TISSUE ENGINEERINGVolume 2, Number 3, 1996Mary Ann Liebert, Inc.

Salient Degradation Features of a 50:50 PLA/PGAScaffold for Tissue Engineering

A.R. SINGHAL, M.D., CM. AGRAWAL, Ph.D., P.E., andK.A. ATHANASIOU, Ph.D., P.E.

ABSTRACT

An implant system that undergoes a gradual, time-dependent, nontoxic degradation processmay offer an efficacious, safe, and desirable alternative to metallic materials used in thetreatment of various musculoskeletal conditions. Such a scaffold may also be a suitable ve-hicle for growing cells and tissue in the laboratory for tissue engineering applications. Wehave used a scaffold of this type previously in animal studies for biological resurfacing oflarge articular cartilage defects.1 This study examined important in vitro degradation char-acteristics of a 50:50 polylactic acid/polyglycolic acid (PLG) implant during an 8-week pe-riod. It was determined that this particular implant degraded in a biphasic fashion. The ini-tial phase occurred during the first 2 weeks with a decrease in molecular weight and surfaceaxial strain, coupled with an increase in percent porosity. The second phase demonstrateda decline in surface axial strain by 4 weeks and an ongoing decline in molecular weight. Lossof gross structural properties was not evident until the start of the second phase and wascomplete at 8 weeks. This study demonstrated the potential uses for this implant as a meansof providing structural support for cells and tissue ingrowth for up to 8 weeks. Further stud-ies need to be conducted in order to determine the biological effects of the degrading poly-mer byproducts on host tissues.

INTRODUCTION

KULKARNI ET AL.2 generated the initial interest in the use of polylactic acid (PLA) and polyglycolic acid(PGA) polymers in the field of medicine. Subsequently, biodegradable sutures made from PLA were

used to repair mandibular fractures in dogs.3 Since that time there has been an enormous amount of furtherresearch in this field. PLA and PGA are attractive for many reasons. They tend to "degrade" slowly, makingthem an attractive means for use in internal fixation. "Degradation" in this sense is used to denote mass lossdue to resorption or dissolution of the biomaterial, precipitated or accompanied by molecular weight reduc-tion, structural changes, and reduction in strength and stiffness. These materials provide a gradual transfer ofloads to the healing tissues, greatly reducing the osteopenia commonly seen in hard tissues secondary to stressshielding by metallic implants. This slow transfer occurs because as the polymer degrades, its mechanicalproperties deteriorate. Thus, it can support a decreasing level of load resulting in a gradual increased loadingof the healing tissue. Perhaps the greatest advantage of using biodegradable implants is that they bioabsorb insitu, making it unnecessary to perform a repeat surgical procedure for their removal.

Department of Orthopedics, The University of Texas Health Science Center at San Antonio, San Antonio, Texas78284-7774.

197

SINGHAL ET AL.

PLA and PGA are bioabsorbable polyesters belonging to the poly a-hydroxy acids group.4 These polymersand their copolymers (hereafter referred to as PLG) degrade by nonspecific hydrolytic scission of their esterbonds.5 The hydrolysis of PLA yields lactic acid, which is a normal byproduct of anaerobic metabolism inhumans and is incorporated in the tricarboxylic acid (TCA) cycle to be finally excreted by the body as car-bon dioxide and water. Polyglycolic acid biodegrades by a combination of hydrolytic scission and enzymatic(esterase) action producing glycolic acid, which can either enter the TCA cycle or is excreted in urine.

There has been a considerable amount of research performed in the field of polymer degradation.Polyglycolic acid implants in an in vivo study commenced hydrolysis within 2 days and their tensile strengthwas gradually reduced to zero within 28-32 days.7 However, 100% PLA rods implanted in the medullarycavity of rabbits demonstrated a lengthy decline in molecular weight (MW) over a period of nearly 18months.8 An in vivo study comparing the degradation rates of various copolymers found that a 50:50 PLGcopolymer with starting MW of 46 kDa degraded the fastest with a half-life of 1 week.9 A recent in vitrostudy demonstrated that 50:50 PLG with an initial MW of 60 kDa completely degraded in a gradual fash-ion over a period of 10 weeks.10 These studies suggest that there is a wide temporal range during whichdegradation occurs depending on the identity of the polymers involved.

