted on industrial knitting machines and formed into circular or rectangular cross sections. In other versions, the wire is knitted over a silicone or neoprene sponge core or over a solid silicone core with a hollow cross section. Other variations include attaching the wire to nonconductive elastomers for environmental sealing to metal frames. Compressed mesh gaskets are also available. The most recent addition to this class of gaskets is a tin-plated steel wire mesh gasket with an air core.

An array of typical EMI shielding gaskets used in protecting medical electronics equipment. Photo courtesy of Chomerics Div., Parker Hannifin Corp. (Woburn, MA)

Finger Stock. These gaskets are formed beryllium-copper spring fingers in various cross sections, including spiral wrapped. They can be tin or nickel plated.

Wires Oriented in Elastomer. Such gaskets are spring-loaded, Monel wires oriented in one direction in either a solid or a sponge silicone matrix.

Metal Screens Impregnated with Elastomer. A woven aluminum screen impregnated with silicone or neoprene or an expanded Monel foil impregnated with silicone elastomer characterizes these gaskets.

Conductive Yarn, Conductive Fabric, or Foil over Nonconductive Foam. Nonconductive foam serves as a common base for gaskets that add the following conductors: silver-plated nylon yarn; nickel-, copper-, or silver-plated fabric; or reinforced aluminum foil. These types of gaskets are most prevalent in commercial electronics because of their low cost, low closure force deflection requirements, and relatively high shielding effectiveness.

Conductive Elastomer over Nonconductive Elastomer. These gaskets usually feature a silver-filled conductive elastomer over a nonconductive silicone core.

Conductive Elastomer. Typically, these gaskets are silicone elastomer filled with electrically conductive particles such as carbon, nickel, nickel-plated graphite, silver-plated aluminum, silver-plated copper, silver-plated nickel, silver-plated glass, or pure silver. Other conductive elastomers include fluorosilicone, fluorocarbon, and ethylene-propylene terpolymer (EPDM) elastomer. It is possible to mold or extrude the elastomer into many cross sections or to robotically apply it as a form-in-place gasket. The latter option is the most recent development in this area.


EMI gaskets generally have uses in two areas: low-impedance grounding and EMI shielding. In low-impedance grounding applications, gaskets provide a low-impedance ground so that the metal structural parts that form the chassis won't be affected by internal electromagnetic fields and therefore won't contribute to the radiated electromagnetic fields within an enclosure.

Using a gasket for EMI shielding is the more traditional application. The gasket seals an otherwise shielded enclosure to prevent electromagnetic leakage from the seams of mating flanges. In addition, EMI gaskets can also protect a device from airflow and other environmental conditions.


Electrical, mechanical, and environmental issues can all affect the design of a device and therefore the selection of device components, including gaskets.

For enclosure flange design, the flange width, fastener types and spacing, flange surface treatment, and means of gasket attachment must all be factored into the gasket selection process. EMI gaskets must be properly deflected to perform. They must also be mated against a conductive surface treatment to avoid being electrically insulated.

Further electrical requirements include conductivity of the gasket itself; shielding effectiveness against electric, magnetic, and plane wave electromagnetic fields; and transfer impedance. The most important thing to remember about electrical requirements is that there is no universally accepted means of measuring electrical performance. Each manufacturer is likely to measure performance using a slightly different procedure, especially when determining shielding effectiveness. It is important that designers understand the gasket performance data presented, and in particular how they relate to a specific design issue. For example, shielding of a handheld medical electronics enclosure must meet certain radiated emission requirements for the European Union. It is often necessary to rely on a manufacturer's technical personnel, EMI test personnel, or EMI consultants for such information. It is also important to note that not all gasket types perform the same way against lower-frequency magnetic fields or at plane wave frequencies above 1 GHz.

Designers also must consider requirements for compression and deflection, the difference in height achieved when a soft gasket is pushed down. EMI gaskets must be properly deflected to perform as intended. Each type of gasket and each cross section of that gasket has a unique compression and deflection curve, which defines the gasket's range of deflection. A design that does not affect the minimum deflection does not allow the gasket to provide the necessary electrical conductivity. In the case of over-deflection, however, the gasket may be permanently damaged or may have an unacceptable compression set.

