Effects of Loading Concentration, Blood and Synovial Fluid on Antibiotic

Release and Anti-Biofilm Activity of Bone Cement Beads

Devendra Dusane a,*,†, Scott Diamond b,*, Cory Knecht b,*, Nicholas Farrar c,*, Casey Peters d,

Robert Howlin e, Matthew Swearingen a, Jason Calhoun f, Roger D. Plaut g, Tanya Nocera c,

Jeffrey Granger h, Paul Stoodley a,h

a Department of Microbial Infection and Immunity, The Ohio State University, Columbus,

Ohio 43210

b Department of Medicine, The Ohio State University, Columbus, Ohio 43210

c Department of Biomedical Engineering, The Ohio State University, Columbus, Ohio 43210

d Department of Biochemistry, The Ohio State University, Columbus, Ohio 43210

e Centre for Biological Sciences, Faculty of Natural & Environmental Sciences & Institute for

Life Sciences, University of Southampton, Southampton, UK

f Department of Muscoskeletal Sciences, Spectrum Health Medical Group, Grand Rapids,

Michigan

g Division of Bacterial, Parasitic, and Allergenic Products, Center for Biologics Evaluation

and Research, Food and Drug Administration, Silver Spring, MD 20993

h Department of Orthopaedics; The Ohio State University, Columbus, Ohio 43210

† E-mail address: Devendra.Dusane@osumc.edu

* Equal contribution

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

ABSTRACT

Antibiotic loaded cement beads are commonly used for the treatment of biofilm related

orthopaedic periprosthetic infections; however the effects of antibiotic loading and exposure

of beads to body fluids on release kinetics are unclear. The purpose of this study was to

determine the effects of (i) antibiotic loading density (ii) loading amount (iii) material type

and (iv) exposure to body fluids (blood or synovial fluid) on release kinetics and efficacy of

antibiotics against planktonic and lawn biofilm bacteria. Short-term release into an agar gel

was evaluated using a fluorescent tracer (fluorescein) incorporated in the carrier materials

calcium sulfate (CaSO4) and poly methyl methacrylate (PMMA). Different fluorescein

concentrations in CaSO4 beads were evaluated. Mechanical properties of fluorescein-

incorporated beads were analyzed. Efficacy of the antibiotics vancomycin (VAN) or

tobramycin (TOB) alone and in combination was evaluated against planktonic and lawn

biofilms of bioluminescent strains of Staphylococcus aureus and Pseudomonas aeruginosa.

Zones of inhibition (ZOI) were measured visually and using an in-vivo imaging system

(IVIS). The influence of body fluids on release was assessed using CaSO4 beads that

contained fluorescein or antibiotics and were pre-coated with human blood or synovial

fluid. The spread from the beads followed a square root of time relationship in all cases. The

loading concentration had no influence on short-term fluorescein release and pre-coating of

beads with body fluids did not affect short-term release or antibacterial activity. Compared to

PMMA, CaSO4 had a more rapid short term rate of elution and activity against planktonic

and lawn biofilms. This study highlights the importance of considering antibiotic loading and

packaging density when investigating the clinical application of bone cements for infection

management.

Running title: Short-term antibiotic release and anti-biofilm efficacy from bone cement

beads

2

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

Keywords: periprosthetic infection; biofilm; bone cement; antibiotic release; zone of

inhibition

1. Introduction

Orthopaedic periprosthetic joint infections (PJI) are difficult to treat with systemic

antibiotic therapy and can lead to severe complications such as removal of the implant with

functional loss of the affected body part or life-threatening conditions [1, 2]. The extent of

infection depends upon several factors, such as health of the patient, length of time of

infection and condition of local soft tissues.

Antibiotic-loaded bone cement beads are a commonly-used option functioning to fill

the dead space and managing infection for the treatment of PJI [3, 4] (Fig. 1). Beads are

made directly in the operating room from poly (methyl) methacrylate (PMMA) or mineral

based formulations such as calcium sulfate (CaSO4) [4-6].

Fig. 1. Surgical site of infection showing antibiotic loaded calcium sulfate (CaSO4) beads.

PMMA was first used in a structural role in order to stabilize the implant or providing

a spacer [7], but by adding antibiotics, surgeons have also used it to control infections.

3

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

However, PMMA is dense, with compressive strength of more than 70MPa, acrylic, and non-

resorbing material, which generally must be removed in a second surgical procedure when its

function has been fulfilled to avoid becoming a nidus for future infection [8, 9]. CaSO4 on the

other hand assists in the regeneration of bone, but is not utilized for structural integrity as it is

a weak inorganic compound [10, 11]. High purity synthesized CaSO4 has the advantage of

less adverse reactions in patients than with cements purified from mined mineral. It is also

biodegradable, so that all antibiotic is released and no material remains and it cures at

relatively low temperatures and is therefore compatible with a wide spectrum of antibiotics

[12].

Biofilms are communities of microorganisms that colonize surfaces or exist as

aggregates and play an important role in the pathogenesis of PJI [13]. In the case of PJI,

biofilms are commonly associated with a foreign body such as a prosthetic [5, 13].  Bacteria

in a biofilm becomes highly tolerant to antibiotics as compared to the planktonic phenotype,

therefore localized antibiotics delivery is needed to provide sustained high concentrations of

antibiotics, that cannot be achieved systemically [14, 15].

However, to achieve these concentrations, whether incorporating the required dose of

antibiotics in few beads or distributing the dose in multiple beads to provide better release

and efficacy is unclear. In a previous study, we reported that the zone of inhibition of

Pseudomonas aeruginosa PAO1 from CaSO4 carrier beads loaded with tobramycin (TOB)

was similar regardless of whether beads were placed singly or in groups of 2, 3 or 4 (Fig. 2A)

[5]. This is important since it shows that in diffusion limited environment, such as might be

found in periprosthetic tissue or quiescent fluid, that the packing density of the beads is an

important consideration in providing adequate coverage of the infected area. Further, multiple

beads placed together also provide a multiplicity of antibiotic at a specific location.

Moreover, when the release is diffusion limited for an absorbable material in which all

4

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

antibiotic will be released it might be expected that the higher bead density would provide

longer coverage.

