The Influence of Eggshell on Bone Regeneration in Preclinical In Vivo Studies

Simple Summary The aim of this study is to review the available information on the use of avian eggshell as bone regeneration material. Five databases were searched up to October 2020. Animal studies with a bone defect model using eggshell as a grafting material were included. Risk of bias and the quality of the papers were assessed. Overall, a total of 581 studies were included in the study, 187 after duplicate removal. Using the inclusion and exclusion criteria 167 records were further excluded. The full text of the remaining 20 articles was assessed for eligibility and included in the review. There were different methods of obtaining eggshell for grafting purposes. Eggshell is a biocompatible grafting material, with bone formation capabilities. It forms new bone similar to other products currently in use in clinical practice. It can be combined with other materials to enhance its proprieties. Eggshell is a promising biomaterial to be used in bone grafting procedures, though further research is needed. Abstract The aim of this study is to systemically review the available evidence on the in vivo behavior of eggshell as a guided bone regeneration substitute material. Five databases (PubMed, Cochrane, Web of Science, Scopus, EMBASE) were searched up to October 2020. In vivo animal studies with a bone defect model using eggshell as a grafting material were included. Risk of bias was assessed using SYRCLE tool and the quality assessment using the ARRIVE guidelines. Overall, a total of 581 studies were included in the study, 187 after duplicate removal. Using the inclusion and exclusion criteria 167 records were further excluded. The full text of the remaining 20 articles was assessed for eligibility and included in the qualitative and quantitative assessment synthesis. There were different methods of obtaining eggshell grafting materials. Eggshell is a biocompatible grafting material, with osteoconduction proprieties. It forms new bone similar to Bio-Oss and demineralized freeze-dried bone matrix. It can be combined with other materials to enhance its proprieties. Due to the high variability of the procedures, animals, production and assessment methods, no meta-analysis could be performed. Eggshell might be considered a promising biomaterial to be used in bone grafting procedures, though further research is needed.


Introduction
Guided bone regeneration (GBR) is the method used in oral surgery to increase the volume of available host bone in sites chosen for dental implant therapy [1]. The original concept that led to size, type of material used, intervention site, clinical and radiological assessment, biopsy (histology), follow-up, complications, excluded subjects, outcome.

Outcome Measures
Primary outcomes: New bone formation can be measured with different techniques such as histomorphometric analysis, radiographic analysis like computer tomography, micro-CT, standard radiograph and residual biomaterial.
Secondary Outcomes: any complications and adverse events related to the biomaterial used. The eggshell derived biomaterials and characterization were also investigated.

Quality Assessment and Risk of Bias Assessment
The quality of the studies was assessed independently by two reviewers (H.O. and G.A.), based on the ARRIVE (Animal in Research: Reporting In vivo Experiments) guidelines [10]. The items considered were the following: ethical statement, experimental procedures, experimental animals, randomization, allocation concealment, sample size calculation, completeness of information, blinding of the evaluator and financial conflict of interest.
SYRCLE's risk of bias tool for animal studies [11] was used for the quantification of risk of bias in ten domains. All items could be judged as yes/no/unclear. Studies were considered at high risk of bias if at least two items were judged as "no". Studies were judged as low risk of bias if at least seven items were judged as "yes", and no item was judged as "no". In other cases, the studies were considered at medium risk of bias. Two reviewers independently assessed the risk of bias for the included articles and if any disagreement occurred, a third reviewer intervened.

Study Selection
The electronic search provided 518 articles that were reduced to 187 after the duplicate removal. No further articles were identified by manual search. Screening of titles and abstracts led to the exclusion of 167 records. The full texts of the remaining 20 articles were obtained. These papers were analyzed systematically and quality-wise. The flow diagram of the search results is shown in Figure 1.

