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.

Sitemap


Articles from 1996 In May


Continuation of MDDI's May 1996 article, "Selecting and Using Protective Gloves: An Overview of the Critical Issues"

Back to previous page

Because the medical device and in vitro diagnostics (IVD) industry produces many thousands of different products, the most efficient way to approach the potential for product contamination from gloves is to categorize the potential contaminants. It is important to remember that if powdered gloves are worn during manufacturing or packaging, products can be contaminated by aerosol particles as well as by direct contact.

Chemicals. The level and composition of the chemical residue on a glove are determined by the raw materials (type and quality), chemical formulation, and manufacturing processes of each manufacturer for each glove type.

Several of these chemicals can be detrimental to medical devices and materials. For example, chemical contamination of IVDs and laboratory products can alter results, leading to misinformation, erroneous conclusions, and costly delays. Final assembly and compounding of ingredients for such diagnostic products often occurs under HEPA filtration to prevent contamination. In view of the potential adverse effects caused by powder-borne contaminants, powdered gloves should never be allowed in or near these areas.

High-performance analytical instruments are often vulnerable to contaminants such as calcium carbonate, chloride compounds, and silicone. Interference has been reported in the curing stages of implant devices, dental polyvinyl siloxane impression materials, and orthopedic adhesives.12 It has been reported that residual sulfur on gloves resulting from its excessive use as a vulcanizing agent, or its presence as a by-product of accelerator degradation, can interfere with activating catalysts used in the manufacture of various products.13 Glove powder also has been identified as affecting the set of dental composites.

Proteins. Natural rubber latex protein contaminants can compromise product safety. Allergens from high-protein latex gloves can be transferred to medical devices during manufacturing, packaging, customer assembly, and use.2,14,15 When a protein-contaminated device comes in contact with a previously sensitized patient with Type I hypersensitivity, the effects can be devastating, with anaphylactic shock as the most extreme reaction. Local responses may include hives, heightened inflammation, delayed healing, and implant rejection.16,17 Special care should be taken to ensure that devices used in the care of patients suffering from spina bifida are not handled with high-protein gloves, as this population is uniquely vulnerable--about 40­70% are serum-positive.

Powder. Known for their abrasive properties, powder particles scratch fine lenses and ruin particle-sensitive instruments. These particles are unwelcome contaminants in the manufacturing environments of parenteral and IV drug-delivery devices. The biocompatibility of cardiac catheters, joint replacements, autotransfusion devices, and peritoneal dialyzers can also be compromised by glove powder.18­20

Pyrogens. Manufacturers whose device labels bear the claim nonpyrogenic routinely test representative samples as part of their release criteria. When products fail, manufacturers attempt to isolate and eliminate the source of contamination. During such an investigation, manufacturers should evaluate the gloves that came into contact with the failed product because gloves can carry and transfer endotoxins (lipopolysaccharides that cause pyrogenicity). Sterility is not synonymous with nonpyrogenicity, so each must be addressed separately during glove evaluations.

Manufacturers should be aware that, in addition to the gloves worn during the manufacture of their devices, the gloves used to handle the devices in the end-user setting can also affect device performance. Because improper glove selection can result in customer complaints, manufacturers should consider the benefit of specifying, in product labeling or instructional courses, the type of gloves that should be used.

Chemicals and Powder. When transferred from a glove or contaminated medical device into a wound, cytotoxic and antigenic chemicals can interfere with healing. As the list below suggests, a repertoire of responses exists with this scenario. When powder particles are coated with chemicals, they function as time-release capsules, slowly releasing contents into the wound and extending the chemical reaction into a chronic condition. The potential consequences of chemicals in a wound that may affect the real or perceived performance of a device include the following:

  • Cell lysis.
  • Augmented inflammation.
  • Complement cascade activation.
  • Stricture formation.
  • Vascular necrosis.
  • Impaired perfusion.
  • Chronic inflammatory syndromes.
  • Scar formation.
  • Granulomas.
  • Adhesions.21
  • Band formation.
  • Graft rejection.
  • Transplant rejection.
  • Dehiscence.
  • Delayed healing.
  • Weak replacement tissue.

The mere physical presence of powder in a wound can cause abrasion of soft tissues. Added to the trauma of abrasion is the grinding in of any bioincompatible substances bound to the powder. If a patient experiences complications that are sufficiently severe to require device removal, both tissue and device should be microscopically studied under polarized light--under appropriate conditions, powder particles will reflect a Maltese cross image, confirmatory for starch.

Similarly, manufacturers should consider the effect transferred chemicals from the gloves of end-users may have on the consistent performance of their diagnostic products. For example, powder contamination has been cited as the cause of false negatives from HIV diagnostic tests (through interference with the polymerase chain reaction), inaccurately low levels of cyclosporin in blood concentration assays, and contaminated tissue culture assays.22­24

Pyrogens. Blood-processing products (e.g., dialyzers), most parenterals, and devices that will come in contact with nerve tissue, spinal fluid, or the eye are tested and labeled as nonpyrogenic to minimize the adverse effects of endotoxin contamination. However, endotoxins from the gloves of health-care providers can be transferred to such devices, compromising their nonpyrogenic condition.25 Technicians in research laboratories and the biotechnology industry must also be cognizant of the potential for endotoxin transfer from gloves to instruments, growth media, and culture vessels, because even nanogram amounts can alter responses in biological systems both in vivo and in vitro. As with any toxin, the concentration, dose delivered, and system contacted or perfused will determine the extent of any biological responses. The partial list of adverse biological consequences from endotoxin contamination that follows may be useful to companies investigating product complaints or providing technical consultation. The biological effects of endotoxins include:

  • Fever.
  • Augmented inflammation.
  • Macrophage activation.
  • Release of vasoactive amines.
  • Disseminated intravascular coagulation.26
  • Complement activation.
  • Cell lysis.
  • Vascular necrosis.
  • Impaired perfusion of essential organs resulting in pathology.
  • Shock

Infections. In addition to transporting chemicals, protein allergens, and endotoxins, powder can act as a fomite, infection resistance suppressor, or abrasive. Nosocomial (hospital-acquired) infections can occur when pathogenic and opportunistic microorganisms are transferred from patient to patient. Medical devices are often assumed to be the fomites (objects of infectious agent transfer), but the possibility of contaminated powder should be investigated. Similarly, in clinical and research laboratories, glove powder can pick up pathogenic microorganisms and contaminate other areas in the work environment. This is an issue not only for assay interference and misdiagnosis, but also for employee safety--laboratory-acquired infections are a serious concern.

The mere presence of glove powder in a wound has been found to reduce resistance to infection.27 Synthetic vascular grafts, tendon materials, and braided sutures readily carry glove powder into the wound, where the local inflammatory response is augmented immediately, wasting defensive action on the coated powder particles.28 If an otherwise inconsequential number of organisms from any source later contaminates the area, an infection can take hold that, under normal conditions, would have been overcome by the immune system. And, because powder particles may not completely degrade for a number of years, such effects can become chronic, reducing local resistance to infection well after surgical closure. Therefore, when investigating the causes of postsurgical infection even months after surgery, glove powder on or in the surgical devices should be considered, if not as the primary fomite, then as a suppressor of infection resistance.

Even in cases of minimally invasive surgeries in which fewer complications are anticipated, gloves may adversely affect device performance. Powder is efficiently transferred to ridges on cannulas and into the nooks and crannies of endoscopic instruments and subsequently deposited in the peritoneum, synovial spaces, or other operative sites, potentially leading to a number of the postsurgical complications detailed earlier.29, 30

This article has addressed glove-related concerns regarding employee exposure, product contamination by gloves during manufacturing, and the effect of gloves on product use. Two final considerations are procurement and disposal.

Procurement. While the selection process should focus on the gloves' composition and performance characteristics in the workplace, supplier performance is also a consideration. For example, it is important to ensure that a vendor will continue to ship the product your company evaluated. Many distributors purchase gloves from several outside sources. Because chemical formulations and processing methods vary greatly among manufacturers, such multiple sourcing may result in product inconsistencies, requiring the purchaser to continually perform incoming inspections and costly reevaluations.

The glove supplier's delivery capability should also be reviewed. Have other companies in the industry experienced continual back-order problems with this supplier? The best gloves created are worthless if they are not available for use when they are needed.

Disposal. The environmental impact of used-glove disposal has become an increasing concern for all industries. The factors that must be considered when choosing incineration or disposal in a landfill vary with material type:

  • Combustion of natural rubber latex gloves is quite clean, although some hydrocarbons, minute quantities of unreacted nitrogen-based chemicals, and sulfur dioxide may be produced at low incineration temperatures. In a landfill, residual chemicals will leach out as the rubber biodegrades.
  • Incineration of vinyl gloves reduces the PVC to hydrochloric acid gas, minor residual chemicals, and ash. Under standard landfill conditions, vinyl is not biodegradable. However, esters of phthalic acid, which are used as plasticizers and make up about 50% of the total volume, will leach from the material if it comes in contact with nonaqueous solvents. Before choosing incineration for PVC gloves, users should ensure that local laws do not ban such a practice.
  • Nitrogen-based reaction products are released minimally during incineration of nitrile gloves; the other chemical by-products are similar to those produced by natural latex. In a landfill, residual chemicals such as accelerators will leach out, as they do in rubber gloves; the nitrile itself will resist degradation.

