Biomaterials 22 (2001) 701}708
Bioactivity in glass/PMMA composites used as drug delivery system D. Arcos, C.V. Ragel, M. Vallet-RegmH * Departamento de Qun& mica Inorga& nica y Bioinorga& nica, Facultad de Farmacia, Universidad Complutense de Madrid, 28040 Madrid, Spain Received 14 February 2000; accepted 4 July 2000
Abstract Gentamicin sulfate has been incorporated in composites prepared from a SiO }CaO}P O bioactive glass and polymethylmethac rylate. Data showed that these materials could be used as drug delivery system, keeping the bioactive behavior of the glass. The composites supply high doses of the antibiotic during the "rst hours when they are soaked in simulated body #uid (SBF). Thereafter, a slower drug release is produced, supplying &maintenance' doses until the end of the experiment. The gentamicin release rate is related with the ionic Ca> and H O> exchange between composite and SBF. The porous structure of the composites allows the growth of hydroxycarbonate apatite on the surface and into the pores. 2001 Elsevier Science Ltd. All rights reserved. Keywords: Bioactive glasses; PMMA}glass composites; Gentamicin release; In vitro bioactivity
1. Introduction The study of biomaterials suitable to be used as "lling bone are one of the most interesting "elds in orthopedic surgery [1}3]. These materials are needed to "ll the defects of dead spaces caused by surgical intervention over traumatized or damaged bone. Porous apatites, b-TCP, biphasic ceramics (OHAp-b-TCP) and bioactive glasses are often used, due to their excellent biocompatibility and integration with the osseous tissue [4}7]. An important trouble associated with the use of materials for bone-"lling is the osteomielitis incidence. Techniques for its treatment include the systemic antibiotic administration, surgical debridement, wound drainage, and implant removal is, therefore, essential for the prevention of further complication, such as loss of function and septicemia [8]. Local drug release in the implanted site appears to be a very interesting alternative. The possibility of introducing drug release systems into the implant site has been widely studied and used. Beads of PMMA containing gentamicin has been one of the "rst systems used [9,10]. Later, alternative systems such as biodegradable mate-
* Corresponding author. Tel.: #34-91-394-1861; fax: #34-91-3941786. E-mail address:
[email protected] (M. Vallet-RegmH ).
rials [11], bioceramics [12] or ceramic/polymer composites [13] were developed. Antibiotics [14}16], growth factor and hormones [17}19], chemiotherapeutic agents [12], antistrogens [20], antiin#ammatory drugs [21}23], etc., have been introduced into the systems mentioned above. The bibliography shows that the systems developed for drug release have been very numerous. However, there have been only a few studies about "lling bone materials showing simultaneously controlled drug release and bioactive behaviour [24]. Actually, it seems to be a very attractive idea to look for materials that could release an antibiotic in a local and controlled way while showing bioactive properties. These materials would prevent infections and also would ensure the bone integration and regeneration. In this work we have synthesized glass/acrylic cement composites, and gentamicin sulfate has been added. These materials are formed by a discrete phase (glass) and a continuous phase (PMMA) in which gentamicin is included. The PMMA hydrophobicity should avoid the instantaneous gentamicin release from the composite to the environment, while the glass should supply the bioactive behaviour and release adequate doses of gentamicin during the "rst hours after implantation. These composites, made up of these components, can represent an advantageous solution to solve problems of bone "lling as well as bone regeneration.
0142-9612/01/$ - see front matter 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 0 ) 0 0 2 3 3 - 7
702
D. Arcos et al. / Biomaterials 22 (2001) 701}708
2. Experimental The glass was prepared by hydrolysis and polycondensation of tetraethyl orthosilicate (TEOS), triethyl phosphate (TEP) and Ca(NO ) ) 4H O to obtain a nominal composition (mol%) of SiO (58), CaO (36) and P O (6) as it is described by Vallet-RegmH et al. [25]. The stabilization treatment was carried out by means of heating the dried gel at 7003C for 3 h. The stabilized glass was ground and sieved, selecting the fraction from 32 to 63 lm. Particle size and particle size distribution were determined with a SediGraph 5100 Particle Size Analyzer. Particles were dispersed in a 0.1% calgon water solution and measurements were carried out at 353C. The mean particle size was 22 lm. Data collected are presented in Table 1. PMMA and MMA were products of Aldrich, and Braun MEDICA S.A. kindly supplied gentamicin sulfate. Glass/PMMA/gentamicin composites in a ratio (wt%) of 55 : 32 : 13, respectively, were synthesized by mixing of the components and radical polymerization of MMA in PMMA. Glass (4 g) and gentamicin sulfate (1 g) were "nely mixed. The mixture was added to a solution of PMMA (2 g) in MMA (4 ml) with benzoyl peroxide (0.5 wt%) as reaction starter. This excess of MMA when compared with the nominal composition was added to compensate the loss of MMA during the polymerization process. The amount of MMA added was estimated by thermogravimetric analysis of previous experiences. This mixture was homogenized in an ultrasonic bath for 5 min, and the resulting paste was introduced into barshaped molds of Te#on of 55;8;4 mm. Polymerization was carried out at 603C for 24 h. No activating agent was used and the residual benzoyl peroxide was estimated to be less than 0.2 wt%. Thermogravimetric analyses were carried out over different parts of the obtained bar under air atmosphere and at heating ratio of 53C min\, with a Seiko TG/ATD 320 thermobalance.