Fig. 2. Influence of single and multiple beads on local high concentration of antibiotic. ZOI

against P. aeruginosa PAO1 by multiple TOB-loaded CaSO4 beads after 24 h is relatively

similar as with single bead [16] (A) ††. Representative images showing fluorescein elution

through multiple beads after 24 h was indifferent than with single fluorescein-loaded CaSO4

(B). ††Adapted from [5].

Antibiotic release from loaded bone cement has been earlier reviewed [17, 18]. The

release of antibiotic from bone cement materials can be affected by factors such as type [19],

preparation [20], surface characteristics [21], and porosity of the cement [22], as well as the

amount and type of the antibiotic. Highly porous cement may elute more antibiotic and for a

longer period of time relative to cement with less porosity [23]. The porous cement material

such as CaSO4 is a better antibiotic carrier material than PMMA for local delivery of

antibiotics [5]. In a recent study by Chang et al (2014) it was shown that adding antibiotics

(such as vancomycin, amphotericin B) powder in distilled water before mixing with bone

cement significantly improves the efficiency of antibiotic release than same dose of antibiotic

5

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

powder [24]. However, the ultimate compressive strength of the beads was significantly

reduced in specimens containing liquid antibiotics.

Release of antibiotic from bone cement has previously been reported and measured in

phosphate buffered saline (PBS) by estimating the zone of inhibition [25] or by using HPLC

[26] or LC-MS [27]. However, the diffusion of antibiotics from a carrier bead into

surrounding soft tissue and quiescent joint fluid is a diffusion limited environment and the

degree of penetration of an antibiotic into the infected site is an important determinant of

therapeutic success [4, 28]. We have earlier reported the use of agar diffusion method to

analyze the zone of inhibition [5] and extended these observations to study the importance of

antibiotic loading density, loading amount, material type and exposure to body fluids on

release kinetics and efficacy of antibiotics against planktonic and lawn biofilm bacteria. The

agar method allows understanding how antibiotics might spread into the surrounding tissues

as we can monitor the elution distance, antibacterial effects and preserve the gradients that

are likely to develop in vivo in areas that are diffusion limited. The diffusion limited agar

technique therefore becomes relevant and clinically important.

To investigate the release kinetics of diffusion from beads into a surrounding gel in

real time we used fluorescein dye tracer which has previously been used as a surrogate to

study antibiotic release based on its hydrophilic characteristics [16]. In the clinical setting,

antibiotic-loaded beads immediately come into contact with the body fluids, such as blood

and synovial fluid that may influence the properties and release from the beads. We

hypothesized that exposure to these body fluids influences the release kinetics of antibiotic

from antibiotic-loaded orthopaedic cements. Calcium is a known clotting factor and it is

possible that blood clots that form around the beads may also influence release of antibiotics.

This might be disadvantageous (by trapping antibiotics within the beads) or advantageous

6

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

(slowing down the release to increase the period over which the beads may effectively release

antimicrobial concentrations inhibitory to the cells).

Therefore the purpose of the present study was to determine whether (i) the bone

cements PMMA and high purity synthetic CaSO4 hemihydrate would have different elution

kinetics and efficacy against planktonic and biofilm bacteria, (ii) the release kinetics of

antibiotics from CaSO4 beads was dependent on the loading concentration and (iii) if

exposure to human blood or synovial fluid influenced the release kinetics and antibiotic

efficacy. To track release, we used an agar model to simulate release into a surrounding

viscous medium by diffusion alone, using fluorescein as a tracer and the extinction of activity

in biofilm lawns grown from bioluminescent strains of Staphylococcus aureus and

Pseudomonas aeruginosa, two common PJI pathogens.

2. Materials and methods

2.1. Preparation of bone cement beads with fluorescein

CaSO4 (Stimulan Rapid Cure, Biocomposites Inc, NC, USA) and PMMA

(PALACOS® R, Zimmer, IN, USA) powders were weighed, and the corresponding volume of

respective liquid (sterile water for Stimulan Rapid Cure and methyl methacrylate, N,N-

dimethyl-p-toluidine in case of PALACOS R) was added and mixed for 30 to 60s to form a

homogeneous paste. The paste was then spread onto a flexible rubber mold (Biocomposites

Ltd) to make hemispherical beads of 4.8 mm diameter. The beads were left undisturbed to set

for 1 h. After the beads solidified, the molds were flexed by hand in a torsional motion to

remove the beads which were loosened from the mold and dropped in sterile petriplate and

stored at room temperature until use. Control beads were prepared without fluorescein.

Fluorescein (Fluka, USA) was incorporated into beads in the amounts of 1.0 g per 10cc pack

(20.0 g) of CaSO4 or 2.33 g per 40.0 g of PMMA were prepared. One pack of Stimulan

7

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

contains 20.0 g of powder and 6 mL of liquid, which upon mixing yields 23.8 g of set

cement. The mass of one set 4.8 mm bead was 97 ± 4 (n=4) mg so that one pack would

theoretically make approximately 245 beads. With 1.0 g of fluorescein per pack of Stimulan,

the average weight of fluorescein per bead was 3.92 mg and made approximately 250 beads.

When 1.0 g of fluorescein was mixed into a Stimulan pack, each bead was infused with 4.2%

(by weight) of fluorescein. Similarly for PMMA, one pack of PALACOS containing 40.0 g

of powder and 20 mL of liquid that made 53.2 g of set cement. The average mass of a single

4.8 mm bead was 59 ± 8 (n=4) mg, so that one pack could potentially make up to 900 set

beads. The mass of an individual bead and the total number of beads that could be made per

pack were used to calculate the amount of additive (fluorescein or antibiotic) on a per-bead

basis to account for density differences between the two materials.

In order to analyze the effect of number of beads on release of fluorescein, single or

multiple CaSO4 beads (1−4) containing fluorescein (1.0 g per 10cc pack of CaSO4, Stimulan

Rapid Cure, Biocomposites) were placed on 1% agar (Fig. 2B). Using fluorescein, similar

elution characteristics were observed as in case of TOB when single and multiple beads were

used (Fig. 2A-B). Furthermore, in order to investigate the effect of loading density on elution

kinetics, various concentrations of fluorescein-containing beads were also prepared. The

weights of fluorescein used were 0.125, 0.25, 0.5, 1.0, 2.0 and 4.0 g per 10cc pack of CaSO4.

