CA1274179A - Bioerodible articles useful as implants and prostheses having predictable degradation rates - Google Patents

Abstract

ABSTRACT OF THE INVENTION
A novel series of articles useful as implants and prostheses and methods for their preparation and use are provided which utilize polyanhydride polymeric matrices as a general class of materials. These articles are biocompatible, non-inflammatory and degrade predictably into non-toxic residues after introduction in-vivo. The articles may be formed in any desired dimensions and configuration and may take specific shape as vascular or skin grafts, as biodegradable sutures or as orthopedic appliances such as bone plates and the like.

Description

`11 FIELD OF T~E INVENTION
l The present invention is concerned with biocompatible

2 articles suitable ~or surgical intr~duction in-vivo and is par-

3 ticularly directed to bioerodible articles which degrade

4 predictably as biocompatible and non~toxic-products.

BACKGROUND OF THE INVENTION
S Synthetic polymeric compositions which are bioerodible 6 and biocompatible have become increasingly important and valuable in recent years. One application for such compositions is as 8 surgically implantable biomaterials or prosthetic articlès ~or 9 human and animal subjects in-vivo. Consequently, such biomat-erials axe articles which serve as implants, a tangible item 11 to be inserted into a living site as for growth or formation 12 of an organic union, or prostheses, artificial devices introduced 13 into living tissues to replace a missing part of the body;
14 these are exemplified by articles such as vascular grafts, bio-degradable sutures and pellets, and orthopedic appliances such 16 as bone plates and the l~ke. In order for an implantable or 17 prosthetic article to be truly useful, it should be composed o~
18 a synthetic polymeric composition having specific characteristics l9 and properties: First, the polymeric composition should have a surface that permits and encourages growth of appropriate 21 mammalian cell types; that is, the growing and maintenance of 22 cells on, and i~ appropriate, within, its matrix, after intro-23 duction into a subject _- iVO. For the purposes o~ graft mater-24 ials, one objective is to maintain n o n - t h r o m b og e n i c surfaces similar to that as exists in living tissues. A

~L~74~1L79 1 failure to initiate and maintain such en~othelial cell 2 surface layers leads to the occurrence of thrombotic 3 events such as occlusion of the blood vessel(s) an 4 ultimate failure of the implanted or prosthetic article.
Second~ the synthetic c~mposition should provide 6 sufficient elasticity and tensile strength over a 7 preselected minimal time period which will vary with the 8 specific application. Third, the synthetic composition g should be non-immunogenic, biocompatible, biodegradable in- ivo and yield degradation products which are 11 themselves non-inflammatory, non-toxic, and 12 non-antigenic. Lastly, the ideal material should 13 degrade within predictable periods of time and be 14 suitable for introduction at multiple tissue sites thereby eliminating the need for surgical removal.
16 Despite continuing research efforts, no class of 17 synthetic polymeric biomaterials has yet been developed 18 which provides all these desired attributes. For 19 example, most of the research concerning synthetic grafting materials has utilized compositions such as 21 Dacron [Graham et al., Arch. ~ . 115:929-933 ~1980);
22 Herring et al., Ann. sur~. 190:84-90 ~1979)]. These 23 materials, however, never develop the desired 24 endothelial cell intima necessary to maintain a non-thrombo~enic surace; as a result, thrombosis -26 blood vessel blockage and diminished blood cell supply 27 to organs in the body - often occurs. Similarly, the 28 use of polygalactin mesh has failed to provide the 29 necessary surface erosion characteristics and thus is unpredictable in degradation time [Bowald et al., 31 Surgerv 86:722-729 ~1979)]. Other presently known I _ 3 _ 1~ ` 12743L7~