complex in the gentamicin determination. Therefore, NaOH was used to "t pH"7.32. Gentamicin release measurements were carried out by means of UV}VIS spectroscopy with a Beckman DU-7 spectrophotometer using the o-phthaldialdehyde method [26]. Absorbance values were taken at a wavelength j"331 nm, at which the gentamicin}phthaldialdehyde complex shows an absorbance maximum. Calibration curve of gentamicin}phthaldialdehyde complex was determined by taking absorbance vs gentamicin concentration between 0 and 30 lg ml\ as parameters. For this interval, the calibration curve "ts the Lambert and Beers' law A"0.02714;C,
(1)
where A is the absorbance and C is the concentration (lg ml\). 2.2. Assessment of in vitro bioactivity The assessment of in vitro bioactivity was carried out by soaking pieces of 10;8;4 mm, vertically held over platinum holders, in 50 ml SBF at 373C in sterile polyethylene containers. The SBF was previously "ltered with a Millipore 0.23 lm system and all operations were carried out in a laminar #ux cabin (Telstar AV-100), avoiding bacterial contamination. The Ca> concentration and pH changes produced in the SBF solution were determined with an Ilyte Na>, K>, Ca>, pH system. Composite surfaces were studied by X-ray di!raction (XRD), Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM) coupled to dispersive energies spectroscopy (EDS), transmission electron microscopy (TEM) and electron di!raction (ED). XRD patterns were obtained with a Philips X'Pert di!ractometer using Cu K radi? ation. FTIR were obtained with a Nicolet Magna-IR 550 spectrometer. The SEM}EDS studies were carried out in a JEOL 6400-LINK AN 10000 microscope, and TEM}ED in a JEOL 200 FX microscope.
2.1. Study of in vitro release For the study of gentamicin release the bars were cut into parallelepiped-shaped pieces of 10;8;4 mm and 280 mg and soaked in 10 ml of simulated body #uid (SBF) at 373C. SBF was modi"ed since one of its components, tris-(hydroxymethyl aminomethane), interferes with the Table 1 Equivalent spherical diameter as a function of cumulative mass percent of the glass particles Cumulative mass percent Equivalent spherical diameter (lm) Mean particle size"22 lm.
100 63
80 29
50 18
20 4
3. Results Fig. 1 shows one of the thermograms obtained for the composite. All thermograms showed an important mass loss between 240 and 4003C due to the decomposition of gentamicin sulfate and PMMA. From 400 to 5703C the mass loss is lower, remaining stable from 5703C until "nal temperature. At the end of the analysis, a mass percentage of 52}55% corresponding to the glass of the composite was observed in all the experiments. Di!erential thermal analysis (DTA) showed an endothermic process at 2443C corresponding to the gentamicin sulfate melting. Two exothermic maxima are also observed: "rst at 3323C corresponding to the
D. Arcos et al. / Biomaterials 22 (2001) 701}708
703
dialdehyde complex concentration as a function of soaking time (Fig. 2), as it is explained in the experimental section. A fast release of the antibiotic can be observed during the "rst 15 h after soaking, reaching gentamicin release values of 62%. Later, a slower release stage occurs, releasing 80% of the drug after 48 h. From this point, the process becomes slower, obtaining values of 90% of released gentamicin after 14 days. The fraction of gentamicin released versus the square root of time can be "tted to a third-order polynomial, corresponding with the model proposed by Cobby et al. [27] for this kind of system (heterogeneous, insoluble matrix-partially in this case*with pores and canals, and parallelepiped-shaped with all the surfaces exposed to the medium). The equation that describes this behavior is (a b #a b #b c ) (a #b #c ) K t ! Kt f" R < < Fig. 1. TG/DTA diagrams of the synthesised composites.