It is well known that the characteristics and kinetics of PLG degradation are a function of monomer ra-tio, molecular linearity, initial MW, and percent crystallinity. Higher MW polymers take longer to degradebecause the size of their molecular chains has to be reduced below a threshold level for the chains to ac-quire enough mobility to diffuse out of the system.10 Furthermore, since PLG degrades by hydrolytic scis-sion, more porous implants degrade faster as they offer a higher surface area for contact with water mole-cules.10 It has also been shown that the rate of decrease in MW is directly proportional to the glycolic acidcontent of the polymeric implant.7

Crystallinity plays a significant role in the rate of polymer degradation. Crystallinity is determined bythe arrangement of the polymers' molecular chains and is affected by molecular chemistry, temperature,and rate of cooling during solidification from a melt.4 Monomer units containing large or complex chemi-cal species render crystalline order difficult to obtain.4 On the other hand, elevated temperatures and a slowrate of cooling enable the chains to be mobile and to reposition themselves in a more ordered crystallinepattern.4 Therefore, the crystallinity of PLG copolymers can be greatly altered as a result of changes in thefabrication process. PGA and L-PLA polymers are best described as being semicrystalline, since they con-tain both crystalline and amorphous regions.1112 The ratio between crystalline and amorphous contents isan important determinant of degradation characteristics because crystallinity is inversely related to the rateof polymer breakdown.13 Crystalline regions provide less access to water molecules than the amorphousregions and, as a consequence, are more resistant to hydrolysis.

As outlined above, the degradation of implants fabricated from PLG polymers occurs by a basic chem-ical phenomenon, but is rendered complex in nature due to the influence of a variety of interacting factors.These parameters depend not only on the chemical makeup of the polymer but also on the fabrication processto a significant degree. PLG implants can be fabricated using a variety of techniques; a comprehensive re-view of these has been provided by Agrawal et al.14 The choice of a fabrication methodology can deter-mine the molecular weight distribution, crystallinity, porosity, permeability, and mechanical properties ofthe implant. Each of these parameters, in turn, can influence the degradation characteristics of the implant.

The authors of the present study have previously described two-phase PLG implants, which could po-tentially be used for osteochondral repair.115 The fabrication process for each of the two phases in theseimplants is different, leading to differences in their physical, mechanical, and morphological properties.Thus the objective of this study was to examine the degradation characteristics of the two phases and thecomposite two-phase implant. Changes in molecular weight, surface axial strain, and percent porosity weredetermined to better elucidate these objectives.

MATERIALS AND METHODS

The overall design objective of our ongoing research is to produce biodegradable implants that can beused to enhance tissue regeneration and repair processes in the treatment of osteochondral defects in di-

198

50:50 PLA/PGA SCAFFOLD FOR TISSUE ENGINEERING

arthrodial joints. To this end, the production methodology involves steps for making a two-phase implantwith mechanical and physical characteristics that mimic those of cartilage and underlying bone at the site.For this study, a total of 180 implants were manufactured and divided equally into three groups: "bone"phase, "cartilage" phase, and two-phase implants. The initial dimensions of the implants were "bone" phase:5 mm height X 7 mm diameter; "cartilage" phase: 2 mm height X 7 mm diameter; and two-phase: 7 mmheight X 7 mm diameter. These implants were fabricated using 50:50 poly-D,L-lactide-co-glycolide(Birmingham Polymers, Birmingham, AL) with an inherent viscosity of 0.71 dl/g, polydisperity of 1.75,and initial weight average molecular weight of 54.5 kDa. To make the "bone" phase section of the 60 two-phase implants, 4.5 g of copolymer was solubilized in 21.5 ml of acetone. This mixture was then mechan-ically stirred for 20 min to ensure that the polymer was completely dissolved in the acetone and then pre-cipitated in 13.5 ml of 200 proof ethanol. The gummy composite was meticulously separated from the liquidphase and then manually kneaded to allow the escape of excess alcohol and acetone. It was then placed ina desiccator at room temperature for 5 min under 25 mTorr vacuum. The polymer was next flattened to anapproximately 0.5-mm-thick film by rolling it with a 0.5-in.-diameter Teflon® (DuPont Chemical Company,Wilmington, DE) rod. The polymer was again placed in desiccator under 25 mTorr vacuum for 3 min. Thefilm was then divided into 60 separate pieces of equal weight. These pieces were packed in a Teflon moldand resubjected to 25 mTorr vacuum at room temperature for approximately 22 min. Care was taken notto handle the polymer except to push any extruded amounts back into the mold. The mold was then placedin a lyophilizer under 37°C and 25 mTorr vacuum was applied for 2 h. It was then placed at room tem-perature and subjected to 25 mTorr pressure in a lyophilizer for 20 h. The implants were removed from themold and placed in a vacuum oven at 47°C for 24 h. Thus, the 5 mm height X 7 mm diameter "bone" phaseof the two-phase implants was produced. Subsequently, the "cartilage" phase was produced in a similarfashion. Following the flattening step with the Teflon® rod, the polymer was divided into masses of equalweight and placed into the Teflon® mold, forming the "cartilage" phase implants. Thus, the "cartilage"phase, unlike the "bone" phase, was not subjected to extensive vacuum or high temperature treatments. Thetwo-phase implants were constructed by placing the "cartilage" phase on top of the "bone" phase. Althoughno additional adhesives, solvents, or other bonding elements were used in this process, the two phases stucktogether due to the presence of organic solvents in the "cartilage" phase. The implants were then placed ina lyophilizer for 24 h and finally stored in a desiccator at room temperature until time of testing.

