In vitro biocompatibility of polyetheretherketone and polysulfone composites L. M. Wenz,* K. Memitt,+ S. A. Brown,+and A. Moet* Departments of Macromolecular Science and Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio 44106 A.D. Steffee Co-Director, Cleveland Spine and Arthritis Center, Cleveland, Ohio
Short carbon fiber reinforced composites could potentially replace some of the metal alloys used in orthopedic implants. In particular, polysulfone and, more recently, polyetheretherketone have been considered as the matrix material for carbon fiber reinforced composite implant materials. ASTM standards F813 and F619 for direct contact cell culture evaluation and extraction were employed to determine the in vitro biocompatibility of a carbon fiber composite
of polyetheretherketone, PEEK, in comparison to a carbon fiber reinforced polysulfone composite. The cell cultures were assessed qualitatively by microscopy and quantitatively using an enzyme assay to determine cytotoxicity. Overall, the cellular response to the PEEK and polysulfone composites were negligible indicating that further in vivo studies with these materials are appropriate.
INTRODUCTION
Several studies have focused on the development of polymer composites as implant materials with the carbon fiber reinforced composites of special interest for use as osteosynthesis plates. '-'Polymer composites offer many advantages over the current orthopedic metallic alloys since they have good corrosion resistance, elicit minimal allergic responses, and are radiolucent. Furthermore, the modulus and strength of a composite implant can be controlled by changing the fiber content and orientation. According to Wolff's law, bone requires a mechanical stimulus to grow.4 Therefore, a polymer composite with a modulus closer to bone, 18-20 GPa, should cause minimal stress shielding as compared to a stiffer metallic implant. This effect has been documented in several s t ~ d i e s . ~ Currently only a few polymer composites have been investigated for orthopedic implants. More polymer composites need to be identified that have good resistance to crazing, autoclave stability, the appropriate mechanical properties, and are biocompatible. A new high-temperature polymer, polyetheretherketone (PEEK), has recently been considered as an implant material. From a mechanical and material perspective, PEEK pos' j 6
Journal of Biomedical Materials Research, Vol. 24, 207-215 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/020207-09$04.00
WENZ ET AL.
208
sesses some very attractive properties for implant applications. In particular, the 30% short polyacrylonitrile (PAN) carbon fiber composite of PEEK has an elastic modulus of 17 GPa (bone 18-20 GPa) and flexural strength of 320 MPa (bone 150-250 MPa).5Furthermore, PEEK exhibits high stress corrosion resistance in most environments as compared to polysulfone comp o s i t e ~However, .~ with the exception of one study by Williams et a1.,6 no published reports have assessed the biocompatibility of PEEK. This report is the first in a series of studies addressing the in vitro and in vivo biocompatibility of PEEK. In this study the in vitro biocompatibility of PEEK was measured using cultures of Souse fibroblast cells. MATERIALS AND METHODS
Both extract and direct contact cell culture tests were used to measure the cellular response to the composites and their extracts. The materials were evaluated for 24 and 96 h in direct contact as per ASTM F813,8 using cultures of mouse fibroblast L929’s cells (ATCC #CCL l). In the extract tests, extract liquids from the composite materials were added to cell cultures to determine if any cytotoxic leachables were released from the composites. Materials The materials evaluated are listed in Table I and were divided into two experimental groups based on material availability. The composites in group 1 were 30% short PAN fiber reinforced polysulfone, PS (GC-1006, Wilson Fiberfill), and PEEK (J 1105/CF/30, LNP Corporation). A polysulfone carbon fiber composite was included since this material has been previously evaluated as a bi~material.~,” Unfortunately, no information was available on any additives or mold release agents that might have been incorporated in these composites. The second group of experiments were conducted with 30% short fiber reinforced PEEK (J 1105/CF/30, LNP Corporation) that TABLE I Materials for Direct Contact and Extract Tests
Direct Contact
Group 1
PS”
Group 2
PEEK‘ Latexd
PEEK^
96 h
Extract
PS PEEK PEEK Latex
PS PEEK PEEK Latex
”30% PAN carbon fiber reinforced polysulfone, Wilson-Fiberfill, 1985. b30% PAN carbon fiber reinforced polyetheretherketone, LNP, 1985. ‘30% PAN carbon fiber reinforced polyetheretherketone, LNP, 1988. dLatex urethral catheter, Bard, 1988.