1 . I biodegradable polymers such as polylactic acid, i polyglycolic acid, polycaproloctones and the various ¦ polyamides also all degrade irregularly and 4 1 unpredictably with a demonstratable loss of permeability and mechanical strength over time rHeller et al., 6 ¦ "Theory and Practice of Control Drug Delivery rom 7 Bioerodible Polymers", in Controlled Release O
8 ¦ Bioactive Material, R.W. Baker Editors, Academic Press, g I New York, 1980, pp. 1-17; Pitt et al., Biomaterials ¦! 2:215-220 (19811; Chu, C.C., J. ~ . PolYm. Sci.
26:1727-1734 (1981)]. Insoar as is presently known, 12 ,¦ therefore, no synthetic polymeric composition offers all 13 !¦ the properties and characteristi~s which would make it 14 ¦I desirable for use as an article for prosthesis or il implantation in-vivo I
SUMMARY OF THE_INVENTION
16 A bioerodible article useful for prosthesis and 17 implantation and methods for its manufacture are 19 .provided which comprises a biocompatible, hydrophobic 19 polyanhydride matrix, prepared in preselected dimensions and configurations, which erodes predictably into 21 non-toxic residues a~ter introduction in-vivo. The 22 method o usiny the article as an implant and prosthesis 23 comprises the step of introducing a specifically 24 configured article into a subject in-vivo at a ¦ predetermined site.

. I DETAILED DESCRIPTION OF T~E DRAWING
26. The present invention may be more completely and 27 easily understood when taken in conjunction with the 28 accompanying drawing, in which:
.

1~'74179 1 Fig. 1 is a graph illustrating the degradation 2 rates for poly [bis (p-carboxyphenoxy) propane 3 anhydride] matrices and its sebacic acid copolymer matrices at 37C;
Fig. 2 is a graph illustrating the degradation rates ~or compression molded poly tbis (p-carboxy-7 phenoxy) propane anhydride] at different pH levels 8 xanging from 7.4 to 10.0;
g Fig. 3 is a set of two photographs illustrating the biocompatibility attributes of the present invention 11 in-vivo as corneal implants in rabbits one week ater 12 introduction;
13 Fig. 4 is a photograph illustrating a magnified 14 view of the rabbit cornea of Fig. 3 in cross section;
Fig. 5 is a set of two photographs visualizing a 16 magnified view of rat skin tissue in cross section 17 ~ollowing a six month period of subcutaneous 18 implantation o~ the present invention: and 19 Fig. 6 is a set of two photographs visualizing a magnified view o ~ixed poly~er-cell complexes cultured 21 in-vitro.

DETAILED DESCRIPTION OF THE PREFE~RED_E BODIMENTS
22 The present invention comprises articles useful as 23 implants or prostheses and methods for their preparation 24 and use. These articles comprise a biocompa~ible, bioerodible, hydrophobic class of synthetic 26 polyanhydride polymeric compositions having the general 27 formula:

l ~

HC~ R--~H

wherein, R is a hydrophobic organic group and n is 2 greater than lo This class of polymeric compositions 3 can be formed in preselected dimensions and specific 4 . configurations. Regardless of the specific application, . these compositions degrade withln predictable periods of ,~ time after introduction in-viYo into non-toxicr non-inflammatory, and non-immunogenic residues.
The preEerred embodiments of the R group within the general formulation given above is exemplified by, but lo is not limited or restrict~d to, the entities given in ; 11 Table I below.
., . "'.
.

: TABLE I

whereln x > 2 and > 16 ~b) , ~

(c)~ -C~

I . wherein 16 > x > 2 l (d) - ~ a ~ 2 ~
l wherëin x > 1 (e) ~~CH~ H2.CH;~ ~

wherein x > 1 and y > 1, (f) ~ ~ ~ ~(0~

~herein R' and R- are organic groups ~ jl I ~,7~

1 The entire class of polyanhydrides can be 2 synthesized using alternative methods of polymerization 3 now known in the art: bulk polymerization [Conix, A., 4 Macro ~y~. 2:9598 (1966)3; solution polymerization [Yoda et al., Bull. Chem. Soc. Japan 32:1120-1129 6 (1959)]; and interfacial polymerization [Matsuda et al., 7 7apanese Patent No. 10,944 (1962)]. Using any of these 8 methods, a variety of different synthetic polymers 9 having a broad range of mechanical, chemical, and 10 ¦ erosion properties are obtained; the differences in 11 I properties and characteristics are controlled by varying 12 ¦ the parameters of reaction temperatures, reactant 13 concentration, types of solvent, and re~ction time.
14 This is true ~r all potential embodiments of R within the general formula stated above as well as the entities 16 listed in Tl and the specific embodiments described in 17 the Examples which follow herein.
18 All of the articles useful as prostheses or 19 implants are synthetic polyanhydride polymeric compositions which share a number of demonstratable 21 qualities in common:
2Z 1. Each of the articles displays ~redictable 23 degradation rates when introduced in-vivo in a subject.
24 ~egardless o the exact composition, size and configuration, these articles erode only at their 26 exterior surfaces without affecting the center of the ~7 matrix in any way. As the surfaces continue to 28 pro~ressively erode, the articles become thinner and 29 smaller and eventually vanish complétely from the tissue.