Sixty implants in each of the three groups ("bone," "cartilage," and two-phase) were studied for theirdegradation characteristics. Degradation was studied at t = 0, 2, 4, 6, and 8 weeks such that each time pe-riod contained 12 implants. The / = 0 week specimens served as the controls. Each implant was placed ina capped centrifuge tube containing 5 ml of phosphate-buffered saline (PBS) at pH 7.4 and 37°C. Every3.5 days, the PBS solution was aspirated using a pipette and discarded. New 5 ml of PBS was then added.After each 2 week interval, 12 implants were removed and placed in a desiccator under 25 mTorr vacuumat room temperature until further testing.

To elucidate the kinetics of polymer degradation, changes in the implants' MW, compressive mechani-cal properties, percent porosity, and morphological structure were examined. Gel permeation chromatogra-phy (GPC) was used to determine temporal changes in weight average molecular weight. Specimens rep-resenting a cross section of the implant, approximately 3 mg in weight, were dissolved in 1 ml of chloroform.This solution was then filtered through a 0.45-^m filter and injected into the GPC system that used chlo-roform at 32°C as the mobile phase and polystyrene standards. Molecular weight was monitored up to t =4 weeks.

The implants' mechanical behavior was determined using creep indentation measurements on an auto-mated creep indentation machine.16 The implant was attached to the specimen holder using cyanoacrylate.A fiberoptic laser positioning system was used to obtain perpendicular alignment of the test site. The testchamber was then filled with PBS and a tare load of 0.02 N was applied for 15 min. A step load of 0.083N was then applied through a 0.5-mm-diameter, porous, rigid, indenter tip to obtain 1 h of creep. The sur-face axial strain at 1 h, <?zz

s (z = 0, r = 0, t = 3600 sec) was calculated by normalizing the equilibriumcreep value with each implant's thickness.

Scanning electron microscope (SEM) photographs at magnifications of 80-650X were taken of both theexterior (uncut) and interior (cut) surfaces of the "bone" phase, "cartilage" phase, and two-phase implants.

199

SINGHAL ET AL.

The cut surface of the implants was obtained by first placing the specimens in 5 ml of liquid nitrogen for30 sec and then using a guillotine-like device to obtain a smooth, even cut through the implant interior. TheSEM micrographs were then scanned into a computer. The boundaries of the pores were outlined using animage analysis software (NIH Image) and the total area encompassed by these pores was determined. Thisarea was then divided by the total area of the SEM micrograph to calculate the percent porosity. For thetwo-phase implants, SEM micrographs of an area 0.5 mm above and below the interface line between thetwo phases were used for porosity measurements. For morphological examinations, 1:1 magnification pho-tographs were taken at each time period to determine macroscopic and gross structural changes.

An effort was made to detect and measure the melting temperature (Tm) or any significant crystallinityin the implants using differential scanning calorimetry (DSC). The implant was cut into multiple smallpieces and 3-5 mg of sample was then analyzed on the DSC system.

Surface axial strain, weight average molecular weight, and percent porosity values were compiled asmean ± standard deviation and were statistically examined with analysis of variance (ANOVA), multiplecomparison tests, and Student's t test as required.

RESULTS

The results show significant temporal changes in both single-phase and two-phase implants. Figure 1shows the change in surface axial strain for the "bone" and "cartilage" phase implants over a period of 4weeks. There is a statistically significant decrease of approximately 65% in surface axial strain during thefirst 2 weeks in both groups of implants. This was followed by a significant increase to near initial valuesin the "bone" phase by 4 weeks. During the same period of time, the surface axial strain for the "cartilage"phase first decreased and then increased to approximately 230% of its initial value. Thus, both phases ofthe implant became stiffer by 2 weeks and then reversed the trend and became more compliant by 4 weeks.

0.08-1

0.07-

0.06-

0.05-

0.02-

0.01-

• "BONE"PHASE

0 "CARTILAGE" PHASh

•m••

TIME (WEEKS)

FIG. 1. Changes in surface axial strain for "bone" phase and "cartilage" phase during the initial 4 weeks of testingin = 6 per group). "Bone" phase is significantly stiffer than "cartilage" phase at t = 0 and 2 weeks. Both groups ex-perience an initial period of stiffening (decrease in surface axial strain) followed by gradual increase in compliance.

200

50:50 PLA/PGA SCAFFOLD FOR TISSUE ENGINEERING

S

oEd

70-,

60 -

50-

40

30-

20-

El "BONE" PHASE

H "CARTILAGE" PHASE

*&&

TIME (WEEKS)

FIG. 2. GPC results show a gradual, statistically significant decrease in molecular weight of the "bone" phase and"cartilage" phase implants over the first 4 weeks (n = 6 per group).