Gloves may affect employee health and productivity. Gloves high in chemicals, proteins, and powder can result in skin irritation and allergies, as well as occupational asthma. Conversely, the selection of gloves low in these elements can save the costs of treatment, lost working time, retraining, and possible litigation. Incorrect selection and care of gloves can result in barrier breakdown, which can also put employees at risk. Proper fit optimizes staff performance by preventing fatigue and providing adequate grip and tactile sensitivity.

Gloves used during manufacturing, assembly, and packaging may affect the quality of finished products. While knowledge of these issues assists in the investigation of the source of contaminants, it should ideally be used during glove selection to avert problems. End-users must also be informed about the potential consequences their glove selection may have on procedural outcome. This is best accomplished by label instructions or in-service courses.


Wava Truscott, PhD, is vice president of scientific affairs at Safeskin Corp. (San Diego).

1.Voeller B, Coulson AH, Bernstein GS, et al., "Mineral Oil Lubricants Cause Rapid Deterioration of Latex Condoms," Contraception, 39(1): 95­102, 1989.

2.Truscott W, and Roley L, "Glove-Associated Reactions: Addressing an Increasing Concern," Dermatology Nursing, 7(5):283­303, 1995.

3.Heese A, von Hintzenstern J, Peters KP, et al., "Allergic and Irritant Reactions to Rubber Gloves in Medical Health Services," J Am Academic Dermatology, 25(5):831­839, 1991.

4.Fay MF, "Hand Dermatitis, the Role of Gloves," Assoc of Operating Room Nurses, 54(3):451­467, 1991.

5.Truscott W, "The Industry Perspective on Latex," in Immunology and Allergy Clinics of North America, vol 15, Fink J(ed), Philadelphia, W.B. Saunders, pp 89­121, 1995.

6.Federal Register, 39(38), February 25, 1974.

7.Gonzales E, "Latex Hypersensitivity: A New and Unexpected Problem," Hospital Proceedings, 27(2):137­151, 1992.

8.Hunt LW, Fransway AF, Reed CE, et al., "An Epidemic of Occupational Allergy to Latex Involving Healthcare Workers," Occupational and Environmental Med, 37(10):1204­ 1209, 1995.

9.Tarlo SM, Sussman G, Contala A, et al., "Control of Airborne Latex by Use of Powder-Free Latex Gloves," J Allergy Clinical Immunology, 93:985­989, 1994.

10.Seaton A, and Cherrie B, "Rubber Glove Asthma," Br Med J, 296:531­532, 1988.

11.Marcos C, Lázaro M, Fraj J, et al., "Occupational Asthma Due to Latex Surgical Gloves," Annals of Allergy, 67:319­323, 1991.

12.Kahn RL, Donovan TE, Winston WL, et al., "Interaction of Gloves and Rubber Dams with a Poly (Vinyl Siloxane) Impression Material: A Screening Test," International J Prosthodontics, 31:140, 1989.

13.Causton BE, Burke FJ, and Wilson NH, "Implications of the Presence of Dithiocarbamate in Latex Gloves," Dental Mat, 9:3, 1993.

14.Beezhold D, Kostyal D, and Wiseman J, "The Transfer of Protein Allergens from Latex Gloves," Assoc of Operating Room Nurses, 59(3):605­613, 1994.

15.Villarroel F, and Ciarkowski AA, "A Survey on Hypersensitivity Reactions in Hemodialysis," Artificial Organs, 9:231­238, 1985.

16.Garred P, Vâge D, Mollnes TE, et al., "Latex Gloves as a Cause of Inflammation and Eczema," Lancet, 335:1469, 1990.

17.Abuck D, Phzybilla B, Enders F, et al., "Latex Allergy and Repeated Graph Rejections," Lancet, 339:1609, 1992.

18.McKee PH, and McKeow EF, "Starch Granulomata of the Endocardium," J Pathology, 12:103­107, 1978.

19.Verkuyl DAA, "Glove Powder Introduced in the Circulation by Autotransfusion and Severe Cardiac Failure," Lancet, 340:550, 1992.

20.Huertas VE, Rosenzweig J, and Weller JM, "Starch Peritonitis Following Peritoneal Dialysis," Nephron, 30:82­84, 1982.

21.Cooke SAR, and Hamilton DG, "The Significance of Starch Powder Contamination in the Aetiology of Peritoneal Adhesions," Br J Surg, 64:410­412, 1977.

22.Lampe AS, Pieterse-Bruins HJ, and Van Wissekerkejcre EJ, "Wearing Gloves as a Cause of False-Negative HIV Test," Lancet, 2:1140, 1988.

23.Hamlin CR, Black AL, and Opalek JT, "Assay Interference Caused by Powder from Pre-Powdered Latex Gloves," Clinical Chemistry, 37(8):1460, 1991.

24.Vâge DI, Garred P, Lea T, et al., "Elutable Factors from Latex-Containing Materials Activate Complement and Inhibit Cell Proliferation: An In Vitro Biocompatibility Study of Medical Devices," Complement Inflammation, 7:63­70, 1990.

25.Kure R, Grendahl H, and Paulssen J, "Pyrogens from Surgeon's Sterile Latex Gloves," Acta Pathologica Microbiologica et Immunologica Scandinavica, Sect B 90:85­88, 1982.

26.Jacobs HS, Craddock PR, Hammerschmidt DE, et al., "Complement-Induced Granulocyte Aggregation, An Unsuspecting Mechanism of Disease," New England Med, 296:769­774, 1977.

27.Jaffray DC, and Nade S, "Does Surgical Glove Powder Decrease the Inoculum of Bacteria Required to Produce an Abscess?" Royal College of Surgeons of Edinburgh, 28:4, 1983.

28.Ruhl CM, Urbancic JH, Foresman PA, et al., "A New Hazard of Cornstarch, an Absorbable Dusting Powder," J Emergency Med, 1:11­14, 1994.

29.Yaffe H, Toby R, Laufer N, et al., "Potentially Deleterious Effect of Cornstarch Powder in Tubal Reconstructive Surgery," Fertility and Sterility, 29(6):699­701, 1978.

30.Singh I, Chow WL, and Chablani LV, "Synovial Reaction to Glove Powder," Clinical Orthopaedics and Related Res, 99:285­292, 1974.

THE DEMISE OF RULE MAKING AND THE RISE OF GUIDANCES

Return to "FDA Regulatory Programs: Cooperation and Common Sense Typify a Record of Success"

Bradley Merrill Thompson, Baker & Daniels (Indianapolis)

With authority from Congress, FDA makes and enforces law. But over the past two decades, the process of regulating the medical device industry has changed profoundly.

Traditionally FDA relied on a process called rule making, which often involves the publication of a proposed rule in the Federal Register, followed by a formal period for receipt of public comments. The process ends after public comments have been integrated and a final rule is published, again in the Federal Register. Although this process can take many months and absorb considerable staff resources, it has the advantage of collecting useful public comment on complex scientific and commercial subjects, while simultaneously educating the public about the agency's regulatory expectations.

But during the past 20 years, FDA has gradually shifted away from rule making toward a surrogate that has neither of the benefits afforded by rule making. As currently practiced this surrogate of rule making--the issuance of guidances--is not nearly so open to public scrutiny. Apart from adhering to a general policy that says FDA will always listen to suggestions offered by the public, the agency rarely seeks out broad public comment on a guidance. The consequent lack of public involvement in guidance development results in a corresponding reduction in the educational effect of the process.

Guidances have grown in popularity because of the agency's need to conserve money and manpower. These guidances may be any of several types, including formal guidelines, compliance policy guides, simple lists of points to consider, or even industrywide letters. And there is no all-encompassing process used to generate them. Because FDA generally does not solicit broad public involvement or take the time to respond to comments that the public might submit, the creation of guidances is generally less costly than rule making in terms of time, money, and staff resources.

Many forms of guidance used by the agency today were virtually nonexistent in the mid- to late-1970s, and only a trickle of such documents were released even in the early 1980s. But by the middle of that decade, the number of guidances began to grow rapidly (see chart) until today CDRH issues dozens of them each year. And these numbers reflect only what CDRH officially calls guidances. Controversially, the agency also has increased its use of speeches, warning letters, and even press releases to make industry aware of FDA regulatory expectations.

Legally, guidances are not binding on the public, whereas certain rules are. But for all practical purposes, there often is little difference between the two. Because of the important role that guidances have grown to play, many people are concerned that they do not measure up to quality standards.

Apparently, FDA has also recognized the need for quality assurance. On March 7, 1996, it announced that it will adopt good guidance practices that ensure public participation at critical points in the initiation, development, and dissemination of guidances.

If successful in applying these practices, FDA may be able to achieve an effective compromise between the decades-old tradition of rule making and the recent surge in issuing guidances--a compromise that balances the need to conserve FDA resources with the advantages of integrating public feedback and education into the regulatory process.

