In vivo efficacy of tobramycin-loaded synthetic calcium phosphate beads in a rabbit model of staphylococcal osteomyelitis
Osteomyelitis is an acute or chronic inflammatory process of the bone following infection with pyogenic organisms like Staphylococcus aureus. Tobramycin (TOB) is a promising aminoglycoside antibiotic used to treat various bacterial infections, including S. aureus. The aim of this study was to investigate the efficacy of tobramycin-loaded calcium phosphate beads (CPB) in a rabbit osteomyelitis model.
Tobramycin (30 mg/mL) was incorporated into CPB by dipping method and the efficacy of TOB-loaded CPB was studied in a rabbit osteomyelitis model. For juxtaposition, CPB with and without TOB were prepared. Twenty-five New Zealand white rabbits were grouped (n = 5) as sham (group 1), TOB-loaded CPB without S. aureus (group 2), S. aureus only (group 3), S. aureus + CPB (group 4), and S. aureus + TOB-loaded CPB (group 5). Groups infected with S. aureus followed by CPB implantation were immediately subjected to surgery at the mid-shaft of the tibia. After 28 days post-surgery, all rabbits were euthanized and the presence or absence of chronic osteomyelitis and the extent of architectural destruction of the bone were assessed by radiology, bacteriology and histological studies.
Tobramycin-loaded CPB group potentially inhibited the growth of S. aureus causing 3.2 to 3.4 log10 reductions in CFU/g of bone tissue compared to the controls. Untreated groups infected with S. aureus showed signs of chronic osteomyelitis with abundant bacterial growth and alterations in bone architecture. The sham group and TOB-loaded CPB group showed no evidence of bacterial growth.
TOB-incorporated into CPB for local bone administration was proven to be more successful in increasing the efficacy of TOB in this rabbit osteomyelitis model and hence could represent a good alternative to other formulations used in the treatment of osteomyelitis.
KeywordsTobramycin Antibacterial activity Calcium phosphate beads Osteomyelitis
American Type Culture Collection
colony forming units
calcium phosphate beads
erythrocyte sedimentation rate
trypticase soy agar
white blood cells
Postoperative and post-traumatic osteomyelitis is one of the most serious complication with high rates of associated morbidity and has been categorized as difficult-to-treat infections in the orthopaedic field [1, 2]. Tremendous progress has been made in the past few years aimed at preventing osteomyelitis, especially by using antibiotics with different spectrums of activity against specific pathogens. Operative techniques like debridement and muscle flaps have also been employed to overcome this complication [3, 4, 5, 6, 7, 8, 9]. In addition to the systemically administered antibiotics alone, the contemporary prevention strategies using biomaterials like antibiotic-loaded poly(methyl methacrylate) (PMMA) cement spacers and antibiotic-incorporated beads for local administration of antibiotics were comparably more effective in increasing the efficacy of locally-administered antibiotics [10, 11, 12, 13].
Treatment of osteomyelitis with local administration of antibiotics using different delivery system, e.g. biodegradable microsphere containing TOB, PMMA beads containing TOB, in combination with systemic antibiotics are some of the choices of treatment [13, 14]. The continued release of TOB from the microsphere and PMMA beads has been demonstrated in the treatment of osteomyelitis . Microspheres containing TOB has also been reported to contain higher local antibiotic concentration, especially in bone and soft tissues compared to systemic levels [14, 15]. While the use of beads have shown considerable effectiveness in the treatment of osteomyelitis, disadvantages such as poor discharge of antibiotics, the inability to incorporate many useful antibiotics because of heat-instability, involvement of secondary surgery in the removal of cements and lack of biodegradable qualities makes them unviable treatment options [16, 17, 18]. Given that current treatment options for staphylococcal infections have become limited, this creates an urgent need for novel antibiotic delivery systems. Therefore, development of new delivery systems for antibiotics using biomaterials like CPB will not only remove S. aureus from infection sites effectively, but can potentially revert the damages inflicted on the bone by subsidizing mineral components, which is advantageous to new bone formation.
A biodegradable drug delivery system would have the merit of avoiding the removal of drug carrier which normally involves a secondary surgery. Furthermore, the discharge rate of antibiotics can be fine-tuned in a biodegradable system, which can be attributed to the properties of the material. The solubilizing property of the carrier does not prevent new bone from growing into the defective site, and the calcium component of the beads serves to replace the bone mineral.
