You are on page 1of 8

Biomaterials 23 (2002) 2519–2526

Controlled release of rhodium (II) carboxylates and their association complexes with cyclodextrins from hydroxyapatite matrix
A.E. Burgos, J.C. Belchior, R.D. Sinisterra*
Departamento de Qu! ımica-ICEx, Universidade Federal de Minas Gerais, Pampulha, (31.270-901), Belo Horizonte, MG, Brazil Received 21 February 2001; received in revised form 5 November 2001; accepted 7 November 2001

Abstract Preparation and characterization of a controlled release system of rhodium (II) citrate, acetate, propionate, butyrate and their inclusion or association compounds with cyclodextrin (CD) are described. The porous hydroxyapatite (HA) was characterized by X-ray powder pattern diffraction, FTIR and solid state 31P NMR. Scanning electron microscopy and gas adsorption analysis (BET) were also performed. Release profiles of rhodium (II) carboxylates and their inclusion or association compounds from HA matrix were obtained at different drug loadings (5% and 10%). These were reasonably consistent with a diffusion model. This analysis, mainly using rhodium (II) citrate and butyrate, showed that the strategy of using CDs with a HA matrix may offer a useful new method for the controlled release of these compounds, and hence an alternative strategy for the controlled release of chemotherapeutic agents containing toxic metals. This may be a valuable new technique for localized anti-tumour chemotherapy that minimizes the side effects of such agents. r 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Calcium phosphate; Hydroxyapatite; Drug delivery systems; Controlled release; Rhodium (II) carboxylates; Cyclodextrins

1. Introduction Considerable attention has been focused in order to develop new strategies for drug-controlled release systems, based on biodegradable polymers and also on bioceramics [1–9]. These efforts have been widely applied to improve the therapeutic effects of the drugs in different pathologies. For example, biodegradable polymers have been used as matrices for the delivery of vaccines [10] and anti-tumor [11,12], antibiotic [13] and anti-inflammatory agents [14]. Bioceramic materials are also used as matrices for drug delivery [8,9,11,15]. Cyclodextrins (CDs) offer a third strategy for the efficient sustained release of drugs [16–18]. Rhodium (II) carboxylates are amongst the most promising of the second generation platinumgroup anti-cancer compounds [19–22]. Howard et al. [23] have noted a correlation between the hydrophobicity of the rhodium (II) carboxylates and their anti-tumor activity and toxicity. On increasing the alkyl chain length of the rhodium (II) carboxylates, one
*Corresponding author. Fax: +55-31-3499-5700. E-mail address: (R.D. Sinisterra).

observes increasing activity but also toxicity, in tumor model systems [24]. The toxicity increase is one of the most important drawbacks of using these compounds in chemotherapy. CDs are oligosaccharides that can modify the physico-chemical and biological properties of drugs [25–29]. They therefore have considerable potential, and possibility for improving the activity of drugs [27]. They can modify the partition coefficient of, e.g. rhodium (II) carboxylates so as to diminish the toxicity or to increase the activity of these complexes. Moreover, apatites such as hydroxyapatite (HA) can be used as promising vehicles for drug delivery [9]. As reported by Netz et al. [8], the advantage of using HA as a matrix for drug delivery systems mainly arises from their low cost, ease of manufacture, biocompatibility, osteoconductivity and ostheophilicity. This paper deals with the evaluation of porous HA as a delivery system for rhodium (II) carboxylates and their inclusion compounds with CDs. It aims to develop a valuable form of local chemotherapy when used as grafts to repair bone defects, as well as a means of reducing the systemic toxicity of these metal complexes [11,15]. Finally, the burst effect of these compounds is

0142-9612/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 2 - 9 6 1 2 ( 0 1 ) 0 0 3 8 6 - 6

