Effects of platelet-rich fibrin produced by three centrifugation protocols on bone neoformation in defects created in rat calvaria

Abstract This study evaluated the potential of Leukocyte-platelet-rich fibrin (L-PRF; fixed angle centrifugation protocol), Advanced-platelet-rich fibrin (A-PRF; low-speed fixed angle centrifugation protocol), and Horizontal-platelet-rich fibrin (H-PRF; horizontal centrifugation protocol) in bone neoformation in critical size defects (CSDs) in rat calvaria. Thirty-two rats were divided into groups: Control (C), L-PRF, A-PRF, and H-PRF. 5 mm diameter CSDs were created in the animals’ calvaria. Defects from group Control (C) were filled with blood clots, while defects from groups L-PRF, A-PRF, and H-PRF were filled with respective platelet-rich fibrin (PRF) membranes. L-PRF, A-PRF, and H-PRF were prepared from animal blood collection and specific centrifugation protocols. At 14 and 30 days, calcein (CA) and alizarin (AL) injections were performed, respectively. Animals were euthanized at 35 days. Microtomographic, laser confocal microscopy, and histomorphometric analyzes were performed. Data were statistically analyzed (ANOVA, Tukey, p < .05). L-PRF, A-PRF, and H-PRF groups showed higher values of bone volume (BV), newly formed bone area (NFBA), and precipitation of CA and AL than the C group (p < .05). The H-PRF group showed higher values of BV, number of trabeculae (Tb. N), NFBA, and higher precipitation of AL than the A-PRF and L-PRF groups (p < .05). Therefore, it can be concluded that: i) L-PRF, A-PRF, and H-PRF potentiate bone neoformation in CSDs in rat calvaria; ii) H-PRF demonstrated more biological potential for bone healing. Plain Language Summary After tooth loss, the alveolar bone (which supports the teeth) undergoes a natural process called bone remodeling, which can lead to significant decreases in bone height and thickness over time. Faced with the need to replace missing teeth, especially when it comes to dental implants, the lack of supporting tissues can compromise their correct positioning, leading to negative impacts on the success and longevity of the treatment. Therefore, over the years, several materials and procedures have been proposed to preserve and regenerate oral tissues. Leukocyte-platelet-rich fibrin (L-PRF) consists of a membrane obtained by centrifuging the patient’s blood in a fixed-angle centrifuge, allowing cells to be available to stimulate tissue regeneration directly at the place of action. Several reports demonstrate high potential in stimulating the formation of new tissues using L-PRF. In recent years, new protocols have been proposed to increase cell concentration and improve the regenerative potential of these membranes, changing the speed and time of centrifugation and introducing horizontal centrifugation. However, there still needs to be concrete evidence of the superiority of the new protocols in relation to the original protocol. In this study, we evaluated the healing of defects created in rat calvaria using platelet aggregates obtained through different centrifugation protocols. Within the limits of this study, it can be concluded that platelet aggregates improve bone healing, and horizontal centrifugation promotes more satisfactory results compared to fixed-angle protocols.


Introduction
One of the biggest challenges of regenerative therapies in Dentistry, especially after the advent of titanium implants, is to replace lost tissues properly or even prevent loss before it occurs so that the introduced treatment results in success and longevity, leading to the search for the development of materials that promote regeneration effectively in defect areas. 1,2n this sense, the use of platelet concentrates made from the patient's blood, like the Leukocyte-platelet-rich fibrin (L-PRF), has been proposed in recent years, obtaining an easily applicable three-dimensional membrane with promising results in promoting tissue regeneration.[8][9][10][11][12][13][14] Initially proposed for the potentiation of bone neoformation after the installation of dental implants, the use of L-PRF currently extends to periodontal therapies and bone reconstructions, such as grafting procedures and post-extraction socket preservation, 15,16 tissue repair in furcation defects, 17,18 root coverage in gingival recessions, [19][20][21] and procedures for maxillary sinus lift. 22,23n recent years, new protocols for the production of platelet concentrates have been proposed, giving rise to the concepts of centrifugation at reduced speed and time, including Advancedplatelet-rich fibrin (A-PRF) and platelet-rich fibrin (PRF) through horizontal centrifugation (H-PRF) to promote greater retention of platelets and leukocytes and consequently greater release of growth factors, enhancing regenerative and healing effects.2][33][34][35][36][37] Thus, the present study aimed to evaluate bone neoformation in critical size defects (CSDs) treated with PRF membranes produced by fixedangle and horizontal centrifugation protocols.

