In 1970, Buchholz and Engelbrecht introduced the concept of local antimicrobial delivery by loading polymethylmethacrylate (PMMA) with gentamicin for treatment and prophylaxis of prosthetic joint infection.4 The activity of antimicrobial-loaded PMMA has been shown14,17; inhibition of bacterial adhesion with antimicrobial agents loaded into PMMA has been suggested.16 Use of antimicrobial-loaded bone cement is common for treatment and prevention17,18 of bone infection, achieving high local doses while avoiding systemic toxicity.23,25
The most common antimicrobial agents loaded into PMMA in clinical practice are aminoglycosides8 and vancomycin.18 These agents have been well studied for compatibility with PMMA, presumably because of their spectrum and relative lack of associated allergic reaction (eg, in comparison with β-lactam agents).12 The spectrum of several other antimicrobial agents (eg, quinolones and rifampin) renders them attractive candidates for loading into PMMA. It is important to ascertain whether their biologic activity is compromised as a result of mixing with PMMA, and to define their kinetics of release from PMMA.
Emergence of antimicrobial resistance among organisms causing bone infections threatens the effectiveness of antimicrobial-loaded PMMA.2 Therefore, antimicrobials other than aminoglycosides and vancomycin may be needed to load into PMMA. We designed this study to determine the in vitro compatibility of antimicrobials which appropriate spectra for loading into PMMA, and to determine the antimicrobial release kinetics of these agents from PMMA beads in a continuous flow chamber.
MATERIALS AND METHODS
We analyzed the kinetic parameters and the amount of antimicrobial agent detected after PMMA polymerization. With this aim, we prepared PMMA beads loaded with different amounts of antimicrobial agent. Each bead was subjected to a continuous flow of Krebs-Ringer bicarbonate buffer (Sigma Chemical Co., St. Louis, MO) and aliquots collected for 48 hours. Concentrations of antimicrobials in each aliquot were determined by bioassay. Release kinetic curves were drawn and analyzed for each study antimicrobial.
The antimicrobials studied included: cefazolin (Ancef®, GlaxoSmithKline, Research Triangle Park, NC); ciprofloxacin (USP, Rockville MD); gatifloxacin (Bristol-Myers, Squibb, Plainsboro, NJ); levofloxacin (Sigma-Aldrich, St Louis, MO); linezolid (Pharmacia & Upjohn Inc, Kalamazoo, MI); and rifampin (Bedford Labs, Bedford, OH).
Beads were prepared by adding antimicrobial powder to Surgical Simplex P® radiopaque bone cement (Stryker Orthopedics, Mahwah, NJ) (3 g PMMA, 15 g methylmethacrylate-styrenecopolymer, and 2 g barium sulfate, USP) and methyl methacry- late liquid (9.75 mL methyl methacrylate, 0.25 mL N, N- dimethyl-para-toluidine, 75 ± 15 ppm hydroquinone).
Beads containing 2.5%, 7.5%, and 15% (weight/weight) antimicrobial agent were prepared by mixing 100 mg, 150 mg, and 200 mg antimicrobial powder with 3756 mg, 1850 mg, and 1132 mg cement powder, respectively. We then added 1.878 mL, 0.925 mL, or 0.626 mL methyl methacrylate liquid monomer, respectively, which was mixed thoroughly with a spatula to form putty-like material. The cement was spread in a sterile-bead mold and allowed to polymerize for 2 hours. The mold made 25 beads 3 mm in diameter (weight, 20.7-23 mg) that were stored at 4°C until studied.
We assessed antimicrobial activity after polymerization of PMMA. Three beads were weighed, crushed with sterile pliers, placed in a vial with 5 mL Krebs-Ringer buffer, pulverized using a homogenizer (Ultra-Turrax® T 25 basic, IKA®-Werke, Staufen, Germany) at 13,500 rpm to ≤ 50-μm particles, and left in the dark at 4°C for 48 hours. Antimicrobial concentrations were determined in 48-hour eluates.