The identity of HA in the samples was confirmed by comparison with the peaks from HA or other phases that might occur with the ICDD cards for HA (9-432). Burgos et al.02 ppm plus weak sidebands. HA pellet preparation and release studies The HA pellets were prepared as follows: 5.36–37] and their association complexes with b-CD can be found elsewhere [38].2520 A. respectively. 2. as used by Rusin et al. 600. IR analysis gave more details about the phase composition of HA and confirmed that the samples only have a single HA phase.06 mol/l). Synthesis of HA HA samples were synthesized by pyrolysis from a precursor solution using a modification of published methods [32–35]. Ca and P contents of the HA were assayed using SEM microprobe. Characterization of the solid complexes X-ray powder diffraction (XRD) patterns were recorded on a Rigaku 2037 X-ray diffractometer using Cu Ka radiation (l ¼ 1:5405 nm). Nazarova et al.d. In addition. The IR spectra were recorded on Mattson Galaxy 3000 FTIR spectrometers in the range of 4000– 400 cmÀ1. a characteristic absorption of the rhodium (II) carboxylates. All samples were analyzed in the triplicate. Analytical grade CaCl2 Á 2H2O and (NH4)H2PO4 (Merck) were the starting reagents. Our sample showed a surface area equal to 4. Thus. [39] and Yesinowski et al. Materials and methods 2. The visible absorption was measured at 590 nm.5 A . [40]. 2. This spectrum has only one 31P resonance at 7. The HA samples were then sintered at 10001C for 2 h in air.1 mol/l) was dissolved in 500 ml of water and this solution was added to an aqueous solution of (NH4)H2PO4 (0. 2. A complementary micropores of about 8.1. This showed all the samples to be a one-phase composite of HA without any admixture of other phosphates. respectively. with H3PO3 as the internal reference. These samples were prepared in a 1 Â PBS phosphate buffer solution using a 1 cm quartz sample cell.2.E. The expected bands at 3581 cmÀ1 are usually masked in the spectra by H2O vibrations between 3300 and 3568 cmÀ1 [39]. 988. 1050 cmÀ1 with a shoulder at 960 cmÀ1.3. In addition. The rhodium (II) carboxylates and their association complexes released were quantified using visible spectroscopy (Beckman DU-640) and inductively coupled mass spectrometry (ICP-MS). Characterization of the HA matrix The phase composition is usually considered to be one of the most important parameters of HA ceramics because it determines their biocompatibility [8]. 3. Results and discussion 3. The rhodium (II) carboxylates were prepared by the methods reported in the literature [34.0% or 10% of rhodium (II) carboxylates or their inclusion or association complexes with b-CD were mixed with powdered HA and then compressed at a 4000 ton loading for 5 s. 585 and 569 cmÀ1 were assigned to PO and PO4 vibrations. Details of the preparation and characterization of rhodium (II) complexes are described by Boyar and Robinson. The release medium was periodically changed at predetermined time points throughout the study and replaced with a fresh release buffer. whether or not b-CD is also presented. The release profile for all the loaded HA pellets were determined as follows: a sample pellet was introduced to 2 ml of phosphate buffer release medium PBS (pH 7) and place on a glass vial at 371C. In the present study. These quantities are also used for the preparation of the association complexes with CDs. at pH 12–13. 1050. sapphire rotors and a spinning speed of 5 KHz. 0. All samples were measured as KBr pellets. in good agreement with the XRD patterns of ICDD card 9-432. and by Howard et al.31].35]. The b-CD was purchased from Cerestar Company and used without further purification. and this is in satisfactory agreement with the above references. The crystallographic equivalence of all the phosphate groups of HA was demonstrated by 31P CP/MAS NMR. The porosimetric measurements analysis of the HA materials revealed the presence of numerous micropores. The CDs do not absorb at this wavelength. [24. our samples were prepared at constant rhodium (II) carboxylates loadings. The 31P NMR spectra in the solid state were obtained on a Bruker-400 Avance spectrometer equipped with a magic-angle spinning CP-MAS probe using 7 mm o.1. Calibration curves were established using known concentrations of rhodium (II) carboxylates and their association complexes in 1 Â PBS. bands at 1130. Scanning electron microscopy (SEM-Jeol-JXA 8900RL) and gas adsorption analysis (BET-Nova-1200 quantachrome corporation) were used to study the microstructure and specific surface area of the porous material. the phase composition of the polycrystalline HA was investigated via XRD. / Biomaterials 23 (2002) 2519–2526 analyzed in the light of the model recently used by Narasimhan and Langer [30. Very sharp and wellresolved peaks of HA were observed. The IR HA spectrum had PO4 bands at 659.8 m2/g and ( diameter.05 mol of CaCl2 Á 2H2O (0.

