Preparation of a biphasic scaffold for osteochondral tissue engineering

Tissue engineering has been developed as a prospective approach for the repair of articular cartilage defects. Engineered osteochondral implants can facilitate the fixation and integration with host tissue, and therefore promote the regeneration of osteochondral defects. A biphasic scaffold with a stratified two-layer structure for osteochondral tissue engineering was developed from biodegradable synthetic and naturally derived polymers. The upper layer of the scaffold for cartilage engineering was collagen sponge; the lower layer for bone engineering was a composite sponge of poly(DL-lactic-co-glycolic acid) (PLGA) and naturally derived collagen. The PLGA–collagen composite sponge layer had a composite structure with collagen microsponge formed in the pores of a skeleton PLGA sponge. The collagen sponge in the two respective layers was connected. Observation of the collagen/PLGA–collagen biphasic scaffold by scanning electron microscopy (SEM) demonstrated the connected stratified structure. The biphasic scaffold was used for culture of canine bone-marrow-derived mesenchymal stem cells. The cell/scaffold construct was implanted in an osteochondral defect in the knee of a one-year old beagle. Osteochondral tissue was regenerated four months after implantation. Cartilage- and bone-like tissues were formed in the respective layers. The collagen/PLGA–collagen biphasic scaffold will be useful for osteochondral tissue engineering.     

Long term results after implantation of tissue engineered cartilage for the treatment of osteochondral lesions in a minipig model

In present study we determined the long term in vivo integration and histological modeling of an in vitro engineered cartilage construct. Tissue engineered autologous cartilagenous tissue was cultured on calcium phosphate cylinders and implanted into osteochondral defects into the femoral condyles in minipigs. Radiological follow-up was performed at 2, 8, 26 and 52 weeks, condyles were harvested 26 and 52 weeks post-implantation. Thickness of cultivated tissue (1.10 ± 0.55 mm) was comparable to in situ cartilage and cells produced in vitro cartilage specific proteins. In vivo, 26 and 52 weeks post-implantation defects were resurfaced with hyaline-like tissue, the implants were well integrated with no gap at the interface between the engineered neocartilage and the adjacent articular cartilage. Synthesis of type II collagen was detected 26 and 52 weeks after implantation. The modified ICRS score increased from 26 to 52 weeks. Histomorphometric evaluation revealed a decrease in cellularity in tissue engineered cartilage from 2.2-fold of native cartilage after 26 weeks to 1.5-fold after 52 weeks. In conclusion, these findings demonstrate the integration and maturation of tissue engineered cartilage pellets attached on a bone substitute carrier implanted in osteochondral defects over a long time. J. P. Petersen, P. Ueblacker, C. Goepfert have contributed equally to this study.     

Intra-articular administration of hyaluronic acid increases the volume of the hyaline cartilage regenerated in a large osteochondral defect by implantation of a double-network gel

Implantation of PAMPS/PDMAAm double-network (DN) gel can induce hyaline cartilage regeneration in the osteochondral defect. However, it is a problem that the volume of the regenerated cartilage tissue is gradually reduced at 12 weeks. This study investigated whether intra-articular administration of hyaluronic acid (HA) increases the volume of the cartilage regenerated with the DN gel at 12 weeks. A total of 48 rabbits were used in this study. A cylindrical osteochondral defect created in the bilateral femoral trochlea was treated with DN gel (Group DN) or left without any implantation (Group C). In both Groups, we injected 1.0 mL of HA in the left knee, and 1.0 mL of saline solution in the right knee. Quantitative histological evaluations were performed at 2, 4, and 12 weeks, and PCR analysis was performed at 2 and 4 weeks after surgery. In Group DN, the proteoglycan-rich area was significantly greater in the HA-injected knees than in the saline-injected knees at 12 weeks (P = 0.0247), and expression of type 2 collagen, aggrecan, and Sox9 mRNAs was significantly greater in the HA-injected knees than in the saline-injected knees at 2 weeks (P = 0.0475, P = 0.0257, P = 0.0222, respectively). The intra-articular administration of HA significantly enhanced these gene expression at 2 weeks and significantly increased the volume of the hyaline cartilage regenerated by implantation of a DN gel at 12 weeks. This information is important to develop an additional method to increase the volume of the hyaline cartilage tissue in a potential cartilage regeneration strategy using the DN gel.     