Animals
Thirty-two rats (Rattus Norvergicus, Albinus, Wistar), aged 14 weeks, with body mass between 350 g and 450 g, were obtained after submission and approval by the Committee on Ethics in the Use of Animals (CEUA) of the School of Dentistry of Ribeirão Preto (FORP) of Universidade de São Paulo (USP), under number 2019.1.930.58.0.The animals were kept in an environment with a 12-hour light cycle per day and a temperature between 22 and 24°C and consumed selected solid chow and water ad libitum.
The animals were randomly divided into four experimental groups (n = 8): Control (C), L-PRF, A-PRF, and H-PRF (Figure 1).The sample size ensured a power of 80% in the statistical analysis of the data obtained in the study, considering the means and standard deviations of the Area of Neoformed Bone (ANB) in the experimental groups of previous studies, 38- 40 adjusting the α value at 0.05.

Blood collection and PRF processing
Three milliliters of blood was obtained from each animal via cardiac puncture for the preparation of platelet-rich fibrin (PRF).Blood was collected in a 5 mL disposable syringe (Descarpack, São Paulo, Brazil) and stored in 5 mL silica-coated plastic tubes (BD Vacutainer®, Becton, Dickinson and Company, Franklin Lakes, NJ, USA) without the presence of anticoagulant substances, and immediately processed.
The cardiac puncture technique was employed to get a sufficient blood sample in a timely manner for the production of PRF membranes, being possible to obtain 3 mL of blood from each animal, according to the average body mass of the animals employed in this study, without causing suffering or extrapolate physiological limits. 41,42RF collection followed three different protocols.In the L-PRF group, blood was centrifuged at a speed of 2700 revolutions per minute (rpm) for 12 minutes (~700 g maximum Relative Centrifugal Force [RCFmax], corresponding to the maximum force applied to the tubes and calculated at the bottom of the centrifugation 43 in an Intra-Spin™ centrifuge (33° of rotor angle, 55 mm of rotor radius at the height of the clot, 86 mm of maximum radius; Intra-Lock® International, Inc, Boca Raton, FL, USA).In the A-PRF group, blood was centrifuged at 1500 rpm for 14 minutes (~200 g RCFmax, 28 also in an Intra-Spin TM centrifuge.Finally, a horizontal centrifuge (Eppendorf 5702 centrifuge, Hamburg, Germany) was used in the H-PRF group.Blood was centrifuged at an RCFmax of 700 g for 8 minutes (~2000 rpm), according to a protocol previously established in the literature. 14fter processing, the intermediate portion, representing the PRF clot, was collected and processed to fill the defects in the calvaria of animals in the L-PRF, A-PRF, and H-PRF groups.

Validation of L-PRF, A-PRF, and H-PRF protocols
To quantify and determine the location of cells along the tube after centrifugation and production of PRF, sequential pipetting analysis was performed using an adaptation of the work by Miron et al. 9 Six rats (two for each PRF protocol) were subjected to the collection of 4 mL of blood in tubes containing EDTA, so that, after centrifugation, the layers were separated.Still, no clot was formed, allowing the layers to be pipetted to laboratory analyses.After centrifugation of the L-PRF, A-PRF, and H-PRF protocols, 350-µL layers were pipetted sequentially, from the most superficial layer to the bottom of the tube, totaling ten samples per tube.These samples were stored in Eppendorf flasks and sent for quantification of platelets, leukocytes, and red blood cells to verify the location of these cells along the tube.

Creation of CSD
After blood collection for PRF preparation, a 5-mm diameter CSD was created on the calvaria of each animal using a trephine bur (5-mm Trephine Burr, Neodent®, Curitiba, PR, Brazil) at low rotation under constant irrigation with sterile saline solution.Defects from group C were filled with blood clots, naturally formed after the creation of CSDs.Defects from groups L-PRF, A-PRF, and H-PRF were filled with respective PRF membranes.The PRF membranes were divided approximately in half with the aid of surgical scissors.The portion closest to the red cells was cut into small pieces and used to fill the defect.The other half of the membrane was used to cover the defect in order to keep the material inside.

