This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Schofield, S.C. Articles by Filiaggi, M.J. Search for Related Content PubMed PubMed Citation Articles by Schofield, S.C. Articles by Filiaggi, M.J. Social Bookmarking                         What's this?

Gelled Calcium Polyphosphate Matrices Delay Antibiotic Release

¹ Department of Applied Oral Sciences, Faculty of Dentistry, Dalhousie University, 5981 University Avenue, Halifax, Nova Scotia B3H 3J5, Canada;
² The Atlantic Region Magnetic Resonance Centre, Dalhousie University; and
³ School of Biomedical Engineering, Dalhousie University

Correspondence: ^* corresponding author, Filiaggi{at}dal.ca

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Introducing a gelling step during antibiotic incorporation has previously been found to delay vancomycin delivery from a calcium polyphosphate matrix intended for local treatment of bone infections. This study examined the general applicability of this approach using cefuroxime, a lower-molecular-weight antibiotic with different charge characteristics compared with those of vancomycin. A calcium polyphosphate/cefuroxime paste was "gelled" in disk form in a humid environment for 5 or 24 hours prior to drying. Antibiotic release in Tris-buffered saline under gentle agitation was monitored over a seven-day period. While non-gelled samples clearly exhibited a burst release, the gelling process significantly retarded early antibiotic release from five- and 24-hour gelled matrices, yielding a constant release rate over the first four days. Cefuroxime incorporation did not appear to alter matrix structure or degradation. Overall, this non-aggressive process effectively trapped cefuroxime and reduced its release rate, suggesting its potential applicability with molecularly diverse therapeutic agents.

Key Words: calcium phosphate • antibiotics • local drug delivery • bone repair

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Local antibiotic delivery systems reduce the circulating concentration of the drug, limiting the exposure of non-target organs to the antibiotic. Carriers developed from bone replacement materials especially can aid in the prophylaxis and treatment of localized bone infections (osteomyelitis), while offering protection to metallic implants against foreign body infections, improving fracture healing and bone-defect filling (Adams et al., 1992).

Calcium phosphates are excellent bone replacement materials, due to their proven osteoconductivity, biocompatibility, and non-toxic degradation products (Pilliar et al., 2001; Dorozhkin and Epple, 2002; Grynpas et al., 2002), and have shown potential in drug delivery applications (Paul and Sharma, 2003). Many calcium phosphate stoichiometries have been investigated (LeGeros, 1991; Dorozhkin and Epple, 2002; Weiss et al., 2003), with calcium polyphosphate the material of interest in the current investigation. This osteogenic material forms linear phosphate chains that degrade to calcium orthophosphate, a non-toxic salt that is readily metabolized by the body (Pilliar et al., 2001). Fukui and colleagues (1977) developed the first calcium polyphosphate for biomaterial applications. More recent in vivo studies have since demonstrated general vascularization and infiltration of connective tissues into these matrices (Nelson et al., 1993; Grynpas et al., 2002).

Dion and colleagues (2005a,b) investigated vancomycin-incorporated calcium polyphosphate matrices for both their structural character and antibiotic release properties. A non-aggressive process—referred to as "gelling"—was used to incorporate a high concentration of vancomycin into the amorphous calcium polyphosphate matrix. Antibiotic release was reduced in the gelled samples, with additional mechanical testing suggesting potential use in non-load-bearing skeletal wound sites. The objective of the current study was to look at the general applicability of this calcium polyphosphate gelling approach for other therapeutic agents. Here, cefuroxime, a lower-molecular-weight antibiotic with different charge characteristics compared with those of vancomycin, was assessed with respect to its ability to be incorporated and subsequently released in a sustainable manner. Any influence of the antibiotic on matrix structure and degradation was also evaluated.

	MATERIALS & METHODS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Matrix Preparation
Calcium phosphate monobasic monohydrate crystals (Ca[H₂PO₄]₂:H₂O) were calcined for 10 hrs at 500°C in a Pt-5%-Au crucible, subsequently melted at 1100°C for 2 hrs to allow for phosphate chain growth, then quenched in ddH₂O at room temperature to produce an amorphous calcium polyphosphate, [Ca(PO₃)₂]n (Pilliar et al., 2001). This frit was pulverized in a ball mill and screened to produce a powder with a particle size below 45 µm.