The first part will concentrate on the release of citrate. Structures with different pore sizes can give the products with different permeabilities creating their potential as drug delivery systems [8]. acetate. and after release of rhodium (II) butyrate inclusion complex with b-CD 10% (E).E. 3. The second part considers the burst effect and a diffusion analysis.31]. [30.A. after release of rhodium (II) butyrate 10% (D). analysis through SEM (Fig.02. The HA devices loading with rhodium (II) carboxylates and their association complexes were also characterized by SEM. Burgos et al. 1). The release profiles with 5% loading of rhodium (II) citrate and butyrate complexes show the highest and . / Biomaterials 23 (2002) 2519–2526 2521 Fig. Release analysis The release results will be divided into two parts. This was performed before and after drug release (Fig.2.12 mm with SD of 70. pure HA (C). 1. with 10% of rhodium (II) butyrate inclusion complex with b-CD drug-loaded before release (B). 1) measurements provided both the median particle size and macropores of about 1. propionate and butyrate rhodium (II) carboxylates and their inclusion complexes in CDs. This figure proves that there is a lower drug loading following the release. Scanning electron micrograph of HA loaded with 10% of rhodium (II) butyrate before release (A). based on the model discussed in Refs.

butyrate (black squares). Comparison of the release profile of the CD compounds at 5% (Fig. In contrast. These observations also confirm the relationship between the solubility and the release profile of the complexes. propionate (squares).8 0. which could result in crystallization of the complex blocking its release.4 0. via hydrogen bonding. This may be attributed to its low solubility as well as supersaturation in the microenvironment of the matrix. Burgos et al.8 0. [26]. 2a and b shows the biggest effect of the CDs with the rhodium (II) citrate and butyrate complexes. 3b) with those in Figs. Each point and error bar represents the mean SD of three samples.2 0 0 (a) 1 Fraction of Drug Released 0.2522 A. and citrate (black triangles) of release 5% (a) and (b) 10%.2 0 (b) 50 100 150 200 250 Time (hours) 300 350 400 Fig. respectively.4 0. Each point and error bar represents the mean SD of three samples. In this reference. this compound is released at a rate independent of drug loading (Figs. As described in Ref. 3a) and 10% (Fig. these compounds were tested for anti-tumor activity against different model cancers. 3.2 0 0 (b) 50 100 150 200 250 300 350 400 Time (hours) 50 100 150 200 250 300 350 400 Time (hours) 1 Fraction of Drug Released 0.8 0. this behavior is explained by the possibility of formation of a supramolecular complex. butyrate inclusion complex with b-CD (black dots) and citrate inclusion complex with b-CD (open square): (a) 5% and (b) 10%. an exception is seen with rhodium (II) butyrate whose release does not depend on the drug loading. / Biomaterials 23 (2002) 2519–2526 lowest burst effects. The propionate compound shows intermediate behavior (Fig. The most pronounced effect is observed for the rhodium (II) citrate complex. the resulting solution concentrations are almost 1000 times greater than in Ref.28] showing a higher solubility of poorly soluble compounds after inclusion in CDs.6 0. A higher release of the compounds in the first 6 h of the experiment is observed for the complexes at 10% (Fig. .2 0 0 (a) 1 Fraction of Drug Released 0. It is important to note that the total concentration of rhodium (II) carboxylates released from HA matrices are very higher indeed. at the end (384 h) of the release of the butyrate complex. 2.4 0. 2a). This modulation may be attributed to the differences in the host–guest interaction of the complexes with the CDs. Release profiles of 5% and 10% of rhodium (II) acetate (circles). one observes a higher release for the first 6 h for the rhodium (II) butyrate complex with CD in comparison with the complex without inclusion. after 16 days.E. which shows a drastic decrease of release upon inclu- 1 Fraction of Drug Released 0. for the first 6 h period.6 0. sion. Interestingly. propionate inclusion complex with b-CD (open down triangle).4 0. [27. This result agrees with the observations in Refs.6 0.8 0. [24]. 0 50 100 150 200 250 Time (hours) 300 350 400 Fig.6 0. 3a and b). so as to decrease the hydrophilicity of the metal complex. 2b) drug loading. Release profiles of 5% and 10% of rhodium (II) acetate inclusion complex with b-CD (open up triangle). However. For example. It is well known that a relationship exists between the carboxylate solubility and the burst effect [24] and this explains our observations. The longest time release is observed with the rhodium (II) butyrate complex (16 days).

