Dental pulp stem cells for reconstructing bone defects: A systematic review and meta-analysis

Background. Bone reconstruction with appropriate quality and quantity for dental implant replacement in the alveolar ridge is a challenge in dentistry. As dental pulp stem cells (DPSCs) could be a new perspective in bone regeneration in the future, this study investigated the bone regeneration process by DPSCs. Methods. Electronic searches for articles in the PubMed, EMBASE, and Scopus databases were completed until 21 April 2022. The most important inclusion criteria for selecting in vivo studies reporting quantitative data based on new bone volume and new bone area. The quality assessment was performed based on Cochrane’s checklist. Results. After the title, abstract, and full-text screening of 762 studies, 23 studies were included. A meta-analysis of 70 studies that reported bone regeneration based on new bone area showed a statistically significant favorable influence on bone tissue regeneration compared to the control groups (P<0.00001, standardized mean difference [SMD]=2.40, 95% CI: 1.55‒3.26; I2=83%). Also, the meta-analysis of 14 studies that reported new bone regeneration based on bone volume showed a statistically significant favorable influence on bone tissue regeneration compared to the control groups (P=0.0003, SMD=1.85, 95% CI: 0.85‒2.85; I2=84%). Conclusion. This systematic review indicated that DPSCs in tissue regeneration therapy significantly affected bone tissue complex regeneration. However, more and less diverse preclinical studies will enable more powerful meta-analyses in the future.


Introduction
Reconstruction of bone defects is often a clinical challenge, especially in dentistry. Maxillofacial bone deficiencies result from tooth loss, periodontitis, trauma, tumor removal, congenital anomalies, and radiationrelated osteonecrosis. Periodontitis and ridge remodeling following tooth loss is the most common cause of the alveolar bone defect. Successful implant placement requires adequate bone quality and quantity to avoid implant failure; therefore, reconstructing the alveolar ridge is a substantial issue for dental implant-supported prostheses. [1][2][3] Autogenous bone grafting is the gold standard for bone regeneration. However, mitigating the complications associated with the harvest of autologous bone was the primary impetus for developing bone graft substitutes. 4 Reconstructing bone defects with tissue engineering using dental pulp stem cells or DPSC is one of the most modern rehabilitation methods that can revolutionize future treatments. 5 DPSCs can include self-renewal capacity, multilineage differentiation capacity, high proliferation potential, and clonogenic efficacy. These features have made them the most promising mesenchymal stem cells (MSCs) for clinical purposes. However, many issues and challenges must be addressed before using these cells in clinical treatment. 6,7 Tissue engineering scaffolds can facilitate the proliferation and differentiation of progenitor cells. Combining osteogenic cells, osteogenic factors, biocompatible scaffolds, and angiogenesis are the elements of bone tissue engineering. Treatment with bone-related factors, gene transfection, and gene overexpression enhances the bone regeneration potential of DPSCs. 5,8 Due to the limited clinical trials conducted in the field of bone regeneration by DPSCs, they have not yet been effectively used in clinical treatments. Further investigation of the studies conducted in this field will lead to achieving suitable study designs and, ultimately, the progress of therapy using DPSCs. Since the quantitative evaluation of bone regeneration by DPSCs has not been carefully evaluated in the previous systematic reviews, this study aimed to evaluate the potential of DPSCs in clinical and preclinical bone regeneration from a quantitative point of view. For this purpose, this review study analyzed the amount of bone volume and bone area regenerated by DPSCs.

Protocol
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement 9 was the protocol of this systematic review.

Focus questions
The objective of this study was to review the literature to answer the focused question systematically: Do DPSCs improve the quantitative results of bone regeneration?
Which samples, scaffold type, final follow-up, and defect type significantly impact bone regeneration?

Eligibility criteria
The inclusion criteria for the selection were: In vivo studies, bone defect regeneration therapy utilizing DPSCs, studies reporting quantitative data in the form of new bone volume percentage or new bone area percentage, and studies published in the English language.
The exclusion criteria for the selection were: Studies reporting only qualitative results of bone regeneration.
In studies where the results of bone regeneration had been reported quantitatively, the exclusion criteria were these items: Standard deviations were not apparent, the numbers of samples were not reported, the created bone defects were not filled by scaffolds seeded with DPSCs in the test group, and an acellular scaffold did not fill the created bone defects in the control group.

