Platelet-Rich Plasma (PRP) is derived from autologous blood and is defined as a certain volume of plasma that has a platelet concentration several fold above the physiologic levels. In the present study we consistently used PRP with a high concentration of platelets (8–10 fold above the physiologic levels). The preparation of PRP, including the amount of blood collected, the centrifugal forces, and the final volume of the obtained PRP was optimized for each bioassay in order to obtain that concentration (see Table 1, Panel B for details). The concentration of platelets in whole blood and in the PRP was determined manually using a Petroff-Hausser counting chamber (improved Neubauer, cell-depth of 0.02 mm)(Hausser Scientific, USA). In all cases, PRP was prepared from autologous blood by a 2-step centrifugation process following previously published protocols 1218 with some modifications (Fig. 1). Briefly, aliquots of whole blood were collected in tubes containing acid-citrate-dextrose as an anti-coagulant (0.163 ml per 1 ml of blood) (Fig. 1, step 1). Immediately after being drawn, blood was centrifuged to separate red blood cells (RBCs) from platelets and plasma (Fig. 1, step 2). Then, the supernatant composed of platelets and plasma was collected and centrifuged again in order to pellet the platelets (Fig. 1, step 3). After this second centrifugation, the platelets were re-suspended in an appropriate volume of plasma (Fig. 1, step 4, 5) to achieve a platelet concentration equal to 8–10 fold above the physiologic levels.

The commercially available powder form of Calcium Sulfate (CS) is the product of partial dehydration of gypsum, which produces CS hemihydrate (CaSO₄·1/2 H₂O). When the hemihydrate is mixed with water in the correct proportions, a suspension is formed that is initially fluid and workable; then, the hemihydrate completely precipitates until crystals of dihydrated solid CS are formed. This is an exothermic setting reaction during which 1 g of hemihydrate CS combines with 186 μl of water 1920. Similarly, as described in previous studies 16, CS-Platelet was prepared by mixing hemihydrate CS (Capset, Lifecore Biomedical, Inc. USA for the animal bioassays, and Surgiplaster, ClassImplant, s.r.l. Italy for the human clinical cases) with Platelet-Rich Plasma (PRP) in place of water. To account for the viscosity of plasma and for the platelet's volume, the powder (CS) to liquid (PRP) ratio was adjusted to 1 g of CS to 240 μl of PRP. By mixing CS with PRP a malleable paste was obtained and applied at the surgical sites (Fig. 1, step 6). After 30 minutes CS-Platelet crystallized becoming a solid and homogeneous compound 16. According to each experiment or clinical case, the required amount of CS-Platelet was prepared by mixing CS with PRP in the ratio previously mentioned.

All animal bioassays presented hereafter were conducted in accordance with the Institutional Animal Care and Use Committee (IACUC) of the University at Buffalo.

A heterotopic bioassay 2122 was used to test the osteoinductive activity of CS-Platelet (Fig. 2). To this end, 120 μl of PRP or 120 μl of double distilled sterile water were mixed with 500 mg of CS to prepare samples of CS-Platelet or CS. Then, 4 samples of CS-Platelet and 4 samples of CS were implanted into the triceps brachii and biceps femoris of a ferret (Fig. 2A). To help with post-surgical localization, a tattoo (1% India ink in 25 ul PBS) was performed at the mesial aspect of each implantation. Management of post-operative pain included subcutaneous administration of buprenorphine (0.01 mg/kg body weight). At 4 weeks post-implantation, x-rays were obtained from each muscle (10 MA, 78 kVp, 1/15 of a second impulse, anode-film distance = 20 cm) (Fig. 2B). Then, histological specimens were retrieved, demineralized in 14% EDTA for 10 days, and processed for tissue sectioning (4 μm) and staining (Hematoxylin and Eosin) (Fig. 2C–L).