aÞ. rhodium (II) citrate. [30]. This behavior may be understood by the differences in the host–guest interaction of these systems. ð 2n þ 1Þ 2 p 2 n¼0 ð 5Þ ð 2Þ where J0 ðxÞ and J1 ðxÞ are Bessel functions of orders zero and one. the diffusion is affected by the association or inclusion phenomena when one goes from acetate to butyrate. the second most soluble system (Rh(II) acetate) has a diffusion coefficient almost three times bigger. (5) Cita 5% 10% a b Acb 7. In the present study. This model was developed for slab and/or cylinder shapes.E.1E-05 6. However. butyrate and their association and/or inclusion compounds with cyclodextrins calculated through Eq.0E-04 2. Table 1 gives the diffusion coefficients calculated via Eq.6E-03 Pr-b-CD 2. On the other hand.5E-04 Bud 2.8E-03 8. respectively.0E-04 Prc 3. one finds that the diffusion coefficient of rhodium (II) butyrate is more than three times smaller than those 5% drugloaded. C is the concentration at any time and r and z are defined within the limits 0prpa and Àl pzpl : The quantities a and l are. Although the present drug release device is slightly different from that proposed in Ref. propionate and butyrate together with their respective association complexes with CDs.5E-04 CitFRh(II)-citrate. these parameters are a ¼ 0:5 mm and l ¼ 0:1 mm. / Biomaterials 23 (2002) 2519–2526 2523 4. .5E-04 3. For rhodium (II) citrate. (5). Burst effect analysis Using these kinetic measurements.1E-03 1. c PrFRH(II)-propionate. has its diffusion coefficient decreased by a factor of almost two. The solution of this partial differential equation can be written as C ¼ C0 cðzÞ fðr. d BuFRh(II)-butyrate and b-CD means b-cyclodextrin. the radius and the length of the pellet. aÞ are given by cðzÞ ¼ N 4X ðÀ1Þn ð 2n þ 1Þ p exp½ÀDð2n þ 1Þ2 p2 t=4z2 Š cos p n¼0 ð2n þ 1Þ 2z ð 3Þ and fðr. where cðzÞ and fðr. [11] has pointed out that this kind of Table 1 Diffusion coefficients of rhodium (II) citrate. [30. The drug is released by diffusion from both the planar ends and the cylindrical portions of the pellets. one can obtain where F defines the fraction of drug released. In this case. ð 1Þ qt qr 2 r qr qz 2 where D is the diffusion coefficient. rhodium (II) citrate.0E-04 Bu-b-CD 3.8E-04 1. These results suggest that CDs may be efficient drug carriers. the analysis of the kinetic results can be performed using the same model. acetate. AcFRh(II)-acetate.5E-06 Cit-b-CD 9. in the alkyl carboxylates. there is a strong hydrogen bonding interaction between the free carboxylate groups and the CDs.0E-04 1. The results for 5% of loading in the absence of CD show that the butyrate compound has the lowest diffusion coefficient while the most soluble system. especially for the release of toxic drugs.0E-05 3. This decreases the diffusion coefficient by modifying the partition coefficient of the drug. The HA device has the property that it retains its dimension throughout the drug release. at 10% of loading but only almost 1.A.0E-05 1. These observations may be explained by the difference in their chemical structures. The same trend is also observed in the case of 10% loading. The concentration of drug released can be calculated at any time by solving the following partial differential equation [30]  2  qC q C 1 qC q2 C ¼D þ þ . propionate. Burgos et al. The most soluble system. a burst effect analysis can be performed via the model described in Refs. expðÀDa2 m tÞ a n¼1 am J 1 ð aam Þ the following equation: F ¼1À8 N X expðÀDð2n þ 1Þ2 p2 t=4l 2 Þ . acetate. however. At this higher drug-loading.0E-03 Ac-b-CD 2.5% for 5% of loading. More interesting. relative to two compounds. respectively.31]. The question that one can now address is the determination of the diffusion coefficients for rhodium (II) citrate. has the highest solubility and hence the highest diffusion coefficient. aÞ ¼ N 2X J0 ð r am Þ . are the diffusion coefficient results upon inclusion. Itokazu et al. In addition. The rhodium (II) butyrate system inclusion complex has an increase of almost five times in its diffusion coefficient. One may assume that the HA device is not consumed and hence that the parameter a will be assumed to approach to infinity as the time increases.