Bioassays were performed in triplicate using Micrococcus luteus ATCC 9341 as the indicator organism for cefazolin; Klebsiella pneumoniae ATCC 10031 as the indicator organism for ciprofloxacin, gatifloxacin, and levofloxacin; and Bacillus subtilis as the indicator organism for linezolid and rifampin. Twenty microliters of each sample was pipetted onto 6-mm absorbent sterile filter paper discs (Schleicher & Schuell, Riviera Beach, FL). Standard concentrations (in μg/mL) of 30, 25, 20, 15, and 10 for cefazolin; 3, 1.5, 0.75, 0.325, and 0.162 for ciprofloxacin; 10, 5, 2.5, 1.25, 0.62, and 0.3 for gatifloxacin; 2, 1, 0.5, 0.25, and 0.125 for levofloxacin; 30, 25, 20, 15, 10, and 5 for linezolid; and 10, 8, 6, 4, and 2 for rifampin were used to draw a line (R2 coefficient > 0.9) using Microsoft® Office Excel 2003 (Microsoft Office 2003 Professional Edition, Redmond, WA). One standard concentration of the antimicrobial being tested was placed on each plate containing eluates of unknown antimicrobial concentration. The plates were incubated at room temperature for 30 minutes and then in room air at 30°C for cefazolin and at 37°C for the other study antimicrobials for 24 hours. The diameters of the zones of inhibition were measured using an electronic caliper and corrected using the plate standard. Antimicrobial concentrations were calculated by linear regression of the mean size of the zone of inhibition against the standard curve.
We used the continuous flow chamber as described by Perry et al.19 Sixty mL of Krebs-Ringer bicarbonate buffer was warmed at 37°C in room temperature for 30 minutes. One antimicrobial loaded PMMA bead was weighed and placed in a 5 mL test tube with a plastic lid (Becton Dickinson Labware, Franklin Lakes, NJ) with 1 mL Krebs-Ringer bicarbonate buffer. Two holes were made in the test tube and the bead-containing test tube was placed in a Thermolyne Type Dri-bath (Barnstead/Thermolyne, Model 0817615, Dubuque, IA) at 37°C. A 45-mL plastic vial reservoir with a hole in the cap was filled with Krebs-Ringer bicarbonate buffer. Tygon® tubing (Saint-Gobain Performance Plastics Corporation, 2.25 and 0.75 mm outside and inside diameters, respectively) was run through each end of a two- channel variable-speed tubing pump (Masterflex® C/L®Vernon Hills, IL). On one side of the pump, the tube ran from the reservoir through the pump to the test tube containing the bead. On the other side of the pump, the tube ran from the test tube containing the bead through the pump to a fraction collector (Model 2110, Bio-RAD, Richmond, CA). Krebs-Ringer bicarbonate buffer passed from the tubing at a rate of 1 mL per hour through an intravenous catheter (Critikon, Tampa, FL) and into 1.5-mL microcentrifuge tubes with caps (Tyco Healthcare Group LP, Mansfield, MA), which rotated once every 60 minutes. Before sampling, Krebs-Ringer bicarbonate buffer was run through the system to flush air from the tubing. Samples were collected hourly for 24 hours and then every 2 hours up to 48 hours. Samples were stored at −70°C until bioassays were performed. Each antimicrobial concentration was tested in triplicate.
The area under the curve, peak concentration, and mean ± standard deviation (SD) were calculated. Data are presented as mean ± standard deviation of three bead eluates. The area under the curve for each antimicrobial and concentration were compared by multiple regression analysis testing cross products.
All antimicrobial agents were detectable after mixing with PMMA (Table 1).
All antimicrobial release curves showed bimodal profiles consisting in rapid initial release followed by slower, sustained release (Figs 1-3).
