Studies on the Sterilization of Human Bone and Tendon Musculoskeletal Allograft Tissue Using Supercritical Carbon Dioxide

Anastasia Nichols, David C. Burns, Renee Christopher

Address for Correspondence:
Renee Christopher
Nova Sterilis Inc., 3109 N. Triphammer Rd, 
Lansing, NY 14882, USA
Phone:    + 1-607-330-2772
E-mail:  rac1@novasterilis.com

Abstract:

The risk of disease transmission from human musculoskeletal allograft tissue distributed by tissue banks is a subject of ongoing concern.  Efforts to minimize disease transmission have generally included donor screening, bioburden assessment, aseptic handling, chemical washes, antibiotics, and gamma irradiation.  Without adequate sterilization, however, viral and bacterial contamination of allograft tissue remains a significant problem.  Achieving terminal sterilization to a sterility assurance level of 10^-6 (SAL6; the standard for medical devices) frequently compromises biomechanical properties in the allograft.  Here we report that SAL6 sterilization of human allografts can be achieved using supercritical CO₂ with a peracetic acid-based additive.  The biomechanical properties of allografts sterilized with supercritical CO₂ are superior to those sterilized by traditional gamma irradiation. These findings suggest that supercritical CO₂ sterilization of human allograft tissue will increase the safety of allograft tissues while maintaining biomechanical properties.

J.Orthopaedics 2009;6(2)e9

Keywords:

sterilization; supercritical carbon dioxide; allograft tissue; bone

Introduction:

The importance of allograft tissue in the treatment of musculoskeletal disorders has grown dramatically since 1990.  Such disorders are the most frequently identified impairments of physical health in the United States (US).  It has been estimated that over 36 million Americans suffer from musculoskeletal conditions that limit their ability to function, with costs to society exceeding one billion dollars annually (1).

Allograft tissue is widely used by transplant surgeons for orthopedic (joint replacement), trauma, and cancer (surgical reconstruction) procedures (2), (3), (4), (5).  In the US, tissue bank distribution of such allografts has increased from 350,000 specimens in 1990 to over 1.6 million in 2005 (1), (6).  The need for safe and effective allografts continues to grow as allogenic transplantation has become the clinical therapeutic strategy of choice to combat musculoskeletal disorders.

US tissue banks that supply allografts from cadavers have traditionally operated with relatively little regulatory oversight.   In the past, bacterial and viral infections derived from transplanted musculoskeletal tissues have been rare events.  However, the potential for disease transmission remains a major concern for clinicians and patients (7).   Now, with demand growing for allografts, the US Food and Drug Administration (FDA) has introduced requirements for the manufacture of allograft tissue [21 CFR parts 1270 and 1271].

Despite more stringent regulations, tissue banks continue to utilize aseptic processing strategies, which may increase the potential risk to patients from non-sterile allografts.   Current allograft processing techniques often involve several disinfectant steps following aseptic donor harvest, including soaking in disinfectants and exposure to a sterilization step (8), (4), (9).   Allograft tissue samples processed by such methods have generally failed to achieve sterility assurance levels of 10-6 (SAL6) as required for the sterilization of medical devices (3), (4), (8), (10).  Moreover, preserving the osteogenic and biomechanical properties of structurally complex allograft tissue has proven to be challenging.

Figure 1 – Phase Diagram of CO₂.  Carbon Dioxide has four distinct phases; the standard solid, liquid and gas as well as the unique supercritical phase.

We have developed a method for sterilizing musculoskeletal allografts using supercritical carbon dioxide (SCCO2) [US patent 7,108,832] that involves the use of low temperature, low pressure and proprietary nontoxic sterilization additive.    Carbon dioxide has a unique critical point, defined by pressure (Pc=1,099 psi) and temperature (Tc=31.1˚C) at which the liquid and vapor phases become indistinguishable (Figure 1).  With its low surface tension, liquid-like density and gas-like diffusivity, SCCO2 is also ideal for temperature sensitive materials. 