Repair of large osteochondral defects in a beagle model with a novel type I collagen/glycosaminoglycan-porous titanium biphasic scaffold

The limited repair potential of articular cartilage, which hardly heals after injury or debilitating osteoarthritis, is a clinical challenge. The aim of this work was to develop a novel type I collagen (Col)/glycosaminoglycan (GAGs)-porous titanium biphasic scaffold (CGT) and verify its ability to repair osteochondral defects in an animal model with bone marrow stem cells (bMSCs) in the chondral phase. The biphasic scaffold was composed of Col/GAGs as chondral phasic and porous titanium as subchondral phasic. Twenty-four full-thickness defects through the articular cartilage and into the subchondral bone were prepared by drilling into the surface of the femoral patellar groove. Animals were assigned to one of the three groups: 1) CGT with bMSCs (CGTM), 2) only CGT, and 3) no implantation (control). The defect areas were examined grossly, histologically and by micro-CT. The most satisfied cartilage repairing result was in the CGTM group, while CGT alone was better than the control group. Abundant subchondral bone formation was observed in the CGTM and CGT groups but not the control group. Our findings demonstrate that a composite based on a novel biphasic scaffold combined with bMSCs shows a high potential to repair large osteochondral defects in a canine model.     

Long-term in vivo response to citric acid-based nanocomposites for orthopaedic tissue engineering

The disadvantages of current bone grafts have triggered the development of a variety of natural and synthetic bone substitutes. Previously, we have described the fabrication, characterization, and short-term tissue response of poly(1,8-octanediol-co-citrate) (POC) with 60 weight % hydroxyapatite nanocrystals (POC-HA) at 6 weeks. In order to better understand the clinical potential, longer term effects, and the biodegradation, biocompatibility, and bone regenerative properties of these novel nanocomposites, POC-HA, POC, and poly-L-lactide (PLL) were implanted in osteochondral defects in a rabbit model and assessed at 26 weeks. Explants were stained with Masson Goldner Trichrome and the fibrous capsule and tissue ingrowth measured. In addition, the bone-implant and bone-cartilage response of POC-HA, POC, and PLL were assessed through histomorphometry and histological scoring. Upon histological evaluation, both POC-HA and POC implants were biocompatible, but PLL implants were surrounded by a layer of leukocytes at 26 weeks. In addition, due to the degradation properties of POC-HA, tissue grew into the implant and had the highest area of tissue ingrowth although not statistically significant. Histomorphometric analyses supported a similar osteoid, osteoblast, and trabecular bone surface area among all implants although the fibrous capsule thickness was the largest for POC. Moreover, histological scoring demonstrated comparable scores among all three groups of the articular cartilage and subchondral bone. This study provides the long-term bone and cartilage response of novel, citric acid-based nanocomposites and their equivalence to FDA-approved biomaterials. Furthermore, we provide new insights and further discussion of these nanocomposites for orthopaedic applications.     

Joint immobilization inhibits spontaneous hyaline cartilage regeneration induced by a novel double-network gel implantation

We have recently discovered that spontaneous hyaline cartilage regeneration can be induced in an osteochondral defect in the rabbit, when we implant a novel double-network (DN) gel plug at the bottom of the defect. To clarify whether joint immobilization inhibits the spontaneous hyaline cartilage regeneration, we conducted this study with 20 rabbits. At 4 or 12 weeks after surgery, the defect in the mobile knees was filled with a sufficient volume of the hyaline cartilage tissue rich in proteoglycan and type-2 collagen, while no cartilage tissues were observed in the defect in the immobilized knees. Type-2 collagen, Aggrecan, and SOX9 mRNAs were expressed only in the mobile knees at each period. This study demonstrated that joint immobilization significantly inhibits the spontaneous hyaline cartilage regeneration induced by the DN gel implantation. This fact suggested that the mechanical environment is one of the significant factors to induce this phenomenon.     

A Three‐Layered Osteochondral Plug: Structural,Mechanical, and in vitro Biocompatibility Analysis

This study aims to produce an osteochondral plug with three distinct layers resembling the naturally occurring cartilage, tidemark, and subchondral zones, for the regeneration of defects of articular cartilage. The bone layer is constructed from a PLLA/PCL polymeric blend using a dual‐porogen approach. The pore surfaces are coated with type‐I collagen and hydroxyapatite. The upper layer, made of PGA nonwoven felt, is combined with the lower using a polymeric blend with a pigment for better visualization during implantation. Vertical channels are formed from the bottom layer to the upper border of the tidemark to facilitate the delivery of stem cells and blood from the bone marrow when implanted. The Young's modulus of the osteochondral plugs is 94.5 ± 9.42 kPa. Cell‐culture studies confirm the biocompatibility.     

