The Role of Bone Grafts in Preventing Medication-Related Osteonecrosis of the Jaw: Histomorphometric, Immunohistochemical, and Clinical Evaluation in Animal Model

Abstract

Objective: To evaluate the effects of inorganic bovine bone graft (Lumina Bone, Criteria, Brazil) and beta-tricalcium phosphate (β-TCP) graft (ChronOS, Synthes, Brazil) in rats with the risk of developing post-extraction medication-related osteonecrosis of the jaw (MRONJ). Methods: Eighteen male Wistar rats weighing 350 to 450 g were induced to develop MRONJ using zoledronic acid for 5 weeks. In the sixth week, the right maxillary first molar was extracted. The animals in Group I (G1) did not receive bone grafts after tooth extraction, while Group II (G2) animals received inorganic bovine bone grafts, and Group III (G3) animals received beta-tricalcium phosphate (β-TCP) grafts. Clinical evaluation and histomorphometric and immunohistochemical analyses were performed. ANOVA and Tukey’s statistical tests were used and a level of significance was considered to be 5%. Results: In the clinical evaluation, animals from G2 and G3 did not present clinical manifestations of osteonecrosis, unlike the control group (G1) animals, which presented necrotic bone tissue exposure in all samples. In the histomorphometric evaluation, animals in G3 showed greater formation of bone tissue (66%) and less formation of bone lacuna (18%) than animals in G1 (58%/32%) and in G2 (59%/27%) (P < 0.05). Moderate (++) immunostaining was observed in G2 and G3 for RANKL, TRAP, and OC, while G1 showed moderate (++) labeling for OC and mild (+) immunostaining for TRAP and RANKL. Conclusions: Greater formation of bone tissue and fewer bone lacunae were found in animals treated with β-TCP. In clinical evaluation, bone graft groups presented with the clinical manifestation of MRONJ and showed higher intensity of immunostaining for TRAP and RANKL. Despite the limitations of experimental animal studies, the results of this work may assist in the development of future clinical research for the prevention of MRONJ.

Introduction

Bisphosphonates (BPs) are drugs that can inhibit bone resorption performed by osteoclasts in the same manner as endogenous pyrophosphates, which are physiological regulators of calcification.[1,2] These drugs bind to hydroxyapatite on the bone surface, inhibiting osteoclasts that attempt to degrade the matrix.[2] These anti-resorptive drugs are used in the treatment of metastatic malignant diseases, Paget’s disease, and osteoporosis.[2,3,4] As these drugs can modify bone remodeling, questions are being raised concerning the influence of the drugs in oral and maxillofacial surgical procedures.[4]

Despite the benefits of BPs, an important complication called medication-related osteonecrosis of the jaw (MRONJ) has been observed in chronically-using patients, which is clinically presented as avascular bone exposure in the maxillofacial region.[4,5,6] One of the main triggers for the manifestation of this disease is tooth extraction.[6,7] Although the first reports of MRONJ were published in 2003, a standard protocol for treatment or prevention has not yet been established, and the American Association of Oral and Maxillofacial Surgery (AAOMS) suggests possibilities of treatment according to the stage of the disease, ranging from antibiotic therapy to more invasive procedures such as sequestrectomy and resections.[6,8,9] Other factors, such as the duration of treatment with BPs, oral or intravenous administration, and the study of biological markers represent important aspects in the preoperative evaluation of these patients.[7,10,11]

Numerous studies have been conducted to understand the risk factors and minimize the occurrence of MRONJ; however, research models in humans are extremely limited, making experimental studies in animals more viable.[12,13] Animal models at risk of developing osteonecrosis have been established for research, and the use of platelet-rich plasma, adipose-derived stem cells, and photodynamic therapy has shown promising results for the treatment of this disease.[14,15,16] Nevertheless, the use of bone grafts to prevent MRONJ after dental extraction has not yet been studied.