A gradual, statistically significant decline in weight average molecular weight of the "bone" and "cartilage"phase implants occurred over a period of 4 weeks (Fig. 2). The molecular weight of "bone" phase implantsdecreased from an initial value of 58 kDa at t = 0 weeks to 8 kDa at t = 4 weeks. Similar results were seenin the "cartilage" phase as the molecular weight decreased from 56 to 16 kDa over this same time period.

DSC measurements on implants at t = 0 weeks indicated a low endothermic peak extending from ap-proximately 90 to 120°C. However, no perceptible peak indicating a Tm was observed up to 4 weeks.

The change in percent porosity over a period of 4 weeks on the uncut outer surface of the implants isshown in Figure 3. This porosity was essentially zero at t — 0 weeks for each of the 3 groups. There wasa marked, statistically significant increase in percent porosity seen in "bone" (26%), "cartilage" (17%), andtwo-phase (32%) groups from t = 0 to t = 2 weeks. The percent porosity continued to increase even fur-ther in all 3 phases from t = 2 to t = 4 weeks.

Figure 4 shows the change in percent porosity on the cut, interior surface of the 3 different groups of im-plants. At t = 0 weeks the initial percent porosity for the "bone" phase was 48.5%, "cartilage" phase 41%,and two-phase 50%. There was a statistically significant increase in percent porosity to 59% over the first2 weeks in the "bone" phase. No significant changes were observed in percent porosity over the initial 2weeks in either the "cartilage" or two-phase implants. In addition, there were significant differences be-tween the "bone" and "cartilage" phase groups at t = 0 and t - 2 weeks. Specifically, at t - 0 the "bone"phase was 18% more porous than the "cartilage" phase. At t = 2 weeks and t = 4 weeks, the "bone" phasewas 40 and 16% more porous, respectively, than the "cartilage" phase.

An SEM micrograph of the exterior surface of a "bone" phase implant at t — 0 weeks is shown in Figure5. There are multiple linear markings visible on the surface but the number of pores is close to zero. Figure6 shows the interior surface of a "bone" phase implant at t = 0 weeks. The pores are considerably increasedin both number and size in the interior of the implant, compared to the exterior. Similar differences be-tween the exterior and interior of the "bone" phase implant were seen at t = 2 weeks (Figs. 7 and 8).

201

SINGHAL ET AL.

au<

O2axw

70 n

60-

40-

53 30 HoIH 20 -\uog IOH

• "BONE"

H "CARTILAGE"

n 2-PHASE

1 |

J L

TIME (WEEKS)

FIG. 3. Percent porosity at the uncut, exterior surface is essentially zero at / = 0 weeks in "bone," "cartilage," and2-phase implant groups (n = 3 per group). A marked, statistically significant increase in percent porosity is seen in thefirst 4 weeks within each of the 3 groups. In addition, "bone" phase is statistically more porous at the surface than the"cartilage" phase during the first 4 weeks.

The gross morphology of the "bone," "cartilage," and two-phase implants did not exhibit any significantdifferences. At t = 0, the implants were smooth, translucent, and had few surface irregularities. At t = 2weeks, the implants were considerably harder to the touch and had changed to a bright, chalkish-white color.At t — 4 weeks, the implant surface had more cracks and cavities but the overall structure was fairly wellmaintained. By / = 6 weeks, the implants had lost considerable structural composition and were extremelysoft and malleable. Implants were almost completely disintegrated by t = 8 weeks.

DISCUSSION

This study focused on the examination of the salient degradation characteristics of a 50:50 PLG copoly-meric implant proposed to be used for the treatment of osteochondral defects. One-phase ("bone" and "car-tilage") and two-phase ("osteochondral") implants were studied until virtually complete degradation, whichoccurred by approximately 8 weeks. In summary, significant differences were found in the degradation char-acteristics of the "bone" phase and "cartilage" phase, suggesting that the fabrication methodology plays apivotal role in determining this implant's functional behavior. Furthermore, it was shown that these 50:50PLG implants degrade in a temporally biphasic manner with significant differences observed between thefirst 2 weeks and the remainder of the degradation process. Additionally, it was interesting to note vast dif-ferences between the surface and interior structural and degradation characteristics of these implants, sug-gesting that the presence of a relatively nonporous "skin" may significantly alter this implant's degradationkinetics and in vivo response.

The initial segment of degradation lasted from t = 0 to 2 weeks and consisted of a decrease in molecu-

202

50:50 PLA/PGA SCAFFOLD FOR TISSUE ENGINEERING

O

3

wu

o.

