Interventions for Dental Implant Placement in Atrophic Edentulous Mandibles: Vertical Bone Augmentation and Alternative Treatments. A MetaAnalysis of Randomized Clinical Trials

Background: The purpose of the current study is to assess which vertical bone augmentation techniques are most effective for restoring atrophic posterior areas of the mandible with dental implants and compare these procedures with alternative treatments. Methods: Electronic literature searches in PubMed (MEDLINE), Ovid, and the Cochrane Library were conducted to identify all relevant articles published up to July 1, 2015. Eligibility was based on inclusion criteria, and quality assessments were conducted. The primary outcome variables were implant and prosthetic failure. After data extraction, meta-analyses were performed. Results: Out of 527 potentially eligible papers, 14 randomized clinical trials were included. Out of these 14 studies, four trials assessed short implants (5 to 8 mm) as an alternative to vertical bone augmentation in sites with a residual ridge height of 5 to 8 mm. No statistically significant differences were found in implant (odds ratio [OR]: 1.02; 95% confidence interval [CI]: 0.31 to 3.31; P = 0.98; I2: 0%) or prosthetic failure (OR: 0.64; 95% CI: 0.21 to 1.96; P = 0.43; I2: 0%) after 12 months of loading. However, complications at treated sites increased with the augmentation procedures (OR: 8.33; 95% CI: 3.85 to 20.0; P <0.001; I 2: 0%). There was no evidence of any vertical augmentation procedure being of greater benefit than any other for the primary outcomes (implant and prosthetic failure). Conclusions: Short implants in the posterior area of the mandible seem to be preferable to vertical augmentation procedures, which present similar implant and prosthetic failure rates but greater morbidity. All the vertical augmentation technique comparisons showed similar intergroup results. J Periodontol 2016;87:1444-1457.


S
ince Bra ˚nemark et al. 1 defined osseointegration in the mid-1960s, oral rehabilitation with dental implants has become a common practice, with reliable and safe long-term results. 2 However, presence of an adequate bone volume is mandatory to achieve optimum treatment outcomes. 3Lack of sufficient bone height to place dental implants in the posterior mandible due to the presence of the inferior alveolar nerve is a common scenario in partially edentulous patients.Hence, an atrophic posterior mandible presents a great challenge for successful rehabilitation. 4lthough removable partial dentures are the most common and simplest option, many patients seek fixed prosthesis treatment.However, atrophy precludes use of standard size implants in many cases.][6] Vertical bone augmentation techniques using guided bone regeneration (GBR), [7][8][9][10][11][12][13][14] alveolar distraction osteogenesis, 3,[15][16][17][18][19][20][21] interpositional block grafts, 19,[22][23][24][25][26][27][28][29][30] or onlay bone grafting 20,29,[31][32][33][34][35] have shown favorable outcomes, both clinically and histologically. 4However, these procedures cannot be considered the standard of care due to the high rate of postoperative complications.Also, information on long-term results ( ‡10 years of follow-up) is scarce. 526][27]31 Although there is still no consensus regarding the cutoff length between short and standard implants, 36,37 Renouard and Nisand 38 defined short implants as devices with an intrabony length of 8 mm or less.Several studies have reported successful outcomes in terms of implant and prosthetic survival as well as implant success rates in short-term followups. 5,39,40Nevertheless, there are still concerns regarding consequences of peri-implant bone loss and its impact on long-term success rate.Also, placement of short implants requires a certain amount of bone above the mandibular canal, which is not always available.
2][43][44][45] However, both temporary and permanent neurosensory dysfunction as well as mandibular fractures may occur postoperatively and overshadow the high survival rates reported. 44,45lthough two meta-analyses have been published on this topic, in 2009 5 and 2014, 40 they did not include some well-designed recently published trials.Furthermore, one of these reviews 40 also analyzed the outcomes of implants placed in the maxilla.Thus, a new meta-analysis of randomized controlled trials centered exclusively on the posterior mandible may add new information.
Aims of the present study are to analyze all relevant data from randomized clinical trials (RCTs), to assess which vertical bone augmentation techniques are most effective for restoring atrophic posterior areas of the mandible with dental implants, and to compare vertical bone augmentation procedures with alternative treatments.