Using microbiologic assays, the lowest detectable concentrations (μg/mL) were 7 for cefazolin, 0.1 for ciprofloxacin, gatifloxacin, and levofloxacin, 4 for linezolid, and 1.5 for rifampin. Linezolid showed the greatest stability after PMMA polymerization (ie, > 80% detectable after polymerization). All antimicrobial agents were detected in the first hour except linezolid at 2.5%, which was not detected throughout. Peak concentrations were achieved within the first 2 hours. After 48 hours in the continuous flow chamber, less than 20% of the initial amount of antimicrobial was detected. Increasing the concentration of antimicrobial loaded into PMMA resulted in increased peak concentration and amount of total exposure as reflected in the area under the curve (p < 0.0001). All three quinolones studied had similar slopes as determined by regression analysis, suggesting similar elution behavior from PMMA. However, gatifloxacin elution concentrations were significantly higher than levofloxacin or ciprofloxacin elution concentrations (p < 0.002). Cefazolin release was the most affected by the concentration loaded into PMMA, and linezolid was the least affected by the concentration as determined by regression analysis (Fig 4). We observed that adding rifampin to PMMA prevented complete curing of PMMA.
Antimicrobial-loaded PMMA is commonly used in orthopaedic surgery as a local delivery system for antimicrobial agents to prevent and treat infections and maintain dead space after debridement.5 Antimicrobial agents mixed into PMMA should be appropriate for the spectrum of organisms likely to cause infection, and should achieve adequate local concentrations. Knowledge of kinetics and confirmation of activity after release from PMMA are essential to define suitability of antimicrobial agents for mixing with PMMA. We used a continuous flow chamber to compare the antimicrobial release from PMMA of six antimicrobials not traditionally mixed with PMMA, but with potentially attractive antimicrobial spectra for such an application.
Our study has some limitations. The limit of detection of our bioassays varied among antimicrobials; we used AUC0-∞ trying to minimize this issue for comparison. We did not assess mechanical properties of antimicrobial-loaded PMMA. Gross observation suggested that the strength of PMMA decreased when mixed with rifampin. Mechanical strength is a consideration when PMMA is used for prosthesis fixation; it is less relevant when PMMA is used for treatment or prevention of bone infection (eg, beads and spacers). It also is unknown how our model correlates with in vivo conditions, and whether local antimicrobial concentrations are enough to result in effective antimicrobial activity. Our data must be compared with data from an in vivo model to answer these questions.
Consistent with prior reports using other antimicrobial agents, kinetics of release of the antimicrobial agents that we studied followed a bimodal curve with high initial concentrations followed by sustained release.1,23,26 The amount of detectable bioactive antimicrobial varied, suggesting partial and variable inactivation as a result of polymerization. Differences between study quinolones and other types of antimicrobial agents13 suggest that each agent may be unique regarding stability after mixture with PMMA. A relationship within quinolone elutions from PMMA was observed by regression analysis (Fig 4).
Elution of antimicrobials from PMMA is not a linear process. Several nonlinear mathematical equations have been described to fit the kinetics of antimicrobial leach from PMMA.24,26 Factors that influence diffusion are brand of bone cement,3 volume and surface area of the bead (or other structure),22 the amount of antimicrobial mixed in PMMA,11 and the turnover of fluid surrounding the bead (or other structure).1 Various mechanisms have been described to explain antimicrobial diffusion from PMMA: the process has been related to diffusion from solid PMMA,3 through voids and pores present after polymerization10 or exclusively from the bead surface.15 We found that antimicrobials under similar conditions have different elution profiles, suggesting that diffusion may be influenced by unique properties of individual antimicrobial agents.
Detectable cefazolin concentrations were greater than the minimal inhibitory concentration breakpoint for susceptible organisms (ie, 8 μg/mL for Staphylococcus species).6 Previous in vitro static models1,13,20,26 showed suitability of cephalosporins for loading into PMMA. Adams et al studied the in vivo release of cefazolin from PMMA beads containing 10% cefazolin w/w in a dog model.1 They found release of concentrations above the breakpoint susceptibility limit for 2 weeks.1 This was similar to in vivo release of 7.5% vancomycin from PMMA beads in a rat model.21
As has been shown with other antimicrobial agents,9,11 we found that increasing antimicrobial concentrations in beads increased the rate of release more than if this process was proportional to concentration.