Euthanasia
The animals were euthanized 35 days after the making of the CSDs, using a CO2 chamber with controlled flow after anesthesia with the association of 10% Ketamine (80 mg/Kg) and 2% Xylazine (10 mg/Kg).Samples of the calvaria containing the area of the original surgical defect and the surrounding tissues were removed en bloc with a diamond disc and fixed in a 10% neutral buffered formalin solution for 24 hours.Then, they were washed in running water for another 24 hours and identified and stored in flasks containing 70° alcohol.

Analysis by micro-computed tomography (micro-CT)
Undecalcified samples of the calvaria were scanned by a SkyScan 1172 cone-bean micro-CT (SkyScan N.V., Kontich, Belgium) to generate three-dimensional (3D) images.Using DataViewer software v. 1.4.3 (SkyScan N.V., Kontich, Belgium), the 3D image was rotated to a standard position for analysis.Then a 5-mm diameter region of interest (ROI) and a volume of interest (VOI) of 0.5 × 5x5 mm was determined.To evaluate the trabecular bone tissue present in the calvaria in each VOI, a grayscale (0-255) was used, adopting the threshold between 80 (minimum) and 170 (maximum).
Using the CT-Analyzer® v.1.13.5.1+ software (Bruker, Kontich, Belgium), the following structural parameters were evaluated in each VOI by a calibrated examiner (L.M.S.P): i) BVthe percentage of VOI filled by bone tissue; ii) BP -the percentage of porosity present in the bone tissue determined in the VOI; iii) Tb.N -number (mmˉ1 of bone trabeculae present in the VOI; iv) Tb.Sp -total spaces (mm) between the bone trabeculae in the VOI; v) Conn.Dn -connectivity density between the bone trabeculae present in the VOI.

Laboratory processing
][46] The pieces obtained were fixed in a 10% formalin solution with pH 7 for 10 days, with the solution being changed every two days.After this period, the pieces were transferred to recipients containing increasing solutions of alcohol for dehydration (70%, 80%, 96%, and 100%), being subjected to agitation at each change.The samples were placed in polyethylene containers containing pure resin (LR White Hard Grade, London, UK) and under agitation for 60 minutes.After this period, the pieces were stored and kept in a refrigerator at 4°C for at least 12 hours.Then, the pieces were kept in a vacuum for one hour, shaken for the same period, and again stored in a refrigerator at 4°C for 24 hours.This routine was repeated for 15 days, changing the resin every 48 hours.On the 15th day, after changing the resin, the pieces were placed in Teflon molds and taken to the oven at 60°C for 12 hours for the resin polymerization.The polymerized resin blocks containing the samples were then cut in half using an Exakt system, using a diamond saw, and continuous irrigation with water.The surfaces containing the samples were ground in the Exakt system using sandpaper with increasing grit (320, 800, 2500, 4000) until the part's surface was smooth.The sanded and polished blocks were then glued to an acrylic sheet using the Exakt system, and the exposed surface of the piece glued to the sheet was sanded and polished again until a thickness of approximately 70 µm was obtained.

Image analysis
The pieces were scanned with a Trinocular Motorized Microscope for Immunofluorescence and Bright Field (model AXIO IMAGER M2, Carl Zeiss®, Germany) with a 10× objective coupled to an AxioCam MRm Rev.3 FireWire camera (Carl Zeiss®).After image reconstruction using AxioVision Rel.4.8 (Carl Zeiss®, Germany), it was possible to visualize the calcein precipitation at 14 days (Figure 2a), evidenced by the green color, representing the deposition of calcium in the old bone, and the precipitation of alizarin at 30 days (Figure 2b), in red color, which demonstrates the deposition of calcium in the newly formed bone.The overlapping of the two layers of fluorochromes was performed, allowing the visualization of calcium deposition in both periods, which represents the conversion of old bone into newly formed bone (Figure 2c).The images were standardized in hue, saturation, and brightness.Image captures were performed in the center of the CSD and the regions close to the original borders of the defect.
The saved images were then transferred to the Image J program (Processing Software and Image Analysis, Ontario, Canada).For each analyzed sample, calcein was delimited in a central region of the CSD with the "freehand" tool, measuring the area in mm 2; the same procedure was performed with alizarin.[46]