E3 "BONE"70-, H "CARTILAGE"

• 2-PHASE

60-

50 -

40 -

H 30-

12 20-

10-

mm•1

ii

i•

J L I L

TIME (WEEKS)

F I G . 4. Percent porosity at the cut, interior surface increases in the first 4 weeks of testing in the "bone," "cartilage,"and 2-phase implant groups (n = 3 per group). "Bone" phase is statistically more porous than the "cartilage" phase att = 0 and 2 weeks.

F I G 5. SEM micrograph at 80X of the exterior surface of a "bone" phase implant at t = 0 weeks. Multiple, linearmarkings are present, but the percent porosity is near zero.

203

SINGHAL ET AL.

FIG. 6. SEM micrograph at 80X of the interior of a "bone" phase implant aw = 0 weeks. An increase in the num-ber and size of pores is evident compared to Figure 5.

lar weight, a decrease in surface axial strain, and an increase in percent porosity at both the outer and in-terior surfaces of the implants. A significant decrease in molecular weight of about 30% in "bone" phaseand approximately 50% in "cartilage" phase was seen over the first 2 weeks of testing. In a study by Agrawalet al.17 similar results were obtained as 50:50 PLG implants with an initial molecular weight of 71 kDa hada 34% decrease in molecular weight over 3 weeks. It has been reported that loss of molecular weight startsimmediately upon contact with water accompanied by a simultaneous decrease in implant strength.18

However, the decrease in surface axial strain in this study suggests a paradoxical increase in mechanicalstrength. An earlier study by Athanasiou et al.10 hypothesized that this could be attributed to a possible in-crease in crystallinity of the polymer by breakdown of the chains in the amorphous regions. As the chainsget smaller in size they increase in mobility and have a higher ability to crystallize.

FIG. 7. SEM micrograph at 230 X of the exterior surface of a "bone" phase implant at t = 2 weeks shows a smallnumber of pores.

204

50:50 PLA/PGA SCAFFOLD FOR TISSUE ENGINEERING

FIG. 8. SEM micrograph at 230X of the interior of a "bone" phase implant at f = 2 weeks shows an increased num-ber of well-defined pores compared to Figure 7.

The DSC measurements in this study failed to detect a melting point during the first 4 weeks of testing.Coombes and Heckman19 obtained similar DSC results for 50:50 PLG implants. They detected an increasein crystallinity only after 4 months of the degradation process under similar testing circumstances. This ab-sence of a true melting point suggests that the polymer was mostly amorphous in nature and perhaps a rea-son other than, or in conjunction with, an increase in crystallinity was probably responsible for the decreasein surface axial strain. The increase in stiffness (both gross and mechanical) could be the result of a plas-ticization phenomenon from solvent trapping by the polymer. The percent porosity results for t — 0 weekssupport this hypothesis because there was high porosity within the implant interior (48.5% for "bone" phase,41% for "cartilage" phase) but minimal porosity (3.3% for "bone" phase, 0% for "cartilage" phase) at theimplant surface. Thus, solvent could have easily been retained within the implant interior during the fabri-cation process. In addition, the presence of short chain oligomers could also have resulted in a plasticiza-tion process. Further studies need to be conducted to better elucidate the intricacies of this finding.

The degradation process from t = 2 to 4 weeks involved a continuous decrease in molecular weight cou-pled with an increase in surface axial strain. The molecular weight decreased by 87 and 67% of initial val-ues by the end of 4 weeks for "bone" phase and "cartilage" phase, respectively. Percent porosity for bothphases increased on the exterior surface, but remained approximately constant in the interior. In addition,it is in this time period that the implants began to lose their structural properties. This process continuedfor the remainder of the testing duration and the implants were nearly completely disintegrated by 8 weeks.While we did not document changes in material weight in this study, our previous studies suggest that sig-nificant mass loss of a similar implant does not occur until after 3^4 weeks of testing.20 This decline in ma-terial weight corresponds to the initial loss of structural properties as seen in this study at around 4 weeks.

It has been reported that degradation of PLG implants initially occurs faster at the center than at the sur-face of the implant due to entrapment of degradation products.2021 The results obtained in this study showedthe presence of a "skin" on the implants' surface, which may contribute to this phenomenon. SEM micro-graphs at t = 0 weeks showed almost no porosity at the surface of the implants, while the interior was filledwith multiple pores allowing rapid hydrolysis to occur at the center. These initial physical characteristicsof the implant (low percent porosity at the surface) and high molecular weight of the chains (responsiblefor steric effects) resulted in a "skin" phenomenon, preventing degradation products generated at the cen-ter of the implant from readily exuding. The exterior surface of the implant may have also provided a bar-rier between the vastly different internal and external degradation milieus. However, continued degradationresulted in increased percent porosity and decreased molecular weight and a gradual loss of this "skin" bar-

205

SINGHAL ET AL.

rier. Once a threshold value was reached there was rapid, generalized decrease in both material and struc-tural properties of the implant.

