Platelet-rich plasma (PRP), containing various growth factors, may speed up wound and bone healing. Using osteoconductive alloplastic materials in reconstructive surgery, the amount of autogenous bone needed can be reduced. The purpose of this experiment was to study the effect of PRP on a mixture of autogenous bone and deproteinized bovine bone mineral (Bio-Oss ® ) particles in goats. Four, round, critical size defects were made in the foreheads of 20 goats. In all goats the defects were filled with a mixture of autogenous particulate cancellous bone and (Bio-Oss ® ) particles, in which 1 ml of PRP was added in two of the four defects. The goats were allocated to four subgroups each containing five goats, which were killed after 1, 2, 6 and 12 weeks. The results of the histological and histomorphometric examination showed that early and late bone healing were not enhanced when PRP was used.
There is a continuing search for bone substitutes to avoid or minimize the need for autogenous bone grafts. A common problem is that an appropriate amount of autologous bone needed for complete filling of a bone defect cannot be harvested. To solve this, the harvested autologous bone can be mixed with an alloplastic or synthetic bone graft to increase the volume of the graft. The graft has to support bone healing. One of the frequently used allogenic grafts consists of deproteinized bovine bone mineral (Bio-Oss ® ). This material contains pores of different sizes: macro pores (300–1500 μm), micro pores (Haversian and vascular marrow canals) and intracrystalline spaces (3–26 nm), which results in a high overall porosity of 70–75% . As a consequence of this highly porous structure, Bio-Oss ® can easily be invaded by blood vessels resulting in subsequent migration of osteoblasts . Bio-Oss ® is considered to be an osteoconductive and biocompatible material evoking no significant inflammatory reaction . J ensen et al. found, in a study on rabbits, that Bio-Oss ® became completely incorporated in newly formed bone. In comparison to three other bone-substitute materials (Endobon, Pro Osteon 500 and Interpore 500 HA/CC), Bio-Oss ® was found to be integrated to a higher degree in the surrounding bone. These findings were confirmed by other studies in beagles . B erglundh & L indhe also concluded that Bio-Oss ® becomes integrated and subsequently, but slowly, replaced by newly formed bone. The results of other studies are in accordance with these findings . These characteristics of Bio-Oss ® are in accordance with the criteria of an osteoconductive material .
For several years platelet-rich plasma (PRP) has been thought to promote bone healing but there are contradictory reports about its clinical efficacy. Several studies in humans show that PRP has a beneficial effect on bone healing . Studies in animals reveal conflicting results, but most are negative . The possible reasons for these results are discussed in a previous study . In summary, the surgical site may influence the results, for example, in contrast with periosteum and dura , respiratory epithelium (sinus floor elevation) can act as a resorptive endosteum . The type of bone used may also play a role, since particulate cortical bone did not to heal in a predictable fashion . In some studies, diffusion of PRP and/or growth factors from the experimental to the control site cannot be excluded. The main reason for the conflicting results may be that the average concentration of blood platelets used in the studies did not take into account the great variation in the baseline platelet values of the animals, and thus there is variation in the concentration of the platelets in the PRP used. It is important that this issue about PRP efficacy is solved, because the use of PRP is associated with increased treatment cost and inconvenience for the patient, such as the necessity to draw blood.
When the current study was designed, the results of the previous study were not known. In an attempt to reduce the amount of autogenous bone in reconstructive surgery, the authors decided to carry out a study on the use of a mixture of an osteoconductive alloplastic material and particulate autogenous cancellous bone with and without PRP. Based on previous positive reports , Bio-Oss ® was thought to be a suitable material.
The hypothesis of the current study was that the healing of a bone defect filled with a mixture of particulate autogenous bone and Bio-Oss ® would be more rapid when PRP was added and that Bio-Oss ® would be integrated in the newly formed bone. A goat model was selected because previous studies at the same institution had shown that a round, 14 mm diameter defect in the forehead of a goat is the critical size, at 12 weeks postoperatively . At 24 weeks, variable amounts of bone deposition, often extending a considerable distance from the margins, were observed. The dimension of the frontal bone also allows the preparation of four separate round defects in the facial part of each skull, which makes it possible to create two experimental and two control sites for each animal. The goat model appears to be a suitable model for using particulate autogenous cancellous bone or a mixture of particulate autogenous cancellous bone and Bio-Oss ® (the current study) to study the effect of PRP on facial bone healing.