As can be seen in Figs 2 and 3. for both loadings.. This behavior might be attributed to the mixture of association and inclusion of this system. Such a strategy could also be used to decrease the toxicity or increase the bioavailability of the rhodium (II) carboxylates in order to improve the application of these compounds in leukemia.4 Fraction of Drug Released 0. Burgos et al. In addition. 4 and 5 provide insights of the burst effect. For both cases.6 0. to reduce the rate of local recurrence of bone cancer. dashed lines. As described in Ref.2 0 0 20 40 60 Time (hours) 80 100 Fig. Finally. Ehrlich ascites. It is interesting to note that the burst effect of rhodium (II) propionate is explained well by the model given in Section 4. which are important in analyzing alternative ways to decrease the toxicity of rhodium (II) complexes. Fitted curves were obtained through Eq.4 0.8 0. 4 and 5 show the release profiles of 5% (top) and 10% (bottom) loading and the calculated diffusion coefficients.8 0. butyrate results. / Biomaterials 23 (2002) 2519–2526 1 0. propionate (black square) and butyrate (black triangle). The results are shown for rhodium (II) citrate. Release profile of 10% (bottom) of rhodium (II) citrate (*). Our results are in agreement with the conclusion pointed out in Ref. as expected. i. the acetate results. acetate.6 0. one can address the role of the burst effect in the kinetic release for these systems. Accordingly.4 Fraction of Drug Released 0. The corresponding diffusion coefficients D are given in Table 1.8 0. Figs. acetate association complex with b-CD (black dot).8 0. dot lines. included and without inclusion. The corresponding diffusion coefficients D are given in Table 1. propionate results and dot dashed lines.4 0.2 0 0. butyrate and their inclusion complexes. 4.6 0. dot lines. one can conclude that the model predicts all the experimental data for rhodium (II) acetate and butyrate. (5). butyrate results. 1 0. using Eq. propionate (black square) butyrate (black triangle). acetate (black dot). is independent of the drug-loaded concentration for rhodium (II) butyrate while it increases for the propionate.2 0 0. the solubility and the diffusion parameters may play an important role in controlling release. (5) where dashed line shows citrate results. The exception of propionate system which shows a higher burst effect when the concentration increases can be explained in the light of the competition between diffusion and solubility of this system as previously discussed. propionate (black square) butyrate (black triangle). the burst .2 0 0 20 40 60 Time(hours) 80 100 technology could be applied as an alternative to cryosurgery with liquid nitrogen. the burst effect.6 0. [30] where the authors showed that the amount of drug release is independent of the diffusion coefficient. the results for the rhodium (II) propionate complex clearly show a quantitative fit only at lower release times. The fitting curves are obtained through Eq. (5) where dashed line shows citrate results. propionate association complex with b-CD (black square). and butyrate inclusion complex with b-CD (black triangle). This could be understood as a competition between the recognition of aliphatic moiety of the metal complex throughout CD cavity and also by increasing the hydrogen bonding discussed above. The results shown in Figs. propionate. Fig.e. dashed lines. Release profile of 10% (bottom) or rhodium (II) acetate (*) acetate (black dot). acetate. [30]. respectively. 5. propionate and dot dashed lines. Release profiles of 5% (top) of rhodium (II) citrate association complex with b-CD (*). Release profile of 5% (top) of rhodium (II) citrate (*).E. acetate (black dot). or sarcoma 180 ascites cancer model systems [22].2524 A.

