structured drug-eluting bioresorbable films: …meitalz/articles/a19.pdfto the engineering of new...

Structured Drug-elutingBioresorbable Films:

Microstructure and Release Profile

M. ZILBERMAN,* Y. SHIFROVITCH, M. AVIV AND M. HERSHKOVITZ

Department of Biomedical Engineering, Faculty of Engineering

Tel Aviv University, Tel Aviv 69978, Israel

ABSTRACT: Bioresorbable drug-eluting films can be used in many bio-medical applications. Examples for such applications include biodegradablemedical support devices which combine mechanical support with drug releaseand antibiotic-eluting film coatings for prevention of bacterial infectionsassociated with orthopedic implants or during gingival healing. In the currentstudy, bioresorbable drug-loaded polymer films are prepared by solutionprocessing. Two film structures are studied: A polymer film with large drugcrystals located on its surface (A-type) and a polymer film with small drugparticles and crystals distributed within the bulk (B-type). The basic mode ofdrug dispersion/location in the film (A or B-type) is found to be determinedmainly by the process of film formation and depends mainly on the solventevaporation rate, whereas the drug’s hydrophilicity has a minor effect on thisstructuring process. Most release profiles from A-type films exhibit a bursteffect of �30% and a second release stage that occurs at an approximatelyconstant rate and is determined mainly by the polymer weight loss rate.An extremely high burst release is exhibited only by a very hydrophilic drug.The matrix (monolithic) nature of the B-type film enables release profiles thatare determined mainly by the host polymer’s degradation profile, with a verylow burst effect in most of the studied systems. In addition to the drug location/dispersion in the film, the host polymer and drug type also strongly affect thedrug’s release profile from the film. It has been demonstrated that appropriateselection of the process parameters and film components (polymer and drug)can yield film structures with desirable drug release behaviors. This can lead

*Author to whom correspondence should be addressed. E-mail: [email protected]

JOURNAL OF BIOMATERIALS APPLICATIONS Volume 00 — 2008 1

0885-3282/08/00 0001–22 $10.00/0 DOI: 10.1177/0885328207088261� SAGE Publications 2008

Los Angeles, London, New Delhi and Singapore

+ [Ver: 8.07r /W] [11.7.2008–12:32pm] [1–22] [Page No. 1] REVISED PROOFS {SAGE_REV}Jba/JBA 088261.3d (JBA) Paper: JBA 088261 Keyword

J Biomater Appl OnlineFirst, published on July 16, 2008 as doi:10.1177/0885328207088261

Copyright 2008 by SAGE Publications. at Tel Aviv University on May 3, 2016jba.sagepub.comDownloaded from

http://jba.sagepub.com/

to the engineering of new bioresorbable drug-eluting film-based implants forvarious applications.

KEY WORDS: bioresorbable films, controlled drug delivery, dexamethasone,gentamicin, metronidazole, cortisone.

INTRODUCTION

Bioresorbable drug-eluting films can be used as basic elements invarious biomedical implants. Poly(�-hydroxy acid) films loaded with

water-soluble and water-insoluble drugs have been developed and studiedfor various applications [1–8]. For example, tetracycline was released fromPLLA barrier films and the system was studied for potential use inperiodontal therapy [1], gentamicin was released from poly(�-hydroxyacid) films for potential use in local treatment of bone infection [2,3],dexamethasone (DM) was released from various poly(�-hydroxy acid)films used as basic elements in tracheal films [4,5], and sirolimus wasreleased from a bi-layer PLLA-PLGA film in order to simulate its releasefrom a stent [6]. Bioactive agents with a relatively high molecular weight,such as albumin and the malaria vaccine (synthetic polypeptide),were released from PLGA films [7]. New combinations of materials,other than the traditional poly(�-hydroxy acid), were tried as hostpolymers [8–12] in order to exhibit appropriate mechanical propertieswith the desired release profile from polymeric bioresorbable films.For example, a combination of alginate and polyethylene glycol and a graftcopolymer of vinyl alcohol and lactic/glycolic acid were tried in paclitaxeleluting films for stent applications [8,9] and segmented poly(ether-ester-amide) films and cross-linked gelatin films were loaded withbioactive agents, such as metronidazole for dental applications [10,11].

Solution casting of polymers is a well-known method for preparingpolymer films. In order to incorporate drugs by this method, the polymeris dissolved in a solvent and mixed with the drug prior to casting.The solvent is then evaporated and the polymer/drug film is created.A method for controlling drug location/dispersion within the film has beenreported earlier [4,5,13,14]. This method is briefly described in thisparagraph. Bioresorbable polymeric films containing drugs were preparedusing the solution casting technique, accompanied by a post-preparationisothermal heat treatment. In this process, the solvent evaporation ratedetermines the kinetics of drug and polymer solidification and thus thedrug dispersion/location within the film. Solubility effects in the startingsolution also contribute to the post-casting diffusion processes, and occurconcomitantly to the drying step. In general, two types of polymer/drug

2 M. ZILBERMAN ET AL.

+ [Ver: 8.07r /W] [11.7.2008–12:32pm] [1–22] [Page No. 2] REVISED PROOFS {SAGE_REV}Jba/JBA 088261.3d (JBA) Paper: JBA 088261 Keyword at Tel Aviv University on May 3, 2016jba.sagepub.comDownloaded from


film structures were created and studied: a polymer film with large drugcrystals located on the film surface (A-type), and a polymer film withsmall drug particles and crystals distributed within the bulk (B-type).The morphology and formation process of these structured films havebeen studied extensively and a detailed model describing the structuringof these films is described elsewhere [4].

However, the promising technique for controlling the drug location/dispersion in the film raised the need to investigate structured filmsloaded with various hydrophilic and hydrophobic drugs and to elucidatethe effect of drug type and polymer type on the film microstructure andon the drug release profile. Four types of drugs were chosen in thecurrent study: dexamethasone (DM, hydrophobic), cortisone (CO,hydrophobic), gentamicin sulfate (GS, hydrophilic), and metronidazole(ME, partially hydrophilic). These drugs were incorporated into variouspoly(�-hydroxy acids) and the resulting polymer/drug films were studied.