Data also are available for ciprofloxacin-loaded PMMA.1,7 To the best of our knowledge, there are no published data for gatifloxacin and levofloxacin release from PMMA. DiMaio et al reported that ciprofloxacin is heat stable; however, they did not determine whether it could be recovered after polymerization.7 We found a considerable amount of antimicrobial inactivation. The same applied to levofloxacin (ie, only 24-28% of the loaded antimicrobial was recovered after PMMA polymerization). In contrast, gatifloxacin was more stable after polymerization.
To the best of our knowledge, kinetics of elution of linezolid or rifampin from PMMA have not been previously reported. After PMMA polymerization, linezolid was the most stable antimicrobial agent studied. Linezolid was not detected in beads containing 2.5% antimicrobial; conversely, 7.5%, and 15% linezolid loaded into PMMA were associated with high peak concentrations and area under the curve. All detectable linezolid concentrations were above the susceptibility breakpoint for Staphylococcus (ie, ≤ 4 μg/mL).6 Rifampin was the antimicrobial with the lowest recovery rate after polymerization, although all detectable rifampin concentrations were greater than the susceptibility breakpoint for Staphylococcus species (≤ 1 μg/mL).6
Previous (unpublished) experiments from our group, following the same method, showed antimicrobial recovery before elution and after polymerization of 100% for gentamicin, 70% for tobramycin, and 90% for vancomycin after PMMA polymerization. These results suggest that some inactivation of study antimicrobial agents may have occurred during polymerization. Substantial concentrations of antimicrobial were released, however, bringing into question the clinical relevance of this observation. When comparing the results of our study with results in previous studies using the same technique,11,19 gatifloxacin or linezolid 7.5% beads showed elution similar to tobramycin, vancomycin or daptomycin beads, and slightly lower elution than gentamicin beads. However, beads containing 15% gatifloxacin or linezolid did not elute as completely as did beads containing 15% daptomycin.
All antimicrobial agents studied eluted from PMMA. Profile characteristics were unique to each antimicrobial agent. Gatifloxacin and linezolid showed heat stability and diffusion rates similar to vancomycin at the same concentration.
We thank James M. Steckelberg for contibution to conception of the study and critical revision of the manuscript; Andras Heijink and Megan Ostrem are acknowledged for their technical assistance.
1. Adams K, Couch L, Cierny G, Calhoun J, Mader JT. In vitro and in vivo evaluation of antibiotic diffusion from antibiotic-impregnated polymethylmethacrylate beads. Clin Orthop Relat Res
. 1992;278: 244-252.
2. Anguita-Alonso P, Hanssen AD, Osmon DR, Trampuz A, Steckelberg JM, Patel R. High rate of aminoglycoside resistance among staphylococci causing prosthetic joint infection. Clin Orthop Relat Res
3. Bayston R, Milner RD. The sustained release of antimicrobial drugs from bone cement: an appraisal of laboratory investigations and their significance. J Bone Joint Surg
4. Buchholz HW, Engelbrecht H. [Depot effects of various antibiotics mixed with Palacos resins] Chirurg
5. Calhoun JH, Mader JT. Antibiotic beads in the management of surgical infections. Am J Surg
6. Clinical and Laboratory Standards Institute. MIC testing:supple- mental tables. Supplemental Tables M100-S14 (M7)
. Wayne, PA: Clinical and Laboratory Standards Institute; 2004.