Antibiotic solution was combined with the particulate in the ratio: 150 mg calcium polyphosphate, 60.2 µL ddH₂O, and 7.5 mg of cefuroxime. The resulting paste (3.44 wt% antibiotic) was gently mixed in the fingertip of a nitrile glove before being transferred to disk-shaped polyvinylsiloxane moulds. Approximately 85% of the paste was incorporated into each mould (6.4 mg cefuroxime), as dictated by the mould size. The samples were left to "gel" for 5 or 24 hrs in ~ 100% relative humidity at 37°C, then were dried for 48 hrs at 37°C in atmospheric air. Additional samples were prepared without the gelling phase (non-gelled) and without antibiotic (blank). For each condition (non-gelled, five-hour-gelled and 24-hour-gelled), 14 matrices with cefuroxime incorporation and 2 blank disks were evaluated.

Elution Protocol
We tracked antibiotic release by placing the disks in cornea viewing chambers containing 15 mL of 0.1 M Tris-buffered saline (pH 7.3), with gentle agitation on a rotating platform (LabRotator, Barnstead/Labline, Dubuque, IA, USA) at 90 rpm. pH was monitored by means of an Accumet Basic AB15 pH meter (Fisher Scientific, Boston, MA, USA). At multiple time-points up to 7 days, a 7-mL quantity of the elution medium was removed for analysis and replaced with fresh Tris-buffered saline. Cefuroxime concentration was assessed spectrophotometrically at 274 nm, while calcium and phosphate levels were determined by established atomic absorption (operation manual; Perkin-Elmer, Wellesley, MA, USA) and colorimetric assay (Halmann, 1972) protocols, respectively. Matrices were dried for 48 hrs at 37°C at the conclusion of the elution experiment, and subsequently dissolved in a 3% Na-EDTA solution at a ratio of 6 mg calcium polyphosphate per milliliter of EDTA (Dion et al., 2005a), so that residual cefuroxime retention could be assessed.

Chemical Analysis
We used Raman spectroscopy and solution ³¹P-NMR to characterize the matrix structure before and after elution, and to determine if the incorporated antibiotic altered the gelling process. For Raman spectroscopy, disks were pulverized and observed with a Bruker FT-Raman Spectrometer (1064 nm Nd:YAG laser; Bruker, Madison, WI, USA) at 300 mW over a range of 100–4000 cm^–1. Solution ³¹P-NMR was necessary to analyze the phosphate chain lengths of these matrices. Disks were dissolved in a 3% Na-EDTA solution as described above, and analyzed in a Bruker AC250 at 101.26 MHz, with a 15° pulse, 7.0-second repetition rate, and 65.5 x 10³ datapoints (Dion et al., 2005a). Average phosphate chain lengths (n) were calculated from the ³¹P-NMR spectra by a peak area method according to the following equation: n = (Q² + Q¹ + Q⁰) / [(Q¹/2) + Q⁰], where Q⁰, Q¹, and Q² are the standardized delimited peak areas corresponding to the ortho, terminal (end), and internal phosphate groups, respectively (Brow et al., 1995; Dion et al., 2005a).

Microbiological Activity Assay
We conducted microdilution assays using S. epidermidis, according to the National Committee for Clinical Laboratory Standards protocol (2000), to compare freshly prepared cefuroxime solution to cefuroxime eluted from these matrices. Samples were taken at 2 hrs (non-gelled matrices) and 24 hrs (five- and 24-hour gelled matrices) from the elution trials. Differences in time point collection were necessary to yield the appropriate cefuroxime concentration in solution necessary for the microdilution assay.

Statistical Analysis
We performed two-way ANOVA tests to determine the influence of gelling and elution time during the elution study, with Bonferroni post hoc tests to compare means of the 3 processing parameters with each other. Results are presented as mean ± standard error (n = 14).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Disks nominally 8 mm in diameter and 2 mm in thickness were routinely produced for all processing conditions. Disk swelling was observed for all sample groups during elution, occurring within 4 hrs for non-gelled disk and within 24 hrs for the gelled samples. The outer shells of the disks remained intact throughout the study, with the eventual extrusion of the inner core from the disk.

Elution studies revealed a burst release of antibiotic from the non-gelled disks that was significantly reduced (p < 0.001) with a five- or 24-hour gelling step. Notably, 69.2 ± 1.7%, 17.9 ± 0.6%, and 20.4 ± 2.1% of incorporated antibiotic was released after 24 hrs for the non-gelled, five-hour-gelled, and 24-hour-gelled samples, respectively (Fig. 1A). By 4 days, the corresponding cumulative release values were 86.6 ± 1.2%, 81.3 ± 3.6%, and 70.7 ± 2.9%. Cefuroxime release rates at the 13 individual time-points are displayed in Fig. 1B, with the graph scaled accordingly to display release rates clearly at all time-points. The burst release of antibiotic from the first two time-points in the non-gelled samples (2.35 ± 0.17 mg/hr, and 1.33 ± 0.11 mg/hr, respectively) has been excluded for clarity. The gelled disks showed a slight burst release of cefuroxime at 30 min, but near-constant release rates from the second time-point (1 hr) until 96 hrs. Analysis of the matrices following elution confirmed the presence of residual cefuroxime; overall, approximately 95% of the initial load was accounted for.