Previous research has shown that proteins and other macromolecules are unreactive towards SCCO2 (11).  SCCO2 successfully inactivates viruses and produces minimal inflammatory reaction and satisfactory integration when used in conjunction with hydrogen peroxide on sheep bone allografts (12), (13).  Here we describe a novel method to achieve SAL6 using musculoskeletal allograft tissue that is terminally sterilized in double Tyvek® packaging.  Moreover, we demonstrate that the sterilization process does not compromise the biomechanical properties of the bone allograft.  The method achieves terminal sterilization while avoiding the damage caused by gamma or other processing methods.  Our experiments further indicate that sterilization by SCCO2 represents a viable solution to the pressing need to terminally sterilize musculoskeletal allograft tissue.

Materials and  Methods:

Allograft tissue samples

Bone-tendon-bone (BTB) and large tendon (Achilles and Tibialis; designated as research tissue) were provided by Community Tissue Services (CTS) (Dayton, OH).  Biomechanical studies utilized nine sets of trisected BTB samples from nine different donors (three BTBs harvested per donor). 

Cortical rings (femur and tibia) 15mm in height were also provided by CTS.  Femoral cortical bone struts measuring 4mm (W) x 4mm (H) x 30mm (L) were used in biomechanical studies.  Twelve groups of three struts per group were procured from a single donor at the same femoral location and with the same orientation. 

Allograft tissue samples used in sterilization protocols were processed either by Allowash™ or rinsed only with normal saline solution (0.9g NaCl).  Tissue designated as “Allowashed” was processed using Allowash™ (LifeNet, VA) at CTS prior to SCCO2 treatment at NovaSterilis.  Tissue designated as “saline rinsed” was rinsed extensively with sterile saline solution at NovaSterilis prior to SCCO2 treatment.

Sterilization Protocols/ Microbiological Assays

Gamma irradiation for biomechanical studies was performed by Sterigenics (Westerville, OH) at ambient temperature using a dose of 15-25 kGy.

SCCO2 sterilization was performed in-house using the 20 L Nova2200™ instrument.  Sterilization runs were conducted using NovaKill™ additive (peracetic acid based additive; 16 mL per run).  Additive was pipetted onto a 1 ½” x 7 ⅞” cellulose pad, which was then secured in the chamber’s lower 1” stainless steel basket using a holder.  Allograft samples terminally sealed in double Tyvek® pouches were arranged in stainless steel baskets (7” and 5”), which were stacked in the chamber. The Nova2200™ containing samples and additive was then charged with CO2­  from ambient conditions in 6-9 minutes to a pressure of 1436 ± 70 psi and a temperature of 35 ± 3˚C with constant stirring (680 ± 20 rpm).  System parameters and run times were maintained as specified, following which the vessel was depressurized over 15-25 minutes. 

Sterilization run times for tendon or bone allografts inoculated with B. atrophaeus spores were determined conservatively by fraction negative analysis, following ISOs 11138-1, 14937, and 11737-1.   Allograft samples were inoculated with Bacillus atrophaeus spore suspensions (>106 CFUs/10μL in aqueous solution) (SGM Biotech, MT) for fraction negative testing.  Inocula (10μL) were allowed to permeate the grafts at room temperature for 15 minutes.  Each allograft sample was transferred to the appropriately sized Tyvek® gas-permeable pouch, which was sealed before being inserted into a second terminally sealed Tyvek® pouch prior to sterilization treatment.

Allograft samples were then exposed to SCCO2 for various time intervals.  To assay for microbiological growth following treatment, allograft tissue samples were aseptically removed from packaging and cultured in bottles containing tryptic soy casein broth (Bacto) at 35˚C.  Each sample culture was observed daily for turbidity over a period of 7 days and scored for no growth (0) or growth (+).  Samples scoring negative for growth after 7 days were spiked with 10-100 CFUs B. atrophaeus to test for bacteriostasis.  If growth was observed in the spiked media, the sterility test was affirmed and the tissue judged sterile. 

The total time or full cycle of the process corresponds to twice the time calculated to result in the 6 log reduction as demonstrated by the time to achieve total kill in microbiological methods.  This method is in compliance with the “overkill” methodology outlined in ISO 11737-1 and as required for SAL6.