Flange designs must allow for the effect of compression set on EMI gaskets. Certain gasket types, specifically wire mesh gaskets, can exhibit severe compression set, which in turn prevents them from providing a repeatable sealing function.

Specific mechanical requirements such as tear resistance, elongation, du-rability, or performance in shear force applications must also be taken into account. If a design subjects the gasket to repeatable shear, it may not survive very long. Therefore, designers should examine gasket solutions that include finger stock. For applications that require increased durability, designers should consider elastomer gaskets reinforced with Dacron.

Of course, environment plays a key role in how a gasket will perform. Designers should consider the temperatures and any liquids, such as saline solution, a gasket may encounter. In the medical field, fluids can include saline solution or other chemicals. Further, electronic products exposed to high humidity or salt air may corrode. To seal a product for use in a corrosive environment, wire mesh gaskets or conductive elastomer gaskets must be designed either with environmental seals or as comolded or coextruded parts. Conductive elastomers can be used with an acknowledged hierarchy of gasket materials that resist corrosion in flanged joints. The three materials in descending order of performance are: a passivated silver-plated aluminum in fluorosilicone, silver-plated aluminum in silicone, and nickel-plated graphite in silicone. For single gasket designs, it is important to choose the materials based on available corrosion resistance data. Designers again face the problem that the industry has no universally accepted corrosion test for EMI gaskets, and therefore must rely on manufacturers' personnel and consultants for guidance. Designers must carefully examine corrosion test data to ensure that the test method properly replicates the gasket design.


Selecting the appropriate EMI gasket is relatively complicated, and the array of gasket choices reflects this complexity. Choosing the correct gasket for a particular problem requires an awareness of gasket design criteria. Although this article has provided some basic guidance for selecting the proper EMI gasket, designers must remember that EMI gasket performance can be verified only when installed in a system and tested accordingly.

Joseph Butler is the market manager for Chomerics Div. of Parker Hannifin Corp. (Woburn, MA).

Copyright ©1998 Medical Device & Diagnostic Industry

Choosing Semiconductor Components for Medical Products

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


The selection decision depends on which type of integrated circuit best meets the product's technical and economic requirements.

The drive to reduce costs by moving patients from hospital to ambulatory care as rapidly as possible has increased the demand for portable medical devices and challenged medical device designers to find semiconductor components that can meet these devices' technical and economic requirements. Portable medical devices must consume less energy and take up less space than nonportable devices. They also may require remote or wireless communication features. Three types of integrated circuits (ICs) are generally used in medical applications: standard ICs, semicustom ICs, and mixed-signal application-specific integrated circuits (ASICs). Ten technical and economic issues should be considered when choosing which component to use: power, size, design flexibility, reliability, functionality, design security, electromagnetic compatibility, cost to design, cost to manufacture, and time to market (Table I).

Variable Standard Semicustom Custom
Power Minimum Average Maximum
Size Minimum Average Maximum
Design flexibility Minimum Average Maximum
Reliability Minimum Average Maximum
Functionality Minimum Average Maximum
Design security Minimum Average Maximum
Electromagnetic compatibility Minimum Average Maximum
Cost to design Maximum Average Minimum
Cost to manufacture Minimum Average Maximum
Time to market Maximum Average Minimum

Table I. Ranking of technical and economic variables for each semiconductor manufacturing option according to the amount of benefit offered.


Standard off-the-shelf analog and digital semiconductor components include microprocessors, analog-to-digital converters (A/Ds), digital-to-analog converters, and operational amplifiers in addition to memory devices such as RAM, ROM, and EEPROM. The chief benefit of using standard semiconductor components is that they already exist, decreasing time to market and design cost. However, design flexibility is low—limited to products that already exist—and the components may not be able to meet the more rigorous miniaturization and power consumption demands of portable medical equipment. Moreover, designs using standard components are easy to reverse engineer, decreasing design security.


Semicustom integrated circuits rectify some of the limitations of standard ICs. They provide designers with more flexibility because digital functions can be designed into a standard gate array, programmable gate array, or digital ASIC. Limited analog functions also can be added. The product's overall power consumption and size can be decreased while the digital portion of the design is tailored to meet precise requirements.

