Introduction
After tooth extraction a cascade of biological events are triggered that typically result in significant local anatomical changes [
1]. Several studies have demonstrated that volume loss after tooth extraction is a natural but irreversible consequence, involving both horizontal and vertical dimension loss, and is most pronounced on the buccal side [
2‐
5]. Without intervention, in the first year the width of the alveolar ridge can be reduced by up to 50% [
2,
5,
6].
The hard- and soft-tissue morphology at the extraction site and adjacent teeth determine the course of dental implant placement [
5]. Extraction defects are classified according to several grading systems. The severity of the defect is usually categorized by the extent of the buccal bony defect [
7,
8], which is the most decisive factor in implant placement. In cases of severe buccal bone loss, alveolar ridge preservation (ARP) might lessen the need for staged surgical rehabilitation. Although alveolar ridge preservation procedures have been used since 1998 there are still debates about its effectiveness [
9,
10]. Using ARP, the horizontal and vertical resorption may be reduced by 16–40% [
6,
11]. A statistically significant difference can be found between ARP and unassisted healing. However, the clinical significance of it is still unclear [
9,
12]. Several techniques and bone grafting materials were advocated for ARP. However, none could fully accomplish the required expectations [
13,
14].
For ARP, either particulate or non-particulate graft materials can be utilized. Non-particulate graft materials can complete remodeling but have lower space maintenance properties. The advantage of particulate graft materials is their ease of use and their space-maintaining effect. Using xenografts with a prolonged resorption time has significantly improved alveolar ridge preservation [
15]. However, at the time of reentry (at implant placement), graft remnants are frequently detected, potentially interfering with autogenous bone formation and osseointegration of the implant. Some authors reported that none of the graft materials could show higher percentage of newly formed bone proportion than unassisted healing alone [
16].
In the 1960s, dentin was evaluated as a biomaterial for inducing bone formation. Bone formation was induced at the tooth extraction sockets and muscles, but only after 8–12 weeks [
17]. Since then, numerous preclinical studies have evaluated the biological properties and effects of autogenous tooth bone grafts [
18‐
20]. However, the first human clinical use was only documented in 2010 [
21]. The idea was based on the anatomical observation that the embryonic origin of dentin is the same as that of alveolar bone, which may explain its bone-forming capacity [
22]. Human dentin and bone are composed of 65% inorganic and 35% organic substances [
23]. The inorganic proportion promotes osteoconductivity and space maintenance [
24]. On the other hand, the organic matrix of mineralized dentin is responsible for the osteoinductive property [
25,
26].
Several protocols have been proposed to produce autogenous tooth bone grafts from extracted teeth, which commonly involve the removal of soft tissues, carious lesions, and fillings after tooth extraction [
27]. In addition, some protocols describe the use of only the root of the removed tooth [
28], while others recommend using both the crown and the root [
29].
According to the degree of demineralization, three main graft types can be distinguished: undemineralized dentin matrix (UDDM), partially demineralized dentin matrix (PDDM), and demineralized dentin matrix (DDM) [
30,
31]. Differences between the graft materials and their effect on the healing processes are still under investigation.
Since the first clinical application of ATB, several clinical trials have revealed its potential benefits for ARP. However, clinical studies were conducted with small sample sizes. Therefore, conclusions rely on weak evidence, including high levels of uncertainty. In addition, no meta-analysis has been conducted to confirm the effectiveness of ATB on the preservation of alveolar ridge width; there is also a lack of histological information on its graft remodeling capacity.
This systematic review and meta-analysis aimed to evaluate the current evidence on ATB's efficacy for ARP, the graft turnover capacity, and the effect of utilizing dentin alone vs. dentin combined with enamel to produce ATB.
Materials and methods
We report our systematic review and meta-analysis based on the recommendation of the PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) 2020 guideline (see Additional file
1: Appendix Table 1), while we followed the Cochrane Handbook. Furthermore, the study protocol was registered on PROSPERO (International prospective register of systematic reviews; registration number CRD42021287890). We made minor deviations compared to the registered study, however it has no effect on the reported data. The program used for the analysis was changed for easier visualization.
Eligibility criteria
We used the PICO framework to formulate our research question. We included studies reporting on (P) patients undergoing ARP with (I) particulate ATB graft. The primary outcome (O) was the ridge width change, measured in millimetres (mm). Regarding the change, we radiographically compared (C) the baseline alveolar ridge width to alveolar ridge width 4–6 months postoperatively. Due to the heterogeneity of the measurement methods vertical dimensions of the alveolar ridge could not be investigated. The secondary outcomes were the histological results: the proportion of residual graft, newly formed bone, and connective tissue. We included case series, randomized and non-randomized clinical trials, in which ATB was used in either arm. We excluded literature and systematic reviews and case studies.