decomposition of gentamicin sulfate and, second at 3873C corresponding to the maximum of PMMA decomposition, although this process is expected to start at about 3003C. The assignations of these maxima are based on previous DTA carried out over the gentamicin sulfate. DTA of the drug showed endothermic and exothermic maxima at 244 and 3323C, respectively. 3.1. Study of in vitro release The gentamicin release from the composite to the SBF was determined by measuring the gentamicin}phthal-
1 # Kt , < where f is the fraction of drug released at time (t), K is R the boundary retreat rate constant, a , b , and c are the parallelepiped dimensions, and < is the parallelepiped volume. The value of the release rate constant, K , obtained from "tting the experimental values to the equation for the "rst 12 days results is K "0.441$0.04 mm h\ . Fig. 3 shows TG/DTA thermograms obtained of the sample after the release studies. TG diagrams show a mass loss of about 15% between 25 and 1003C, as a consequence of H O vaporization. A fast mass
Fig. 2. Gentamicin released from composites as a function of soaking time in SBF. The inset shows a magni"cation of the "rst 24 h.
704
D. Arcos et al. / Biomaterials 22 (2001) 701}708
Fig. 3. TG/DTA diagrams of the synthesized composites after 14 days in SBF.
loss occurs from 3003C corresponding to the PMMA decomposition, obtaining a "nal residual mass of 47%. DTA diagrams show the exothermic signal of the PMMA decomposition between 300 and 4403C, with the maximum shifted at 4303C.The endothermic signal at 2443C corresponding to the gentamicin sulfate melting or the maximum at 3323C corresponding to its decomposition are not observed. 3.2. Assessment of in vitro bioactivity 3.2.1. Growth of an apatite-like layer Fig. 4 shows the changes produced in the SBF during the soaking of the composites. A continuous Ca> release from composite to SBF is observed and two di!erent kinetics occur: a fast release of Ca> until the "rst 2 days and a second stage of slower Ca> release. After 24 days, the composites kept on releasing Ca> to SBF. The pH changes produced in the SBF as a function of soaking time are plotted in Fig. 4b. Simultaneously with the Ca> release, a fast pH increase occurs during the "rst two days from 7.32 to 7.62. Later, a more gradual pH increase takes place, reaching values of 7.85 after 24 days. Fig. 5 shows FTIR spectra obtained from the surface of the samples before and after soaking in SBF. Spectra obtained for composites before soaking show the vibrational bands associated to the functional groups of the components. Spectra obtained from the composites soaked for 1, 3, 7 and 14 days show an intensity enhancement of the bands at 1085, 955, 603 and 565 cm\,
Fig. 4. Changes produced in SBF as a function of soaking time: (a) variation of Ca> content, and (b) variations of pH.
corresponding to the phosphate groups [28], and bands at 1535, 1441 and 872 cm\ corresponding to carbonate groups [29]. The intensity increase of such bands is associated with the soaking time in SBF. XRD patterns of the as-cast composites show that it is an amorphous material. When composites are in contact with SBF for di!erent periods (Fig. 6), a wide and weak maximum could be observed which could be assigned to the (2 1 1) re#ection (the most intense) of an apatite-like phase. Data obtained by means of FTIR and XRD point to the formation of HCA on the surface of the composites when they are soaked in SBF. However, in order to con"rm this point, characterization by other techniques such as SEM, EDS, TEM and DE is necessary. 3.2.2. SEM study Micrographs obtained from composites before and after soaking in SBF are collected in Fig. 7. The micrograph of the glass before soaking (Fig. 7a) shows a heterogeneous surface formed by the composite components. After soaking for 1 day (Fig. 7b) the growth of white particles of less than 1 lm size on the composite surface can be observed. After 3 days (Fig. 7c) the composite surface appears to be covered by a layer. This layer is formed by 2 lm rounded particles, which are constituted by acicular sub-particles. After 7 days (Fig. 7d) the layer turns to be a more compact one and with the rounded particles enhanced in size. After 14 days
D. Arcos et al. / Biomaterials 22 (2001) 701}708
705
Fig. 5. FTIR spectra obtained of the composite surfaces before and after soaking in SBF.