Return to "FDA Regulatory Programs: Cooperation and Common Sense Typify a Record of Success"

METAL INJECTION MOLDING: AN OVERVIEW

Medical Plastics and Biomaterials Magazine | MPB Article Index

Originally published May 1996

Metals

As the medical profession increases its demands for small, intricate, precision metal components to be used in increasingly sophisticated procedures, the process of metal injection molding (MIM) has provided an alternative for the manufacture of these small, complex parts. MIM is sometimes referred to in its broader sense as "particulate injection molding"--a term comprising several processes that offer the freedom of design associated with close- tolerance injection molding of plastics but form components in a wide range of metal and ceramic compositions. This article will provide an introduction to metal injection molding by briefly reviewing the basic process steps, the competitive position of MIM in relation to other metalworking technologies, the relative economics of the process, and material and part characteristics.

History

While MIM is certainly a new technology when compared with traditional metalworking processes, it has been a commercial reality for about 25 years. During that time, the technology has spawned an industry that is now international in scope. The Metal Injection Molding Association (MIMA) was formed in 1987, under the auspices of the Metal Powder Industries Federation (MPIF); current membership includes representatives from six countries. In 1993, a guide to specifying MIM materials, "Materials Standards for Metal Injection Molded Parts," was published and is available from MPIF.

Markets

Any industry requiring small, highly configured metal components represents a potential market for metal injection molding. The medical and dental industries clearly fit this description, and are indeed among the primary market segments currently being served. Components for the medical market are commonly specified in applications such as endoscopic and laparoscopic devices. Performance characteristics of MIM components include wear and chemical resistance, structural strength, biocompatibility, and controlled expansion.

The Mim Process

Metal injection molding uses very fine metal powders, generally in the range of 2 to 20 µm. At the lower limit, powders are so fine as to approach the "dust" category. Alloy compositions are developed through blends of elemental powders, or supplied directly as prealloyed powders. In the case of prealloyed powders--which are generally used for stainless-steel compositions--each powder particle is in reality an ingot of wrought material.

The metal powders are thoroughly mixed with various waxes, thermoplastics, and other ingredients, and the blend is granulated to form a feedstock. The polymeric binder materials may comprise as much as 40% of the mixture by volume. The feedstock is then fed into a conventional injection molding machine, with molding temperatures generally ranging from 300° to 500°F (149°­260°C). Once the feedstock has attained a toothpaste-like consistency, it can be injected into cavities to form precision components. As with plastic injection molding, multiple-cavity tooling can be used to reduce manufacturing costs.

After the parts are removed from the mold, they exhibit excellent strength, with a consistency similar to that of wax crayons. Most, but not all, of the binders and additives are then removed from the molded (green) components by a low-temperature thermal treatment or solvent extraction--or combination of both--in a process known as debinderizing. The exact method chosen depends on the binders used and the cross sections of the part. Binder ingredients are designed to separate sequentially, without disturbing part geometry; just enough binder is left so that the shape integrity of the part is maintained. Following debinderizing, parts are thermally processed (sintered) in either atmosphere or vacuum furnaces, during which the remaining binder is removed and the metal particles bonded into a coherent mass. Sintering temperatures are usually in excess of 2300°F (1260°C). The very fine powders used for MIM provide a strong thermodynamic driving force, and parts undergo shrinkage of as much as 20%. Repeatability is achieved through close control over the feedstock and molding variables. Final densities are generally from 95 to 98% of theoretical projections, with interconnected porosity of less than 1%. If necessary, parts can be processed to be pressure-tight and to satisfy hermeticity requirements.

Materials

MIM offers a significant variety of metal compositions. An abridged listing of metals being processed comprises a number of nickel-iron alloys (ranging from 2% nickel for structural applications to 36­42% nickel for glass/metal sealing and 50% nickel for soft magnetic requirements); alloy steels 4340 and 4650; 3% silicon-iron alloys; and stainless steels, including 304L, 316L, 410, and 17-4PH. Medical applications in current production are generally produced from 316L or 17-4PH stainless.

In addition to these standard materials, there are several other material families that are undergoing developmental work to determine their appropriateness for MIM applications. The first group comprises cobalt alloys, and especially cobalt chrome, an implantable biomaterial that is of special interest to designers of orthopedic devices. A significant amount of experimental work has also been done toward the goal of producing titanium parts with MIM.

Part Characteristics

MIM offers the freedom of design associated with the injection molding of plastic parts, or with investment- or die-casting techniques. As with any manufacturing process, however, the limits of applicability need to be understood. First, MIM processing is best suited for relatively small parts. On a volume basis, parts typically should be smaller than approximately the size of a tennis ball; the process is generally most cost-effective when parts are smaller than a golf ball. Irregularly shaped parts can be produced with a major axis measuring up to about 100 mm (3.9 inches) in length.

The cross section or wall thickness of a part actually has a greater influence on determining feasibility for MIM production than does overall size, as the maximum thickness of a component must be kept relatively low to allow for effective removal of the binders. It is generally recommended that parts have cross sections of less than 6 mm (0.24 in.), although components with cross sections up to 10 mm (0.39 in.) are commonly in production. A practical lower limit is 1 mm (0.040 in.). Still thinner sections are possible, but depend on the flow length or height of the segment. Thin-walled tubular sections are not feasible with MIM.

The weight of a part will obviously be related to its size and section thickness--in other words, to the volume of material present. Components that benefit the most from MIM appear to be those weighing less than 100 g (with a lower limit of less than 1 g). Parts of up to 200 g, or even greater, can be produced, although technically feasible large parts may not be cost-effective. The majority of parts produced are probably within a range of 1 to 50 g.

MIM processing is normally carried out to a tolerance of ±0.3%, but specific dimensions may be held to accuracies of ±0.1%. In order to achieve tighter tolerances, it is desirable to keep cross sections as uniform as possible. The surface finish of sintered MIM parts is typically 32 µin. MIM parts also exhibit those surface features often associated with the injection molding process. Gate vestiges, parting lines, and knockout or ejection-pin marks will sometimes be present and should be taken into account when considering part function. However, many secondary machining operations can be eliminated because of the dimensional accuracy, high density, and surface finish of MIM parts.

Competitive Position

Experience in the marketplace has shown MIM to be competitive with investment casting and discrete machining within the component size and weight range described previously. Whereas the price of an MIM part is often equivalent to that of an investment casting, the process's similarities to injection molding result in parts with better surface finish, closer tolerances, greater freedom of formed holes and--in some instances--sections with thinner walls. Conventional multicomponent assemblies can often be formed as one piece with MIM, reducing material, labor, and inventory costs.

The process compares favorably with conventional powder metallurgy (P/M) when machining of a P/M part is required to generate additional geometry. Within the perspective of a wider range of metalworking techniques, MIM normally will not compete in cost with drop-off screw machining, stamping, or die casting (see Figure 1).

An economic benefit of MIM is that the process often employs multi -cavity tooling, which makes it suitable for meeting medium- to high-volume fabrication requirements (see Figure 2). Most MIM projects are probably within the 20,000 to 200,000 pieces/year range, though there are numerous applications demanding annual production runs in excess of one million parts. Lower-run quantities, which are typically specified during the development phase of a medical device, are also common.

Conclusion

Metal injection molding offers an effective alternative for manufacturing the kind of small, intricate, precision metal components typically required by the medical device industry. The process is well suited to the high-volume production of geometrically complex parts, and offers a freedom of design equivalent to that of injection molded plastic parts or investment castings. Parts can be produced from a wide selection of alloy compositions, including medical-grade stainless steels. Close-tolerance production often permits the formation of net-shaped parts directly by the MIM process.

With any manufacturing technology, the greatest benefit is derived when parts are designed from their inception to be produced by that particular process. Such is certainly the case with metal injection molding. Medical product engineers should "design for the process" whenever possible, since converting from existing designs may not permit full exploitation of the advantages of MIM.

Acknowledgments

This paper was initially conceived as part of a forum presented by the Metal Injection Molding Association of the Metal Powder Industries Federation (Princeton, NJ).

J. Robert Merhar is director of market development at Parmatech Corp. (West Chester, PA). With more than 30 years' experience in the field of powder metallurgy, he has been involved with metal injection molding since 1981, and is a past president of the Metal Injection Molding Association. He currently serves on the industry development board of the Metal Powder Industries Federation.

THE MAKING OF A REGULATORY PROFESSION

Medical Device & Diagnostic Industry Magazine | MDDI Article Index

Originally published May 1996

Richard E. Greco

President, Regulatory Associates, Inc. (Clarence, NY)

In 1972, the executive management of my company assigned me the task of forming a corporate regulatory affairs department. The goals I adopted as part of this task were as lofty as they could be, and ranged from ensuring that all company departments consistently complied with regulatory requirements to promoting the company as one that satisfied both the letter and the spirit of those requirements. I imagined that this new depart- ment would act as the regulatory adviser for all corporate operations while also serving as the liaison to regulatory agencies, which--because our company made drugs and special infant dietary supplements as well as medical devices--included not only FDA but the Department of Agriculture, the Drug Enforcement Agency, and the Bureau of Alcohol, Tobacco, and Firearms.