Biomechanical Testing

Biomechanical testing was performed by IMR Test Labs (Lansing, NY).  Creep testing was performed on an Instron Dynamite fatigue tester (Model 8841).  Each BTB allograft was clamped on bone sections using standard wedge grips.  The samples were manually cycled three times from 50N to 80N for preconditioning and loaded to 8N with the width of thickness of the tendon in the midpoint measured using digital calipers.  To determine creep measurements, the gap between the substantial bone segments was also measured. 

Samples were then subjected to tension cycles (1 sec) from 50N to 250N in a sinusoidal waveform, followed by a hold (15 sec) at 50N for a total of 100 cycles.  The final length between the bone sections was measured at 8N and the creep of the samples calculated as in Equation 1.

Creep (%) = (final length – original length)/original length x 100            Equation 1

The samples were loaded to failure using an Instron Tensile Tester (Model 5584).  The tibia section of the BTB was mounted inside plastic cylinders using epoxy resin.  A 3/8” diameter dowel was inserted between the bone end and the tendon to prevent tearing.  The midpoint of each tendon was trimmed to 5 mm.  The patellar bone section was placed in standard wedge grips.  The reduced section width was measured and the samples were then pulled at a crosshead speed of 0.5 inches per minute until a significant load drop was recorded.  From the load deflection data, the peak stress, elongation at break (based on 1 mm gauge length) and modulus were calculated.

For tendon biomechanics, 9 sets of trisected BTBs consisting of 3 BTBs harvested per donor (27 total samples) were divided into three groups.  SCCO2 sterilization was carried out for 4 hours (overkill from half cycle 90 min; SAL6).  All BTBs were loaded onto an Instron tensile testing machine and forced to failure at a rate of 1 inch elongation/minute.  Tensile strength was calculated as the maximal load divided by the cross sectional area.

For bone biomechanics, 36 bone struts (4mm x 4mm x 30mm) consisting of 3 cortical struts harvested per donor from the same location and orientation were divided into 3 groups with 12 struts each (one per donor).  These groups consisted of untreated, traditional gamma irradiated, and SCCO2 sterilized.  SCCO2 sterilization was carried out for 70 minutes (overkill, strut half cycle 20 minutes).  Subsequent to the respective treatments all bone struts were measured for density, 3 point bending, ash fraction, and collagen cross-linking.

Bone density measurements were performed according to ASTM D 792-00.  The struts were subsequently incubated at room temperature in phosphate-buffered saline (PBS) for 48 hours and kept moist during the testing.  Three point bending was carried out in accordance with ASTM D 790-03 using an Instron Model 5584 Universal Tester controlled by Instron Merlin software.  The specimens were supported using a 20mm span (L) with a 4.8mm loading diameter.  The load was applied to the center of the specimen (L/2) at a rate of 1 mm/min.  Flexural modulus was calculated from the slope of the loading curve.  Flexural strength was calculated following ASTM D 790-03 by integrating the area under the stress-strain curve.

After the bending tests were completed, the two halves of the failed specimens were examined to determine mineral content and collagen cross-linking, respectively.  Ash fraction was measured by ashing the specimens (ash mass/dry mass) (ASTM D 4630-01 with slight modification).  Bone samples were dehydrated by incubating at 100˚C for 48 hours, weighed, and then incubating at 800˚C for 24 hours.  The remaining ash was weighed and divided by the dry weight to determine the ash fraction.

Collagen cross-linking

Collagen cross-linking analysis was performed by Articular Engineering, LLC (Northbrook, IL).  Samples were first powdered using liquid nitrogen, demineralized using ethylenediaminetetraacetic acid (EDTA), and papain digested.  For each digest, two aliquots (2 mL each) were hydrolyzed using hydrochloric acid (HCl).  Hydroxy-lysyl pyridinoline and lysyl pyridinoline were separated using cellulose column chromatography and lyophilized samples were resuspended in 400 μL of 1% heptafluorobutyric acid in double distilled H2O (ddH2O).  Aliquots (100 μL) were assayed using high performance liquid chromatography (HPLC).

Calculations

Statistically significant differences between experimental and control groups were determined using analysis of variance (ANOVA) and paired T-testing using Prism graphpad software (Graphpad Software, CA).

