**4. Discussion**

The objective of this systematic review and NMA was to investigate the influence of different GTR techniques used as adjuncts to endodontic surgery and analyze their efficacy, assessed in terms of success rates. The results of the NMA show that the success rate of

endodontic surgery can be improved using GTR techniques as adjuncts, and combined therapy with bone grafts plus membranes results in a higher success rate.

Since the NMA did not show heterogeneity or inconsistency (Q = 1.16; *p* = 0.2821), the present NMA satisfied the assumption of transitivity, indicating that there were no systematic differences among the compared techniques other than the GTR techniques being compared [33]. Evaluating the transitivity assumption is critical, because the existence of intransitivity will bias treatment effect estimates [12]. Therefore, the calculated OR (3.6; *p* < 0.05) for the comparison between membrane plus bone grafting and endodontic surgery alone indicates that the success rate of this combination was almost four times higher than that of surgery without an adjunct GTR technique.

Most authors highlighted the relevance of membranes to promoting the healing of bone defects and preventing adjacent soft tissue ingrowth. The use of a membrane alone, without a bone graft, was 1.02 times more effective than endodontic surgery without a GTR technique (control), and more effective than platelet-rich plasma techniques. However, the membrane plus bone graft combination was 3.6 times more successful than membrane only.

The success rate of the combined membrane plus bone graft was 3.7 times higher than that of endodontic surgery alone (control). Parmar et al. (2019) reported a nonsignificant radiographic reduction in periapical bone defects regenerated using a resorbable collagen membrane. Complete periapical healing was observed in the control group, with rates of 60 to 80% and 53.3 to 73.3% of those of the membrane group, depending on the radiodiagnostic technique [32].

Marin-Botero et al. (2006) also reported that polyglactin-910 resorbable membranes had little influence on the complete healing of periapical bone defects after endodontic surgery (40%) compared with the control treatment (60%) [26]. Garret et al. (2002) reported that resorbable membranes did not show a statistically significant (*p* > 0.05) radiographic reduction in periapical bone defects after endodontic surgery. Additionally, they did not recommend the use of resorbable membranes for bone defects with four walls that are confined to the apical region [34]. Santamaria Zuazua et al. (1998) analyzed the bone density and radiographic reduction in periapical bone defects after endodontic surgery using resorbable and non-resorbable membranes, and found no statistically significant difference (*p* > 0.05) in bone density at 6 months after surgery between the two types of membranes.

These results sugges<sup>t</sup> that GTR techniques using membranes do not contribute to increased periapical bone regeneration regardless of the membrane type [35]. However, Taschieri et al. (2011) retrospectively analyzed clinical and radiographic periapical bone healing after endodontic surgery procedures using a collagen resorbable membrane and recommended its application for through-and-through lesions [36]. Goyal et al. (2011) analyzed the impact of membranes and platelet-rich plasma on the complete periapical healing of periapical bone defects after endodontic surgery. They found no statistically significant differences (*p* > 0.05) among the membrane alone, platelet-rich plasma alone, and the two combined [30]. Dhiman et al. (2015) reported no statistically significant difference (*p* > 0.05) in the clinical and radiographic reduction in periapical bone defects after endodontic surgery using platelet-rich plasma techniques with respect to the control group [32].

Most authors reported that bone grafts stimulate bone defect healing and prevent adjacent soft tissue collapse [4,23,37–39]. Kattimani et al. (2014) highlighted the use of bovine-derived and synthetic hydroxyapatite bone grafts for the radiographic reduction in periapical bone defects after endodontic surgery. They found no statistically significant difference (*p* > 0.05) in radiographic reduction between the two bone graft materials [38]. Kattimani et al. (2016) also compared the clinical and radiographic outcomes of bovinederived and synthetic hydroxyapatite bone grafts after endodontic surgery. They found no statistically significant (*p* > 0.05) difference between the two bone graft materials at 6-month follow-up [39]. Stassen et al. (1994) also analyzed the clinical and radiographic effects of bovine-derived hydroxyapatite bone grafts and did not recommend their use as

adjuncts in endodontic surgery [23]. However, Sreedevi (2011) reported complete clinical and radiographic periapical bone healing after endodontic surgery using hydroxyapatite bone graft material with respect to the control group [4].

Other bone graft materials have been used as adjuncts to GTR techniques in endodontic surgery. Pantchev et al. (2009) retrospectively analyzed the clinical and radiographic outcomes of a synthetic bioactive glass material used as a bone graft after endodontic surgery. They found a statistically significant difference (*p* < 0.05) at short-term follow-up (9–24 months), but no statistically significant difference (*p* > 0.05) at long-term follow-up (33–48 months) [37]. It is more difficult to apply endodontic surgery using a GTR technique to 4-wall defects and through-and-through lesions because of the higher risk of soft tissue collapse and decreased stability of the bone regeneration material. Pecora et al. (2001) demonstrated that the addition of calcium sulfate as a bone graft material in GTR techniques for the treatment of through-and-through lesions improves the clinical outcome [24]. However, Taschieri et al. (2007, 2008) showed no statistically significant difference (*p* > 0.05) after endodontic surgery when using resorbable collagen membrane and bovine-derived hydroxyapatite bone graft material for through-and-through lesions [27] and four-wall defects [29]. Tobon et al. (2002) reported that the simultaneous use of nonresorbable membrane and bovine-derived hydroxyapatite bone graft material produced complete clinical and radiographic periapical bone healing after endodontic surgery [25].

In addition, the wound healing scales and indices used in oral surgery do not capture the relationships between outcome parameters; therefore, Hamzani et al. (2018) proposed a novel scale that allows the assessment of wound healing phases [40]. Recently, Haj Yahya et al. (2020) described a novel procedure for measuring the healing process after surgical extraction based on an inflammatory proliferative remodeling scale that could also be used in further studies for the assessment of wound healing following endodontic surgery [41].

A limitation of this systematic review and meta-analysis is the possibility that not all articles related to the selection criteria were identified, although the risk was decreased because three databases were searched. In addition, most of the studies were of poor quality, according to the Cochrane Collaboration tool [16]. Furthermore, the most effective GTR technique (MB + Os) was only included in a single study. Therefore, further, better designed clinical studies with higher quality are necessary.