It soon became apparent, however, that creating such an expert, credible, and effective department would require benchmarks and resources that had not yet been created. There was no forum in which to network with other professionals in the field, nor were there any schools to teach the subject. Equally important, there was almost no interaction between industry and the regulatory agencies, such as FDA, that might have provided guidance.

Throughout most of industry in the years leading up to the Medical Device Amendments of 1976, regulatory affairs were a corporate afterthought. Many companies lacked personnel who were formally designated to handle regulatory affairs, and employees were shuttled in and out according to who had the time available. As a result, despite the voluminous amount of work and critical responsibilities assigned to regulatory affairs personnel, they were not recognized as professionals. Regulatory affairs personnel often "got no respect" from company management, and frequently there was no truly rewarding career path available to them.

These were the circumstances that led to the establishment of the Regulatory Affairs Professionals Society (RAPS). Despite the apparent need, however, the group did not have a very auspicious start. Only 60 device manufacturers indicated an interest in helping to create the society, and only 6 company representatives attended the first meeting in December 1974. Nevertheless the organizers were determined to go forward, and the society was finally incorporated in 1976.

Those who started RAPS understood the enormous responsibility that was being placed on the shoulders of their fledgling profession. At the time, however, many in top management underestimated the time, effort, and expertise required for their companies to meet regulatory requirements. For many of these companies, the need for an organization to support regulatory affairs professionals was only brought into focus with the enactment of the Medical Device Amendments of 1976.

The goals that RAPS adopted at its founding remain basic to its mission: to educate members and promote them as regulatory professionals, to provide a forum for networking, to enhance members' career opportunities, and to improve members' status within their companies. To meet these goals, the society established a certification program that has helped to create a profession where none existed before.

Over its 20-year history, RAPS and its members have participated in bringing about many changes in the device industry. Interaction with FDA has improved, in part because staff from the agency joined the society and began to give presentations to its membership. Many regulatory affairs professionals have become part of their companies' executive management. And women, who were previously absent from the field, have found enormous opportunities in it. To meet the needs of a rapidly globalizing industry, the society has also branched out to hold meetings overseas.

These milestones mark the growth of RAPS from a regional association to an international professional society of more than 6000 members who are today recognized as professionals pursuing well-paying careers. The work of those professionals has helped to give shape to the device industry over the past 20 years, and will continue to be an important factor in the future.

Richard E. Greco was the founder of the Regulatory Affairs Professionals Society (Rockville, MD).

Laser Micromachining of Medical Devices

Medical Plastics and Biomaterials Magazine | MPB Article Index

Originally published May 1996

PROCESSING

RONALD SCHAEFFER

Lasers provide unique processing capabilities for the manufacture of many products, including disposable medical devices. In some cases, the use of laser-micromachining technology allows more cost-effective solutions compared with other available methods, even though competing technologies may offer similar technical capabilities. In many other instances, laser-based manufacturing may provide the only technical solution to unique fabrication problems. It is the job of laser integrators to review all customer

requirements--including tangible elements such as processing speed, throughput, feature size, and budget, as well as intangibles such as "feature quality" and "user-friendliness"--and propose the most sensible, effective approach. Rigid attention to detail is necessary both in the actual production and in the quality control necessary to ensure product conformity to specifications. This article reviews the fundamental physics of carbon dioxide (CO2), excimer, and solid-state lasers as related to material interaction and discusses considerations relevant to optimizing a laser-based system for a particular manufacturing environment. Finally, factors relevant to quality control in manufacturing are discussed.

COMMON INDUSTRIAL LASERS

Excimer, CO2, and solid-state (primarily Nd:YAG and Nd:YLF) lasers represent the three most common types of lasers used in industrial micromachining applications. Each has its own unique characteristics, and together they provide very complementary capabilities (see Table I). (Tables and figures are not yet available on-line.)

Carbon dioxide lasers are the most familiar and widely used photon sources. Employed in industry since the 1960s, they are still quite popular because of their wide acceptability, ease of use, and simplicity. While some varieties of these lasers are available at multi-kilowatt power levels, most medical-related micromachining requires less than 150 W of average power. The wavelength of emitted photons with CO2 lasers is about 10 µm, which means that they are infrared photon sources (see Figure 1).

The fundamental physics of material interaction with infrared photons is primarily between ro-vibrational energy levels of the molecules. In essence, absorption of 10-µm photons by most materials causes an increase in the vibrating frequencies of molecules, which leads to a temperature increase. Therefore, the primary interaction is thermal. In most materials, these thermal effects are quite pronounced on a microscopic scale (see Figure 2). If these effects can be tolerated, the CO2 laser provides a high-speed method of material processing.

In addition, since the ultimate achievable feature resolution is related to the wavelength of the photons, the CO2 laser, while theoretically capable of producing 10-µm feature resolution, is limited to 50-µm resolution in the majority of cases. Finally, in most (but not all) applications, the CO2 laser is set up in a focal-point machining configuration (see Figure 3), which means that it is normally possible to machine only one feature at a time.

Solid-state lasers, first used industrially in the 1970s, are best represented by the more recognizable Nd:YAG (neodymium:yttrium, aluminum, and garnet) sources, although some different processing capabilities are currently being investigated using Nd:YLF (neodymium: yttrium, lithium, and fluoride) lasers.1 Both of these lasers emit photons with about a 1-µm wavelength--which also places them in the infrared region of the spectrum--so the considerations discussed previously regarding thermal material interaction would again apply. Solid-state lasers can also be frequency converted using nonlinear crystals to produce photons at double, triple, and quadruple the fundamental frequency, which gives flexibility for material processing at visible and UV wavelengths. Most of the

current processing work is being done at fundamental frequencies, but it is probable--given the development of diode pumping of the rod at high efficiencies and high repetition rates--that more work will soon be done at frequency-converted wavelengths, especially in the UV region, and that solid-state UV lasers will take the place of traditional UV sources for many applications. Ultimate feature resolution depends on the wavelength used, but is approximately 2 µm for the fundamental frequencies, while the ultimate practical resolution is from 10 to 25 µm. Like CO2 lasers, solid-state lasers are generally used in a focal-point configuration.

Excimer lasers are comparatively newer photon sources, having been first used in industrial settings in the 1980s. The utility of these laser sources is that they directly emit photons at high average powers in the UV region without frequency conversion. Also, because of the incoherent and highly divergent nature of the beam, excimer lasers are used in an imaging configuration, allowing multiple, simultaneous feature machining and easy splitting of the main beam for multiple part processing. Discreet spectral lines are emitted, depending on the gas mixture employed (see Table II). Longer-wavelength, 308-nm excimer lasers are primarily used for marking purposes. This wavelength produces a photochemical color change in materials like plastics and ceramics, and also produces visible and indelible marks on some metals. The 193-nm lasers are used primarily for specialty applications in which materials do not absorb well at other wavelengths. These lasers are not routinely used because of the difficulties associated with working at UV wavelengths below 200 nm, which include atmospheric absorption of the photons, short laser gas lifetime, low laser output powers, high maintenance costs, and color-center formation in optics.

More than 90% of all applications involving excimer lasers use 248-nm photons because of the aggressive material interaction and highly reliable laser operation at this wavelength. Practical feature resolution of 1 µm is achievable, with most applications having feature resolutions greater than 10 µm. Below 10 µm, extreme attention to detail in equipment design is necessary to avoid mechanical, thermal, vibrational, and optical aberrations. Because of the halogen gases used in excimer lasers and the high cost and relative difficulty of operation, these laser sources are usually reserved for applications for which neither CO2 nor solid-state sources are practical. Nevertheless, excimer lasers fulfill an important niche processing area and frequently are the best--or the only--technology capable of performing certain operations.

Several other comments apply to laser micromachining in general. First, the production of straight-walled features is usually not possible, since some degree of inherent taper is normally present and increases as the part aspect ratio (feature size to depth) increases. Second, debris from the lasing process may be a problem in some instances. In order to maximize cleanliness--and sometimes to enhance processing speed or to give higher quality results--assist gases are frequently used. It is important to optimize the gas type, pressure, nozzle shape, and direction of gas flow in order to achieve the best possible processing conditions. Finally, facility requirements or safety requirements may, in some cases, override technical considerations in the choice of laser sources. An example is the use of a 308-nm (XeCl) excimer laser chosen instead of a 248-nm (KrF) excimer laser primarily because chlorine sources are somewhat less toxic then fluorine sources, even though the shorter-wavelength light gives better processing results.

SYSTEM CONSIDERATIONS

Once the appropriate laser has been chosen, it must be integrated into a complete processing system. The first step is to identify the environment in which the system will be located. It is important that the system be built in accordance with available facilities, especially as concerns electrical and cleanroom requirements: most medical products require at least Class 100,000 clean conditions or postprocess cleaning. It is also necessary to decide whether the system will be used as a stand-alone or on-line operation. The integration of laser-based systems into automated serial production lines, most often using conveyor-belt carriers, requires a detailed knowledge and understanding of how the line as a whole functions. The production line should offer centralized control of all line stations and be designed so that the product will flow smoothly through each operation.