Dexamethasone is a steroid anti-inflammatory drug. It has beenincorporated into bioresorbable systems for several uses: DM-containingsystems were developed and studied for intraocular application in thetreatment of inflammation following cataract surgery [14,15], DM-containing microspheres and tablets were investigated for treatment ofinflammatory bowel disease [16,17]. Local DM release was also used forvascular applications. For example, microspheres and nanoparticles wereused for local delivery of DM to the arterial wall in order to reduceneointimal formation after balloon angioplasty [18,19]. DM was alsoreleased from an intravascular eluting stent in order to preventrestenosis [20]. The authors have previously prepared poly(�-hydroxyacid) solution cast films loaded with DM using their structuringtechnique in order to control the drug location/dispersion in the film.The mechanical properties of the films and the expandable bioresorbabletracheal stents developed from the films have been studied [5].Dexamethasone was incorporated into the film-based stent to permitits controlled local release. In addition to its anti-inflammatory activity,DM has been demonstrated to inhibit the fibrotic response and theauthors thought that the addition of this drug may help preventproliferative reactions, such as induced airway stenosis. Cortisone (CO) isalso a steroid anti-inflammatory drug. DM and CO are both hydrophobic.

Gentamicin sulfate (GS) is a water soluble antibiotic drug used totreat bacterial infections, especially Gram-negative ones. GS moleculesreact with the bacterial ribosomes and consequently interfere withprotein synthesis. GS was incorporated in poly(methyl methacrylate)(PMMA) to form GS-containing PalacosTM PMMA bone cement [21,22].GS was also incorporated into fracture fixation devices for the

Drug-eluting Bioresorbable Films 3



prevention of bacterial infections associated with orthopedic implants.For example, researchers have coated metal implants, such as plates andwires with a polymer/gentamicin layer, using the dipping technique[23–25]. The obtained release profile demonstrated that most of the drugwas released within several hours after exposure to an aqueousenvironment. Melt processing techniques, such as compression moldingand extrusion, were used to develop bioresorbable implants loaded withgentamicin to serve as an additional part of the fracture fixation device.Cylinders, films, and disks were developed and studied [26–28]. Based onthe polymer/drug film structuring technique, gentamicin-loadedbioresorbable films were prepared that can be ‘bound’ to orthopedicimplants (by slightly dissolving their surface before attaching them tothe implant surface) and prevent bacterial infections by controlledrelease of gentamicin for at least one month. It is expected that thesesystems will provide desired drug delivery profiles and will not requirean additional implant.

Metronidazole is a partially hydrophilic drug, which effectivelyinhibits anaerobic microorganisms and protozoan infections [29,30].This drug is used for the treatment of many infections, includingperiodontal and vaginal infections [31–33]. This drug was incorporatedin drug delivery systems for various applications, such as tablets fortreatment of peptic ulcers [34], microspheres for the treatment ofdiseases associated with the colon and the gastric mucosa [35],mucoadhesive gel and tablets for the treatment of periodontal diseases[36,37], alginate gel beads for gastric applications [38], and films for thetreatment of periodontal pockets [32].

The aim of the present study was to examine polymer/drug films basedon amorphous and semicrystalline poly(�-hydroxy acids) and todetermine the effects of drug type, film type, and polymer weight losskinetics on the film structure (drug location/dispersion in the film) andon the resulting drug release profile from the film.

EXPERIMENTAL DETAILS

Materials

Bioresorbable polymersTwo types of poly(L-lactic acid) were used:

1. Poly(L-lactic acid), Resomer L210, Inherent viscosity (I.V.)¼ 3.9 dL/gin CHCl3 at 308C, M.W.¼�450,000 Da, Boehringer Ingelheim,Germany. This polymer will herewith be designated PLLA.




2. Poly(L-lactic acid), Medisorb 100 L, I.V.¼ 1.6 dL/g in CHCl3 at 308C,M.W.¼�200,000 Da, Alkermes, USA. This relatively low molecularweight polymer will herewith be designated PLLA(L).

Two types of 75/25 poly(DL-lactide-co-glycolide) were used:

1. 75/25 poly(DL-lactic-co-glycolic acid), A123-12, I.V.¼ 0.65 dL/g inCHCl3 at 308C, M.W. ¼�97,100 Da, Absorbable PolymerTechnologies Inc., USA. This polymer will herewith be designated75/25 PDLGA.

2. 5/25 poly(DL-lactic-co-glycolic acid), A123-03, I.V.¼ 0.24 dL/g inCHCl3 at 308C, M.W. ¼�23,800 Da, Absorbable PolymerTechnologies Inc., USA. This relatively low molecular weight polymerwill herewith be designated 75/25 PDLGA(L).

50/50 Poly(DL-lactic-co-glycolic acid), I.V.¼ 0.60 dL/g in CHCl3 at308C, M.W.¼�80,000 Da, Birmingham Polymers, USA. This polymerwill herewith be designated 50/50 PDLGA.

Poly(DL-lactic acid), I.V.¼ 1.07 dL/g in CHCl3 at 308C, M.W.¼�93,000 Da, Birmingham Polymers, USA. This polymer will herewithbe designated PDLLA.

The inherent viscosity, initial molecular weight, and approximatedegradation time of all polymers used in this study are presented inTable 1.

Drugs:Dexamethasone (DM) USP, 9�-fluoro-162-methylprednisolone, Sigma

D-9184.Gentamicin sulfate (GS), (cell-culture tested), 590mg gentamicin base

per mg, Sigma G-1264.Cortisone (CO), 17�,21-dihydroxy-4-pregnene-3,11,20-trione, Sigma

C-2755.

Table 1. Properties of the bioresorbable polymers used in this study.