Following the analysis given in Ref. Young RA. J Control Release 1997. Proc Soc Exp Biol Med 1974. Katagiri Y. Alves de Carvalho. Erck A. Smith ML. Matsuda Y. Hirayama F. Moreover. Prankerd R.7:575. [4] Otsuka M. Fox JL. [30]. [21] Sinisterra RD. [22] Rao PN. 4 and 5 and those diffusion coefficients given in Table 1. the system may be used as an efficient strategy for providing a controlled release system.A. Spadoro ACC. Nakahigashi Y. this contribution showed that the interaction of rhodium (II) carboxylates and CDs increases the diffusion coefficient. [25] Sinisterra RD. Takano M. Sugiyama T. Sugiyama Y. Chem Rev 1998. [30. Leen C. as D increases. US Patent 5.39:536. [9] Slosarczyk A.47:13. Zhu J. Lloyd DK. Our results are in fair agreement with this model as can be seen in Figs. Yotsuyanagi T. Langer R. Nakahigashi Y. In: Lehn JM. J Biomed Mater Res 1998. [8] Netz DJA. Matsuda Y.9:227.13:120. using the condition e51 and using t ¼ e=k. J Mater Chem 1997. such a model showed accurate results for the diffusion coefficients for the short period of time. editors. Whileyman IM. [28] Loftsson T.22:91. J Bone Jt Surg 1992. 1999. J Biomed Mater Res 1992. this technology could be an alternative to cryosurgery with liquid nitrogen to reduce the rate of local recurrence of bone cancer.21:1215. [14] Dhillon B. For rhodium (II) carboxylates.392:5(suppl. Bear JL. J Control Release 1998. Kimball AP.204: 1050. Kimball AP. Brazilian J Med Biol Res 1994. [7] Langer R.E. (6). J Amer Chem Soc 1980. Najjar R. Comprehensive supramolecular chemistry. Atwood JL. Int J SDT AIDS 1998. [23] Howard J. [10] Walduck AK. Brazil.263:1715. as pointed out by Narasimhan and Langer [30].341:206. Irie T. da Silva AC. [27] Szejtli J. Int J Pharm 2001. Robinson SD. Yano H. the release profiles of these compounds have been measured in order to analyze the effect of different drug-loadings and also inclusion of the drugs in CDs. Langer R. Rainen L. Szejtli J. [20] Erck A. [11] Yapp DTT. Fox JL. Mycek B. [3] Yamamura K. one obtains dQ ¼ 2 pC 0 a 3 k : dt ð 8Þ Hence. J Nat Cancer Inst 1980. editors. Uekama K. For short time release. J Radiat Oncol Biol Phys 1997. Bear JL.50:109. one defines a parameter e ¼ kt in Eq. Shastri VP. 5. Clin Orthop Relat Res 1997. dt ð 6Þ Acknowledgements This work was supported by CNPq. [15] Shinto Y. [2] Kay MI.64:905. Mader JT.).39:497. Finally. the analysis of the solubility and diffusion contributions can be performed as given by t¼ e a2 ð 6 þ B Þ : 16D ð 9Þ According to this equation. Bear JL.51:269. Med Chem 1977. Higuchi IW. J Incl Phen 1995. FAPEMIG and UFMG-FUNDEP. [26] Sinisterra RD. Iwata H. Biomaterials 2000.98:2045. Wada E. k¼ 2 a ð6 þ BÞ ð 7Þ where D is the diffusion coefficient and B is proportional to the solubility. Bentley MVLB. Pandolfelli VC. Macnicol DD. Nature 1998. Posner AS. Science 1994. [5] Otsuka M. C0 is the initial concentration