A one-step treatment for chondral and osteochondral knee defects: clinical results of a biomimetic scaffold implantation at 2 years of follow-up

The increasing interest in the role of subchondral bone with regard to articular surface disease led to the development of new bioengineered strategies. Aim of this study is to evaluate the clinical and MRI outcome after the implantation of a nanostructured biomimetic three-phasic collagen–hydroxyapatite construct for the treatment of chondral and osteochondral defects of the knee in a large cohort of patients. Seventy-nine patients (63 M, 16 W), affected by grade III–IV femoral condyle or trochlea chondral lesions or osteochondritis dissecans (OCD) were consecutively treated. Mean age was 31.0 ± 11.3 years, mean lesion size was 3.2 ± 2.0 cm². Fifty patients underwent previous surgeries, concurrent procedures were necessary in 39 cases. The clinical outcome was evaluated using the IKDC and Tegner scores at 12 and 24 months of follow-up. At follow-up times an MRI was performed and evaluated with the MOCART score. All the scores improved significantly from the baseline. IKDC subjective score showed a further increase between 12 and 24 months of follow-up, and 82.2 % of the patients improved their symptoms at the final evaluation. Patients affected by OCDs had better results than those with degenerative lesions. Some abnormal MRI findings were present, even though no correlation was found with the clinical outcome. This one-step biomimetic approach developed to favor osteochondral tissue regeneration is effective in treating knees affected by damages of the articular surface, leading to a significant clinical improvement. However, abnormal MRI findings were present, even if not correlated with the clinical outcome.     

Novel alginate biphasic scaffold for osteochondral regeneration: an in vivo evaluation in rabbit and sheep models

Current therapeutic strategies for osteochondral restoration showed a limited regenerative potential. In fact, to promote the growth of articular cartilage and subchondral bone is a real challenge, due to the different functional and anatomical properties. To this purpose, alginate is a promising biomaterial for a scaffold-based approach, claiming optimal biocompatibility and good chondrogenic potential. A previously developed mineralized alginate scaffold was investigated in terms of the ability to support osteochondral regeneration both in a large and medium size animal model. The results were evaluated macroscopically and by microtomography, histology, histomorphometry, and immunohistochemical analysis. No evidence of adverse or inflammatory reactions was observed in both models, but limited subchondral bone formation was present, together with a slow scaffold resorption time.The implantation of this biphasic alginate scaffold provided partial osteochondral regeneration in the animal model. Further studies are needed to evaluate possible improvement in terms of osteochondral tissue regeneration for this biomaterial.     

Evaluation of a press-fit osteochondral poly(ester-urethane) scaffold in a rabbit defect model

The purpose of this study was to evaluate the impact on osteochondral healing of press-fitted multiphasic osteochondral scaffolds consisting of poly(ester-urethane) (PUR) and hydroxyapatite into a cylindric osteochondral defect in the distal non-weight bearing femoral trochlear ridge of the rabbit. Two scaffolds were investigated, one with and one without an intermediate microporous membrane between the cartilage and the bone compartment of the scaffold. A control group without a scaffold placed into the defect was included. After 12 weeks macroscopic and histomorphological analyses were performed. The scaffold was easily press-fitted and provided a stable matrix for tissue repair. The membrane did not demonstrate a detrimental effect on tissue healing compared with the scaffold without membrane. However, the control group had statistically superior healing as reflected by histological differences in the cartilage and subchondral bone compartment between control group and each scaffold group. A more detailed analysis revealed that the difference was localized in the bone compartment healing. The present study demonstrates that an elastomeric PUR scaffold can easily be press-fitted into an osteochondral defect and provides a stable matrix for tissue repair. However, the multi-phasic scaffold did not provide a clear advantage for tissue healing. Future investigations should refine especially the bone phase of the implant to increase its stiffness, biocompatibility and osteoconductive activity. A more precise fabrication technique would be necessary for the matching of tissue organisation.     