A laser-beam-delivery system performs three primary tasks. First, it propagates photons onto the workpiece. Second, optical elements within the system are used to shape and condition the beam in order to maximize the efficiency of photon use. This can include beam splitting, beam homogenization, or beam motion (such as a galvanometer-driven head). Finally, the system must protect operators from any dangerous light or surface reflections.

Many laser processing systems use a fixed beam--moving the parts rather than the beam. In addition to conveyor-belt motion, x-y-z-* stages can be used to position the parts properly during lasing. Systems can have many axes of motion--all controlled from a centralized host computer--including z-axis adjustment of the focus head and automated demagnification. In some instances, robotics are used for efficient part handling. For actual part holding, dedicated tooling integrated into the motion control is designed for quick and accurate part positioning. Vacuum chucks with solenoid control are commonly employed when the part being lased is a thin film, especially for roll-to-roll applications.

Even though some production systems do not require the use of cameras for visual inspection, some form of part viewing is usually necessary or recommended. Systems can be designed for viewing the part either while it is being lased or once it is off the lasing axis. By using both optical and electronic magnification (perhaps incorporating zoom) and video crosshairs for targeting, images of 500* or higher are easily achievable. For highly automated and extremely accurate alignment, more sophisticated vision systems can be selected that grab an image, digitize it, compare stored data to the target site, and align to predetermined fiducials.

Computer control of the motion system is accomplished using a PC486 or equivalent with appropriate drivers, boards, and software. Integrated CAD/CAM capability is common. State-of-the-art manufacturing systems include keyboard, mouse, and touch-screen control. The latter feature is especially useful, allowing full programmability for senior engineers using the keyboard and mouse but limiting others' access to the software through the use of a touch screen programmed only with operator-level commands.

Finally, the entire system--including areas for beam propagation, part location, and laser output--must be fully interlocked and enclosed to ensure operator safety. Excimer lasers require further attention because of their use of high-pressure, halogen-containing gas mixtures. Additional accessories such as gas processors, gas cabinets, water chillers, and air-purification units may also be desirable.

PROCESSING CONSIDERATIONS

While disposable medical devices can be manufactured out of ceramics, glass, and metals, by far the most common material is some form of plastic. Catheters and injection-molded plastic parts make up the majority of disposable devices that require laser micromachining. The first step in processing any component is a detailed analysis of the material to determine its interactions with various laser sources, keeping in mind the requirements for feature size, processing area, and desired finish quality and expected production volume. Table III gives information on etch rates for different materials. There are many cases in which the technical objectives can be met by laser processing, but financial considerations do not allow the use of lasers because of limited throughput or unacceptably high amortized piece cost.

A very important decision for the customer is whether to incorporate laser systems into existing manufacturing environments or to rely on the use of laser contract manufacturing services or "job shops." This choice is usually based solely on the portability of the product and the relative costs involved. For instance, very-high-volume products could justify capital-equipment purchases (assuming that operating costs are also acceptable), whereas lower-volume jobs, perhaps requiring only a fraction of the available time, would not. Unless there are good reasons for capital purchases--such as a proprietary process or simply to retain full control of manufacturing--consideration should be given to contract manufacturing. There are a large number of CO2 and Nd:YAG contract manufacturers, but only a few shops capable of using excimer lasers in any reasonable volume. For medical projects, a critical consideration is the ability of the laser shop to provide levels of product cleanliness and traceability that are acceptable to both the customer and FDA. The best partnerships are long-term relationships based on personal contact, audits, and historical evidence such as quality and on-time delivery, which assure the customer that "ship-to-stock" processing is possible. The importance of selecting a reputable contract manufacturer with a long history of involvement in medical device manufacturing cannot be emphasized enough.

DOCUMENT CONTROL AND TRACKING

Another important aspect of medical product manufacturing is documentation control, which starts with accepted engineering drawings. Most laser processing is done on raw material or molded parts, and is delivered in bulk. In general, there is a lot number associated with incoming material. If possible, it is a good idea to preserve this number, perhaps with additional qualifying characters, in order to retain traceability to the material supplier.

A typical lot-tracking document of a laser-services provider is shown in Figure 4. This document is attached to the incoming product by the receiving department and accompanies material through every stage of processing. In this case, the product must be manufactured in a Class 100,000 cleanroom environment, and associated information and instructions also appear. Additionally, this particular job requires using a split-beam approach, so lot-tracking numbers differentiate the beamlets. A six-digit designator is attached to the original lot number and identifies the month (first two digits), day (third and fourth digits), and beamlet number (fifth and sixth digits). All personnel handling the parts during any phase of the operation log onto this tracking document, and tallies are kept on both incoming and outgoing products. Note also that the outgoing part number carries a numerical suffix ("-1") indicating that the part has in fact been laser processed, since this is not always obvious to the naked eye because of the small size of the holes in typical molded parts. Copies of the tracking document ship with the product and are also kept in the job files.

Additional required documentation includes cleanroom operational and maintenance procedures, cleanroom certification (usually done once per year), job-related setup and operational procedures, and certificates of compliance. Cleanroom operational and maintenance procedures must be written and kept on file, and should be required reading for the restricted number of employees authorized to enter the area: signed statements from each involved employee verify that they have read and understood the procedures. Each long-run job should have an associated manual comprising a complete set of operational procedures, setup instructions, explanations of any required tooling, and specifications (including tolerances). When incoming QC at the level following laser processing is minimal or nonexistent, certificates of compliance are sometimes used to ensure "lot-to-stock" production.

QUALITY CONTROL AND ASSURANCE

Whenever possible, laser systems should be set up with fail-safe mechanisms, automated monitoring and recording of important process parameters, and periodic quality checks. Just as there are many different types of laser processing, so will the quality system depend on the material, the type of job being undertaken, and the particular circumstances of the project. For example, with flow devices, the actual flow of gas or liquid through the orifice under constant conditions of viscosity and pressure is the desired result. In situ, one might measure drilled hole size and correlate that size to a flow empirically. But in the case of using gas flow as a calibrant for liquid flow (liquid flow tests are usually destructive), care must be taken to avoid confusion induced by different geographic and atmospheric conditions.

If parts are being shipped to various locations for additional processing, it is best to have accepted, duplicate standards at each location. Actual measurements of physical feature sizes and the positions of the features (relative to part fiducials) can be performed using microscopes, optical comparators, and a whole host of other analytical equipment. It is sometimes more difficult to judge feature "quality" or cleanliness: the laser process will naturally leave debris, and processing conditions must be optimized to give the best quality while retaining manufacturability. However, in some cases--whether for technical or fiscal reasons--a secondary cleaning process may be necessary.

Process monitoring must be done on all conditions vital to the success of the run. For example, most laser systems monitor and record output energy as an important parameter affecting feature size and quality as well as processing speed. In some instances, however, the process depends less on total output energy than on some other factor perhaps considered secondary in nature, such as beam homogeneity.

All production jobs should have accepted quality programs documented in the job manual, offering explicit instructions on how to check the work both in progress and after processing. The final authority to ship product or move it to the next processing step should rest with someone other than the operators, and the entire job should be reviewed by the person with shipping authority to ensure final product conformation.

CONCLUSION

The growing use of laser micromachining in disposable medical device fabrication--whether conducted in-house or contracted through service bureaus--is providing opportunities for unique and cost-effective manufacturing solutions. Continued developments in laser systems have made the technology a viable commercial processing tool, with applications in R&D, prototyping, short-run or low-volume production, and high-volume, high-throughput manufacturing. Through familiarization with the technology and interaction with skilled and reputable subcontractors, medical device manufacturers can cross the comfort threshold and use laser-based processing as a valuable enhancement to traditional manufacturing techniques.

REFERENCE

1. Scheffer R, and Angell J, "Novel High-Power Nd:YLF Laser for CVD-Diamond Micromachining," presented to the SPIE Micromachining and Microfabrication '95 Conference, Austin, TX, October 1995.

BIBLIOGRAPHY

Pippert K, and Zaal G, "Excimer Lasers Carve Out Industrial Market Niches," Indust Las Rev, April, pp 13­16, 1995.

Kincade K, "Industrial-Laser Job Shops: The Road to Success," Indust Las Rev, April, pp 13­18, 1993.

Scaggs M, Sowada U, and Andrellos J, "Excimer Laser Micromachining/Marking," in SAE Technical Paper Series, Warrendale, PA, Society of Automotive Engineers, pp 7­13, 1989.

Ogura G, Andrew R, and Schaeffer R, "Practical Consequences of Matching Real Laser Sources to Target Illumination Requirements." To be published in Vol 2703 of the SPIE Photonics West '96 Conference Proceedings, Bellingham, WA, Society for Photooptical Instrumentation Engineers, 1996.

Ronald Schaeffer, PhD, has been with Resonetics Inc. (Nashua, NH)--a provider of laser micromachining services and systems--for the past four years, holding positions in both corporate and technical management. He currently serves as the company's director of corporate development. Schaeffer holds a doctorate in physical chemistry from Lehigh University.

Partnership Efforts Take Roots-Up Approach

Medical Device & Diagnostic Industry Magazine | MDDI Article Index

Originally published May 1996

FDA Reform

Judging from a number of developments at both the national and regional levels, FDA appears to be undergoing a veritable transformation--a shift in culture that is bringing many in the agency to view device manufacturers as partners rather than as adversaries. While the watchdog agency has sporadically paid lip service and little else to such a change, there are now clear signs that the idea has taken root.