Histomorphometric analysis
After performing Confocal Microscopy, the slides already processed were stained with Stevenel Blue and Alizarin Red for analysis with light microscopy.In each section, the histopathological characteristics of the newly formed bone tissue were analyzed.
Histometric analysis was performed by a blinded and calibrated examiner (D.S.F.S.) using a computer image assessment system and image acquisition and analysis software (LAS EZ version 4.1.0,Leica Mycrosystems®).One histological section of the central area of the surgical defect of each specimen was selected (in each experimental group, eight images were analyzed).Each histological section was photographed using a trinocular microscope for brightfield and fluorescence (DMLB, Leica Mycrosystems® Wetzlar GmbH, Heidelberg, Germany) with a 1.6× objective coupled to a DFC300FX camera (Leica Mycrosystems®).In each image, a delimitation of the analyzed area was performed, corresponding to the region of the calvaria bone where the defect was created, called Total Area (TA).Within the TA, the Area of Neoformed Bone (ANB) was selected and delimited.The TA value was considered 100% of the analyzed area, and the ANB value was calculated as a percentage of TA.
The histopathological analysis was performed by observing the same selected histological sections using the same microscope, with 10× and 20× objectives coupled to the same camera used in the histometric analysis.

Statistical analysis
Analyzes were performed using GraphPad Prism software (GraphPad Software, Inc, v. 5.01, San Diego, CA, USA).The animal was considered as the statistical unit.A significance level of 5% was chosen (p < .05).Data distribution was verified by the Shapiro-Wilk test.The significance of differences between groups for all variables was determined by analysis of variance (ANOVA), followed by Tukey's post-hoc test.

Validation of platelet concentrates protocols
The quantification and location of platelets and leukocytes throughout the tube after centrifugation and production of PRF by three protocols (L-PRF, A-PRF, and H-PRF) were investigated by means of sequential pipetting analysis and represented in Table I.
It was observed that in the L-PRF protocol, platelets and leukocytes were significantly concentrated in layer 5, in the region of the interface between the portion corresponding to the PRF and the red cell layer, corresponding to 62.6% and 40.2% of the total of these cells after centrifugation, respectively.It was not possible to observe the presence of leukocytes in the first 4 layers of the tube.A large number of leukocytes was also found in the red cell layer below the region corresponding to the PRF.
In turn, in the A-PRF protocol, there was a higher concentration of platelets and leukocytes in layers 4 and 5, totaling ~ 55% and almost 40% of the total of these cells after centrifugation, respectively.Compared to the L-PRF, platelets, and leukocytes were more homogeneous and distributed throughout the tube.However, leukocytes were also not observed in the first two layers.
In the H-PRF protocol, it was possible to observe a high concentration of platelets in the first 4 layers of the tube, evenly distributed in these layers, corresponding to 87.4% of the total of these cells in the tube after centrifugation.A more homogeneous distribution of leukocytes was also found between layers 2 and 6, representing 80.7% of the total leukocytes after centrifugation.