Information sources
Electronic searches were completed for articles in MEDLINE (PubMed), EMBASE, and Scopus databases until 21 April 2022. Also, related systematic references were added.

Selection process
The studies were evaluated by two reviewers separately (NM and FO), and the third reviewer (FE) reviewed the differences.

Data collection process
Two reviewers (NM and FO) collected data from each report independently.

Data items
The list and definition of outcomes are as follows: "Author-year" specifies the author and year of publication.
"Sample" specifies the type of animals with bone defects. "Number" specifies the total number of test and control group samples.
"Site and size of bone defects" specifies the type of the created bone defects and their dimensions or the dimensions of the bur used.
"Final follow-up" specifies the final duration of treatment by week.
"Laboratory method" specifies the laboratory method. "Scaffold" specifies the scaffold used. "Regenerated bone area" specifies the study outcome based on bone area percentage. The results of the test and control groups for each study are stated in this column. In the test group, the defect was filled by scaffolds seeded with DPSCs, and in the control group, the defect was supplied with an acellular scaffold.
"Regenerated bone volume" specifies the study outcome based on bone volume percentage. The results of the test and control groups for each study are stated in this column. In the test group, the defect was filled by scaffolds seeded with DPSCs, and in the control group, the defect was supplied by an acellular scaffold.

Study risk of bias assessment
Cochrane's risk of bias tool was used to assess the risk of bias in the included studies. 10 The criteria used were as follows: random sequence generation (selection bias), allocation concealment (selection bias), blinding of personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other sources of bias.
The classification of studies based on seven criteria of risk of bias assessment was as follows: A study had a low risk of bias if it had none of the types of preferences, a study had a moderate risk of bias if it had one of the types of bigotry, and a study had a high risk of bias if it has more than one type of bias.

Statistical analysis
Review Manager (RevMan, Computer program, Version 5.4, The Cochrane Collaboration, 2020) was used for statistical analysis. Separate meta-analyses were performed according to regenerated bone volume and area. In addition, subgroup analyses were performed according to the sample, defect, follow-up, and scaffold types.
Meta-analyses were performed using the Z test with random effects weighted inverse variance method. The effect size was measured using standardized mean differences (SMDs) and 95% confidence intervals. SMD < 0.2 was considered a 'small' effect size, SMD between 0.2 and 0.8 represented a 'medium' effect size, and SMD > 0.8 was considered a 'large' effect size. The results were considered significant when P < 0.05. The heterogeneity was assessed using the I 2 test. If I 2 was > 75%, it was interpreted as highly heterogeneous.

Study selection
A total of 762 records were identified through database searches. After removing duplicate articles, in vitro articles, review articles, and articles that did not investigate bone tissue regeneration, the full texts of 78 studies were reviewed. The reasons for excluding studies after a fulltext assessment were as follows: The quantitative data were not reported (n = 27)  ; the percentage of new bone volume or new bone area was not reported (n = 9) 38-46 ; the number of the samples was not reported (n = 2) 47,48 ; stem cells from human exfoliated deciduous teeth were used as MSCs (n = 2) 49,50 ; the created bone defects in the test group were not filled by scaffolds seeded with DPSCs (n = 10) [51][52][53][54][55][56][57][58][59] ; and the created bone defects in the control group were not filled by an acellular staging (n = 5). [60][61][62][63][64] Finally, 23 studies were included in the meta-analysis.   Figure 1 shows the flow chart.