PolymerInherent

viscositya (dL/g)Initial molecular

weight (Da)Approximate degradation

time (months) in water [40]

PLLA 3.90 450,000 424PLLA(L) 1.60 200,000 42475/25 PDLGA 0.65 97,100 4–575/25 PDLGA(L) 0.24 23,800 N/A50/50 PDLGA 0.60 80,000 2–3

aThe inherent viscosity values were measured by the manufacturers.




Metronidazole (ME), 2-Methyl-5-nitroimidazole-1-ethanol, SigmaM-3761.

The molecular weight and water solubility of all drugs used in thisstudy are presented in Table 2.

Film preparation

Polymer films (0.12–0.15 mm thick) consisting of poly(�-hydroxy acid)and drug were prepared by a three-step solution processing method:

Step 1. Components were mixed in a solvent at room temperature untilpolymer dissolution. A constant quantity of 1 g polymer and 0.1 gdrug (10% w/w) were chosen for all experiments. Both a dilutedand a concentrated solution were prepared for each polymer/drugsystem. Chloroform was used as a solvent for all polymer/drugsystems, except those containing ME, in which methylenechloride was used.

Dilute solutionsDilute solutions were prepared by using relatively large volumes of

solvent (polymer concentration: 0.01 g/mL, drug concentration: 0.6 mg/mL). Both the polymer and the drug were totally dissolved.

Concentrated solutionsConcentrated solutions were prepared by using relatively small

volumes of solvent (polymer concentration: 0.05 g/mL, drug concentra-tion: 3.0 mg/mL). In these solutions the polymer was totally dissolved,whereas the drug powder was only broken into small particles(aggregates), which yielded an opaque solution.

Step 2. Solution casting into a Petri dish and solvent drying underatmospheric pressure at room temperature. Two solventevaporation rates were used. They could not be measured precisely

Table 2. Properties of the drugs used in this study a.

DrugMolecular weight

(g/mole)Solubility in H2O

(mg/100 mL)

Dexamethasone (DM) 392.46 25Cortisone (CO) 360.44 28Gentamicin sulfate (GS) 477.6 5000Metronidazole (ME) 171.15 1000

aThe data were taken from the Merck index.




but were evaluated as a relatively slow rate of 2–5 mL/h and arelatively fast rate of 10–20 mL/h. In order to get slowevaporation the Petri dish was covered with aluminum foil and inorder to get slow evaporation rate the Petri dish was not covered.

Step 3. Isothermal heat treatment at a temperature higher than the glasstransition temperature of the polymer for at least 1 h in a vacuumoven. This heat treatment enabled disposal of residual solvent.

Morphological characterization

Polarized light microscopy (LM) was performed using an OlympusBHS compound microscope or a Leica microscope (transmission mode).

In vitro weight loss studies

The polymer films (weight range 0.2–0.3 g) were weighed and thenimmersed in sterile water (50 mL, in a Petri dish) at 378C in order todetermine their weight loss profiles. Samples (also those broken intosmall pieces) were removed every week, dried in a vacuum oven, andweighed. If samples started to break down into small pieces due todegradation, the pieces were first collected. The weight loss wascalculated as:

Weight loss ð%Þ ¼ 100�w0 �wf

w0, ð1Þ

where w0 and wf are the weights of the dried films before and afterexposure to water, respectively. Five samples were tested for each point.The mean values and standard deviations are presented in the graphs.

In vitro drug release study

Polymer/drug films (0.08–0.1 g) were immersed in sterile water (onlypolymer/GS films were immersed in PBS) at 378C in order to determinethe kinetics of drug release from the films. Samples, 1.5 mL, werecollected every week and replaced with a fresh medium of sterile water.The drug content in each sample was measured. DX500 Dionex HighPerformance Liquid Chromatography (HPLC) was used to measureDM and CO contents and an AD-20 Ultraviolet detector was used tomonitor absorbance of DM at 254 nm and CO at 245 nm. The GSconcentration was measured by spectrophotometry after derivatization




with o-phthalaldehyde using Sampath and Robinson’s procedure [39],as follows: O-phthaldialdehyde reagent was prepared by adding 2.5 go-phthaldialdehyde, 62.5 mL methanol, and 3 mL 2-hydroxyethylmer-captan to 560 mL 0.04 M sodium borate in distilled water. Two-hundredmicroliters gentamicin solution, 200 mL isopropanol (to preventsedimentation), and 200 mL o-phthaldialdehyde reagent were reactedfor 45 min at room temperature. The absorbance, which corresponds tothe gentamicin concentration, was then measured at 333 nm. MEcontents were measured directly at 320 nm, without using any specificreaction. The GS and ME measurements were performed using a UV/visscanning spectrophotometer (Anthos, Zenyth 200rt). At least threesamples were examined for each film type. The mean values andstandard deviations are presented in the graphs.

RESULTS AND DISCUSSION

The morphology of polymer/drug films

Polymer drug films were prepared by solution casting. In this process,the solvent evaporation rate strongly affects the kinetics of drug andpolymer crystallization, and thus the mode of drug dispersion in thepolymer [4]. Two types of film structures were created for most polymer/drug films in this study:

(a) A polymer film with large crystalline drug particles located onits surface, as presented in Figure 1(a) for PLLA films loadedwith DM. This structure was derived from a dilute solution of bothpolymer and drug and was obtained using the slow solventevaporation rate which enables prior drug nucleation and growthon the polymer solution surface. This skin formation is accompaniedby a later polymer core formation/solidification. This structure wasnamed the ‘A-type’.

(b) A polymer film with small drug particles and crystals distributedwithin the bulk, as presented in Figure 1(b) for PLLA films loadedwith DM. This structure was derived from a concentrated solutionand was obtained using the fast solvent evaporation rate andresulted from drug nucleation and segregation within a densepolymer solution. Solidification of drug and polymer occurredconcomitantly. This structure was named the ‘B-type’. It has beennoted earlier that in semi-crystalline based films, such as PLLA/DM,the drug is located in amorphous domains of a semi-crystallinematrix, around the spherulites [4].