q4~g 1 Predictable times for degradation is a consequence 2 ' of biomaterial~ which erode by sur~ace (or 3 heterogeneous) erosion rather than by bulk (or 4 homog~neous) erosion. Surface erosion is degradation which occurs only at the exterior surfaces of ~he 6 composition which, in turn, become progressively thinner 7 with the passage of time. Accordingly, by careful 8 selection oE the specific composition and control of the g physical dimensions of the article, the user can preselect predictable times for complete degradation to 11 occurO
12 2. The rate(s) of degradation for all entities 13 , within the class of polyanhydride polymeric compositions 14 as a whole is not only predictable, but may also be controlled by varying the hydrophobicity of the polymer~
16 The mechanism of predictable degradation requires that , 17 the polyanhydride polymers be hydrophobic in nature 18 thereby preventing water from entering into the interior 19 o~ the matrix in any appreciable degree. This may be achieved alternatively either by substituting one 21 monomer (the R group) in the general formula for another 22 or by combining two monomerîc units as a copolymer and 23 then utilizing,the copolymer as the functional R group 24 within the composition. This is exempliied by the use and attributes of the monomer poly (carboxyphenoxy) 26 propane alone and in combination with sebacic acid as a 2~ copolymer. Although each of these compositions contains 2~ individual hydrophobic properties~ each of these 29 polyanhydride polymerlc compositions contains water-labile linkages between its monomer (or copolymer~
31 R units which ater introduction into the subject either iL27~

l react with or become hydrolyzed by the subject's tissues 2 and body fluids.
3 3. The rates of degradation for each individual 4 poly~eric composition within the general class of s polyanhydride polymers are predictable and constant at a 6 single pH level and present different and distinctive 7 rates of degradation with small changes of pH~ This 8 permits the articles to be ;ntroduced into the subject 9 at a variety o~ tissue sites; this is especially lo valuable in that a wide variety of articles and devices ll to meet diferent but specific applications may be 12 composed and configured to meet specific demands, 13 dimensions, and shapes - each of which offers 14 individual, but different, predictable periods for degradation.
16 4. The entire class of polyanhydride polymers are 17 biocompatible and bioerodible. In view of their 18 intended function as an implant or prosthesis to be l9 introduced into a subject in-vivo, it is absolutely required that these compositions be non-inflammatory, 21 non-toxic, and non~immunogenic; that is biocompatible 22 with the subject's tissues and body fluids in all ~3 respec~s.
24 5. Implantable articles and prostheses formed of polyanhydride polymers do not measurably affect or 26 influence living cells or tissues in any degree.
27 Various polyanhydride compositions may be combined with 28 large vessel endothelial cells and/or smooth muscle 29 cel7s without inhibiting or affecting cell growth.
In-vitro studies show that the endothelial or muscle 31 cells maintained a non-overlapping, contact-inhibited l - 10 -:"~`