Confocal scanning laser microscopy
Images obtained by Confocal Microscopy allowed the observation of bands of calcein (injected 14 days after creating the CSDs, green color) and alizarin (injected 30 days after making the CSDs, red color), formed by the precipitation of calcium in the organic matrix (Figure 4).In group C, in the central region (Figure 4a), small markings of calcein and alizarin were observed, which were superimposed, with a thickness much smaller than that found in the region close to the margins of the original defect.In the area close to one of the original borders (Figure 4b), a thin layer of calcein and alizarin was observed, demonstrating the slow process of bone neoformation.In the L-PRF and A-PRF groups, greater precipitation of calcein and alizarin was observed in the central region (Figure 4c,e) and in the region close to one of the original borders (Figure 4d,f), manifested by the greater intensity in fluorescence.In the H-PRF group, the central region (Figure 4g) showed intense labeling of calcein and alizarin, evidencing a higher rate of bone neoformation and lower rates of resorption of old bone.In addition, the thickness observed in the central region of the defect was very close to that present in the region close to the margins of the original defect.In the region close to one of the original borders (Figure 4h), more evident bands of calcein and alizarin were also observed compared to group C.
Higher statistically significant values of calcein were found in the L-PRF, A-PRF, and H-PRF groups compared to the C group (Figure 5b).No statistically significant differences were found between the L-PRF, A-PRF, and H-PRF groups regarding calcein precipitation.The H-PRF group showed the highest alizarin labeling, with statistically significant differences in relation to the C, L-PRF, and A-PRF groups (Figure 5c).The L-PRF and A-PRF groups had statistically significant higher values of alizarin in relation to the C group.There were no statistically significant differences in alizarin between the L-PRF and A-PRF groups.

Histometric analysis
The means and standard deviations of ANB and the result of the intergroup comparisons are represented in Figure 6.The H-PRF group (n = 8; ANB = 12.67 ± 9.247) showed the highest statistically significant values of ANB (p < .01;p < .05 and p < .01,respectively) compared to groups C (n = 8; 0.2125 ± 0.3934), L-PRF (n = 8; 3.682 ± 3.176) and A-PRF (n = 8; 2.129 ± 1.783).The L-PRF and A-PRF groups showed higher ANB values than the C group (p < .01 and p < .05,respectively).No statistically significant differences in ANB were found for the L-PRF and A-PRF groups (p > .05).

Histopathological analysis
Control group.A connective tissue composed of collagen fibers oriented parallel to the wound's surface was observed along almost the entire length of the surgical defect.In the central part of the defect there was a thickness much smaller than that found in the original bone tissue of the cap.Along the margins of the surgical defect, a small amount of newly formed bone tissue was observed, showing a small number of osteoblasts on its edges (Figure 7a,b).In all specimens, a small number of blood vessels and a moderate number of fibroblasts were found.A generally mild inflammatory infiltrate was also observed, with lymphocytes and plasma cells, distributed along the defect.Different protocols of PRF and bone healing 5 L-PRF group.The newly formed bone tissue exhibited a small number of osteoblasts at its edges.The connective tissue presented numerous bundles of collagen fibers arranged parallel to the wound surface, but more organized and thicker than those observed in specimens from Group C, although still smaller than the thickness of the surgical defect margin.Most specimens from the L-PRF Group had a greater amount of newly formed bone tissue near the margins of the surgical defect when compared to specimens from Group C (Figure 7c,d).In some specimens, bone neoformation was also observed along the defect, forming islets.In addition, a significant presence of blood vessels was observed along its entire length.
A-PRF group.The newly formed bone tissue exhibited a small number of osteoblasts at its edges.The connective tissue had numerous bundles of collagen fibers arranged Different protocols of PRF and bone healing 7 parallel to the wound surface but were more organized and thicker than those observed in specimens from Group C. Most samples from the A-PRF Group had a more significant amount of newly formed bone tissue along the margins of the surgical defect when compared to specimens from Group C (Figure 7e,f).In some specimens, bone neoformation was also observed along the defect, forming islets.In addition, a significant presence of blood vessels was observed along its entire length.
H-PRF group.The newly formed bone tissue exhibited numerous osteoblasts at its edges.The connective tissue had numerous bundles of collagen fibers arranged parallel to the wound surface but were more organized and thicker than those observed in specimens from Group C. Most specimens from the H-PRF Group had a greater amount of newly formed bone tissue along the margins of the surgical defect when compared to specimens from Group C, L-PRF, and A-PRF.In most specimens, intense bone neoformation was also observed along the defect, forming islets (Figure 7g,h).In addition, a significant presence of blood vessels was observed along its entire length.