Amounts above 4.0 g per 10cc pack of CaSO4 were not used, since at higher amounts the

beads did not set well.

2.2. Elution kinetics of fluorescein through bone cement beads

CaSO4 and PMMA beads containing fluorescein were placed on 1% pure agar

(Sigma, USA) plates using sterile forceps. The plates were incubated at 37°C, and at different

time intervals (0−8 h) fluorescent images were captured using a gel documentation system

8

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

(Gel Doc™ XR Bio-Rad). The images were analyzed using a freeware program, Fiji

(Fiji.sc/fiji). Images at various time points were aligned to create a set of images that were

normalized to pixel size. Plate diameter was used to set the scale to pixel ratio, the images

were stacked and automatic intensity threshold was applied to determine fluorescein elution

distance. The fluorescein release experiments were performed in triplicate.

2.3. Mechanical properties of cement beads

Increasing the concentration of antibiotics can impair the setting and mechanical

properties of the bone cements [15, 29]. Weakening of cement material is of high concern,

thus in order to avoid mechanical weakening, low concentration of the compound is usually

recommended [15, 30]. The strength of beads containing fluorescein was measured using the

compression test. A bead was placed between two parallel plates and was crushed with a

constant strain. The test frame (100Q Series Mechanical Test Frame, Test Resources) was

activated by turning on the device and running a compressive test profile with a constant

positional change in the parallel plates. After optimizing the strain, 1 mm/min was selected

due to the uniformity and consistency of the load (Newtons) as a function of time (seconds).

The top compression plate was manually lowered until it was flush with the top point of the

spherical section of the bead, and this position was set to zero. Upon initiation of the constant

strain, the threaded top shaft was forced into the apex of the bead, and the force was

measured in real-time by the load cell. The plots of load as a function of time were

transmitted from the test frame to a laptop using WinCom™. Load as a function of time for a

CaSO4 bead was plotted where the bead exhibited a clear first crack point, which was termed

its ultimate compressive strength. Thus, the applied force recorded was when the

microstructure of the bead began to permanently deform (termed as the first crack strength).

9

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

Maximum load values of the impregnated beads were exported to Excel and compared to

their respective control beads.

Fluorescein-impregnated CaSO4 beads were compressed until their ultimate

compressive strengths were observed on the plots of applied load as a function of time.

Student t-tests were performed on each of the concentrations of fluorescein compared to the

non-impregnated bead.

2.4. Preparation of CaSO4 and PMMA beads impregnated with antibiotics

The efficacy of antibiotic-impregnated bone cements was measured by preparing

beads in a similar way as mentioned for fluorescein. Briefly, CaSO4 (20.0 g per pack) or

PMMA (40.0 g per pack) beads with or without the antibiotic(s) [vancomycin hydrochloride

(VAN; Fluka USA) 1.0 g per pack of cement powder or tobramycin sulphate (TOB; Sigma-

Aldrich) 240 mg per pack of cement powder] were prepared. The loading amount of

antibiotic (VAN) per bead in CaSO4 was 3.9 mg and TOB was 0.97 mg. PMMA had 3.47 mg

of VAN and 0.9 mg of TOB per bead [14]. The percent weight of antibiotic impregnated

bead was normalized based on 1.0 g antibiotic added to a 10cc pack of Stimulan per the

manufacturer's recommendation. The levels of antibiotic addition were representative of

levels reported in the literature [5]. The powders were evenly blended with or without the

antibiotics and beads were prepared as described earlier. Controls consisted of beads with no

antibiotic (negative control) and plates with no beads (positive control). All the experiments

were performed in triplicate.

2.5. Bacteria and growth conditions

Bioluminescent strains of Staphylococcus aureus SAP231 [31] and Pseudomonas

aeruginosa Xen41 (PerkinElmer, USA) were used. S. aureus SAP231 is a multidrug resistant

10

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

strain (MRSA) of pulsed-field gel electrophoresis type USA300 containing an integrated

plasmid carrying luxBADCE, [31] whereas P. aeruginosa Xen41 is a P. aeruginosa PAO1

luminescent strain that harbors the luxCDABE cassette inserted in a constitutively expressed

manner [32, 33]. These bacteria were cultured in brain heart infusion (BHI) and lysogeny

broth (LB) respectively (Sigma Aldrich, USA).

2.6. Determination of MIC of VAN and TOB against bacteria

The MIC of antibiotics (VAN and TOB) was determined using the E-test strip

method. S. aureus SAP231 or P. aeruginosa Xen41 was grown in the appropriate growth

media overnight and diluted 100 times. The diluted culture was spread (100 µL) onto freshly

prepared BHI or LB agar plates, and E-test strips (BioMerieux, KY, USA) were placed on the

surface of the plates. The plates were incubated for 24 h at 37°C, and the MIC was

determined based on the area of no growth corresponding to the lowest antibiotic

concentration [34].

2.7. Inhibition of planktonic bacterial growth using antibiotic eluting beads

Agar diffusion assays were used to determine release and inhibitory activity of

antibiotics from the beads (CaSO4 or PMMA) over time. For inoculating the agar plates, a

colony of bacteria was removed from a cultured petri dish to culture overnight at 37°C in 15

mL BHI or LB broth for respective bacterial cultures. The grown culture was diluted to an

optical density (OD) corresponding to ~106 cells/mL and plated on respective BHI or LB

solid agar media. The antibiotic-loaded beads containing VAN, TOB or combinations (VAN

+ TOB) were placed in the center of the plates immediately after inoculation. The plates were

incubated at 37oC at 5% CO2 for 24 h. Images of the plates were taken, and the diameter of

the zones of inhibition (ZOI) formed around the beads was measured using Vernier calipers

11

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

and the in-vivo imaging system (IVIS). Measurements of ZOI were taken 24 h after

placement of the bead. Assays were performed in experimental triplicate and the ZOI data

were expressed as the mean ± standard error.

2.8. Killing pre-existing lawn biofilms using antibiotic eluting beads

Bacterial cultures were spread on respective agar media as described earlier for

planktonic bacterial growth and were incubated for 24 h at 37°C. CaSO4 or PMMA beads

with or without antibiotics were placed onto pre-grown lawn biofilms of S. aureus SAP231 or

P. aeruginosa Xen41 on respective agar plates. Images of the plates were taken and the area

around the antibiotic beads showing the dark zone where bioluminescence was suppressed by

the spreading antibiotic under IVIS imaging system was measured and designated as the

“zone of inactivation” (ZOIn). The ZOIn formed around the beads was measured every 24 h

following placement of the bead on the plate. The spread of inactivation was measured as the

radius of the dark circle. The distance was plotted as a power law relationship function of the

square root of time to linearize the data (all R2 > 0.90). The slope (power law constant) was

used as a comparative release rate coefficient.