The time to market of products using semicustom ICs is generally longer than those using standard components. Digital ASICs are almost always production qualified within one year from start of design, although designers using a field-programmable gate array can significantly decrease this time. The design cost will also be higher because the semicustom IC must be designed and fabricated for the specific application. However, new digital design software is available that allows designers to work at the circuit behavioral level using high-level hardware description languages. This software allows designers to simulate the behavior of their high-level designs, verifying circuit behavior and performance. Once this is done, designers can run synthesis software to compile the behavioral descriptions and produce a list of cells and their interconnections (a netlist) ready for ASIC layout. Other software can draw a useful schematic from the synthesized netlist. Because all of the device's digital functions are integrated into a semicustom IC, it is more difficult to reverse engineer, increasing design security.

Semicustom ICs solve many medical device performance, size, design flexibility, and design security problems. However, semicustom ICs are generally limited to digital functions, which may not be sufficient for a device's requirements.


Custom, mixed-signal ASICs combine analog and digital circuits on a single chip. Designers can opt for exactly what they need—specifying an 11-bit A/D, for example, instead of an off-the-shelf 12-bit A/D. Mixed-signal ASICs can reduce the number of components used in a circuit board from 20 or 30 to as few as 5 or even 1. The accompanying reduction in the number of long-wire connections and board interconnects makes devices using ASICs more electromagnetically compatible, i.e., less likely to generate and less susceptible to electromagnetic fields. Further, end-product reliability increases when the number of components making up the circuitry decreases, and manufacturing costs fall due to fewer manufacturing steps and components purchased and inventoried.

Because mixed-signal ASICs permit designers to specify exact pinout configurations, connections and packaging are simpler and more direct. In addition, if ASICs are obtained as flip chips, they—and thus the product—can be smaller. Custom ICs provide smaller size, greater energy efficiency, and better design flexibility than standard or semicustom ICs. Moreover, attempts to reverse engineer the product are almost impossible because most or all of the circuitry is included on one IC.

These benefits come at a price. The design cycle for products using custom mixed-signal ASICs is significantly longer. A minimum of one year from design start to receiving qualified production parts is the norm, but the time can be much longer depending on the ASICs' complexity. Engineering costs will be higher because—unlike all-digital ICs, which frequently can be designed in-house—mixed-signal ASICs must be created by experienced analog designers who are typically only available through outside sources. The complexity of an ASIC and, therefore, its development time and cost, becomes greater as a result of increased digital gate count and analog functions. Thus, design work must be initiated with enough lead time to compensate for these factors.

Identifying ASIC Designers. Mixed-signal ASIC development requires the manufacturer to match the needs of the medical device with the capabilities of potential partners and existing technology.

Questions to ask at this stage include:

  • What are the main functional requirements of the ASIC?
  • Are there any special testing or screening requirements?
  • What are the trade-offs between size and power?
  • Will the device benefit from having gold or solder bumps for flip-chip assembly?
  • Must the die be backlapped to make it thinner?
  • What are the voltage requirements?
  • Are there low- and high-voltage requirements on the same IC?
  • What are the electrical noise requirements?
  • What are the key functions, such as A/D conversion and memory requirements?
  • What passive components (resistors and capacitors) should be integrated into the ASIC?
  • What are the electrostatic discharge protection requirements?
  • Throughout what temperature range must the part function?
  • How will the ASIC be tested, both at the vendor's facility and in the final product?

The answers provided by potential partners will help the manufacturer identify qualified sources for designing custom ASICs.

Budgetary and Technical Information. Once the list of potential ASIC designers has been narrowed down to those who are technologically qualified, the manufacturer should meet with each potential vendor for an in-depth review of the requirements and end product and ask for a budgetary quotation.

All economic aspects—such as exploring cost reduction trade-offs between bare die, flip chips, and packaged parts—should be reviewed. Manufacturers should compare manufacturing costs using standard or semicustom ICs against those using ASICs. A higher cost for an ASIC may be offset by reductions in procurement, inventory, manufacturing, or assembly expenses.

Low annual volumes and the potential liability for medical product ASICs may preclude some vendors from responding to the request for quotation. Those quotations that are received should be accompanied by a discussion of pertinent technical issues and information on application-specific experience.