Considerable differences were present in the degradation characteristics of "bone" phase vs. "cartilage"phase implants, probably as a result of differences in the fabrication processes. The "bone" phase was sig-nificantly stiffer compared to the "cartilage" phase at t = 0 and 2 weeks. However at t = 4 weeks, bothphases exhibited similar stiffness. Although the molecular weights of the two phases were similar initially,significant differences were observed in the temporal decreases of molecular weight at t = 2 and 4 weeks.For example, at t = 2 weeks, the molecular weight of "bone" phase was 30% greater than that of the "car-tilage" phase. In contrast at / = 4 weeks, the molecular weight of "bone" phase was 50% less than that ofthe "cartilage" phase. The percent porosity of the "bone" phase was significantly greater than that of the"cartilage" phase at all time groups and is probably secondary to differences in the fabrication process. Thecomplex interplay between changes in stiffness, molecular weight, and percent porosity resulted in the var-ious degradation outcomes and patterns seen in this study.

Our goal in this study is to design an osteochondral implant with the "bone" phase similar to underlyingcancellous bone and "cartilage" phase similar to articular cartilage for use in the repair of osteochondraldefects. The results obtained suggest that as in cancellous bone and articular cartilage, the "bone" phasewas stiffer and more porous than the "cartilage" phase. Additional work needs to be performed to make the"bone" phase even stiffer and to match the permeability values of the two phases to those of cartilage andbone. In conclusion, this particular implant may offer a new method for the treatment of various muscu-loskeletal conditions, such as in biological resurfacing of large articular cartilage defects.

ACKNOWLEDGMENTS

Image analysis was performed using the public domain NIH Image program (developed at the U.S.National Institutes of Health and available from the Internet by anonymous FTP from zippy.nimh.nih.govor on floppy disk from the National Technical Information Service, Springfield, Virginia). The authorswould like to thank C. Zhu, D. Lanctot, M. Landry, and M. Singhal for their assistance on this project.

REFERENCES

1. Athanasiou, K.A., Schmitz, J.P., Schenck, R.C., Clem, M., Aufdemorte, T., and Boyan, B.D. The use of biodegrad-able implants for repairing large articular cartilage defects in the rabbit. Transact. Orthop. Res. Soc. 17(1), 172,1992.

2. Kulkarni, R.K., Pani, K.C., Neuman, C , and Leonard, F. Polylactic acid for surgical implants. Arch. Surg. 93, 839,1966.

3. Cutright, D.E., Hunsuck, E.E., and Beasley, J.D. Fracture reduction using a biodegradable material, polylactic acid.J. Oral Surg. 29, 393, 1971.

4. Agrawal, C, Niederauer, G., Micallef, D., and Athanasiou, K. The use of PLA-PGA polymers in orthopaedics. InEncyclopedic Handbook of Biomaterials and Bioengineering. New York: Marcel Dekker, 1995, pp. 2081-2115.

5. Hollinger, J.O., and Schmitz, J.P. Restoration of bone discontinuities in dogs using a biodegradable implant. J. OralMaxillofac. Surg. 45, 594, 1987.

6. Williams, D.F., and Mort, E. Enzyme-accelerated hydrolysis of polyglycolic acid. J. Bioeng. 1, 231, 1977.7. Reed, A.M., and Gilding, D.K. PLA/PGA copolymers. In vivo and in vitro degradation. Transact. Soc, Biomat. 2,

15, 1978.8. Nakamura, T., Hitomi, S., Waranabe, S., Shimizu, Y., Jamshidi, K., Hyon, S.H., and Ikada, Y. Bioabsorption of

polylactides with different molecular properties. J. Biomed. Mat. Res. 23, 1115, 1989.9. Miller, R.A., Brady, J.M., and Cutright, D.E. Degradation rates of oral resorbable implants (polylactates and poly-

glycolates): Rate modification with changes in PLA/PGA copolymer ratios. J. Biomed. Mat. Res. 11, 711, 1977.10. Athanasiou, K.A., Singhal, A.R., Agrawal, CM., and Boyan, B.D. In vitro biodegradation and release character-

istics of biodegradable implants containing trypsin inhibitor. Clin. Orthop. Rel. Res. 315, 272, 1995.11. Gilding, D.K., and Reed, A.M. Biodegradable polymers for use in surgery-polyglycolic/poly (lactic acid) homo-

and copolymers: 1. Polymer 20, 1459, 1979.

206

50:50 PLA/PGA SCAFFOLD FOR TISSUE ENGINEERING

12. Jamshidi, K., Hyen, S.H., Nakamura, T., Ikada, Y., Shimizu, Y., and Teramatsu, T. In Christel, P., Meunier, A.,and Leein, A.J.C., eds. Biological and Biomechanical Performance of Biomaterials. Amsterdam: Elsevier SciencePublishers, 1986, pp. 227-232.

