A systematic review and meta-analysis on the effectiveness of xenograft to prevent periodontal defects after mandibular third molar extraction

Background To evaluate the use of guided bone regeneration with xenograft to prevent periodontal defect in the distal aspect of the second molar after the surgical removal of the mandibular third molar. Material and Methods Three electronic databases (Pubmed, Cochrane Library and Scopus) were searched in April 2020. Randomized clinical trials in non-smokers and healthy patients, with at least six months follow-up, comparing periodontal probing depth, clinical attachment level, alveolar bone level and adverse events were selected by two independent investigators. The risk of bias assessment of the selected studies was evaluated by means of the Cochrane Collaboration’s Tool. Finally, a meta-analysis of the outcomes of interest was performed. Results Despite 795 articles were found in the initial search, only three randomized controlled clinical trials were included. Pooled results favoured the use of the xenograft plus collagen membrane over the spontaneous healing in terms of periodontal probing depth gain (MD=2.36; 95% CI 0.69 to 4.03; P=0.005) and clinical attachment level gain (MD=2.52; 95% CI 0.96 to 4.09; P=0.002). No other statistically significant differences were found. Conclusions Within the limitations of the present review, the xenograft plus collagen membrane exhibited better periodontal results than spontaneous healing without increasing postoperative complications. However, future well-designed studies with larger samples are required to confirm our results. Key words:Third molar, tooth extraction, bone regeneration, xenograft.


Introduction
Extraction of mandibular third molar (M3M) is a very widespread practice in dentistry. Although its indication is clear when provoking symptoms or disease (e.g. infection, non-restorable caries, periodontal disease, root resorption), currently the prophylactic extraction remains a controversial issue (1). Indeed, the decisionmaking for removal of wisdom teeth has been discussed in the literature and some countries such as Finland, France, The United Kingdom or Spain, have made their own clinical practice guidelines, exhibiting discrepancies between them, especially about the prophylactic extraction of the M3M (2)(3)(4). Periodontal disease on the mandibular second molar (M2M) is one of the primary reasons for the treatment of M3M (5). There are predisposing factors associated to the appearance of bony periodontal defects in the distal aspect of the M2M after the surgical removal of the M3M such as patient's age (older than 25 years), position of the wisdom tooth or pre-existing periodontal defect. Knutsson et al. (6) described that mesioangular or horizontal M3M with a large contact with M2M had a greater risk of periodontal postoperative complications. Additionally, other studies have shown that the surgery itself can also cause a residual intrabony defect behind the M2M (7)(8)(9). To prevent periodontal defects after the M3M extraction, various treatment modalities have been suggested, including different flap designs, soft-tissue suturing, and different bone and tissue regeneration techniques. In the context of periodontal regeneration therapy, bone substitutes such as autologous bone, al-lografts, xenografts or alloplastic grafts and occlusive membranes have been broadly studied (10,11). Each material is associated with some advantages and disadvantages, so their selection should depend on the clinical scenario, as well as, the preferences of the clinician and the patient (12). Among these bone substitutes, the xenograft has been widely used in the field of bone reconstruction since it is a safe and a well-documented osteoconductive material with a low resorption rate. Due to its chemical composition and its trabecular structure, the xenograft has proven to be a good scaffold for cell growth, and thus, for bone regeneration (13,14). Despite previous systematic reviews have been published on this topic, none of them compared solely the xenograft to the spontaneous healing. Thus, the aim of the present systematic review and meta-analysis was to gather the published randomized clinical trials to determine whether bone regeneration with xenograft is useful to prevent periodontal defects in the distal aspect of the M2Ms after the surgical extraction of the M3M.

Material and Methods
This systematic review and meta-analysis was conducted in accordance with the statements of "Preferred Reporting Items for Systematic Reviews and Meta-Analyses" (PRISMA) (15). -Eligibility criteria The inclusion criteria were depicted in Table 1. We included articles that met the following eligibility criteria: (P) Population: Non-smokers and healthy patients that underwent a M3M extraction.
Conclusions: Within the limitations of the present review, the xenograft plus collagen membrane exhibited better periodontal results than spontaneous healing without increasing postoperative complications. However, future welldesigned studies with larger samples are required to confirm our results.  characteristics, surgical interventions, postoperative follow-up and the outcomes. Finally, we contacted with the authors of the selected studies for clarification when data were missing or incomplete.
-Risk of bias assessment Two independent reviewers (V.R-R. and J.T-S.) evaluated the risk of bias of each article by means of "Cochrane Handbook for systematic reviews of interventions, version 5.1.0" (16). We evaluated as low, unclear or high risk of bias the following six quality criteria: random sequence generation, allocation concealment, patient blinding, outcome blinding, incomplete outcome data and selective reporting. Finally, a third independent reviewer (MÁ.S-G.) resolved any disagreement during this step.
-Statistical analysis Odds ratio (OR) with 95% confidence intervals (CI) was used for adverse events outcome. In order to estimate the size of the effect, mean difference (MD) and standard deviation (SD) were used for PPD, CAL and ABL. A pairwise meta-analysis was conducted using RevMan software (Review Manager version 5.3; The Cochrane Collaboration, Copenhagen, Denmark) using M3M as the statistical unit in split-mouth studies. We selected the random effect model due to methodological and clinical heterogeneity expected across the included studies (17). In addition, significant heterogeneity was interpreted when I2 value was >50 (18). Statistical significance was defined as P < 0.05 for all analyses.