The literature argues that the practice of bone grafting and soft tissue covering results in benefits for the alveolus, such as preserving the volume of the alveolar rim, enabling new bone formation, and better soft tissue healing.[17,18] Bone substitutes have osteoconductive properties, which favor the process of bone formation. In this scenario, xenogenous bone grafts have become the preference for alveolar preservation due to their availability and osteoconductive properties.[19,20] These biomaterials are derived from another species, usually bovine, and demonstrate successful bone regeneration and composition similar to that of human hydroxyapatite.[19,20,21] Other bone substitutes also show good results during bone augmentation, such as btricalcium phosphate (β-TCP). β-TCP graft is a microporous bone substitute with a homogeneous, porous structure, which facilitates bone growth.[22,23,24]

Given the need for a post-extraction MRONJ prevention protocol and due to the few studies evaluating the role of bone grafts in this process, this in-vivo study aimed to evaluate the effects of inorganic bovine bone grafts and β-TCP graft in rats at risk of developing post-extraction MRONJ.

Materials and Methods

Study Design and Ethics

All experimental protocols involving animals performed in this study were approved by the Ethical Committee for Animal Care of the Arac¸atuba School of Dentistry e FOA/UNESP (protocol 009402017) and were performed in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC) as well as of the Brazilian Society of Laboratory Animal Science (COBEA) and Arrive Essential 10 guidelines.

Eighteen albino Wistar male rats, mean age of 2 months and weight of 350 to 450 g, were previously induced to develop MRONJ using zoledronic acid (Novartis Pharma Stein AG, Stein, Switzerland). The sample number was selected based on previous similar studies and power analysis on the website http://www.lee.dante.br. Level of significance of 5% and power test of 95% were adopted, and a group size of 5 animals was suggested.[25]All animals underwent surgical procedures for extraction of the upper right first molar and were randomized into 3 groups of 6 albino Wistar rats per group to compensate for the lost follow-up. The groups were selected according to the following treatments: control group (G1)—no graft was performed; experimental group II (G2)—inorganic bovine bone graft (Lumina Bone, Criteria, São Paulo, Brazil); experimental group III (G3)—β-TCP graft (chronOS; DePuy Synthes, Paoli, CA, USA) (

Table 1. Distribution of Animals by Groups and Grafting Materials Used.

Group	Grafting material	Number of animals
1	Control (Clot)	6
2	Inorganic bovine bone graft (Lumina Bone®)	6
3	β-TCP(chronOS®)	6
Total		18

).

MRONJ Rat Model

All animals received IV injections of 0.04 mg/kg zoledronic acid (Novartis Pharma Stein AG, Stein, Swiss) once a week for 5 weeks in the tail vein. While there is no unanimous protocol of MRONJ induction in an animal model, this dose was adapted from Hokugo et al in 2010 and is already used in the experimental treatment of MRONJ in rats.[12,13,15,26]

Surgical Procedure

In the sixth week of treatment, surgical extraction was performed. At the beginning of the surgical procedures, the animals were sedated with intraperitoneal administration of 1% ketamine (0.20 ml/kg) and 2% xylazine (0.30 ml/kg) (Francotar, Virbac Ltda., Sao Paulo, Brazil). The animals were positioned in the supine position to facilitate oral cavity access. Antisepsis of the oral mucosa was performed with topical 1% polyvinylpyrrolidone aqueous solution before tooth extraction. Animals first had their gingiva detached using specific retractors, and the teeth were dislocated from the alveoli with adapted dental forceps. In G1, no treatment was performed. In G2, the alveoli were implanted with an inorganic bovine bone graft (Lumina Bone, Criteria, São Paulo, Brazil), and in G3 the alveoli were implanted with β-TCP (chronOS, DePuy Synthes) (Figure 1). The wounds were sutured with nylon (Mononylon 5-0, Ethicon, São Paulo, Brazil), and all animals were subcutaneously injected with 0.075 mg/kg tramadol every 24 hours (h) for 3 days.

Euthanasia

Euthanasia was performed 28 days after dental extraction. The rats were euthanized with an anesthetic overdose (Sodic Thiopental, 150 mg/kg), and the right maxilla was removed and immediately immersed in 10% buffered formalin.

