Enhanced osseointegration through direct energy deposition porous coating for cementless orthopedic implant fixation

Specimen preparation

For comparison, we prepared three types of test coupons based on different manufacturing technologies (DED, PBF and TPS). Each test coupon was cylindrical in shape and consisted of a solid substrate and a porous coating surface layer. For in vitro tests, the solid substrates were made with a diameter of 14.6 mm and a height of 4.0 mm, and for in vivo tests, with a diameter of 6 mm and a height of 2.0 mm. The PBF porous coating surface layer was designed to have a 1.0 mm thickness, which was also the target for the DED and TPS layers (Supplementary Fig. S5).

To fabricate DED specimens, a porous surface layer was laminated on CoCr solid substrate using a DED metal 3D printer (MPC-mini, Insstek, Korea) with pure Ti powder (grade 2, ASTM F1580). The process parameters were set as follows: 70 W laser power, 1500 mm/min scan speed and 2.2 g/min powder feeding rate. To design the porous part of PBF specimens, we scanned an actual cancellous structure in the proximal tibia of a rabbit using a micro computed tomography (micro-CT) system (TVX-IMT225-RC-S2, Tech Valley, Republic of Korea) and employed commercial reconstruction software (Mimics 22.0 & 3-matic 14.0, Materialise, Belgium). Then, the specimen was fabricated using a PBF metal 3D printer (SLM280HL, SLM solution, Germany) with Ti-6Al-4V powder (grade 9, ASTM F136). The process parameters were as follows: 350 W laser power and 1400 mm/s scan speed. In the TPS, the porous layer was laminated on a CoCr solid substrate using a commercially available TPS coating with pure Ti powder (grade 2, ASTM F1580). The Ti powder was injected into the plasma gas stream generated with an electric arc gun with a temperature of about 20,000 °C. Then, the molten Ti particles impacted onto the CoCr substrate with high kinetic energy and were coated.

Characterization of porous structure

Cross-sectional images of specimens were obtained via micro-CT scanning (SkyScan1173, Bruker, Belgium). The pore size of each porous surface coating type was measured by manually segmenting the pores from the cross-sectional images using ImageJ software (National Institutes of Health, USA). Then, based on these 2D cross-sectional images, we reconstructed 3D shapes to measure the porosity and layer thickness of the specimens using Mimics 22.0 and 3-matic 14.0 software (Materialise).

Sample size calculation for in vitro and in vivo studies

For reliable in vitro and in vitro studies, we determined the minimum sample size using G*Power 3.1 software (Heinrich-Heine University, Dusseldorf, Germany), based on similar studies that we have performed previously^41,42,43. The minimum sample size for each in vitro study was calculated (α = 0.05, β = 0.2) based on the mean and standard deviation of the OD values of the CCK-8 assay, and the minimum sample size for each in vivo study was determined based on the mean and standard deviation of BIC. For each in vitro study, six experiments were needed as the minimum requirement to ensure 80% power; hence, to meet the minimum requirement, every in vitro experiment was repeated approximately nine times. For each in vivo study, five experiments with each test group were needed as the minimum requirement to ensure 80% power. Therefore, five rabbits were randomly allocated to six different experimental conditions: DED 6 weeks, DED 12 weeks, PBF 6 weeks, PBF 12 weeks, TPS 6 weeks, and TPS 12 weeks.

Fatigue cycle test

The fatigue cycle was measured via three-point bending tests using an Instron^® 8872 system (Instron, USA). The specimens were mounted to a jig attached to the Instron^® 8872 with the porous layer facing down. The fatigue tests were carried out at a frequency of 10 Hz and a stress ratio R of 0.1 (minimum stress 103.33 MPa and maximum stress 1033.33 MPa).

Cell preparation for in vitro study

The human osteoblast cell line, Saos-2, was purchased from the Korean Cell Line Bank (KCBL). Saos-2 cells were cultured in Minimum Essential Medium (MEM, Welgen) supplemented with 10% fetal bovine serum (FBS, #S001-01, Welgen, Republic of Korea) and 1% anti-anti (Gibco, Thermo Fisher Scientific, Inc., Waltham, MA, USA).