Discussion
In recent years, literature has focused on proposing new formulations of platelet concentrates obtained by centrifugation of autologous blood, with the objective of improving the structural and molecular characteristics of these matrices, such as mechanical strength and retention and release rates of growth factors.
Variations in Relative Centrifugal Force (RCF) and centrifugation time, as well as the introduction of new centrifuges and tubes for blood collection, led to changes in the original method proposed by Choukroun et al. 7 and the emergence of new protocols such as  A-PRF, and PRF obtained by horizontal centrifugation (H-PRF). 27However, the biological impact in in vivo studies of these modifications is still an obscure field in the literature.The present study is the first to evaluate bone neoformation in CSDs created in rat calvaria and treated with platelet concentrates of autogenous origin, obtained by three different protocols involving Different protocols of PRF and bone healing 9 high and low-speed centrifugation as horizontal and fixed-angle centrifugation equipment.The results obtained showed, in short, that platelet concentrates potentiate bone neoformation in CSDs in rat calvaria.However, the horizontal centrifugation protocol produced platelet concentrates with greater biological potential for bone regeneration.This study used three different centrifugation protocols to obtain PRF membranes.The PRF membranes produced with rat blood showed similar characteristics to the PRF produced with human blood in terms of cell number and distribution, according to findings from previous studies. 1,14,47It is possible to observe that the protocols of low-speed centrifugation and protocols using horizontal centrifugation were more effective in concentrating platelets and distributing them more evenly along the PRF matrix than the original L-PRF protocol.Regarding leukocytes, horizontal centrifugation promoted a 2-fold increase in the number of cells, as well as a more uniform distribution between the PRF layers, when compared to fixed-angle centrifugation protocols.
Fluorescence analysis by confocal microscopy has been widely used in studies involving the evaluation of bone regeneration due to the ability of fluorochromes to bind to calcium ions at the time of their precipitation in the matrix, and the amount of fluorescence allows direct visualization of the amount of bone present.5][46] This study observed increased bone turnover in the L-PRF and H-PRF groups, demonstrated by the higher alizarin labeling and lower calcein labeling.Furthermore, the greater presence of calcein in the L-PRF, A-PRF, and H-PRF groups compared to the control group indicates lower rates of bone resorption.In contrast, the more expressive presence of alizarin in the H-PRF group highlights its potential to promote neoformation bone.In general, all PRF matrices stimulated the process of bone neoformation in the present study.[10][11][12][13] Previous studies have already demonstrated the potential of L-PRF and A-PRF in accelerating bone neoformation of CSDs created in rat calvaria.Higher amounts of newly formed bone were demonstrated in microtomographic and histomorphometric analyses, in statistically significant values, compared to filling the CSDs with clot alone (control group). 37Furthermore, the association of L-PRF with bovine xenograft promoted higher rates of neoformed bone in microtomographic and histomorphometric parameters against experimentally induced osteoporosis and greater expression of essential growth factors for the promotion of bone regeneration, compared to the control group. 48he present study produced PRF matrices using high and lowspeed centrifugation.In vitro, studies have demonstrated better cellular and molecular characteristics in low-speed centrifugation protocols. 1,28,49,50It is thus assumed that structural changes in PRF clots may alter their regenerative potential.In all parameters analyzed in this study, no statistically significant differences were found between the L-PRF and A-PRF groups, which supports the findings of Silva et al. (2022), 37 where the two different types of PRF matrices promoted greater bone neoformation compared to the control group, in microtomographic and histomorphometric parameters, but there were no significant differences between them.Higher rates of release of important growth factors for the promotion of bone neoformation have been demonstrated in low-speed centrifugation protocols compared to the original high-speed centrifugation protocols. 24,28 the other hand, Dohan Ehrenfest et al. 51 observed that the slow release of some growth factors from the original L-PRF membranes was significantly more potent at all experimental times than the release from the A-PRF membranes.It is important to emphasize that more critical than the number of platelets or growth factors present in the PRF samples is the dynamics in the interrelationship between platelets, leukocytes, fibrin, and growth factors. 28It is not just the amount that matters, but how the cells/ molecules are distributed in the PRF matrix, their retention level, and their release pattern.
In an unprecedented way, the present study evaluated with proper statistical power the biological impact of PRF matrices produced from different centrifugation angles on bone regeneration.Membranes produced by horizontal centrifugation provided higher values of BV, Tb.N, ANB, and higher expression of Alizarin, compared to those produced by centrifuges with fixedangle rotors.In fact, over the years, new protocols have been proposed based on concepts of low speed and reduced time centrifugation, such as A-PRF and i-PRF, in addition to the introduction of horizontal centrifugation (H-PRF), to improve the structural characteristics of PRF membranes and increase the bioavailability of cells and growth factors, which could potentiate their regenerative effects. 1,14,24,25,28,47,52,53n the present study, H-PRF samples showed a more homogeneous distribution of leukocytes and a 7.7-fold increase compared to peripheral blood samples.This increase was also higher than those observed in the L-PRF and A-PRF samples.This information supports findings from previous studies, where horizontal centrifugation allowed a 4-fold increase in the number of cells (especially leukocytes) present in the layers corresponding to PRF compared to fixed-angle centrifugation. 14][56][57][58][59] In summary, the present study demonstrated that using fixedangle centrifuges, the centrifugation speed in the production of platelet concentrates did not alter their biological potential.Both L-PRF and A-PRF promoted comparable results in hard tissue regeneration.However, H-PRF increased BV by 464%, 238%, and 206% in relation to control, L-PRF, and A-PRF groups.The differences between H-PRF and other platelet concentrates reported in several in vitro studies 14,26,47,57 may explain these results.It is important to highlight that the literature is still controversial regarding predictability in PRF treatment, especially in bone regeneration.High-quality randomized controlled clinical studies and meta-analyses are required to reverse this limitation of the literature.Therefore, it is essential to standardize the protocols for preparing and applying PRF matrices, minimizing the occurrence of variables that can compromise the effectiveness, treatment, and final result. 30,51ata from this study should be interpreted with caution.The preparation of autologous platelet concentrates in rats required adaptations of the original protocols regarding the total volume of blood collected in each tube.A recent publication has standardized the PRF preparation process for animal studies, emphasizing the importance of a consistent method for evaluating these platelet concentrates. 60This standardization is crucial for ensuring the reliability and comparability of results in animal-based PRF research.Although the present study did not focus on measuring growth factor levels in PRF samples, addressing this aspect in future investigations is necessary to understand the potential impact on experimental outcomes better.Furthermore, isolated PRF matrices were used in the creation of CSDs.Future studies evaluating the association of these matrices with biomaterials, as well as evaluating the correlation between the levels of growth factors, number of platelets, and amount of newly formed bone, are important.Longer follow-up periods, in addition to the use of animals with systemic compromises and animals with a larger phylogenetic scale, should also be performed to elucidate better the role of platelet concentrates in promoting tissue neoformation. 37