FDA headquarters will soon put into effect a pilot program that will test new procedures for GMP inspections. If implemented permanently, the new procedures could smooth the process and ease the burden on medical device companies undergoing inspections.

The pilot program is designed to test the feasibility of three major procedural improvements. First, rather than every inspection being a surprise, FDA will give qualified companies advance notice of upcoming site visits, enabling them to make preparations for a smooth and efficient inspection. Second, inspectors will tell the company about violations as they are uncovered and, if those violations are corrected prior to completion of the inspection, the corrective action will be noted on the FDA-483 list of observations. Third, FDA will provide a closure document after an inspection is completed, formally noting when issues listed on an FDA-483 are resolved.

The program was announced on March 15 in a letter jointly issued by FDA associate commissioner for regulatory affairs Ron Chesemore and device center director Bruce Burlington. Although no date was given for formal promulgation of the program in the Federal Register, implementation is expected to move rapidly. "FDA has already been moving ahead with the program internally, doing some training--basically greasing the skids," says Wendell Gardner, senior vice president at Cobe Laboratories, Inc. (Lakewood, CO), and a key participant in the development of the initiative. "There is no more review of this program left to do."

According to Gardner, FDA will not wait for industry comments about the program before beginning implementation, but will instead consider comments arising from the Federal Register publication at the same time as it evaluates the results of the pilot program, which is scheduled to last until the end of this year. "This is a really good program and the agency has taken a very reasonable approach," Gardner says. "We have all the FDA districts onboard and ready to work with it."

This nationwide program is only the most visible effort at improved industry-FDA relations to grow out of last year's series of regional "grassroots initiative" meetings, which brought together industry and FDA representatives to discuss their differences. In the wake of those meetings, FDA field offices appear to be taking a genuine interest in developing ways to improve communications with medical device companies. This interest has taken a number of different forms.

The Southwest Region has scheduled a meeting of frontline regulators and representatives of the medical device industry to examine how the two sides can work together more effectively during the FDA inspection process. The idea is "to gear the process more toward problem solving and not confrontation," says Marie Falcone, small business representative for the FDA Southwest Region (Dallas).

More such joint training sessions are expected in the future in each of the three districts in the Southwest Region. In some, company project engineers and FDA staff, including inspectors, will tell each other about their jobs and the pressures they are under. "We don't want to get into touchy-feely things, but we do want to have a mutual understanding of where we are coming from," Falcone says. "The end purpose is to build trust." Other training sessions will be more technical, with industry experts bringing FDA investigators up to speed on new manufacturing technologies, particularly in computers and metals testing.

This shift toward FDA cooperation may be a combination of several forces acting independently yet having a cumulative effect. Ed Esparza, Southwest Region director, credits agency's growing recognition that the American medical device industry is competing on a global stage. "We recognize that FDA has an impact on the worldwide economy," Esparza says. "We must ensure that we are not an impediment to our industry's getting new products to market and being competitive in the world trade arena."

The rising emphasis on cooperation may even have come from individuals such as Ron Johnson, former director of the device center's Office of Compliance, who recognized problems in the field that were beyond his reach before becoming FDA's Pacific Region director (San Francisco). "When I was at the Center for Devices and Radiological Health, I had a pretty good sense for industry concerns about field activities, but I didn't have any control over them," Johnson says. "When I came out here, I was able to address some of them."

The Pacific Region has been a pioneer in the current effort to improve communications. Even before the nationwide inspection initiative was proposed, the Pacific Region had instituted a policy of advising companies about violations as they were uncovered during an inspection. The region's policy also called for field offices to respond to written requests about the status of an FDA-483. "We acknowledge receipt of letters asking about FDA-483s, and we say whether it looks like the company responded reasonably," Johnson says. Other aspects of the national initiative--such as annotating FDA-483s during an inspection to verify that violations have been fixed--"were more than we were able to do locally," he explains. But opening the communications process was a step in the right direction, he says.

Since then, the Pacific Region has taken several more steps toward improving relations with industry. FDA staff have contacted the companies attending the grassroots meetings in the region "and we have continued the dialogue," Johnson says. The Pacific Region has also publicized the names of district directors along with their direct phone numbers. "Just knowing who the players are and sharing information on a regular basis helps," he says. These directors have also assured Johnson that they are committed to meeting with industry representatives without putting any restrictions or conditions on the topics to be discussed. "Any company that has a need to talk to top management of the districts or Pacific Region can now do so," Johnson says.

Another step along the path to better relations with industry was the identification of a person at the Pacific Regional Office who will handle requests from companies seeking their current regulatory status. "A company that is getting ready to submit an application to CDRH and wants to make sure there is nothing in GMP compliance that will hold it up, can find out very quickly," Johnson says. In the near future, the Los Angeles district is expected to begin experimenting with an ombudsman project in which a staff person is assigned to help companies that are experiencing regulatory problems. "That person will shepherd them through the system to resolve the issue, whatever it might be, whether they have a PMA hung up or something else has gone wrong," he says.

While there may be many individual forces behind the agency's sudden interest in corporate concerns, the undercurrent is probably the FDA reform effort launched by the Republican leadership of Congress last year. "There is no doubt in my mind that the pressure from Capitol Hill was one of the major motivating factors toward the reform now being seen at FDA," says George Burditt, a principal in the Chicago law firm of Burditt and Radzius, which represents medical device companies. "But I also believe there are a lot of very strong feelings within the agency that they want to reform--that they want to do things better."

Gardner agrees with that assessment. "I think at least some in FDA believe that they need to change their culture," he says.

Pacific Region director Johnson appears to be one of them. "We're trying to get both industry and government to deal with broader issues in collaborative ways," he says, "so we can solve problems before they become nightmares."--Greg Freiherr *

THE EVOLUTION OF BIOMATERIALS LIABILITY

Medical Device & Diagnostic Industry Magazine | MDDI Article Index

Originally published May 1996

An Interview with James S. Benson

Senior Vice President for Technology and Regulatory Affairs, Health Industry Manufacturers Association (Washington, DC)

On March 11, James S. Benson visited the White House to encourage President Clinton to sign legislation that would protect the makers of biomaterials from liability. In those discussions, Benson was a representative of the medical device industry, but he brought to bear experience gained in a number of top-level positions at FDA and at the Health Industry Manufacturers Association. Benson has sought for years to put liability issues surrounding biomaterials into perspective. In this interview with MD&DI, Benson discusses the evolution of those issues over the last 20 years.

What is the central point at issue for biomaterials liability, and how did it develop?

One of the most important problems in the area of biomaterials is the issue of safety and what methods should be used for determining degrees of safety. No device is entirely risk-free for all patients at all times, so the relationship between risk and safety is important for both FDA and industry. And the agency's approach to these questions has changed dramatically over the years.

An example of that is the use of silicone in breast implants. Right after the 1976 amendments were passed, the classification of breast implants was on the agenda at FDA. But the issues back then were not toxicological; they involved the ethics of using breast implants and the mechanical aspects of these implants--their rupture rate and how long they would last.

Isn't silicone still considered relatively safe by the scientific community?

Not only by the scientific community but by many different industries. Silicone continues to be used ubiquitously in foods and many other products. But in the medical industry, class action suits have allowed junk science to determine the risk of using this material.

What do you mean by junk science?

When a doctor testifies that three out of four breast implant patients he has examined show signs of autoimmune disease, that implies that three out of four women with breast implants will develop this disease. But no matter how many such stories are told, the plural of anecdote is not data. Unfortunately, people sometimes get sick and encounter health problems. And it is human nature in those situations to look for something to blame other than aging or some naturally occurring disease process. So, a person who has something unique in his or her medical history may tend to blame that something else.

When that happens, there is an opportunity for class action suits. And then a lot of money is spent trying to produce evidence that can be convincing to a lay jury. That is the kind of situation we find ourselves in with biomaterials.

How has the fallout from breast implant litigation affected the biomaterials community in general?

The litigation itself, and the junk science that has played a part in proving the lawyers' cases, have raised questions about the use of all biomaterials. Just as significantly, these cases have led FDA to become more zealous in its regulatory approach to the point of seeking absolute risk determination.

There is now a movement to find out exactly what every biomaterial does. That is like counting stars. Can you ever be sure you have counted them all? When that approach is applied to biomaterials, the question becomes one of degree--how much testing is enough?

What is the greatest concern that has come from this evolving interest in the toxicity of biomaterials?

FDA's zeal for absolute risk determination, combined with junk science, has put us in a situation that is very dangerous. There has been a kind of indictment and conviction of silicone that has rubbed off on all biomaterials.

As a result, about 15 companies that supply raw materials of one type or another for the manufacture of medical devices have said that they will no longer do so because they can't risk the litigation. In short, over the last 20 years we have evolved from not worrying enough about toxicity to worrying too much.

James S. Benson was formerly deputy director (1982­1988) and director (1991­1992) of FDA's Center for Devices and Radiological Health, deputy FDA commissioner (1988­1990 and 1991), and acting FDA commissioner (1990).