Swabs from the inner and outer zones for both PMMA and CaSO4 beads placed on the

lawn biofilm were collected and spread on fresh agar plates to determine whether the

bacterial cells in the zone of inactivation had been killed. Assays were performed in

experimental triplicate and the ZOIn data were expressed as the mean ± standard error.

The colony forming units (CFU/cm2) were calculated from the 24 h grown cultures on

the agar plates. Agar plugs from different areas of the plates containing S. aureus or P.

aeruginosa (24 h grown) were immersed in sterile PBS, homogenized, serially diluted and

plated on respective growth media. The counts for S. aureus and P. aeruginosa were 1.33 ×

12

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

107 and 1.21 × 106 CFU/cm2 respectively. One tailed student’s t-test was used to compare

between CaSO4 and PMMA beads.

2.9. Influence of host factors on release kinetics

To determine whether host factors influence fluorescein release, we used human

blood and bovine synovial fluid (Lampire Biological Laboratories, PA). Blood from healthy

human donors was collected in accordance with the approved IRB protocol 2015H0121.

Individual beads (with or without fluorescein) were placed in wells of a 96-well microtitre

plate and fresh unadulterated blood (~200 µL) was immediately added to the well after

collection. In addition to unadulterated blood which formed a clot on the beads we also

exposed beads to heparinized blood. Similarly, bovine synovial fluid (~200 µL) was added to

the microtitre plate wells containing fluorescein in CaSO4 beads. The beads were removed

after 20 min and were placed on the surface of 1% agar plates using sterile forceps. The

plates were incubated for 24 h at 37°C and every hour fluorescent images were collected and

processed using Fiji.

2.10. Influence of host factors on antibacterial activity

To compare the effect of unadulterated human blood and bovine synovial fluid on the

ZOI, CaSO4 beads with and without TOB antibiotic (240 mg in 10cc pack) were placed in

microtitre plate wells and fresh blood from healthy donors (~200 µL) was immediately

added. The beads were removed after ensuring blood clots and coats around the beads (20

min) and were placed on the surface of P. aeruginosa spread LB agar plates. The beads were

inserted inside the agar surface, plates were incubated at 37°C, and images (using camera and

IVIS) were collected after 24 h.

13

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

3. Results

3.1. Short-term release from bone cements

Fluorescein was used as a tracer compound to study short-term release (8 h) through

1% agar using varying concentration of fluorescein-impregnated CaSO4 beads. Fluorescence

intensity increased over time along the radial distance travelled (Fig. 3A-B).

Fig. 3. Short-term release kinetics of fluorescein from CaSO4 beads over a period of time (t in

minutes). Panel A shows fluorescent images of release of fluorescein from single CaSO4

beads (containing 1g/10cc fluorescein) in 1% agar, and panel B shows the fluorescence

intensity (FI) profile at different time intervals and the radial distance travelled from the

fluorescein incorporated CaSO4 bead.

The spreading distance (d) conformed to a square root of time (t) power law

relationship according to:

d = rt0.5

where we termed the coefficient “r” as the release coefficient for comparative purposes. This

half-order relationship was derived from Fick’s second law on the basis that antibiotic elution

14

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

through a media undergoes unsteady-state molecular diffusion. We observed significant

differences in the rate of fluorescein elution when PMMA and CaSO4 beads were tested

(P<0.05) (Fig. 4A-B). CaSO4 had a faster short-term release rate than PMMA.

Fig. 4. (A) Fluorescein elution radial distance from CaSO4 beads and PMMA; (B) Influence

of loading concentration of fluorescein in CaSO4 and PMMA beads on short-term release as a

function of time (t1/2); (C) Mechanical properties of fluorescein-loaded CaSO4 beads.

3.2. Mechanical properties of fluorescein-containing CaSO4 beads

Initially, the ultimate compressive strength of the CaSO4 beads strength increased

with fluorescein weights of upto 0.5 g (Fig. 4C). However, at weights of 2.0 g and 4.0 g of

fluorescein, the beads were significantly weakened. Above 4.0 g, the beads did not set.

Previous reports with PMMA bone cement showed no significantly change as the antibiotics

15

323

324

325

326

327

328

329

330

331

332

333

334

335

336

are eluted and the percentage drop compressive strength was much higher with larger

amounts of antibiotics [35, 36].

3.3. Determination of minimum inhibitory concentration

The MIC of both VAN and TOB against S. aureus SAP231 was 2.0 µg/mL. For P.

aeruginosa Xen41, the MIC of TOB was 1.5 µg/mL. VAN had no effect on the growth of P.

aeruginosa.

3.4. Inhibition of planktonic bacteria by antibiotic elution through beads

CaSO4 beads demonstrated faster killing than PMMA beads against both P.

aeruginosa and S. aureus (Table 1). The ZOI was measured surrounding CaSO4 beads loaded

with or without VAN or TOB on agar plates. ZOI by CaSO4 loaded with TOB against P.

aeruginosa was larger (41 ± 0.6 mm) than against S. aureus (30.6 ± 0.8 mm). Moreover,

VAN had lower ZOI (23.9 ± 0.8) against S. aureus than did TOB (Table 1).

Table 1. Effect of pre-coating of antibiotic-incorporated cement beads with blood or synovial

fluid on the ZOI diameter of S. aureus SAP231 or P. aeruginosa Xen41 after 24 h incubation.