Study Characteristics
Only qualitative data was extracted from each study and it was synthetized in analytic tables. Tables 1 and 2 summarize the definition of the critical sized defect in the selected studies. In eight of the included papers Wistar Rats were used [5,6,[12][13][14][15][16][17], six studies used Sprague-Dawley rats [8,[18][19][20][21][22] and seven New Zealand rabbits [6,[23][24][25][26][27][28]. The calvaria critical-sized bone defect was used in 90% of the included papers to assess new bone formation (Tables 1 and 2).      microCT: there is a statistically significant difference between the control group and the grafted groups. The nHA+ silk fibroin showed more bone formation than the nHA group. Histomorphometry: nHA group showed good bone formation with well-organized lamellar bony islands. The space formed by silk degradation was replaced by new bone but the most area was occupied by poorly degraded biomaterial.
None of the studies included the use of eggshell scaffold with bone marrow-derived mesenchymal stem cells. Non-Resorbable membranes such as ePTFE (expanded polytetrafluoroethylene) were used for GBR in two studies [13,14]. Collagen resorbable membranes were used for guided bone regeneration in one study [17]. A paper [23] attempted to use eggshell membranes as a resorbable alternative in bone grafting. Histology was used in all studies to assess bone healing (n = 20), histomorphometry in 30% of papers (n = 6), microCT in 25% of papers (n = 5) and contact radiograph in 30% of papers (n = 6). Other less often used investigations include immunohistochemistry (n = 1) and fluorescent labeling (n = 1). Follow-ups ranged from 2 to 24 weeks. A single observation interval was reported for eight of the included studies [5,6,12,13,15,22,23,28], while the rest had multiple observation points.
Because chemical composition and processing technology are considered important factors for determining the benefit of using the biomaterial, they were analyzed and summarized in Table 3 for the hen and ostrich eggshell. The most employed methods of assessing the materials are scanning electron microscopy (55%), x-ray diffractometry (40%), Fourier-transform infrared spectroscopy (30%) and energy-dispersive x-ray spectroscopy (10%). SEM: particles of N-HA showed rectangular shape, some were aggregated; The silk fibroin scaffold was web shaped, with highly porous structures with a round shape at the end. When the N-HA was precipitated into the silk fibroin, the particles were evenly distributed to the surface of the scaffold.

Ref. Biomaterial Production Method and Sterilization PROPRIETIES Eggshell Originating from Struthio Camelus (Ostrich)
[6] Block 2 mm thick The shell was immersed in 10% sodium hypochlorite for 24 h. They were washed and autoclaved.
- [23] 375 Eggshell was ground, washed, dried, and sterilized using ethylene oxide. - [22] 300-500 Particles were obtained using mill and sieve, followed by immersion in sodium hypochlorite solution for 48 h. After it was washed. Treatment: Alkaline etching (microroughened-OES) Immersion in supersaturated calcification solution (CaP coated OES) All particles were sterilized using gamma irradiation.
SEM: alkaline etching increased surface area; calcium phosphate coating showed platelet-like morphology of crystals.
[24] 700-1500 Small pieces of eggshells were put into Petri dish with glutaraldehyde for 24 h. After they were washed and ground. They were sterilized using ethylene oxide.
- [15] 20-150 or Block 7 mm Ø Inner and outer membranes were removed. The eggshells were broken into small pieces of 7 mm diameter. Powder was then prepared with electrical burr. Sterilization was done with ethylene oxide. A variety of production methods are described for the eggshell grafts, starting with eggshell blocks [6,15]. 75% of the included studies produced the biomaterial by crushing and milling eggshell into a powder. Calcination is being used in 35% of the cases to treat eggshell to improve its bone regeneration capabilities (n = 7). Hydrothermally treating the grafts greatly improves the porosity and structure of the eggshell [8,18]. A study [25] designed the use of silk fibroin scaffolds with eggshell to try to increase the proprieties of the biomaterial. Others [21] used a combination of carrageenan gel, xanthan gum gel to manufacture new biomaterials. Pure brushite cement was used [17] to find the best cement as bone augmentation material.
One study employed a commercially available nutritional supplement derived from eggshells used to promote healthy bones [28]. One study used a scaffold with carboxymethyl chitosan and BMP2 to enhance bone regeneration [12]. The ostrich eggshells were used in particles ranging from 20 to 1500 µm size. Sterilization of the products was done by either autoclaving, gamma irradiation or ethylene oxide.