Using the 8 mm rat calvaria critical size defect (CSD) model of bone regeneration 23, the efficacy of CS-Platelet for in vivo bone regeneration was compared to the efficacy of other biomaterials (Fig. 3 and Fig. 4). The 8 mm diameter craniotomy defects were created with a trephine, avoiding perforation of the dura mater. Then, the created bone defects were filled with the chosen biomaterial. Briefly, 35 female adult Sprague-Dawley rats (275 g) were randomly divided into 7 groups of 5 rats each: 1) no filling biomaterial (Negative); 2) 100 mg of CS with 24 μl of double distilled sterile water (CS); 3) 24 μl of PRP activated by thrombin and CaCl₂ in accordance with previously published protocols 1224 (PRP); 4) 100 mg of CS and 24 μl of PRP (CS-Platelet); 5) 6.8 mg of collagen 25 (Coll); 6) 5 μg of human recombinant BMP2 (Cell Sciences, Inc. USA) and 6.8 mg of collagen 2526 (rhBMP2); and 7) no surgery performed (External control). The rhBMP2 group (5 μg of rhBMP2 per cranial defect) represents a gold standard treatment for the regeneration of the 8 mm rat calvaria defect model 2126 and in this study is used as comparative positive control. 100 mg of CS powder was determined to be the amount of CS required to completely fill the 8 mm cranial defect. After implantation of the biomaterials, the soft tissues were closed with skin staples. Management of post-operative pain included subcutaneous administration of buprenorphine (0.01 mg/kg body weight). Rats were euthanized by CO₂ asphyxiation 8 weeks after surgery. The craniotomy sites with 10 mm contiguous bone were recovered from the skull and placed in a fixative solution (20 ml of RNA-Later, Ambion Inc, USA). All explanted calvaria were scanned using a Scanco Medical AG (Bassersdorf, Switzerland) μCT 40 system. Specimens were scanned using an 18 μm isotropic voxel size (FOV: 36.8; 2048 × 2048 image matrix) and 2000 projections/360 degrees 26. Mineralized bone tissue was segmented from non-mineralized tissue by applying a gauss filter (sigma:1.6; support:3.0) and using a threshold of 21% of maximum gray scale value. The new bone formation was analyzed by extracting the region of new bone in the critical size defect using semi-automated contouring. The volume (mm³), the surface (mm²), and the density (mg/mm³) of the newly regenerated bone were measured by a blinded operator. Statistical power was calculated on previous bone regeneration calvaria defect studies 26 and differences among groups were statistically analyzed by ANOVA followed by the Fisher LSD multiple comparison test (significance level of 0.01). After tomography quantification, the specimens were processed for non-demineralized tissue sectioning (7 μm) and staining (Goldner's Trichrome).

To test its clinical features and the histological outcomes, CS-Platelet was compared to demineralized freeze-dried bone allograft (DFDBA) (American Red Cross, Tissue Services, USA) (Fig. 5). One adult male non-human primate (Macacus rhesus) was used. Prior to the surgical extraction of the maxillary and mandibular first premolars, first molars, and third molars the animal was properly anesthetized. 30 ml of blood were collected for the preparation of 1.5 ml of PRP (see Table 1 for details). Then, using a split-mouth design, the right extraction sockets were each filled with CS-Platelet (1 g of Calcium Sulfate and 240 μl of PRP) and the left extraction sockets with human DFDBA (0.5 cc, American Red Cross – Tissue Services, Arlington VA). Management of post-operative pain included intramuscular administration of buprenorphine (0.01 mg/kg body weight) every 6–12 hrs for 4 days. Eight weeks after surgery, the animal was sacrificed and block sections of the grafted sites were collected, demineralized in 14% EDTA for 10 days, and processed for tissue sectioning (4 μm) and staining (Hematoxylin and Eosin).

To test its clinical potentials, CS-Platelet was used in five human clinical dental cases (Fig. 6). In all patients treated (age group: 35–55 year old), the platelet count was within the normal limits and the medical and dental history revealed no significant contraindications to bone regenerative therapy. In all cases presented regeneration of bone was needed in order to either augment the amount of the preexisting bone for subsequent implantation of titanium implants (Augmentation cases: cases 1–3) or to augment the amount of bone surrounding a preexisting titanium implant to preserve the implant (Preservation cases: cases 4 and 5). Clinical cases consisted of: 1) Augmentation of extraction socket upon extraction of tooth # 8 (Fig. 6A–D), 2) Ridge augmentation therapy in area of teeth # 5–6 (Fig. 6E–H), 3) Sinus augmentation therapy (induced bone formation within the right maxillary sinus) prior insertion of a titanium implant in area of teeth # 3–4 (Fig. 6I–L), 4) Regeneration of bone around the exposed threads of a titanium implant positioned are of tooth # 9 (Fig. 6M–P), and 5) Peri-implantitis around titanium implant in position of tooth # 30 (Fig. 6Q–T). Prior to implantation of CS-Platelet the surgical recipient sites were freed of soft tissue tags and contaminated bone (tissue debridement). In the case of the sinus augmentation therapy, the schneiderian membrane was carefully elevated and CS-Platelet was implanted between the membrane and the floor of the sinus. In all other cases, after tissue debridement, CS-Platelet was simply applied in order to completely fill the bone defect. Then, the surgical wounds were sutured in order to achieve healing by first intention. Management of post-operative pain included administration of ibuprofen (400 mg t.i.d. p.r.n.). Clinical and radiographical re-evaluations were performed at 6 months. The human clinical cases were conducted in private dental practices on volunteers who gave written informed consent. Each subject was also informed that data concerning the case would be submitted for publication.