Zone of inhibition (ZOI) diameter

(mm)

Antibiotic Coating

(Blood/synovial)

S. aureus SAP231 P. aeruginosa Xen41

CaSO4 beads only No 0 0

TOB No 30.6 ± 0.8 41 ± 0.6

VAN No 23.9 ± 0.8 ND

CaSO4 beads only Blood 0 0

TOB Blood 30.8 ± 0.5 41.2 ± 0.5

VAN Blood 24.2 ± 1.0 ND

CaSO4 beads only Synovial 0 0

16

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

TOB Synovial 30.5 ± 0.7 40.7 ± 0.9

VAN Synovial 23.9 ± 0.9 ND

ND = not determined; TOB = Tobramycin; VAN = Vancomycin

3.5. Biofilm inactivation by antibiotics elution from CaSO4 and PMMA beads

The zone of inactivation (ZOIn) of growth of lawn biofilms of P. aeruginosa and S.

aureus with different antibiotics and carrier materials is shown in Fig. 5 and Fig. 6.

Fig. 5. Killing kinetics of pre-existing lawn biofilms of (A) P. aeruginosa Xen41 by TOB

incorporated in PMMA and CaSO4 beads and of (B) S. aureus SAP231 by VAN and VAN +

TOB incorporated in PMMA and CaSO4.

17

354

355

356

357

358

359

360

361

362

363

Fig. 6. Killing of lawn biofilms of S. aureus and P. aeruginosa by antibiotic beads showing

(A) PMMA loaded with VAN (●) and VAN + TOB ( ) and (B) CaSO4 beads containing

VAN (●) and VAN + TOB ( ) against S. aureus SAP231 and (C) PMMA loaded with TOB

(●) and VAN + TOB ( ) and (D) CaSO4 beads containing TOB (●) and VAN + TOB ( )

against P. aeruginosa Xen41.

The spread of inactivation was higher in the case of CaSO4 than PMMA (Fig. 6 A-D).

Rate of inactivation was higher as evidenced by the slopes using CaSO4 than with PMMA

against S. aureus and P. aeruginosa Xen 41 (Fig. 6 A-D). PMMA containing VAN + TOB

had higher slope [2.58 ± 0.13 (r2=0.95)] as compared to VAN alone [1.39 ± 0.09 (r2=0.96)]

against S. aureus suggesting increased rate of inactivation. CaSO4 on the other hand had

higher rate of inactivation with VAN + TOB [3.36 ± 0.07 (r2=0.90)] than VAN alone [1.88 ±

0.01 (r2=0.96)] against S. aureus. Against P. aeruginosa, the zone of inactivation with

PMMA containing TOB [3.59 ± 0.12 (r2=0.95)] was comparable with TOB + VAN [3.62 ±

18

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

0.14 (r2=0.98)]. Similarly, the difference was non-significant with CaSO4 containing TOB

antibiotic [4.59 ± 0.26 (r2=0.99)] and TOB + VAN [4.20 ± 0.89 (r2=1.00)].

The spread rate constants were higher for CaSO4 (P<0.05) than for PMMA,

demonstrating a faster release rate. TOB + VAN significantly increased the spread rate of

inactivation of lawn biofilms in the case of S. aureus from both cements than VAN alone,

suggesting a possible synergistic effect of the two antibiotics. However, the combination of

antibiotics had no significant effect on increasing the ZOIn with either of the cement

materials against P. aeruginosa (Fig. 6 C-D).

Areas within the ZOIn were swabbed from plates with P. aeruginosa and S. aureus

lawns containing antibiotic beads after day 3. S. aureus showed viable cells present in the

ZOIn with both cement beads containing antibiotics, whereas no growth of P. aeruginosa

Xen41 in the ZOIn was observed. Although this suggests complete eradication of P.

aeruginosa biofilms with antibiotics VAN and/or TOB, a longer antibiotic contact time may

be required to eradicate S. aureus SAP231.

3.6. Body fluids do not influence release from bone cement

The rate of elution of fluorescein from PMMA when coated with blood was

significantly slower (P<0.05) than with no fluids (Fig. 7). Heparinized blood had no influence

on the short-term (8 h) elution of fluorescein from CaSO4 (P˃0.05, data not shown).

19

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

Fig. 7. Release efficiency of fluorescein from CaSO4 (●) and PMMA (▲) beads coated with

or without blood ( ) or synovial fluid ( ).

3.7. Body fluids do not influence ZOI

Antibiotic-incorporated CaSO4 beads had the same ZOI regardless of whether they

were uncoated or were coated with blood or synovial fluid (Fig. 8 A-B).

Fig. 8. White light (A) and no light (luminescence) (B) images showing the influence of

blood or synovial fluid on ZOI of antibiotic (tobramycin) against the bioluminescent strain of

P. aeruginosa Xen41 after 24 h.

20

398

399

400

401

402

403

404

405

406

407

408

409

The ZOI by TOB in the absence of body fluids was 41 ± 0.6 mm against P.

aeruginosa Xen41 and 30.6 ± 0.8 mm against S. aureus SAP231 (Table 1). VAN showed a

ZOI of 23.9 ± 0.8 mm against S. aureus. In the presence of blood or synovial fluid, the ZOI

for TOB against S. aureus was 30.8 ± 0.5 or 30.5 ± 0.7 mm respectively, whereas for VAN

the ZOI was 24.2 ± 1.0 or 23.9 ± 0.9 mm respectively. Against P. aeruginosa, the ZOI with

TOB was 41 ± 0.6 that was statistically non-significant (P=0.39) when the beads were coated

with blood (41.2 ± 0.5mm) and synovial fluid (40.7 ± 0.9 mm) suggesting no influence of

body fluids on antibacterial activity.

4. Discussion

The delivery of local antibiotics for the management of musculoskeletal infection has

become increasingly popular [17, 37]. Initial high local concentration of antibiotic is

beneficial to eradicate resistant bacteria and biofilms that may be present at the site of

infection, even after debridement. High local concentrations of antibiotics facilitate diffusion

to avascular areas. These areas could remain inaccessible to systemic antibiotics and

organisms that are resistant to drug concentrations achieved by systemic antibiotic. Due to the

higher localized drug concentrations from antibiotic-loaded beads, these resistant organisms

may be more susceptible [37-39]. Fluorescein sodium, a freely water soluble and hydrophilic

compound was used as a fluorescent tracer for elution kinetic studies [16]. In this study,

fluorescein-loaded CaSO4 had a greater short-term rate of elution than PMMA (Fig. 4A-B).

The higher rate of elution could be due to differences in micropore structure and the

absorbable feature of CaSO4 as compared to the non-absorbable PMMA [5, 14].

