4.1. Bone Volume Gain
Data retrieved by the analysed studies report an average successful outcomes for GBR procedures with titanium meshes and nonresorbable membranes with and without titanium reinforcement. The newly generated bone volume is always sufficient for a correct final implant placement, although in one of the studies  by Cucchi et al. two cases required an additional grafting procedure at the time of the implant placement.
The selected RCT compare or the use of titanium conventional meshes with and without coverage with resorbable membranes or the use of customized titanium devices, but no comparisons are made in the same study between the two different types of meshes.
Despite this last consideration, three different articles [25, 26, 28] report an average of vertical bone gain comparable between customized and non customized titanium meshes. These studies have been conducted by the same group of operators and, although the authors declared a randomization process of the treatments, the risk of bias seems quite relevant and it remains unclear when the choice of a customized device gives additional benefits to the procedure. None of the included studies report the different thickness between conventional and SLM fabricated titanium meshes, which might affect the final volume of regenerated bone, but mostly the chemical composition of the alloy used for the fabrication of Grade 4 customized meshes is different from the non customized ones, classified as Grade 5 .
Torres J. et al.  report the use of autogenous PRP to cover the titanium mesh and the final volume of bone seems to be higher in the group where the platelet derived membrane was employed. Beside the better protection of the graft and the wound, this results might be also related to a direct biological stimulation due to the increased amount of local growth factors [4, 10], but it doesn’t clarify whether the choice of the barrier is the real benefit to the procedure.
Mounir M. et al.  introduced polyethilenchetone (PEEK) as an alternative material for the fabrication of customized devices, but neither in the final bone volume, nor in the healing phase statistical differences where noticed between the two materials. One possible benefit could derive from the cost of the raw material and the consequent faster and simpler 3D printing process of a plastic material versus a metal alloy .
Titanium mesh shows excellent mechanical properties, its high strength and stiffness enable space maintenance and support for osteogenesis, its stability is mandatory to maintain bone graft volume during the healing, and the elasticity can reduce the oppression of oral mucosa . Due to its plasticity, titanium mesh can be adapted to different bone defects through bending and shaping. These features allow GBR with titanium mesh to show a high stable osteogenesis effect, and achieve instantaneous bone augmentation in horizontal and vertical directions .
Therefore the costs - benefits ratio seems to be directed more on the titanium rather than other non resorbable polymers for customized devices, since the second surgery is still needed for the device removal.
An interesting concept has been introduced by Cucchi et al. 2019  related to the pseudo-periostium classification. In essence, it consists on the quantification of the layer of connective tissue,that can be observed above the newly formed bone. In this study, the authors aim to evaluate the clinical and histologic features and to suggest a classification of this connective tissue after GBR with nonresorbable membrane (Group A) or titanium (Ti)-mesh plus resorbable membranes (Group B). Pseudo-periosteum was classified into Type 1 (no tissue or tissue < 1 mm); Type 2 (regular tissue between 1 and 2 mm); and Type 3 (irregular tissue or tissue > 2 mm). Results showed that the vertical bone gain was 4.2 ± 1.0 mm in Group A and 4.1 ± 1.0 mm in Group B. Group A had a higher bone density and greater amounts of type 1 periosteum than Group B (P = .01 for both). The results of this study show that both d-PTFE membranes and Ti-mesh plus collagen membranes are two valid options for bone augmentation in the mandible. However, nonresorbable membranes achieve higher bone density and a thinner pseudo-periosteum layer above the newly formed bone, which, in other words, can favor the total amount of newly generated bone.
4.2. Incidence of Complications
All the included studies evaluated the incidence of complications, which, for a better comprehension of the reader, have been analysed following the classification proposed by Fontana et al. in 2011 .
Basically, there are two types of complications: surgical, which may happen during the intervention, such as flap ruptures, damage to anatomical structures, and complications which occur during the healing phase, mostly related to the exposure of the barrier. Moreover, these latter events can be furtherdivided into four classes, according to
the presence and dimension of exposure, as well as the presence of a purulent exudate.