7. DiMaio FR, O'Halloran JJ, Quale JM. In vitro elution of ciprofloxacin from polymethylmethacrylate cement beads. J Orthop Res
8. Donati D, Biscaglia R. The use of antibiotic-impregnated cement in infected reconstructions after resection for bone tumours. J Bone Joint Surg
9. Faber C, Stallmann HP, Lyaruu DM, de Blieck JM, Bervoets TJ, van Nieuw Amerongen A, Wuisman PI. Release of antimicrobial peptide Dhvar-5 from polymethylmethacrylate beads. J Antimicrob Chemother
10. Frutos Cabanillas P, Diez Pena E, Barrales-Rienda JM, Frutos G. Validation and in vitro characterization of antibiotic-loaded bone cement release. Int J Pharm
11. Hall EW, Rouse MS, Jacofsky DJ, Osmon DR, Hanssen AD, Steckelberg JM, Patel R. Release of daptomycin from polymethylmethacrylate beads in a continuous flow chamber. Diagn Microbiol Infect Dis
12. Henry SL, Galloway KP. Local antibacterial therapy for the management of orthopaedic infections: pharmacokinetic considerations. Clin Pharmacokinet
13. Hughes S, Field CA, Kennedy MR, Dash CH. Cephalosporins in bone cement: studies in vitro and in vivo. J Bone Joint Surg
14. Josefsson G, Kolmert L. Prophylaxis with systematic antibiotics versus gentamicin bone cement in total hip arthroplasty: A ten-year survey of 1,688 hips. Clin Orthop Relat Res
15. Kuechle DK, Landon GC, Musher DM, Noble PC. Elution of vancomycin, daptomycin, and amikacin from acrylic bone cement. Clin Orthop Relat Res
16. Oga M, Arizono T, Sugioka Y. Inhibition of bacterial adhesion by tobramycin-impregnated PMMA bone cement. Acta Orthop Scand
17. Ostermann PA, Seligson D, Henry SL. Local antibiotic therapy for severe open fractures: a review of 1085 consecutive cases. J Bone Joint Surg
18. Ozaki T, Yoshitaka T, Kunisada T, Dan'ura T, Naito N, Inoue H. Vancomycin-impregnated polymethylmethacrylate beads for methicillin-resistant Staphylococcus aureus (MRSA) infection: report of two cases. J Orthop Sci
19. Perry AC, Rouse MS, Khaliq Y, Piper KE, Hanssen AD, Osmon DR, Steckelberg JM, Patel R. Antimicrobial release kinetics from polymethylmethacrylate in a novel continuous flow chamber. Clin Orthop Relat Res
20. Picknell B, Mizen L, Sutherland R. Antibacterial activity of antibiotics in acrylic bone cement. J Bone Joint Surg
21. Rouse MS, Piper KE, Jacobson M, Jacofsky DJ, Steckelberg JM, Patel R. Daptomycin treatment of Staphylococcus aureus experimental chronic osteomyelitis. J Antimicrob Chemother
. 2006;57: 301-305.
22. Seeley SK, Seeley JV, Telehowski P, Martin S, Tavakoli M, Colton SL, Larson B, Forrester P, Atkinson PJ. Volume and surface area study of tobramycin-polymethylmethacrylate beads. Clin Orthop Relat Res
23. Torholm C, Lidgren L, Lindberg L, Kahlmeter G. Total hip joint arthroplasty with gentamicin-impregnated cement: a clinical study of gentamicin excretion kinetics. Clin Orthop Relat Res
. 1983;181: 99-106.
24. Torrado S, Frutos P, Frutos G. Gentamicin bone cements: characterization and release (in vitro and in vivo assays). Int J Pharm
25. Wang J, Calhoun JH, Mader JT. The application of bioimplants in the management of chronic osteomyelitis. Orthopedics
. 2002;25: 1247-1252.
26. Wilson KJ, Cierny G, Adams KR, Mader JT. Comparative evaluation of the diffusion of tobramycin and cefotaxime out of antibiotic- impregnated polymethylmethacrylate beads. J Orthop Res
. 1988;6: 279-286.