Questions to be asked at this stage include:

  • Has the vendor designed medical circuits in the past?
  • Did the vendor identify any risks?
  • Did the vendor discuss the method to be used to design the ASIC (i.e., use previous cells or design new cells)?
  • Did the vendor delineate responsibilities? For example, will the vendor perform all design and test development? If not, how will these steps be handled?
  • Has the vendor designed similar circuits?
  • Does the vendor have reusable blocks and cells already designed?
  • Are the vendor's internal quality systems certified against an internationally recognized standard such as ISO 9000?

The manufacturer should be wary of budgetary quotations that do not address these technical issues and provide only a price. Unless the ASIC is extremely simple, some technical discussion should always be included with the quotation.

Manufacturing and Financial Considerations. The manufacturer should also study each vendor's financial strength and access to manufacturing facilities. The length of time that the vendor has been in business and the portion of the manufacturer's business devoted to ASICs and to the medical market are all-important. In addition, the device manufacturer should ask whether the vendor has its own fabrication facility (fab) and, if so, how many. If the vendor is simply a design house that can use any fab, the manufacturer should question how committed the fab will be to the manufacturer's business now and in the future. A vendor with limited financial or manufacturing capabilities will probably not be a successful partner.

The manufacturer's overriding objective should be to select a vendor who will be involved with current and future products and ASICs. Price can be most beneficially addressed in the final stages of evaluation, when only fully qualified candidates are being considered. Paying more for an ASIC may make sense if the main cells have already been designed and proven. A higher nonrecurring engineering expense may make sense if a new architecture saves space and decreases the amount of software code needed. In short, price is just one of many parameters to consider when selecting an ASIC vendor.

Management Objectives and Philosophies. Once a vendor has been selected, the manufacturer should arrange regular meetings throughout the design process. Because some product specifications may have changed during the quotation period, all requirements should be reviewed before the design starts. If some specifications are more flexible, that fact should be noted. Adequate resources and technical support should be provided for the project, even if the entire design is being completed by the ASIC partner. Some trade-offs probably will have to occur as the design takes shape; a slight change in a noncritical specification may result in significant yield improvement, decreased design time, or decreased risk.

The manufacturer should be an integral part of all design reviews and insist on periodic updates of milestones and development schedules. There should be advance agreement on the definition of a prototype or first article part; what comprises design validation, preproduction, and production; and what is required by both parties before releasing the final ASIC to production.

If the end assembly will be produced at a contract manufacturer, the manufacturer should meet with the contractor during the design process. Mechanical samples from the ASIC vendor should be taken to the contract manufacturer as early as possible.

The device manufacturer should know how much testing of the ASIC is required and how to test the end product in which the ASIC will be used. Furthermore, the manufacturer should be prepared to relax testing requirements where possible to keep piece part prices low.


The demand for portable medical products will continue to grow, increasing the requirements for lower-power operation, smaller size, better performance, greater reliability, and design flexibility. Medical device manufacturers should consider their options when choosing ICs for their products and take care to consider a variety of technical and economic issues when looking for a vendor.

Jim Gentile is manager of ASIC products at Mitel Semiconductor (San Diego).

Copyright ©1998 Medical Device & Diagnostic Industry

A Guide to AAMI's TIR for EtO-Sterilized Medical Devices

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


A step-by-step approach to applying ANSI/AAMI/ISO 10993-7:1995 and AAMI TIR-19 to EtO-sterilized medical devices.

While sterilization is a necessary element in the medical device manufacturing process because it destroys viable organisms, one should not be surprised that all sterilization procedures are potentially hazardous to humans. Limits for residues remaining on devices sterilized with ethylene oxide (EtO) have been set by a number of national pharmacopoeias and by FDA in a proposed rule published in 1978.