DED, PBF, and TPS test coupons were prepared and cleaned in six steps for in vitro experiments. Firstly, the coupons were sonicated twice in a solution with 1% Solujet (Alconox) at 45 °C. Then, the mixture was sonicated in distilled water twice at 45 °C. After sonication, the coupons were autoclaved at 120 °C for 30 min and dried in a dry oven at 60 °C. Then, each experimental coupon was placed in a 24-well plate for cell culturing. To minimize the number of cells that adhered to the surface of the plastic wells instead of the surface of the coupons, a non-surface-treated 24-well plate (#32024, SPL Life Science) was used and the coupons were designed to fit tightly into the 24-well plate hole. For the positive control, cells were cultured directly on a surface-treated 24-well plate (#30024, SPL Life Science). Saos-2 cells were seeded at a density of 2.4 × 10⁴ cells/well and incubated in a CO₂ incubator at 37 °C.

Cell proliferation and alkaline phosphatase (ALP) activity assays

Cell proliferation rates were assessed using CCK-8 (Dojindo Molecular Technologies, USA). Cell proliferation was tested at 3, 5, 7, 10, or 14 days of culture. At each time point, CCK-8 reagent (50 μL) was added to each well of 24-well plates, and the plates were incubated in a CO₂ incubator at 37 °C for 90 min. The OD value was measured using microplate reader (BioTek™ Eon™ Microplate Spectrophotometers, USA) at 450 nm.

An ALP assay kit was purchased from Abcam (UK). After 5 or 14 days of cell culturing, cells were lysed twice with ALP assay buffer on ice for 10 min. The lysate was collected and centrifuged at 3000 rpm and 4 °C for 15 min. Then, the ALP assay kit was used, in accordance with the kit manual. The lysate was injected into a 96-well plate and 5 mM p-Nitrophenyl Phosphate (pNPP) solution was added. The Lysate-pNPP mixture was incubated at 25 °C for 60 min. The reaction was stopped by adding stop solution. The OD was measured using a microplate reader at 405 nm wavelength.

Immunofluorescence imaging

After 14 days of culturing, cells were washed twice with phosphate-buffered saline (PBS), pH 7.4, fixed with 3.7% formaldehyde/PBS solution for 20 min at room temperature (RT), permeabilized with 0.2% Triton X-100 (Sigma, USA) for 15 min, and washed twice with PBS. To reduce non-specific binding, 5% normal goat serum/PBS was added to the sample, which was then incubated for 1 h at RT^44,45. Rhodamine phalloidin (Invitrogen, Grand Island, NY, USA) in 5% normal goat serum/PBS was used to label actin stress fibers, and 5 μg/mL Hoechst 33342 (Sigma, USA)/PBS solution were used to label nucleus⁴⁶. Since an inverted confocal microscope was used, we dropped the mounting solution (Vectashield^® Vector Labs., USA) and placed the coupons upside down on the cover glass bottom dish (SPL Life Science) to ensure that the cell-attached-surface was facing downward. The cells were visualized using a wide-field fluorescence microscope (DMi8, Leica, Germany) and a confocal microscope (TCS SP5, Leica, Germany). Using confocal microscopy, we detected the Hoechst-33342-stained nuclei with 450 nm emission/400 nm excitation wavelengths, and the topography of the specimen surface with the laser light reflected (580 nm emission/580 excitation) from the surface.

Quantification of cell adhesion at initial stage

To compare the properties of different types of porous surfaces at the initial stage of cell-surface adhesion, the cells attached to the specimens were quantified 6 h after seeding. Saos-2 cells were seeded on DED, TPS, PBF, and well plate (as positive control) surfaces at a density of 7.2 × 10⁴ cells/well, and incubated for 6 h. After incubation, the coupon surfaces and well plate surfaces were washed with PBS vigorously to remove the non-adherent cells. The coupons were then placed in a new 24-well plate, and each well was filled with the cell culture medium and CCK-8 reagent. All the specimens were incubated with CCK-8 reagent for 90 min in a CO₂ incubator at 37 °C, and the OD values of the media were measured using a microplate reader at 450 nm.

Animal preparation

Thirty-six-week-old female New Zealand white rabbits with an average weight of approximately 4 kg were used in this study. The rabbits were individually housed in a temperature- (23 ± 2 °C), humidity- (60% ± 10%), and light-controlled environment, and provided with food and water ad libitum under a 12 h light cycle⁴¹. Animal care, surgical procedures, and all experimental protocols with in vivo animal model were performed in accordance with the National Institutes of Health guide for the care and use of laboratory animals. The study protocol was approved by the Ethics Committee on Animal experimentation at Samsung Medical Center (SMC 2018-0713-002). Also, this study was carried out in compliance with the ARRIVE guidelines. In total, thirty experiment animals were used for six different experiments: 6 weeks of DED (n = 5), TPS (n = 5), PBF (n = 5) and 12 weeks of DED (n = 5), TPS (n = 5), PBF (n = 5).