The concept of ‘film structuring’ was investigated and it was foundthat A-type and B-type DM loaded films can be developed for both typesof polymers, i.e., amorphous and semi-crystalline. Amorphous A-typefilms based on PDLLA and 50/50 PDLGA are presented (as examples) inFigure 1(c) and (e) and B-type films based on PDLLA and 50/50 PDLGAare presented (as examples) in Figure 1(d) and (f). In these films largerectangular DM crystals are located on the film surface of the A-typefilms, whereas small DM particles and crystals are spread within thebulk of the B-type films. Thus, it is demonstrated that the host polymertype affects the film morphology but has almost no effect on the drug

(a) PLLA/DM-A type

(c) PDLLA/DM-A type (d) PDLLA/DM-B type

(e) 50/50 PDLGA/DM-A type (f) 50/50 PDLGA/DM-B type

100µm 100µm

100µm 100µm

100µm 100µm

(b) PLLA/DM-B type

Figure 1. Polarized light micrographs of various polymer/DM films. ((a) and (b) are takenfrom an earlier publication by the authors [5] and (c) and (d) from [4]).




location/dispersion within the polymeric film. The latter is determinedmainly by the kinetics of film formation [4].

Three additional drugs were used in order to study the effect of thedrug on the film structure: cortisone (CO) – a water insoluble drug,gentamicin sulfate (GS) – a water soluble drug, and metronidazole (ME)– a partially water soluble drug. Both A and B-type structures werecreated for these polymer/drug films. The film structures (based onPLLA) are presented in Figure 2. ME, similarly to DM, also created largerectangular crystals on the surface of the film when a slow solventevaporation rate was used to create the A-type film (Figure 2(a)).In contradistinction, when a fast evaporation rate was used most of the

(a) PLLA/ME- A type

50µm 50µm

50µm 50µm

50µm100µm

(b) PLLA/ME- B type

(d) PLLA/CO- B type(c) PLLA/CO- A type

(f) PLLA/GS- B type(e) PLLA/GS- A type

Figure 2. Light micrographs of various PLLA/drug films.




ME was spread in particles and small crystals within the film, but someof the drug was located on the surface (Figure 2(b)). Large drug crystalson the surface of the polymeric films were also observed when CO wasused. However, they were arranged in star-like structures (Figure 2(c))rather than in rectangular structures, as with DM and ME. PLLA/GSfilms also created an A-type structure when a slow solvent evaporationrate was used. However, in this case it is difficult to see the drugparticles, probably because of the irregularity of the host polymer(Figure 2(e)). B-type films were also created for PLLA/CO and PLLA/GSsystems when a high solvent evaporation rate was used (Figure 2 (d) and(f)). While observing the film morphologies it was noticed that partof the drug particles in B-type films loaded with ME and GS are locatedon the surface of the film. This was probably due to the hydrophilicnature of the drug. Although a concentrated solution and fast solventevaporation rate were used, some of the drug molecules are not ‘trapped’in the viscous crystallizing medium and tend to diffuse out to the surfacebecause of their hydrophilic nature (the polymer solution is hydro-phobic). In conclusion, the basic mode of drug dispersion/location in thefilm (A or B-type) is determined mainly by the film formation processand depends mainly on the solvent evaporation rate. The drug’shydrophilicity has some effect on this structuring process.

In vitro Drug Release from Bioresorbable Polymer/Drug Films

In the current study, the effect of film structure (drug location/dispersion), drug type, polymer type and initial molecular weight on thedrug release profile was examined. Five host polymers of various chain(chemical) structures were chosen for this study. Their inherentviscosities, initial molecular weights (evaluated by the manufacturers),and approximate total degradation times are presented in Table 1.PLLAs are the only semicrystalline polymers used in this study.50/50 PDLGA and 75/25 PDLGA are amorphous. The neat films wereexposed to aqueous medium and their weight loss was measuredat weekly intervals. The weight loss profiles are presented in Figure 3.The weight of the polymers that degrade relatively fast could not bemeasured adequately for long periods of time and in these cases theweight loss study was performed only for a few weeks.

Both PLLAs degrade slowly. The relatively high M.W. PLLA showedalmost no weight loss for 28 weeks in aqueous medium and the relativelylow M.W. PLLA lost only 7% of its initial weight after 20 weeks ofincubation. Both 75/25 PDLGA and 50/50 PDLGA contain glycolic acidin addition to lactic acid monomers. Lactic acid degrades slower than




glycolic acid due to the additional CH3 group which makes it morehydrophobic and also provides steric hindrance to water attack on esterbonds. Both PDLGAs therefore degraded much faster than PLLA andpractically could not be handled after several weeks of degradation.The low M.W. 75/25 PDLGA exhibited the fastest weight loss profile.

Effect of Drug Location/Dispersion in the FilmThe cumulative release profiles of DM from both A-type and B-type

PLLA films are presented in Figure 4(a). Both PLLA based filmsexhibited an almost linear release profile, i.e., constant release rate(slope). Such systems are desirable for many controlled releaseapplications. The A-type film contains large DM crystals on the polymerfilm surface (Figure 1(a)). The A-type film therefore exhibited a ‘bursteffect’, as expected, i.e., �30% of the drug was released during the firstweek of immersion in an aqueous medium. In contradistinction, mostof the DM in the B-type film is located within the polymer film(Figure 1(b)) and only a small portion is located on the surface. TheB-type PLLA/DM film therefore did not exhibit a burst effect, and itsrelatively slow rate of release results from the low rate of degradationand weight loss (Figure 3). It is clear that in this system degradationof the matrix polymer is essential in order to enable water penetrationinto the film and DM release from the film. Surprisingly, only 30% of theDM was released from the A-type PLLA/DM film during the first 3 daysof degradation. This release behavior probably results from DM’s veryhydrophobic nature (water solubility of 25 mg/100 mL, Table 2).Furthermore, PLLA contains ester groups and each DM molecule

14

12

10

8

6

4

2

00 5 10 15 20

Time (weeks)

Pol

ymer

wei

ght l

oss

(%)

25 30 35

Figure 3. Polymer weight loss of various films as a function of degradation time: x – 50/50PDLGA, m– PLLA, �– PLLA(L), g – 75/25 PDLGA, �– 75/25 PDLGA(L).




contains two carbonyl and three hydroxyl groups. Relatively strongspecific interactions, namely hydrogen bonds, may have formed betweenthe carbonyl oxygen atoms in the PLLA chains and the hydroxylhydrogen atoms in DM. Such interactions may prevent intensive releaseof DM to the aqueous medium. Degradation of the polymer must occurin order to enable further release of drug located in the film. The factthat the constant DM release rate from the B-type PLLA/DM film is verysimilar to that from the A-type film supports the latter suggestion.