Risk of bias in studies
In this category, 2, 8, and 13 articles showed a low, medium, and high risk of bias, respectively. Figures 2 and 3 show reviewing authors' judgments about each risk of bias item presented.
There were no significant differences in final followup and sample types. The most positive impact on bone regeneration occurred in groups where the final followup was six weeks, and the sheep were used as samples (Figures 8 and 9). The outcome of sample types subgroup analysis without considering the subgroups including less than two studies was as follows: The biggest SMD occurred in the rat subgroup (SMD = 2.19 [95% CI: 1.05-3.33], studies = 6), and the smallest SMD occurred in the rabbit subgroup (SMD = 0.97 [95% CI: 0.10-1.84], studies = 5) ( Figure 6).
Subgroup analyses showed a significant difference in bone regeneration with all the subgroups. However, all   the subgroups had only one to two studies, indicating low statistical power. In addition, subgroup analyses showed high heterogeneity (Figures 12, 13, 14, and 15). Subgroup analyses showed that the most positive impact on bone regeneration occurred in groups where bone defects were created in the alveolar bone, and the pigs were used as samples (Figures 12 and 13). More precisely, the outcomes of subgroup analyses without considering subgroups including less than two studies were as follows: The SMD in the alveolar bone defect subgroup (SMD = 8.48 [95% CI: -4.03-20.96], studies = 2) was higher than that in the calvarium bone defect subgroup (SMD = 1.58 [95% CI: 0.47-2.69], studies = 10) ( Figure 13). In the sample type subgroups, the SMDs in the rat, rabbit, and mice subgroups The most significant positive impact on bone regeneration occurred in groups where the final follow-up was 12 weeks, and hydroxyapatite/tricalcium phosphate (HA-TCP) was used as a scaffold (Figures 14 and 15).

Reporting biases
There was a possibility of bias due to the small number of studies. Also, the funnel plots of the new bone area and new bone volume indicated an asymmetrical shape. The asymmetrical shape might have been caused by publication bias, study heterogeneity, and methodological anomaly ( Figures 5 and 11).

Discussion
Bone tissue engineering by DPSCs has been the subject of many studies as a method that could have a promising future in alveolar ridge reconstruction. However, despite many advances in this field, the high heterogeneity of studies and the few studies with complete statistical data make high-power statistical analysis impossible and the clinical application and effectiveness of stem cell utilization unclear.
This meta-analysis evaluated the impact of tissue engineering by DPSCs on bone regeneration based on the new bone volume and new bone area formation. Two previous systematic reviews 6,88 mentioned a positive impact of tissue engineering by DPSCs on bone regeneration based on qualitative data. The present study is the first to evaluate the effect of tissue engineering by DPSCs on bone regeneration based on quantitative data.
According to this review, DPSCs and scaffold complexes significantly increase bone regeneration. Clinical diversity and high methodological heterogeneity should be considered in the interpretation of the metaanalysis. Analyses were performed in the subgroups of sample type, scaffold type, final follow-up and defect types. Although heterogeneity decreased in the majority of subgroup analyses, a few studies in subgroups caused the low statistical power of meta-analysis. Nonetheless, In addition, it is necessary to pay attention to this point that this meta-analysis has not reviewed the impact of other factors, such as growth factors, on bone regeneration. Bone-related factors, gene transfection, and gene overexpression enhance the bone regeneration potential of DPSCs. 5,8 However, meta-analysis was impossible due to the high heterogeneity of methodology in the studies examining these factors' impact on bone tissue engineering. Therefore, we can expect a more significant amount of bone regeneration by DPSCs with the application of growth factors, gene transfection, and gene overexpression.
Compared to other stem cells, studies comparing the ability of DPSCs to bone marrow MSCs did not show a significant difference in bone regeneration. 47,50,66,89 However, evaluation of the osteogenic potential of adipose tissue-derived stem cells 67 and amniotic fluid stem cells 76 showed a significant increase in bone regeneration compared to DPSCs. Overall, using DPSCs with appropriate scaffold, growth factor, and gene therapy will result in the maximum bone regeneration percentage. Finally, as mentioned in previous systematic reviews, bone tissue engineering can be expected to result in a favorable clinical outcome. 6,88 Few clinical studies examined bone reconstruction in Mansfield. [38][39][40]43,90,91 Most of these studies have reported new bone regeneration based on probing depth and clinical attachment loss. Future clinical trials should also evaluate the extent of bone regeneration in other ways, such as micro-computed tomography.
Future research should concentrate on humans or samples closer to humans, such as dogs and sheep, than on mice and rats. The results should be in the form of statistical data such as bone volume, trabecular number, bone mineral density, and mineral content.
The new bone formation could include maxillary or mandibular bone defects rather than cranium or subcutaneous ones. Future research should compare the effect of different growth factors, scaffolds, and gene overexpression on bone regeneration.

Conclusion
Bone tissue engineering by DPSC is one of the promising ways for bone regeneration in the future. This study was designed in response to the question of whether the current clinical studies quantitatively indicate the ability of DPSC to regenerate bone properly. In this review article, the meta-analysis conducted on the results of the studies showed a significant increase in the amount