The Effect of Molecular Orientation on the Radiation Stability of Polypropylene

Medical Plastics and Biomaterials Magazine | MPB Article Index

Originally published May 1996

GEOFFREY HEBERT

Polypropylene (PP) is widely used in the medical industry to produce syringes, vials, and numerous other devices, often through injection molding. Many manufacturers employ gamma or electron-beam (E-beam) processing to sterilize these varied products, even though PP is known to undergo excessive oxidative degradation upon irradiation in air. Without intervention, the severe embrittlement that can occur will essentially destroy the polymer's physical properties, and thus the usefulness of the molded part.1,2 However, sta-bilization packages have been developed to combat this phenomenon.2,3

One effect of the injection molding process is that it imparts molecular orientation to the plastic in the direction of polymer flow.4 The present investigation is designed to determine the effect of this orientation on the radiation resistance of PP, with the goal of aiding in the design, manufacture, and testing of injection-molded, radiation-sterilized medical devices.

EXPERIMENTAL

Materials. Table I outlines the six PP samples used in this study, of which three are homopolymers and three are random copolymers (with ethylene). The samples were not stabilized, apart from the addition of approximately 200 ppm of in-process antioxidant. Samples PP2, PP3, PP5, and PP6 were chemically visbroken through the addition of peroxide and extruded, under nitrogen, at 230°C on a Davis-Standard 2-in. extruder. Samples PP1 and PP4 were not extruded prior to molding. The melt-flow rate of each sample was measured according to ASTM D 1238, with 1% BHT added to stabilize the polymer.(Tables and figures not yet available on-line.)

Sample Preparation. Plaques measuring 76 mm on a side and 1.0 mm thick were injection molded on a Van Dorn 75-tn machine under conditions outlined in Table II. Samples measuring 38 mm x 13 mm were then cut from these plaques (as shown in Figure 1), and the crystallinity of the samples determined via density-gradient column measurements on small pieces cut from the centers of the specimens.

Irradiation and Testing. The polypropylene samples were mounted on racks, which left the entire surface of each specimen exposed to air. Samples to be measured across orientation were dosed to 53 kGy of E-beam, whereas those tested with orientation were dosed somewhat higher, at 68 kGy of E-beam.5 The samples were stored at 23°C in air. Infrared dichroism was employed to measure the crystalline (fc) and average (favg) orientation of the samples prior to irradiation, using a Nicolet 60-SX FTIR and a KRS-5 supported-wire grid polarizer.6

Figure 2 depicts the flex-test apparatus, which was operated at a cross-head speed of 254 mm/min.7 Flexural strength was determined using Equation (3) in ASTM D 790, tangent flexural modulus derived from Equation (5) in the same standard, and break-angle results calculated using the formula shown in Figure 3. An Instron 1114 tensile tester equipped with a 50-kg load cell was used for flexural testing, conducted at 2, 5, 24, 50, 100, 250, 500, and 1000 hours after irradiation. Five specimens were tested and averaged for each data point.

RESULTS AND DISCUSSION

Molecular Orientation. Crystalline, amorphous, and average molecular orientation for the six samples can be found in Table I. Amorphous orientation, which is not directly measured via IR dichroism, can be calculated by the following equation:

favg = Vcfc + (1­Vc)fam,

where favg = average orientation, Vc= volume fraction crystallinity, fc = crystalline orientation, and fam = amorphous orientation.6

Orientation is given by the Hermans orientation function fp:

fp = 1/2 (3 cos2e ­1),

where p can be crystalline, amorphous, or average.6 If all molecules are oriented in the flow direction, fp = 1. If all molecules are oriented in the cross-flow direction, fp = -1/2. In a nonoriented sample, fp = 0.

The results in Table I indicate that all of the samples are oriented in the flow direction in varying degrees. The degree of orientation measured in these samples is in the same range as results obtained from measurements on PP syringes with a wall thickness of 1 mm.

Flex-Test Results. Table III presents the flexural properties of the six samples prior to irradiation. Clearly in evidence is the anisotropy that developed in the samples with higher molecular orientation. Figure 4 compares the flexural strength measured with orientation divided by the flexural strength measured across orientation (anisotropy) versus crystalline orientation.

Break-angle results with and across orientation after irradiation can be found in Figures 5 and 6, respectively. Note that samples that did not break before a 90° bend are recorded as a 90° break. Figure 5 shows that when the specimens were measured in the direction of flow, they were all quite ductile; in fact, the only failure was found in PP1 at 1000 hours. However, Figure 6 clearly demonstrates the loss of ductility in virtually all the samples when tested in the cross-flow direction. Only sample PP6 survived the entire 1000-hour test period.

Flexural-strength results outlined in Figure 7 (with flow) and Figure 8 (across flow) reveal a similar trend. Specimens tested in the flow direction retained their strength over the 1000-hour test, whereas most specimens tested in the cross-flow direction showed substantial losses in flex strength.

The above data demonstrate the extreme effect of molecular orientation on the retention of the ductility and strength of injection-molded PP after exposure to ionizing radiation. These results would not necessarily have been predicted by the values listed in Table III. Indeed, only sample PP1 exhibited extreme anisotropic behavior prior to irradiation. The rapid loss of ductility after radiation sterilization in the cross-flow direction underscores the need for product designers and processors to be aware of possible areas of molecular orientation in injection-molded PP articles. Unfortunately, it has been the author's experience that many of the standard tests for radiation stability--such as tensile, flexural, and impact testing--use injection molded, ASTM-type end-gated tensile bars and generally measure sample performance after irradiation only in the direction of the polymer flow.

The data also point to a possible effect of crystallinity on the retention of ductility. Figure 9 shows the break angle of specimens measured across orientation versus crystallinity (for various time periods), and suggests that by decreasing the crystallinity of the polymer, ductility retention can be increased. From a molecular standpoint, decreasing a polymer's crystallinity has been shown to make radical recombination occur at a faster rate,8 which should in turn lead to a lesser degree of oxidative degradation and an improvement in the retention of ductility. Although not the main point of the present work, this potential effect is nonetheless interesting to note.

CONCLUSION

It has been shown that the apparent radiation resistance of injection-molded PP samples is determined to a large extent by the direction of testing in relation to molecular orientation developed during the injection molding process. This reality becomes increasingly important as the drive to thin-wall injection-molded medical devices continues, and may be surprising in light of the rather small effect of molecular orientation on predose flexural properties in most of the samples. It has also been suggested that a reduction in polymer crystallinity may result in increased retention of ductility after irradiation.

REFERENCES

1. Encyclopedia of Polymer Science and Engineering, 2d ed, New York, Wiley, 1985.

2. Dunn TS, and Williams JL, "Radiation Stability of Polypropylene," J Indust Irrad Tech, 1:33­49, 1983.

3. Horng P, and Klemchuck P, "Stabilizers in Gamma-Irradiated Polypropylene," Plastics Eng, April, pp 35­37, 1984.

4. Rubin II, Injection Molding Theory and Practice, New York, Wiley Interscience, 1972.

5. Results of a screening experiment run at 30- and 50-kGy doses showed that a higher dose would indeed be needed to ensure that samples tested in the flow direction would fail within the 1000-hour time period. Unfortunately, we underestimated the effect of molecular orientation, as increasing the dose to approximately 70 kGy did not improve our situation (i.e., we were not able to distinguish between the materials when tested in the flow direction, and virtually none of the samples failed within the 1000-hour time period).

6. Huber JE, and Samuels RJ, in Interrelations between Processing Structure, and Properties of Polymeric Materials, Seferis JC, and Theocaris PS (eds), Amsterdam, Elsevier, 1984.

7. Dziemianowicz TS, Himont Development Report, August, 1984.

8. Dunn TS, Williams EE, and Williams JL, "The Dependence of Radical Termination Rates on Percent Crystallinity in Gamma-Irradiated Isotactic Polypropylene," Radiat Phys Chem, 19:287­290, 1982.

Geoffrey Hebert is leader of market intelligence for the polystyrene division of Novacor, Inc. (Leominster, MA), where he currently manages marketing support for the injection molding business segment.

Electronic Submissions Aim at Faster FDA Product Reviews

Medical Device & Diagnostic Industry Magazine | MDDI Article Index

Originally published May 1996

Device Approvals

FDA has worked out the kinks in its process for reviewing electronic submissions, opening the door for industry to submit in electronic form every type of medical device application from investigational device exemptions (IDE) to premarket approvals (PMA).

The holy grail of electronic submissions is faster review times, and it appears that the technology implemented by FDA has brought the agency to the verge of achieving that objective. "Clearly, for the last few submittals--from the reviewer's perspective and certainly from the administrative perspective--the review process has speeded up," says Neil Goldstein, deputy associate director of the Office of Systems and Management at FDA's Center for Devices and Radiological Health and one of the agency's prime movers in the electronic submissions process.

