Human allograft tissues of tendon, bone, or bone and tendon combined are referred to as musculoskeletal allografts. These allografts are commonly used in orthopedic and sports medicine to repair and maintain analogous functions in the patient.
Safety and sterility assurance risks have previously limited the scope of allograft tissues as materials in medical device design. However, newer processes that enable higher gamma doses by using radioprotectants and carefully controlling conditions can achieve sterility assurance levels compatible with medical device specifications.
Devices combining allograft materials with polymers or metals can now be terminally sterilized in sterile barrier packaging with a sterility assurance level (SAL) of 10–6. This SAL enables the finished product to meet specifications commonly used in the medical device industry. Standards from AAMI and ISO now also allow a wide range of gamma dose levels to be used with medical devices. The selection of the appropriate dose, however, must be evaluated for every product and must take into account the material origins, processing, and risks inherent to the product.
For many treatments, orthopedic and cosmetic medical device implants made from metals and polymers are often used as an alternative to musculoskeletal allografts. Clinical considerations often influence a surgeon's choice. As biomaterials, allografts offer several advantages over artificial materials. One advantage is that bone and tendon are devitalized. Devitalization is a process of mechanical and chemical cleaning that removes most lipids, blood, and marrow, along with most of the antigens that could trigger a reaction in the patient. Another advantage is that bone implants promote osteointegration, or the ingrowth of a patient's own tissue. A protein called bone morphogenic protein (BMP), which is found in the bone matrix itself, stimulates a patient's body to infiltrate the structure with living cells called osteoblasts, which create bone. This fusion by the patient, called remodeling, speeds healing and can restore function to a high degree. Tendons are also remodeled by the body, with limited vascularization observed.
Comparing Artificial Materials and Allograft Biomaterials
Unlike artificial materials, musculoskeletal tissue allografts have a structure virtually identical to the target tissue. Microscopically, the matrix structure exhibits similar void space and surface characteristics to a patient's tissue. This similarity enables remodeling. Macroscopically, the density, strength, and elasticity (modulus) of allograft materials impart similar stress and flexibility characteristics, thereby yielding natural function, mobility, and appearance to many surgical interventions.
Devitalized tendon allograft tissue generally has a clean, smooth surface appearance. Bone allograft tissue that has been devitalized and lyophilized has a clean, dry, off-white surface appearance. It is shelf stable, making it easy to work with as a device biomaterial. Bone and tendon are usually soaked in sterile water or physiological saline immediately prior to use by the surgeon in order to rehydrate the tissue.
One difference for a device designer to consider between an artificial material, such as a medical-grade polyethylene polymer, and an allograft biomaterial, such as devitalized bone, is the inherent variability of the materials. Cadaveric biomaterials have a higher variability with respect to strength (tensile and compression), texture, elasticity (flexural modulus), and bioburden than artificial materials have.
Material Differences Can Compromise Sterility
Table I. (click to enlarge) Comparison of artificial materials and allograft
biomaterials. The differences in history, environment, structure, and fabrication between the types of materials create different risks to the final sterility of an assembled device.
Unlike for artificial materials, the history, environment, treatment, and structure of allograft materials before and during fabrication significantly affect the bioburden of the assembled device. These differences, with examples, are shown in Table I.
Allograft tissue presents an environment better suited to bacterial pathogens than the relatively hostile environment created by industrial molding and forming processes. Simply stated, eons of evolution have made bacterial pathogens better at infecting people than injection molding machines, steel mills, or a company's production line. Different environments do not just affect the numbers of microorganisms they harbor. The environment also affects the species and characteristics they exhibit.
All radiation sterilization methods specify a dose level, which is adjusted in response to the quantity, or bioburden, of organisms found on process raw materials, on intermediate assemblies, and in the environment. To determine a probability of sterility, an assumption of how much radiation is needed to kill a microorganism is required. A commonly used parameter to describe the organism's susceptibility is the D10 value, which defines the dose of radiation, in kilograys, required to kill 90%, or one log, of the organisms present at any time.
Table II. (click to enlarge) Susceptibility of organisms to gamma radiation. Values were determined in bovine serum albumin, with the five most resistant organisms repeated in freeze-dried, human bone. Source: Clearant Inc. results by Grieb et al.4
D10 values vary greatly from organism to organism. Spore-forming bacteria, for example, tend to have a higher D10 value than nonsporulating bacteria because the dormant spores tend to be more radioresistant. Gram-negative bacteria tend to have lower D10 values than gram-positive bacteria because they have weaker cell walls.1 Examples of bacterial susceptibility to gamma radiation are shown in Table II.