In 1990, the International Organization for Standardization (ISO) through its Technical Committee (TC) 194, Biological Evaluation of Medical Devices, established Working Group (WG) 11 to develop a standard for EtO residuals, resulting in the publication of ISO 10993-7 in 1995. In 1992, the Association for the Advancement of Medical Instrumentation (AAMI) EtO Sterilization Residuals Working Group, a technical working group of the AAMI Sterilization Standards Committee, formed a task group to write a technical information report (TIR) that would guide manufacturers and regulators responsible for EtO-sterilized medical devices as they applied ISO 10993-7. At the same time, FDA convened an internal committee to consider the application of the ISO standard to the review of submissions of EtO-sterilized devices and how (and whether) to use the ISO standard to replace the 1978 proposed rule.

The ISO standard for EtO residues was accepted through parallel ballot by ISO and the European Committee for Standardization (CEN) and was quickly accepted as an American National Standard.1,2 However, in a letter to Medical Device & Diagnostic Industry, Donald E. Marlowe, director of the Office of Science and Technology at FDA's Center for Devices and Radiological Health (CDRH), indicated that FDA still had a number of questions concerning the application of the ISO standard and was thus postponing the implementation of ISO 10993-7:1995 as a replacement for the FDA 1978 proposed rule.3,4 FDA and others had expressed concerns about the manner in which certain issues, including multiple device use and use of devices on neonates, were addressed in the 1995 ISO standard.

ISO continues to refine ISO/DIS 14538, which is a general procedure for deriving permissible limits for sterilization and process residues using health-based risk assessment, and this activity probably will take another two to three years to complete. The ISO committee responsible for the preparation of the EtO residue standard plans to revise the current standard by applying ISO 14538 to the generation of the allowable limits for EtO.


In its quest to guide manufacturers and regulators through this new standard, in August 1996 AAMI's TIR task group concluded that a flowchart with an accompanying text was the most practical approach. The group also developed test protocols for manufacturers to use to simulate product use and show conformity with the requirements of the standard. AAMI's EtO Sterilization Residuals Working Group reviewed the TIR flowchart and simulated-use extraction protocol in January 1997, and the ISO/TC 194/WG 11 examined the materials in April.

Some staff members of CDRH's Office of Device Evaluation who are responsible for the review of premarket approval applications and premarket notification (510(k)) submissions raised some concerns about the specific language of the accompanying text at a joint meeting between FDA and the TIR task group in August. The revised text from this meeting, with the flowchart and simulated-use extraction protocol, has since been circulated to the full EtO Sterilization Residuals Working Group, which agreed to some minor editorial changes to the document at its meeting in September 1997 and recommended to the AAMI Sterilization Standards Committee that the TIR be circulated for ballot to establish final consensus before publication of AAMI's TIR.

At the September 1997 meeting, Donald Marlowe indicated that FDA would use ANSI/AAMI/ISO 10993-7 accompanied by the guidance documents provided in the TIR as the basis for evaluating the EtO residue requirements for EtO-sterilized medical devices. He noted that the agency had not yet determined precisely how it would proceed since the 1978 proposal also applies to drugs, which are regulated by another center at FDA and are not addressed by ANSI/AAMI/ISO 10993-7. There are a number of options available to CDRH, including reopening the proposed rule making or publishing a blue book memorandum on this topic. The agency will decide on the appropriate course to take once the TIR is published early this year.


AAMI TIR-19 is intended to assist manufacturers in applying the standards of the ISO 10993 series to the biological evaluation of EtO-sterilized medical devices. The international standard ISO 10993-7, "Biological Evaluation of Medical Devices—Part 7: Ethylene Oxide Sterilization Residuals," specifies the requirements for establishing allowable limits for EtO residues and the analytical methods to show that an EtO-sterilized device is in compliance with the allowable limits.

The standard specifies maximum allowable residues for ethylene chlorohydrin (ECH); however, no exposure limits for ethylene glycol (EG) are set. When ISO/TC 194/WG 11 experts conducted risk assessment for EG, they found that when EtO residues are controlled, it is unlikely that biologically significant residues of EG will be present. Note that "dose to patient" is the basis for establishing the allowable limits and the reference method for showing compliance with ANSI/AAMI/ISO 10993-7.

The flowchart outlining the steps necessary to apply the standard is shown on pages 74—75 (editor's note: the flowchart could not be reproduced in electronic format). The numbers appearing within parentheses in the flowchart indicate specific clauses within the TIR text. When the TIR states "reduce EtO," manufacturers should extend the aeration time for the medical device and/or raise the aeration temperature.