and k is given by [30] 16D . Szymura-Oleksiak J. Science 2001. Chang IM. 1984. Burgos et al. Kimball AP. 1995. In: Lehn JM. Atwood JL. Shastri VP. 4 and 5 are well described by the model proposed in Refs. Langer R. Nature (London) 1964. J Mater Chem 1997. London: Academic Press. Hence. [16] Szejtli J. Sherwood E.52:281. .89:1486. [24] Boyar EB.31]. J Control Release 1998. In addition. Kamal A. [30] Narasimhan B. Coord Chem Rev 1983. Alves OL. Udo K.954. [29] Sinisterra RD. References [1] Rothwell WP. [19] Hughes RG. one expects that the burst time should decrease. In addition.293:5527.74:600. Maria DA. Ohno T. [11]. Santos PS.43:115. Benson HE.102:2637. 1984. Hildgen P.26:1053. [17] Hirayama F. London: Academic Press. Joward RA. Najjar R. PCT/US99/ 11981 Patent International Application. Davies JED.213:117. which characterizes the burst effect. our results in Figs. Pathak S. Proc Am Assoc Cancer Res 1972. Couto LG. Opdebeeck JP. one can write [30] dQ ¼ 2pC0 a3 kð1 þ ktÞ2 .88(5):574. Ono K. Lehnert SM. If e{1 in Eq.7:575. [18] Uekama K. Langer R. Langer R. Sugiyama Y. Davies JED. Riebscheid EM Zyngier S. Practical applications of this approach could be valuable for treating metastatic bone tumors or minimizing the chance of recurrence or localized benign bone tumors such as giant cell tumors [11]. Alencastre JB.20:943. Korkusuz F. [6] Peppas NA. Najjar R. J Control Release 1997. Sepulveda P. Waugh JS. as also pointed out in Ref. Szejtli J. J Pharm Sci 1999. (6). Araki N. Ogata T. J Pharm Sci 2000. [12] Itokazu M. Arima H.472.27:91. Marchetti JM. Comprehensive supramolecular chemistry. Uchida A. where a is the radius of the slab. Macnicol DD. Bear JL. [13] Calhoun JH. Higuchi IW. / Biomaterials 23 (2002) 2519–2526 2525 time effect can be analyzed following the work of Narasimhan and Langer [30]. Yesinowski. Concluding remarks This work has dealt with the preparation and characterization of a controlled release system of rhodium (II) carboxylates and their inclusions using CDs from a HA matrix.145:1278.

42–65. J Am Chem Soc 1987. 1984. I. 1995. Condrate SR. Zh Neorg Khim 1965. . Elliott JC. [31] Langer RS. [38] Burgos AE. J Mater Sci 1998. [40] Yesinowski JP. Bear JL.9:779.81:1421. [39] Rusin RP.2526 A.10:539. [33] Merry JC. p. Sinisterra RD. Eckert H. [32] Morgan H. Anderson P. vol. J Am Ceram Soc 1998. Morozova SA. Fischman GS.38:1015. Biomaterials 2000. / Biomaterials 23 (2002) 2519–2526 [36] Nazarova AL. Atsuo I. Wilson RM. Unpublished results. 1–115.21:617. Best SM. Bonfield W.33:4111. Burgos et al. Bioceramics: materials and applications II. Medical applications of controlled release. Wynne AM. Wise DL. p. [34] Jillavenkatesa AS. Dowker SEP. Tetsuya T.109:6274. [35] Layrolle P. Gibson IR. Chenyaev II. J Mater Sci 1998. J Inorg Nucl Chem 1976. [37] Howard RA.E. Westerville: American Ceramic Society. Wendlant WW. Boca Raton: CRC-Press.