Eligible studies included patients over 18 years old with good oral hygiene. ARP was performed with any powder type of autogenous tooth bone graft application with or without membrane coverage, with minimally 3 months of healing. We excluded studies including patients (1) with uncontrolled systemic or infectious diseases, (2) undergoing previous radiotherapy, (3) current or previous bisphosphonate therapy, or (4) heavy smokers (> 5 cigarettes/ day). Studies without linear alveolar ridge width measurement on CBCT or without histomorphometric measurements, immediate implant placement, and those with incomplete data were also excluded.
A systematic literature search was conducted in Cochrane Central Register of Controlled Trials (CENTRAL), Embase, MEDLINE (via PubMed), and Scopus for studies published from inception to 31 November 2021. The search key attached to the supplementary material was applied (Additional file
1: Appendix Document 1). The literature search was limited to articles in English.
Furthermore, scanning the bibliographies of all publications selected for our review for inclusion and also the search of the gray literature (expert contact) were accomplished for potentially relevant articles.
Selection process
Duplicate removal of yielded articles was performed by EndNote X9 (Clarivate Analytics, Philadelphia, PA, USA). Two independent researchers (ES, ESz) followed the Cochrane Handbook’s recommendation and simultaneously screened the titles, abstracts, and full texts of the included studies based on predetermined criteria. The degree of agreement between the review authors was measured using Cohen’s kappa. In case of any disagreement, a consensus was reached after discussion with a third author (BM).
Data collection process and data items
Data were extracted from the included articles into a pre-defined Excel sheet (Office 365, Microsoft, Redmond, WA, USA) by two authors (ES, ESz) independently. A third party (BM) settled any discrepancies. The following data were collected from each study: first author, article title, study design, demineralization process methods, additional material used (membrane), processing method (root part of tooth vs. whole tooth), measurement method of the preoperative defect morphology.
Primary outcome: mean horizontal ridge width preoperatively and postoperatively, or ridge width changes were measured in mm using CBCT. Secondary outcomes: the proportion of residual graft, newly formed bone and connective tissue in the histological sample expressed in percentage and patient follow-up period.
Study risk of bias assessment and quality of evidence
Based on the recommendations of the Cochrane Prognosis Methods Group, the ROB-2 (Risk of Bias assessment tool) was used for randomized control trials, and the ROBINS-I (Risk Of Bias In Non-randomized Studies—of Interventions) for non-randomized clinical trials. The methodological quality of the included studies was assessed separately by two authors (ES, ESz). Any disagreement was resolved by arbitration by a third reviewer (BM).
For each analyzed outcome, the certainty of evidence (certainty in the estimates of effect) was determined with the GRADE approach [
32].
Effect measures and statistical analysis
All statistical analyses were made with a preset alpha value of 0.05 using the R (R Core Team 2022; v4.1.1) software and its
meta (Schwarzer 202; v5.2.0) package. The detailed statistical analysis is presented in the supplementary material (Additional file
1: Appendix Document 2).
We calculated means and mean differences (MDs) with 95% confidence intervals (CIs) from the means and the mean changes of the alveolar crest width and from the histological parameters.
Subgroup analysis
For the primary outcome a subgroup analysis was conducted according the linear measurement level of the alveolar crest width (subgroup
crest and
1 mm apically from crest) (Additional file
1: Appendix Table 3).
For secondary outcomes a subgroup analysis was performed according to the ATB processing methods (Additional file
1: Appendix Table 4). DDM, PDDM and UDDM groups were defined according to the degree of ATB demineralization.
Another subgroup analysis was conducted according to the composition of ATB. The
root subgroup which originates only from the root portion of the tooth, is composed only of dentin, while the
whole subgroup which originates from both the root and crown portions of the tooth, is composed of both dentin and enamel (Additional file
1: Appendix Figs. 6–9).
Certainty of evidence and additional analyses
For each meta-analysis, the certainty of evidence (certainty in the estimates of effect) was determined with the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach (GRADEpro, 2021) [
33]. However, due to the low number of studies, precise outlier and influence analyses could not be carried out. Therefore, we visually inspected funnel plots (Additional file
1: Appendix Figs. 2–5).
Discussion
Alveolar ridge resorption is an inevitable process, although the extent of tissue breakdown is reducible with appropriate interventions [
41,
42]. Several approaches for ARP are available using different graft materials, but none of them deliver ideal outcomes. Since the first use of ARP techniques the popularity of it is still arising. Around 29% of the procedures using grafting materials are ARP surgeries, and it tends to increase approximately 11.4% per year [
9]. The cost and also the enviromental footprint of the used graft materials are growing proportionally. Most recently, ATB has been considered an autologous, cost effective, sustainable alternative because it is easily retrievable, safe, and has minimal risk of rejection or infection [
43].
All recent systematic reviews on the topic [
27,
30,
44,
45] summarized that ATB has a beneficial effect on alveolar ridge preservation, but to the best of our knowledge none of them could statistically demonstrate these findings. Our meta-analysis aimed to statistically confirm this observation regarding changes in alveolar ridge width and histological outcome. Unfortunately, vertical dimensional changes could not be analyzed in this study due to the large differences in measurement methods.
