Review of the Literature on the Current State of Periosteum-Mediated Craniofacial Bone Regeneration

Abstract

Study Design: This is an article review on the current state of periosteum-mediated bone regeneration (PMBR). It is a known mandibular reconstruction option in children, and though poorly understood and unpredictable, the concerns of developmental changes to donor and recipient tissues shared by other treatment options are nonexistent. The definitive role of periosteum during bone regeneration remains largely unknown. Objective: The objective is to review the literature on the clinical and molecular mechanism evidence of this event. Methods: Our search methodology was modeled after the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) guidelines. Search strategies were categorized into search 1 for clinical evidence of mandibular regeneration and search 2 for gene expression review for craniofacial regeneration. The quality assessment of each publication was undertaken, and inclusion criteria comprise mandibular continuity defect for search 1 and use of gene expression assay propriety kit for search 2. Results: 33 studies were selected for search 1 while four studies with non-human subjects were selected for search 2. Monitoring of PMBR onset was advised at 2 weeks post-operative, and the gene expression results showed an upregulation of genes responsible for angiogenesis, cytokine activities, and immune–inflammatory response in week 1 and skeletal development and signaling pathways in week 2. Conclusions: The results suggest that young periosteum has a higher probability of PMBR than adult periosteum, and skeletal morphogenesis regulated by skeletal developmental genes and pathways may characterize the gene expression patterns of PMBR.

1. Introduction

Several techniques are used in the reconstruction of childhood mandibular defects and till date, about 29 cases of pediatric mandibular reconstruction has been reported in the English literature of which 13 were free flap reconstructions, 11 were non-vascularized reconstructions, and 5 used both techniques [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29]. Due to developmental changes of bone and soft tissue donor and recipient sites in children, secondary mandibular reconstruction is usually preferred [26,30]. However, the clinical event, periosteum-mediated bone regeneration (PMBR), is thought to sometime occur and is today considered a third pediatric bone reconstruction option [31,32]. Mandibular PMBR is not clearly understood but is characterized by new bone formation in a defect from resection of the mandible. Indications for mandibular PMBR include defects from periosteum sparing benign jaw tumors, cystic enucleation, osteomyelitis, and traumatic avulsions. Despite its poor understanding and unpredictability, there are no concerns about changes in donor and recipient tissues during growth and development, often seen in other treatment options. Similarly, bone induction during both bone healing and repair comprises four successive phases: an initial inflammatory response and recruitment of osteo-progenitor cells, formation of a cartilaginous template, replacement of template with immature (spongy) bone, and finally remodeling into mature bone, with the periosteum being the main driver [33]. While others have reported a mandibular PMBR onset of ∼14 days [34], our group reported 21 days and a bi-mandibular regenerative span of ∼8 cm in length. Our group also reported that mandibular PMBR can be categorized as “complete”, that is, regenerated bone occupying >50% of the radiographic defect, or “incomplete” which is irregular in outline and occupying <50% of the radiographic defect [32]. While there have been numerous research studies on bone healing and repair, the definitive role of periosteum and the molecular pathways that regulate bone regeneration in mammals remain largely unknown [35]. As an important first step, a review of PMBR genetic determinant would be beneficial in this quest to unraveling this and an analysis of gene expression that comprises RNA transcription and protein translation, which provides an insight into normal cellular processes, would be considered accurate for interrogating genetic association of this event [36,37]. The techniques used for evaluating gene expressions include DNA arrays and realtime polymerase chain reaction for RNA expression and Western blotting and gel electrophoresis for protein expression [36,37].

The aim of this study is to review the literature on the clinical evidence of mandibular PMBR and further review its molecular mechanism through gene expression studies.

2. Methodology

Two searches were conducted, the first (search 1) focusing on the clinical evidence of mandibular PMBR and the second (search 2) being on the molecular mechanism of craniofacial PMBR.

2.1. Search Strategy

The methodology was modeled after the PRISMA (Preferred Reporting Items for Systematic Review and MetaAnalysis) guidelines [38]. Cochrane style MeSH terms and keywords were used for the initial searches. Search 1 terms comprised mandible, spontaneous bone regeneration, periosteum-induced bone regeneration, and periosteum, mandible, and (or) bone regeneration. Search 2 terms comprised craniofacial and (or) bone regeneration, periosteum-induced bone regeneration, gene expression studies, and techniques, that is, DNA array and real-time polymerase chain reaction. The search tools such as PubMed, Ovid Medline, and Web of science were used.

