Comparison of Bovine Bone-Autogenic Bone Mixture Versus Platelet-Rich Fibrin for Maxillary Sinus Grafting: Histologic and Histomorphologic Study

In the posterior maxilla, a dental implant treatment may be challenging due to insufficient bone height.¹  Maxillary sinus floor augmentation (MSFA) is a technique for increasing the volume and quality of bone in this region to provide long-term success for dental implants.^2,3  Autogenous bone, bone substitutes, or a mixture of both can be used as grafting materials for MSFA.

Today, autogenous bone grafts are considered as the gold standard for their osteoconductive, osteoinductive, and osteogenic properties.^1,4  However, they may lead to donor site problems, may require hospitalization and general anesthesia, and may cause postoperative pain.⁵  Xenografts are also used as a reliable material in sinus-lifting procedures without such disadvantages. However, these graft materials have the disadvantage of high cost and longer periods of time for healing.⁶  Therefore, new filling materials are still being evaluated to overcome the abovementioned drawbacks of the current bone substitutes.³ 

The addition of growth factors to the grafting materials have gained popularity for their potential to promote bone formation.⁷  Platelet-rich fibrin (PRF) is an autologous fibrin matrix that serves as a resorbable membrane that contains various growth factors.^8,9  PRF is considered as a healing biomaterial that acts as a biological connector between the bone particles, fastens neo-angiogenesis, and supports the survival of bone grafts.⁹ 

The aim of the present histologic study was to evaluate the effects of PRF membrane as the sole filling material to enhance bone regeneration in maxillary sinus floor elevation in sheep and to compare its results with bovine bone-autogenic bone mixture.

The Local Animal Ethics Committee of Erciyes University approved the study protocol. A total of 24 healthy adult sheep were used in the present study. The animals were female, full-grown, aged 2–3 years, not gestating females, and ranged in weight between 40 and 55 kg. General guidelines for care and use of animals in research have been followed. The sheep had free access to drinking water and food ad libitum. Their general condition was checked 3 times a day for pain monitoring to detect variations in well-being and injuries of the musculoskeletal system.

Platelet-rich fibrin membrane was prepared according to Dohan et al.¹⁰  Ten milliliters autologous blood was withdrawn from each sample and centrifuged at 3000 rpm for 10 minutes.^9, The fibrin clot was separated and compressed between 2 moistened sponge gauzes to obtain a resorbable membrane.

The animals fasted for 12 hours preoperatively. Intermuscularly, xylazine (Rompun 2%, Bayer, Istanbul, Turkey) and ketamine HCl (Ketalar, Eczacibasi-Warner-Lambert, Istanbul, Turkey) were used for general anesthesia. Antibiotics (Sulcid 1 g intermuscularly, IE Ulugay, İstanbul Turkey) were administered preoperatively and for 5 consecutive days postoperatively to prevent infection. Local anesthesia was administered for postoperative pain control and hemostasis at the operating sites. Analgesics were also administered postoperatively.

A split face model was used to apply 2 different filling materials for maxillary sinus floor elevation in 22 sheep, and 2 sheep were used as healthy controls. All surgical procedures were performed bilaterally using an extraoral approach to the maxillary sinus. The maxillary sinus area was shaved and disinfected with iodine solution. A bony window, diameter of 1.5 × 1 cm, was created on the lateral aspect of the maxilla, using electro-motors and round burrs. Then the sinus membrane was elevated with special care from the buccal and caudal bony walls and displaced cranially (Figure 1). An autogenous bone block was harvested from the iliac crest, converted to bone chips, and mixed with demineralized bone matrix (Bio-Oss, GeistlichPharma AG, Wolhusen, Switzerland), both of which were 0.6 mL in volume. This bone mixture of 1.2 mL was placed into one of the subsinus cavities (group 1), whereas PRF membrane was placed into the other side (group 2). The PRF membrane was cut into an appropriate size to fit into a 1.2-mL volume. All bony windows were closed using collagen membranes (Bio-gide, Geistlich Pharma AG) (Figure 2). The sinuses were randomly assigned to either group 1 or 2. Randomization was conducted with a shift in assignment for every other sheep.

Experimental groups were euthanized using an overdose of pentobarbital 3, 6, and 9 months after the surgical procedure. The augmented sites of the sinus cavity were harvested and subjected to histologic and histomorphometric examination.