E-PTFE in rabbit knee-joints

The healing of expanded polytetrafluoroethylene (e-PTFE) in articular cartilage and bone was studied. A 1×4 mm osteo-chondral defect was created in the medial femoral condyle in 10 rabbits (20 knee-joints). A correspondingly broad strip of e-PTFE was placed in the defects and pulled through two drilled channels to the dorso-lateral side of the condyle. The contra-lateral knee-joint served as control. The animals were not immobilized and allowed to move about freely together in a room. The animals were killed by perfusion fixation after 14 months, the implants and tissues retrieved en bloc and examined with scanning electron microscopy (SEM) and light microscopic (LM) morphometry. No macroscopic signs of inflammation were detected in the knee-joints. Observations with SEM in control joints showed that the articular surface ranged from smooth to irregular with superficial crevices and fibrillations at the site of the defect. The smooth articular surface of the surrounding articular cartilage partly overlapped the e-PTFE membrane. The surface of the e-PTFE membrane had a nodular character and was surrounded by fibrocartilage with clusters of chondrocytes. A consistentobservation was the large amount of bone around and in direct contact with the surface of e-PTFE membrane. LM morphometry of intace e-PTFE-tissue specimens in three different section planes showed that 73.1% and 8.8% of the implant surface was in contact with bone and bone marrow, respectively. Our morphological observations of e-PTFE in the cartilage and bone of the rabbit knee-joint after a 14-months healing period indicate that e-PTFE could be a useful material in reconstructive surgery of smaller non-weight-bearing joints.     

Correlation between primary stability and bone healing of surface treated titanium implants in the femoral epiphyses of rabbits

The aim of this study was to analyse the stability and osseointegration of surface treated titanium implants in rabbit femurs. The implants were either grit-blasted and acid-etched (BE Group), calcium phosphate (CaP) coated by using the electrodeposition technique, or had bioactive molecules incorporated into the CaP coatings: either cyclic adenosine monophosphate (cAMP) or dexamethasone (Dex). Twenty four cylindrical titanium implants (n = 6/group) were inserted bilaterally into the femoral epiphyses of New Zealand White, female, adult rabbits for 4 weeks. Implant stability was measured by resonance frequency analysis (RFA) the day of implantation and 4 weeks later, and correlated to histomorphometric parameters, bone implant contact (BIC) and bone growth around the implants (BS/TS 0.5 mm). The BIC values for the four groups were not significantly different. That said, histology indicated that the CaP coatings improved bone growth around the implants. The incorporation of bioactive molecules (cAMP and Dex) into the CaP coatings did not improve bone growth compared to the BE group. Implant stability quotients (ISQ) increased in each group after 4 weeks of healing but were not significantly different between the groups. A good correlation was observed between ISQ and BS/TS 0.5 mm indicating that RFA is a non-invasive method that can be used to assess the osseointegration of implants. In conclusion, the CaP coating enhanced bone formation around the implants, which was correlated to stability measured by resonance frequency analysis. Furthers studies need to be conducted in order to explore the benefits of incorporating bioactive molecules into the coatings for peri-implant bone healing.     

Effect of FGF-2 and melatonin on implant bone healing: a histomorphometric study

Melatonin influences the release of growth hormone and cortisol in humans, and it was recently reported that it promoted bone formation. On the other hand, fibroblast growth factor-2 (FGF-2) was reported to facilitate the proliferation of osteoblasts. In the present study, we examined the effect of recombinant human FGF-2 and melatonin on the promotion of osteogenesis around titanium implants. Twenty-four 10-week-old female rats of the Wistar strain received titanium implants in both tibiae. In the experimental groups, 100 mg/kg body weight of melatonin was administered by intraperitoneal injection for 4 weeks after implantation and 10 microg of FGF-2 was locally injected around the implant sites 5 days after implantation. The control groups were administered saline only. In the control group, few newly formed bone could be seen around the implants. It was observed to be in direct contact with the implant surface, but otherwise unmineralized connective tissue was occasionally interposed. In the experimental group, newly formed bone was observed around the titanium implant. In addition, in contrast to the control group, abundant bone trabeculae were seen in the medullary canal region. Bone trabeculae were directly connected to existing cortical bone. These results strongly suggested that melatonin and FGF-2 have the potential to promote osseointegration.     