The differences in initial defect morphology, surgical techniques, preparation procedures of ATB, and follow-up time made comparisons of primary studies difficult. Nevertheless, some important findings were made.
In a recent meta-analysis, the change in alveolar ridge width using the xenograft Bio-Oss® material was -0.88 mm [
16]. Our data provided similar results (-0.72 mm), suggesting that ATB is as effective in preserving alveolar ridge width as the most studied particulate graft material, although RCTs are needed to directly compare the two materials statistically. An increased heterogeneity is observed, caused by different measurement methods and patient populations with different initial defect morphologies. For example Joshi et al. included extraction defects with four walls, which might have better healing potential than extraction defects with fewer bony walls.
Due to the heterogeneity of the measurement techniques, the present MA cannot compare the reduction in socket dimensional changes following ARP using ATB with extraction alone. However, Del Canto-Díaz et al. found a significant ridge preservation effect in their split-mouth study, with a mean bone loss of vestibular width of 0.46 mm in the ARP group using ATB compared with 1.91 mm in the unprovided extraction sockets group measured from the vertical line to the buccal cortical bone at 1 mm apical from the alveolar crest.
Other hot issues are the rate of graft material resorption and whether intra-socket grafts compromise the normal healing process of the tooth extraction socket. Regarding the resorption rate of ATB graft material, it can be said that ATB may also has greater graft remodelling capacity compared to other particulate graft materials. De Risi et al. in a meta-analysis compared different grafting materials for ARP and concluded that the newly formed bone proportion of xenografts was 23%, the residual graft proportion was 37%, and the connective tissue proportion was 32% [
46]. According to our findings, the mean newly formed bone proportion of ATB was 40%, the residual graft proportion was 12% and the connective tissue proportion was 45%.
Due to the low number of studies, a conclusion regarding the efficacy of the processing methods resulting in different levels of ATB mineralization (UDDM, PDDDM, DDM) cannot be drawn, but a slight difference between the outcomes can be observed.
Our meta-analysis shows that the newly formed bone proportion was highest in the PDDM group (51%) and lowest in the DDM group (31%) –a statistically and clinically significant difference. We also found that the proportion of connective tissue was the lowest in the PDDM group (39%) and highest in the DDM group (51%). However, this difference was not significant. These data suggest that partial demineralization may positively affect the rate of new bone formation. This is likely due to increased osteoinductivity resulting from the more exposed collagenous and non-collagenous proteins and growth factors, and increased osteoconductivity resulting from the increased porosity and surface area [
43]. However, aggressive demineralization can cause a depletion of growth factors and lead to collapse of the 3D structure [
47]. This correlates well with previously conducted preclinical [
47] and clinical studies [
21,
48‐
50].
Mazor et al. used UDDM for ARP in combination with a non-resorbable membrane. After seven months, 63% of newly formed bone could be detected. Minetti et al. used DDM for ARP and a xenogenic resorbable membrane for the graft coverage. After four months the total bone volume was 41%. The heterogeneity of measurement techniques means that the current MA cannot use these data either. However, considering the differences in the study designs and measurement parameters, this data suggests that the demineralization of the graft material tends to decrease the newly formed bone proportion [
28,
51] and this is in line with our findings.
No statistically significant difference was observed between the subgroups in terms of residual graft remnants. However, the difference in the residual graft proportion between DDM (9.5%), PDDM (9.8%) vs. UDDM groups (14.5%) may be clinically relevant.
According to our analysis, the origin and composition of ATB, i.e., root composed by denin only or root and crown, composed by dentin and enamel, can affect the quantity of newly formed bone. We found a statistically significant difference between the subgroups in favor of the whole tooth group. There was no statistical difference between the subgroups in the graft turnover or connective tissue proportions. The combined application of root and crown also resulted in significantly higher graft volume, increasing the cost/benefit ratio of the treatment.
Strengths and limitations
Previous systematic reviews collected the available literature data but have not been able to analyze the effects of ATB in alveolar ridge preservation [
27,
30,
44,
45,
52]. Regarding the strengths of our analysis, we followed a strict protocol, which was registered in advance. A rigorous methodology was applied. Moreover, the full spectrum of currently available data was analyzed.
However, the results should be interpreted with caution due to the following limitations. The heterogeneity of preparation methods resulting in different ATB composition and mineralization, the difference of initial defect morphology and follow-up times between studies prevents a complete overview of the entire healing process. The presence of a moderate risk of bias in some of the domains is another limitation.
Clinical and research implications
Based on our results, we suggest more detailed inclusion criteria, randomization procedures, standardized dentin matrix processing methods, and a pre-specified analysis plan. Due to its cost effectiveness and sustainability ATB can be an alternative graft material for ARP.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.