The specimens were decalcified in 10% formic acid solution and embedded in paraffin. Five-micrometer-thick serial and coronally parallel sections (1/50) were taken, and each section was stained with hematoxylin-eosin (HE) and Masson trichrome. Serial sections from the area of interest were selected for the evaluation of new bone formation and PRF remnants under light microscopy (BX50, Olympus, Tokyo, Japan) at ×4 and ×20 magnifications, respectively. The histomorphometric analysis of the digital images of the sections was evaluated with the Adobe Photoshop 7.0 (Adobe, Adobe Systems Software Ireland Ltd, Dublin, Ireland) software program. A scoring system was used to determine the criteria for bone formation including the amount of hyaline cartilage (HC) formation, bony trabecules (BT), connective tissue infiltration into the bone, and PRF remnants (Table 1). Volume calculations were carried out using the principles of Cavalieri, which is a volume calculation method for equally and parallelly sectioned structures. A dotted field measurement scale was placed randomly on the sectional images, and the number of the dots that fall on the area of interest was recorded (Figure 3). The volumes of HC, new bone formation, and the remnants of PRF membrane were then calculated by using the following formula: V = Σp × [a(p)] × t (V, volume of the cross-sectional view; Σp, total number of dots per area; [a(p)], the area represented by a dot; t, section thickness).

Histomorphometric analyses were performed using 3 parameters including the volumes of the HC, new bone formation, and PRF remnants (Table 2).

Statistical analyses were performed using SPSS 17.0 for Windows (SPSS, Chicago, Ill). The one-way analysis of variance (ANOVA) test for variables and Student-Newman-Keuls test for multiple comparisons were used. Unless otherwise indicated, the data are expressed as mean ± SD.

In 4 sinuses in the control group, the sinus membrane kept its original structure consisting of a bilaminar membrane with ciliated columnar epithelial cells on its cavernous side and the periosteum on the osseous side. The thickness of the lamina propria was normal, with serous and mucous glands. The thicknesses of the epithelium and periosteum were normal, and they had a smooth structure. There was no formation of edema and cartilage tissue, and the bone lacunes were regular (Figure 4).

At the third month, the volume of the lamina propria was increased in all specimens, and edema and HC tissue formation were observed. Edema space was surrounded by fibrous capsules, and the collagen structures were regular. The number of blood vessels increased in the connective tissue, and the periosteum maintained its original constitution. Epithelial damage was apparent, and the thickness of the periosteum increased in certain samples. Some samples showed newly formed bone and extensive fibrous cartilage formation that was expected to differentiate in new BT.

At the sixth month, the sinus epithelium reached its original structure with a bilaminar membrane with ciliated columnar epithelial cells, and goblet cells were observed in the lamina propria. The cartilage tissue gradually replaced with new BT surrounded by new osteoblastic cells in connective tissue. Also, the connective tissue was observed to be enlarged.

At the ninth month, the sinus epithelium was normal, consisting of serous and mucous glands. The thicknesses of the connective tissue and periosteum were normal and contained no cartilage tissue. The new bone could no longer be distinguished from the host bone (Figure 5).

The thickness of the lamina propria increased, and the connective tissue advanced into the bone tissue in some of the subjects in PRF group at the third month. The PRF particles that were surrounded by the compact fibrous capsules were seen in the connective tissue. Also, HC formation and damage of the sinus epithelium were observed. The collagen structure was more irregular compared with the graft groups.

The sinus epithelium was observed to be normal without edema at the sixth month. Goblet cells were evident, and HC formation continued. Connective tissue enlargement did not have a normal appearance compared with the graft group. Newly formed bone was seen between the connective tissue and the host bone; however, the amount of BTs was lower than the graft group. Connective tissue volume decreased around the newly formed BTs, and HC volume was observed to be increased.

The connective tissue enlargement into the host bone could be observed in the PRF group, but connective tissue thickness and the epithelium were normal at the ninth month. The new bone formation was still continuing, and healing and bony formation could be seen (Figure 5).

In the graft group, the maximum amount of new bone formation was detected at the sixth month, and HC formation gradually reduced with time (P < .001). In the PRF group, new bone and HC formation were significantly higher at the sixth month and were reduced at the ninth month (P < .001 and P < .05).

New bone formation was significantly greater in the graft group at the third and sixth months (P < .01). At the ninth month, however, new bone formation was significantly greater in the PRF group (P < .01). Hyaline cartilage formation was significantly greater in the graft group at the third month; however, it was greater in the PRF group at the sixth and ninth months.