Results :

The term “Sterility Assurance Level” (SAL), is routinely used in relationship to sterilization and is the probability of one microbial survivor is one in one million. B. atrophaeus endospores have been previously shown to be the most resistant microorganism to sterilize using our SCCO2 technology (11).  Inactivation kinetics of B. atrophaeus was examined by construction of a survivor curve by plotting the percent survival of B. atrophaeus as a function of time.  Ideally, the survivor curve is linear over the full range of inactivation.  Using the fraction negative method we show linear inactivation of greater than 106 CFUs B. atrophaeus spores on saline rinsed tendon using SCCO2 sterilization (Figure 2).  Similar data was obtained for bone, but is not shown. Samples of Allowashed™ tendon exposed to SCCO2 for 90 minutes cultured negative for B. atrophaeus growth, while also passing the bacteriostasis testing.  Similar sterilization of saline-rinsed tissue required 120 minutes.  Both times of inactivation are representative of half cycles.  SAL6, full cycle sterilization times were 180 minutes and 240 minutes, respectively.

Figure 2.  Survival curve for inactivation of Bacillus atrophaeus endospores inoculated saline rinsed tendon or allowashed preprocessed tendon. Survival curve for inactivation of Bacillus atrophaeus endospores  as a function of time.  Saline rinsed tendon (square) or allowashed tendon (diamond) was inoculated with greater than 106 CFUs and subjected to sterilization using SCCO2 and NovaKill additive at different times.  The results were calculated as positive growth/total growth as a function of time.

Biomechanical

Figure 3 depicts differences in the biomechanical properties of differently processed BTBs.  In samples exposed to gamma radiation, the data reveal variation in elongation and creep for several donors.  Since elongation is a factor in creep, these observations were unsurprising.  Although some individual donor differences in the treatment groups were apparent, ANOVA statistical analysis reveals no significant difference in BTB biomechanics between the groups (Table 1).

Table 1. Results of ANOVA calculations, Mean, and standard error for untreated control, Gamma irradiated and supercritical CO2 treated BTBs. Newton (N); Mega Pascal (MPa); Percent (%).

Figure 3.  Graphs comparing effects of sterilization methods of biomechanical properties of BTBs for all donors

Of the biomechanical parameters monitored in bone testing, no statistically significant differences in density, ash fraction, and Young’s modulus were observed in the three groups (Table 2).  Three-point bending revealed that gamma sterilization negatively impacted the biomechanical properties with respect to flexural strength, maximum load, and overall toughness (Table 2 and Figure 4).

Figure 4. Box plots of those measures showing significant differences between groups. A. Flexural strength, B. Toughness, and C. Max load all are unchanged by supercritical CO2 sterilization but compromised by gamma treatment.

Table 2.  Results of ANOVA calculations for Untreated, Gamma irradiated, and supercritical CO2 treatment groups with respect to the indicated measures. *one sample was removed due to experimental error, ** Data from 3 independent samples per group were excluded because clean breaks were not observed during testing resulting in artificially inflated toughness.

Paired T-testing confirmed no significant differences between control (untreated) and SCCO2 sterilized cortical bone (Table 3).  By contrast, statistically significant differences were noted between control and gamma irradiated bone (Table 2).

Table 3 . Results of paired T-testing for those measures with significant differences between groups.

Collagen Cross-Linking

Analysis of collagen cross-linking through measurement of hydroxylysylpyridinoline and lysylpyridinoline revealed no significant differences between control and test groups (Table 2).  In addition, no significant correlations were detected between any of the biomechanical measures and the level of collagen cross-links.

BTBs

Photographs of sterilized BTBs (Figure 5) indicated that SCCO₂ treatment produced significantly brighter and cleaner tendon samples than either untreated or gamma irradiated samples.  Pre and post-sterilization weighing revealed that BTB allografts lost approximately 14% of their mass.  That loss was attributed to extraction of lipids and excess fluids during the sterilization process.  In fact, extracted lipid could be detected in the oily residue present in the package after sterilization.