1 monolayer of living cells throughout the testing period 2 of two weeks. Subsequently made histological studies of 3 these cultured specimens revealed attenuated endothelial 4 monolayers over the exterior sur~aces o~ the polymeric matrices which strongly resemble intimal endothelial 6 monolayers in-vivo.
7 6. The articles comprising the present invention 8 degrade (erode~ into residues or moieties which are 9 themselves biocompatible and non-toxic. As evidenced by ~he Examples which ~ollow, each of the articles, 11 regardless of specific polyanhydride formulation used, 12 may be implanted into ~he cornea of rabbits without 13 causing inflammation even in a minor degree; this is in 14 stark contrast to presently known compositions (such as polylactic acid matrices) which d em on s t r a te 16 at least minor inflammation of such corneal tissues.
17 Moreover, articles comprising polyanhydride compositions 18 are demonst~atably biocompatible as will be de~cribed in 19 a study involving subcutaneous implantation of such articles in rats. Despîte their presence in the living 21 tissues over a period of weeks, no inflammatory cell 2~ infiltration (polymorphonuclear leukocytes, macrophages, 23 and lymphocytes) is s~en in the tissues adjacent to the 24 implant~ Equally important, as the article predictably degrades, the degradation products are demonstrably 26 non~mutagenic, non-cytotoxic, and non-teratogenic.
27 It will be appreciated that these properties and ~8 ¦ characteristics, as well as the mechanical and chemical 29 attributes, identify and distinguish such articles as being singularly suitable as implants or prostheses.
31 The Examples which ~ollow merely serve to illustrate one 32 or more o the above described characteris~ics, which 33 are representative o the entire class as a whole.

1.'~74~79 . ~
1 The articles prepared and tested were forms of poly 2 [bis (p-carboxyphenoxy) propane anhydride] ~hereinafter "PCPP"~ and its copolymers with sebacic acid 4 (hexeinafter "PPCP - SA"~. These poly lbis (p carboxy-S phenoxy) alkane anhydrldes] were synthesized by meit 6 polycondensation following the method of Conix [Macro 7 Synthe. 2:95-98 ~1966)]. Briefly summarized~ the _ .
8 dicarboxylic acid monomers (or copolymers) were 9 converted to the mixed anhydride by total reflux in acetic and anhydride. Caution was taken to avoid 11 excessive reaction, which would yield a highly insoluble 12 prepolymer difficult to purify; 30 minutes was deemed 13 sufficient. The prepolymers isolated were further 14 recrystallized in a 50:50 (v/v) mixed solvent of acetic anhydride and dimethylformamide. A recrystallization 16 period of several weeks was sometimes necessary to 17 obtain a reasonable yield (30~)~ The prepolymers then 18 underwent melt polycondensation In vacuo under nitrogen 19 sweep. The prepolymers were prepared by reacting different monomer ratios with acetic anhydride.
21 Attempts were also made to obtain the copolymers by 22 polycondensing the mixture o~ individually prepared 23 pxepolymers. In terms of controlling the final 24 composition and purity of the product, the latter approach was found to be superior; however, because of the 26 difficulty of isolating the sebaci~ acid anhydride 27 prepolymer, the former method was deemed more conven-28 ient. Regardless of the methodology used, the composi-29 tion of the copolymers were determined by ultraviolet spectrometry and weight analysis after decomposition in 31 1 M NaOH at 70C overnight.

-~2-I

1 The resulting polymers were purified by extraction with anhydrous ether in a Soxhlet E extractor for 3 several hours and then stored in a dessicator over 4 calcium chloride. The purified polymers, obtained as a crystalline solid, were then ground in a Micro Mill 6 Grinder and sieved into a particle si~e ranging from 7 90-150 micrometers (hereinafter "um-). The polymer 8 particles were then pressed into circular disks using a 9 Carver Test Cylinder Outit at 30 ~PSI at 5C above the polymer's glass transition temperatu;^e, TGV for ten 1~ ~ minutes. Those polymeric compositions that had glass 12 1 transition temperatures below 30C were molded at room 13 ¦ temperature. The dimensions of the circular disks were 14 14 millimeters ~hereinafter lmmn) in diameter, O~9-lol mm thick, and weighed between 140 and 160 16 milligrams (hereinater "mg").
17 The degradation characteristics of such synthetic 18 polyanh-ydrides are demonstrated by hydrophobic disks 19 comprising PCPP; PCPP and PCPP-SA in 85:15 ratio;
PCPP-SA in a 85:15 ratio; PCPP-SA in a 45:S5 ratio; and 21 PCPP-SA in a 21:79 ratio. Circular disks comprising 22 each of these compositions were placed in 0.1 M
23 phosphate buffer, pH 7.4 at 37C for a time period of up to 14 weeks. The results are given in Fig. 1 as degradation profiles whose erosion kinetics were 26 followed by measuring the ultravio~et absorbance at 250 27 nanometers (hereinafter "nmn) of periodically changed 2a buffer solutions using a Perkin-Elmer W
29 Spectrophotometer model 553. As is readily seen, the more hydrophobic polymers - PCPP and PCPP-S~ (85:15) -31 displayed constant erosion kinetics over several months.
I.