2.2. Eligibility Criteria

English-only publications were reviewed and studies selected were cross-sectional studies, case-controlled studies, and controlled clinical trials. Publication titles and abstracts were initially reviewed to identify articles suitable, and selected manuscripts were proofread by the author. Additional publications from the reference list of this initial search were retrieved and reviewed to identify potential papers that met these study criteria. All studies for search 1 must have bony continuity defect while those for search 2 must use proprietary named kits for gene expression studies according to manufacturer’s protocol. In addition, purely clinical and pathology-related studies were included for search 1, while clinical, pathology-related, conference proceedings, and abstract-only articles were excluded for search 2.

2.3. Quality Appraisal

The quality assessment of each publication was undertaken through pre-existing assessment tools or their modifications as written below:

Search 1. Since these would be majorly observational studies, that is, case reports and series, the current quality assessment tools provide limited assessment of these types of studies [39].

Search 2. The quality assessment of each publication was undertaken through a modification of a pre-existing tool: STrengthening the REporting of Genetic Association studies (STREGA) [40]. Assessment would be achieved by evaluating in each article the following: characterization of a periosteum-mediated craniofacial bone regeneration, description of case and control screening population, inclusion of gene or variant ID (eg, NCBI rs identification, evaluation of gene function, or gene ontology identifiers), and measure of genetic association. In addition, the difference in gene expressions between 7 and 15 days was also reviewed and this was arrived at based on the scientific report that, the first phase (inflammatory response and recruitment of osteoprogenitor cells) of bone induction during bone repair and healing occurs in the first week [33].

2.4. Data Extraction and Analysis

All results were extracted onto a data form for tabulation. SPSS 27 software package (IBM Company, Armonk, NY, USA) was used for statistical analysis. Sample size, age range, and median of subjects with paired sample data during expression analysis were captured for descriptive statistical information. Ages of participants for search 1 were categorized as children—between 0 and 11 years—pre-adolescence—between 12 and 18 years—and adult >18 years [41]. Where microarray was used in search 2 for differential gene expressions, a fold change threshold according to manufacturer’s protocol was used to consider a gene expression as significant.

3. Results

3.1. Search 1

The electronic search yielded 101 citations, of which 68 publications were removed for lack of correlation to search objectives, that is, 39 were unrelated clinical and pathological-based studies, 15 were observational studies, and 14 were conference proceedings; thus, 33 [31,32,34,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71] publications were selected for full-text review and all met the criteria for study inclusion (Chart 1). The 33 studies comprised 8 case series, and 25 case reports all utilizing 65 participants (

Table 1. Search 1 Publications With Clinical Characteristics.

). Geographically, 24% (8/ 33) of all studies utilizing 43% (28/65) of cases were reports originating from Africa, [32,42,50,58,60,62,63,65] an observation probably conditioned by the limited facilities mitigating against primary or vascularized reconstruction. 61% (36/59) of cases were males while 39% (23/59) were females and 6 cases did not specify gender. Ages were categorized as 37% (22/59) children, 30% (18/59) preadolescence, and 32% (19/59) adults. Concerning PMBR indications, 74% (44/59) were benign pathologies (Figure 1 and Figure 2), 8% (5/59) were malignant pathologies, 7% (4/59) were high-velocity traumas, 4% (2/59) were TMJ ankylosis, and 2% (1/65) were congenital condylar atresia, medication-induced osteonecrosis, and osteomyelitis while 12% (7/65) were unspecified. Concerning time of onset of PMBR, a range of between 2 and 128 weeks was recorded; however, most of the study observations were based on follow-ups compared to periodical monitoring, thus a median of the data would be an accurate characterization of this event. A median of 3 weeks for PMBR onset was recorded in our earlier study[32] based on periodic radiographic monitoring; thus, a period of <2 weeks up or down this threshold period was considered an appropriate period for PMBR onset (Figure 3). No correlation was observed in histology type and immobilization. There was also no correlation to defect span and complete or incomplete PMBR.

3.2. Search 2

The electronic search yielded 69 citations, of which 57 publications were removed for lack of correlation to search objectives, that is, 38 descriptive studies. 9 conference proceedings, 8 observational (clinical) studies, and 2 papers for study duplicity; thus, 12 publications were selected for full-text review. Upon further review, eight additional studies were removed for being purely clinical (observational) studies and having no gene expression correlation, resulting in four studies [72,73,74,75] being considered for inclusion (Chart 2). All four studies used nonhuman subjects, comprising twelve Wistar rats, six Mini pigs, and three Calves’ calvaria transplanted on mice. In summary, the result analysis showed an upregulation of genes responsible for angiogenesis, cytokine activities, immune–inflammatory response, and skeletal development. Cytokine and immune–inflammatory response genes like cytokines (IL-1 beta and IL-6), cathepsin-K (CTSK), and receptor activator of nuclear factor kappa-B ligand (RANKL) were mostly implicated at day 7 while osteoblastspecific genes like runt-related transcription factor 2 (Runx2), collagen type 1, osteonectin, and osteocalcin/osteopontin genes were implicated at day 15 (

Table 2. Search 2 Publications With Specimens and Percentage Genes Expression at Weeks 1 and 2.