Surgical procedure

General anesthesia was induced by intramuscular injection of ketamine (700 μL/kg) and xylazine hydrochloride (200 μL/kg). The right knee of each rabbit was shaved and sterilized with povidone-iodine (Fig. 6a). In the supine position, the right legs were incised longitudinally from 2 cm above the knee joint to 1.5 cm below (Fig. 6b).

On the superomedial side of the patella, the vastus medialis muscle was incised through the medial side patella and patella tendon to the proximal end of the tibial tuberosity. This exposed the trochlear groove and the condyle of the femur, with the patella sliding to the lateral side. A hole in the proximal side of the trochlear groove was created with a 6 mm trephine burr, taking care to ensure that the hole was gently reamed (Fig. 6c). Normal saline was sprayed during the reaming procedure to prevent thermal injuries around the bone and soft tissue. An experimental specimen was placed in the hole of the trochlear groove with the porous surface facing the cancellous bone (Fig. 6d). We gently impacted the specimen to facilitate contact with the cancellous bone (Fig. 6e). After implantation, patella reduction was performed. After checking the reduction status and knee motion, we repaired the joint capsule and subcutaneous tissue with Vicryl 2-0 (Fig. 6f). Finally, the wound was disinfected with povidone–iodine⁴¹.

Postoperative care and sacrifice

After surgery, experimental rabbits were administered 0.6 mL/kg of cefazoline (Chongkundang, Seoul, Korea) and 1.8 mL/kg ketoprofen (UNIBIO tech, Seoul, Korea) intramuscularly thrice daily for 3 days. The rabbits were allowed to act freely within their cages after surgery. Subsequently, the rabbits were sacrificed after the planned number of weeks following implantation surgery. We injected ketamine (700 μL/kg) and xylazine hydrochloride (400 μL/kg) intramuscularly; this was followed by an intravenous injection of potassium chloride. Then, the right side of the distal femur was harvested, and the specimens were fixed in 10% neutral buffered formalin (Sigma-Aldrich Corp., St. Louis, MO, USA) for 2 weeks..

Histologic slide manufacturing and staining

Specimens were cleaned with distilled water, and decalcification was performed using ethylenediaminetetraacetic acid (EDTA) solution (pH 9.0) (Zytomed systems GmbH, Berlin, Germany) for 5 weeks. After confirming the removal of calcium, the specimens were embedded in paraffin and sectioned to a thickness of 50 µm with a hard tissue slicer (Struers, Willich, Germany)⁴⁷. The sections were stained with hematoxylin and eosin (Sigma-Aldrich) and Masson’s trichrome (Sigma-Aldrich) stain to visualize the contact surface and osseointegration. General specimen imaging and histomorphometric analyses were conducted at 40 × magnification (Supplementary Fig. S6)⁴¹.

Bone histomorphometry

Light microscopy images were obtained using 12.5 × and 100 × objective lenses (BX 51, Olympus, Tokyo, Japan). The images were captured using a digital camera (CC-12, Soft Imaging System GmbH, Munster, Germany) attached to the microscope⁴⁸. Specimens from different implants were analyzed using (1) Bone to implant contact (BIC): the percentage of direct contact surface between mineralized bone and the Ti porous coating surface (Supplementary Fig. S7a); (2) absent area: the percentage of non-contact area within the total area in a 1000-µm region (Supplementary Fig. S7b); (3) bone area (500 µm): the percentage of new bone formation and neovascularization area within the total area in a 500-µm region; (4) bone area (1000 µm): the percentage of new bone formation and neovascularization area within the total area in a 1000-µm region; and (5) bone area (2000 µm): the percentage of new bone formation and neovascularization area within the total area in a 2000-µm region (Supplementary Fig. S7c)^47,49,50.

Statistical analysis

The Kruskal–Wallis test was used to compare differences between experimental groups. For comparison between each experimental group, multiple Mann–Whitney U tests were used and adjusted with the Benjamini–Hochberg procedure⁵¹. For analysis within each group from 6 to 12 weeks, the Wilcoxon signed rank sum test was used. All analyses were performed using SPSS^® 25.0 software (SPSS, Chicago, IL, USA). A p value < 0.05 was considered significant.