The HA/α-TCP composite materials were kindly provided by Geistlich Pharma AG (Wolhusen, Switzerland). HA/α-TCP is a composite bilayered material composed of an α-TCP core epitactically coated with a biomimetic nanocrystalline HA layer of a defined thickness (bilayered biphasic calcium phosphate (BBCP)). The sintered core material is obtained by mixing powders of calcium and phosphate donors in appropriate ratios, and sintering the mixture at high temperature. The coating is obtained through the transition of the surface of the sintered TCP core material into biomimetic HA. The biomimetic HA coating of BCP has a crystal size of 18 × 8 x 38 nm, similar to that of natural bone mineral. The following prototypes were produced 3%:97% HA:α-TCP (BBCP1), 12%:88% HA:α-TCP (BBCP2), and 23%:77% HA:α-TCP (BBCP3). All BBCP prototypes were provided in particulate form. The size of all tested particles ranged from 0.7 to 2.0 mm (Fig. 1a). The surface of all particles presented numerous macro-pores. Each bone substitute was separately packed for each defect and sterilized before surgery.
In total, 24 adult female New Zealand white rabbits (3.0–4.0 kg) were purchased from Charles Rivers Laboratories (Romans, France) and subject to calvarial defect assays. The rabbits were housed 2 weeks prior to surgery for the acclimatization and throughout the whole experiment in the Central Animal Care Facility at the University of Berne with an adjusted climate (temperature 19–21 °C, humidity 45% ± 10%, a light:dark cycle of 12:12 h), without excessive or disturbing noises. The animals were fed a standard diet and given water ad libitum. The study was designed as a prospective, randomized study following the ARRIVE guidelines for preclinical in vivo studies (NC3Rs, UK), approved by the Committee for Animal Research, State of Berne, Switzerland (BE 89/17).
Anesthesia and surgery
Rabbits were premedicated with ketamine 65 mg/kg s.c (Narketan®, Vetoquinol AG, Bern, Switzerland), xylazine 4 mg/kg (Xylapan®, Vetoquinol AG, Bern, Switzerland), and buprenorphine 0.03 mg/kg (Temgesic®, Reckitt Benckiser, Wallisellen, Switzerland) in the neck area mixed in the same syringe and left undisturbed for 10–15 min. After reaching the appropriate depth of sedation, the eyes were lubricated (Bepanthen Augen und Nasensalbe, Bayer Vital GmbH, Leverkusen, Germany), and pure oxygen administered by means of a facemask. A 22 G IV catheter (Vasofix® Safety, B. Braun Melsungen AG, Sempach, Switzerland) was inserted in the auricular vein. A size R3 or R4 V-Gel® (Docsinnovent Ltd, London, UK) was inserted and the anesthesia maintained with 0–1% isoflurane (Forene, Abbvie AG, Baar, Switzerland) vaporized in pure oxygen through a Jackson Rees modified T-piece breathing system. Rabbits were spontaneously ventilating throughout the procedure. Monitoring included clinical assessment of the anesthetic depth by a pulse oximetry, capnography, gas analysis, and rectal temperature (Datex-Ohmeda S3, Helsinki, Finland). Clinical variables were recorded at 5 min intervals. Meloxicam 0.3 mg/kg s.c. (Metacam®, Boehringer Ingelheim, Ingelheim, Germany) and penicillin 150,000 IU/mL + benzathine penicillin 150,000 IU/mL s.c. (Duplocillin®, MSD Animal Health, Luzern, Switzerland) 0.01 mL/kg were administered before surgery. NaCl 0.9% (B. Braun Medical AG) was administered at 5 mL/kg/h throughout the procedure.
The surgical area was shaved and disinfected. A local anesthesia was introduced with 0.5 mL of 1% lidocaine HCl (Lidocaïne, Bichsel, Interlaken, Switzerland). A 3.5-cm incision was made from the nasal bone to the mid-sagittal crest. The parietal bone was exposed following the elevation of the periosteum, and two 10-mm diameter bone defects were prepared with a trephine and diamond round burs under copious irrigation with sterile saline. Maximal care was taken to avoid injury of the dura mater.