-Study selection and description
The initial electronic and manual search rendered 795 references. After the removal of the duplicates and the irrelevant articles based on their title and abstracts, 7 full texts were screened. Inter-reviewer agreement between the investigators (V.R-R. and J.T-S.) was 100% with a Cohen's kappa index of 1 (perfect agreement). The reasons for rejecting four articles were as follows: an insufficient follow-up (19), duplicates studies (20,21) and included smoker patients (22). Finally, for the present review three articles (23-25) were selected (Fig. 1).
-Risk of bias assessment As shown on Fig. 2 (25) showed that the xenograft plus the collagen membrane group had a significantly better results in terms of PPD and CAL than the group that used the xenograft alone (P<0.05). None of the papers revealed statistically significant differences between groups with regard to adverse events. Out of the two studies that reported this outcome, three postoperative infections occurred in the grafted group and one in the control group (23). -Quantitative synthesis The same studies included in the qualitative synthesis were used to perform a pairwise meta-analysis comparing the use of xenograft covered by a collagen membrane after the removal of the M3M (23-25). We were unable to meta-analyse the adverse events outcome due to lack of data. The results of two articles (24,25) were pooled for PPD and CAL analysis. These studies involved 73 M3Ms in total. Quantitative analysis favoured the use of the xenograft plus collagen membrane over the spontaneous healing in terms of PPD gain (MD= 2.36; 95%CI 0.69 to 4.03; P=0.005; I2=97%) ( Fig. 3, Table 3) and CAL gain (MD=2.52; 95%CI 0.96 to 4.09; P=0.002; I2=95%) ( Fig.  3, Table 3). No statistically significant differences were found in terms of ABL changes (Fig. 3, Table 3). -Qualitative synthesis Across the three included trials in the present review (23)(24)(25), one of these studies had a multi-arm design (25). All included papers compared xenograft plus collagen membrane versus spontaneous healing (23)(24)(25), while the multi-arm study had also a group comparing the xenograft without membrane (25). In two trials (24,25), the xenograft covered or not by a

Discussion
The purpose of the present study was to answer the following clinical question: in M3M post-extraction sites, what benefit does the use of bone regeneration with xenograft in terms of PPD, CAL, ABL and adverse events when compared to spontaneous healing have? After performing the meta-analysis, our results revealed a significant PPD reduction and CAL gain comparing guided bone regeneration (xenograft plus collagen membrane) versus spontaneous healing. Moreover, the bone filling with xenograft and the spontaneous healing resulted in similar ABL gain and number of postoperative complications.
Periodontal defect in the distal site of the M2M is a common finding in patients undergoing M3M extraction. In fact, through the selected studies, up to 50% of the cases exhibited PPD of at least 7mm before the intervention (24,25). This resembles the results of Garaas et al. (26) in which 65% of the patients had a PPD ≥4 mm at the distal site of the M2M. The age of patients has their own relevance in bone regeneration of M3M sites. Kugelberg (27) demonstrated that patients older than 25 years old have a poor periodontal healing, which might cause periodontal pockets behind the M2M. In relation to this, the maximum age range of the patients included in our systematic review  Regarding the use of resorbable or non-absorbable membrane there are no statistically significant differences between them, however, second surgery is avoided when the resorbable membranes are used (35,36). In this review, only one study (25) compared the xenograft with or without membrane and the best outcomes were for the membrane group. Generally, bone regeneration increases the risk of postoperative complications (29,32,34), however, among the included studies, we did not obtain significant complications (23).
Although it would be interesting to obtain histological studies to observe whether tissue regeneration is formed, it is not clinically relevant since the objective of bone regeneration is that the patients do not have periodontal defects, being able to maintain stable over time.
Across the included studies, only one of them (25) provided histological results showing that with the use of a collagen membrane the level of xenogeneic particles was lower and more mature osteoid matrix (better bone quality) was observed at 6 months. Nevertheless, it is not essential since we have not evaluated this outcome.
In this review, all included studies (23-25) performed a scaling and root planning either in experimental or control groups. This procedure has been shown to remove plaque and calculus behind M2M and it consequently improves periodontal healing (37) therefore, the included studies could have been benefited from this procedure. There were several limitations related to the present study that must be mentioned. Firstly, only three papers which compared the guided bone regeneration with xenograft and the spontaneous healing were able to be included in our meta-analysis. There were no studies to compare by a meta-analysis the effectiveness of the xenograft without a collagen membrane. Moreover, the limited number of patients and M3M included together with the fact that only one paper of the selected studies had a low risk of bias, did not allow to make robust conclusions. Another possible drawback of this metaanalysis was the substantial heterogeneity across the selected studies. Thus, authors recommend being cautious with the results of the present review.

Conclusions
Within the above-mentioned limitations, it can be concluded that guided bone regeneration with xenograft and collagen membrane exhibited greater PPD reduction and CAL gain in the distal aspect of the M2M after the surgical extraction of the M3M than spontaneous healing. However, to confirm our results well-conducted investigations with larger samples and with a longer follow-up are needed.