MATERIALS AND METHODS
This meta-analysis complies with the Quality of Reporting of Meta-Analyses Statement. 46udy Selection Criteria Inclusion criteria were: 1) RCTs (including split-mouth designs) that considered the effect of two different vertical bone augmentation procedures, or two different biomaterials for the same vertical bone augmentation technique, on the outcome of atrophic posterior mandible implant rehabilitation; and 2) RCTs (including split-mouth designs) comparing vertical bone augmentation with alternative surgical treatments, such as short implants (length £8 mm 38 ) or inferior alveolar nerve transposition/lateralization.

J Periodontol • December 2016
Camps-Font, Burguen ˜o-Barris, Figueiredo, Jung, Gay-Escoda, Valmaseda-Castello ´n The posterior area of the mandible was classified as atrophic when bone height from the alveolar crest to the inferior alveolar nerve canal did not allow placement of standard length dental implants (length >8 mm).
The present review excluded trials with <1 year of follow-up after loading the implant-supported prosthesis.Implant placement, abutment connection, and yearly follow-up visits after prosthetic loading were used as time points.The predefined study population, intervention, control group, and outcome parameters for eligibility of studies are summarized in Table 1.
Primary outcome measures were as follows: 1) Implant survival: biologic failures were defined as implant mobility or removal of stable implants caused by progressive marginal bone loss (MBL) or infection.Mechanical failures were considered to comprise any mechanical complication, such as implant fractures or platform deformations, which rendered the implant unusable.Biologic failures were classified as early (failure to establish osseointegration before prosthetic loading) or late (failure to maintain the established osseointegration). 82) Prosthesis survival: prosthetic failures were defined as failure to position the planned prosthesis due to implant failure(s) or loss of the prosthesis secondary to implant failure(s). 8econdary outcome measures were as follows: 1) postoperative complications at augmented and/or donor sites (e.g., infection, nerve injury, hemorrhage) before prosthetic loading; 2) augmentation procedure failure: inability to position implants planned, not affecting survival of implant actually inserted; 3) biologic complications: implant function disturbances characterized by involvement of supporting tissues (e.g., periimplantitis); 4) technical complications: mechanical damage to implants, implant components, and/or suprastructures (e.g., fractures of implants, screws, or abutments, fractures or deformations of the framework or veneers, or screw or abutment loosening); 5) radiographic bone gain (expressed in mm or as a percentage); 6) radiographic peri-implant MBL: marginal bone level changes over time, from baseline to last follow-up appointment (expressed in mm or as a percentage); and 7) patient preference (only in split-mouth trials).

Search Strategy
An electronic search of PubMed (MEDLINE), Ovid, and the Cochrane Library databases up to July 1, 2015 was conducted to identify all relevant human RCTs without year or language restrictions.
The following search terms were applied: [(vertical bone augmentation OR vertical ridge augmentation OR vertical ridge regeneration OR vertical bone regeneration OR guided bone regeneration OR bone graft OR block graft OR interpositional bone graft OR distraction osteogenesis) AND (''Dental Implants''[Mesh] OR short dental implants OR inferior alveolar nerve lateralization OR inferior alveolar nerve transposition) AND posterior mandible].
The search was completed by manual screening of references cited in the selected articles and reviews.

Selection of Studies
Two examiners (OC-F and GB-B) independently selected studies in accordance with inclusion criteria.Any disagreements were resolved by consensus.
Initially, duplicates or irrelevant publications (based on title) were excluded, and abstracts were examined.Finally, full texts of all remaining papers were assessed.Studies removed at this stage and reasons for their exclusion were recorded.
When multiple reports on the same patients were identified, the publication with the longest follow-up was included.

Data Extraction and Method of Analysis
Two reviewers (OC-F and GB-B) independently extracted data using data-extraction tables.Whenever possible, the following data were retrieved from the selected papers: 1) author(s); 2) year of publication; 3) country of origin; 4) study design; and 5) details of participants, intervention(s), and outcomes.

Risk of Bias Assessment
Two reviewers (OC-F and GB-B) independently assessed risk of bias of the RCTs included as part of the data extraction process, using the Cochrane Collaboration tool for assessing risk of bias, suggested in the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0). 47The following items were evaluated: 1) sequence generation; 2) allocation concealment; 3) examiner and patient masking; 4) outcome masking; 5) incomplete outcome data addressed; 6) selective reporting; and 7) other sources of bias, such as conflict of interest.Publications were grouped into the following categories: 5 A) low risk of bias (possible bias not seriously affecting results) if all criteria were met; B) high risk of bias (possible bias seriously weakening reliability of results) if one or more criteria were not met; C) unclear risk of bias when too few details were available for classification as high or low risk.Authors were contacted for clarification of missing or unclear information when necessary.