Biodegradable materials like calcium sulfate, collagen, polymers and other microcapsules have been reported to be suitable drug delivery systems in the treatment of osteomyelitis . In our earlier investigation, we have shown that in vitro incorporation of tobramycin and gentamycin with CPB resulted in a slow residual release of the antibiotic from 30 min to 1344 h (8 weeks) with dissolution of calcium phosphate . However, incorporating TOB with CPB has not been tested in vivo in an animal model for the prevention of osteomyelitis. Therefore, the aim of the present study was to investigate the efficacy of TOB incorporated CPB in a rabbit osteomyelitis model.
Materials and methods
Preparation of CPB loaded with TOB
Commercially available calcium phosphate beads (JectOS® liquid and powder, Kasios®, LʹUnion, France) was prepared according to the methods described previously . Briefly, the beads were prepared by pouring 5 mL of jectos liquid and 3 mL of 1% soluble starch (BDH, United Kingdom) into a sterile bowl. Then 10 g of jectos powder was added and mixed vigorously with spatula until the mixture turned to a smooth homogenous liquid (~ 30 min). Then, using a 1 mL syringe, the liquid paste was filled and slowly extruded on a petri dish to form ~ 0.2 g bead. The prepared beads were allowed to dry overnight and sterilized by gamma irradiation (Malaysia Nuclear Agency) at a measured dose of 25.1 kGy. Tobramycin antibiotic (Sigma Aldrich, St. Louis, MO) was then loaded into CPB by dipping the beads into TOB solution at room temperature (21 °C) and allowing it to stand for 24 h at 4 °C.
Twenty-five specific pathogen free adult male New Zealand White rabbits (Sapphire Enterprise, Selangor, Malaysia) weighing between 2.5 and 3 kg were used in this study. The animals were allowed to acclimatize for 3 weeks before surgery in the animal housing facility of Universiti Putra Malaysia (UPM), in a 12:12 light–dark cycle. All experiments were carried out in accordance with the guidelines of the Institutional Animal Welfare Committee; the study protocol was approved by the Animal Ethical Committee with approval number: UPM/IACUC/AUP-R030/2013. Animals were housed in individual cages and were provided with antibiotic free rabbit diet (Perternakan Hong Lee Sdn Bhd, Malaysia) and tap water ad libitum. Rabbits were divided into five groups with four animals in each group and the details of grouping are as follows: group 1—normal control; group 2—untreated group implanted with TOB-loaded CPB; group 3—untreated group inoculated with S. aureus; group 4—treated group implanted with CPB and inoculated with S. aureus and group 5—treated group implanted with TOB-loaded CPB and inoculated with S. aureus.
Bacterial strain and inoculum preparation
Reference strain of S. aureus ATCC 12600 (S. aureus strain Xen 29, ATCC Manassas, VA, USA) was used in this study . The strain was grown in tryptic soy broth (TSB, Becton–Dickinson and Company) at 37 °C. Prior to surgery, bacterial inoculum containing ~ 107 CFU/mL in phosphate buffered saline (PBS) was prepared. Aliquot of ~ 5 × 107 CFU/mL were obtained and re-suspended to a final volume concentration of 2 × 107 CFU/mL. Cell suspensions were formulated on the day of surgery and held on ice until implantation. From this suspension, a volume of 100 µL of inoculum was injected into the midshaft of the tibia. After surgery, the bacterial load present in 1 g of bone tissue was serially diluted and plated on TSA. After 28 days post-surgery, the presence of bacteria in bone tissues was evaluated based on the colony forming units (CFU) in 1 g of bone.
Osteomyelitis of the tibia was induced by following the procedure described previously . Briefly, the rabbits were weighed and anaesthetized with a combination of 5 mg/kg xylazine (Rompun, Bayer Vital GmbH, Leverkusen, Germany) and 35 mg/kg of ketamine (Ketaset, Zoetis, New Jersey, USA) via subcutaneous injection. Anaesthesia was maintained with free flow inhalation of 1.5% isoflurane (Primal Critical Care Inc., Pennsylvania, USA). Then the rabbits were placed on recumbence and the right knee and hock area was shaved and surgically prepared with chlorhexidine, 70% alcohol and finally with tincture iodine. Prior to surgery, the surgical area was isolated from the rest of the body by placing a sterile surgical drape. A small skin incision of about 3–5 mm was made at the anterior surface of the tibia. The periosteum was sharply incised and bluntly elevated from the midshaft. Using an 18 guage needle, a hole of ~ 0.2 cm was made at the midshaft of the tibia. Subsequently, the midshaft of 12 rabbits were inoculated with 100 µL of S. aureus inoculum containing 2 × 107 CFU/mL by microinjection using a sterile pipette tip with an exterior diameter of 0.6 mm directly into the medullary canal. Immediately after challenging with S. aureus at the surgical site, a prepared CPB was inserted and placed directly over the hole at the midshaft of 12 rabbits (see Additional file 1: Fig. S1). The fascial layer was then replaced to its initial position and closed immediately with a coated VICRYL® 4-0 suture (Polyglactin 910, Ethicon Inc., New Jersey, USA) to prevent leakage of the inoculum and to secure the CPB in place. The control group (n = 4) received plain CPB without TOB. The TOB treated group (n = 4) received a total of 30 mg/mL of TOB. Osteomyelitis infection group (n = 4) were inoculated with S. aureus ATCC 12600 only. Test group (n = 4) were implanted with TOB-loaded CPB without bacteria. And the sham group (n = 4) was operated but left untreated. After surgery, the surgical sites were closed with stitches and disinfected, and all rabbits received 0.05 mg/kg of buprenorphine (PharmaForce Inc., Ohio, USA) as a post-operative analgesic, subcutaneously. Osteomyelitis of the right tibia was allowed to develop and was confirmed by X-ray after 28 days post-surgery. Animals were monitored daily for 28 consecutive days for locomotion status, food and water intake and signs of localized and systematic infection. Changes in body weight and body temperature were also recorded for their correlation to infection.