The following four treatment modalities were randomly allocated: (i) sham, (ii) BBCP1, (iii) BBCP2, and (iv) BBCP3. The weight of all tested materials was about 0.15 g per defect. Allocation of the applied treatment modalities was randomized according to the systematic random protocol (www.randomization.com). The 600 µL of blood was collected from auricular artery per an animal. The 300 µL of blood was used to mix with each granule and implanted into a defect and the 300 µL filled up into the negative control (sham) (Fig. 1b). After implantation of the materials, the 12.5 mm × 13.0 mm-sized resorbable collagen membrane (Geistlich Bio-Gide®, Geistlich Pharma AG, Wolhusen LU, Switzerland) was used to cover the defect site. The periosteum and skin were closed with interrupted sutures in layers using 4–0 Vicryl® and 4–0 Monocryl® sutures (Ethicon, Somerville, NJ, USA). The wound surface was sealed with a spray film dressing (OPSITE® SPRAY, Smith & Nephew, London, UK).
Once the surgery was completed, the rabbits were left to recover under infrared lights and administration of oxygen. Temperature, pulse rate, and respiratory rate were regularly monitored. When fully recovered, the rabbits were returned to the animal facility. Postoperative analgesia included buprenorphine 0.03 mg/kg s.c. every 8 h for 2–3 days and meloxicam 0.3 mg/kg once daily for 5 days. Pain was assessed at regular intervals (composite pain scale and grimace scale) and rescue analgesia was administered if necessary. Water and food consumption, fecal and urine production as well as postoperative weight were monitored.
Following the premedication with ketamine 65 mg/kg and xylazine 4 mg/kg s.c. in the neck area, Esconacron 120 mg/kg i.v. (Streuli Pharma AG, Uznach, Switzerland) was injected for the euthanasia.
The defect sites were removed and fixed in 10% neutral formalin for 7 days at room temperature, then replaced in 70% ethanol at 4 °C, followed by PBS rinsing. Thereafter, the specimens were subjected to micro-CT scans using a desktop Cone-Beam scanner (micro-CT 40, Scanco Medical AG, Brüttisellen, Switzerland). The X-ray source was set at 70 kV with 114 μA at an isotropic voxel size of 18 µm, which showed an image matrix of 2048 × 2048 pixels. The micro-CT images were then analyzed and reconstructed by using 3D structural analysis software (Amira, Visualization Sciences Group, Düsseldorf, Germany). The volume of interest was selected corresponding to the dimensions of the defect sites, with a diameter of 10-mm full-thickness cylinders. Bone volume (BV, mm3), bone density (BD, relative % to initial bone at the defect site), residual material volume (RMV, mm3), and mineralized tissue volume (MTV, mm3; BV + RMV) were calculated after the segmentation.
Histology and histomorphometry
Histology was performed on all prototypes mixed with blood and on all defects. The specimens were trimmed, dehydrated in ascending concentrations of ethanol, and embedded in methylmethacrylate without decalcification. The embedded tissue blocks were cut in a sagittal direction at the middle of the defects into approximately 800-µm thick ground sections using a slow-speed diamond saw (VC-50; LECO, St. Joseph, MI, USA). After mounting on acrylic glass slabs, the sections were ground and polished to a final thickness of 200 µm (Knuth Rotor-3; Struers, Ballerup, Denmark). The labeled bone was digitally photographed under a fluorescent microscope (Nikon Eclipse E800; Nikon, Tokyo, Japan). The sections were stained with toluidine blue combined with fuchsin. The images were photographed under a Wild Heerbrugg M400 ZOOM Makroskop for overviews and a light microscope (Nikon Eclipse E800) equipped with a digital imaging system (NIS Elements; Nikon). Morphometric analysis was performed by a graphic software (Photoshop CS6; Adobe, San Jose, CA, USA) using the corresponding 10-mm initial defect area as the region of interest 1 (ROI 1). Furthermore, each ROI 1 was divided into ROI 2 (central area of a defect) and ROI 3 (peripheral area of a defect) (Fig. 2). The parameters including horizontal defect closure (%), new bone area (NBA, relative % to total augmentation area), bone marrow area (relative % to total augmentation area), residual material area (RMA, relative % to total augmentation area), mineralized tissue area (MTA, NBA + RMA; relative % to total augmentation area), and connective tissue area (CTA, relative % to total augmentation area) were calculated.
For all the quantitative data, the means and all plots of values were represented. The statistical analysis was done by one-way analysis of variance with Tukey test by a statistical program (GraphPad Prism 7.0 software; GraphPad Software, Inc., La Jolla, CA, USA). The p values < 0.05 were considered significant.