Conclusion
Within the limits of this study, it can be concluded that: i) the platelet concentrates L-PRF, A-PRF, and H-PRF potentiate bone neoformation in CSDs in rat calvaria; ii) the horizontal centrifugation protocol produces a PRF with greater biological potential for bone regeneration.

Figure 1 .
Figure 1.Flowchart of the study design.

Figure 2 .
Figure 2. Images of the center of the CSD obtained by confocal microscopy.(a) Calcein precipitation (green); (b) Alizarin precipitation (red); (c) Overlap of images, showing both fluorochromes.

Figure 4 .
Figure 4. Images obtained by confocal microscopy showing the overlapping of calcein (green) and alizarin (red) in the experimental groups.Representative images of the center of the CSD (a, c, e, g) and the region close to margins of the original surgical defect (b, d, f, h).Groups: C (A, B); L-PRF (c,d); A-PRF (E,F); H-PRF (g,h).Arrows represent the margins of the original surgical defect.Original magnification = 10×.

Figure 6 .
Figure 6.Means and standard deviations for ANB for groups C, L-PRF, A-PRF and H-PRF, with the results of comparisons between groups.Different letters indicate statistically significant differences (p < .05).

Figure 5 .
Figure 5. Means and standard deviations for calcein and alizarin precipitations for groups C, L-PRF, A-PRF and H-PRF, with the results of intragroup and intergroup comparisons.* indicates statistically significant differences (p < .05).

Table I .
Percentual distribution of platelets and leukocytes in all layers (A1-A10) after blood centrifugation to prepare L-PRF, A-PRF and H-PRF.