The maximum amount of PRF remnants was observed at the third month and gradually decreased with time. Connective tissue infiltration increased until the sixth month in both groups but still continued at the ninth month in the PRF group. Connective tissue infiltration had minimal volume in the graft group (Table 3).

The present study was designed to evaluate new bone formation after maxillary sinus floor augmentation either by using a bone graft or PRF membrane. The histomorphometric examinations revealed significant differences between the groups with respect to the predetermined criteria for bone formation in favor of a bone graft. Histologic findings also confirmed its efficacy over the PRF membrane as a sinus-lifting material.

Autogenous bone has been considered the gold standard for bone formation as it has osteoconductive, osteoinductive, and nonimmunogenic properties.¹⁴  Autogenous bone grafts can be harvested from the mandibular ramus, coronoid process, symphysis, maxillary tuberosity, or iliac crest. However, it has certain drawbacks, such as increasing morbidity, a longer operation time, the need for general anesthesia, and prolonged pain. In addition, when autogenous bone is used alone for maxillary sinus grafting, a considerable amount of bone resorption occurs over time.^1,2  For this reason, autogenous grafts are usually combined with allografts, xenografts, and/or alloplastic materials.¹¹  Deproteinized bovine bone (DBB) is a cell-free grafting material with osteoconductive properties.¹²  When combined with autogenous bone, DBB provides many advantages. It increases the bone graft volume, eliminating the need for harvesting large amounts of autogenous bone; shortens the duration of new bone formation¹³ ; adds osteoconductive properties,¹⁴  leading to a more pronounced bone formation; and slows the resorption rate.¹⁵  Hatano et al suggested the use of DBB and autogenous bone mixture in a rate of 2:1 for better long-term results.¹⁶  In the present study, we used the same bone mixture in a rate of 1:1 due to the small volume of the maxillary sinus of the sheep and observed excellent bone formation and remodeling in long-term histologic examinations. This result confirmed the reliability of DBB and autogenous bone mixture as a filling material for maxillary sinus lifting.

The potential of the new bone formation beneath the lifted sinus membrane has previously been shown in the literature.^17–19  This potential is in part due to the osteogenic potential of the Schnederian membrane,^17,20,21  and the concept of guided bone regeneration contributes to the bone regeneration in a bony cavity that is created between the membrane and the floor of the maxillary sinus. This cavity is traditionally filled with various bone grafts and isolated from the nonosteogenic tissues by using a collagen membrane. Based on the abovementioned osteogenic property, recent studies have described graftless solutions, simply elevating the sinus membrane, with simultaneous implant placement, or using space maintainers. In a recent clinical study, Atef et al⁵  used titanium micromesh to maintain the elevated sinus membrane in place without using bone grafts. Six-month postoperative cone beam computerized tomography (CBCT) findings revealed an average of a 6-mm increase at the residual alveolar ridge height, and the technique was concluded to be reliable by the authors.⁵  Despite the use of space-maintaining devices in sinus-lifting procedures, the volume of the newly formed tissue may decrease²³  before the placement of the implants. A potential handicap of this method is that the titanium mesh may need to be removed due to infection. Also, the use of a titanium mesh may raise the treatment cost for the patients. Palma et al¹⁷  compared histologic outcomes of sinus membrane elevation and simultaneous placement of implants with and without adjunctive autogenous bone grafts in primates. To keep the membrane elevated, they used 2 implants, acting as tent poles. The results showed no differences between membrane-elevated and grafted sites regarding implant stability, bone-implant contacts, and bone area within and outside of the implant threads.¹⁷  In addition, they showed the contact of the Schneiderian membrane with the apical surface of the implants in both treatment groups. Accordingly, Scala et al²⁴  demonstrated that the sinus membrane collapsed onto the implant apex at a distance about half of the length of the implants, after the sinus-lifting procedure without bone fillers. Therefore, supporting the sinus membrane with additional materials other than grafts is controversial. In the present study, the tested grafting materials acted as space maintainers without additional supporting materials for the sinus membrane.²⁴  We evaluated new bone formation at the grafted sites histologically; however, no radiographic examinations were performed to determine the bone height after sinus lifting.

