Magnetic hyperthermia enhance the treatment efficacy of peri-implant osteomyelitis

All chemicals, including 1,2-tetradecanediol, dibenzyl ether solution, oleic acid (C₁₇H₃₃COOH, technical grade 90%), oleylamine (C₁₈H₃₅NH₂, technical grade 70%), Iron(III) acetylacetonate were obtained from (Fe(acac)₃, Sigma-Aldrich, Inc., St. Louis, MO, USA). Fe₃O₄ magnetic nanoparticles were synthesized according to the following procedure. First, 2.0 mmole of Fe(acac)₃, 10 mmole 1,2-tetradecanediol solution was added into 15 mL of dibenzyl ether solution, mixed with stabilizer and dispersant [containing 6 mmole oleic acid (C₁₇H₃₃COOH) and 6 mmole oleylamine (C₁₈H₃₅NH₂)], stirring for 30 min at 130 °C and then then under a blanket of nitrogen, heated to reflux (290 °C) for 1 h. The dibenzyl ether which containing 0.03 M Fe³⁺ acetylacetonate mixture solution was added drop-wise under vigorous mechanical stirring (2000 rpm) for 30 min. The color of the suspension turned to black immediately. The suspension was cooled to room temperature by removing the heat source, ethanol (40 ml) was added to the mixture and a black material was precipitated and separated via centrifugation. The black product was dissolved in hexane in the presence of oleic acid (~0.05 mL) and oleyl amine (~0.05 mL). Centrifugation (6000 rpm, 10 min) was applied to remove any undispersed residue (almost none). The product, 6 nm Fe₃O₄ nanoparticles, was then precipitated with ethanol, centrifuged (6000 rpm, 10 min) to remove the solvent and redispersed into hexane. The precipitate was washed three times with doubly distilled water and then several times with ethanol. The supernatant solution was removed from the precipitate after decantation. Finally, the 6 nm Fe₃O₄ magnetic nanoparticles were dispersed with small amount of oleic acid (~0.05 mL) and oleylamine (~0.05 mL) and stored with hexane in a stoppered bottle for further use. The size and morphologies of products were characterized by transmission electron microscopy (TEM) of HITACHI TEM H-7500 microscope (HITACHI, Japan) with 200 mesh carbon film coated copper grid. The crystal structures of samples were recorded using a Rigaku X-ray powder diffractometer (Geigerflex; Tokyo, Japan).

An AMF was generated by a vertical coil with an inner diameter of 6.5 cm (Coil 1.5 A, 30 turns), driven by a transistor inverter (HOT SHOT; Ameritherm, New York, USA) operated at a frequency of 100 kHz and electric current (EC) 250 A which is capable of generating a 20G magnetic field. Thermal images were taken using a thermograph (infrared thermal imaging camera InfReC R300SR; Nippon Avionics, Tokyo, Japan). Temperature was also measured by using a thermograph.

The magnetic properties of Fe₃O₄ nanoparticles were analyzed by superconducting quantum interference device (SQUID, MPMS-XL7, Quantum Design, USA) at temperature 2 and 300 K with an applied field between −20 and 20 kOe; moreover, zero field cooling (ZFC) and field cooling (FC) measurement were carried out at 500 Oe to determine the blocking temperature (T_b). The magnetic properties analyses were used to determine the blocking temperature (Tb), the temperature above which the magnetic property will change from ferrimagnetism to superparamagnetism, and magnetization curve which showed the saturation magnetization value for examining the feasibility of temperature increase within the AMF.

To evaluate the heating effect in vitro, the magnetic nanoparticles were placed in the bottom of a 1 mL eppendorf tube filled with 0.5 mL PBS. Each loaded eppendorf tube was set in the center of a two-turning copper coil (25 mm in diameter) in an alternating magnetic field (AMF, Power cube 64/900, ~ 750–1150 kHz, Ceia Co., Arezzo, Italy). Heat generation from the samples was determined by measuring the temperature change of PBS every 10 s with an optical fiber probe (STF-1 m) connected with the fluoroptic thermometer (I652, Luxtron, USA). The in vitro hyperthermia effect of the samples were carried out in the AMF as mentioned above and the eppendorf tubes were filled with 0.5 mL FBS. The heating process was performed for 30 s and the temperature was maintained at 74–75 °C by automatically switching an on/off device of the AMF.