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
209
was known not to contain any additives. Sections of a Bard urethral latex catheter, abbreviated LATEX, were included in the second group of experiments as a positive material control. All materials were tested in triplicate. Extract tests The extract tests were conducted following ASTM standards F619 and F813.8r"An extraction vehicle of minimum essential medium with 10% horse serum and sodium bicarbonate was chosen so that the extract liquids could be added directly to cell cultures. The surface areas of the PEEK and PS materials, 12 cm', and the LATEX, 6 cm', although smaller than what is recommended by the standard, were chosen to better approximate the surface area to volume ratios achieved in the direct contact tests. The materials were steam (PEEK and PS) or gas (latex) sterilized, each aseptically added to 20 mL of the extract liquid, and extracted for 120 h at 37°C. The extract liquids were stored at room temperature and then warmed to 37°C and added to 35 x 10 mm culture dishes containing monolayers of mouse fibroblast cells. Four cultures were given extract liquid from a control tube that did not contain any material. The cultures were incubated at 37°C in 5% CO, for 96 h then qualitatively graded on the basis of cellular appearance using a scale of 1-5 as indicated in Table II.12 Direct contact test and LDH assay The direct contact tests were conducted following ASTM F813. The steam sterilized PEEK and PS materials and the gas sterilized LATEX were carefully placed directly onto monolayers of fibroblast cells in 35 x 10 mm culture dishes, The surface area of the PEEK and PS materials in contact with the cells was 100 mm'. The specimen geometry of the latex, a cylindrical tube, and the culture dish diameter, restricted the latex surface area in contact with the cells to 15 mm2. The separate experiments were conducted with the materials in culture for 24 and 96 h at 37°C. Four control cultures without materials were included in each experiment. After incubation the two direct contact experiments were analyzed using qualitative and quantitative techniques. TABLE I1 Scale for Qualitative Cell Indexing 0 1 2 3
4 5
No observable lysis < 20% of cell culture lysed < 40% of cell culture lysed < 60% of cell culture lysed < 80% of cell culture lysed > 80% lysis in cell culture
210
WENZ ET AL.
A quantitative method was also employed to measure the cellular response to the extract liquids and the media from the direct contact tests. Lactate dehydrogenase, LDH, is an enzyme that is released from the cell when the cell membrane is ruptured. Therefore, the amount of LDH present in the media at the completion of the test correlates to the amount of cell death in the culture. LDH has been previously used as a quantitative measure of cytotoxicity by Rae and is easily measured using Sigma Chemical kit #3#-W.'3 The four control cultures included in the experiments were used as controls for LDH analysis. After the 96-h incubation, two of the four control cultures were given 100 p L of Triton X100 and held at 37°C for 2 h. These cultures were labeled Ct and served as positive controls for LDH testing since triton causes complete cell death. The other two control cultures were labeled C- and served as negative controls for LDH analysis. The amount of LDH in each culture is measured in Units, U, and is corrected for background activity by subtracting the average activity of the negative controls, Uc-. The LDH activity of the culture is then expressed as a percentage of the total possible activity in the cultures as given in the formula. % LDH Activity = 100 X (USpec - Uc-)/(Uc+ - Uc-)
Therefore, by definition, the positive control cultures, Ct, have 100% LDH activity and the negative control cultures, C-, have 0% LDH activity. Expressing the data in this manner allows for comparison between the different experiments. RESULTS
All the cell cultures were very healthy after the 96-h exposure to the extract liquids. The cultures with extract liquids from the PEEK and latex materials did not differ in appearance from the two negative control cultures, C-'s. All the C-, PEEK, PS, and latex cultures in the extract experiment were graded as 0 for no observable cell lysis. A monolayer representative of these cultures with attached and rounded fibroblasts is shown in Figure 1. The round, unattached fibroblast cells were evidence that the culture had matured beyond a monolayer consequently leaving no more sites for cell attachment. Complete cell lysis was observed in the positive control cultures exposed to Triton x 100. These C' cultures were graded as 5 for greater than 80% cell lysis. The combined average results in % LDH activity for the PS, PEEK, and LATEX extracts are presented in Table 111. The LDH values are expressed as % LDH activity divided by the surface area of the materials to correct for surface area differences. The % LDH activity for the cultures exposed to PEEK, PS, and LATEX extracts was at or slightly above the background LDH activity in the negative controls. No differences between the materials were statistically significant using the two sample T-test ( p < 0.05).