Platelet-derived products and growth factors are currently used in oral implantology procedures to promote bone healing and reduce the overall treatment period. Choukroun's PRF is an autologous fibrin matrix that contains large quantity of platelets and leucocyte cytokines.²⁵  Growth factors are released from this fibrin matrix and stimulate cell migration, aiding the neoangiogenesis and vascularization.⁹  The use of PRF as the sole grafting material for maxillary sinus lifting simultaneously with implant placement has recently been described.^26–28  Mazor et al performed 25 sinus elevation procedures with simultaneous installation of implants using Choukroun's PRF as the sole filling biomaterial in 20 patients. Six-months panoramic X rays showed the final bone gain from 7 to 13 mm. The bone was well organized and vital on histologic examinations.²⁶  In a similar clinical study, Simonpieri et al used PRF membranes to cover the Schneiderian membrane and PRF clot to fill the subsinus cavity. The patients were followed up for a minimum of 2–6 years. The final radiologic examination showed that the level of the new sinus floor was in continuation with the implant tips, and the mean vertical bone gain differed from 8.5 to 12 mm. Therefore, the authors defined their method of bone augmentation to be a reliable surgical option.²⁷  The present study evaluated the bone healing potential of a subsinus cavity when filled solely with PRF membrane. Unlike the previous studies, we did not use implants to support Schneiderian membrane for two reasons. First, the sinus membrane may collapse and cause the implants to be introduced into the sinus cavity. This phenomenon was demonstrated in the study of Jeong et al,²⁹  in which the space filled with PRF membrane fell down onto the implants at the 6-month CT images. Second, dental implants cannot always be installed immediately after the elevation of the sinus membrane, due to the insufficient residual alveolar bone height. Although the vertical height of the grafted site was not determined by using advanced imaging techniques in the present study, it can be predicted that the PRF membrane may degrade rapidly, resulting in a decrease in the volume of the subsinus cavity. However, an additional internal lifting can be performed immediately before the implant placement, when the vertical height of the grafted cavity is observed to be reduced. Nonetheless, this speculation is beyond the results of our study, and further studies are needed to evaluate the long-term vertical bone height effect of PRF membrane without the use of dental implants. The lack of using cone beam or plain CT for the evaluation of sinus pathologies or vertical bone height after graft healing is a drawback of our study. Although previous studies favor a CBCT evaluation before sinus lifting procedures, it increases the cost, and it is very difficult to take CBCT images from sheep in clinical settings.

The use of PRF membrane as the sole filling material offers the advantages of the ease of procurement and application, lack of adjunctive grafting materials,²⁹  lack of immunogenic reactions, and promotion of bone formation. However, our results demonstrated that bone regeneration was better with DBB and the autogenous bone mixture. Histologically, new bone formation was detectable at the third and sixth months, whereas it could not be distinguished from the host bone at the ninth month in the graft group. In contrast, new bone formation was still evident at the ninth month in the PRF group, along with residual PRF remnants. The newly formed bone was denser and firmer than the PRF group. Histomorphologically, HC formation, which indicates ongoing bone regeneration, and BTs were higher in the graft group than in the PRF group at the third month. At the sixth month, however, HC formation reduced as the amount of BTs increased. In the PRF group, the maximum amount of HC was seen at the sixth month and reduced to the level of the 6-month results of the graft group at the ninth month. BT formation in the PRF group was also slower than the graft group. These results indicated the slower regeneration potential of PRF membrane compared with DBB and an autogenous bone mixture as a grafting material for maxillary sinus floor elevation.

In conclusion, the autogenous bone and bovine bone mixture represented a superior regeneration potential than PRF membrane grafting of the maxillary sinus. However, our long-term results supported that the Choukroun's PRF is a simple and inexpensive biomaterial, and its application during sinus lifting may be considered an alternative to bone grafts, particularly when minimal bone is needed around the implants. Future studies concerning vertical alveolar bone height after using PRF as the sole filling material for maxillary sinus lifting are recommended.

Abbreviations

BT

bony trabecules

CBCT

cone beam computerized tomography

CT

connective tissue

DBB

deproteinized bovine bone

HC

hyaline cartilage

HE

hematoxylin and eosin

MSFA

maxillary sinus floor augmentation

PRF

platelet-rich fibrin

The authors thank Assistant Professor Ferhan Elmalı (Department of Biostatistics, Erciyes University Faculty of Medicine) for excellent assistance on the statistical analysis. The authors also thank the Scientific Research Council of Erciyes University for financial support of our study with Project TSA-11-3596.

The authors declare no conflicts of interest.