View larger version (19K): [in this window] [in a new window]   Figure 1. Antibiotic release and matrix degradation of calcium polyphosphate disks in 0.1 M Tris-buffered saline solution over 7 days from non-gelled disks (•), 24-hour-gelled disks (), and five-hour-gelled disks (). Cefuroxime release shown as a percent of the total antibiotic incorporated into the calcium polyphosphate disks (A), and the rate of cefuroxime release (mg/hr) at 13 time-points over the 7 days (B). Calcium release as a percent of the total available in the calcium polyphosphate matrix (C), and the rate of calcium release (mg/hr) at individual time-points (D), along with phosphate release as a percent of the total available in the calcium polyphosphate matrix (E), and the rate of phosphate release (mg/hr) at individual time-points (F) describe the matrix degradation of the calcium polyphosphate disks. Change in pH over time from the collected 7-mL samples was tracked (G). Cumulative release of cefuroxime (mg) is shown for the first 4 days (H) of the elution study, to highlight the linear portion of the 24- and five-hour-gelled curves. All samples for each time-point are presented as mean ± standard error (n = 14); blank samples (n = 2) did not significantly differ from their gelled and non-gelled equivalents and, thus, were not included in graphs C through G.

Raman spectroscopic analysis of the disks before and after the elution study suggested no chemical interaction between cefuroxime and the calcium polyphosphate matrices, as would be observed through spectral peak shifts (Fig. 2). In addition, little change in the matrix structure arising from antibiotic incorporation was observed by solution ³¹P NMR (Fig. 3). An overall decrease in phosphate chain length with increasing gelling time was noted (Fig. 4), with a further decrease during elution consistent with expected continued hydrolysis of the longer phosphate chains in the aqueous elution medium. However, there were no apparent chain-length differences between blank calcium polyphosphate matrices and their cefuroxime-incorporated counterparts.

View larger version (18K): [in this window] [in a new window]   Figure 2. Raman spectra of five-hour-gelled calcium polyphosphate matrices before elution. A blank five-hour-gelled disk (A) and a corresponding cefuroxime-loaded disk (B) showed characteristic peaks for a linear polyphosphate (704 and 1170 cm–1) (Jager and Heyns, 1998) and cefuroxime (1482 and 1600 cm–1), with no apparent interaction (no peak shifts) evident with antibiotic incorporation or compared with the starting compounds (not shown). Note that elevated "noise" or background levels seen between 2500 and 1400 cm–1 in the blank disk spectrum could not be adequately resolved with the available background subtraction software. Non-gelled and 24-hour-gelled disks did not have spectra significantly different from those of the five-hour-gelled disks and, thus, are not shown (n = 1).

View larger version (19K): [in this window] [in a new window]   Figure 3. Solution 31P NMR spectra are shown of the amorphous calcium polyphosphate (A), and five-hour-gelled disks with cefuroxime incorporated before (B) and after the elution study (D), and five-hour-gelled disks without cefuroxime incorporated before (C) and after the elution study (E). No significant peak shifts can be seen between the amorphous calcium polyphosphate (A) and the five-hour-gelled disks before the elution study, either with (B) or without (D) cefuroxime incorporation. A chemical phase shift at ~ 1–2 ppm corresponds to an orthophosphate group having no bridging oxygen atoms (Q0), whereas the peak at ~ –9.5 ppm describes a terminal phosphate group with only 1 oxygen atom bridging to a neighboring tetrahedral (Q1). An internal phosphate tetrahedral bond involving 2 bridging oxygen atoms (Q2) is representative of either linear polyphosphate chains (~ –20 ppm) or of metaphosphate ringed structures (~ –22 ppm). Dramatic spectral changes are observed in the post-elution calcium polyphosphate matrices (C&E), showing very few internal phosphate bonds. Spectra of non-gelled and 24-hour-gelled disks did not significantly differ from those of the five-hour-gelled disks and, thus, are not shown (n = 1).