Studies in Rabbits-Main Features
The characteristics and the main features of the studies in rabbits are summarized in Table 2. The included papers reported uneventful healing outcomes and no relevant adverse reactions. All included studies used the calvaria bone defect model with a range of 6 to 15 mm diameter, all except one [24] being full thickness. The number of defects per animal varied from one to six. Ostrich eggshell particles of different sizes were used in two studies [23,24]. No inflammatory reaction was noted in either study, bone regeneration seems to begin from the margins of the defect. Smaller graft particles resorb faster than larger ones. When compared to the demineralized bone matrix (DBM) graft, ostrich eggshell resorbs slower and produces less bone than DBM. Ostrich eggshell block was used in one study as an interposition graft [6]. When analyzed after the follow-up period, the interposition grafts seem to be delineated from the surrounding bone with no signs of remodeling.
Another study [26] compared the eggshell hydroxyapatite with the synthetic hydroxyapatite in the rabbit model and found no difference between the groups regarding new bone formation, both having low inflammatory response.
One of the papers [28] employed a nutritional supplement commercially available product from eggshell as a guided bone regeneration material-Membrell's Bonehealth Plus. After two weeks of follow-up there was no sign of inflammatory response, the graft tissue was completely resorbed and there was deposition of newly formed bone. Eggshell particles combined with silk fibroin in a rabbit calvaria bone defect model showed similar results to eggshell particles alone regarding new bone formation [25].

Studies in Rats-Main Features
The characteristics and the main findings of the studies in rats are provided in Table 1. None of the studies reported adverse effects of the biomaterials used. To evaluate the bone regeneration capabilities of the biomaterial, 11 of the studies used the calvaria model, two studies [5,21] worked on a mandible/maxilla defect model and one used a periodontal bone defect [16]. Defect size ranged from 4 to 5 mm diameter.
Only one study [5] employed the use of the split mouth model, with one defect left empty and one with the graft. The biomaterial was found to have a uniform distribution in the defect site and be surrounded by a thin fibrous layer.
Dupoirieux et al. [6] used an ostrich eggshell implant as an interposition graft to compare it to an empty defect. The in vivo model showed that the block had little resorption and there was a radiolucency surrounding the implant, mainly formed of fibrous tissue.
Two studies by Dupoirieux et al. [13,14] used expanded polytetrafluoroethylene (ePTFE) membranes for guided bone regenerating in the calvaria bone defect model. The first paper [13] used carboxymethyl cellulose and pentosan polysulphate alongside eggshell powder. Bone healing started from the margins with centripetal bone formation. The group containing eggshell showed the most resorption with interposition fibrous tissue. In the second study [14] the non-resorbable membranes were used and compared to a periosteal graft and to eggshell powder. The results show no resorption of the eggshell powder and no complete closure with bone in this group.
Only one study [16] used a split mouth periodontal defect model to compare the eggshell particles to demineralized freeze-dried bone matrix (DMBM), with the use of collagen membranes on all grafts. The tissues healed with new bone forming from the margins of the defect with more connective tissue in the graft group.
Another paper [12] used eggshell particles to create a scaffold for bone morphogenetic proteins 2 (BMP2). The results were excellent with new bone formation, complete cover of the defect and enhanced bone formation capabilities in the scaffold group.
Using ostrich eggshell powder and an eggshell implant to compare them to DMBM, Uygur et al. concluded that there was no difference regarding new bone formation, but the DMBM still produces significantly more new bone [15].
Three studies [8,18,19] compared the eggshell with Bio-Oss. Park et al. [8] compared surface-modified eggshell particles with calcium phosphate with Bio-Oss. In the Bio-Oss group, bone healing was incomplete and there was no resorption of the graft. In the eggshell group complete bony closure was seen (40%), whereas in the surface-modified eggshell lot complete bone bridging was observed more often (80%). The authors have reported superior new bone regeneration in the surface-modified eggshell particles group. In another publication [19] the eggshell particles were compared with the Bio-Oss and eggshell hydroxyapatite enhanced with calcium sulphate. The most newly formed bone was with the nanohydroxyapatite derived from eggshell. Comparison of the use of deproteinized eggshell particles with hydrothermally treated eggshell particles and Bio-Oss revealed that there is new bone formation for all the groups, but the hydrothermally treated eggshell powder had almost complete bone healing with significantly greater new bone formation, even than Bio-Oss [18].
Another paper combined the use of eggshell derived biomaterials with brushite cement to produce an eggshell brushite cement [17]. This guided bone regeneration model uses collagen membranes on all grafted sites. The newly developed material resorbs faster than pure brushite and the newly formed bone starts from the margins of the defect and is present in a greater quantity. The pure brushite on its own is surrounded by inflammatory cells, unlike the eggshell brushite cement.
In another paper, Biocoral is compared to microroughened ostrich eggshell particles and calcium phosphate coated ostrich eggshell particles [22]. There were no significant differences between the grafted groups with new bone being formed from the margins of the defect and around the particles in the center. It seems that the Biocoral and the eggshell derived biomaterial have both similar bone regeneration proprieties.
Only one study [20] compared the eggshells to synthetic hydroxyapatite in which more new bone resulted especially around the grafted particles. Interestingly, around the synthetic hydroxyapatite there were more foreign body multinucleate cells. Comparing the bone volume on a microCT scan, the eggshell group showed significantly more volume than the synthetic hydroxyapatite.
Alternative biomaterials derived from eggshell were produced and a combination of carrageenan gel and xanthan gel have been used in the bone defect model [21] to compare it with the eggshell particles. The combination between the eggshell particles and carrageenan gel determined complete defect healing at the end of the study period.