13. Chu, C.C., and Browning, A. The study of thermal and gross morphologic properties of polyglycolic acid upon an-nealing and degradation treatments. J. Biomed. Mat. Res. 22, 699, 1988.

14. Agrawal, CM., Niederauer, G.G., and Athanasiou, K.A. Fabrication and characterization of PLA-PGA orthopaedicimplants. Tissue Eng. 1(3), 241, 1995.

15. Athanasiou, K.A., Schenck, R.C., Constantinides, G., Sylvia, V., Aufdemorte, T., and Boyan, B.D. Biodegradablecarriers of TGF-b in rabbit osteochondral defects. Transact. Orthop. Res. Soc. 18(1), 288, 1993.

16. Athanasiou, K.A., Agrawal, A., and Dzida, F.J. Comparative study of intrinsic mechanical properties of human ac-etabular and femoral head cartilage. J. Orthop. Res. 12(3), 340, 1994.

17. Agrawal, CM., Kennedy, M.E., and Micallef, D.M. The effects of ultrasound irradiation on a biodegradable 50-50%copolymer of polylactic and polyglycolic acids. J. Biomed. Mat. Res. 28, 851, 1994.

18. Dauner, M., Mueller, E., Wagner, B., and Plank, H. In vitro degradation of polylactides depending on differentprocesses. In Plank, H., Dauner, M., and Renardy, M., eds. Degradation Phenomena on Polymeric Biomaterials.New York: Springer, 1992, pp. 107-121.

19. Coombes, A.G.A., and Heckman, J.D. Gel casting of resorbable polymers 1. Processing and applications.Biomaterials 13(4), 217, 1992.

20. Thompson, D.E., Agrawal, CM., and Athanasiou, K.A. The effects of dynamic compressive loading on biodegrad-able implants of 50-50% polylactic acid-polyglycolic acid. Tissue Eng. 2(1), 61, 1996.

21. Li, S., Garreau, H., and Vert, M. Structure-property relationships in the case of the degradation of massive poly(x-hydroxy acids) in aqueous media. Part 3: Influence of the morphology of poly (L-lactic acid). J. Mat. Sci. Mat.Med. 1, 198, 1990.

Address reprint requests to:Kyriacos A. Athanasiou, Ph.D., P.E.

Department of OrthopedicsThe University of Texas

Health Science Center at San Antonio7703 Floyd Curl Drive

San Antonio, TX 78284-7774

207

This article has been cited by:

1. Xiaoming Li, Yu Yang, Yubo Fan, Qingling Feng, Fu-zhai Cui, Fumio Watari. 2013. Biocomposites reinforced by fibers or tubesas scaffolds for tissue engineering or regenerative medicine. Journal of Biomedical Materials Research Part A n/a-n/a. [CrossRef]

2. Tissue Engineering of Articular Cartilage 165-248. [CrossRef]3. Orapin V. Horst, Miquella G. Chavez, Andrew H. Jheon, Tejal Desai, Ophir D. Klein. 2012. Stem Cell and Biomaterials Research

in Dental Tissue Engineering and Regeneration. Dental Clinics of North America 56:3, 495-520. [CrossRef]4. Sophie C. Cox, Kajal K. MallickPreparation of Porous Hydroxyapatite Scaffolds using Yeast as a Pore Forming Agent 15-25.

[CrossRef]5. Yi Yi Gong, Ji Xin Xue, Wen Jie Zhang, Guang Dong Zhou, Wei Liu, Yilin Cao. 2011. A sandwich model for engineering cartilage

with acellular cartilage sheets and chondrocytes. Biomaterials 32:9, 2265-2273. [CrossRef]6. Lin Lu, Qingwei Zhang, David Wootton, Peter I. Lelkes, Jack Zhou. 2010. A novel sucrose porogen-based solid freeform

fabrication system for bone scaffold manufacturing. Rapid Prototyping Journal 16:5, 365-376. [CrossRef]7. Yih-Lin Cheng, Meng-Long Lee. 2009. Development of dynamic masking rapid prototyping system for application in tissue

engineering. Rapid Prototyping Journal 15:1, 29-41. [CrossRef]8. David Eglin, Derek Mortisen, Mauro Alini. 2009. Degradation of synthetic polymeric scaffolds for bone and cartilage tissue repairs.