Statistical Analyses
Statistical analysis was carried out with statistical software.† † For dichotomous outcomes, odds ratios (ORs) with 95% confidence intervals (CIs) were used to estimate effect of an intervention.Parametric and non-parametric tests (Pearson x 2 , Fisher, and Mann-Whitney tests) were used to compare groups.For continuous outcomes, mean differences and standard deviations (SDs) were used to summarize data for each group.The level of significance was set at a P value <0.05.
In parallel group studies, the statistical unit was the patient, not the augmentation procedure or implants.In split-mouth designs, augmentation procedures or prostheses used in each pair were the unit of analysis. 48 meta-analysis was only performed when there were studies comparing similar techniques and reporting the same outcome measures.ORs and mean differences were combined for dichotomous and continuous data, respectively, using random-effects models.Data from split-mouth studies were combined with data from parallel group trials using the generic inverse variance method. 49atistical heterogeneity was estimated by means of x 2 (Q value) and I 2 analyses.A x 2 P value of <0.05 and an I 2 value of >50% were interpreted as significant heterogeneity. 50ad there been a sufficient number of meta-analyzed trials (more than 10), publication bias and clinical heterogeneity assessment, as well as sensitivity analyses, would have been performed according to Higgins and Green. 47

Study Selection and Description
The initial electronic database search yielded 527 references, and three additional papers 26,27,34 were included after hand searching reference lists for pertinent articles and reviews.After duplicate removal and assessment of both title and abstract, a total of 26 articles were eligible for full-text analysis.Reviewer agreement was 95.2%, with a k index of 0.90 (almost perfect agreement).
Twelve publications were excluded after applying study criteria: seven were removed as more recent data were available; [51][52][53][54][55][56][57] two trials were excluded as the followup after prosthetic loading was <1 year; 58,59 and another three papers were rejected because of retrospective design, 31 absence of a control group, 35 and not presenting vertical bone augmentation procedures, 60 respectively.

Risk of Bias Assessment
All studies included were considered to have a high risk of participant and clinician/researcher masking bias due to difficulties in masking the selected treatment. 61ence, results of the present systematic review and meta-analysis should be interpreted with caution.Flowchart illustrating the study selection process.
Extraction Data Qualitative synthesis.3,19,20,24-30,34 None of the studies revealed significant differences between groups in terms of implant and prosthesis failure rates (P >0.05).