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
211
Figure 1. Monolayer after testing, 50X, from the C- culture that represents the appearance of all the cultures exposed to extract liquids.
TABLE 111 Average % LDH Activity per cm2 Surface Area for the Extract Test Materials. Standard Deviations Are Given in the Parenthesis ~
~~~
Material PEEK PS Latex
%LDH/cm*
0.23 (0.16)
4.04 (3.26) 0.02 (0.02)
In the direct contact tests, the cells surrounding and underneath the materials were examined and compared to the healthy monolayers in the negative control cultures. In all the PEEK and PS cultures, healthy cell monolayers with several unattached fibroblasts were observed adjacent to and surrounding the materials. Figure 2 illustrates that the cells were growing directly adjacent to a PEEK composite after 24 h in culture. The same cell-material contact was seen in the PS and PEEK cultures after 96 h. Furthermore, a healthy cell monolayer existed directly underneath the PEEK and PS composites as shown in Figure 3. Some elongated fibroblasts were observed at the composite edges as shown in Figure 4. No qualitative differences were observed between the 24- and 96-h cultures for either PEEK or PS. However, the cell cultures exposed to latex for 24 and 96 h were very unhealthy. After 24 h, cell lysis dominated in the immediate vicinity of the LATEX
212
WENZ ET AL.
Figure 2. Corner of PEEK composite, 50X, in culture after Direct Contact 24-h testing. Healthy cells are in direct contact with the material edge.
Figure 3. Monolayer under PEEK composite, 50 X, after Direct Contact 24-h testing.
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
213
Figure 4. Cells at the PEEK composite edge, 50x, after Direct Contact 24-h testing.
material. However, in areas farther away from the material, a healthy cell monolayer was present. After a 96-h exposure to the latex, almost complete cell lysis had occurred. The cultures at 24 and 96 h were qualitatively graded as 4 and 5, respectively. The LDH results for the direct contact tests also correlated well with the qualitative observations. The direct contact results presented in Table IV are expressed as the % LDH activity divided by the material area in contact with the culture. The data is corrected for material contact area because the cylindrical shape of the LATEX catheter significantly reduced the material contact with the cells. The PEEK and PS cultures exhibited negligible cell death at both 24 and 96 h. However, the LATEX cultures had statistically significant ( p < 0.05) higher cell death, two samples T-test, when compared to the composites. DISCUSSION
From the results of the direct contact and extract tests it is apparent that the carbon fiber reinforced PEEK and PS composites exhibited excellent in vitro biocompatibility with cultures of mouse fibroblast cells. The qualitative cell indexing corresponded with the quantitative LDH results presented. The cellular compatibility of PS that was observed supports the in vivo work
WENZ ET AL.