Post mortem and sample acquisition
On 28 days post infection (dpi), rabbits were euthanized with intravenous injection of 75 mg/kg pentobarbital sodium (Dolethal, Vetoquinol S.A., Lure cedex, France). The tibiae of each rabbit were subjected to radiographical, bacteriological, histological and macroscopic analysis. The infected bone and surrounding soft tissues of the tibia were removed. The right tibia was retrieved and stored in 10% buffered formal saline in airtight glass containers before it was processed for histological analysis. Prior to euthanasia, blood samples were drawn for hemogram analysis and erythrocyte sedimentation rate (ESR) evaluation at 0, 7, 14, 21 and 28 dpi. Bone samples were also obtained from experimental rabbits for bacteriological and histological analysis. The tibiae of the experimental rabbit was radiographed at three stages of the study i.e., before surgery, immediately after surgery and 28 dpi in order to monitor the progression of osteomyelitis.
Two transverse slices of ± 0.5 cm (~ 0.9 g) from the tibia were sawn; the proximal section was used for bacteriological determination, while the distal section was used for histological analysis. The proximal section was homogenized in 20 mL of PBS and the suspension was serially diluted in trypticase soy agar (TSA) plate. After 24 h incubation, the bacterial colony counts were enumerated to represent the CFU/g of bone. The average detection limit was set at 2500 CFU/g of bone and a full-blown infection was characterized by a bacterial load that reached 10% of the initial inoculum dose (> 105 CFU/g). Colonies formed on TSA were further subcultured onto mannitol salt agar for the presumptive detection of S. aureus (Oxoid, Malaysia).
Odekerken’s scoring system was used to confirm the signs of osteomyelitis, where a maximum score of four signifies severe osteomyelitis . The specific morphological changes around the infected midshaft of the tibia were graded as follows: 0—no radiologic abnormalities; 1—mild periosteal reaction with mild osteolysis; 2—periosteal reaction and evident osteolysis; 3—periosteal reaction with calcification in the subperiostium, extension of osteolysis to diaphysis and cortical thickening; 4—extensive metaphyseal osteolysis.
Osteomyelitis scoring system-histology (Masson Goldner’s trichrome stain)
No histopathological changes
Mild periosteal inflammation
Severe periosteal inflammation
Acute intra-medullar inflammation
Chronic intra-medullar inflammation
Microanatomy scoring system (Masson Goldner’s trichrome stain)
Decrease bone formation
Destruction of bone architecture
Increase bone formation
Retention of bone architecture
Hematological data and the changes in body weight and temperature were analyzed by two way analysis of variance (ANOVA) followed by Tukey’s multiple comparison of groups. The differences in bacterial loads between groups were determined by Mann–Whitney U-test for non-parametric two-tailed significance. The radiology, histology and microanatomy data were analyzed by Kruskal–Wallis test followed by Dunnett’s or Tukey’s multiple comparison test. A p value of < 0.05 was considered statistically significant and graphical representation of the data was performed using GraphPad Prism® version 6.01 (GraphPad, San Diego, CA, USA).
Tobramycin-loaded CPB prevented experimental rabbits from developing chronic osteomyelitis
Tobramycin-loaded CPB exhibited strong antibacterial activity against S. aureus
Tobramycin-loaded CPB groups exhibited enhanced new bone formation and remodeling of bone defects
In the present study, we used a prophylactic and a therapeutic model which was partially modified based on previously published models [14, 25]. Moreover, many of the most relevant infection parameters were combined in a single animal model and these parameters were measured on a daily basis for 28 consecutive days. To additionally ensure a valid osteomyelitis model with a full blown infection in the S. aureus inoculated groups, the inoculum size was slightly increased. A 100 µL inoculum of S. aureus containing 106 CFU/mL is ideal to cause high rate of infection in the presence of carrier matrices like PMMA which is much similar to CPB [22, 25]. However, in this study 107 CFU/mL was used which resulted in a full blown infection in the contaminated bead implant groups.