Materials and methods
In vivo study
This study was carried out on 20 skeletally mature female Saanen goats, weighing about 65 kg. Protocols according to the national guidelines for the care and use of laboratory animals were respected. Previous studies confirmed that the goat model used caused negligible discomfort for the animals so permission was granted by the Animal Ethical Committee of the University of Nijmegen for further studies using this model.
Based on the method described by C lemmons et al. and in consultation with the transfusion laboratory, the currently used PRP was prepared by the transfusion laboratory as described below. Platelets are highly (but reversibly) activated shortly after collection of whole blood and show spontaneous aggregation. These aggregates disappear after a resting period of 16–20 h at room temperature . These platelets have equal in vitro quality compared with platelets produced from whole blood that had been stored for 6 h . For logistic and uniform processing of whole blood, cooling the units to room temperature within 2 h and a subsequent storage period of about 16–20 h until separation into components is preferred . Two hundred and fifty cubic centimetre of autogenous blood was drawn 1 day before surgery. This blood was treated by centrifugation in various cycles at the transfusion laboratory. For logistical reasons it was decided to perform the first spin (2000 G, 10 min) 3 h after collection of the blood and to store the buffy coat and plasma separately at 22 °C. On the day of surgery, three other cycles took place: 250 G (5 min), 3000 G (5 min) and 3000 G (5 min). After this, plasma was removed until 10 cm 3 were left with the remaining platelets on the bottom of the bag. From this mixture, a sample was taken to count the amount of platelets. At the time the current experiment was performed goat platelets could not be automatically counted by the Coulter Onyx ® , in the authors’ institute, so this was carried out with a haematocytometer. The concentration of platelets varied from 800 to 1000 × 10 9 /l. Starting from a baseline value of 171 (±19.6) × 10 9 /l , a 4.2–6.6-fold increase of the baseline value was achieved. According to M arx , a sufficient cellular response to platelet concentration begins when a four to five-fold increase to baseline platelet numbers is achieved.
The surgical procedure has been described previously . In 20 goats, four full-thickness critical bone defects were made with a hollow cylindrical trephine drill (standard outer diameter 14 mm) in the frontal bone. Dental pins (Stabilok ® , Wimbledon Village, London, UK) were inserted in the bone adjacent to each defect to facilitate orientation, one at defect A and C, and two at defects B and D each. Defects A and C were subsequently filled with a composite graft consisting of 0.3 g of autogenous cancellous bone (iliac crest) and 0.2 g of Bio-Oss ® spongious granules of 1–2 mm (Geistlich-Pharma, Wolhusen, Switzerland). The other two defects (D and B) were filled with the same composite graft mixed with 1 ml PRP ( Fig. 1 ). The PRP suspension was activated at the time of surgery with 1 cm 3 10% calcium chloride solution and 300 i.u. of bovine thrombin (Fibriquick ® Organon Teknica, Durham, USA) to form a gel , while the erythrocytes were given back . After activation and just prior to forming a gel, the PRP was added ( ex vivo ) to the mixture of Bio-Oss ® and the autogenous bone particles. In this way, the PRP gel was distributed optimally in the mixture without blending, thus keeping the gel structure unimpaired.
The animals were divided in four groups: Group 1 ( N = 5 goats; materials were left in place for 1 week); Group 2 ( N = 5 goats; materials were left in place for 2 weeks); Group 3 ( N = 5 goats; materials were left in place for 6 weeks); and Group 4 ( N = 5 goats; materials were left in place for 12 weeks). This resulted in 10 specimens of each graft type per implantation time. At the end of the implantation times, all animals were killed with an overdose of pentobarbital.
After death, the frontal bone of each goat was resected and stripped of soft tissue leaving a thin layer of connective tissue over the experimental sites. These specimens were stored in a 4% phosphate-buffered formaldehyde solution (pH 7.4) for fixation. After fixation, the frontal bone was separated into four parts, each containing one experimental site. The parts were dehydrated in graded series of alcohol and embedded in methylmethacrylate. After polymerisation, each part was divided in such a way that the experimental site was split into two halves. One half was used to make three 10 μm transverse sections using a modified microtome technique . The sections were etched with hydrochloric ethanol and subsequently stained with methylene blue and basic fuchsin.