With AAMI/ISO/ANSI 11137, sampling or dose setting, depending on the method chosen, is based upon the standard distribution of resistance (SDR), which was established by analyzing 397 isolates from a medical device factory.2 Most raw-material manufacturing environments have low variability due to facility environmental controls and protocols (see Table I). Artificial raw materials, such as polymers, metals, and adhesives, often originate from petrochemical or metallurgical processes that are not likely to support bacterial or viral growth. The bioburden present is often located on the material surfaces, because the dense structure and the lack of available moisture of the material will not support microorganism growth deep within the material. The modification and assembly processes performed to create devices are carried out in controlled environments suitable to the device.3
As an incoming biomaterial entering the device factory, tissue has a much different history from artificial materials. A study that identified isolates from organ donors postmortem found a 27% occurrence of Clostridium species in blood and a 22% occurrence of Clostridium species in bone from donors with a positive blood-screening culture.5 The study also found a 7.3% Clostridium occurrence in bone from donors with a negative blood-screening culture. When compared with the SDR, Clostridium species (gram positive, mostly spore forming) can have a higher resistance to gamma radiation exposure in this population of isolates. Gram-positive, spore-forming organisms, such as Clostridium and Bacillus species, were in 30.1% of the isolates. These species tend to have higher D10 values (see Table II) and have been reported in transmission to recipients of allografts.6
To achieve an SAL of 10–6 using allograft tissue, a level typically used for medical devices, a higher dose of radiation is required than that used for metals or synthetic materials if preirradiation cleaning cannot deliver materials with the same bioburden and bacterial population as the artificial materials. An SAL is the probability that a single medical device, drug dose, or biologic unit will be contaminated with one or more viable microorganisms. It is a functional definition of sterility and is used in industry because the absolute definition—the complete absence of viable organisms—is impossible to reliably verify.
Simply stated, it's like running an inverted lottery that results in at least a million winners to every one loser. The patients are the winners by receiving treatment with a sterile product. Similar to a lottery, probability is used to calculate the dose necessary prior to processing based on the knowledge that bacteria tend to respond in a log-linear manner to radiation exposure.
To achieve a 10–6 SAL using radiation, it is necessary to first kill the bioburden, as well as to provide a dose of irradiation equivalent to six additional logs of lethality. In a hypothetical example of a device that harbors a bioburden of 1000, or 103, Clostridiumsordelli per unit, the dose would be
Dose (SAL 10–6) = 3 × D10 organism + 6 × D10 organism. (1)
Using D10 C. sordelli (see Table II), the dose would be
Dose (SAL 10–6) = (3 × 4.3 kGy) + (6 × 4.3 kGy) = 38.7 kGy. (2)
The first equation illustrates that nine logs must be removed, or six logs more bacteria than is suspected to be on the product, to achieve the sterile ratio if this single microorganism were the only one to deal with. The variety of organisms found in most manufacturing environments makes it necessary to estimate how an entire population of organisms (e.g., mixture of low and high D10 values) would respond to a given radiation dose, as opposed to a single resistant species.
Complicating matters further, some microorganisms are not killed by radiation until the dose delivered exceeds a limit, at which point they exhibit a log-linear response, effectively increasing the total dose required by the limit. The population calculations, which are beyond the scope of this article, are based upon the SDR and are employed by the device industry successfully in several standard methods (e.g., AAMI/ISO 11137 method 1 and VDmax extrapolated to derive method 2).7
Using the SDR as an assumption of the bacterial load in tissue is problematic because it does not represent the environment of the tissue before and during collection (see Table I). SDR samples were taken in predominantly aerobic conditions. A large percentage of cadaveric isolates, however, were Clostridium species, an anaerobic spore-forming genus of bacteria.5 When compared with open-air environments, human deep tissues have a low oxygen concentration and are considered anaerobic. In addition, treatment of the allograft tissue donor prior to death differs depending on the clinical events. Treatment can involve extended administration of antibiotics that selectively kill some types of bacteria, allowing others to increase. Fatal trauma often involves the rupture of internal membranes combined with intravenous administration of fluids, which can also spread and place selection pressure on bacteria. These variable clinical factors are not reflected in the SDR applied to artificial materials during manufacture.
When combining cadaveric materials and low-dose irradiation to manufacture medical devices, it is imperative to mitigate such risks by validating SDR equivalence of the cleaned and processed allograft material with those used to create the standards. Lacking a suitable SDR, using the inactivation kinetics of the most radioresistant organisms that pose a threat is an alternative method to calculate both the lethality and the SAL.