214
TABLE IV Average % LDH Activity per cm2 Surface Area in the 24- and 96-Hour Direct Contact Tests. Standard Deviations Are Given in the Parenthesis. Statistically Significant Values Are Indicated by an Asterisk 96 Hours
24 Hours
Material
PEEK PS Latex
%L D H / ~
0.00 (0.00) 1.56 (1.87) 144.02* (22.83)
% LDH/cm2
6.04 (3.03) 6.46 (4.29) 385.37 (86.42)
conducted by S p e ~ t o rSince . ~ the chemical formulas of PEEK and PS are very similar as shown in Figure 5, it is not surprising that PEEK exhibited good in vitro biocompatibility results. The cytotoxic response to the LATEX was expected since it is considered a positive material control for the direct contact ASTM standard. The lack of a cytotoxic response to any of the extract liquids tested indicates that no detectable cytotoxic leachables were extracted from the materials during the 120-h 37°C condition. These cell culture tests have served to screen PEEK and PS composites for potential toxicity in vivo. Because of the simplified and stagnant conditions in cell culture experiments, the in vitro results cannot completely predict in vivo results. However, these promising results do indicate a strong potential for PEEK biocompatibility and in vim studies are warranted. CONCLUSIONS
In vitro cell culture studies using mouse fibroblast cells in direct contact and extract experiments demonstrated that PEEK and PS exhibit excellent in vitro cellular biocompatibility. Further work is being conducted to assess
CH3 (b)
Figure 5. Clemical formulas for a) polyetheretherketone and b) polysulfone.
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
215
the in vim biocompatibility of PEEK and the biomechanical compatibility of the composite as an osteosynthesis plate. This work was supported by Acromed Corporation, Cleveland, Ohio.
References 1. J. S. Bradley, G. W. Hastings, and C. Johnson-Nurse, ”Carbon fibre reinforced epoxy as a high strength, low modulus material for internal fixation plates,” Biomaterials, 1, 38-40 (1980). 2. L. Claes, W. Huttner, and R. Weiss, “Mechanical properties of carbon fibre reinforced polysulfone plates for internal fracture fixation,“ in Biological and Biomechanical Performance of Biomaterials, P. Christel, A. Meunier, and A. J. C. Lee (Eds.), Elsevier Science, Amsterdam, 1986, 81-86. 3. N. Gillett, S. A. Brown, J. H. Dumbleton, and R. P. Pool, ”The use of short carbon fibre reinforced thermoplastic plates for fracture fixation, ” Biomaterials, 6, 113-121 (1985). 4. S. A. Brown, ”Biomechanical compatibility,” in Biocompatibilify of Orthopedic Implants, Vol. 1 , D. F. Williams (Ed.), CRC Press, Boca Raton, 1982, pp. 75-110. 5. R. S. Hastings, S. A. Brown, and A. Moet, “Characterization of short fiber reinforced polymers for fracture fixation devices,” presented at the Thirteenth Annual Meeting of the Society for Biomaterials, New York, June 3-7, 1987. 6. D. F. Williams, A. McNamara, and R. M. Turner, ”Potential of polyetheretherketone (PEEK) and carbon fibre-reinforced PEEK in medical applications,” J. Mater. Sci. Lett., 6(2), 188-190 (1987). 7. F. N. Cogswell and M. Hopprich, “Environmental resistance of carbon fibre-reinforced polyetheretherketone,” Composites, 14(3), 1983, 251-253. 8. ASTM F813 ”Standard practice for direct contact cell culture evaluation of materials for medical devices” in Annual Book of A S T M Standards 13.01. ASTM, Philadelphia, 1987. 9. C. A. Behling and M. Spector, ”Quantitative characterization of cells at the interface of long-term implants of selected polymers,” J. Biomed. Mater. Res., 20, 1986, 653-666. 10. M. Spector, “A high modulus polymer for porous orthopedic implants: biomechanical compatibility of porous implants,” I. Biomed. Mafer. Res., 12, 1978, 665-677. 11. ASTM F619 ”Standard practice for extraction of medical plastics” in Annual Book of A S T M Standards 13.01. ASTM, Philadelphia, 1987. 12. ASTM F895 ”Standard test method for agar diffusion cell culture Screening,” in Annual Book of A S T M Standards 13.01. ASTM, Philadelphia, 1987. 13. T.Rae, ”Tissue culture techniques in biocompatibility testing,” in CRC Techniques of Biocompafibility Testing, Vol. 2 , D. F. Williams (Ed.), CRC Press, Boca Raton, 1986, pp. 81-93. Received January 23, 1989 Accepted August 22, 1989
Short carbon fiber reinforced composites could potentially replace some of the metal alloys used in orthopedic implants. In particular, polysulfone and, more recently, polyetheretherketone have been considered as the matrix material for carbon fiber reinforced composite implant materials. ASTM standards F813 and F619 for direct contact cell culture evaluation and extraction were employed to determine the in vitro biocompatibility of a carbon fiber composite
of polyetheretherketone, PEEK, in comparison to a carbon fiber reinforced polysulfone composite. The cell cultures were assessed qualitatively by microscopy and quantitatively using an enzyme assay to determine cytotoxicity. Overall, the cellular response to the PEEK and polysulfone composites were negligible indicating that further in vivo studies with these materials are appropriate.