DISCUSSION
The present study, which used recommended methods for systematic reviews and meta-analyses, aimed to compare effectiveness of different vertical ridge augmentation procedures and then compare these procedures with treatment alternatives.However, due to the lack of data on many therapeutic approaches, the only comparison that could be made was between alveolar ridge augmentation and short implants or inferior alveolar nerve transposition.][26][27] However, these results should be treated with caution since all four trials had a potential risk of bias.In addition, they were all conducted by the same research group, using a similar augmentation protocol, and with limited follow-up and sample sizes.What is more, their internal validity might be compromised since they were all conducted mainly in multiple private practices, and in two of the studies, operations were not performed by the same surgeon, leading to potential operator-dependent   bias. 26,27This issue could affect reliability and quality of the studies.In addition, the small number of participants in all these studies might have led to a Type 2 error (failure to reject a false null hypothesis).Indeed, if implant failure is defined as the primary outcome and a 0.2 difference between groups is considered clinically significant (as proposed by Felice et al. 25 ), only one study had a statistical power greater than 70%. 27Furthermore, due to the small number of papers available for review, no evaluation of publication bias (i.e., funnel plot) could be made. 47Finally, all but one of the selected articles reported on studies performed in Italy, so their external validity would seem to be threatened as well. 30Another possible limitation of the present paper is that gray literature and hand searching of related journals were not included in the search strategy, so some studies might have been neglected.Pen ˜arrocha-Oltra et al. 31 described similar survival and success rates for implant treatment in sites with vertical bone defects involving autogenous onlay block grafts or short dental implants, but the rate of complications was significantly higher in the augmented group (this article was excluded because of its retrospective design).
On the other hand, a recent meta-analysis of RCTs showed no differences among the groups regarding implant survival/success and complications. 40A possible explanation for this difference could be related to the fact that none of the four articles selected for that paper were considered in the present review: two trials 51,52 were replaced by others with more recent data, 24,25 one study included augmentation techniques in the posterior maxilla, 62 and one paper was rejected because standard implants were placed in native bone. 63reatment duration and cost may also play an important role in the decision and willingness of the patient to undergo vertical ridge augmentation. 64Undeniably, bone regeneration techniques entail additional biologic and financial costs associated with one additional surgical procedure, a bone substitute, and a barrier membrane, and at least an additional 4 months to complete treatment.In the studies reviewed, almost twice the number of patients preferred short implants to augmentation procedures, although the difference was not significant, probably due to the small number of meta-analyzed studies reporting patient-based outcomes. 24,276][67] However, it seems reasonable to assume that peri-implant bone loss around short implants can be more critical than in standard implants, as loss of osseointegration can occur in a short time span. 65Therefore, it is crucial to control main risk factors for peri-implant diseases and apply strict maintenance programs for long-term performance of these implants.
Information on inferior alveolar nerve transposition is scarce, highlighting the need to perform RCTs to assess whether this approach offers advantages over other surgical techniques in the posterior area of the mandible.Thus, in the opinion of the authors, until data from well-designed RCTs become available for analysis, other options such as use of short implants or augmentation techniques seem preferable.
Distraction osteogenesis allows more bone gain 19 and a reduction in treatment time. 3,19,20However, other procedures such as GBR or onlay block grafting techniques may permit simultaneous bone widening if needed. 5utogenous bone is often considered the ''gold standard'' material for bone augmentation procedures. 5evertheless, three trials compared autogenous grafts with bone substitutes and observed no differences for any clinical outcomes registered. 12,28,30Indeed, Felice et al. 28 reported eight out of 10 patients preferred augmentation procedures with a bone substitute, probably due to lower postoperative morbidity.On the other hand, from a histomorphometric point of view, two of the studies revealed more residual grafted material at implant placement in the group treated with bone substitutes. 28,30Moreover, implants placed in bone augmented with substitutes showed a tendency to increased MBL. 12,28Further research is needed to clarify which graft material is most cost-effective.
Similarly, non-resorbable titanium-reinforced expanded polytetrafluoroethylene (ePTFE) membranes are traditionally considered the benchmark for vertical GBR. 8,9,68,69However, an increased rate of soft tissue complications after premature membrane exposure has been reported as a major disadvantage of these barriers. 70Once exposed to the oral cavity, their porous surface (100-to 300-mm porosity) 13 is rapidly colonized by bacteria, leading to infections of adjacent tissues that require early membrane removal, resulting in impaired bone regeneration. 71,72Another disadvantage of non-resorbable membranes is the need for re-entry surgery and membrane removal, which is associated with patient morbidity.To overcome such drawbacks and simplify surgical protocols, resorbable membranes have been proposed. 73Merli et al. 11 revealed similar outcomes with fewer postoperative complications when using resorbable barriers in comparison with non-resorbable ones after a 6-year follow-up period.Nevertheless, when ePTFE was used, a higher bone gain was recorded, and less periimplant MBL was registered over time.

CONCLUSIONS
Placement of short implants (5 to 8 mm) seems to be the best option for treating atrophic posterior areas of the mandible, since this approach is less invasive and has a significantly lower complication rate when compared with more demanding grafting procedures.Furthermore, survival rates and marginal bone level changes after 1 year of loading seem similar.To confirm these results, large-sample studies involving several centers and countries, different surgical protocols, and patient-centered outcomes should be conducted.
Different surgical protocols for vertical bone augmentation seem to provide similar intergroup results.Bone substitutes usually entail less postoperative morbidity and may be a valid alternative to autogenous bone.Well-designed RCTs are needed to determine which bone augmentation techniques are more effective, simpler, and safer and have better long-term results.
Figure 2 summarizes the quality of RCTs included.

Figure 2 .
Figure 2.Risk of bias assessment of selected studies.+ = low risk of bias; -= high risk of bias; ?= unclear risk of bias.
Forest plots (OR) for implant failure (A), prosthetic failure (B), and postoperative complications (C) comparing short implants (test group) with interpositional block xenografts covered with a resorbable membrane to receive long implants (control group) at 12 months' follow-up.IV = independent variable.

Table 3 .
Comparison of Selected Studies