Meeting the biological testing requirements for each individually designed medical device as indicated in ANSI/AAMI/ISO 10993-1, combined with satisfying the EtO-sterilization process residue limits, forms the justification that an EtO-sterilized device is acceptable for use with regard to its biocompatibility.


Informative annex A of TIR-19 provides an outline to enable users to develop a simulated-use extraction procedure. Water should be used for simulated-use extraction of EtO residues.5 Devices that contact the body in any way during use should be extracted at 37°C, and the conversion of EtO to EG should be evaluated. Devices having no immediate body contact during use (e.g., hypodermic syringes) should be extracted at 25°C.

In determining the appropriate extraction time for a device, testers should consider the expected, reasonable worst-case use time the device would encounter. In addition, it may be useful to refer to clause in ISO 10993-7:1995, which suggests that analysts establish extraction rates for EtO from various devices at various use temperatures. The minimum extraction time is 1 hour.

Any pretreatments a device must undergo prior to use, e.g., priming, should also be performed before a device is extracted. If the device is filled with water before the extraction, care should be taken to avoid air pockets. If use of the device involves circulation of fluids (e.g., blood, dialysis fluid), the extraction process should simulate the fluids circulating in a manner consistent with product use. Note that where blood is returned from the device to the patient it must be assumed that any EtO will stay in the body. Manufacturers must document the rationale for using the conditions established. Devices of similar design but different sizes may be grouped, with the worst-case conditions for the group selected for testing.

When evaluating EtO residues on device kits and trays, manufacturers should determine residue levels for each EtO-absorbing patient-contact device, and then choose a worst-case device to use in testing.


We hope that manufacturers and regulators who are responsible for producing and evaluating EtO-sterilized medical devices will find the guidance contained in AAMI TIR-19 a useful adjunct to ANSI/AAMI/ISO 10993-7:1995. We also hope that annex A to TIR-19 will enable analysts and others responsible for evaluating EtO residues on medical devices to develop adequate test protocols. These documents should enable those responsible for the biological evaluation of EtO-sterilized medical devices submitted for regulatory approval to make appropriate, timely decisions.

In a follow-up article this spring, we will examine the mechanism by which FDA will implement ANSI/AAMI/ISO 10993-7:1995, and we will evaluate the similarities and differences between the ISO standard and the FDA 1978 proposed rule, which has been used by both industry and the agency. In addition, we will discuss the application of ISO 10993-7:1995 to certain devices, including hemodialyzers, custom kits, and surgical gowns and drapes.


1. ANSI/AAMI/ISO 10993-7:1995, "Biological Evaluation of Medical Devices—Part 7: Ethylene Oxide Sterilization Residuals," Arlington, VA, Association for the Advancement of Medical Instrumentation, 1995.

2. Page B, "ISO Standard Redefines Limits for EtO Residuals," Med Dev Diag Indust, 18(6):68—73, 1996.

3. Marlowe DE, "ISO Standard on EtO Sterilization," Med Dev Diag Indust, 18(7):32—33, 1996.

4. "Ethylene Oxide, Ethylene Chlorohydrin and Ethylene Glycol-Proposed Maximum Residue Limits and Maximum Levels of Exposure," Federal Register, 43FR:27474-27483, 1978.

5. Kroes R, Bock B, and Martis L, "Ethylene Oxide Extraction and Stability in Water and Blood," personal communication to the AAMI committee, January 1985.

Barry F.J. Page is an industry consultant, a member of the AAMI Sterilization Standards Committee, cochair of the AAMI Ethylene Oxide Sterilization Residuals Working Group, and was convener of ISO/TC 194/WG 11 when ISO 10993-7 was published. W. Howard Cyr, PhD, is a research biophysicist in CDRH's Division of Life Sciences, Office of Science and Technology.

Illustration by Brad Hamann

Copyright ©1998 Medical Device & Diagnostic Industry

Cooperation Leads to Rapid Development of Global Cleanroom Standards

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI  February 1998 Column


ISO TC 209 anticipates completing documents relating to cleanrooms and controlled environments before the year 2000.