Four weeks after implantation of CS or CS-Platelet in the triceps brachii and biceps femoris of a ferret (Fig. 2A), the x-ray evaluations showed the presence of radio-opaque formations only at the sites of the CS-Platelet implantations (Fig. 2B). Upon histological evaluation, bone formation in the form of an ossicle was noted (Fig. 2C and 2E–H). Highly vascularized trabecular bone associated with highly vascularized bone marrow rich of reticular cells was observed (Fig. 2E and 2G, red arrows). In addition, at the periphery of the bone nodule endochondral bone formation was noticeable (Fig. 2F, blue arrow), testifying a peripheral functional remodeling of mesenchymal tissue. No encapsulation of the bone nodule by chronic inflammation was observed. The presence of numerous active osteoblasts forming a rim of cells in the process of laying down osteoid (Fig. 2G and 2H, purple arrows), numerous osteocytes incorporated in the newly formed bone (Fig. 2G and 2H, black arrows), and osteoclasts forming Howship's lacunae (Fig. 2H, green arrow) confirm the high vitality of the generated bone nodule. No bone formation was observed at the site of the implantations of CS alone (Fig. 2D). CS was not completely resorbed and residual small crystals of CS were seen within the tissue. (Fig. 2L, grey arrow). The amount of residual CS is modest and the crystals are too dispersed to be shown by our x-ray evaluation (Fig. 2B). A foreign body reaction with both chronic (multinucleated-giant cells) and acute inflammatory response (polymorphonucleated) were also noticed (Fig. 2K,L, yellow arrows). Importantly, in both the CS-Platelet and CS cases, no signs of malignancy were observed.

Eight weeks after implantation in the 8 mm rat critical size defects, only CS-Platelet and rhBMP2 showed a regeneration of bone extended at the center of the defects (Fig. 3). In both these groups, areas of vascularized bone marrow are evident, and numerous osteocytes are present within the newly formed bone, testifying the vitality of the regenerated tissue. Osteoblasts present at the osteoid interface (osteoid tissue is stained in red with Goldner Trichrome) are in the process of laying down osteoid matrix. In all other groups, a limited regeneration of bone was seen only at the periphery of the defects (Fig. 3). As expected, in these cases bone regenerated only at the margins with fibrotic tissue always present at the center of the defects. In both the CS-Platelet and CS groups the implanted biomaterial is not completely resorbed (Fig. 3). Upon quantification and statistical analysis of the volume of bone regenerated within the 8 mm defect (Fig. 4A), only CS-Platelet, rhBMP2, and the External control showed a statistically significant difference from the negative control (p < 0.001). Evidently, CS alone (CS), activated PRP alone (PRP), and Collagen alone (Coll) are able to regenerate the same amount of bone that is regenerated when the defect is left unfilled (Negative)(non statistical significant difference). More specifically, CS-Platelet regenerated a volume of bone approximately 2/3 the volume regenerated by rhBMP2 (statistically significant difference, p < 0.001) and approximately 1/2 the volume present in an area of 8 mm of the normal calvaria (External control) (statistically significant difference, p < 0.001). Upon quantification and statistical analysis of the surface area of the bone regenerated within the 8 mm defect (Fig. 4B), only CS-Platelet, rhBMP2, and the External control showed a statistically significant difference from the negative control (p < 0.001). However, the surface area of the bone regenerated by CS-Platelet was not statistically different from the one regenerated by rhBMP2 (p = 0.26) and the one normally present in an area of 8 mm of the normal calvaria (External control) (p = 0.68). Again, CS alone (CS), activated PRP alone (PRP), and Collagen alone (Coll) are able to regenerate the same surface area of bone that is regenerated when the defect is left unfilled (Negative)(non statistically significant difference). Upon quantification and statistical analysis of the density of the bone regenerated within the 8 mm defect (Fig. 4C) all treatment groups showed a density of the regenerated bone statistically different from the density of the bone normally present in an area of 8 mm of the normal calvaria (External control)(p < 0.001).

Eight weeks after implantation of CS-Platelet or demineralized freeze-dried bone allograft (DFDBA) in the tooth-extraction sockets, healing was shown to be uneventful and no bone loss or recession was reported at the adjacent teeth. In both cases, histologically, the overlying mucosa was fully formed with stratified squamous epithelium (Fig. 5A and 5E, blue arrows). Moreover, all grafted sites, whether implanted with CS-Platelet or with DFDBA, showed regeneration of trabecular bone (Fig. 5B–D and 5F–H, red arrows) associated with fibrofatty vascularized bone marrow (Fig. 5C–D and 5G–H, green arrows). Importantly, the grafted sites with DFDBA showed some residual biomaterial (Fig. 5A, black arrows), whereas no residual biomaterial was found in the sites grafted with CS-Platelet. In both cases, no signs of malignancy and no signs of chronic or acute inflammation were evident.

Six months after implantation of CS-Platelet, the patients were seen for re-evaluation of the bone regenerative therapy (Fig. 6). In all augmentation cases (Cases 1–3), the surgical re-opening showed complete bone repair within the treated area so that placement of titanium implant was achievable and successful (Fig. 6D,H and 6L respectively). The radiographical exam, when present (Fig. 6L), also confirmed the clinical evaluation, showing an area of radio-opaque bone repair within the bone defect. The Exposed threads therapy and the Peri-Implantitis (Preservation cases: cases 4 and 5) showed a successful bone repair with the achievement of healthy clinical status that allowed for the preservation of the previously implanted titanium implants (Fig. 6P and 6T respectively).