- Class I: Membrane exposure < 3 mm, no purulent exudate.
- Class II: Membrane exposure > 3 mm, no purulent exudate.
- ClassIII: Membrane exposure, with purulent exudate.
- ClassIV: Abscess, without membrane exposure.
Healing complications can be also divided into major or minor, depending on the influence on the regenerative process for newly formed bone . Two studies Cucchi et al. 2017 and Cucchi et al. 2021 [24, 25] follow these criteria; the study conducted in 2017 reports four surgical complications and seven healing complications. Surgical complications were related to neurological paresthesias but no vascular or flap damages. Out of the seven healing complications, two cases in the group treated with PTFE membranes affected the amount of new bone or the success of the bone augmentation surgery and they were, therefore, classified as major complications. In the group treated with Titanium meshes, four cases of healing complications were observed, leading to an overall complication rate of 21.1% . Of these, three (Class III and IV) were classified as major complications and one (Class II) was classified as a minor complication. No statistically significant difference was observed between the two groups regarding the healing complication rates (P = .69) or the major or minor healing complication rate (P = .99).
In the other study  Surgical/technical complication rates were 13.3% and 26.7% for the analysed groups, respectively, and no statistically significant differences were observed (p-value = .65). Regarding healing complications, three early and two late exposures of the meshes occurred for a total of 5 (two class 2; three class 3), and two infections without exposure (two class 4) were observed during the healing time. In these seven cases, complications were managed as follows: early exposed meshes were removed from one to three months after surgery; in cases of late exposure or infection, the meshes were removed within seven days after the complications were observed. In all cases, implants were placed as planned, but in two cases an adjunctive GBR procedure was required.
The other studies included in this review do not report a detailed classification of the complications and generally attribute unfavourable events to exposure, which, eventually, did not jeopardize the final implant placement.
These results are in accordance with a previously published systematic review  on the use of titanium meshes for bone augmentations, where three outcome variables were defined: a) horizontal and vertical bone regeneration, b) complication rate, defined as the percentage of membrane exposures and c) evaluation of implant survival, success and failure rate. The final results were comparable with those reported in case of bone regeneration obtained through other types of non-resorbable membranes. An advantage in favour of the titanium mesh was found in terms of bone loss after exposure, as implant placement was not jeopardized in almost all of the cases.
One of the advantages for customized devices is the precision and the better fitting onto the defect, therefore it can be expected that surgical complications are less than those where conventional barriers are employed. However, it doesn’t seem from the analysed studies, that healing complications are reduced with the use of customized titanium meshes. In a retrospective study published in 2020  the aim was to evaluate a new protocol for customized bone augmentation in a digital workflow. Patients and augmentation sites were retrospectively analysed based on defect regions, demographic factors, healing difficulties and potential risk factors. In 25% of the cases, exposures of the meshes were documented. Within this exposure rate, most of them were slight and only punctual (A = 16.2%), like one tooth width (B = 1.5%) and completely (C = 7.4%).
Data extracted from the included studies and those reported from the above -mentioned one do not show a lesser exposure rate of the customized meshes when compared to conventional ones.
One of the possible explanations can be related to two factors: the alloy used for the fabrication of customized devices and the thickness.
The titanium alloy used for the production of the meshes is Grade 4, which has a different composition compared to the Grade 5 alloys, used for customized devices; the unexpected presence of carbon retrieved in two studies [15, 16] demonstrates a different chemical composition of the alloy and can be related to a different host response.
Another important aspect is the thickness; infact traditional meshes are produced with an average thickness of 0,3mm, while customized devices are thicker (from 0,5mm to 0,8mm according to the manufacturer). As discussed above, the layer of soft tissues used for the surgical closure is extremely thin and represented only by the epithelial part, thus an unexpected trauma or simply a sudden movement of the lip or the chin may generate tension on the flap, which can easily relapse.