INTRODUCTION
Several studies have focused on the development of polymer composites as implant materials with the carbon fiber reinforced composites of special interest for use as osteosynthesis plates. '-'Polymer composites offer many advantages over the current orthopedic metallic alloys since they have good corrosion resistance, elicit minimal allergic responses, and are radiolucent. Furthermore, the modulus and strength of a composite implant can be controlled by changing the fiber content and orientation. According to Wolff's law, bone requires a mechanical stimulus to grow.4 Therefore, a polymer composite with a modulus closer to bone, 18-20 GPa, should cause minimal stress shielding as compared to a stiffer metallic implant. This effect has been documented in several s t ~ d i e s . ~ Currently only a few polymer composites have been investigated for orthopedic implants. More polymer composites need to be identified that have good resistance to crazing, autoclave stability, the appropriate mechanical properties, and are biocompatible. A new high-temperature polymer, polyetheretherketone (PEEK), has recently been considered as an implant material. From a mechanical and material perspective, PEEK pos' j 6
Journal of Biomedical Materials Research, Vol. 24, 207-215 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/020207-09$04.00
WENZ ET AL.
208
sesses some very attractive properties for implant applications. In particular, the 30% short polyacrylonitrile (PAN) carbon fiber composite of PEEK has an elastic modulus of 17 GPa (bone 18-20 GPa) and flexural strength of 320 MPa (bone 150-250 MPa).5Furthermore, PEEK exhibits high stress corrosion resistance in most environments as compared to polysulfone comp o s i t e ~However, .~ with the exception of one study by Williams et a1.,6 no published reports have assessed the biocompatibility of PEEK. This report is the first in a series of studies addressing the in vitro and in vivo biocompatibility of PEEK. In this study the in vitro biocompatibility of PEEK was measured using cultures of Souse fibroblast cells. MATERIALS AND METHODS
Both extract and direct contact cell culture tests were used to measure the cellular response to the composites and their extracts. The materials were evaluated for 24 and 96 h in direct contact as per ASTM F813,8 using cultures of mouse fibroblast L929’s cells (ATCC #CCL l). In the extract tests, extract liquids from the composite materials were added to cell cultures to determine if any cytotoxic leachables were released from the composites. Materials The materials evaluated are listed in Table I and were divided into two experimental groups based on material availability. The composites in group 1 were 30% short PAN fiber reinforced polysulfone, PS (GC-1006, Wilson Fiberfill), and PEEK (J 1105/CF/30, LNP Corporation). A polysulfone carbon fiber composite was included since this material has been previously evaluated as a bi~material.~,” Unfortunately, no information was available on any additives or mold release agents that might have been incorporated in these composites. The second group of experiments were conducted with 30% short fiber reinforced PEEK (J 1105/CF/30, LNP Corporation) that TABLE I Materials for Direct Contact and Extract Tests
Direct Contact
Group 1
PS”
Group 2
PEEK‘ Latexd
PEEK^
96 h
Extract
PS PEEK PEEK Latex
PS PEEK PEEK Latex
”30% PAN carbon fiber reinforced polysulfone, Wilson-Fiberfill, 1985. b30% PAN carbon fiber reinforced polyetheretherketone, LNP, 1985. ‘30% PAN carbon fiber reinforced polyetheretherketone, LNP, 1988. dLatex urethral catheter, Bard, 1988.