Fig. 7. SEM micrographs of the composites before and after soaking in SBF for di!erent periods. Original magni"cation ;5000.
Fig. 6. XRD patterns of the composites before and after soaking in SBF for di!erent periods. An XRD pattern of hydroxyapatite is included for reference purposes.
(Fig. 7e), this situation becomes even clearer, with the enhancement of both the volume of rounded particles and acicular sub-particles size.
EDS spectra obtained for the sample surfaces before and after soaking for 7 days in SBF are shown in Fig. 8. After 7 days of treatment, the silicon line almost disappears and the Ca/P rate is approximately 1.23. The line corresponding to sulfur appears in the spectrum before soaking in SBF due to gentamicin sulfate. After 7 days of soaking, such line does not appear, con"rming the drug release from the composite. The apatite-like layer does not grow only on the external surface of the composites, that is, the surface which is in direct contact with the SBF. Fig. 9 shows a
706
D. Arcos et al. / Biomaterials 22 (2001) 701}708
Fig. 10. SEM micrograph of a cross section of the composite after soaking in SBF for 14 days. Original magni"cation ;1000.
Fig. 8. EDS patterns of the composite surface before and after soaking in SBF for 7 days.
Fig. 11. (a) TEM micrograph of a particle scrapped from the composite surface after 7 days in SBF. (b) ED pattern of the particle.
The transverse section of the piece allows to view the thickness of the layer formed on the surface. Fig. 10 shows that the growth layer has a thickness of 5}10 lm after being soaked for 14 days. Fig. 9. SEM micrograph of a cross section of the composite soaked in SBF for 14 days. The arrow indicates a big pore covered by a new grown phase. Original magni"cation ;200.
micrograph obtained of a transverse section of the sample after soaking for 14 days in SBF. Growth of this new-formed phase into the inner part of the piece can be observed. This growth is produced at those sites in the material where large macropores exist with access to SBF. Besides these macropores, a high concentration of Ca> is released leading to the apatite growth, similar to the way it is produced on the external surface. This mechanism is in agreement with the mechanism proposed by Kokubo [30].
3.2.3. TEM and ED studies In order to perform a more complete characterization of these particles, ED and TEM have been carried out. Fig. 11a shows a highly magni"ed micrograph of a particle, taken from the composites after 7 days in SBF. As can be seen, it is formed by hundreds of small crystallites with a size of about 20 nm. The electron di!raction pattern shown in Fig. 11b corresponds to a typical polycrystalline material. Measuring the diameter of the rings observed in Fig. 11b, and taking into account the camera constant, the interplanar spacing of such rings was calculated: 3.65, 2.88, 1.99, 1.78 As that correspond to the (0 0 2), (2 1 1), (2 2 2) and (2 1 3) re#ections of an apatitelike phase.
D. Arcos et al. / Biomaterials 22 (2001) 701}708
4. Discussion The gentamicin concentration data in SBF as a function of soaking time shows that drug release is produced fundamentally in a "rst stage, in which after 48 h 80% of gentamicin is released. Subsequently, the release rate decreases and after 14 days, 90% of the drug load is present in the SBF. EDS and TG/DTA analyses con"rm the process of gentamicin release after 14 days in SBF. EDS spectra obtained after 7 days of soaking do not show the line corresponding to sulfur. On the other hand, DTA diagrams did not show the endothermic signal at 2443C corresponding to gentamicin sulfate fusion. PMMA decomposition temperature is sited between 300 and 4403C, but the maximum changes from 3873C before the drug release, to 4303C after the experiment. These data indicate an interaction drug-PMMA, since the presence of both components in the composite leads to changes in the decomposition temperature of the PMMA without drug. The drug release process shows a very similar kinetics to those of Ca> ions release and the pH increase. These data seem to point out that drug release is not a single di!usion process, since ionic exchange between the composite and SBF also in#uences it. The Ca> di!usion and pH enhancement are in agreement with an apatite-like phase growth over materials containing SiO }CaO in their composition. In such mate rials, an ionic exchange between Ca> of material and H O> of the media takes place. This exchange leads to the formation of silanol groups (Si}OH) on the surface, leading to the apatite nucleation. The subsequent layer growth is produced at the cost of the ions contained into the solution. FTIR data show a clear intensity enhancement of the bands corresponding to CO\ and PO\ with the soaking time in SBF. FTIR data together with XRD patterns point out that hydroxycarbonate apatite (HCA) grows on the surface of our composites. However, in order to carry out a more exhaustive structural analysis, complementary techniques such as SEM, EDS, TEM and ED are necessary. Correlating the SEM and TEM results, we can conclude that the grown particles observed by SEM are crystalline aggregates formed by several thousands of nanocrystallites of about 20 nm in size. ED diagrams would be coherent with the existence of an apatite-like phase. Data obtained by XRD and FTIR, combined with the chemical changes in SBF, reveal that the initial nucleation sites of the new apatite-like phase take place during the "rst 24 h. After 3 days, this phase covers the entire composite surface.