Present and previous studies have shown that packing density is important when

considering the use of antibiotics in bone cements (PMMA and CaSO4) [5, 14], highlighting

the importance that initial release of antibiotics into a tissue or fluid may not be strongly

21

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

influenced by concentration. However, it is important that the packing density is adequate to

ensure spatial distribution of the antibiotic, particularly in areas that are diffusion-limited. If

impregnated CaSO4 beads initially release at the same rate regardless of antibiotic

concentration, it can be concluded that the higher the concentration of antibiotic, the longer

the bead will elute. Though, at high antibiotic impregnation, faster degradation may ensue

due to changes in the mechanical structure and properties of the bone cement beads [15, 29,

40] may lead to faster degradation. It is evident that the low concentration of fluorescein in

CaSO4 increases the short term ultimate compressive strength. At 2.0 g and 4.0 g (per 10cc)

fluorescein, CaSO4 became increasingly brittle and harder to mix, thus forming weaker beads.

When 1.0 g of fluorescein was incorporated in CaSO4, no difference in the ultimate

compressive strength was observed (Fig. 4C). Interestingly, even at 4.0 g there was little

influence of the mechanical strength on short-term release kinetics, even though the

significantly weaker strength suggested changes to the microstructure.

Biofilms are of major concern in PJI infections [13]. Due to the link between the

biofilm phenotype and the establishment of periprosthetic infection, additional studies were

undertaken to determine how effective antibiotic-loaded beads were at killing planktonic and

preformed lawn biofilms. Vancomycin and tobramycin were utilized because these two

antibiotics are commonly used in the treatment of PJI [5, 14]. Vancomycin is a popular

antibiotic due to its potential to prevent Gram-positive bacterial infections, which are mostly

due to methicillin resistant S. aureus (MRSA) [41]. Tobramycin is effective against both S.

aureus and the opportunistic pathogen, P. aeruginosa. When incorporated in a porous carrier

material (CaSO4) the short-term release and efficacy of antibiotics was significantly enhanced

as compared to PMMA. Our study showed that there was a faster initial release from CaSO4

beads than PMMA, as evidenced by the ZOIn and fluorescein spreading patterns (Fig. 4 and

5). This suggests higher killing of lawn biofilms from antibiotic-impregnated CaSO4 beads

22

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

(Fig. 6), which is consistent with the previous reports [5, 39, 42]. Furthermore, in previous

studies, delivery of vancomycin resulted in low release efficiency when delivered using non-

biodegradable materials such as PMMA [43, 44]. The lower antibiotic release from PMMA

could be because of the compact matrix capable of taking up only small quantities of

dissolution fluid into its outermost layers [23, 45]. Gradually, the dissolution fluid can

penetrate the bone cement to dissolve antibiotics within superficial regions of the bone

cement and fails to reach the interior of the bone cement. For example, gentamicin was only

released from the outer 100 μm of PMMA bone cement [45]. CaSO4 on the other hand is

more porous and absorbable than PMMA [5, 42, 46], therefore results in higher rate of

elution of antibiotics. The sustained release of antibiotics from bone cements is largely

affected by the penetration of fluids into the polymer matrix. Host factors such as in-vivo

interaction of the beads with blood, synovial fluid or clotting at the site of surgery could

affect the release and efficacy of antibiotics. Therefore considering the clinical relevance, the

influence of blood and synovial fluid on release kinetics and antibiotic activity from CaSO4

and PMMA beads was studied. Blood and synovial fluid showed no significant effects on

short term release kinetics or antibiotic activity.

All the elution studies in the present investigation were performed by assessing

diffusion into a solid hydrogel agar medium instead of the aqueous ‘infinite medium’

solution. In this case, the advantage of agar method for understanding how antibiotics might

spread into the surrounding tissue is two-fold since the elution distance and antibacterial

effects can be easily monitored, and it preserves the gradients that are likely to develop in-

vivo in areas that are diffusion-limited. However, there are certain limitations of this study,

wherein this is an in-vitro study, and did not evaluate the in-vivo elution kinetics or in-vivo

inhibition of bacterial biofilms. These issues will be examined in future studies with more

focus on clinical applications.

23

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

Antibiotic-infused bone cements are promising materials in fighting and preventing

PJIs. A greater short term rate of elution and antibiotic efficacy was observed with CaSO4

beads as compared to PMMA. Interestingly, this was independent of the loading density and

was not affected by exposure to body fluids. This study highlights the importance of

considering antibiotic loading and packaging density when investigating the clinical

application of bone cements for infection management.

Acknowledgements

We thank Biocomposites Ltd. for providing the Stimulan.

Figure Legends

Fig. 1. Surgical site of infection showing antibiotic loaded calcium sulfate (CaSO4) beads.

Fig. 2. Influence of single and multiple beads on local high concentration of antibiotic. ZOI

against P. aeruginosa PAO1 by multiple TOB-loaded CaSO4 beads after 24 h is relatively

similar as with single bead [16] (A) ††. Representative images showing fluorescein elution

through multiple beads after 24 h was indifferent than with single fluorescein-loaded CaSO4

(B). ††Adapted from [5].

Fig. 3. Short-term release kinetics of fluorescein from CaSO4 beads over a period of time (t in

minutes). Panel A shows fluorescent images of release of fluorescein from CaSO4 beads in

1% agar, and panel B shows the fluorescence intensity (FI) profile at different time intervals

and the radial distance travelled from the fluorescein incorporated CaSO4 bead.

Fig. 4. (A) Fluorescein elution radial distance from CaSO4 beads and PMMA; (B) Influence

of loading concentration of fluorescein in CaSO4 and PMMA beads on short-term release as a

function of time (t1/2); (C) Mechanical properties of fluorescein-loaded CaSO4 beads.

24

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

Fig. 5. Killing kinetics of pre-existing lawn biofilms of (A) P. aeruginosa Xen41 by TOB

incorporated in PMMA and CaSO4 beads and of (B) S. aureus SAP231 by VAN and VAN +

TOB incorporated in PMMA and CaSO4.

Fig. 6. Killing of lawn biofilms of S. aureus and P. aeruginosa by antibiotic beads showing

(A) PMMA loaded with VAN (●) and VAN + TOB ( ) and (B) CaSO4 beads containing

VAN (●) and VAN + TOB ( ) against S. aureus SAP231 and (C) PMMA loaded with TOB

(●) and VAN + TOB ( ) and (D) CaSO4 beads containing TOB (●) and VAN + TOB ( )

against P. aeruginosa Xen41.