Light-microscopical analysis consisted of a description of the tissue response within the grafted area and was carried out by two investigators using an optical microscope (Leica BV, Rijswijk, The Netherlands).
Apart from a subjective description, each experimental site was quantitatively assessed using a grading scale for bone formation. Specimens were evaluated for: bone formation at the edges of the defect; and the amount of bony bridging of the defect ( Table 1 ). Two blinded reviewers evaluated at least three sections for each specimen. The reviewers reached a consensus on the score of each section and then assigned a final score to each specimen. The scores of the specimens in each material group were averaged to determine the overall score for the material group.
|Bone formation at the edges of the defects|
|Bridging of the defect|
In all groups, histomorphometry was also carried out on the sections to quantify the bone fill percentage of newly formed bone and grafted bone and the Bio-Oss ® fill percentage in a plane perpendicular to the grafted area, called ‘the surface of the defect’. For this purpose a Leica Qwin Image Processing and Analysis System (Cambridge, UK) was used. The sections were digitized at low magnification.
In each slide, the outline of the surface of the defect was marked ( Fig. 2 B ). The newly formed bone, the grafted bone and the Bio-Oss ® were marked in an interactive manner ( Fig. 2 C and D). The computer calculated the area of the surface of the defect, the bone surface (newly formed bone and grafted bone) and the surface of the Bio-Oss ® particles. From these data, the bone fill percentage and the Bio-Oss ® fill percentage in the surface of the defect were calculated by dividing the bone surface area and the surface of the Bio-Oss ® particles, respectively, by the area of the surface of the defect.
The results were statistically evaluated, the quantitative grading scale measurements using the Pearson χ 2 test and the results of the histomorphometry using Student’s t -test. Differences were considered statistically significant if P < 0.05. Calculations were performed in GraphPad Instat, Version 3.05 (GraphPad Software, San Diego, CA, USA). All charts show error bars depicting standard deviation.
Recovery and healing were uneventful in all goats with no signs of infection or wound dehiscence.
Light-microscopic examination of the sections revealed no differences between the PRP and the non-PRP group. At 1 week, there was a limited inflammatory reaction with sporadic giant cells. Fibrous tissue was rarely seen. No newly formed bone and no resorption of the grafted particles could be observed.
At 2 weeks, there was an increase of inflammatory cells although no further increase in the number of giant cells. There were moderate signs of resorption of the grafted particles, marked by irregularities and some osteoclast-like cells, in particular around the autogenous bone particles. Newly formed bone was hardly visible at the borders of the defect, Bio-Oss ® particles, and autogenous bone particles.
At 6 weeks, the number of inflammatory cells had decreased considerably. Extensive bone formation was seen at the surface of the autogenous bone particles. Bio-Oss ® particles show less newly formed bone compared with the autogenous bone particles ( Fig. 3 ). It was difficult to judge the amount of resorption of the autogenous particles because of the newly formed bone. Resorption of the Bio-Oss ® particles was very limited.
At 12 weeks, there were no signs of inflammatory reaction. More newly formed bone was observed in greater quantities at the surface of the autogenous bone particles compared with the Bio-Oss ® particles. Similar to the samples retrieved after 6 weeks, resorption of the Bio-Oss ® particles was very limited.
Empty lacunae, caused by the absence of osteocytes, were rarely observed at each time point. Fibrous tissue was hardly seen in the first week. After 2 weeks, the transplanted autogenous particles and Bio-Oss ® particles were surrounded by a fibrous tissue layer. At weeks 6 and 12, the amount of fibrous tissue remained the same.
Comparing bone formation at the edge of the defect and the amount of bone bridging the defect, no significant differences were seen between the sites with and without PRP for the different implantation times. In all cases P > 0.05 (Pearson χ 2 test) ( Table 2 ). After 1 week postoperatively, no bone formation was seen at the edges of the defect. At 2 weeks, in five sites bone formation was seen at one edge and in 14 sites at two edges. Bridging was seen at a maximum of 75% of the distance of the two edges in nine cases. At 6 weeks, 17 of 20 sites, and at 12 weeks, 18 of 20 sites showed bone formation at two edges. At 6 weeks, 75–100% bridging was seen in six of 20 cases, at 12 weeks this was the case in 12 of 20 cases.
|Bone formation||Bridging of the defect|
|Weeks||None||One edge||Two edge||Sites examined||0–25%||25–50%||75–100%||75–100%||Sites examined|