The actual practice of determining these kinetics involves spiking known amounts of bacteria into processed allograft tissue followed by the precise application of gamma radiation at different dose levels. Although tedious, this testing has been done and is available from commercial tissue sterilization services.
Another option for device designers is the use of high dose sterilization methods, such as the levels available in AAMI TIR-33 (e.g., VDmax35), which provides audit plans for maintenance of a validated state based on the bioburden with a dose sufficient to handle radiation-resistant microorganisms.
Viral Inactivation Using Gamma Irradiation
Donor screening and testing effectively reduce the risk of viral transmission from the source. In the event that virally contaminated tissue gets past the screening and processing steps, high-dose controlled irradiation has been shown to provide robust viral inactivation.
Inactivation in excess of four logs has been demonstrated for human immunodeficiency virus 1, hepatitis A, and West Nile virus when a 50-kGy dose is delivered in albumin solutions.6 Using enveloped (Sindbis) and nonenveloped (porcine parvovirus) models, high doses of radiation applied to viruses spiked into radioprotected tissue demonstrated robust inactivation.8,9 These viruses are thought to be good models for HIV, hepatitis C, West Nile virus, human B-19, and hepatitis A.4,9
Maintaining Strength with High-Dose Gamma Irradiation
Free radicals generated during irradiation damage tissue structure. This potential for damage has previously discouraged the use of high doses (e.g., exceeding 25 kGy) of gamma radiation to treat allograft materials. New high-dose irradiation processes substantially preserve the structure of the tissue by protecting it with chemicals that sacrificially quench the free radicals. These chemicals, called radioprotectants, are even more effective if the conditions are carefully controlled during irradiation.
Table III. (click to enlarge) Tendon collagen damage by protease digestion assay. Results shown are from chymotrypsin digestion of human tendon segments. Source: Clearant Inc. internal documents.
Intact collagen is the structural building block in bone and tendon. Gamma radiation reduces the strength of tissue by disrupting, or denaturing, its structure irreversibly. A study has measured collagen damage using a protease digestion assay (see results in Table III). The previously unpublished results show substantial (95%) collagen denaturation when 50 kGy of radiation is applied at ambient conditions.
The addition of a controlled low-temperature environment and other controls during irradiation can preserve more than half of the collagen structure. Combining the controlled irradiation process with radioprotectants increases the protection further: in the study, only 8% of the collagen structure was denatured, even though a 50-kGy dose was delivered to the tendon.
Table IV. (click to enlarge) Biomechanical properties of bone grafts tested in compression. Source: Grieb et al., “Preservation of the Structural and Functional Properties of Irradiation Using the Clearant Process,” unpublished paper available from Clearant Inc.
The compressive strength of lyophilized human bone in Table IV shows no statistically significant reductions in strength when a sterilizing dose (50 kGy) of controlled gamma radiation with radioprotectants is applied to different bone types compared with 0-kGy controls or low-dose uncontrolled gamma samples. These results correspond to the denaturation results in Table III.
The efficacy of optimized irradiation conditions combined with radioprotectants has shown similar strength preservation results using different types of bone, such as dense cortical bone or spongy cancellous bone. Substantial preservation of tensile strength and elasticity (modulus) has also been observed using tendon materials. Both tendons and bones originating from different anatomical locations have shown similar results, thereby making a variety of tissue allograft materials available for specific device applications.
Device Design Considerations
Device designers and engineers must take many issues into consideration during product design and development phases when evaluating the use of sterile allograft materials. Final product storage and stability are influenced by material choice. Irradiated lyophilized bone materials are generally shelf stable at ambient conditions, whereas tendons are commonly kept frozen until surgical use.
Final specification requirements determine tissue selection. Cancellous bone, for example, has a spongy appearance with high void space and low mechanical strength, while cortical bone is denser with low void space. Its mechanical strength is nearly 10 times higher than that of cancellous bone.
The anatomical source of tissue limits the dimensions of a component. A human femur (upper leg bone), for example, is much larger than an ulna (small lower arm bone). Another type of material combines crushed bone and artificial materials, such as binders and cements, which can enhance strength and fabrication properties and overcome dimensional limitations for some applications.
A design using safe tissue materials starts by understanding tissue-banking practices. To reduce the risk of disease transmission, tissue banks screen donors with assays and culture samples taken postmortem. Clinical events leading to donor expiration and lifestyle factors of the donor are also included in the screening process.