Antibiotic-impregnated beads, including CPB are commonly used to decrease the adverse systemic reactions and obtain local therapeutic levels of antibiotics . In this study, TOB-incorporated CPB, irrespective of the relatively large inoculum (107 cfu), did not result in sepsis development in the animals. Another significance of this study is the excellent placement of the CPB in the tibial cavity. The placement of antibiotic beads has been shown to affect antibiotic tissue levels in the bone . Localized treatment of osteomyelitis involves insertion of antibiotic beads into the tibial cavity thus adding to the clinical significance of this model .
With respect to the antibacterial potency of the TOB-loaded CPB, the results indicate that 30 mg/mL of TOB incorporated CPB was sufficient in inhibiting the growth of S. aureus in vivo, as evidenced by the lower bacterial load in comparison to the controls (p = 0.0014) (Fig. 4). The rabbit group inoculated with S. aureus only demonstrated a trend (p ≤ 0.001) towards increased bacterial growth per gram of bone. This in turn suggests a dose response effectiveness of TOB in vivo (Fig. 4). On the other hand, the group implanted with TOB incorporated CPB without bacterial inoculums showed no bacterial growth in their tibia (Figs. 4, 5a). Similar findings were observed in the sham group.
The dose of TOB incorporated in the CPB (30 mg/mL) was sufficient enough to prevent infection in the experimental rabbits. In our earlier investigation on the in vitro elution and dissolution studies using TOB, incorporating TOB with CPB resulted in a slow residual release of the antibiotic from 30 min to 1344 h (8 weeks) and dissolution of calcium phosphate. Moreover, the mean TOB release in the TOB-incorporated CPB displayed a gradual increase from 10.0 µg/mL (168 h) to 60.7 µg/mL (1344 h) which suggests that the rate of drug release was exponentially related to the release time . This slow and steady release of TOB from CPB can be attributed to the bactericidal action of TOB against S. aureus as observed in this study. Furthermore, the dose of TOB released over time resulted in an amount that was optimal in killing the bacterial load present in 1 g of bone. It is therefore difficult to relate the very high release characteristics of TOB in vitro to the results obtained in vivo. Further confirmation is needed on the steady release and rapid clearance of TOB in the animal system. An earlier investigation by Den Hertog et al.  showed that TOB antibiotic is susceptible to degradation by proteolytic enzymes. This could possibly reduce local tissue levels of TOB in vivo . In the present study, a consolidation of these factors could have resulted in diminished local complementary or active tissue concentrations of tobramycin, feasibly compromising the in vivo efficacy. This underlines the relatively better efficacy of TOB incorporated CPB in preventing osteomyelitis.
Previous studies have shown the importance of antibiotics incorporated to cement spacers or PMMA and their elution properties. One major advantage of these implants is that antibiotics are released from the carriers in such a way that the local levels of antibiotic vastly exceed the minimum inhibitory (MIC) or bactericidal (MBC) concentrations needed to treat most pathogenic organisms, and that these levels are much higher than those achieved with parenteral therapy . Such high concentrations and a steady drug release are important, especially when TOB is used for impregnation of bone cement/CPB, because this aminoglycoside has a peak dose effect for bactericidal activity [30, 31]. A steady release and detectable concentration of TOB from carrier beads for 28 days has also been reported elsewhere .
It should be noted that a slight reduction in CFUs (0.1 log10CFU reduction) was observed in the additional control i.e., group 2 (group receiving CPB with TOB). CPB is not known to possess any antibacterial activity; however, further investigation is required in order to confirm whether the slight reduction in CFUs observed in this group was exerted by CPB or due to immunological factors in the host. Besides, we only evaluated one bacterial species (S. aureus strain Xen 29), which means that the results obtained here may not be extrapolated in the scenario of osteomyelitis caused by clinical S. aureus in hospital settings. One limitation of this study is the use of mannitol salt agar only for the presumptive identification of S. aureus. The use of mass spectrometry MALDI-TOF for the confirmation of the presumptive S. aureus was not carried out in this study and would have confirmed the S. aureus identified. Furthermore, the use of an antibiotic resistant clinical isolate of S. aureus from a osteomyelitis positive patient would have had ample acerbity to steadily generate osteomyelitis in vivo which further broaden the applicability of this acute model. Thus, the data obtained here must therefore be considered only as indicative. We note that, similar in vitro observations have also been made for tobramycin alone and in combination with vancomycin [33, 34, 35], suggesting that detailed efficacy studies may be of interest.