Histomorphometric and histologic evaluation of titanium–zirconium (<Emphasis Type="Italic">a</Emphasis>TiZr) implants with anodized surfaces

The choice of implant surface has a significant influence on osseointegration. Modification of TiZr surface by anodization is reported to have the potential to modulate the osteoblast cell behaviour favouring more rapid bone formation. The aim of this study is to investigate the effect of anodizing the surface of TiZr discs with respect to osseointegration after four weeks implantation in sheep femurs. Titanium (Ti) and TiZr discs were anodized in an electrolyte containing dl-α-glycerophosphate and calcium acetate at 300 V. The surface characteristics were analyzed by scanning electron microscopy, electron dispersive spectroscopy, atomic force microscopy and goniometry. Forty implant discs with thickness of 1.5 and 10 mm diameter (10 of each-titanium, titanium–zirconium, anodized titanium and anodized titanium–zirconium) were placed in the femoral condyles of 10 sheep. Histomorphometric and histologic analysis were performed 4 weeks after implantation. The anodized implants displayed hydrophilic, porous, nano-to-micrometer scale roughened surfaces. Energy dispersive spectroscopy analysis revealed calcium and phosphorous incorporation into the surface of both titanium and titanium–zirconium after anodization. Histologically there was new bone apposition on all implanted discs, slightly more pronounced on anodised discs. The percentage bone-to-implant contact measurements of anodized implants were higher than machined/unmodified implants but there was no significant difference between the two groups with anodized surfaces (P > 0.05, n = 10). The present histomorphometric and histological findings confirm that surface modification of titanium–zirconium by anodization is similar to anodised titanium enhances early osseointegration compared to machined implant surfaces.     

Osteochondral tissue engineering using a PLGA–collagen hybrid mesh

Canine bone marrow stromal cells were cultured in a PLGA–collagen hybrid mesh in osteogenic medium in vitro and laminated to construct an osteo layer. Canine articular chondrocytes were cultured in the hybrid mesh in DMEM containing 10% FBS and laminated to construct a chondral layer. The osteo and chondral layers were sutured together and implanted subcutaneously in nude mice. The original round disc shape of the osteochondral constructs was preserved during the implantation. The osteo and chondral layers appeared red and glistening white, respectively. Histological examination of the implant specimens indicated that stromal cells and chondrocytes were evenly distributed throughout the scaffold. The laminated meshes were bound together and the two layers had a distinct interface between them. The cells showed a round morphology in the chondral layer and a spindle morphology in the osteo layer. In the chondral layer, spherical chondrocytes were surrounded by an abundant cartilaginous extracellular matrix. The round morphology and positive stain by safranin-O and toluidine blue, together with the expression of genes encoding type II collagen and aggrecan suggested the formation of neocartilage in the chondral layer. Expressions of genes encoding type I collagen and osteocalcin were detected in the osteochondral implant. These results indicate the formation of osteochondral-like tissue, and the hybrid mesh and lamination method may be useful for osteochondral tissue engineering.     

A histological and histomorphometrical evaluation of screw-type calciumphosphate (Ca-P) coated implants; an in vivo experiment in maxillary cancellous bone of goats

The bone response to different calcium phosphate (Ca-P) coated implants was evaluated in a goat animal model. Two types of plasma spray coatings were applied to a commercially pure titanium (cpTi) tapered, conical screw-design implant (BioComp^®); hydroxyapatite (HA-PS) and a dual coating, consisting of FA and HA (FA/HA-PS). In addition an amorphous RF magnetron sputter coating (Ca-P-a) and uncoated implants were investigated. Forty-eight implants were inserted in the maxilla of 12 adult female goats. After implantation periods of 3 and 6 months, the bone implant interface was evaluated histologically and histomorphometrically. After both implantation periods all plasma spray coated implants were maintained. On the other hand three Ca-P-a and two cpTi implants were lost. Histological examination revealed a better bone response to both plasma spray coated implants. Histomorphometrical evaluation confirmed this finding. At 3 and 6 months significantly higher percentages of bone contact (p<0.001, ANOVA) were measured for both plasma spray coated implants than for the cpTi and Ca-P-a implants, while no significant difference (p<0.05) existed between both implantation periods. Degradation of both plasma spray coatings was observed. Supported by the results, it is concluded that, although Ca-P coatings can improve the performance of dental implants, the presence of a Ca-P coating is not the only important factor for bone healing around implants placed in low density trabecular bone.     