Il - 13 -~z~

l Moreover, as the sebacic acid content increased in the 2 polymeric composition, the polymers became more 3 hydrophilic and demonstrated increased rates of erosion.
4 The rate of erosion for the composition in fact S increased 800 times when the sebacic acid content of the polymer reached 80 percent~ It was also noted that the 7 more hydrophilic copolymer compositions, PCPP-5A (45:55) a and (21:79~ tended to crumble in the later stages of 9 degradation.
lo A separate study investigated the affect of 11 increasing the size of the alkane monomer and its affect 12 on degradation rate in the homologous poly [bis 13 (p-carboxyphenoxy) alkane] seriesO It was found that as 14 the number of methylene units in the R group backbone was increasedr the polymeric composition became more 16 hydrophobic and the degradation rates decreased several 1~ orders of magnitude. Specifically, as the methylene 18 units were increased from 1 to 6 in the composition, the 19 degradation rate underwent a decrease of 3 orders of magnitude. By extrapolation of the data in ~ig. 1, the 21 PCPP polymer is predicted to degrade completely and 22 slightly over 3 years' time; it is also thereore 23 apparent that the predictable time for degradation can 24 be increased or decreased at ~lill by several orders of magnitude by either increasing or decreasing the alkane 26 size.

Example_2 27 The degradation rates for PCPP at a variety of pH
28 levels ranging from 7.4-lO.0 were determined~ The PCPP
29 was prepared, purified, and compression molded into ~ - 14 -1 circular disks as described earlier in Example 1. Each 2 disk was placed in 10-50 milliliters (hereinaf~er ~mln) 3 of phosphate buffer held at 37~C which was prepared at 4 specific pH levels. The erosion of the disk was again followed by W absorbance at 250 nm over a test period 6 of 350 hours. The results are illustrated in Fig~ 2 7 which demonstrates that the rate of degradation is 8 increased by a ~actor o~ 20 as the pH is increased fr~m 9 7.4 to 10Ø It is also noteworthy that the rates of degradation remain stable and constant during the entire 11 two wee~ testing period.

Example 3 12 The biocompatibility and toxological effects were 13 investigated using the representative polyanhydride 14 polymers PCPP, PCPP-SA (45:55) poly (terephthalic acid S anhydride) hereinafter "PTA"~ and poly 16 (terephthalic-sebacic acid anhydride) hereinafter 17 "PTA-SA" in a 50:S0 ratio. The polymers PCPP and 18 PCPP SA ~45:55) were prepared in the manner described in 19 Example 1. PTA and PTA-SA (50:50) were synthesized in solution by an adaptation o~ the method of Yoda and 21 Matsuda tBull. Chem. Soc~ Japan 32:1120-1129 (1959);
22 J~panese Patent No. 10944]. This technique of solution 23 polymerization utilizes a dehydrative coupling reaction 24 between an acyl chloride and a carboxyl group to obtain the polymerl This polymerization technique is 26 pre~errable because the high melting point (372C) of 27 the PTA polymer and its instability (charring) at ~his 28 temperature made the melt-polycondensation methodology 29 unsuitable. Preferrably, 0.02 mole of terephthalic acid l - 15 -~ ~4~.7~
, 1 (or 0.02 mole of sebacic acid) was dissolved in 400 ml 2 of chloro~orm in the presence of 0.04 mole of triethyl~mine. Terephthaloyl chloride (0.0~ mole3 4 previously dissolved in benzene was added through a dropping funnel over a 30 minute period under vigorous 6 agitation. This mixure was allowed to react for three 7 ! hours at room temperature and was held under a nitrogen 8 ¦ sweep at all times. The resulting polymer, PTA or 9 PTA-SA (50:50) was then purified by extraction with I anhydrous ether in a Soxhlet Extractor for ~-3 hours and 11 was then stored in a dessicator over calcium chloride.
12 1 All samples for toxicological studies were prepared 13 ~ by placing each polymer ~n O.l M phosphate buffer, 14 1 pH 7.4, ~or several days and allowing the polymer to 1 degrade. The concentration of the degradation products 16 , was then determined by ultraviolet spectrophotometry.
17 ¦ The cytotoxicity and mutagenicity of the respective 1~ degradation products for the copol~mer PCPP-SA (45:55) 19 were determined by a forward mutation assay in Salmonella typhimurium using 8-azaguanine resistance as 21 a genetic marker as described in Skopec et al., Proc.
22 NatlO Aca. Sci. U~S.A. 75:410-414 (1978). This 23 mutagenicity assay also includes a test for toxicity so 2~ that the mutagenicity, i~ present, can be expressed quantitatively as the number of mutants per surviving 26 cell. In this way, an independent measurement of 27 toxicity for this bacterial species can be obtained.
28 Samples of PTA-SA (50:50) were tested at 1 mg/ml both 29 with and without the addition of mammalian metabolism enzymes. The results o this assay demonstrated that 31 j the degradation products of this polymer were I