Tissue banks and processors use aseptic processing with protocols designed to prevent cross-contamination between donors and to prevent introduction of additional bioburden to the tissue. The specific practices and regulations are determined by both nonprofit organizations such as the American Association for Tissue Banking (AATB) and by government regulations (e.g., 21 CFR 1271) and implemented by the tissue banks. See the sidebar, “Tips for Risk Mitigation.”
When using lower-dose sterilization protocols such as VDmax15, the tissue-cleaning processes used prior to terminal sterilization must be validated using methods that assess conditions on the surface as well as those conditions within the tissue matrix.
For processes using sterilization doses exceeding 25 kGy, methods of selection and introduction of radioprotectants into the tissue matrix have been designed requiring minimal specialized equipment. Radioprotectant solution is available premixed, filtered, aseptically filled, and ready to use with tissue. The design process must also take into account the interaction of other device materials, such as metals, polymers, and adhesives, with the radioprotectant and with the tissue itself.
The effect of having processing steps between the introduction of radioprotectant and the irradiation step also requires consideration. Choosing sterile barrier packaging materials already qualified with the sterilization method can save development time and cost. Lyophilization and storage conditions during this period must be carefully controlled (generally not an issue at most existing device or tissue bank facilities).
Finally, the irradiation process itself requires added engineering controls. Factors such as low-temperature maintenance and control, loading to control dose variance, and validation of dose delivery to the device must be considered. Many commercial sterilization service providers have already worked out these issues for similar applications and should be consulted early in the design process to avoid problems and accelerate the overall development process.
Enabling New Possibilities
New terminal sterilization technologies combined with the benefits of allografts should enable new medical devices and improved treatments by increasing safety, reducing both patient risk and manufacturing challenges. The differences between artificial materials and allografts have created a different kind of bioburden from that found in the SDR applied to artificial materials.
Dose levels used in the new technology can provide a 10–6 SAL with these organisms. Safer orthopedic devices using human allograft tissue are possible and readily implemented. Applications beyond orthopedic devices may also be possible. Similar safety and risk mitigation advantages using high-dose gamma radiation with xenograft (from other species) tissue are also being studied.
An organization with little prior allograft experience should begin seeking information from national nonprofit organizations such as the AATB for guidance. Any organization can handle allograft material issues by following sound design control and risk management practices already in use for other devices. Working with sterilization service suppliers can reduce development time of terminally sterilized devices.
Steve Burns is director of technology transfer and manufacturing for Clearant Inc. (Los Angeles). He can be reached at [email protected].
1. B Ouattara et al., “Combined Effect of Gamma Irradiation, Ascorbic Acid, and Edible Coating on the Improvement of Microbial and Biochemical Characteristics of Ground Beef,” Journal of Food Protection 65, no. 6 (2002): 985.
2. JL Whitby and AK Gelda, “Use of Incremental Doses of Cobalt 60 Radiation as a Means to Determine Radiation Sterilization Dose,” Journal of the Parenteral Drug Association 33 (1979): 3.
3. Code of Federal Regulations, 21 CFR 820.70 C.
4. T Grieb et al., “High-Dose Gamma Irradiation for Soft Tissue Allografts: High Margin of Safety with Biomechanical Integrity,” Journal of Orthopaedic Research 24 (2006): 1011–1018.
5. O Martinez et al., “Postmortem Bacteriology of Cadaver Tissue Donors: An Evaluation of Blood Cultures as an Index of Tissue Sterility,” Diagnostic Microbiology and Infectious Disease 3, (1985): 193–200.
6. LK Archibald, “Update: Allograft-Associated Bacterial Infections—United States,” Morbidity and Mortality Weekly Report 51, no. 10 (2002): 207–210.
7. KW Davis, WE Strawderman, and JL Whitby, “The Rationale and a Computer Evaluation of a Gamma Irradiation Sterilization Dose Determination Method for Medical Devices Using Substerilization Incremental Dose Sterility Test Protocol,” Journal of Applied Bacteriology 57 (1984): 31–50.
8. Clearant Inc. internal documents and D Mann et al., “High Doses of Gamma Radiation Are Both Necessary and Feasible When Sterilizing Allograft Tissues with Gamma Irradiation” (poster presented at the AATB National Meeting, Hollywood, FL, September 17–20, 2005).
9. WH Burgess and RW Amareld, “Controlled Gamma-Irradiation Mediated Pathogen Inactivation of Human Urokinase Preparations with Significant Recovery of Enzymatic Activity,” Biologicals 31 (2003): 261–264.