The experimental protocol was approved by the Institutional Animal Care and Use Committee of Medical College, National Taiwan University (Taipei, Taiwan). Sixty male 12 weeks-old Wistar rats (300 ± 20 g were purchased from the Laboratory Animal Center, Medical College, National Taiwan University (Taipei, Taiwan) and acclimated under standard laboratory conditions at 22 ± 2 °C and 50 ± 10% humidity. Standard rat chow and water were available ad libitum during the acclimation period. Animals were anesthetized by Zoletil (15-18 mg/kg) and Rompun (0.05 ml/kg). Surgical interventions were performed under strict sterile conditions. The surgical area was shaved and disinfected with iodine, and a midline knee incision was made, followed by medial parapatellar arthrotomy and lateral patella subluxation to expose the knee joint. In the pilot study (24 rats), the osteomyelitis model was built by implanting metallic 18G needle into the bone marrow cavity of distal femur after injecting 10⁵ CFU/50 μl bacterial suspension of Methicillin-sensitive Staphylococcus aureus BCRC10451 (MSSA); while in the sham-operated negative control rats, metallic 18G needle was implanted without injection of MSSA (Fig. 1.). The strain used in this study, S. aureus BCRC10451 (MSSA), was cultivated overnight at 37 °C in brain heart infusion (BHI) broth (Sigma-Aldrich, Inc., St. Louis, MO, USA). The arthrotomy was closed using non-resorbable sutures in an interrupted pattern and then the skin was closed with resorbable suture. Intramuscular injection of cefazolie (20 mg/kg, Purzer Pharmaceutical Co., Ltd., Taipei, Taiwan) was used for postoperative prophylaxis. The formation of biofilm were confirmed by histological staining and scanning electron microscopy. In the pilot study, the rats were randomized to either 10, 20, 30 or 40 days endpoints to check the progression of histopathological change of osteomyelitis. Since the biofilm dormation was quite clear at 20 days after MSSA injection, we selected this time point for the following experiment to evaluate treatment efficacy of hyperthermia.

The evaluation for the treatment efficacy of magnetic hyperthermia, different treatment modalities (total volume 0.5 cm³) was implanted into the bone marrow cavity of distal femur by a metallic 23G needle. All of the 30 animals were divided into 5 groups (Fig. 2): Group I: osteomyelitis positive control without treatment; Group II: osteomyelitis treated with intramuscular injection of vancomycin [VC (20 mg/ kg, i.m., Sandoz Inc., Princeton, NJ, USA)]; Group III: osteomyelitis treated with both intramuscular and femur cavity injection of vancomycin [VC (20 mg/kg, i.m.) + VC (20 mg/ kg, f.c)]; Group IV: osteomyelitis treated with both intramuscular injection of vancomycin and magnetic particles [VC(20 mg/kg, i.m.) + M]; Group V: osteomyelitis treated with intramuscular injection of vancomycin and magnetic particles hyperthermia [VC(20 mg/kg, i.m.) + M + IOHA]. Animals from each group were sacrificed by overdose of pentobarbital at 40 days after surgery.

Bacteriology swabs from the subcutaneous tissue and the implant site were obtained from all operation sites at the time of sacrifice. Swabs were moistened with one drop (20 μl) of sterile PBS, evenly streaked onto agar plate. The samples were incubated for 48 h at the 37 °C under anaerobic conditions. After 48 h’ incubation, the media were analyzed by conventional bacteriological techniques.

The morphology of biofilm was observed by scanning electron microscopy (SEM; Topcon ABT-60, Tokyo, Japan). Briefly, the specimens were fixed with 4% para-formaldehyde (PFA) for 2 h and 2% osmium tetroxide (OsO₄) solution for 1 h, dehydrated in a graded series of ethanol, dried by critical-point drying (CPD) method, and sputter-coated with gold to a thickness film before observation.

For hematoxylin and eosin (HE) staining, randomly chosen femur from the osteomyelitis group was fixed in 5% formaldehyde for two days, decalcified by 5% nitric acid solution, and then embedded with paraffin. The tissues were cut and coronally sliced into 7 μm sections until use. Slices were stained with haematoxylin/eosin. Congo red staining and gram’s staining were also performed using the same sliced tissues. Six visual fields from each section were randomly selected and observed under light microscopy.

For radiographic evaluation, the soft tissue attached to the femur was remove, the rat femurs were examined by X-rays (Siemens, Germany). After radiologic examination, implants were removed and then assessed three dimensionally using Micro-CT (SkyScan 1176; Bruker Micro-CT, Kontich, Belgium) [24]. Datasets were reconstructed using CTvox 2.4 software for fast volumetric reconstruction, 2D/3D quantitative analysis and realistic 3D visualization. The tissue volume (TV: mm³), bone volume (BV: mm³) were recorded and then percent bone volume (BV/TV: %) was analyzed.