One company has already used electronic submissions extensively. Cardiac Pacemakers, Inc. (CPI; St. Paul, MN), which is part of the cardiac rhythm management group at Guidant (Indianapolis), has sent 12 electronic applications to FDA--two IDEs, nine PMA supplements, and an original PMA. Karen Peterson, manager of regulatory affairs for CPI, has capitalized on the process to build an electronic library of submissions that has helped the company to standardize its submission process. Peterson says feedback from FDA indicates that electronic submissions allow the reviewers to search for specific information in the applications. That ability can be enormously helpful, Peterson says. "Our submissions are huge," she explains. "One of our supplements alone was 2208 pages."

The first electronic applications were submitted on floppy disk, but the size of the documents has made CD-ROM the preferred method. Receiving submissions in this form means a reduction in the amount of paper that must be handled and a consequent improvement in document flow and document tracking. "Instead of having to get a half dozen copies to reviewers in different buildings, we put the electronic submission up on a server and reviewers can download it whenever they are ready," Goldstein says. Reviewers working with electronic documents can cut and paste materials electronically to create a review log that includes their comments. With paper-based submissions that process had to be done by hand.

As yet there are no written guidances on how to prepare electronic submissions, and reviewers have been trained selectively in the technology and methods involved. For those reasons, Goldstein says, "we would like sponsors to give us as much notice as they can before submitting an electronic application, just to be sure that we understand each other. We also need to identify reviewers, to make sure they have the necessary software and hardware, and more importantly, if they haven't done an electronic submission before, to be certain they have the right training." A workshop planned for summer might smooth the process. As it is now conceived, representatives of industry and FDA would use the workshop to examine and record the tenets of electronic submissions.

The pilot project that developed the current methodology for electronic submissions began in June 1994 with two PMA supplements sent in by CPI and another group within Guidant, Advanced Cardiovascular Systems (Santa Clara, CA), which focuses on vascular intervention technology. Hard lessons in software incompatibility and communications were learned in the early stages of these submissions.

The companies and FDA initially decided to use WordPerfect as a common software package, but it was soon apparent that the software was not up to the task. Reviewers found inconsistencies between the on-screen text and the paper copy submitted along with the electronic version, particularly differences in pagination. There were printing problems caused by different fonts. And graphics, created using such common software as AutoCad and Illustrator, could not be seen on-screen.

There were several other problems that seemed to accompany the medium itself. Reviewers wanted to make the equivalent of marginal notes on-screen, but couldn't. Even reading the document was a problem, because only a quarter of the text appearing on a printed page appeared on the computer screen at one time.

Communications were also flawed. FDA and the companies had agreed to use an off-the-shelf electronic mail and fax service. E-mail, it was later learned, was unworkable because WordPerfect files could not be sent using the software chosen--and the agency rejected the idea of sending non-ASCII files because of the potential for transmitting computer viruses.

Solutions to each of these problems, however, were found in the second phase of the pilot program. Software was acquired that allowed the electronic document to be viewed as it appeared on paper, independent of the word processing and graphics packages that were being used. The software used for document translation and display was Adobe Exchange and Adobe Acrobat.

Various documents can be distilled into portable document format using Adobe Acrobat, and the new document can be read with Adobe Exchange on any system--IBM-compatible PC, Macintosh, or even a UNIX-based platform. Janice Kates, regulatory affairs associate at Hybritech (San Diego), successfully used these software packages to create and read an electronic submission regarding an in vitro diagnostic product now being reviewed at FDA. Acrobat also provided the means for Kates and her colleagues to input intelligent navigational tools. "We can install links indicated by highlighted pieces of text," she says. "By clicking on them, the reviewer is taken to another place in the document. So we are able to help reviewers navigate around the document."

Another convenience of the software is the ability to insert electronic post-it notes containing dialogue at specific points in the document. Hybritech uses these notes to help the reviewer find related information or understand why information appears in certain places. Reviewers can also place their own post-it notes in the text, notes that bear reviewer names and comments. The notes can then be collated into a separate document with each one indicating the time and date created, who created it, and specific comments. Ultimately, video might be embedded at various points in the text. "I can see down the road where a manufacturer might have a hard time describing a process in narrative but might easily be able to demonstrate it on an embedded video," Goldstein says.

The E-mail problem of sending messages between reviewers and company staff was solved by using the Internet. Initially, messages sent over the Internet were encrypted to ensure security. Hybritech eliminated the need for encryption by simply agreeing with FDA not to send sensitive information by that route.

In the future, videoconferencing may be integrated into the process. Potentially, FDA reviewers could discuss an electronic submission among themselves, interactively pointing to text on the screen while voice and visual communications are simultaneously transmitted. That capability might be extended to include the companies submitting the applications, raising the potential of interactively making changes in the document, says Goldstein.

But even without videoconferencing, the agency is satisfied that a feasible methodology has been identified for electronic submission and review of medical device applications, Goldstein says. The pioneers of this effort agree. Both CPI and Hybritech plan to continue submitting applications electronically to FDA. Peterson, whose department handles CPI submissions to the agency, has set a goal of electronically submitting at least 75% of the company's 1996 applications. She and her colleagues are well ahead of schedule. All of the submissions so far this year have been electronic, "so we are definitely heading down that path," she says.--Greg Freiherr

THE CHANGING ROLE OF EDUCATION IN FDA'S REGULATION OF MEDICAL DEVICES

Medical Device & Diagnostic Industry Magazine | MDDI Article Index

Originally published May 1996

John C. Villforth, President, The Food and Drug Law Institute (Washington, DC)
Joseph S. Arcarese, Vice President, The Food and Drug Law Institute

FDA has for years struggled with the merits of voluntary compliance and education as complements to its traditional regulatory role. Within the agency, the Center for Devices and Radiological Health (CDRH) was for many years one of the chief promoters of the educational approach to protecting public health. Other FDA components argued for a stricter interpretation of the law, seeing the agency's mission in purely regulatory terms and viewing collaborative problem-solving with industry quite skeptically. Now, in the 1990s, with some exceptions, the educational approach seems to be waning, and the device center appears to have adopted a more traditional regulatory approach in dealing with industry.

Prior to joining FDA in 1971, the Bureau of Radiological Health (BRH) implemented a strategy involving both education and regulation in the control of unnecessary radiation exposure from electronic products. The bureau's concept of education was very broad, incorporating information exchange, scientific research, and problem-solving in collaboration with manufacturers, health professionals, and consumer organizations.

This concept, combining regulation, research, collaboration, and education, was codified in the Radiation Control for Health and Safety Act of 1968, a model law amending the Public Health Service Act. This multifactorial approach was very successful, especially when certain health problems it confronted were not amenable to straightforward regulation directed only at the equipment and the manufacturer. Having proven the approach's utility, BRH continued to use education as a public health tool along with regulatory compliance.

Anticipating passage of the Medical Device Amendments of 1976, FDA established the Bureau of Medical Devices (BMD) in 1974. Despite testimony at the congressional hearings leading to the amendments, where the strong role of the user in device safety and effectiveness was recognized, these amendments laid almost total responsibility on the manufacturer. BMD complied with the amendments' mandate and implemented its program accordingly. However, it did include a small educational group, the Office of Small Manufacturers Assistance (OSMA) (now known as the Division of Small Manufacturers Assistance), whose focus was on helping device manufacturers understand how to comply with the law. BMD also had a consumer affairs group, which promoted educational programs for the public, and a health affairs office that communicated with health professional organizations.

In 1982, BMD and BRH were merged to become CDRH. The new center incorporated the broader BRH vision of education into its structure by including the offices of Training and Assistance (into which it folded OSMA and the consumer affairs group), Science and Technology, Health Affairs, and Standards and Regulation. What distinguished CDRH from most other FDA centers, however, was not its activities to disseminate information to manufacturers, health professionals, and the public--which many centers did to one extent or another--but rather its programs to achieve consensus among these groups regarding various device-related health-care problems. In meetings with representatives of appropriate organizations, CDRH staff facilitated problem-solving discussions that developed cooperative programs to decrease the risk and increase the effectiveness of certain risk-laden device technologies. Anesthesiology, dialysis, central venous catheters, prenatal ultrasound, and contact lenses were some of the areas of discussion that led to the establishment of cooperative programs.

The center's anesthesia program, for example, has been remarkably successful, and large reductions in avoidable anesthesia deaths and injuries have been reflected in lower malpractice premiums for anesthesiologists and fewer product liability cases. It has demonstrated the efficacy of a win-win philosophy, according to which government can work collaboratively with other affected elements of society to achieve public health goals.

Unfortunately, such success stories are not likely to be repeated. The 1990s have seen FDA continue with its more traditional view of education as one-way communication. Although the agency continues to make full use of its advisory committees, special public meetings, and notice- and-comment rule-making procedures as means of obtaining outside input, it rarely if ever cedes its prerogatives by participating in cooperative problem-solving with other organizations.

FDA is now under withering fire from its critics, especially regarding product review. It is successfully implementing a number of internal administrative changes to increase its productivity. But whether these influences will open the agency as a whole and CDRH in particular to reconsider a broad understanding of education in its mission strategy remains to be seen.

John C. Villforth is a former director of FDA's Center for Devices and Radiological Health (1982­1990) and of its predecessors. Joseph S. Arcarese is a former director of the device center's Office of Training and Assistance (1984­ 1993). The views expressed are those of the authors and do not represent the Food and Drug Law Institute.