Titanium alloy implants in the form of cylindrical dowels (6 × 25 mm) were used in this study. The implants were vacuum plasma sprayed Ti or grit blasted and coated with 80 microns of hydroxyapatite. One implant from each group was used for surface characterisation. Stereozoom images were taken at 1-10x magnifications using an M125C Encoded Stereo Microscope (Leica microsystems, Wetzlar, Germany) and electron microscopy following sputter coating Emitech K575X Sputter Coater (Quorum Technologies, Lewes, United Kingdom) with FEI Nova NanoSEM 230 field-emission scanning electron microscope (FE-SEM) (Thermo Fisher Scientific, Waltham, USA). Implant surface roughness was performed with Olympus DSX510 Digital Microscope (Olympus Corp, Tokyo Japan). A portion of the HA coating was removed with a scalpel blade and absorbance spectra generated between 400 and 4000 cm− 1 based on 64 background subtracted scans using a Spectrum Two FT-IR Spectrometer (PerkinElmer, Waltham, USA). A hydroxyapatite standard was obtained from a local supplier (Sigma-Alderich, CAS: 1306-06-5) and analysed under the same conditions for comparison.
All surgical procedures were performed following institutional ethical clearance on 8 adult cross-bred wethers (2 years old). Animals were received from our open paddock farm and acclimatized for a minimum of 7 days prior to surgery in pairs on deep litter in climate-controlled facilities. Pre-emptive analgesic was provided using transdermal fentanyl patches 24 hours before surgery and to provide smoother sedation and anaesthetic induction [13]. Animals were sedated with an intramuscular (IM) injection of Xylazine (0.2 mg/kg) followed by Ketamine IM (6 mg/kg) 15 minutes later. All animals received 1 g of Cephalothin (18 − 22 mg/kg) intravenously and 5 mL oxytetracycline (200 mg/mL) at 18 to 22 mg/kg intramuscularly. Benacillin (Procaine penicillin 150 mg/mL) 1 mL/10 kg was given IM. The transdermal fentanyl patches were replaced with new ones (to provide a minimum of 72 hours of postoperative analgesia [31]) and Carprofen (Rimadyl 50 mg/mL) at 3 to 4 mg/kg IM given before surgery. Animals were transferred to the operating room table and anaesthesia maintained using on isoflurane (1.5%−3%) and oxygen (2 L/min) throughout the procedures. Animals were allowed weight bearing immediately following recovery from anaesthetic. Animals were monitored and recorded daily for the first 7 days. After 7 days they were monitored daily but only recorded weekly.
This bilateral model allows two cancellous and three bicortical implants per side. Ten dowels (five each side) were implanted using an established osseointegration model in two year old adult cross bred wethers [7, 8, 12, 13, 25–30]. The sample size for this study was 8 implants per group in cortical sites and 5 implants per group in cancellous sites. This sample size has been shown to provide adequate power (Beta error 10%) at Alpha set to 0.05 to detect approximately a 20% difference between groups with a standard deviation of 15%. All implants were randomized in the cortical and cancellous sites. Sites were prepared with saline irrigation during drilling to minimise any thermal damage. Bicortical sites in the anteromedial aspect of the tibia were prepared with a 4.5 mm three-fluted drill (Surgibit, Orthopedic Innovations, Sydney) to create a pilot hole followed by a 6 mm diameter drill-bit for line to line implantation of the dowels. Dowels in the cancellous sites were implanted using a gap model [26] in the cancellous bone of medial distal femoral condyles and proximal tibias. A 4.5 mm three-fluted drill pilot hole was created, over-drilled with 5.5 mm drill to create a press fit for the dowels with the site. A step drill (6, 8 and 10 mm) was used to create a 6 mm hole for the line to line, 8 mm hole for the 1 mm gap and 10 mm hole for a 2 mm gap. The implants were inserted using an impactor into the press fit drilling scenario and centralized in the hole. The periosteum, soft tissues, and dermis were closed in layers using 3 − 0 and 2 − 0 resorbable suture, respectively.
Animals were euthanized following sedation at 4 weeks and 12 weeks. The surgical sites were examined for signs of adverse reaction or infection. The harvested bones were X-rayed in the anteroposterior and lateral views using a Faxitron (Faxitron, Wheeling, IL) and digital plates (AGFA CR MD4.0 Cassette). Radiographs in the anteroposterior and lateral views to determine implant placement, adverse bony reactions as well as evidence of radiographic changes at the implant bone interface. Cancellous sites were isolated using a saw, fixed in cold phosphate buffered formalin and processing using routine polymethylmethacrylate (PMMA) embedding. The cortical sites were isolated using a saw in the axial plane. These samples were sectioned in the sagittal plane to isolate the medial and lateral specimens for push out testing followed by PMMA hard-tissue histology. Prior to mechanical testing the specimens were polished using a Buehler polisher perpendicular to the long axis of the implant to remove any periosteal bone overgrowth.
Implants were tested for implant-bone interface shear strength using a standard push-out test. Specimens were tested at 0.5 mm/min on a calibrated servo-hydraulic testing machine (MTS Mini Bionix®, MTS Systems Inc., Minneapolis, Minnesota, USA). The cortical thickness was obtained from histology images and was used in shear stress calculations following formalin fixation and PMMA embedding. Peak Load, stiffness and energy to failure was determined by plotting of the load-deformation curve and calculated using MatLab script (MATLAB R2016a, MathWorks, Natick, Massachusetts, USA).
The shear stress was calculated according to the following relation:
Equation 1: Shear stress calculation where σ is shear stress, c1 and c2 are cortical thickness on each side of the implant in the histology section and d is implant diameter
Formalin-fixed samples were sequentially dehydrated in increasing concentrations of ethanol before infiltration in methylmethacrylate and polymerization using established techniques. Embedded cortical and cancellous dowels were sectioned along the long axis of the implants using a Leica SP 1600 Microtome (Leica Biosystems, Nussloch, Germany). A minimum of two thin (~ 15 − 20 micron) sections were cut from each dowel. The sections were briefly etched in acidic ethanol (98 mL ethanol 96% and 2 mL HCl 37%) and stained with methylene blue followed by basic fuchsin. The stained slides were reviewed under low magnification to provide an overview of the section and histomorphometry. The implant-bone interface and local reactions were carefully examined at higher magnification for the presence of inflammatory cells or local particulate in the cancellous and cortical sites. The cancellous sites were also examined based on the implant conditions at press-fit, line to line, 1- and 2-mm gaps for local reactions.
PMMA images at the bone implant interfaces were used to determine bone ongrowth and percentage of bone-contact with the implant [12, 13, 26, 27]. The proximal and distal bone – implant interfaces of the cortical bone were evaluated for each slide and a mean value based on 2 slides per site was used for statistical analysis. Similarly, the proximal and distal bone – implant interface in the cancellous sites was used to provide a mean bone ongrowth value for each implantation condition.
Mechanical and histomorphometric data was analysed using a two-way Analysis of Variance (implant, and time) for cortical as well as cancellous sites and post-hoc testing when appropriate using SPSS (version 25, IBM, Armonk, New York, USA).