Fig. 7. Release efficiency of fluorescein from CaSO4 (●) and PMMA (▲) beads coated with

or without blood ( ) or synovial fluid ( ).

Fig. 8. White light (A) and no light (luminescence) (B) images showing the influence of

blood or synovial fluid on ZOI of antibiotic (tobramycin) against the bioluminescent strain of

P. aeruginosa Xen41 after 24 h.

REFERENCES

[1] S.M. Kurtz, E. Lau, J. Schmier, K.L. Ong, K. Zhao, J. Parvizi, Infection burden for hip and knee arthroplasty in the United States, The Journal of arthroplasty, 23 (2008) 984-991.[2] S.M. Kurtz, E. Lau, H. Watson, J.K. Schmier, J. Parvizi, Economic burden of periprosthetic joint infection in the United States, The Journal of arthroplasty, 27 (2012) 61-65.e61.[3] D.N. Fish, H.M. Hoffman, L.H. Danziger, Antibiotic-impregnated cement use in U.S. hospitals, American journal of hospital pharmacy, 49 (1992) 2469-2474.[4] D.A. Wininger, R.J. Fass, Antibiotic-impregnated cement and beads for orthopedic infections, Antimicrobial agents and chemotherapy, 40 (1996) 2675-2679.[5] S.J. McConoughey, R.P. Howlin, J. Wiseman, P. Stoodley, J.H. Calhoun, Comparing PMMA and calcium sulfate as carriers for the local delivery of antibiotics to infected surgical sites, Journal of biomedical materials research. Part B, Applied biomaterials, 103 (2015) 870-877.[6] C.L. Nelson, The current status of material used for depot delivery of drugs, Clinical orthopaedics and related research, (2004) 72-78.[7] A.E. Gross, The role of polymethylmethacrylate bone cement in revision arthroplasty of the hip, The Orthopedic clinics of North America, 36 (2005) 49-54, vi.

25

509

510

511

512

513

514

515

516

517

518

519

520

521

522

523524

525526527528529530531532533534535536537538539540541

[8] O.S. Kluin, H.C. van der Mei, H.J. Busscher, D. Neut, Biodegradable vs non-biodegradable antibiotic delivery devices in the treatment of osteomyelitis, Expert opinion on drug delivery, 10 (2013) 341-351.[9] S. Gitelis, G.T. Brebach, The treatment of chronic osteomyelitis with a biodegradable antibiotic-impregnated implant, Journal of orthopaedic surgery (Hong Kong), 10 (2002) 53-60.[10] A.S. Coetzee, Regeneration of bone in the presence of calcium sulfate, Archives of otolaryngology (Chicago, Ill. : 1960), 106 (1980) 405-409.[11] K.A. al Ruhaimi, Effect of calcium sulphate on the rate of osteogenesis in distracted bone, International journal of oral and maxillofacial surgery, 30 (2001) 228-233.[12] J.Y. Ferguson, M. Dudareva, N.D. Riley, D. Stubbs, B.L. Atkins, M.A. McNally, The use of a biodegradable antibiotic-loaded calcium sulphate carrier containing tobramycin for the treatment of chronic osteomyelitis: a series of 195 cases, The bone & joint journal, 96-b (2014) 829-836.[13] V.K. Aggarwal, M.R. Rasouli, J. Parvizi, Periprosthetic joint infection: Current concept, Indian Journal of Orthopaedics, 47 (2013) 10-17.[14] B.M. Howlin RP, Webb JS, Cooper JJ, Aiken SS, Stoodley P, Antibiotic-loaded synthetic calcium sulfate beads for prevention of bacterial colonization and biofilm formation in periprosthetic infections, Antimicrobial agents and chemotherapy, 59 (2015) 111-120.[15] G. Lewis, S. Janna, Estimation of the optimum loading of an antibiotic powder in an acrylic bone cement: gentamicin sulfate in SmartSet HV, Acta orthopaedica, 77 (2006) 622-627.[16] S.N. A. Seidlitz, B. Semmling, S. Petersen, T. Reske, W. Schmidt, N. Grabow, K. Sternberg, W. Weitschies, In vitro estimation of drug loss during the implantation procedure of drug-eluting stents Biomedical Engineering, 57 (2012) 403-406.[17] W.A. Jiranek, A.D. Hanssen, A.S. Greenwald, Antibiotic-loaded bone cement for infection prophylaxis in total joint replacement, The Journal of bone and joint surgery. American volume, 88 (2006) 2487-2500.[18] A. Bistolfi, G. Massazza, Vern, #233, E. , Mass, #232, A. , D. Deledda, S. Ferraris, M. Miola, F. Galetto, M. Crova, Antibiotic-Loaded Cement in Orthopedic Surgery: A Review, ISRN Orthopedics, 2011 (2011) 8.[19] K.E. Marks, C.L. Nelson, E.P. Lautenschlager, Antibiotic-impregnated acrylic bone cement, The Journal of bone and joint surgery. American volume, 58 (1976) 358-364.[20] M.J. Askew, M.F. Kufel, P.R. Fleissner, Jr., I.A. Gradisar, Jr., S.J. Salstrom, J.S. Tan, Effect of vacuum mixing on the mechanical properties of antibiotic-impregnated polymethylmethacrylate bone cement, Journal of biomedical materials research, 24 (1990) 573-580.[21] B.A. Masri, C.P. Duncan, C.P. Beauchamp, N.J. Paris, J. Arntorp, Effect of varying surface patterns on antibiotic elution from antibiotic-loaded bone cement, The Journal of arthroplasty, 10 (1995) 453-459.[22] A.S. Baker, L.W. Greenham, Release of gentamicin from acrylic bone cement. Elution and diffusion studies, The Journal of bone and joint surgery. American volume, 70 (1988) 1551-1557.[23] M.J. Penner, B.A. Masri, C.P. Duncan, Elution characteristics of vancomycin and tobramycin combined in acrylic bone-cement, The Journal of arthroplasty, 11 (1996) 939-944.[24] Y.H. Chang, C.L. Tai, H.Y. Hsu, P.H. Hsieh, M.S. Lee, S.W. Ueng, Liquid antibiotics in bone cement: an effective way to improve the efficiency of antibiotic release in antibiotic loaded bone cement, Bone & joint research, 3 (2014) 246-251.