Investment in cleanrooms has greatly increased during the past decade, largely because of the financial and quality-related benefits provided by controlled environments. The increased usage of such environments is expected to continue as the emphasis on the miniaturization of components and on the control of biocontamination persists.

International companies using controlled environments—as well as those considering such uses—have been aware of the apparent need for global standards in this area. It is this apparent need that has led to the creation of the International Organization for Standardization's (ISO) Technical Committee 209 and its subsequent accomplishments.


In 1992, at the urging of the Institute of Environmental Sciences, the American National Standards Institute petitioned ISO to create a technical committee on cleanrooms and associated controlled environments. This new committee, ISO TC 209, was formally established in May 1993. Its mission is to develop an international standard for cleanrooms and associated controlled environments that encompasses the standardization of equipment, facilities, and operational methods, while also defining procedural and operational limits and testing procedures to minimize contamination.

Manufacturers require varying levels of controlled environments, as seen in these class 100,000 (top), 10,000 (circle) and class 1 (bottom) facilities. Photos courtesy of Richard A. Matthews


Thirty-four countries are currently active in ISO TC 209. Voting members, referred to as "P" nations, are Australia, Belgium, China, Denmark, Finland, France, Germany, Italy, Jamaica, Japan, Korea, The Netherlands, Norway, Portugal, Russia, Sweden, Switzerland, the United Kingdom, and the United States. Nonvoting members, referred to as "O" nations, are Bulgaria, the Czech Republic, Egypt, India, Ireland, Malaysia, New Zealand, the Philippines, Poland, Saudi Arabia, South Africa, Thailand, Turkey, Ukraine, and Yugoslavia.

In accordance with ISO procedures, all work by the technical committee is performed in working groups for which specific parameters have been defined and established. Additionally, committee work is to be voted by consensus, trade barriers are to be eliminated, and criteria for cleanrooms and related environments are to be standardized. The committee will not define cleanrooms by user-specific application nor by microbial limits, and no standard will be instituted if it has a major negative economic impact on a particular nation. Table I lists ISO TC 209's seven working groups and their respective convenors.

Working Group, Convenor
WG-1: Classes of Air Cleanliness, United Kingdom
WG-2: Biocontamination, France
WG-3: Metrology and Testing Methods, Japan
WG-4: Design and Construction, Germany
WG-5: Cleanroom Operations, United States
WG-6: Terms, Definitions, and Units, Switzerland
WG-7: Minienvironments and Isolators, United States

Table I. ISO TC 209's seven working groups, their subject areas, and convening countries.


Since the committee's inception in 1993, cooperation among P nations has been exceptional, resulting in the rapid development of global cleanroom standards. Below are brief summaries outlining the status of each working group as of December 31, 1997.

Working Group 1. ISO 14644-1, "Classification of Air Cleanliness," is in final draft status (FDIS), and it's conceivable that it will become the first formal international standard produced by ISO TC 209. Three new classes have been added to the existing standard, Federal Standard 209, two cleaner and one dirtier (Tables II and III).

Airborne particulate cleanliness classes comparison
ISO F.S. 209
1 ­ ­
2 ­ ­
3 1 M1.5
4 10 M2.5
5 100 M3.5
6 1,000 M4.5
7 10,000 M5.5
8 100,000 M6.5
9 ­ ­
Class limits comparison
at 0.5 µm
ISO ISO 14644-1 F.S. 209E
1 ­ ­
2 4 ­
3 35 35.3
4 352 353
5 3,520 3,530
6 35,200 35,300
7 352,000 353,000
8 3,520,000 3,530,000
9 35,200,000 ­

Tables II and III. TC 209 comparisons showing that three classes have been added in ISO 14644-1.

ISO 14644-2 on specifications for testing and monitoring is currently a draft international standard (DIS) and is available for comment. This document specifies the requirements for monitoring a cleanroom or clean zone to provide evidence of its continued compliance with ISO 14644-1 for the designated classification of airborne particulate cleanliness. A schedule of normative and informative tests is included in the document. Depending on the backlog and production schedules at ISO's headquarters in Geneva, the formalization of ISO 14644-1 and 14644-2 will probably occur late this year.