View larger version (20K): [in this window] [in a new window]   Figure 4. Average phosphate chain lengths of the calcium polyphosphate matrices are shown for all 3 parameters (24-hour-gelled, five-hour-gelled, and non-gelled disks) and amorphous calcium polyphosphate, as calculated from the 31P NMR spectra by a peak area method (see text). No differences were seen between the cefuroxime-loaded matrices and their blank counterparts for any parameter, before or after the elution study. The gelling process changed the average phosphate chain length, such that a longer gelling time caused more phosphate chain lysis. All of the disks after the seven-day elution study revealed pyrophosphates (2 phosphates bonded together) in the calcium polyphosphate matrices, another indication of matrix degradation throughout the elution process (n = 1).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
In local drug delivery systems, an initial burst release of drug is often observed within the first 20–60 min as the solution infiltrates the matrix. The most significant result in the current investigation was that the gelling process of the calcium polyphosphate matrices delayed antibiotic release and caused a discharge of the drug in a linear fashion over 4 days at clinically relevant concentrations (Scott et al., 2001). This suggested that the degree of hydration, which was established in the gelling phase, and structural rearrangements with drying effectively trapped the molecule. Earlier reports of vancomycin release from calcium polyphosphate, by the same matrix fabrication process, showed that gelling reduced the antibiotic release rate, though not to the same extent as seen with cefuroxime (Dion et al., 2005b). Cefuroxime is a relatively small molecule with a pK_a of around 2.5; in contrast, a vancomycin molecule with approximately 3 times the molecular weight possesses several charge domains dominated by higher pKa values, with an overall formal charge of +1 at neutral pH (Svensson et al., 1998; Loll and Axelsen, 2000; Trissel, 2000). These results suggest that individual molecular properties somewhat influence drug release kinetics from these matrices, though systematic studies are needed to determine the relative contributions of these properties to the overall release behavior.

Antibiotic release did not directly correlate with matrix degradation, indicating that this is not a singular dominating factor. While all groups exhibited consistent calcium and phosphate release, contradictory trends in total calcium or phosphate release between groups were noted, a result possibly attributed to incomplete breakdown of short-chain phosphates to orthophosphates or subsequent chelation of calcium ions in solution. A small peak in phosphate release with the gelled samples at around 56 hrs of elution did coincide, however, with corresponding increases in calcium and cefuroxime, though the reason for this sudden increase remains unexplained. Overall, these matrices degraded largely by bulk rather than by surface erosion, indicating that macroscopic properties such as size and shape likely have a strong effect on drug-release kinetics and could be used to enhance the therapeutic time frame. It is important to note, however, that these are in vitro results. In vivo degradation rates for this ceramic have been shown to be an order of magnitude greater, due in part to cellular and enzymatic activity (Pilliar et al., 2001; Grynpas et al., 2002).

Cefuroxime failed to retain its antimicrobial activity when maintained in Tris-buffered saline solution at 37°C, a result consistent with a reported short in vivo half-life of 1.2–1.4 hrs (Scott et al., 2001). Additional microdilution activity assays performed on cefuroxime recovered from matrices processed and handled at room temperature confirmed that this antibiotic’s relative instability at 37°C, rather than the incorporation protocol explicitly, was responsible for the loss of antimicrobial activity seen with the eluted antibiotic. An ability to retain antimicrobial activity with a more robust vancomycin molecule, reported in earlier studies (Dion et al., 2005b), further emphasizes the relatively benign conditions for this incorporation strategy that make it suitable for most molecules. Importantly for this study, our ability to detect cefuroxime was not impaired over the time frame in question, despite a loss in activity, since absorbance values for a fixed concentration of cefuroxime in Tris-buffered solution were found to remain essentially unchanged over a seven-day period. Notably, this antibiotic was chosen not so much for its therapeutic value, but rather for molecular properties that were sufficiently different from those of vancomycin. Ultimately, other β-lactams, such as Penicillin V, which generally have lower pKa values while exhibiting greater overall stability, may provide a more logical alternative for studying molecular property effects on delivery from these matrices, though ease of detection and relevance to osteomyelitis treatment may be of some concern.

Local drug delivery systems are receiving significant attention, because of their abilities to dispense high drug concentrations to the area of interest without the need for high systemic levels. Calcium phosphate ceramics as drug carriers may function additionally as potential bone substitutes, or may accelerate the bone-healing process (Weiss et al., 2003). Relatively simple processing that allows matrices of various sizes and shapes to be developed, without losing any therapeutic efficacy, makes the approach presented here very alluring. Its potential applicability with other therapeutic agents, including growth factors, suggests a broad array of dental applications for these calcium polyphosphate devices, including a role in craniofacial and maxillofacial reconstruction and in the treatment of periodontal defects.