Il - 16 -1 non-mutagenic with or without the addition of a 2 mammalian metabolizing system. In ail instances, the 3 induced mutant fraction was essentially 0 being indistinguishable from the spontaneous background ! control. Similarly, there was no toxicity in any 6 ¦ sample ~ithout the addition of metabolic enzymes and a 7 slight, but not significant, toxicity in samples with 8 the metabolizing system~ ¦
g The teratogenic potential was measured using a ¦ newly developed in-vitro assay in which the attached 11 efficiency of ascetic mouse ovarian tumor cells to 12 plastic surfaces coated with concanavalin A was 13 1 determined [Braun et al., Proc. Natl. Aca. Sci~ U.S.A.
14 79:2056-2060 (1982)]. In general with this assay, non-teratogens do not inhibit attachment of the tumor 16 cells to the plastic surfaces. The degradation products 17 of each respective polymer in phosphate buffer solutions 18 was titrated to p~l 7.4 with NaOH before ~the tests were 19 conducted since the attachment is sensitive to acidic pH
levels. The reaction mixture of degradation products 21 and tumor cells was allowed to react for 2 hours at room 22 temperature and a cell suspension was bufered with 50 23 mM of HEPES bufer during the two hour incubation 24 period. The preferred concentration for each sample was 2S 0.035 mg/ml. The teratogenicity test results indicated 26 an average decline in attachment efficiency of 35 -~3~.
27 Given that the criteria for potential teratogenicity is 28 an inhibition of cell attachment by more than 50%, the 29 degradation products of these polyanhydr;de compositions are considered non-teratogenic.

, ` 11 lZ~41~9 1 Testing in-vivo for the localized tissue response 2 to the polymers was determined by implantation in the 3 corneas of rabbits and subcutaneous introduction in 4 rats. For this series of tests, PCPP and PTA 5A (50:S0) S were fashioned into pellets whose dimensions were 1 mm x 6 1 mm x 0.5 mm and implanted surgically in rab~it 7 intracorneal pouch~s within the corneal stroma following 8 the methodology described in Langer et al., J. Biomed.
g Mat. Res. 15:267-277 (1981). The rabbit corneas were then inspected twice weekly by stereomicroscopy for 11 signs of inflammation as manifested by edema, cellular 12 infiltration, and neovascularization for a period of six 13 weeks. The host response to the polymeric pellets 14 implanted in the rabbit corneas is seen in Fig. 3a and 3b. Fig. 3a is the stereomicroscopic observation of the 16 cornea one week after implantation; Fig. 3b shows the 17 same cornea after the polymeric pellet had complet~ly 18 degraded after six weeks. No inflammatory response or 19 characteristics were observed at any time over the entire six week implantation and degradation period;
21 urthermore, the clarity o~ the corneas was maintained 22 and the prolife~ation oE new blood vessels absent in all 23 instances. In addition, a~ter the six week testing 24 period, each cornea was surgically removed and prepared ~or histological examination to confirm and verify the 26 accuracy of the stereomicroscopic data. A
Z7 representative cross section of rabbit cornea is seen in 28 Fig. 4 which provides a highly magnified view of the 29 epithelial cell layer EP, the cornneal stroma ST, and the endothelial cell layer EN. As is apparent there is 31 a total absence of inflammatory cells throughout the 32 corneas and this tissue appears normal in all respects.