Neutrophils are the first line of response in inflammation. In most bacterial infections, there is an accompanying leucocytosis with neutrophila . Neutrophils recruited to the site of infection release chemotactic factors that attract other neutrophils and leucocytes to the site of injury. In this study, apparent leucocytosis with neutrophila was observed in all the infected groups at 7 dpi, and subsequently in the infected untreated groups at 14 and 21 dpi. However, the group treated with TOB incorporated CPB had lower total WBC and neutrophil count in comparison to the infected untreated groups; this is due to the effect of TOB which resulted in reduced bacterial population at the site of injury and thus lower systemic leucocyte response.
The use of TOB-incorporated CPB at the sites of bone infection can be adopted as a standard care since CPB enabled the localization of supra-MIC levels which might have been difficult to achieve with other substitutes. Tobramycin-incorporated CPB effectively promoted soft tissue and bone healing, and prevented early bacterial colonization by S. aureus in infected tibia. CPB showed promising antistaphylococcal activity in treating chronic osteomyelitis when combined with TOB, with long periods of sustained efficacy. This profile is consistent with the properties required for the development of a potential antibacterial for staphylococcal osteomyelitis and supports continued investigation of this formulation. Future interests in the development of TOB as a traditional ‘drug of choice’ antibiotic will largely be focused on addressing issues related to its pharmacokinetics and inhibition of enzymatic proteolysis. Data obtained from the present study will serve as a basis for more progress in the development of TOB for the treatment of MDR pathogens in hospital settings.
Conceived and designed the experiments: GAL, AK, LMY, FO. Performed the experiments: GAL, AK, LMY. Analyzed the data: GAL, AK, YA. Contributed reagents/materials/analysis tools: GAL, LMY, FO. Wrote the paper, and critically revised the manuscript for intellectual content: GAL, AK, YA, FMF. All authors read and approved the final manuscript.
The authors thank the staffs at Department of Orthopaedics, Traumatology and Rehabilitation, Faculty of Medicine, International Islamic University Malaysia for their assistance with this project.
The authors declare that they have no competing interests.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Ethics approval and consent to participate
This study was approved by the Animal Welfare Committee and the Animal Ethical Committee of Universiti Putra Malaysia (UPM/IACUC/AUP-R030/2013).
This work was supported Universiti Putra Malaysia through the Fundamental Research Grant Schemes (FRGS) 04-01-12-1131 FR funded by the Ministry of Higher Education (MOHE), Putrajaya, Malaysia.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 2.Ambrose CG, Gogola GR, Clyburn TA, Raymond AK, Peng AS, Mikos AG. Antibiotic microspheres: preliminary testing for potential treatment of osteomyelitis. Clin Orthop Relat Res. 2003;415:279–85. https://doi.org/10.1097/01.blo.0000093920.26658.ae.CrossRefGoogle Scholar
- 14.Ambrose CG, Clyburn TA, Louden K, Joseph J, Wright J, Gulati P, et al. Effective treatment of osteomyelitis with biodegradable microspheres in a rabbit model. Clin Orthop Relat Res. 2004;421:293–9. https://doi.org/10.1097/01.blo.0000126303.41711.a2.CrossRefGoogle Scholar
- 27.Den Hertog AL, Wong Fong Sang HW, Kraayenhof R, Bolscher JGM, Vant Hof W, Veerman ECI, et al. Interactions of histatin 5 and histatin 5-derived peptides with liposome membranes: surface effects, translocation and permeabilization. Biochem J. 2004;379(3):665–72. https://doi.org/10.1042/bj20031785.CrossRefGoogle Scholar
- 31.Boulanger S, Mitchell G, Bouarab K, Marsault É, Cantin A, Frost EH, et al. Bactericidal effect of tomatidine-tobramycin combination against methicillin-resistant staphylococcus aureus and Pseudomonas aeruginosa is enhanced by interspecific small-molecule interactions. Antimicrob Agents Chemother. 2015;59:7458–64. https://doi.org/10.1128/AAC.01711-15.CrossRefPubMedPubMedCentralGoogle Scholar
- 32.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;278:244–52.Google Scholar
- 36.Benson KG, Paul-Murphy J. Clinical pathology of the domestic rabbit: acquisition and interpretation of samples. El Paso: Veterinary Clinic of North America: Exotic Animal Practice; 1999. p. 539–51.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.