). Several biological signaling pathways like BMP-2, Hedgehog, PDGF, Notch, and Wnt were also implicated at day 15.

4. Discussion

While bone regeneration is believed to be periosteal driven, there remains little understanding on the precise role played by the periosteum [76,77]. It is the position of this paper that an improved understanding of this role might proffer countless benefits to bony defect reconstruction with advantages like low cost and reduced morbidity from a non-existent donor site, with its unpredictability and lack of consensus amongst clinicians, remaining its only drawback. The number of studies that met this study’s inclusion criteria suggests the dearth of publications on this topic, with search 2 revealing no report on the underlying molecular mechanism of periosteum in bone regeneration amongst humans. Empirically, three conditions must be met for bone induction to take place: presence of osteoprogenitor cells, an inducing agent, and an environment permissive to osteogenesis [78]. The periosteum acts as an osteogenic progenitor cell and provides vascular supply for the newly formed bone while its intact state is thought to provide a barrier to infiltration of granulation tissue, thus creating a permissive environment for osteogenesis [52,79]. Infection and residual bone segments have been suggested as the inducer agent for mesenchymal cells in the surrounding tissues [42,50], but these are in converse to reports of zero prophylactic antibiotics used during PMBR cases and PMBR of the entire mandible [32,60].

A recurring decimal in most of the reports reviewed was the relatively young periosteum, that is, less than 18 years in age, which constituted 67% of the PMBR cases while 18 years and above constituted 33% of cases. This was always considered a pre-condition for PMBR with the explanation being their osteogenically active periosteum [31]. While some researchers have carried out subperiosteal dissection (mucoperiosteal flap without damage to the periosteum) for adults (>18 years) and reported no PMBR, few have done same and reported PMBR [43,48,65], thus suggesting the ability of a young periosteum to retain its osteocompetency with increasing age. However, childhood/preadolescence continues to be a key predisposing factor in promoting PMBR. Concerning PMBR onset, most of the studies used post-operative radiography for follow-ups rather than monitoring for PMBR. However, for the few that recorded early radiographic images [31,32,34,45,46,47,62,64,67], the PMBR onset ranged from 2 to 8 weeks with a median of 5 weeks. However, this study suggests radiographic monitoring from as early as 2 weeks especially for children <12 years. The use of PMBR for treating malignant cases was another aspect recorded as this study surprisingly recorded 5 cases of malignant cases with all being sarcomas including one post-irradiated case [51,66,67,69,71]. All cases were early-stage sarcomas and were all intra-bony which allowed for subperiosteal dissection. However, the author suggests that the use of PMBR for malignant cases should be exercised with extreme caution.

With little or no genetic association studies on PMBR in humans, one of the four selected animal model studies recorded no gene ontology identifier [75], thereby negating the quality assessment of this publication. However, the skeleto-developmental and immune–inflammatory responses were the major events implicated in bone regeneration according to this review, and it further suggested that the inflammatory–immune response was unique and different from that of the usual soft tissue wound healing [80,81,82]. The upregulation of inflammatory and immune response was not unexpected at 7 days as this supports the initial inflammatory phase of bone induction, but the added responses of I-kB kinase/NF-kB signaling pathway were interesting as this is a known pathway reported to induce inflammation-induced bone loss [83]. In week 2, the upregulated pathways such as TGF beta/Bmp, Wnt, and Notch are known to upregulate skeletal developmental gene expression. While these pathways are wellknown skeletal development pathways, Bmp signaling is the only pathway suspected to be associated with bone regeneration while BMP2 is suggested to regulate PMBR through periosteum-based target cells [84,85]. In addition, Wnt signaling pathway was another upregulation response noted at week 2; however unlike the Bmp target cells, the Wnt target cells are resident in fractured bones and despite showing similarity to PMBR event, this study suggests that the Wnt pathway might influence more of bone repair than regeneration albeit in a different mechanism from the Bmp signaling [86,87].

In conclusion, this review tried to harmonize the various reports on the complex events of PMBR of the mandible through clinical evidence and gene expression studies. The results suggest that PMBR has a higher probability when young periosteum in children and pre-adolescence are considered than adults and periodic monitoring should commence as early as 2 weeks post-surgery. The immune– inflammatory genes appear to regulate inflammatory response early while the skeletal developmental genes and related pathways characterize the gene expression patterns of PMBR much latter in the event. Being deliberate in anticipating PMBR during secondary pediatric reconstruction would be beneficial. The limitation of the study would include the observational nature of the clinical evidence on PMBR.