Figure 5. Photographs of bisected BTBs from a single donor with respective sterilization treatments.

Discussion :

Modest, but significant, differences were noted in the SCCO2 sterilization of Allowashed and saline-rinsed tendon.  In each case, samples were inoculated with 106 CFUs of B. atrophaeus spores, adding to whatever natural (presumably smaller) bioburden was already present in the tissue samples prior to washing.  A likely explanation of these modest differences is that Allowashed samples had an already diminished bioburden prior to inoculation, as compared to saline-washed samples.  However, the more important conclusion is that no preprocessing is required to achieve sterilization.  Regardless of their preprocessing history, tendon samples exposed to SCCO2 in the presence of NovakillTM afforded sterile tissue, free of cellular or sporular bacterial contamination. 

Results from this study also established that SCCO2 sterilization preserved the biomechanical properties of the cortical bone.  No significant difference in all measures between SCCO2 sterilized and untreated (Table 2, Table 3 and Figure 4) were apparent.  Preservation of these properties is important due to the fact that grafts are largely used in load bearing orthopedic applications.  SCCO2 sterilization is effective at achieving medical device levels of sterilization while maintaining essential biomechanical properties.

Interestingly, while some loss of lipid content was noted during tendon sterilization, lipid extraction from the graft has been shown in other studies with bone to increase graft incorporation following transplant (14).  Delipidation enhances accessibility to microporous structures in the bone and enhances osteoconduction following transplant (15), (16), (17).  These data demonstrated that besides achieving sterilization (SAL6), SCCO2 treatment also preserved the biomechanical properties of tendon.

Results reported here show that effective (SAL6) terminal sterilization of bone and tendon allograft tissue can be achieved while preserving essential allograft properties and maintaining essential biomechanical properties.  Moreover, SCCO2 sterilized bone is 254% tougher, can withstand 68% greater load, and has 66% more flexural strength than that of gamma irradiated samples. 

SCCO2 sterilized cortical bone is similar to untreated controls with respect to pre-yield (elastic) and post yield (plastic) properties.  This observation is consistent with the preservation both of mineral content, which is important for elastic properties, and collagen, which is thought to preserve bone plasticity. No significant differences in collagen cross-linking levels relative to controls were noted in any of the test groups.  Moreover, the observation that gamma irradiated samples maintain inherent Young’s modulus strength (an elastic property), but exhibit reduced plasticity (i.e. toughness, flexural strength) confirms earlier reports in the literature (5), (8).

A remaining concern with respect to SCCO2 sterilization of bone allografts is the osteoinduction and osteoconduction properties after sterilization.  The effect of sterilization on these processes are not within the scope of the current study but will be investigated in future studies.  However, protein profiles and content of Salmonella typhimurium inactivated using SCCO2 technology were unaffected when compared to protein profiles of non-sterilized (control) cells (11).  This coupled with our current observations strongly suggests the gentle nature of this process will preserve the essential properties of bone allografts.            

Tendon transplants restore movement and flexion to patients suffering from loss of function.  This study demonstrates that SCCO2 sterilization can be effective in achieving SAL6 (medical device level) sterilization while maintaining the quality of the tendon tissue.  A remaining concern with respect to supercritical CO2 sterilization is the vascularization and cellular properties of the tendon following transplant.  Examination of cellular structure by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) revealed no significant change in cellular structure.  Studies that definitively demonstrate competence of tendon allografts in vivo are currently in progress.

Adoption of SCCO2 sterilization by tissue banks will likely offer several advantages, including the ability to perform terminal sterilization in-house, preserve biomechanical properties of the allograft, and reduce the need for microbiological quantification of incoming bioburden.  SCCO2 also makes possible the use of industry standard validation methodologies, as with medical devices, and possible eventual parametric release of treated tissues.  These advantages will facilitate continued innovation and safe, high quality allografts by clearing existing safety and biomechanical concerns associated with current practices of tissue processing.  Ultimately, the ability to choose from a variety of sterilization options will increase patient safety as well as positive surgical outcomes.