The 75/25 PDLGA film is amorphous and degrades relatively fast(Figure 3). The release profiles of DM from A and B-type 75/25 PDLGAfilms are presented in Figure 4(b). Unexpectedly, the DM release profilefrom the 75/25 PDLGA (A-type) film is similar to the release profile fromthe 75/25 PDLGA (B-type). A difference was observed mainly during thefirst 4 weeks of the study, and in both cases most of the drug wasreleased after 20 weeks. This phenomenon is attributed to the relativelyhigh rate of matrix polymer degradation and weight loss.

The cumulative release profiles of GS from PLLA and 75/25 PDLGAfilms are presented in Figure 4(c) and (d). The effect of drug location/

(a)

(c)

0 4 8

Time (weeks)

12 16 20

0

GS

cum

ulat

ive

rele

ase

(%)

DM

cum

ulat

ive

rele

ase

(%)

DM

cum

ulat

ive

rele

ase

(%)

GS

cum

ulat

ive

rele

ase

(%)

0

20

40

60

80

100

60 9080706050403020100

50

40

30

20

10

0

0

20

40

60

80

100

5 10

Time (weeks)

B type films

B type filmsB type films

B type film

A type film

A type film

A type film

A type film

15 20 25 0 3 6

Time (weeks)

9 12

0 5 10

Time (weeks)

15 20

(b)

(d)

Figure 4. In vitro cumulative drug release from A-type and B-type polymer/drug films:(a) PLLA/DM films, (b) 75/25 PDLGA/DM films, (c) PLLA/GS films, and (d) 75/25 PDLGA/

GS films. œ – A-type films, g – B-type films.




dispersion in the film is similar to that observed for DM loaded films.However, since GS is a very hydrophilic drug (water solubility of5000 mg/100 mL, Table 2), it tends to diffuse from the host polymer andthe release rates of GS from the PLLA films (Figure 4(c)) are thereforehigher than those of DM (Figure 4(a)). When a host polymer whichdegrades very rapidly (such as 75/25 PDLGA) is used, a very highundesired burst release of hydrophilic drugs is obtained (Figure 4(d)).Drug location/dispersion in the polymeric film thus has a dominanteffect on the release profile of the drug from the film when thedegradation rate of the host polymer is slow. When the host polymerexhibits a fast degradation rate, a weight loss of several weeks probablyenables some porosity which accelerates the diffusion of the drugmolecules and the drug location/dispersion in the film therefore hasa minor effect on the drug release profile.

Effect of Drug TypeThe effect of the drug on its release profile from 75/25 PDLGA films

is presented in Figure 5. A-type films loaded with DM, CO, and MEexhibited very similar release profiles, with a burst release of 30%followed by a moderate drug release (Figure 5(a)), as expected for A-typefilms with drug located on the surface of the film. Only GS-loaded filmsreleased most of the drug during the first day, due to the hydrophilicnature of the drug molecules. B-type films loaded with DM and COexhibited a very low burst release followed by a relatively slow releaseat constant rate during weeks 2–20 (Figure 5(b)). B-type films loadedwith ME also exhibited a relatively low burst release, but continuedto release the ME molecules at a relatively high rate during the first4 weeks of release. ME is partially water soluble (water solubility of

(a) 75/25 PDLGA/drug films A-type

10080604020

0

10080604020

00 5 10

Time (weeks)

Cum

ulat

ive

drug

rele

ase

(%)

Cum

ulat

ive

drug

rele

ase

(%)

15 20 0 5 10

Time (weeks)

15 20

(b) 75/25 PDLGA/drug films B-type

Figure 5. Effect of the drug type on the cumulative drug release from 75/25 PDLGA/drugfilms: (a) A-type films, (b) B-type films. Drug: g – Cortisone, œ – Dexamethasone,

� – Metronidazole, m– Gentamicin.




1000 mg/100 mL, Table 2) and therefore diffuses out of the bioresorbablefilms faster than the hydrophobic DM and CO (water solubility of 25 and28 mg/100 mL, respectively). The relatively high rate of ME release fromthe B-type films may also result from the film structure, i.e., in thesefilms some of the drug particles are located on the surface of the film(Figure 2(b)). B-type 75/25 PDLGA/GS films released 65% of the drugduring the first day of release and a second phase of release started after4 weeks (Figure 5(b)). The high burst release results from thehydrophilic nature of the drug and the second phase of release probablyresults from intensive degradation of the host polymer (8% weight lossafter 4 weeks), which enables further diffusion of the remaining drugmolecules. It can be concluded that the drug’s hydrophilicity has a majoreffect on its release rate, in addition to the drug location/dispersionin the film.

Effect of the Host Polymer and its Initial Molecular WeightA series of three polymers: 50/50 PDLGA, 75/25 PDLGA, and PLLA,

were used to elucidate the effect of the host polymer on the releaseprofile. The DM release profiles from the B-type films (Figure 6(a))indicate that the weight loss rate of the matrix polymer (Figure 3) hasa significant effect on the rate of drug release. This behavior is expected,since most of the drug is located within the polymeric film, and supportsthe conclusion that degradation processes play a major role in thesecontrolled release systems and that release does not occur solely bysimple diffusion. Actually, these B-type films are monolithic (matrix)systems based on bioresorbable polymers. When monolithic systems arebased on non-resorbable polymers, the release rate will usually decreasewith time. In this research, the systems are based on bioresorbablematrix polymers. Degradation processes and the resulting increasein the system’s porosity therefore enable more effective diffusion of drugmolecules from the film, which in most cases give rise to release profilesthat are closer to a straight line (constant release rates).