` 1;2~ 7~
1 The in-vivo tissue response was determined by 2 subcutaneously implanting RCPP pellets in the abdominal 3 region of Sprague-Dawley rats following the procedure o 4 Brown et al., J. Pharm. Sci~ 72:1181-1185 (1983)~ After surgical introduction of the polymer pellet into the 6 rat, the animals were maintained normally for a six 7 month period ~ithout special attention. The rats were 8 then sacrificed and histological sections of the g abdominal tissues surrounding the site of implantation prepared and studied. The results are illustrated in 11 Fig. 5a and 5bo Fig. 5a is a magnified cross-sectional 12 view of the abdominal muscle wall in cross section 13 showing s~eletal muscle and blood vessels; Fig. 5b is 14 the identical tissue seen at a greater degree of magnification. It is apparent in each instance that no 16 inflammatory cell infiltration, that is the appearance 17 of polymorphonuclear leukocytes, macrophages and 18 lymphocytes, is seen in the tissues adjacent to the 19 implantation site. Similarly, gross post mortem inspection also did not reveal any abnormalities of any 21 kind at the împlantation site.

Example 4 22 To demonstrate the biocompatibility and non-toxic 23 properties of polyanhydride polymeric compositions as 24 a whole, additional in-vitro tissue culture studies were performed using PCPP-SA (45:55), PTA-SA (50:50) and PTA, 26 each of which was prepared as previsouly described.
21 The tissue culture studies relied on the ability of 28 endothelial cells and smooth muscle cells to grow and be 29 maintain in culture on t e s~rfsce of these polymers.