Histomorphometric and Histological Analysis

The specimens were washed in running water for 24 h and demineralized in ethylenediaminetetraacetic acid (EDTA) solution for 4 weeks. When demineralization was achieved, samples were embedded in paraffin to maintain the apical orientation. Longitudinal 5-mm thick sections were cut at 60 mm intervals, placed on slides, and stained with hematoxylin and eosin. The biopsies were evaluated by light microscopy, and the images were captured using an attached digital camera (JVC TK1270 color video camera) at 12.5× magnification. New bone formation, connective tissue, and bone lacunae were analyzed by histomorphometry using a Merz grid added to the images in PowerPoint forMac (Microsoft, Redmond, WA, USA)[27] (Figure 2). The parameters analyzed were: empty bone gaps, inflammatory process in the connective tissue, peripheral resorption, bacterial colonization in the medullary space, healthy bone tissue, and necrotic bone tissue.

Immunohistochemical Analysis

Primary polyclonal antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA) anti-RANKL, anti-TRAP, and antiOC were used in the immunohistochemical assays with the concentration of 200 mg/ml. The objective of using these markers was to analyze cellular responses regarding the processes of bone mineralization (OC), bone remodeling (RANKL), and osteoclastic activity (TRAP).

The data analyses were performed using a singleevaluator, semi-quantitative approach, with the scores ranging from “—” to “+++” (“—”: absent, “+”: mild, “++”: moderate, and “+++”: intense), according to previous studies.[22,23,27]

Clinical Evaluation

The extraction sites were evaluated at the time of euthanasia (10 weeks) using a single evaluator, and the following diagnostic criteria for MRONJ were considered: presence of bone exposure, fistulae, suppuration, and signs of soft tissue inflammation.

Statistical Assessment

Data obtained from histomorphometric assessment were organized in a table (Excel, Microsoft Office Excel, Redmond, WA, USA) and analyzed in relation to their normal distribution using the Shapiro-Wilkes test. Comparisons between groups were performed using analysis of variance (ANOVA) followed by Tukey’s multiple comparison test. (SigmaPlot 12.3; Systat Software, San Jose, CA, USA). A P-value <0.05 was used for all tests.

Results

Clinical Evaluation

The presence of necrotic bone exposure was found in 5 G1samples. One animal in G1 presented with a fistula at the extraction site. The animals in G2 and G3 did not show any signs of MRONJ and presented normal soft tissue healing (

Table 2. Macroscopic Evaluation of the Surgical site Demonstrating Normal soft Tissue Healing in all Animals of G2 and G3, and Necrotic bone Exposure and Unhealing Soft Tissue in all Samples of G1.

Group	Necrotic Bone exposure	Normal soft tissue healing	Total of Animals
G1: Control Group	06	0	6
G2: Inorganic bovine bone graft (Lumina Bone®)	0	6	6
G3: b TCP(chronOS®)	0	6	6

and Figure 3).

Histological Analysis

Group 1. The control group presented with intense bone necrosis associated with an infectious process. The presence of large colonies of bacteria on the surface of alveolar bone and a moderate quantity of inflammatory cells (lymphocytes and macrophages) in bone tissue and connective tissue was found. All samples showed focal areas of osteonecrosis with empty bone lacunae as well as some areas of healthy bone tissue with the presence of osteocytes. There was an interruption of epithelial tissue coverage at all extraction sites (Figure 4).

Groups 2 and 3. The experimental groups had minor inflammatory processes in the connective tissue and had healthy bone tissue with osteocytes present. In addition, highly cellular connective tissue was observed in the presence of vessels and mature new bone formation. Areas of bone lacunae were found in G2 and were found in G3. The presence of remaining biomaterial could be observed in G2 and less so in G3, with new bone formation on the periphery. There was epithelial tissue with moderate thickness completely covering all G2 and G3 extraction sites (Figure 4).

Histomorphometric Analysis

The rate of bone formation was 58 + 4.2% in G1, 59 + 5.7% in G2, and 66 + 4.7% in G3. Statistical differences occurred between groups 1 and 3 and between groups 2 and 3 (P < 0.05), no statistical difference was found between groups 1 and 2 (Figure 5 and Figure 6). During bone lacunae assessment, G1 presented with 32 + 5.2%, G2 with 27 + 5.6%, and G3 with 18 + 4.2%. Statistical differences for bone lacunae occurred between groups 1 and 3 and between 2 and 3 (P < 0.05), no statistical difference was found between groups 1 and 2 (Figure 4).