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
209
was known not to contain any additives. Sections of a Bard urethral latex catheter, abbreviated LATEX, were included in the second group of experiments as a positive material control. All materials were tested in triplicate. Extract tests The extract tests were conducted following ASTM standards F619 and F813.8r"An extraction vehicle of minimum essential medium with 10% horse serum and sodium bicarbonate was chosen so that the extract liquids could be added directly to cell cultures. The surface areas of the PEEK and PS materials, 12 cm', and the LATEX, 6 cm', although smaller than what is recommended by the standard, were chosen to better approximate the surface area to volume ratios achieved in the direct contact tests. The materials were steam (PEEK and PS) or gas (latex) sterilized, each aseptically added to 20 mL of the extract liquid, and extracted for 120 h at 37°C. The extract liquids were stored at room temperature and then warmed to 37°C and added to 35 x 10 mm culture dishes containing monolayers of mouse fibroblast cells. Four cultures were given extract liquid from a control tube that did not contain any material. The cultures were incubated at 37°C in 5% CO, for 96 h then qualitatively graded on the basis of cellular appearance using a scale of 1-5 as indicated in Table II.12 Direct contact test and LDH assay The direct contact tests were conducted following ASTM F813. The steam sterilized PEEK and PS materials and the gas sterilized LATEX were carefully placed directly onto monolayers of fibroblast cells in 35 x 10 mm culture dishes, The surface area of the PEEK and PS materials in contact with the cells was 100 mm'. The specimen geometry of the latex, a cylindrical tube, and the culture dish diameter, restricted the latex surface area in contact with the cells to 15 mm2. The separate experiments were conducted with the materials in culture for 24 and 96 h at 37°C. Four control cultures without materials were included in each experiment. After incubation the two direct contact experiments were analyzed using qualitative and quantitative techniques. TABLE I1 Scale for Qualitative Cell Indexing 0 1 2 3
4 5
No observable lysis < 20% of cell culture lysed < 40% of cell culture lysed < 60% of cell culture lysed < 80% of cell culture lysed > 80% lysis in cell culture
210
WENZ ET AL.
A quantitative method was also employed to measure the cellular response to the extract liquids and the media from the direct contact tests. Lactate dehydrogenase, LDH, is an enzyme that is released from the cell when the cell membrane is ruptured. Therefore, the amount of LDH present in the media at the completion of the test correlates to the amount of cell death in the culture. LDH has been previously used as a quantitative measure of cytotoxicity by Rae and is easily measured using Sigma Chemical kit #3#-W.'3 The four control cultures included in the experiments were used as controls for LDH analysis. After the 96-h incubation, two of the four control cultures were given 100 p L of Triton X100 and held at 37°C for 2 h. These cultures were labeled Ct and served as positive controls for LDH testing since triton causes complete cell death. The other two control cultures were labeled C- and served as negative controls for LDH analysis. The amount of LDH in each culture is measured in Units, U, and is corrected for background activity by subtracting the average activity of the negative controls, Uc-. The LDH activity of the culture is then expressed as a percentage of the total possible activity in the cultures as given in the formula. % LDH Activity = 100 X (USpec - Uc-)/(Uc+ - Uc-)
Therefore, by definition, the positive control cultures, Ct, have 100% LDH activity and the negative control cultures, C-, have 0% LDH activity. Expressing the data in this manner allows for comparison between the different experiments. RESULTS
All the cell cultures were very healthy after the 96-h exposure to the extract liquids. The cultures with extract liquids from the PEEK and latex materials did not differ in appearance from the two negative control cultures, C-'s. All the C-, PEEK, PS, and latex cultures in the extract experiment were graded as 0 for no observable cell lysis. A monolayer representative of these cultures with attached and rounded fibroblasts is shown in Figure 1. The round, unattached fibroblast cells were evidence that the culture had matured beyond a monolayer consequently leaving no more sites for cell attachment. Complete cell lysis was observed in the positive control cultures exposed to Triton x 100. These C' cultures were graded as 5 for greater than 80% cell lysis. The combined average results in % LDH activity for the PS, PEEK, and LATEX extracts are presented in Table 111. The LDH values are expressed as % LDH activity divided by the surface area of the materials to correct for surface area differences. The % LDH activity for the cultures exposed to PEEK, PS, and LATEX extracts was at or slightly above the background LDH activity in the negative controls. No differences between the materials were statistically significant using the two sample T-test ( p < 0.05).