Quality and Risk of Bias Assessment
A quality assessment was undergone according to ARRIVE guidelines [10] with nine scoring criteria (Supplementary Table S1). The result of the evaluation is provided in the Supplementary Table S2. A percentage of 70% (n = 14) of all of the studies reported data on the ethical statements with the exception of five which have failed to provide any information [5,6,13,14,18].
In all studies complete and adequate information was reported regarding the experimental procedures and the experimental animals. Regarding the allocation concealment and blinding the evaluator, the information was incomplete in all the papers. Most studies (70%) have incomplete financial information and lack the financial acknowledgment except for 6 studies (30%). Eight of the research papers (40%) randomly assigned the animals in different treatment groups. Only one [16] (5%) calculated the sample size using G Power software.
The risk of bias assessment of the selected studies using SYRCLE tool [11] is provided in the Supplementary Table S3. None of the studies have reported on blinding of the care giver, the investigator, or the assessor. Using the tool mentioned we assessed all papers to have a high risk of bias.

Discussion
The aim of this systematic review was to investigate the role of eggshell biomaterials used in guided bone regeneration in critical sized bone defects on experimental animal models compared to an empty defect or other filling materials. Overall, the results show that eggshell induces new bone formation compared to an untreated empty defect. There was little spontaneous bone regeneration in the control groups. When DMBM [16] or Bio-Oss [8,18,19] was used, similar results concerning new bone formation was observed.
Studies conducted on animal models demonstrated the beneficial use of bioceramic scaffolds in guided bone regeneration procedures. The architecture of the material is the key to conduct proper bone healing. An interconnected porous structure similar to natural bone should be present facilitating cell ingrowth, proliferation, and differentiation [29]. The biomaterial should possess adequate mechanical proprieties necessary for a functional loaded area such as the alveolar bone [30].
Resorption rate of the biomaterial should be matched with the osteogenesis rate occurring in the new bone [31]. One way of modifying these rates is to use composite materials alongside biodegradable polymers [32].
Recent studies using PLA (Polylactic acid) and human bone marrow stem cells as a bone substitute on a rodent calvaria bone defect model showed good results with new bone formation at 8 and 16 weeks with no residual material. The groups containing stem cells showed more bone formation at the end of the observation time [33]. Another study using the same in vivo model, compared PLA+ HA (hydroxyapatite) with demineralized bone matrix (DBM) and beta tricalcium phosphate. The beta tricalcium phosphate groups showed more new bone formation followed by the PLA+ HA and DBM groups [34].
Comparison of the included articles with human studies using eggshell as a bone regeneration substitute material shows similar results concerning bone regeneration, resorption rate and lack of immune response of the host [7]. The human studies use different bone defect model (apicectomy, cystectomy, third molar extraction site), but none use clearly defined guided bone regeneration model which does not heal spontaneously [35].
Other methods of producing biomaterials derived from eggshell can be the use of femtosecond laser processing to produce calcium carbonate 3D nanofibrous structures from eggshells [36]. Arslan et al. [37] used an in vitro model to produce a bio composite scaffold from eggshell waste, a collagen-keratin-nanohydroxyapatite with osteoinductive proprieties.
Polyether ether ketone (PEK)-biphasic bioceramic was used in a rabbit bone regeneration model with vascular endothelial growth factor with very good results at the end of the observation period with a significant new bone formation [38].
A combination of nanohydroxyapatite derived from eggshell enhanced with ZrO 2 and Al 2 O 3 was researched by Naga et al. [39] to increase porosity and decrease bulk density thus increasing the bone regeneration proprieties of the material. Further research is needed for this material using a bone defect model to test the in vivo proprieties.
The studies included with the rabbit model all have the calvaria bone defect. Other research has used other sites such as the mandibular bone [38,40], femoral defect [32] and radial defect [41].
The most frequent reason for exclusion in the systematic review was the lack of a bone defect model (64 studies excluded) and the fact that the papers did not present experimental models, but in vitro studies (66 papers excluded). Numerous eggshell biomaterials have been extensively developed and tested in vitro, but few studies developed in vivo bone defect models to test the tissue reaction and the bone formation.