Soft Matter 5:5, 938. [CrossRef]9. SW Dean, CM Agrawal, J Carter, JL Ong. 2006. Basics of Polymeric Scaffolds for Tissue Engineering. Journal of ASTM

International 3:9, 100428. [CrossRef]10. Yang Cao, Tristan I. Croll, Justin G. Lees, Bernard E. Tuch, Justin J. Cooper-White. 2005. Scaffolds, Stem Cells, and Tissue

Engineering: A Potent Combination!. Australian Journal of Chemistry 58:10, 691. [CrossRef]11. Tejas S. Karande, Joo L. Ong, C. Mauli Agrawal. 2004. Diffusion in Musculoskeletal Tissue Engineering Scaffolds: Design

Issues Related to Porosity, Permeability, Architecture, and Nutrient Mixing. Annals of Biomedical Engineering 32:12, 1728-1743.[CrossRef]

12. Pierre Sarazin, Xavier Roy, Basil D. Favis. 2004. Controlled preparation and properties of porous poly(l-lactide) obtained from aco-continuous blend of two biodegradable polymers. Biomaterials 25:28, 5965-5978. [CrossRef]

13. Mona K. Marei, Samir R. Nouh, Mohamed M. Fata, Adham M. Faramawy. 2003. Fabrication of Polymer Root Form Scaffoldsto Be Utilized for Alveolar Bone Regeneration. Tissue Engineering 9:4, 713-731. [Abstract] [Full Text PDF] [Full Text PDFwith Links]

14. M. Endres, D.W. Hutmacher, A.J. Salgado, C. Kaps, J. Ringe, R.L. Reis, M. Sittinger, A. Brandwood, J.T. Schantz. 2003.Osteogenic Induction of Human Bone Marrow-Derived Mesenchymal Progenitor Cells in Novel Synthetic Polymer–HydrogelMatrices. Tissue Engineering 9:4, 689-702. [Abstract] [Full Text PDF] [Full Text PDF with Links]

15. Michael A. Slivka, Neil C. Leatherbury, Kris Kieswetter, Gabriele G. Niederauer. 2001. Porous, Resorbable, Fiber-ReinforcedScaffolds Tailored for Articular Cartilage Repair. Tissue Engineering 7:6, 767-780. [Abstract] [Full Text PDF] [Full Text PDFwith Links]

16. C. Mauli Agrawal, Robert B. Ray. 2001. Biodegradable polymeric scaffolds for musculoskeletal tissue engineering. Journal ofBiomedical Materials Research 55:2, 141-150. [CrossRef]

17. Gabriele G Niederauer, Michael A. Slivka, Neil C Leatherbury, Donna L Korvick, H Hugh Harroff, Jr, William C Ehler, ColinJ Dunn, Kris Kieswetter. 2000. Evaluation of multiphase implants for repair of focal osteochondral defects in goats. Biomaterials21:24, 2561-2574. [CrossRef]

18. C.M Agrawal, J.S McKinney, D Lanctot, K.A Athanasiou. 2000. Effects of fluid flow on the in vitro degradation kinetics ofbiodegradable scaffolds for tissue engineering. Biomaterials 21:23, 2443-2452. [CrossRef]

19. R.G LeBaron, K.A Athanasiou. 2000. Ex vivo synthesis of articular cartilage. Biomaterials 21:24, 2575-2587. [CrossRef]20. C.H. Lohmann, Z. Schwartz, G.G. Niederauer, D.L. Carnes, D.D. Dean, B.D. Boyan. 2000. Pretreatment with platelet derived

growth factor-BB modulates the ability of costochondral resting zone chondrocytes incorporated into PLA/PGA scaffolds to formnew cartilage in vivo. Biomaterials 21:1, 49-61. [CrossRef]

21. Jane S. Tjia, Prabhas V. Moghe. 1999. Analysis of 3-D microstructure of porous poly(lactide-glycolide) matrices using confocalmicroscopy. Journal of Biomedical Materials Research 43:3, 291-299. [CrossRef]

22. Padma Narayan, Margaret A. Wheatley. 1999. Preparation and characterization of hollow microcapsules for use as ultrasoundcontrast agents. Polymer Engineering & Science 39:11, 2242-2255. [CrossRef]

23. Dr. K.A. Athanasiou, J.P. Schmitz, C.M. Agrawal. 1998. The Effects of Porosity on in Vitro Degradation of Polylactic Acid–Polyglycolic Acid Implants Used in Repair of Articular Cartilage. Tissue Engineering 4:1, 53-63. [Abstract] [Full Text PDF] [FullText PDF with Links]

24. Kyriacos Athanasiou, Donna Korvick, Robert Schenck , Jr.. 1997. Biodegradable Implants for the Treatment of OsteochondralDefects in a Goat Model. Tissue Engineering 3:4, 363-373. [Abstract] [Full Text PDF] [Full Text PDF with Links]

25. C.M. Agrawal, D. Huang, J.P. Schmitz, K.A. Athanasiou. 1997. Elevated Temperature Degradation of a 50:50 Copolymer ofPLA-PGA. Tissue Engineering 3:4, 345-352. [Abstract] [Full Text PDF] [Full Text PDF with Links]