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
211
Figure 1. Monolayer after testing, 50X, from the C- culture that represents the appearance of all the cultures exposed to extract liquids.
TABLE 111 Average % LDH Activity per cm2 Surface Area for the Extract Test Materials. Standard Deviations Are Given in the Parenthesis ~
~~~
Material PEEK PS Latex
%LDH/cm*
0.23 (0.16)
4.04 (3.26) 0.02 (0.02)
In the direct contact tests, the cells surrounding and underneath the materials were examined and compared to the healthy monolayers in the negative control cultures. In all the PEEK and PS cultures, healthy cell monolayers with several unattached fibroblasts were observed adjacent to and surrounding the materials. Figure 2 illustrates that the cells were growing directly adjacent to a PEEK composite after 24 h in culture. The same cell-material contact was seen in the PS and PEEK cultures after 96 h. Furthermore, a healthy cell monolayer existed directly underneath the PEEK and PS composites as shown in Figure 3. Some elongated fibroblasts were observed at the composite edges as shown in Figure 4. No qualitative differences were observed between the 24- and 96-h cultures for either PEEK or PS. However, the cell cultures exposed to latex for 24 and 96 h were very unhealthy. After 24 h, cell lysis dominated in the immediate vicinity of the LATEX
212
WENZ ET AL.
Figure 2. Corner of PEEK composite, 50X, in culture after Direct Contact 24-h testing. Healthy cells are in direct contact with the material edge.
Figure 3. Monolayer under PEEK composite, 50 X, after Direct Contact 24-h testing.
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
213
Figure 4. Cells at the PEEK composite edge, 50x, after Direct Contact 24-h testing.
material. However, in areas farther away from the material, a healthy cell monolayer was present. After a 96-h exposure to the latex, almost complete cell lysis had occurred. The cultures at 24 and 96 h were qualitatively graded as 4 and 5, respectively. The LDH results for the direct contact tests also correlated well with the qualitative observations. The direct contact results presented in Table IV are expressed as the % LDH activity divided by the material area in contact with the culture. The data is corrected for material contact area because the cylindrical shape of the LATEX catheter significantly reduced the material contact with the cells. The PEEK and PS cultures exhibited negligible cell death at both 24 and 96 h. However, the LATEX cultures had statistically significant ( p < 0.05) higher cell death, two samples T-test, when compared to the composites. DISCUSSION
From the results of the direct contact and extract tests it is apparent that the carbon fiber reinforced PEEK and PS composites exhibited excellent in vitro biocompatibility with cultures of mouse fibroblast cells. The qualitative cell indexing corresponded with the quantitative LDH results presented. The cellular compatibility of PS that was observed supports the in vivo work
WENZ ET AL.