All release profiles of the hydrophobic DM from A-type films exhibita burst effect of �30% during the first 3 days (Figure 6(b)), indicatingthat 30% of the drug is probably located on the surface of the film butis not bound. The second release phase from the A-type film occurs at anapproximately constant rate, which depends on the rate of polymerdegradation. This phenomenon supports the suggestion that part of thedrug may be physically bound to the polymer molecules within the filmvia hydrogen bonds. In contradistinction, the release profiles of thehydrophilic GS from A-type films based on PDLGAs exhibit a high bursteffect of at least 65% (Figure 6(c)). Only PLLA/GS films showed a more




(a)

(b)

(c)

80

60

40

20

0

80

60

40

20

0

80

100

60

40

20

0

0 5 10Time (weeks)

DM

cum

ulat

ive

rele

ase

(%)

DM

cum

ulat

ive

rele

ase

(%)

GS

cum

ulat

ive

rele

ase

(%)

15

PLLA

PLLA

–PLLA

–75/25 PDLGA

–50/50 PDLGA

75/25 PDLGA

75/25 PDLGA

50/50 PDLGA

50/50 PDLGA

20

0 5 10

Time (weeks)

15 20

0 3 6

Time (weeks)

9 12

Figure 6. Effect of the host polymer type on the cumulative drug release from A-typeand B-type polymer/drug films: (a) B-type polymer/DM films, (b) A-type polymer/DM films,

and (c) A-type polymer/GS films. The host polymer is indicated.




moderate release profile. It is therefore suggested that if controlledrelease of water soluble drugs, for example GS, is desired, they should beloaded only onto films based on polymers with a slow rate of degradation(preferably B-type films).

The effect of the initial molecular weight of the host polymer on therelease profile of GS from 75/25 PDLGA and PLLA films is presentedin Figure 7. When 75/25 PDLGA was used as a host polymer, a higherinitial molecular weight resulted in a lower burst release from the B-typefilm (Figure 7(b)), but did not affect the release profile of GS from theA-type film (Figure 7(a)). When PLLA was used as a host polymer,a higher initial molecular weight resulted in a lower burst release anda moderate release profile (Figure 7(c) and (d)). As expected, this trendis more significant for B-type films than for A-type films. Films basedon higher molecular weight polymers exhibited lower burst effects dueto a greater amount of hydroxylic and carboxylic edge groups,which make the polymer more hydrophobic. Furthermore, a highermolecular weight results in a higher glass transition temperature,which facilitates slower drug release from the polymer. A relatively high

(a) (b)

(c)

PLLA(L)

PLLA(L)

PLLA

PLLA

00

20406080

100

0

20

40

60

80

100

0

20

40

60

80100

020406080

100

3 6

Time (weeks)

GS

cum

ulat

ive

rele

ase

(%)

GS

cum

ulat

ive

rele

ase

(%)

GS

cum

ulat

ive

rele

ase

(%)

GS

cum

ulat

ive

rele

ase

(%)

9

75/25 PDLGA75/25 PDLGA (L)

75/25 PDLGA

75/25 PDLGA (L)

12

0 5 10

Time (weeks)

15 20 25 0 5 10

Time (weeks)

15 20 25

0 3 6

Time (weeks)

9 12

(d)

Figure 7. Effect of the host polymer’s initial molecular weight on the cumulative release

from polymer/GS films: (a) A-type 75/25 PDGLA/GS films, (b) B-type 75/25 PDGLA/GSfilms, (c) A-type PLLA/GS films, and (d) B-type PLLA/GS films. g – Low M.W. host

polymer, œ – Relatively high M.W. host polymer.




initial molecular weight is thus favorable when bioresorbable filmsare loaded with hydrophilic drugs which tend to diffuse out easily fromthe host film. It is more effective in B-type films and when the hostpolymer is more hydrophobic.

Finally, the results demonstrate that in addition to the drug location/dispersion in the film, the host polymer and drug type also stronglyaffect the drug’s release profile from the film. The ability to control thedrug release profile by choosing the right process parameters andpolymer type makes it possible to fit the release profile to theapplication. For example, if relatively high release rate is needed,a system based on polymer with fast degradation rate should be chosenand A-type film should be considered when burst release is needed.In contradiction, when relatively low release rate is needed, a systembased on polymer with slow degradation rate should be chosen andB-type film should be considered when high burst release is notfavorable. In certain cases, the drug type (hydrophobic/hydrophilic)can also be chosen so that it will help to get a desired profile. It shouldalso be mentioned that large drug crystals on the surface of thepolymeric film may affect the degree of foreign body reactions and asa result, also the drug release profile. This point should also beaddressed when the system for a specific application is considered.

SUMMARY AND CONCLUSIONS

Bioresorbable drug-loaded polymer films were prepared by solutionprocessing. Investigation of the films focused on cumulative drug releaseas affected by film morphology (drug location/dispersion in the film),drug type, the host polymer, and its initial molecular weight. Two filmstructures were studied: a polymer film with large drug crystals locatedon its surface (A-type), and a polymer film with small drug particles andcrystals distributed within the bulk (B-type). The basic mode of drugdispersion/location within the film (A or B-type) is determined mainly bythe process of film formation and depends mainly on the solvent’sevaporation rate. The drug’s hydrophilicity has some effect on thisstructuring process.

In general, the drug location/dispersion in the film, the host polymer,and the drug type determine the drug’s release profile from the film.The initial molecular weight of the host polymer can help achieve thedesired drug release profile in certain cases. Most release profiles fromA-type films exhibited a burst effect of �30% and a second release stagefrom these films occurs at an approximately constant rate, which isdetermined mainly by the polymer weight loss rate. Only a very




hydrophilic drug exhibited an extremely high burst release. In mostof the studied systems, the matrix (monolithic) nature of the B-type filmenabled release profiles that were determined mainly by the degradationprofile of the host polymer, with a very low burst effect.