Working Group 2. ISO 14698-1, "Biocontamination Control General Principles," was elevated to DIS status last October and will be available for comment early this year. This document describes the principles and basic methodology for a formal system to assess and control biocontamination. It will include the general requirements of a sampling plan; target, alert, and action levels; qualification; and reporting.

ISO 14698-2, "Evaluation and Interpretation of Biocontamination Data," was also elevated to DIS status last October and will be available for comment early this year. This document describes the basic principles and methodological requirements for all microbiological data evaluation and the estimation of biocontamination data obtained from sampling for viable particles. It will also include evaluation of the initial monitoring plan and of the data resulting from routine monitoring, as well as analysis of the data, trending, and record keeping.

Working Group 3. This working group on metrology and testing methods expects to have a final draft to the technical committee for review in the first quarter of 1998. Performance tests are specified at operational phases—as-built, at-rest, and operational. The items to be measured are categorized as either primary or user-optional tests. Primary tests include particle count, airflow velocity, airflow volume, pressure differential, and installed filter leakage. User-optional tests include flow visualization, airflow parallelism, airflow turbulence, temperature, humidity, molecule contamination, electrostatic charge, particle fallout, recovery, and integrity.

Working Group 4. ISO TC 14644-4, "Design and Construction," was elevated to DIS status in October 1997 and will be available for comment early this year. This document specifies requirements for the design and construction of cleanroom and clean air devices, as well as requirements for start-up and qualification. It also provides guidance on the basic elements of design and construction.

Working Group 5. The cleanroom operations working group estimates that a final draft will be completed by the fourth quarter of this year. Topics to be addressed include entry/admittance, procedures, and cleaning, as well as maintenance as it relates to equipment, materials, and people.

Working Group 6. In detailing terms, definitions, and units, the definitive document of this working group must include all definitions from all approved documents of ISO TC 209. Consequently, it will be the final ISO TC 209 document.

Working Group 7. The minienvironments and isolators working group expects a final draft to be completed by the fourth quarter of this year. This document will specify the performance requirements for minienvironments, isolation systems, and associated transfer devices, with a focus on how these systems differ from conventional cleanrooms in the areas of monitoring, design, testing, material compatibility, integrity, biocontamination, and so on.


While standards in Europe have traditionally been organized on a national basis, the development of a single market in the European Union has made it essential to have European—as opposed to national—standards. The European Committee for Standardization (CEN) was established for this purpose, and CEN standards now take precedence over existing national standards. In many cases, a national standard may be withdrawn when a CEN standard is developed.

With the formation of ISO TC 209, an agreement was reached with CEN to accept these standards by allowing parallel voting within CEN and ISO. ISO 14644-1 has been accepted by CEN and is in the final stages of ISO approval. Thus, this document will become a requirement for all companies selling into Europe.

From a U.S. perspective, Congress has passed a law stating that it is not necessary to maintain a federal standard, such as Federal Standard 209, if a national standard exists. As noted earlier, the formalization of ISO 14644-1 and 14644-2 will probably occur in late 1998, after which time they will become American National Standards and there will be no reason to maintain Federal Standard 209.


ISO TC 209's schedule for completing global cleanroom standards is ambitious. To date, 50% of the standards are complete, and completion of the full family of documents is expected by the year 2000. The committee and many of its working groups will meet in Phoenix during the annual meeting of the Institute of Environmental Sciences and Technology (IEST) in April. This meeting is expected to be the largest technical event of the decade, combining the IEST meeting with the International Committee of Contamination Control Societies' International Symposium on Contamination Control and the Symposium on Cleanrooms for the Healthcare Industry (which is jointly sponsored by IEST and the Parenteral Drug Association).


Companies are increasingly willing to invest in cleanrooms to take advantage of their controlled environments. The popularity of cleanrooms accentuates the need to develop global standards for their components as well as biocontamination and quality issues.

Anne Marie Dixon is managing partner at Cleanroom Management Associates, Inc. (Carson City, NV). Richard A. Matthews is president of Filtration Technology, Inc. (Greens-boro, NC). For more information on TC 209 or the symposia in April, contact IEST, Mt. Prospect, IL, at 847/255-1561.

Copyright ©1998 Medical Device & Diagnostic Industry