Clinical Reference Strategy for the Selection of Treatment Materials for Maxillofacial Bone Transplantation: A Systematic Review and Network Meta-Analysis

Abstract

Bone graft materials have mixed effects of bone repair in the field of oral maxillofacial surgery. The qualitative analyses performed by previous studies imply that autogenous odontogenic materials and autogenous bone have similar effects on bone repair in clinical jaw bone transplantation. This retrospective systematic assessment and network meta-analysis aimed to analyze the best effect of clinical application of autogenous odontogenic materials and autogenous, allogeneic, and xenogeneic bone grafts in bone defect repair. A systematic review was performed by searching the PubMed, Cochrane Library, and other journal databases using selected keywords and Medical Subject Headings search terms. 10 Papers (n = 466) that met the inclusion criteria were selected. The assessment of heterogeneity did not reveal any overall statistical difference or heterogeneity (P = 0.051 > 0.05), whereas the comparison between autogenous and allogeneic bone grafts revealed local heterogeneity (P = 0.071 < 0.1). Risk of bias revealed nine unclear studies and one high-risk study. The overall consistency was good (P = 0.065 > 0.05), and the local inconsistency test did not reveal any inconsistency. The publication bias was good. The confidence regarding the ranking of bone graft materials after GRADE classification was moderate. The effects on bone repair in the descending order were as follows: autogenous odontogenic materials, xenogeneic bone, autogenous bone, and allogeneic bone. This result indicates that the autogenous odontogenic materials displayed stronger effects on bone repair compared to other bone graft materials. Autogenous odontogenic materials have broad development prospects in oral maxillofacial surgery.

Keywords: Allogeneic bone; Autogenous bone; Autogenous tooth; Network meta-analysis; Xenogeneic bone.

Fig. 1

The PICOS principle

Fig. 2

Systematic literature search and selection process

Fig. 3

Document visualization. Different colors represent different research fields and the size of the circle represents the number of studies. The density of the lines represents the research density between various research directions

Fig. 4

Network diagram of intervention measures included in the meta-analysis. Each circle corresponds to an object of research included in the analysis and the area is proportional to the number of events. Each line represents a direct comparison between the objects of research and the thickness corresponds to the number of comparisons within the direct trials. The color of the line represents the intensity of the risk of bias

Fig. 5

Forest plot. The horizontal line represents the confidence interval of the results. The black dots represent the effect size of individual studies. The gray squares represent the weight of individual studies. The diamond represents results of the combination. The solid vertical line in the middle is the invalid line. SMD, standardized mean difference; CI, confidence interval (A, Autogenous tooth; B, Autogenous bone; C, Allogeneic bone; D, Xenogeneic bone)

Fig. 6

Loop inconsistency. Evaluation of inconsistency using loop-specific heterogeneity estimates. A loop-specific approach was used to investigate consistency within every closed triangular loop in every network, showing the difference between direct and indirect estimates in the loop. IF, inconsistency factor; CI, confidence interval (A, Autogenous tooth; B, Autogenous bone; C, Allogeneic bone; D, Xenogeneic bone)

Fig. 7

Sensitivity analysis. The middle vertical line is the line of 0.20, which is the total combined effect size. Furthermore, the circles corresponding to each experiment represent the combined effect size of the remaining studies to judge the impact of this study on the combined effect size. The name of each experiment is omitted

Fig. 8

Evidence contribution plot. The evidence contribution plot shows the direct, indirect, and mixed processing information regarding the overall network. The size of each square is proportional to the weight and the weight is the influence of the direct summary effect (horizontal axis) on the summary network effect (vertical axis). Each weight is expressed as a percentage

Fig. 9

Study limitations summary graph. Study limitations of pairwise comparisons and global network results. The bar chart depicts the ROB assessment of each direct comparisons and the study limitations of each comparison. The pie chart represents the weight of each direct comparison on the entire network

Fig. 10

Forest plot of pairwise comparison

Fig. 11

Plots of the SUCRA for all treatments in the primary outcome network. A larger area under each line indicates a higher percentage. The percentage represents the efficacy of each intervention; a large percentage indicates a more effective intervention