It has been shown that appropriate selection of the processparameters and film components (polymer and drug) can yield filmstructures with desirable drug release behaviors. This can lead to theengineering of new bioresorbable drug-eluting film-based implants forvarious applications.

REFERENCES

1. Park, Y.J., Lee, Y.M., Park, S.N., Lee, J.Y., Ku, Y., Chung, C.P. and Lee, S.J.(2000). Enhanced Guided Bone Regeneration by Controlled TetracyclineRelease from Poly(L-lactide) Barrier Membranes, Journal of BiomedicalMaterials Research, 51(3): 391–397.

2. Lee, K.B., Kang, S.B., Kwon, I.C., Kim, Y.H., Choi, K. and Jeong, S.Y.(1997). Controlled Release and in vitro Antimicrobial Activity of Gentamicinfrom Poly-�-hydroxy Acid Films, In: Proceedings of the 24th InternationalSymposium on Controlled Release of Bioactive Materials, pp. 571–572.

3. Mauduit, J., Bulkh, N. and Vert, M. (1993). Gentamicin/Poly(lactic acid)Blends Aimed at Sustained Release Local Antibiotic Therapy AdministratedPer-operatively. III The Case of Gentamicin Sulfate in Films Prepared fromHigh and Low Molecular Weight poly(DL-lactic Acids), Journal of ControlledRelease, 25(1–2): 43–49.

4. Zilberman, M., Schwade, N., Meidell, R. and Eberhart, R. (2001). StructuredDrug Loaded Bioresorbable Films for Support Structures, Journal ofBiomaterials Science, Polymer Ed., 12: 875–892.

5. Zilberman, M., Eberhart, R.C. and Schwade, N.D. (2002). In-vitro Studyof Drug Loaded Bioresorbable Films and Support Structures, Journal ofBiomaterials Science Polymer, 13: 1221–1240.

6. Wang, X., Venkatraman, S.S., Boey, F.Y.C., Loo, J.S.C. and Tan, L.P. (2006).Controlled Release of Sirolimus from a Multilayered PLGA Stent Matrix,Biomaterials, 27: 5588–5595.

7. Dorta, M.J., Santovena, A., Liabres, M. and Farina, J.B. (2002). PotentialApplications of PLGA Film-implants in Modulating In-vitro Drugs Release,International Journal of Pharmaceutics, 248: 149–156.

8. Livnat, M., Beyar, R. and Seliktar, D. (2005). Endoluminal Hydrogel FilmsMade of Alginate and Polyethylrnr Glycol: Physical Characteristics andDrug-eluting Properties, Journal of Biomedical Materials Research Part A,75A: 710–722.

9. Westedt, U., Wittmar, M., Hellwig, M., Hanefeld, P., Greiner, A., Schaper,A.K. and Kissel, T. (2006). Paclitaxel Releasing Films Consistingof Poly(Vinyl Alcohol)-Graft-Poly(Lactide-co-Glycolide) and Their Potentialas Biodegradable Stent Coatings, Journal of Controlled Release,111: 235–246.




10. Cetin, E.O., Buduneli, N., Athhan, E. and Kurilmaz, L. (2005). In-vitroStudies of a Degradable Device for Controlled-release of Meloxicam, J. Clin.Periodontol., 32: 773–777.

11. Quaglia, F., Vignole, M.C., De Rosa, G., La Rotonda, M.I., Maglio, G.and Palumbo, R. (2002). New Segmented Copolymers ContainingPoly("-caprolactone) and Etheramide Segments for the Controlled Releaseof Bioactive Compounds, Journal of Controlled Release, 83: 263–271.

12. Tarvainen, T., Karjalainen, T., Malin, M., Perakorpi, K., Tuominen, J.,Seppala, J. and Jarvinen, K. (2002). Drug Release Profiles from andDegradation of a Novel Biodegradable Polymer, 2,2-bis(2-oxazoline) LinkedPoly("-caprolactone), European Journal of Pharmaceutical Sciences,16: 323–331.

13. Zilberman, M. (2005). Dexamethasone Loaded Bioresorbable Films used inMedical Support Devices: Structure, Degradation, Crystallinity, and DrugRelease, Acta Biomaterialia, 1(6): 615–625.

14. Kodama, M., Numaga, J., Yoshida, A., Kaburaki, T., Oshika, T., Fujino, Y.,Wu, G.S., Rao, N.A. and Kawashima, H. (2003). Effects of NewDexamethasone-delivery System (Surodex) on Experimental IntraocularInflammation Models, Graefes Archive for Clinical and ExperimentalOphthalmology, 241(11): 927–933.

15. Kagaya, F., Usui, T., Kamiya, K., Ishii, Y., Tanaka, S., Amano, S. andOshika, T. (2002). Intraocular Dexamethasone Delivery System for CornealTransplantation in an Animal Model, Cornea, 21(2): 200–202.

16. Leopold, C.S. and Eikeler, D. (2000). Basic Coating Polymers for the Colon-specific Drug Delivery in Inflammatory Bowel Disease, Drug Developmentand Industrial Pharmacy, 26912: 1239–1246.

17. Nakase, H., Okazaki, K., Tabata, Y. and Chiba, T. (2003). BiodegradableMicrospheres Targeting Mucosal Immune-regulating Cells: New Approachfor Treatment of Inflammatory Bowel Disease, Journal of Gastroenterology,38: 59–62.

18. Dev, V., Eigler, N., Fishbein, M.C., Tian, Y., Hickey, A., Rechavia, E.,Forrester, J.S. and Litvack, F. (1997). Sustained Local Drug Delivery to theArterial Wall via Bioresorbable Microspheres, Catheterization andCardiovascular Diagnosis, 41(3): 324–332.