5. Conclusions (1) Glass/polymer/gentamicin composites suitable to be used as biomaterials have been synthesized.
707
(2) The composites supply high doses of the antibiotic during the "rst hours when soaked in SBF. Thereafter, a slower drug release is produced, supplying a &maintenance' doses until the end of the experiment. The gentamicin release rate is related with the ionic Ca> and H O> exchange between composite and SBF. (3) Nanocrystalline HCA grows on the composite surface when it is in contact with SBF. The presence of an acrylic cement and gentamicin does not inhibit the bioactive behavior of the glass. Therefore, good bone integration is expected.
Acknowledgements Financial support of CICYT, Spain, through research projects MAT98-0746C02-01 and MAT99-0466 is acknowledged. We also thank A. RodrmH guez (Electron Microscopy Center, Complutense University), and F. Conde (C.A.I. X-ray Di!raction, Complutense University) for valuable technical and professional assistance. B. Braun Medical SA kindly supplied gentamicin sulfate.
References [1] Kohn DH. Structure-property relations of biomaterials for hard tissue replacement. In: Wise D, Reantolo D, Altobelli DE, Yaszemski M, Gresser JD, Schwartz ER, editors. Encyclopedic handbook biomaterials and bioengineering. New York: Marcel Dekker, 1995. p. 83}122. [2] Shapo! CA, Alexander DC, Clark AE. Clinical use of bioactive glass particulate in the treatment of human osseus defects. Compend Contin Educ Dent 1997;18:352}63. [3] Itokazu M, Matsunaga T, Ishii M, Yang W. Use of arthroscopy and interporous hydroxiapatite: a bone substitute an tibial plateau fractures. Arch Orthop Trauma Surg 1996;115:45}8. [4] Shors EC, Holmes RE. Porous hydroxyapatite. In: Hench LL, Wilson J, editors. An introduction to bioceramics. Singapore: World Scienti"c, 1993. p. 181}98. [5] Jarcho R, Salsbury RL, Thomas MB, Doremus RH. Synthesis and fabrication of b-tricalcium phosphate (whitlockite) ceramics for potential prosthetic applications. J Mater Sci 1979;14:142}50. [6] Daculsi G, Goyenvalle E, Aguado E. Spongious and cortical bone substitution kinetics at the expense of macroporous biphasic calcium phosphate: animal and human evidence. In: Ohgushi H, Hastings GW, Yoshikawa T, editors. Bioceramics, vol. 12. Nara: World Scienti"c, 1999. p. 287}90. [7] Hench LL. Bioceramics: from concept to clinic. J Am Ceram Soc 1991;74:1487}510. [8] Bucholz HW, Heinet K, Foerster GW. Infected prostheses: the role of antibiotic cement. In: D'Ambrosia RD, Marier RL, editors. Orthopaedic infections. New Jersey: Slack Inc, 1989. p. 477}88. [9] Wahlig H, Dingeldein E, Bergmann R, Reuss K. The release of gentamicin from polymethylmethacrylate beads. An experimental and pharmacokinetic study. J Bone Jt Surg Br 1978;60-B(2):270}5. [10] Vecsei V, Barquet A. Treatment of chronic osteomyelitis by necrectomy and gentamicin}PMMA beads. Clin Orthop 1981; 159:201}7.