'9_ ' ~1274179 1 Endothelial cells were isolated from bovine aortas by 2 collagenase digestion and then cultured in-vitro using 3 the procedures described in Jaffe et al., J. Clin~
4 Invest. 52:2745-2756 (1973~. The identity of these endothelial cells was confirmed by staining ~or ~he 6 presence of Factor VIII antigen lJaffe et al., J. Clin.
7 Invest. 52:2757-2764 (1973). Smooth muscle cells were 8 obtained ~y explantation oE bovine aortic medial tissue 9 ¦ following the methods of Ross, R., J. Cell. Biol.
lo 50:172-186 (1971). At confluence, each cell type grew 11 in a char~cteristic "hill and valleyr morphology.
12 Approximately 1.0 x 104/cm2 cells which had been 13 previously passed in culture 4-15 times were plated 14 directly onto the circular pieces of pol~mer (having dimensions of approximately 1.5 cm2 x 1 mm) in a 0.25 ml 16 drop of culture medium and allowed to react for one 17 hour's duration at room temperature~ The polymeric 18 disks were then flooded with 15 ml of Dulbeccio's 19 Modified Eagle's medium containing 10% calf serum in 100 mm containers. The culture medium was changed daily 21 to avoid accumulation of the acidic polymer degradation 22 products. After two weeks in culture at room 23 temperature, the polymeric disks and adheren~ cells ~ere 24 rinsed with phosphate buffered saline, pH 7.2, and fixed ~or one hour with a 1~ buffered glutaraldehyde solution.
26 Examination of the cultured cells and polymers 27 reveal the following: the bovine aortic endothelial 28 cells grew normally both over the surfaces of the 29 polymers and in the Petri dish containing media and the degradation products of the polymer. In both instances 31 the cells displayed a normal morphology of polygonal I - 2~ -1 cells in a monolayer conformation typical of endothelial 2 cells. ~here was no evidence of any toxic ef~ect 011 3 these cells whatsoever~ Similarly, smooth muscle cells 4 also grew normally over the surfaces of the pol~mers and in the presence of the degradation products. These 6 smooth muscle cells were examined also for abnormalities 7 on the basis of typcial cell morphology or inhibition of 8 their ability to proliferate. Deleterius effects we're g absent and growth was normal for all samples tested.
Moreover, it was found that both the endothelial 11 cells and the smooth muscle cells grew and reproduced in 12 approximately the same amount of time in the control 13 samples utilizing polystyrene tissue culture plastic and 14 using the polyanhydride polymeric substanceO The 1S doubling times for endothelial cells were 16 ~2.5 hours 16 and 15 ~3.0 hours for the polystyrene plastic and the 11 polyanhydride polymers respectively: similarly, the '~
18 doubling,growth times for smooth muscle cells was 8 +1.5 19 hours and 8 ~2.0 hours using smooth muscle cells in the polystyrene plastic and polyanhydride polymers 21 respectively. Gross visual observation of the 22 attachment of both kinds o~ cells to the polyanhydride 23 polymers was dificult if not impossible because of the 24 opa~ue character of the polymers. For this reason, histiological sections were prepared and stained to 26 , confirm the normal appearance and growth of each type of 27 cell. Cross-sections o~ the fixed polyanhydride 28 polymer-complexes are seen in Figs. 6a and 6b which 29 reveals the presence of growing endothelial cells a~ a flattened monolayer similar in character and appearance 31 to endothelial cells growth as an intima layer on blood ~Z741~9 1 vessels in-vivo. There is no evidence of any enlarged 2 cells, there are no vacuoles, and there is no loss of 3 the normal growth pattern which would be visible if cell growth were co~pact-inhibited by the polyanhydride test s samples or their degradation products~
6 Bioerodible and biocompatible polyanhydride 7 polymeric compositions may be prepared ;n a wide variety 8 of specific formulations and be formed into preselected 9 dimensions and desired configurations as articles useful lo for implantation or prosthesis. Any of the presently 11 known compression molding techniques? casting 12 techniques, or other manufacturiny processe~ may be used 13 to fashion the article into the desired dimensions and 14 configurations to meet the specific application. The variety and multiplicity of uses for such articles are 16 exemplified by vascular graft materials, bioerodible 17 sutures, artificial skin surfaces, and orthopedic 18 devices generally such as bone plates and the like. In 19 particular applications the articles may also contain additional substances such as collagen, elastin, 21 proteins and polypeptides which also aid and promote 22 healing and repair o body tissues and organs.
23 A~ter the articles have been ~ormed into the 24 desired dimensions and configurations, it is introduced into the subject _-vivo at a predetermined tissue site~
26 It is expected that presently known surgical procedures 27 will be used for introduction of the article as these 23 are presently the most efficient and most reliable 29 techniques. In certain instances it can be seen that alternate methods of introduction will also be useful~
31 The speci c site for introduction of the article is --` ~ ~7 a matter of the user's need or choice and may be either 2 subcutaneous, superficial, or deeply imbedded into the 3 tissues or organs of the sub~ect as the need arises. It ~ will be recognized that the precise method of introduction and the particular site of implantation are 6 of no consequence and are merely selected at will from 7 the many procedures and applications suitable for use.
8 The present invention is not to be restricted in 9 form nor limited in scope except by the claims appended hereto.
'. .' l - 23 -,1 .

Claims (7)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows.

1. An article useful for implantation or otherwise placed in or on the body comprising a biocompatible, bioerodible, hydrophobic polyanhydride composition of the formula wherein R is an organic group and n is at least 2, wherein said polyanhydride is polymerized from prepolymers formed from dicarboxylic acids, and the prepolymers and the polymer are purified to remove materials provoking a severe tissue inflammatory response, said article degrading by hydrolysis with approximately zero order kinetics into non-toxic residues after introduction in vivo.

2. The article as recited in claim 1 wherein said R group is a hydrocarbon comprising from 2 - 16 carbon atoms.

3. The article as recited in claim 1 wherein said R group is

Claim 3 continued

4. The article as recited in claim 1 wherein the R group is wherein x is not less than 2 and not more than 16.

5. The article as recited in claim 1 wherein said R group is wherein x is at least 1.

6. The article as recited in claim 1 wherein R
group is wherein x is at least 1 and y is at least 1.

7. The article as recited in claim 1 wherein said R group is selected from the group consisting of