	ACKNOWLEDGMENTS

 
Funding for this study was provided by the National Science and Engineering Research Council (NSERC) of Canada and by a Canadian Institutes of Health Research (CIHR) NORTH grant.

Received for publication December 1, 2004. Revision received February 20, 2006. Accepted for publication March 24, 2006.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 

• Adams K, Couch L, Cierny G, Calhoun J, Mader JT (1992). In vitro and in vivo evaluation of antibiotic diffusion from antibiotic-impregnated polymethylmethacrylate beads. Clin Orthop Relat Res 278:244–252.[Medline] [Order article via Infotrieve]
• Brow R, Tallant D, Myers S, Phifer C (1995). The short-range structure of zinc polyphosphate glasses. J Non-Crystal Solids 191:45–55.
• Dion A, Berno B, Hall G, Filiaggi MJ (2005a). The effect of processing on the structural characteristics of vancomycin-loaded amorphous calcium phosphate matrices. Biomaterials 26:4486–4494.
• Dion A, Langman M, Hall G, Filiaggi M (2005b). Vancomycin release behaviour from amorphous calcium polyphosphate matrices intended for osteomyelitis treatment. Biomaterials 26:7276–7285.
• Dorozhkin SV, Epple M (2002). Biological and medical significance of calcium phosphates. Angew Chem Int Ed Engl 41:3130–3146.[Medline] [Order article via Infotrieve]
• Fukui H, Taki Y, Abe Y (1977). Implantation of new calcium phosphate glass-ceramics (annotation). J Dent Res 56:1260.
• Grynpas MD, Pilliar RM, Kandel RA, Renlund R, Filiaggi M, Dumitriu M (2002). Porous calcium polyphosphate scaffolds for bone substitute applications in vivo studies. Biomaterials 23:2063–70.
• Halmann M (1972). Spectrophotometric determination of phosphorus by heteropolary blue method. In: Analytical chemistry of phosphorus compounds. New York: John Wiley & Sons Inc.
• Jager HJD, Heyns AM (1998). Study of the hydrolysis of sodium polyphosphate in water using Raman spectroscopy. Appl Spectrosc 52:808–814.
• Jarcho M (1981). Calcium phosphate ceramics as hard tissue prosthetics. Clin Orthop Relat Res 157:259–278.[Medline] [Order article via Infotrieve]
• LeGeros R (1991). Calcium phosphate biomaterials in preventative and restorative dentistry. In: Calcium phosphates in oral biology and medicine. Monogr in Oral Sci. Basel, Switzerland: Karger.
• Loll PJ, Axelsen PH (2000). The structural biology of molecular recognition by vancomycin. Annu Rev Biophys Biomol Struct 29:265–289.[CrossRef][Medline] [Order article via Infotrieve]
• National Committee for Clinical Laboratory Standards (2000). Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. 5th ed. Approved standard M7-A5. Wayne, PA: NCCLS.
• Nelson SR, Wolford LM, Lagow RJ, Capano PJ, Davis WL (1993). Evaluation of new high-performance calcium polyphosphate bioceramics as bone graft materials. J Oral Maxillofac Surg 51:1363–1371.[Medline] [Order article via Infotrieve]
• Paul W, Sharma CP (2003). Ceramic drug delivery: a perspective. J Biomater Appl 17:253–264.[Abstract/Free Full Text]
• Pilliar RM, Filiaggi MJ, Wells JD, Grynpas MD, Kandel RA (2001). Porous calcium polyphosphate scaffolds for bone substitute applications—in vitro characterization. Biomaterials 22:963–972.
• Scott LJ, Ormrod D, Goa KL (2001). Cefuroxime axetil: an updated review of its use in the management of bacterial infections. Drugs 61:1455–1500.[Medline] [Order article via Infotrieve]
• Svensson LA, Karlsson KE, Karlsson A, Vessman J (1998). Immobilized vancomycin as chiral stationary phase in packed capillary liquid chromatography. Chirality 10:273–280.
• Trissel LA (2000). Trissel’s stability of compounded formulations. 2nd ed. Washington, DC: American Pharmaceutical Association.
• Weiss DD, Sachs MA, Woodard CR (2003). Calcium phosphate bone cements: a comprehensive review. J Long Term Eff Med Implants 13:41–47.[Medline] [Order article via Infotrieve]

Journal of Dental Research, Vol. 85, No. 7, 643-647 (2006)
DOI: 10.1177/154405910608500712


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Schofield, S.C. Articles by Filiaggi, M.J. Search for Related Content PubMed PubMed Citation Articles by Schofield, S.C. Articles by Filiaggi, M.J. Social Bookmarking                         What's this?