19. Guzman, L.A., Labhasetwar, V., Song, C., Jang, Y., Linkoff, A.M., Levy, R.and Topol, E.J. (1996). Local Intraluminal Infusion of BiodegradablePolymeric Nanoparticles. A Novel Approach for Prolonged Drug Deliveryafter Balloon Angioplasty, Circulation, 94(6): 1441–1448.

20. Lincoff, A.M., Furst, J.G., Ellis, S.G., Tuch, R.J. and Topol, E.J. (1997).Sustained Local Delivery of Dexamethasone by Novel IntravascularEluting Stent to Prevent Restenosis in the Porcine Coronary InjuryModel, Journal of the American College of Cardiology, 29(4): 806–816.

21. Kanellakopoulou, K. and Giamarellos-Bourboulis, E.J. (2000). CarrierSystems for the Local Delivery of Antibiotics in Bone Infections, Drugs,59(6): 1223–1232.

22. Yang, J.M., You, J.W., Chen, H.L. and Shih, C.H. (1996). CalorimetricCharacterization of the Formation of Acrylic Type Bone Cements,J. Biomed. Mater. Res., 33(2): 83–88.




23. Price, J.S., Tencer, A.F., Arm, D.M. and Bohach, G.A. (1996). ControlledRelease of Antibiotics from Coated Orthopedic Implants, Journal ofBiomedical Materials Research, 30: 281–286.

24. Gollwitzer, H., Ibrahim, K., Meyer, H., Mittelmeier, W., Busch, R. andStemberger, A. (2003). Antibacterial Poly(D,L-lactic acid) Coating ofMedical Implants using a Biodegradable Drug Delivery Technology,J. Antimicrobial Chemotherapy, 51: 585–591.

25. Lucke, M., Schmidmaier, G., Sadoni, S., Wildemann, B., Schiller, R., Haas,N.P. and Raschke, M. (2003). Gentamicin Coating of Metallic ImplantsReduces Implant-related Osteomyelitis in Rats, Bone, 32: 521–531.

26. Schmidt, C., Wenz, R., Nies, B. and Moll, F. (1995). Antibiotic in vivo/in vitroRelease, Histocompatibility and Biodegradation of Gentamicin ImplantsBased on Lactic Acid Polymers and Copolymers, J. Controlled Release,37: 83–94.

27. Gursel, I., Korkusuz, F., Turesin, F., Alaeddinoglu, N.G. and Hasirci, V.(2001). In vivo Application of Biodegradable Controlled Antibiotic ReleaseSystems for the Treatment of Implant-related Osteomyelitis, Biomaterials,22: 73–80.

28. Zhang, X., Wyss, U.P., Pichora, D. and Goosen, M.F.A. (1994).Biodegradable Controlled Antibiotic Release Devices for Osteomyelitis:Optimization of Release Properties, J. Pharm. Pharmacol., 46(9): 718–724.

29. Pahkla, E.R., Koppel, T., Saag, M. and Pahkla, R. (2005). MetronidazoleConcentrations in Plasma, Saliva and Periodontal Pockets in Patients withPeriodontitis, Journal of Clinical Periodontology, 32: 163–166.

30. Lamp, K.C., Freeman, C.D., Klutman, N.E. and Lacy, M.K. (1999).Pharmacokinetics and Pharmacodynamics of NitronidazoleAntimicrobials, Clinical Pharmacokinetics, 36(5): 353–373.

31. Perioli, L., Ambrogi, V., Rubini, D., Giovagnoli, S., Ricci, M., Blasi, P. andRossi, C. (2004). Novel Mucoadhesive Buccal Formulation ContainingMetronidazole for the Treatment of Periodontal Disease, Journal ofControlled Release, 95(3): 521–533.

32. Golomb, G., Friedman, M., Soskolne, A., Stabholz, A. and Sela, M.N. (1984).Sustained Release Device Containing Metronidazole for Periodontal Use,Journal of Dental Research, 63(9): 1149–1153.

33. Austin, M.N., Meyn, L.A. and Hillier, S.L. (2006). Susceptibility of VaginalBacteria to Metronidazole and Tinidazole, Anaerobe, 12(5–6): 227–230.

34. Wu, Y. and Fassihi, R. (2005). Stability of Metronidazole, Tetracycline HCland Famotidine Alone and in Combination, International Journal ofPharmaceutics, 290(1–2): 1–13.

35. Chourasia, M.K. and Jain, S.K. (2004). Design and Development ofMultiparticulate System for Targeted Drug Delivery to Colon, DrugDeliver, 11(3): 201–217.

36. Varshosaz, J., Tavakoli, N. and Saidian, S. (2002). Development andPhysical Characterization of a Periodontal Bioadhesive Gel ofMetronidazole, Drug Delivery, 9(2): 127–133.

37. Perioli, L., Ambrogi, V., Rubini, D., Giovagnoli, S., Ricci, M., Blasi, P. andRossi, C. (2004). Novel Mucoadhesive Buccal Formulation Containing




Metronidazole for the Treatment of Periodontal Disease, Journal ofControlled Release, 95(3): 521–533.

38. Murata, Y., Kofuji, K. and Kawashima, S. (2003). Preparation of FloatingAlginate Gel Beads for Drug Delivery to the Gastric Mucosa, Journal ofBiomaterials Science, Polymer Edition 2003, 14(6): 581–588.

39. Sampath, S.S. and Robinson, D.H. (1990). Comparison of New and ExistingSpectrophotometric Methods of Analysis of Tobramycin and OtherAminoglycosides, J. Pharm. Sci., 79: 428–431.

40. Lewis, D.D. (1990). Controlled Release of Bioactive Agents from Lactide/Glycolide Polymers, In: Chasin, M. and Langer, R.S. (eds), BiodegradablePolymers as Drug Delivery Systems, pp. 1–41, Marcel Dekker, New York.




structured drug-eluting bioresorbable films: …meitalz/articles/a19.pdfto the engineering of new...

Documents