214
TABLE IV Average % LDH Activity per cm2 Surface Area in the 24- and 96-Hour Direct Contact Tests. Standard Deviations Are Given in the Parenthesis. Statistically Significant Values Are Indicated by an Asterisk 96 Hours
24 Hours
Material
PEEK PS Latex
%L D H / ~
0.00 (0.00) 1.56 (1.87) 144.02* (22.83)
% LDH/cm2
6.04 (3.03) 6.46 (4.29) 385.37 (86.42)
conducted by S p e ~ t o rSince . ~ the chemical formulas of PEEK and PS are very similar as shown in Figure 5, it is not surprising that PEEK exhibited good in vitro biocompatibility results. The cytotoxic response to the LATEX was expected since it is considered a positive material control for the direct contact ASTM standard. The lack of a cytotoxic response to any of the extract liquids tested indicates that no detectable cytotoxic leachables were extracted from the materials during the 120-h 37°C condition. These cell culture tests have served to screen PEEK and PS composites for potential toxicity in vivo. Because of the simplified and stagnant conditions in cell culture experiments, the in vitro results cannot completely predict in vivo results. However, these promising results do indicate a strong potential for PEEK biocompatibility and in vim studies are warranted. CONCLUSIONS
In vitro cell culture studies using mouse fibroblast cells in direct contact and extract experiments demonstrated that PEEK and PS exhibit excellent in vitro cellular biocompatibility. Further work is being conducted to assess
CH3 (b)
Figure 5. Clemical formulas for a) polyetheretherketone and b) polysulfone.
BIOCOMPATIBILITY OF PEEK AND PS COMPOSITES
215
the in vim biocompatibility of PEEK and the biomechanical compatibility of the composite as an osteosynthesis plate. This work was supported by Acromed Corporation, Cleveland, Ohio.
References 1. J. S. Bradley, G. W. Hastings, and C. Johnson-Nurse, ”Carbon fibre reinforced epoxy as a high strength, low modulus material for internal fixation plates,” Biomaterials, 1, 38-40 (1980). 2. L. Claes, W. Huttner, and R. Weiss, “Mechanical properties of carbon fibre reinforced polysulfone plates for internal fracture fixation,“ in Biological and Biomechanical Performance of Biomaterials, P. Christel, A. Meunier, and A. J. C. Lee (Eds.), Elsevier Science, Amsterdam, 1986, 81-86. 3. N. Gillett, S. A. Brown, J. H. Dumbleton, and R. P. Pool, ”The use of short carbon fibre reinforced thermoplastic plates for fracture fixation, ” Biomaterials, 6, 113-121 (1985). 4. S. A. Brown, ”Biomechanical compatibility,” in Biocompatibilify of Orthopedic Implants, Vol. 1 , D. F. Williams (Ed.), CRC Press, Boca Raton, 1982, pp. 75-110. 5. R. S. Hastings, S. A. Brown, and A. Moet, “Characterization of short fiber reinforced polymers for fracture fixation devices,” presented at the Thirteenth Annual Meeting of the Society for Biomaterials, New York, June 3-7, 1987. 6. D. F. Williams, A. McNamara, and R. M. Turner, ”Potential of polyetheretherketone (PEEK) and carbon fibre-reinforced PEEK in medical applications,” J. Mater. Sci. Lett., 6(2), 188-190 (1987). 7. F. N. Cogswell and M. Hopprich, “Environmental resistance of carbon fibre-reinforced polyetheretherketone,” Composites, 14(3), 1983, 251-253. 8. ASTM F813 ”Standard practice for direct contact cell culture evaluation of materials for medical devices” in Annual Book of A S T M Standards 13.01. ASTM, Philadelphia, 1987. 9. C. A. Behling and M. Spector, ”Quantitative characterization of cells at the interface of long-term implants of selected polymers,” J. Biomed. Mater. Res., 20, 1986, 653-666. 10. M. Spector, “A high modulus polymer for porous orthopedic implants: biomechanical compatibility of porous implants,” I. Biomed. Mafer. Res., 12, 1978, 665-677. 11. ASTM F619 ”Standard practice for extraction of medical plastics” in Annual Book of A S T M Standards 13.01. ASTM, Philadelphia, 1987. 12. ASTM F895 ”Standard test method for agar diffusion cell culture Screening,” in Annual Book of A S T M Standards 13.01. ASTM, Philadelphia, 1987. 13. T.Rae, ”Tissue culture techniques in biocompatibility testing,” in CRC Techniques of Biocompafibility Testing, Vol. 2 , D. F. Williams (Ed.), CRC Press, Boca Raton, 1986, pp. 81-93. Received January 23, 1989 Accepted August 22, 1989