708
D. Arcos et al. / Biomaterials 22 (2001) 701}708
[11] Yagmurlu MF, Korkusuz F, GuK rsel I, Korkusuz P, OG rs UG , Hasirci V. Sulbactam-cefoperazone polyhydroxybutarate-co-hydroxyvalerate (PHBV) local antibiotic delivery system: in vivo e!ectiveness and biocompatibility in the treatment of implantrelated experimental osteomyelitis. J Biomed Mater Res 1999; 46:494}503. [12] Itokazu M, Sugiyama T, Ohno T, Wada E, Katagiri Y. Development of porous apatite ceramic for local delivery of chemotherapeutic agents. J Biomed Mater Res 1998;39:536}8. [13] Vallet-RegmH M, Gordo M, Ragel CV, Caban as MV, San RomaH n J. Synthesis of ceramic}polymer}drug biocomposites at room temperature. Solid State Ionics 1997;101}103:887}92. [14] Van Wachem PB, Van Luyn MJA, de Wit AW, Raatjes D, Hendriks M, Verhoeven MLPM, Cahalan PT. Tissue reactions to bacteria-inoculated rat lead samples. I. E!ect of local gentamicin release through vicinal sponge of solution-diping. J Biomed Mater Res 1997;35:217}32. [15] Takechi M, Miyamoto Y, Ishikawa K, Nagayama M, Kon M, Asaokaand K, Suzuki K. E!ects of added antibiotics on the basic properties of anti-washout-type fast-setting calcium phosphate cement. J Biomed Mater Res 1998;39:308}16. [16] Fujimoto K, Yamamura K, Osada T, Hayashi T, Nabeshima T, Matsushita M, Nishikimi N, Sakurai T, Nimura Y. Subcutaneus tissue distribution from a "brin glue/Dacron graft carrier. J Biomed Mater Res 1997;36:564}7. [17] Nicoll SB, Radin S, Santos EM, Tuan RS, Ducheyne P. In vitro release kinetics of biologically active transforming growth factor-b1 from a novel porous glass carrier. Biomaterials 1997;18:853}9. [18] Goodwin CJ, Braden M, Downes S, Marshall NJ. Investigation into the release of bioactive recombinant human growth hormone from normal and low-viscosity poly(methylmethacrylate) bone cements. J Biomed Mater Res 1997;34:47}55. [19] Guicheux J, Grimandi G, TreH cant M, Faivre A, Takahashi S, Daculsi G. Apatite as carrier for growth hormone: in vitro characterization of loading and release. J Biomed Mater Res 1997; 34:165}70.
[20] Kortesuo P, Ahola M, Karlsson S, Kangasniemi I, Kiesvaara J, Yli-Urpo A. Sol}gel-processed sintered silica xerogel as a carrier in controlled drug delivery. J Biomed Mater Res 1999;44:162}7. [21] Vallet-RegmH M, Granado S, Arcos D, Gordo M, Caban as MV, Ragel CV, Salinas AJ, Doadrio AL, San RomaH n J. Preparation, characterization, and in vitro release of ibuprofen from Al O /PLA/PMMA composites. J Biomed Mater Res 1998;39:423}8. [22] Arcos D, Caban as MV, Ragel CV, Vallet-RegmH M, San RomaH n J. Ibuprofen release from hydrophilic ceramic}polymer composites. Biomaterials 1997;18:1235}42. [23] Granado S, Ragel V, Caban as V, San RomaH n J, Vallet-RegmH M. In#uence of a-Al O morphology and particle size on drug re lease from ceramic/polymer composites. J Mater Chem 1997; 7(8):1581}5. [24] Ragel CV, Vallet-RegmH M. In vitro bioactivity and gentamicin release from glass-polymer}antibiotic composites. J Biomed Mater Res 2000;51:424}9. [25] Vallet-RegmH M, Arcos D, Pe`rez-Pariente J. Evolution of Porosity during in vitro hydroxycarbonate apatite growth in sol}gel glasses. J Biomed Mater Res 2000;51:23}8. [26] Sampth SS, Robinson DH. Comparison of new and existing spectrophotometric methods for the analysis of tobramicin and other aminoglycosides. J Pharm Sci 1990;79:428}31. [27] Cobby J, Mayerson M, Walker GC. In#uence of shape factors on kinetics of drug release from matrix tables. I. Theoretical. J Pharm Sci 1974;63:725}32. [28] Fowler BO. Infrared studies of apatites. I. Vibrational assignements for calcium, strontium and barium hydroxiapatites utilising isotopic substitution. Inorg Chem 1974;13:194}207. [29] Rey C, Collins B, Goehl T, Dickson IR, Glimcher MT. The carbonate environment in bone mineral: a resolution-enhanced Fourier transform infrared spectroscopy study. Calcif Tissue Int 1989;45:157}64. [30] Kokubo T, Kushitani H, Sakka S, Kitsugi T, Yamamuro T. Solutions able to reproduce in vivo surface-structure changes in bioactive glass-ceramic A-W. J Biomed Mater Res 1990; 24:721}34.