The percentage of soft tissue was 10 + 1.6% in G1, 13 + 4.7% in G2, and 15 + 3.7% in G3. No statistically significant difference was found between groups during the evaluation of epithelial tissue (P = 0.1280) (Figure 5).

Immunohistochemical Analysis

OC showed moderate (++) labeling in G1; however, mild (+) immunostaining was found for TRAP and RANKL.

Moderate (++) immunostaining was observed in G2 and G3 for RANKL, TRAP, and OC (

Table 3. Scores of Immunostaining for TRAP, RANKL, and OC Distributed in Groups 1, 2, 3. “—”: absent, “+”: mild, “++”: moderate, and “+++”: intense immunostaining.

	TRAP	RANKL	OC
Group 1	+	+	++
Group 2	++	++	++
Group 3	++	++	++

and Figure 7).

Discussion

MRONJ was first reported in 2003, when little was known about this pathology. Since then, various studies have reported the use BPs as being risk factors for this disease.[4,6,10] Even with knowledge spreading, the indiscriminate use of BPs and the lack of a prevention protocol for MRONJ represents a major problem contributing to the increase in these cases. The eminent need for dental extraction in a patient at risk of MRONJ poses a great challenge for all surgeons.

An ideal model for MRONJ studies in rats has not been established due to differences in doses, drugs, and duration of MRONJ induction in animals. The induction protocol used in this study was based on converting the therapeutic dose used in humans to animals, previously confirmed by Biasotto et al, and has already been used for experimental treatments of MRONJ in rats.[12,13,14,15,16,26]

In 2015, Howie et al were able to perform bisphosphonate removal from the bone matrix using chelating agents, and in 2017 Zandi et al obtained good results by using lowdose teriparatide for the stimulation of osteoblastic and osteoclastic activity in rats. Despite being promising, these methods require further studies and are expensive.[28,29]

In order to identify new methods of prevention, this work carried out an experimental model to investigate the prevention of post-extraction MRONJ using bone grafts. The formation of healthy soft tissues without the presence of necrotic bone exposure is the primary objective of MRONJ management.[5,6] The clinical evaluation of animals treated with bone grafts in this study showed normal soft tissue healing for all samples. Barba-Recreo et al also had the same results, but with the use of platelet-rich plasma.[16] Cardoso et al found bone exposure and bone sequestration after dental extraction in rats treated with zoledronic acid.[26] In the present study, we also found the clinical manifestation of MRONJ in animals in G1, which had necrotic bone exposure 4 weeks after dental extraction.

Sarkarat et al performed a similar study with the use of platelet-rich plasma. They found a greater formation of bone tissue and fewer bone lacunae in the platelet-rich plasma group.[15] These results are in line with our experimental study, where we found greater bone formation and significantly fewer bone lacunae in G3. Despite the fact that inorganic bovine bone grafts have biological properties similar to those of β-TCP, G2 animals did not show significant results when compared to the control group.

Statkievicz et al analyzed the immunolabeling of TNFa, IL-1b, and IL-6 and showed that zoledronate prevented total connective tissue repair. In addition, they showed that the increased expression of pro-inflammatory cytokines and the negative effect on collagen fiber maturation contributed to severe impairment of bone repair.[14]

Hassumi et al demonstrated a significant function of the OPG/RANK/RANKL system in the osteoclast and osteoblast responses during alveolar bone healing.[30] Patients with MRONJ show decreased osteoclastic activity and bone remodeling, and this finding makes the evaluation of TRAP (active osteoclast) and RANKL (bone remodeling) immunostaining an important assessment for this disease. In our study, G2 and G3 presented moderate labeling of these proteins 28 days after dental extraction, while the control group showed only mild labeling. Despite these limitations, these results can be interpreted as greater bone remodeling in the experimental groups, which represents a favorable factor in cases of MRONJ.

Even with the limitations of experimental animal studies, the results of this work may assist in the development of future clinical research for the prevention of MRONJ.

Conclusion

Within the limitations of this study, it can be concluded that greater formation of bone tissue and fewer bone lacunae were found in animals treated with β-TCP. In clinical evaluation, bone grafts prevented the clinical manifestation of MRONJ and bone graft groups showed higher intensity of immunostaining for TRAP and RANKL in immunohistochemical analysis. More studies are needed to establish the clinical application of bone grafts in the prevention of MRONJ.