In the included studies no extensive physico-chemical and mechanical characterization was undertaken. Regarding the fabrication of the biomaterials, summarized in Table 3, milling, calcination and hydrothermal treatment were the main procedures implied in fabrication. One study [28] used a commercially available nutritional supplement derived from eggshell as a bone regeneration material which was not indented for such specific use.
There were no studies on other animals closer to the human, like dogs, pigs, sheep, non-human primates, which could provide better insight due to the resemblance to the human metabolism. None of these studies included in the search had a bone defect model to match the inclusion criteria for this review.
Regarding the incorporating active molecules, only one study [12] achieved incorporating BMP2 into the eggshell graft with good results of newly formed bone.
The xenogenic bone substitutes of hydroxyapatite have been manufactured extensively and can be of synthetic origin, derived from corals or algae, or originated from natural bone mineral. It is considered that this material offers biocompatibility and osteoconduction properties (scaffold for the new bone formation). Depending on the particle size and the three-dimensional structures, it can exhibit different integration and resorption rates [42]. When used in alveolar ridge preservation, they offer good width and interproximal bone preservation [43]. Although the production and development are very different from one study to another, eggshells generally showed good biocompatibility, slow resorption rate (in inverse relationship with the size of the particles), osteoconductive proprieties (bone generally tends to heal from the margins of the defect as the graft resorbs).
The eggshell has several other uses in the medical field and the most important are the following: osteoporosis and joint mobility as a supplement, calcium supplement, orthopedics, cancer patients to boost muscle gain and hair thickening, sports nutrition to enhance performance [44].
Further limitations of this study are the fact that it includes only papers in English. Studies with ectopic tissue models were excluded. These could have provided only insight into the host soft tissue reaction to the grafts. When assessing the risk of bias all studies were evaluated as having high risk of bias according to the SYRCLE's tool. There is a lot of debate about the lack of quality information in the animal experiments and because of which a representative risk of bias can be assessed. The risk of bias thus assessed with SYRCLE's tool has to be viewed with its inherent limitations [45].
The calvaria model, which is the most often used in the papers included (85% of articles), is a very widespread model for the evaluation of bone regeneration materials. Nonetheless, its assessment does not offer insight to the biological response to the physiological biomechanical loading occurring in the alveolar bone [46].
There is an ongoing debate on what a critical size bone defect is for each model. In the calvaria model for rodents there are numerous advantages of a 5 mm diameter defect [47], with little spontaneous healing from the bone margins at 12 months [48]. Regarding the rabbit calvaria model, numerous studies have found that there is a direct correlation between the size of the defect and the time it can be considered a critical sized bone defect. Defects of 6 mm, 8 mm and 10 mm width are critical up to 12 weeks [49].
Due to the lack of standardized quantitative measurements, a meta-analysis could not be elaborated. Furthermore, there is a lack of homogeneity in reporting histomorphometry data. There is no standardized quantitative measurement applicable for all cases. However, due to the high variability of the included studies, the composition and the use of the biomaterials, their production methods as well as follow-up duration, no decisive statement can be issued regarding the clinical effectiveness of eggshell as a bone regeneration material. There is still place for further research, to design an in vivo animal model with standardized parameters (adoption of critical sized defects, empty control group, analysis), to allow further comparison with similar studies. There is still a need for further research regarding the optimal evaluation for each defect type for the different animal models.

Conclusions
Within the limitations of this review, eggshell derived grafting materials demonstrated to be osteoconductive in a variety of small animal bone regeneration models, showing better results compared to an empty defect and similar results with Bio-Oss or DMBM. The results show that these biomaterials are promising candidates as bone space fillers. Eggshell particles could be routinely applied to bone defects, to promote tissue healing. Regarding its use as a scaffold for stem cells or growth factors, there is still place for further study.
Supplementary Materials: The following are available online at http://www.mdpi.com/2079-7737/9/12/476/s1, Table S1. Categories to assess the quality of the included studies; Table S2. Study quality assessment-ARRIVE guidelines; Table S3. Risk of bias assessment of the selected studies according to the SYRCLE tool. Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.