26

542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590

[25] C.M. Stevens, K.D. Tetsworth, J.H. Calhoun, J.T. Mader, An articulated antibiotic spacer used for infected total knee arthroplasty: a comparative in vitro elution study of Simplex and Palacos bone cements, Journal of orthopaedic research : official publication of the Orthopaedic Research Society, 23 (2005) 27-33.[26] J. Jackson, F. Leung, C. Duncan, C. Mugabe, H. Burt, The use of bone cement for the localized, controlled release of the antibiotics vancomycin, linezolid, or fusidic acid: effect of additives on drug release rates and mechanical strength, Drug delivery and translational research, 1 (2011) 121-131.[27] S.S. Aiken, J.J. Cooper, H. Florance, M.T. Robinson, S. Michell, Local release of antibiotics for surgical site infection management using high-purity calcium sulfate: an in vitro elution study, Surgical infections, 16 (2015) 54-61.[28] T. Rimmele, E. Boselli, D. Breilh, S. Djabarouti, J.C. Bel, R. Guyot, M.C. Saux, B. Allaouchiche, Diffusion of levofloxacin into bone and synovial tissues, The Journal of antimicrobial chemotherapy, 53 (2004) 533-535.[29] G. Lewis, Properties of antibiotic-loaded acrylic bone cements for use in cemented arthroplasties: a state-of-the-art review, Journal of biomedical materials research. Part B, Applied biomaterials, 89 (2009) 558-574.[30] A.D. Hanssen, M.J. Spangehl, Practical applications of antibiotic-loaded bone cement for treatment of infected joint replacements, Clinical orthopaedics and related research, (2004) 79-85.[31] M.C. Plaut RD, Prabhakara R, Merkel TJ, Stibitz S, Stably Luminescent Staphylococcus aureus Clinical Strains for Use in Bioluminescent Imaging, PloS one, 8 (2013) 3.[32] T. Babrowski, C. Holbrook, J. Moss, L. Gottlieb, V. Valuckaite, A. Zaborin, V. Poroyko, D.C. Liu, O. Zaborina, J.C. Alverdy, Pseudomonas aeruginosa virulence expression is directly activated by morphine and is capable of causing lethal gut derived sepsis in mice during chronic morphine administration, Annals of Surgery, 255 (2012) 386-393.[33] G.S. Hirt H, Antimicrobial peptide GL13K is effective in reducing biofilms of Pseudomonas aeruginosa., Antimicrobial agents and chemotherapy, 57 (2013) 4903-4910.[34] R.P. Rennie, L. Turnbull, C. Brosnikoff, J. Cloke, First Comprehensive Evaluation of the M.I.C. Evaluator Device Compared to Etest and CLSI Broth Microdilution for MIC Testing of Aerobic Gram-Positive and Gram-Negative Bacterial Species, Journal of clinical microbiology, 50 (2012) 1147-1152.[35] R.E. Duey, A.C. Chong, D.A. McQueen, J.L. Womack, Z. Song, T.A. Steinberger, P.H. Wooley, Mechanical properties and elution characteristics of polymethylmethacrylate bone cement impregnated with antibiotics for various surface area and volume constructs, The Iowa orthopaedic journal, 32 (2012) 104-115.[36] M.G. Ficklin, K.A. Kunkel, J.T. Suber, P.D. Gerard, M.P. Kowaleski, Biomechanical evaluation of polymethyl methacrylate with the addition of various doses of cefazolin, vancomycin, gentamicin, and silver microparticles, Veterinary and comparative orthopaedics and traumatology : V.C.O.T, 29 (2016) 394-401.[37] A.D. Hanssen, Local antibiotic delivery vehicles in the treatment of musculoskeletal infection, Clinical orthopaedics and related research, (2005) 91-96.[38] A. Connaughton, A. Childs, S. Dylewski, V.J. Sabesan, Biofilm Disrupting Technology for Orthopedic Implants: What's on the Horizon?, Front Med (Lausanne), 1 (2014) 22.[39] G.W. Bowyer, N. Cumberland, Antibiotic release from impregnated pellets and beads, The Journal of trauma, 36 (1994) 331-335.[40] Y. He, J.P. Trotignon, B. Loty, A. Tcharkhtchi, J. Verdu, Effect of antibiotics on the properties of poly(methylmethacrylate)-based bone cement, Journal of biomedical materials research, 63 (2002) 800-806.

27

591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639

[41] L.J. Lee SH, Baek WY, Lim JO, Regional delivery of vancomycin using pluronic F-127 to inhibit methicillin resistant Staphylococcus aureus (MRSA) growth in chronic otitis media in vitro and in vivo., J controlled release, 96 (2004) 1-7.[42] P. Udomkusonsri, S. Kaewmokul, S. Arthitvong, N. Phaochoosak, Elution profiles of cefazolin from PMMA and calcium sulfate beads prepared from commercial cefazolin formulations, The Journal of veterinary medical science / the Japanese Society of Veterinary Science, 74 (2012) 301-305.[43] B.K. Li B, Wenke JC, Guelcher SA, Sustained release of vancomycin from polyurethane scaffolds inhibits infection of bone wounds in a rat femoral segmental defect model. J Control Release., J controlled release, 145 (2010) 221-230.[44] C.J. Mader JT, Cobos J. , In vitro evaluation of antibiotic diffusion from antibiotic-impregnated biodegradable beads and polymethylmethacrylate beads. , Antimicrobial Agents and Chemotherapy. , 41 (1997) 415-418.[45] D.J. Schurman, C. Trindade, H.P. Hirshman, K. Moser, G. Kajiyama, P. Stevens, Antibiotic-acrylic bone cement composites. Studies of gentamicin and Palacos, The Journal of bone and joint surgery. American volume, 60 (1978) 978-984.[46] P. Frutos, E. Diez-Pena, G. Frutos, J.M. Barrales-Rienda, Release of gentamicin sulphate from a modified commercial bone cement. Effect of (2-hydroxyethyl methacrylate) comonomer and poly(N-vinyl-2-pyrrolidone) additive on release mechanism and kinetics, Biomaterials, 23 (2002) 3787-3797.

28

640641642643644645646647648649650651652653654655656657658659660661