Reference :

1. United States Bone and Joint Decade. Fast Facts on the Bone and Joint Decade.  2006.

2. DePaula CA, Truncale KG, Gertzman AA, Sunwoo MH, Dunn MG. Effects of hydrogen peroxide cleaning procedures on bone graft osteoinductivity and mechanical properties. Cell Tissue Bank 2005;6(4):287-98.

3. Arizono T, Iwamoto Y, Okuyama K, Sugioka Y. Ethylene oxide sterilization of bone grafts. Residual gas concentration and fibroblast toxicity. Acta Orthop Scand 1994 Dec;65(6):640-2.

4. Thoren K, Aspenberg P, Thorngren KG. Lipid extracted bank bone. Bone conductive and mechanical properties. Clin Orthop Relat Res 1995 Feb;(311):232-46.

5. Currey JD, Brear K, Zioupos P, Reilly GC. Effect of formaldehyde fixation on some mechanical properties of bovine bone. Biomaterials 1995 Nov;16(16):1267-71.

6. Organ Transplant and Grafts 1990 to 2009. Washington, DC: Census Bureau; 2009. Report No.: 171.

7. Wang S, Zinderman C, Wise R, Braun M. Infections and human tissue transplants: review of FDA MedWatch reports 2001-2004. Cell Tissue Bank 2007;8(3):211-9.

8. Akkus O, Rimnac CM. Fracture resistance of gamma radiation sterilized cortical bone allografts. J Orthop Res 2001 Sep;19(5):927-34.

9. Lomas RJ, Gillan HL, Matthews JB, Ingham E, Kearney JN. An evaluation of the capacity of differently prepared demineralised bone matrices (DBM) and toxic residuals of ethylene oxide (EtOx) to provoke an inflammatory response in vitro. Biomaterials 2001 May;22(9):913-21.

10. Cornu O, Banse X, Docquier PL, Luyckx S, Delloye C. Effect of freeze-drying and gamma irradiation on the mechanical properties of human cancellous bone. J Orthop Res 2000 May;18(3):426-31.

11. White A, Burns D, Christensen TW. Effective terminal sterilization using supercritical carbon dioxide. J Biotechnol 2006 Feb 20.

12. Fages J, Poirier B, Barbier Y, Frayssinet P, Joffret ML, Majewski W, et al. Viral inactivation of human bone tissue using supercritical fluid extraction. ASAIO J 1998 Jul;44(4):289-93.

13. Fages J, Poddevin N, King MW, Marois Y, Bronner J, Jakubiec B, et al. Use of supercritical fluid extraction as a method of cleaning anterior cruciate ligament prostheses: in vitro and in vivo validation. ASAIO J 1998 Jul;44(4):278-88.

14. Lomas R, Drummond O, Kearney JN. Processing of whole femoral head allografts: a method for improving clinical efficacy and safety. Cell Tissue Bank 2000;1(3):193-200.

15. Fages J, Marty A, Delga C, Condoret JS, Combes D, Frayssinet P. Use of supercritical CO2 for bone delipidation. Biomaterials 1994 Jul;15(9):650-6.

16. Kakiuchi M, Ono K. Preparation of bank bone using defatting, freeze-drying and sterilisation with ethylene oxide gas. Part 2. Clinical evaluation of its efficacy and safety. Int Orthop 1996;20(3):147-52.

17. Frayssinet P, Rouquet N, Mathon D, Autefage A, Fages J. Histological integration of allogeneic cancellous bone tissue treated by supercritical CO2 implanted in sheep bones. Biomaterials 1998 Dec;19(24):2247-53.

This is a peer reviewed paper 

Please cite as: Anastasia Nichols: Studies on the Sterilization of Human Bone and Tendon Musculoskeletal Allograft Tissue Using Supercritical Carbon Dioxide

J.Orthopaedics 2009;6(2)e9

URL: http://www.jortho.org/2009/6/2/e9

ANNOUNCEMENTS

Arthrocon 2011


Refresher Course in Hip Arthroplasty

13th March,  2011

At Malabar Palace,
Calicut, Kerala, India

Download Registration Form

For Details
Dr Anwar Marthya,
Ph:+91 9961303044

E-Mail:
anwarmh@gmail.com

Powered by
VirtualMedOnline