Titanium levels in rats implanted with Ti6Al4V treated samples in the absence of wear

The effect of implantation time and implant nitriding on titanium ion concentration in several tissues of rats carrying Ti6Al4V implants was studied by means of inductively coupled plasma-mass spectroscopy (ICP-MS). Histological studies were also performed in order to check for tissue degeneration due to the Ti6Al4V implantation. The animals were divided into four groups: one received Ti6Al4V implants, the second received nitrided Ti6Al4V implants, the third group received nitrided and descaled Ti6Al4V implants and the last one was the control group. Half the animals of the implanted groups received the Ti6Al4V implant for 30 days, while the other half received the implant for 120 days. Spleen, muscle, kidney, lung, brain and bone samples were retrieved from these rats as well as the control group. Ion concentration measures did not show significant differences between control and implanted rats for the studied period of time, although histological studies showed minor differences, especially on liver tissue samples.     

A novel osteochondral scaffold of ceramic–gelatin assembly for articular cartilage repair

A novel ceramic–gelatin assembly (CGA) has been designed as an osteochondral scaffold for articular cartilage repair. The CGA scaffold consists of four layers, that is, a porous ceramic layer as osseous component and also as anchor, a dense ceramic layer to prevent blood vessel penetration and also to stand shear stress, a porous ceramic layer for fixation of bone to cartilage, i.e. for joining the ceramic part to the porous gelatin layer, the latter being used as cartilaginous component. The joining was done by the infiltration of gelatin solution into the porous ceramic layer, gelling and crosslinking. This CGA scaffold can offer solutions to the so-far not satisfactorily resolved issues of the osteochondral scaffold, i.e. anchoring, blood vessel penetration, shear stress distribution during articular joint motion, and enough strength to join the cartilaginous component to the osseous component to prevent delamination. This novel scaffold was tested by in vitro cell culture with Wistar rat's joint chondrocytes. DNA assay, GAGs assay, RT-PCR, and histological evaluations with hematoxylin–eosin and Safranin-O staining were carried out to show that cartilage tissue can be developed in four weeks.     

Effect of potassium and magnesium doping on mechanical properties and in vitro degradation behavior of calcium polyphosphate

Porous crystalline calcium polyphosphate (CPP) is under investigation as a candidate bone substitute/augmentation material including incorporation in implants intended for repair of osteochondral defects. Previous studies of biphasic implants (i.e., cartilage-CPP constructs) for osteochondral defect repair have shown that porous CPP has the required features for this application including the bone substitute portion of the biphasic implant, but the porous CPP degradation rate is lower than preferred. This study investigated the effect of doping with MgCO₃, MgCl₂, K₂CO₃, or KCl at a molar ratio of M/Ca = 0.02 on properties and in vitro degradation behavior of CPP. Doping with magnesium or potassium resulted in changes in the crystallization and melting temperatures, which required adjustment of the sintering conditions for forming samples of the desired porosity level. This, in turn, resulted in higher compressive and diametral compressive (i.e., tensile) strengths of the porous-doped CPP samples compared with undoped CPP prepared to a given porosity level. For samples prepared in this study, the chemical degradation rate of porous Mg-doped CPP samples was the fastest while K doping resulted in a lower degradation rate than undoped CPP.     

Fabrication of an rhBMP-2 loaded porous β-TCP microsphere-hyaluronic acid-based powder gel composite and evaluation of implant osseointegration

Methods to improve osseointegration that include implantation of rhBMP-2 with various kinds of carriers are currently of considerable interest. The present study was conducted to evaluate if the rhBMP-2 loaded β-TCP microsphere-hyaluronic acid-based powder-like hydrogel composite (powder gel) can act as an effective rhBMP-2 carrier for implantation in host bone with a bone defect or poor bone quality. The release pattern for rhBMP-2 was then evaluated against an rhBMP-2-loaded collagen sponge as a control group. Dental implants were also inserted into the tibias of three groups of rabbits: an rhBMP-2 (200 µg) loaded powder gel composite implanted group, an implant only group, and a powder gel implanted group. Micro-CT and histology of the implanted areas were carried out four weeks later. The rhBMP-2 powder gel released less rhBMP-2 than the collagen sponge, but it continued a slow release for more than 7 days. The rhBMP-2 powder gel composite improved osseointegration of the dental implant by increasing the amount of new bone formation in the implant pitch and it improved the bone quality and bone quantity of new bone. The histology results indicated that the rhBMP-2 powder gel composite improved the osseointegration in the cortical bone as well as the marrow space along the fixture. The bone-to-implant contact ratio of the rhBMP-2 (200 µg) loaded powder gel composite implanted group was significantly higher than those of the implant only group and the powder gel implanted group. The powder gel appeared to be a good carrier and could release rhBMP-2 slowly to promote the formation of new bone following implantation in a bone defect, thereby improving implant osseointegration.